Endocrinology

Research Quantitative heel ultrasound as a predictor for osteoporosis Vasi Naganathan, Lyn March, David Hunter, Nick A Pocock, Joanna Markovey and Philip N Sambrook MJA 1999; 171: 297-300 For related articles, see Prince, Maguire & Lobb et al Abstract - Introduction - Methods - Results - Discussion - References - Authors' details - - More articles on Rheumatology Abstract Objective: To determine the diagnostic value of quantitative ultrasound (QUS) to predict bone mineral density (BMD) categories as defined by dual-energy x-ray absorptiometry. Design: Cross-sectional survey. Setting: Rheumatology department of a tertiary care hospital (Royal North Shore Hospital, Sydney, NSW), 1997-1998. Subjects: 326 healthy women aged 45-80 years who had volunteered for a twin study. Our study included both members of non-identical twin pairs but only one randomly selected member of identical twin pairs. Main outcome measures: BMD categories as defined by dual-energy x-ray absorptiometry of lumbar spine and left hip, and QUS of calcaneus; sensitivity, specificity and likelihood ratios (LRs) of QUS parameters to diagnose osteoporosis as defined by BMD. Results: The sensitivity of QUS to diagnose BMD osteoporosis varied between 9% and 47%, depending on the QUS parameter. The specificity of QUS was high (88%-100%). If all QUS parameters were normal, osteoporosis was unlikely (LR, 0-0.2). One QUS parameter, broadband ultrasound attenuation (BUA), was highly predictive of osteoporosis by BMD when in the osteoporotic range (LR, Infinity), but had low sensitivity (9%). QUS results in the osteoporotic range for other parameters and all QUS results in the osteopenic range were less predictive (LR, 1.0-5.2) of osteoporotic BMD. Conclusion: These results suggest that, for most of those tested for osteoporosis by QUS in the community, uncertainty remains about expected BMD. Introduction Quantitative heel ultrasound has recently been introduced in Australian pharmacies as a "screening" tool for osteoporosis. The technology is relatively cheap, radiation-free and portable, but its accuracy in diagnosing osteoporosis is unclear. Bone mineral density (BMD), measured by dual energy x-ray absorptiometry (DEXA), is the best predictor of fracture risk and is currently considered the "gold standard" for diagnosing osteoporosis. Although prospective studies have shown that quantitative ultrasound (QUS) predicts future fracture risk independently of BMD,1,2 most women with abnormal results will proceed to formal BMD measurement to assess the need for therapeutic intervention; women with normal QUS results may be reassured they do not need BMD measurement. It is unclear how many women have unnecessary further investigations or are falsely reassured. When used in this way, QUS has diagnostic value only if it can accurately predict BMD categories as determined by DEXA. Our aim was to examine the role of QUS in predicting BMD diagnostic categories. We determined conventional sensitivity and specificity, as well as likelihood ratios (LRs). These have the advantage of allowing test results to be assessed for several diagnostic categories rather than only at a single cut-off between "normal" and "abnormal".3 Methods Subjects and setting The study was a cross-sectional survey of healthy women aged 45-80 years who had volunteered to take part in a twin study. They were recruited from the Australian Twin Registry and media advertising. Our ultrasound study included both members of each non-identical twin pair and one randomly selected member of each identical twin pair. The study was conducted in the Rheumatology Department of the Royal North Shore Hospital, Sydney, NSW (a tertiary care hospital), in 1997 and 1998. It was approved by the hospital's Human Research Ethics Committee. Assessment Subjects had BMD measurements of their lumbar spine (L1-L4) and left hip (neck of femur and total hip) by DEXA using a Hologic QDR450 instrument (Hologic Inc, Waltham, Mass, USA). The same machine was used on all patients. QUS of the left calcaneus was performed on the same day using a CUBA Mark II ultrasound instrument (McCue Ultrasonics, London, UK). The two most commonly used QUS parameters were measured: broadband ultrasound attenuation (BUA), which is thought to reflect bone mass and architecture, and velocity of sound (VOS), which reflects mass and elasticity of bone.4 Analyses Each BMD and QUS value was converted to a T score (number of standard deviations from the population mean for young, healthy, sex-matched adults). This population mean was estimated from measurements in 50 women aged 20-30 years who also took part in the twin study. T scores were used to categorise BMD values as normal (T > -1) or indicating osteopenia (T, -2.5 to -1) or osteoporosis (T < -2.5), as proposed by a working party of the World Health Organization.5 QUS values were classified in the same way. Although no consensus has been reached on what T-score cut-offs and diagnostic categories to use with QUS, the instrument used commonly in Australian pharmacies uses the WHO criteria and cut-off values. Subjects were classified as having osteoporosis if at least one of the three BMD measurements (lumbar spine, neck of left femur or total left hip) indicated osteoporosis, and as having osteopenia if at least one measurement indicated osteopenia but none indicated osteoporosis. BUA and VOS results were combined as a cQUS category: this was defined as normal if both results were normal, as osteopenic if either indicated osteopenia but neither indicated osteoporosis, and as osteoporotic if either indicated osteoporosis. Some QUS scanners calculate a stiffness parameter (unrelated to mechanical stiffness) from a linear combination of normalised BUA and VOS values. We calculated stiffness in an analogous manner,6 and categorised it as normal, osteopenic or osteoporotic based on T scores in the same way as other QUS values. We calculated the sensitivity and specificity of QUS parameters in predicting BMD-defined osteoporosis and osteopenia. We also calculated the likelihood ratio (LR) for each QUS result (sensitivity/1 - specificity), defined as the ratio of the probability of the particular QUS result (normal, osteopenic or osteoporotic) in women with BMD-defined osteoporosis or osteopenia to the probability of the same result in women with normal BMD.7 As there is no consensus on what QUS cut-off values should be used to diagnose osteoporosis, the statistical analyses were repeated using a range of QUS T-score cut-off values for osteoporosis between -2.5 and -1.0. Results Subjects and osteoporosis There were 326 subjects, with mean age 58.5 years; 255 (78%) were postmenopausal. Of the 326, 47 (14%) had a BMD measurement indicating osteoporosis at one or more of the three sites where BMD was measured, and a further 160 (49%) had a measurement indicating osteopenia. QUS results are compared with BMD results in Box 1. The percentage of women with values in the osteoporotic range varied between QUS parameters (1% for BUA, 17% for VOS and the combined BUA-VOS category, and 14% for stiffness). Sensitivity and specificity of QUS Sensitivity and specificity of QUS for predicting BMD diagnostic categories are shown in Box 2. Sensitivity and specificity varied between QUS parameters. A BUA result in the osteoporotic range (T < -2.5) had very low sensitivity for predicting BMD-defined osteoporosis (9%), but high specificity (100%). In contrast, VOS, cQUS and stiffness results in the osteoporotic range had sensitivities of almost 50%, and specificities that were again high. For predicting either osteoporosis or osteopenia, stiffness had the best combination of sensitivity (77%) and specificity (81%). Positive and negative predictive values are also shown in Box 2. Negative predictive values were high (87%-91%) for QUS as a predictor of BMD-defined osteoporosis versus osteopenia/normal BMD. This indicated that a woman with BMD-defined osteoporosis was unlikely to have a QUS result in the normal-osteopenic range. Likelihood ratios of QUS LRs for different QUS results to predict BMD-defined osteoporosis are summarised in Box 2 and interpreted in Box 3. Normal QUS result: A BUA, VOS, cQUS or stiffness result in the normal range had a low LR (0-0.2) (ie, a normal result significantly lowered the odds or probability of the woman's having BMD-defined osteoporosis). Osteoporotic QUS result: A BUA result in the osteoporotic range had an LR approaching infinity and so was highly predictive of BMD-defined osteoporosis. In contrast, a VOS, cQUS or stiffness result in the osteoporotic range had a much lower LR (4.0-5.2), increasing the odds of BMD-defined osteoporosis, but to a lesser extent than a BUA result in the osteoporotic range. Osteopenic QUS result: QUS results in the osteopenic range were less predictive, as LR ranged from 1.0 to 2.4. Between 37% and 50% of subjects (depending on the QUS parameter) had results in this range (Box 1). LRs for predicting low BMD (ie, osteoporosis or osteopenia; BMD T score < -1) are also shown in Box 2, and followed a similar pattern to LRs for predicting BMD-defined osteoporosis. When QUS T-score cut-off values for osteoporosis were increased from -2.5 to -1.0, sensitivity increased, but at the expense of decreasing specificity and LR (data not shown). For example, a BUA cut-off of -1.0 gave 83% sensitivity, 69% specificity and LR, 2.6. Corresponding values for a VOS cut-off of -1.0 were 96%, 41% and 1.6. Discussion We found that QUS had variable usefulness in predicting BMD categories. Specificity for predicting BMD-defined osteoporosis was high for all QUS parameters (88%-100%), but sensitivity was low and variable (9%-47%). A BUA result in the osteoporotic range was highly predictive of BMD-defined osteoporosis (LR, Infinity), but had low sensitivity (9%). Results in the osteoporotic range for other QUS parameters and in the osteopenic range for all QUS parameters were less predictive of BMD-defined osteoporosis (LRs, 1.0-5.2). In the light of our study, how can we interpret QUS results? If a BUA result is in the osteoporotic range (LR, Infinity), then BMD-defined osteoporosis is almost certain (predictive value, 100%). However, the low sensitivity of BUA (9%) means that many women with osteoporosis would be missed if BUA alone were used. If results are normal for both BUA and VOS (cQUS normal; LR, 0), we can confidently rule out BMD-defined osteoporosis. If LR is 0, then, no matter what the pre-test probability of osteoporosis, the post-test probability will be < 5% (Box 3). All results in the osteopenic range, and VOS and stiffness results in the osteoporotic range, are less predictive of BMD category. Therefore, if QUS were performed on a population similar to ours (14% prevalence of osteoporosis), then (from Box 1) 1% would have a BUA result in the osteoporotic range (likely to have osteoporosis) and 33% would have both BUA and VOS results in the normal range (osteoporosis could fairly confidently be ruled out, with a post-test probability < 5%). However, there would be a degree of uncertainty about the remaining 66%, who would then need a DEXA scan to identify those with osteoporosis. Previous studies of QUS as a predictor of BMD have generally used conventional sensitivity and specificity analyses only, not LRs, and have not used the WHO BMD definitions. For example, two community-based cross-sectional studies on 700 postmenopausal10 and 1000 perimenopausal women,11 respectively, found that there was a 40%-50% overlap in the number of women in the lowest quartile of both DEXA and QUS measurements. Two other studies found QUS parameters to have a sensitivity of 65%-70% for BMD in the lowest quartile.6,12 Only one study other than ours has evaluated QUS in terms of WHO BMD definitions. It found BUA and VOS to have higher sensitivities, of 77% and 69%, respectively, for diagnosing osteoporosis in 100 women aged 60-69 years.13These higher sensitivities may have been due to use of higher BUA and VOS cut-off values. As expected, specificities were lower than in our study. The ultrasound instrument used in this study, the McCue Cuba Mark II, is not identical to the Achilles ultrasound instrument (Lunar, Madison, Wis, USA) used in Australian pharmacies. Nevertheless, a comparison of the two machines found that BUA measurements on a Cuba Mark II instrument were highly correlated with "stiffness" measurements on a Lunar Achilles instrument (r = 0.906; 95% CI, 0.873-0.931).14 Another study compared measurements of 30 women between the Cuba Mark II used in our study and an Achilles, finding an r value of 0.8.15 Therefore, it is unlikely that the Achilles instrument would be a significantly better predictor of BMD than our Cuba Mark II. There is no consensus on what cut-off values to use with QUS to diagnose osteoporosis. We found that changing the cut-off could achieve higher sensitivity, but only by accepting higher rates of false positives (lower specificity) and less discriminating LRs. Although there is enough evidence to support the use of QUS as an independent predictor of fracture risk,1,2 our study shows that QUS should not been seen as a substitute for BMD measurement. The results of our study suggest that, when women in the community are "screened" for osteoporosis using QUS, a few women will be confidently identified with BMD-defined osteoporosis. Another small group will be able to be reassured that they are unlikely to have osteoporosis. However, for the great majority, the presence or absence of osteoporosis will remain uncertain. References Hans D, Dargent MP, Schott AM, et al. Ultrasonographic heel measurements to predict hip fracture in elderly women: the EPIDOS prospective study. Lancet 1996; 348 (9026): 511-514. Bauer DC, Gluer CC, Cauley JA, et al. Broadband ultrasound attenuation predicts fractures strongly and independently of densitometry in older women. A prospective study. Study of Osteoporotic Fractures Research Group. Arch Intern Med 1997; 157: 629-634. Sackett DL, Haynes BR, Guyatt GH, et al. Clinical epidemiology. A basic science for clinical medicine. 2nd ed. Boston: Little, Brown and Company, 1991. Gluer CC, Wu CY, Jergas M, et al. Three quantitative ultrasound parameters reflect bone structure. Calcif Tissue Int 1994; 55: 46-52. World Health Organization Study Group. Assessment of fracture risk and its application to screening for prostmenopausal osteoporosis. World Health Organ Tech Rep Ser 1994; 843: 1-129. Herd RJ, Blake GM, Miller CG, et al. The ultrasonic assessment of osteopenia as defined by dual X-ray absorptiometry. Br J Radiol 1994; 67: 631-635. Fletcher RH, Fletcher SW, Wagner EH. Clinical epidemiology: the essentials. 3rd ed. Baltimore: Williams & Wilkins, 1988: 65. Fagan TJ. Normogram for Bayes theorem [letter]. N Engl J Med 1975: 293; 257. Australian National Consensus Conference 1996. The prevention and management of osteoporosis. Consensus statement. Med J Aust 1997; 167 Suppl: S1-S15. van Daele Burger H, Algra D, et al. Age-associated changes in ultrasound measurements of the calcaneus in men and women: the Rotterdam Study. J Bone Miner Res 1994; 9: 1751-1757. Massie A, Reid DM, Porter RW. Screening for osteoporosis: comparison between dual energy X-ray absorptiometry and broadband ultrasound attenuation in 1000 perimenopausal women. Osteoporos Int 1993; 3: 107-110. Young H, Howey S, Purdie DW. Broadband ultrasound attenuation compared with dual-energy X-ray absorptiometry in screening for postmenopausal low bone density. Osteoporos Int 1993; 3: 160-164. Langton CM, Ballard PA, Bennett DK, Purdie DW. A comparison of the sensitivity and specificity of calcaneal ultrasound measurements with clinical criteria for bone densitometry (DEXA) referral. Clin Rheumatol 1997; 16: 117-118. Greenspan SL, Bouxsein ML, Melton ME, et al. Precision and discriminatory ability of calcaneal bone assessment technologies. J Bone Miner Res 1997; 12: 1303-1313. Harris ND, Griffiths MR, Nguyen TV, et al. Quantitative ultrasound of the heel: a comparison of Lunar and McCue instruments [abstract]. Proceedings of the Australian and New Zealand Bone and Mineral Society 7th Annual Scientific Meeting. 1997. 29 Sept-1 Oct; Canberra, ACT: 62. (Received 24 Dec 1998, accepted 9 Jul 1999) Authors' details Department of Rheumatology, Royal North Shore Hospital, Sydney, NSW. Vasi Naganathan, FRACP, Research Scholar; Lyn March, FRACP, FAFPHM, Staff Specialist; David Hunter, MB BS, Advanced Physician Trainee in Rheumatology; Joanna Markovey, MSc, Research Bone Densitometry Technician; Philip N Sambrook, FRACP, MD, Head of Department. Department of Nuclear Medicine, St Vincent's Hospital, Sydney, NSW. Nick A Pocock, FRACP, MD, Senior Staff Specialist. Reprints will not be available from the authors. Correspondence: Dr V Naganathan, Department of Rheumatology, Royal North Shore Hospital, St Leonards, NSW 2065. Email: vasinATmed.usyd.edu.au 1: Association between quantitative heel ultrasound (QUS) results and bone mineral density* in 326 women aged 45-80 yearsBone mineral densityQUS resultNormal (n = 119)Osteopenia (n = 160)Osteoporosis (n = 47)Total (n = 326)Broadband ultrasound (BUA) Normal103918202 (62%) Osteopenia166935120(37%) Osteoporosis0044 (1%)Velocity of sound (VOS) Normal81332116 (36%) Osteopenia369623155 (47%) Osteoporosis2312255 (17%)Combined category (cQUS) Normal79280107 (33%) Osteopenia3810125164 (50%) Osteoporosis2312255 (17%)Stiffness Normal96444144 (44%) Osteopenia209421135 (41%) Osteoporosis3222247 (14%) * Measured by dual energy x-ray absorptiometry. Combined result for broadband ultrasound and velocity of sound: normal if both normal; osteopenic if either osteopenic and neither osteoporotic; and osteoporotic if either osteoporotic. Back to text 2: Use of ultrasound parameters to predict osteoporosis or osteopenia defined by dual energy x-ray absorptiometry (DEXA) measurement of bone mineral densityPredictive valuesLikelihood ratio for ultrasound measurement (95% confidence interval)Sensitivity SpecificityPositiveNegativeNormalOsteopeniaOsteoporosisTo predict osteoporosisBroadband ultrasound (BUA)9%100%100%87%0.2 (0.11-0.38)2.4 (1.9-3.1)InfinityVelocity of sound (VOS)46%88%40%91%0.1 (0.03-0.4)1.0 (0.7-1.4)4.0 (2.6-6.2)Combined category* (cQUS)47%88%40%91%0(1.1 (0.8-1.5)4.0 (2.6-6.2)Stiffness47%91%46%91%0.2 (0.08-0.5)1.1 (0.8-1.6)5.2 (3.2-8.4)To predict osteoporosis or osteopeniaBroadband ultrasound (BUA)52%87%87%51%0.6 (0.5-0.7)3.7 (2.3-6.0)InfinityVelocity of sound (VOS)83%68%82%70%0.25 (0.18-0.35)1.9 (1.4-2.6)15.0 (3.7-60.5)Combined category* (cQUS)86%66%82%74%0.20 (0.14-0.29)1.9 (1.4-2.5)15.2 (3.8-61.3)Stiffness77%81%87%67%0.3 (0.2-0.4)3.3 (2.2-5.0)8.4 (2.7-26.5) DEXA = dual energy x-ray absorptiometry. * Combined broadband ultrasound attenuation and velocity of sound category. Back to text 3: Interpretation of likelihood ratios The likelihood ratio (LR) indicates how much a test result raises or lowers the probability of an individual's having "disease". It can be used to determine post-test probability of a disease from the estimated pre-test probability using a normogram (Figure). Thus, if an early postmenopausal woman (aged 60-64 years) had a pre-test probability of osteoporosis of 15%,9 then, from the normogram, a cQUS result in the osteoporot ic range (LR, 4.0) would increase her probability to 40%. If she had a very high pre-test probability of osteoporosis, for example 50%, because of multiple risk factors (eg, including family history of osteoporosis and recent wrist fracture after a fall), then a cQUS result in the osteoporotic range would increase her probability to 80%. A BUA result in the osteoporotic range (LR, Infinity) would make BMD-defined osteoporosis highly likely, no matter what the pre-test risk. However, a cQUS result in the normal range (LR, 0) would make BMD-defined osteoporosis unlikely. QUS results in the osteopenic range, with LRs closer to unity (LR, 1.0-2.4), would make BMD category far less certain. Normogram for applying likelihood ratios, adapted from Fagan.8 Lines show post-test probabilities when LR = 4.0 and pre-test probabilities are 15% and 50%, respectively. Back to text

Vasi Naganathan · Lyn March · David Hunter · Nick A Pocock · Joanna Markovey · Philip N Sambrook

Clinical Update Androgen treatment in women Susan R Davis MJA 1999; 170: 545-549 Many women, both before and after menopause, may have symptoms of androgen deficiency: unexplained fatigue, lack of well-being and diminished libido. If plasma levels of bioavailable testosterone are low, these symptoms will mostly be relieved by judicious administration of testosterone. The addition of testosterone to postmenopausal hormone replacement regimens is becoming more widespread, and other potential uses include prevention and treatment of bone loss, treatment of spontaneous or iatrogenic androgen deficiency in premenopausal women, and, possibly, management of the premenstrual syndrome. Introduction - Physiological effects - Androgen deficiency - Measurement of - Clinical indications - Administering testosterone - Conclusions - References - Authors' details - - Articles on similar material Introduction No longer can it be said that androgens make boys as boys, and oestrogens, girls as girls. It is now known that high levels of oestrogen in the male brain in early life are necessary for male sexual imprinting,1 and that oestrogen has a fundamental role in spermatogenesis2 and maintenance of bone mineralisation in men.3 The reverse also applies. Androgens have important and varied physiological actions in women. Physiological effects of androgens in women During the reproductive years androgens are produced by the adrenal glands and the ovaries (Figure). Androgens act directly via the androgen receptor in tissues, such as bone, skin fibroblasts, hair follicles and sebaceous glands,4 and also have a vital role as the precursor steroids for oestrogen biosynthesis in the ovaries and extragonadal sites, including bone, brain, cardiovascular and adipose tissues. Hence, maintenance of physiological circulating androgen levels is important for an adequate supply of substrate hormone for oestrogen production at these sites. The physiological significance of this is best exemplified by osteoporosis in men, with a mutation in the aromatase enzyme gene which affects the conversion of androgens to oestrogens; oestrogen replacement increases bone mineral density.3It seems well established that testosterone is an important determinant of female sexuality,5-9 and that it has a physiological role in the development and maintenance of bone mineralisation.10,11 Other aspects of testosterone action in women currently being investigated include variations in testosterone level during the menstrual cycle and the behavioural changes in the premenstrual syndrome,12 as well as the effect of androgens on the immune response and autoimmune diseases.13,14 Androgen deficiency in women The prevalence of "androgen deficiency" in women has never been systematically evaluated, and there is no consensus clinical definition of this condition in women. Furthermore, a biochemical definition of androgen deficiency has been hampered by the insensitivity of most testosterone assays at the lower end of the normal range in women in their reproductive years. Women most likely to respond to androgen therapy have the following features: low libido, blunted motivation, fatigue and lack of well-being, associated with normal plasma oestrogen levels and low levels of bioavailable testosterone. Symptoms of "androgen deficiency" are often attributed to psychosocial and environmental factors, and many affected women, unaware that their problems may have a biological basis and apprehensive about the response such problems will elicit, often do not report them. Moreover, the basis of each of the symptoms listed above is likely to be multifactorial, making it important for treating physicians to evaluate and deal with other factors before considering androgen replacement. In general, the concept of androgen deficiency has been most widely accepted for women who have had bilateral oophorectomy. However, women who have undergone a natural menopause not infrequently experience "androgen deficiency" symptoms, as do a subset of women in their late reproductive years. Young women who have suffered either primary or secondary ovarian failure may also experience low libido in association with low blood androgen levels. A general approach to evaluating women with symptoms suggestive of androgen deficiency, as well as the possible causes of androgen deficiency, are outlined in Boxes 1 and 2, respectively. Measurement of androgen level Before commencing testosterone therapy in any woman, levels of testosterone and sex hormone binding globulin (SHBG), and the free androgen index (calculated to adjust for variations in SHBG), should be evaluated. Low bioavailability of testosterone is indicated by either a low ratio of levels of total testosterone to SHBG, or a free testosterone level in the lower third of the normal range in women in their reproductive years. A diagnosis of symptomatic androgen deficiency would be highly questionable with a total testosterone level in the upper third of the normal reproductive age range and a normal free androgen index. However, it is not uncommon for a midrange level of testosterone to be associated with androgen deficiency because of a very high SHBG level secondary to exogenous oestrogen replacement in postmenopausal women. Although DHEA-S and androstenedione are important precursors of testosterone, their measurement does not aid in diagnosis. Clinical indications for androgen therapy in women Sexual dysfunction There are multiple influences on libido and frequency and enjoyment of sexual activity in women. However, androgens appear to be important determinants of female sexuality and low circulating levels are associated with diminished libido. The relationship between androgens and the female sexual response has been reviewed recently.22 Bilateral oophorectomy: Anecdotal accounts suggest that the women most likely to respond to testosterone are those who have undergone bilateral oophorectomy. Premature menopause: Testosterone replacement should also be considered part of the management of young women with premature menopause, particularly those with Turner's syndrome (45,XO). In general, women who undergo menopause in their reproductive prime suffer considerably from symptoms related to androgen deficiency, particularly diminished libido. Alternatively, young women with premature menopause who have not previously been sexually active should be made fully aware of the availability of androgen replacement and, in some instances, offered low dose androgen therapy as part of their hormone replacement. Premenopausal women: It is not uncommon for premenopausal women to complain of diminished libido, and, when other potential influences on sexual dysfunction can be excluded and they have low levels of bioavailable testosterone, androgen replacement therapy is likely to be beneficial. Postmenopausal women: Most women do not report loss of sexual desire after spontaneous menopause, but there is generally an age-related reduction in sexual frequency associated with the menopausal transition.23 In a study of sexagenarian women, the only hormone to positively correlate with sexual desire was circulating free testosterone.9 Although oestrogen replacement improves vasomotor symptoms, such as vaginal dryness and possibly general well-being, it has little effect on libido.24 In contrast, the addition of testosterone to a hormone replacement regimen results in improvement in several aspects of sexuality in postmenopausal women.5-7,25 As a general rule, testosterone replacement should not be administered to postmenopausal women who are not taking concurrent oestrogen replacement. Oestrogen alone may relieve other postmenopausal symptoms, alleviate vaginal dryness and enhance sexuality, obviating the need for androgen therapy. Furthermore, suppression of SHBG with testosterone alone may increase the possibility of adverse side effects. The only exception to this rule is the use of nandrolone decanoate (see below) in postmenopausal women for the prevention of osteoporosis. Prevention and treatment of bone loss In premenopausal women: Bone loss (particularly in the hip) is associated with low total and free testosterone levels.11 Increased circulating androgen levels are associated with higher bone mineral densities.26 In postmenopausal women: Low circulating free testosterone is predictive of subsequent height loss (a surrogate marker of vertebral compression fracture), and hip fracture.27,28 Treatment with either oral or parenteral oestrogen-plus-testosterone therapy results in beneficial effects on bone mineral density over and above those seen with oestrogen alone.22,29 Oral esterified oestrogen with methyltestosterone not only increases spinal bone mineral density but also suppresses biochemical markers of bone resorption, with an increase in markers of bone formation over two years.30 Circulating levels of DHEA and DHEA-S are positively correlated with bone mineral density in older women.31,32 The daily application of a 10% DHEA cream resulted in an increase in bone mineral density of the hip in older women.33As yet, no studies have addressed the impact of androgen therapy on fracture incidence, although the effects of androgens on the mechanical properties of bone have been studied in female cynomolgus monkeys:34 testosterone therapy resulted in increases in intrinsic bone strength and resistance to mechanical stress, as well as increases in bone mineral density, bone torsional rigidity and bending stiffness.34 Potential and more controversial uses for androgen therapy are described in Box 3. Administering testosterone to women Availability: Testosterone has been available as oral methyltestosterone on prescription in North America for many years, and testosterone implants were approved for replacement therapy in postmenopausal women in the United Kingdom in the early 1990s. These and all other available testosterone preparations have primarily been formulated for use in men. Currently, the use of testosterone for hormone therapy in women is not approved in Australia. Despite the lack of approval, specialist menopause clinics in Australia have had more than a decade of experience of testosterone use in menopausal women, and hence management advice based on clinical experience is available. Nandrolone decanoate: Nandrolone decanoate (Deca-Durabolin, Organon) is a very weak aromatisable androgen, available in Australia on authority for treating postmenopausal osteoporosis, and administered intramuscularly. The dose should not exceed 50 mg, with the frequency of administration being titrated against the patient's build (ie, it is recommended that it is given 6 weekly, but 8-12 weekly in women with a body mass index lower than 20 kg/m2, otherwise virilising effects such as hirsutism and voice deepening are not uncommon). This drug will result in cessation of bone loss in most older postmenopausal women and, in some women, in an absolute increase in bone mineral density. When given 6-8 weekly this therapy does not usually improve libido. Testosterone implants: Women experiencing diminished libido are usually treated with testosterone implants and, less commonly, with mixed testosterone esters. Subcutaneous testosterone pellet implants (fused crystalline implants 4.5 mm in diameter) are the most common form of androgen therapy in women in Australia. The implant is usually inserted subcutaneously in the lower anterior abdominal wall under local anaesthesia using a trochar and cannula. A dose of 50 mg, obtained from a 100 mg implant, is extremely effective in enhancing libido and improving bone mineral density without generating unwanted virilising side effects.7,22 It is usually effective for between three and six months, but, because of marked individual variation, testosterone levels should be measured before each subsequent implant is inserted. Rarely are testosterone implants of 100 mg necessary to achieve adequate clinical effects. Indeed, circulating testosterone levels about three times the upper limit of normal have been reported four weeks after insertion of a 100 mg testosterone pellet,40 and six weeks after insertion of a 50 mg implant mean circulating testosterone levels were just above the upper limit of normal for ovulating women.22 A 100 mg dose may be needed in young women with premature ovarian failure or after early oophorectomy. Mixed testosterone esters: Although there are no published studies to support their use in women, mixed testosterone esters 50-100 mg (Sustanon, Organon) are occasionally administered 4-6 weekly as an intramuscular injection to women with androgen-deficiency symptoms. Anecodotally, this therapy results in a much more rapid onset of effects; women report enhanced libido after 2-3 days of treatment, compared with after 7-10 days with testosterone implants. The pharmocokinetics of mixed testosterone esters in women have not been studied, but women more commonly report an increase in acne and other virilising effects due to apparent peaks in testosterone levels after injection. Transdermal testosterone matrix patch: A transdermal testosterone matrix patch intended specifically for use in women has been developed and is currently undergoing early clinical trials. The patch is designed to deliver 150 µg of testosterone per day with twice-weekly application, resulting in an average increase in circulating testosterone levels of about 1 nmol/L. For more information see Box 4 Adverse effects Clinical experience suggests that, to achieve a good response in terms of libido, the testosterone level often needs to be restored to at least the upper end of the normal physiological range in young ovulating women. However, the dose needs to be titrated to keep circulating testosterone close to physiological levels to avoid adverse masculinising effects. Side effects of testosterone in women are rare when the hormone is appropriately administered. However, with excessive dosage, virilisation and fluid retention may occur. Potentially adverse lipoprotein-lipid effects (eg, reductions in high density lipoprotein cholesterol and apolipoprotein A1 levels) may occur with excessive oral administration, but have not been reported with parenteral therapy.22 Clinical data to hand do not indicate that testosterone therapy, with testosterone levels kept close to and within the normal physiological range for women, has any undesirable metabolic consequences.22,41 It is not known whether there is any relationship between exogenous androgen therapy and the incidence of breast cancer, as epidemiological studies have shown both positive and negative associations between endogenous androgen levels and risk of breast cancer. Androgen receptors are found in over 50% of breast tumours,42 and are associated with longer survival in women with operable breast cancer and a favourable response to hormone treatment in advanced disease.43 There are also some data to suggest that the therapeutic effect of high dose medroxyprogesterone acetate on breast cancer is mediated via the androgen receptor.44 Contraindications Pregnancy and lactation, as well as known or suspected androgen-dependent neoplasia, are absolute contraindications to testosterone therapy. Relative contraindications include moderate to severe acne, hirsutism, androgenic alopecia and any circumstance in which enhancement of libido would be undesirable. It is now recognised that the treatment of the postmenopausal woman with testosterone replacement may result in an ethical dilemma if the woman is a participant in older-age competitive sport. This is a controversial issue that is yet to be resolved. Conclusions Women reporting loss of libido may find physicians insufficiently empathetic, and a biological cause for sexual dysfunction in women is rarely sought. However, it is gradually becoming more accepted that androgen deficiency in women may underpin a variety of symptoms and pathophysiological conditions and that, in selected women, androgen replacement therapy is of clinical benefit. References Honda S, Harada N, Ito S, et al. Disruption of sexual behavior in male aromatase-deficient mice lacking exons 1 and 2 of the cyp19 gene. Biochem Biophys Res Commun 1998; 252: 445-449. Sharpe RM. Do males rely on female hormones? Nature 1998; 390: 447-448. Morishima A, Grumbach MM, Simpson ER. Aromatase deficiency in male and female siblings caused by a novel mutuation and the physiological role of estrogens. J Clin Endocrinol Metab 1995; 80: 3689-3698. Colvard DS, Eriksen EF, Keeting PE. Identification of androgen receptors in normal human osteoblast-like cells. Proc Natl Acad Sci USA 1989; 86: 854-857. Studd JWW, Colins WP, Chakravarti S. Estradiol and testosterone implants in the treatment of psychosexual problems in postmenopausal women. Br J Obstet Gynaecol 1977; 84: 314-315. Burger HG, Hailes J, Menelaus M. The management of persistent symptoms with estradiol-testosterone implants: clinical, lipid and hormonal results. Maturitas 1984; 6: 351-358. Burger HG, Hailes J, Nelson J, Menelaus M. Effect of combined implants of estradiol and testosterone on libido in postmenopausal women. BMJ 1987; 294: 936-937. Hickok LR, Toomey C, Speroff L. A comparison of esterified estrogens with and without methyltestosterone: effects on endometrial histology and serum lipoproteins in postmenopausal women. Obstet Gynecol 1993; 82: 919-924. Bachmann GA, Leiblum SR. Sexuality in sexagenarian women. Maturitas 1991; 13: 45-50. Nilas L, Christiansen C. Bone mass and its relationship to age and the menopause. J Clin Endocrinol Metab 1987; 65: 697-699. Slemenda C, Longcope C, Peacock M, et al. Sex steroids, bone mass, and bone loss. A prospective study of pre-, peri- and postmenopausal women. J Clin Invest 1996; 97: 14-21. Rubinow DR, Roy-Byrne P. Premenstrual syndromes: overview from a methodological perspective. Am J Psychiatry 1984; 141: 163-172. Booij A, Biewenga-Booij CM, Huber-Bruning O, et al. Androgens as adjuvant treatment in postmenopausal female patients with rheumatoid arthritis. Ann Rheum Dis 1996; 55: 811-886. Cutolo M, Seriolo B, Sulli A, Accardo S. Androgens in rheumatoid arthritis. In: Bijlsma JWJ, Linden S van der Barnes CG, editors. Rheumatology highlights 1995. Rheumatol Eur 1995; 24: 211-214. Zumoff B, Strain GW, Miller LK, Rosner W. Twenty-four hour mean plasma testosterone concentration declines with age in normal premenopausal women. J Clin Endocrinol Metab 1995; 80: 1429-1430. Zumoff B, Rosenfeld RS, Strain GW. Sex differences in the 24 hour mean plasma concentrations of dehydroisoandrosterone (DHA) and dehydroisoandrosterone sulfate (DHAS) and the DHA to DHAS ratio in normal adults. J Clin Endocrinol Metab 1980; 51: 330-334. Mushayandebvu T, Castracane DV, Gimpel T, et al. Evidence for diminished midcycle ovarian androgen production in older reproductive aged women. Fertil Steril 1996; 65: 721-723. Mathur RS, Landgreve SC, Moody LO, et al. The effect of estrogen treatment on plasma concentrations of steroid hormones, gonadotropins, prolactin and sex hormone-binding globulin in post-menopausal women. Maturitas 1985; 7: 129-133. Krug R, Psych D, Pietrowsky R, et al. Selective influence of menstrual cycle on perception of stimuli with reproductive significance. Psychosom Med 1994; 56: 410-417. Abraham GE. Ovarian and adrenal contribution to peripheral androgens during the menstrual cycle. J Clin Endocrinol Metab 1974; 39: 340-346. Anasti JN, Kalankaridou SN, Kimzey LM, et al. Bone loss in young women with karyotypically normal spontaneous premature ovarian failure. Obstet Gynecol 1998; 91: 12-15. Davis SR, McCloud PI, Strauss BJG, Burger HG. Testosterone enhances estradiol's effects on postmenopausal bone density and sexuality. Maturitas 1995; 21: 227-236. Frock J, Money J. Sexuality and the menopause. Psychother Psychosom 1992; 57: 29-33. Campbell S, Whitehead M. Oestrogen therapy and the menopausal syndrome. Clin Obstet Gynecol 1977; 4: 31-47. Sherwin BN, Gelfand MM, Brender W. Androgen enhances sexual motivation in females: a prespective, crossover study of sex steroid administration in surgical menopause. Psychosom Med 1997; 47: 339-351. Simberg N, Titinen A, Silfrast A, et al. High bone density in hyperandrogenic women: effect of gonadotropin-releasing hormone agonist alone or in conjunction with estrogen-progestin replacement. J Clin Endocrinol Metab 1995; 81: 646-651. Jassal SK, Barrett-Connor E, Edelstein S. Low bioavailable testosterone levels predict future height loss in postmenopausal women. J Bone Miner Res 1995; 10: 650-653. Davidson BJ, Ross RK, Paganni Hill A, et al. Total free estrogens and androgens in postmenopausal women with hip fractures. J Clin Endocrinol Metab 1982; 54: 115-120. Watts NB, Notelovitz M, Timmons MC. Comparison of oral estrogens and estrogens plus androgen on bone mineral density, menopausal symptoms and lipid-lipoprotein profiles in surgical menopause. Obstet Gynecol 1995; 85: 529-537. Raisz LG, Witta B, Artis A, et al. Comparison of the effects of estrogen alone and estrogen plus androgen on biochemical markers of bone formation and resorption in postmenopausal women. J Clin Endocrinol Metab 1995; 81: 37-43. Nawata H, Tariaka S. Aromatase in bone cell: association with osteoporosis in postmenopausal women. J Steroid Biochem Molec Biol 1995; 53: 165-174. Nordin BEC, Robertson A, Seamark RF, et al. The relation between calcium absorption serum DHEA and vertebral mineral density in postmenopausal women. J Clin Endocrinol Metab 1985; 60: 651-657. Labrie F, Diamond P, Cusan L, et al. Effect of 12-month dehydroepiandrosterone replacement therapy on bone, vagina and endometrium in postmenopausal women. J Clin Endocrinol Metab 1997; 82: 3498-3505. Kasra M, Grynpas MD. The effects of androgens on the mechanical properties of primate bone. Bone 1995; 17: 265-270. Engelson ES, Goggin KJ, Rabkin JG, Kotler DP. Nutrition and testosterone status of HIV positive women [Abstract]. Proceedings of the XI International Conference on AIDS, Vancouver, 1996. Miller K, Corcoran C, Armstrong C, et al. Transdermal testosterone administration in women with acquired immunodeficiency syndrome wasting: a pilot study. J Clin Endocrinol Metab 1998; 83: 2717-2725. Bloch M, Schmidt PJ, Su T-P, et al. Pituitary-adrenal hormones and testosterone across the menstrual cycle in women with premenstrual syndrome and controls. Biol Psychiatry 1998; 43: 897-903. Masi AT, Feigenbaum SL, Chatterton RT. Hormonal and pregnancy relationships to rheumatoid arthritis: convergent effects with immunological and microvascular systems. Semin Arthritis Rheum 1995; 25: 1-27. van Vollenhoven RF, Morabito LM, Engleman EG, McGuire JL. Treatment of systemic lupus erythematosus with dehydroepiandrosterone: 50 patients treated up to 12 months. J Reheumatol 1998; 25: 285-289. Buckler HM, Robertson WR, Wu FCW. Which androgen replacement therapy for women? J Clin Endocrinol Metab 1998; 83: 3920-3924. Davis SR, Burger HG. The rationale for physiological testosterone replacement in women. Baillieres Clin Endocrinol Metab 1998. In press. Recchione C, Venturelli E, Manzari A, et al. Testosterone, dihydrotestosterone and oestradiol levels in postmenopausal breast cancer tissues. J Steroid Biochem Mol Biol 1995; 52: 541-546. Bryan RM, Mercer RJ, Rennie GC, et al. Androgen receptors in breast cancer. Cancer 1984; 54: 2436-2440. Birrell SN, Roder DM, Horsfall DJ, et al. Medroxyprogesterone acetate therapy in advanced breast cancer: the predictive value of androgen receptor expression. J Clin Oncol 1995; 13: 1572-1577. Authors' details The Jean Hailes Foundation Research Unit, Melbourne, VIC. Susan R Davis, FRACP, PhD, Director of Research; and Senior Lecturer Department of Preventive Medicine and Epidemiology, Monash Medical School, Alfred Hospital, Melbourne. Reprints will not be available from the author. Correspondence: Dr S R Davis, The Jean Hailes Foundation Research Unit, 173 Carinish Road, Clayton, VIC 3168. Email: suedavis@netlink.com.au ©MJA 1999 Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/> © 1999 Medical Journal of Australia. Back to text 1: Evaluation of androgen deficiency in women Clinical suspicion of androgen deficiency Gradual loss of sexual desire in otherwise satisfying sexual relationship Persistent fatigue with no clear cause Premature ovarian failure Bilateral oophorectomy Exclusion of other causes of symptoms Full psychosocial history Assess adequacy of oestrogen therapy in postmenopausal women Exclude other causes of fatigue (eg, iron deficiency, hypothyroidism) Tests to establish androgen deficiency Total testosterone level Sex hormone binding globulin (SHBG) level Free androgen index Dehydroepiandrosterone-sulfate (DHEA-S) level Consider androgen therapy for women with: Symptomatic testosterone deficiency after natural menopause Symptomatic testosterone deficiency following oophorectomy, chemotherapy or radiotherapy Premature ovarian failure -- primary or secondary Premenopausal loss of libido with low level of bioavailable testosterone. Back to text 2: Causes of androgen deficiency in women Age-related Physiological circulating androgen levels (total and free testosterone, dehydroepiandrosterone [DHEA], and dehydroepiandrosterone-sulfate [DHEA-S]) fall continuously with age,15,16 commencing in the decade preceding the average age of natural menopause. This is a consequence of the concurrent decline with age in adrenal production of the preandrogens DHEA, DHEA-S and androstenedione, and diminished testosterone production by the ovaries.16,17 Iatrogenic Oophorectomy -- bilateral oophorectomy results in a 50% fall in testosterone and androstenedione. Chemical oophorectomy results from administration of gonadotropin-releasing hormone antagonists, chemotherapy or radiotherapy. Administration of exogenous oestrogen -- combined oral contraceptive pill or oral postmenopausal oestrogen therapy increases sex hormone binding globulin (SHBG) levels (thus reducing free testosterone), and suppresses pituitary luteinising hormone production (hence lessening stimulation of ovarian androgen biosynthesis).18,19Administration of exogenous glucocorticosteroids -- glucocorticosteroids reduce adrenal androgen production by suppressing ACTH.20 This appears to contribute to the pathogenesis of osteopenia and osteoporosis, the side effects of long term glucocorticosteroid therapy. Pathological Hypothalamic amenorrhoea or hyperprolactinaemia in premenopausal women. Premature primary or secondary ovarian failure. Bone loss complicates each of these conditions and appears to progress despite adequate oestrogen-progestin therapy.21 Young women with these conditions may also require testosterone replacement to prevent progressive bone resorption. Back to text 3: Potential indications for androgen use in women Postmenopausal loss of muscle mass: In postmenopausal women testosterone therapy is associated with an increase in fat-free mass and a reduction in the fat mass to fat-free mass ratio.22 As this gain in fat-free mass probably reflects increased muscle mass, and ageing is associated with loss of muscle mass, testosterone therapy is beneficial in older women. Management of wasting in HIV infection: Testosterone levels are lower in HIV-positive premenopausal women.35 Augmentation of testosterone levels with a transdermal testosterone patch is associated with increased mean body weight and body mass index as well as improved quality of life.36Testosterone and the premenstrual syndrome: Significantly lower levels of testosterone throughout the menstrual cycle have been reported in women who suffer premenstrual syndrome compared with controls.12,37 Testosterone is being used in selected patients with premenstrual syndrome in specialised centres in the United Kingdom and Australia, and randomised trials evaluating the effects are under way. Testosterone and autoimmune disease: Sex differences in the pattern of autoimmune disease are well recognised, and may be related to higher testosterone levels in men, with some studies indicating that androgens suppress both cell-mediated and humeral immune responses.14,38 Reports in postmenopausal women with rheumatoid arthritis indicate symptomatic improvement with testosterone replacement,13 and reductions in disease activity with DHEA therapy.39 However, apart from its use to counteract the side effects of long term glucocorticosteroid therapy (muscle wasting and bone loss) in autoimmune disease, much more substantial evidence is required before testosterone can be advocated as adjunctive therapy in autoimmune diseases. Back to text 4: Prescribing androgen replacementNandrolone decanoateApproved for use in postmenopausal women with osteoporosis, on authorityDose range:25-50 mgRoute:IntramuscularFrequency:6-12 weekly Testosterone implantsApproved for use in women in the UK, but not in AustraliaDose range:50 mg (rarely, 100 mg)Route:SubcutaneousFrequency:3-6 monthly Mixed testosterone estersNot approved for use in women. No published data pertaining to use in womenDose range:50-100 mgRoute:IntramuscularFrequency:4-6 weekly Testosterone undecanoateLimited data in women indicate high circulating peak levels. Not approved for use in womenDose range:40 mgRoute:OralFrequency:Alternate days/daily MethyltestosteroneIn combination with esterified oestrogen, approved for women in USADose range:1.25-2.5 mgRoute:TransdermalFrequency:Daily Transdermal testosterone matrix patchUndergoing clinical trialDose range:150 µgRoute:TransdermalFrequency:Changed twice weeklyBack to text

Susan R Davis

Editorial Designer insulins: moving closer to convenient and physiological replacement of insulin Heralding a new horizon for people with diabetes MJA 1999; 170: 349-350 Since Banting and Best's development of insulin treatment in 1921, several major advances in production and delivery of insulin have revolutionised the care of people with diabetes. From this beginning, when crude animal insulin extracts and uncomfortable syringes and needles were used, the late 20th century has seen the development of pure genetically engineered human insulin delivered by convenient and almost painless pen delivery devices. These technological advances have been coupled with recent clinical trials that have confirmed the importance of euglycaemia to avoid the chronic complications of diabetes. The Diabetes Control and Complications Trial (DCCT)1 and the most recent United Kingdom Prospective Diabetes Study (UKPDS)2 have clearly shown the benefit of intensive compared with conventional insulin therapy. However, for many people with diabetes, intensive insulin therapy means multiple insulin injections and frequent blood sampling, as well as an increased risk of hypoglycaemia. Insulin analogues, or designer insulins, promise a new era of greater convenience and mastery for people with diabetes by replacing insulin in a more convenient and physiological manner. The first of these analogues available in Australia is insulin lispro (Figure). To make insulin lispro, the order of the amino acids proline and lysine on the B chain at positions 28 and 29 has been reversed.3 This subtle but important conformational change improves the solubility and absorption of the insulin molecule. Unlike conventional regular insulin, the peak insulin response is more rapid, occurring within 60-90 minutes, and the total duration of action is shorter, lasting 5-6 hours. The greatest potential benefits of these new types of insulins are improved blood glucose control with less risk of hypoglycaemia, as the peak of insulin action is more likely to coincide with peak glucose levels. Furthermore, as insulin lispro may be administered immediately before meals, it promises to provide greater convenience and flexibility. In this issue of the Journal, Stocks reports the use of this new insulin in a large cohort of 150 patients with type 1 diabetes managed in a private practice setting; results in 125 patients could be fully analysed.4 Before commencing the study all patients were receiving intensive insulin therapy, with at least four injections of insulin per day. The reported findings are similar to those of several randomised studies comparing regimens of multiple daily doses of lispro and regular insulin in patients with type 1 diabetes.5 As with Stocks' study, most of these have, of necessity, been unblinded, because of the different times of administration of the two insulins, and many have used a crossover design to minimise bias. In Stocks' study, despite the use of an intensive insulin regimen, more than half the patients had glycohaemoglobin values (expressed as %HbA1c) above the acceptable level of 8%. This is not an unusual finding in most clinical reports of type 1 diabetes and reflects the imprecise or unphysiological nature of insulin therapy. So does insulin analogue therapy live up to expectations? After changing to insulin lispro more than half the patients reported less glucose fluctuation. In addition, half experienced a reduction in HbA1c, and in those with initial HbA1c values above 8% this change was highly significant. In most other reported studies, lower HbA1c values are not usually achieved with insulin lispro compared with regular insulin.5 This difference may be explained by the appropriate adjustments made to overnight basal insulin in Stocks' study to ensure that fasting blood glucose levels remained within the target range. Coupled with improved glucose control, there was also a reduction in the number of hypoglycaemic episodes, both during the day and the night. The effect on hypoglycaemic episodes in other lispro studies has varied from no change to a 12% reduction. Intensive insulin therapy has been associated with weight gain,1 but in Stocks' study improvement in blood glucose control was not accompanied by changes in weight. The unblinded nature of this and other studies of lispro, and the use of a non-validated questionnaire, complicates the assessment of its impact on patient quality of life. Nevertheless, most patients reported that they felt better when taking insulin lispro. John Main recently described the nature of clinical practice as the "disorderly world of real medical practice".6 The evidence-based-medicine purists among us may undervalue the importance of the reported findings, as the highest level of evidence -- the double-blind randomised controlled trial -- was not used. Nevertheless, Stocks' study is valuable for several reasons, and the findings may be generalisable to other clinical settings. Firstly, it provides "real world" local outcome data on a large number of patients who would be typically seen in routine practice. Secondly, the study reminds us that biochemical outcomes are only one aspect of good diabetes management. Lifestyle issues, flexibility and convenience must also be considered. The perspective of some physicians may, however, be quite different, focusing on biochemical outcomes that may not be easily achieved without significant detriment to quality of life. With the rise of consumerism, the incorporation of patient values into decision-making is likely to gain further momentum. Lastly, in an age of audit, quality and accountability, as well as good clinical governance, it is heartening to see that serious health-outcomes research can and is being conducted in private practice settings. In the near future there is likely to be a vast array of new designer insulins as well as new methods of delivery.7 Other rapid-acting analogues, such as insulin aspart (aspartic acid replaces proline at position 28), are currently being tested. A neutral protamine lispro has also been developed and mixtures of lispro and neutral protamine lispro are being tested. Research is also progressing towards true basal insulin analogues, long-acting insulins with a prolonged action profile that will be reproducibly absorbed to maintain intermediate metabolism. These designer insulins, coupled with new developments such as inhaled insulin delivery, herald a new horizon for people with diabetes as we approach the 80th year of insulin treatment.7 Steven C Boyages Clinical Associate Professor Department of Diabetes and Endocrinology, Westmead Hospital, Sydney, NSW (Presently, Director Research and Clinical Policy, NSW Health, Sydney, NSW) The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long term complications in insulin dependent diabetes mellitus. N Engl J Med 1993; 329: 977-986. UK Prospective Diabetes Study Group: Intensive blood glucose control of sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998; 352: 837-853. Barnett AH, Owens DR. Insulin analogues. Lancet 1997; 349: 47-51. Stocks AE. Insulin lispro: experience in a private practice setting. Med J Aust 1999; 170: 364-367. Lee W, Zinman B. From insulin to insulin analogs: progress in the treatment of type 1 diabetes. Diabetes Rev 1998; 6: 73-88. Main J. Doctors advocating evidence based medicine may be out of touch with real medicine. BMJ 1999; 318: 332. Marks J. Diabetes management in the future: a whiff and a long shot? Clin Diabetes 1998; 16: 3. Reprints: Professor S C Boyages, Department of Diabetes and Endocrinology, Westmead Hospital, Westmead, NSW 2145. Make a comment Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company. Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/> We appreciate your comments. Figure: Structure of insulin lispro Back to text

Steven C Boyages

Healthcare Insulin lispro: experience in a private practice setting Alan E Stocks MJA 1999; 170: 364-367 For editorial comment, see Boyages Abstract - Introduction - Methods - Results - Discussion - References - Authors' details - - More articles on Endocrinology Abstract Objective: To assess the benefits (or otherwise) of changing patients with unstable type 1 diabetes to insulin lispro (Humalog, Aza Research). Participants and setting: 150 patients with type 1 diabetes treated in a private metropolitan physician practice. Main outcome measures: Glycohaemoglobin (%HbA1c) value; frequency of hypoglycaemia; patients' perceptions of the two regimens. Results: 125 of 150 patients completed the study; 12 (8%) withdrew because they were dissatisfied with lispro. Including these 12 patients, 80/137 patients (58%) felt better physically and 98 (72%) were more satisfied with treatment with lispro. Of the 125 patients who completed the study, 65 (52%) achieved lower HbA1c values, with 22 (18%) experiencing a fall of 1% or more; 78 patients (62%) noted less daytime hypoglycaemia and 103 (82%) experienced less nocturnal hypoglycaemia. Thirty-nine patients (31%) with less daytime hypoglycaemia and 52 (42%) with less nocturnal hypoglycaemia also experienced a fall of more than 0.2% in HbA1c value (mean, 1.0%; range, 0.3%-3.2%). There was no change in body weight or in hypoglycaemia unawareness. Conclusion: Many patients with unstable type 1 diabetes, when changed to insulin lispro, experience significant clinical and biochemical benefits. Introduction Subcutaneous injection of conventional soluble or regular insulin in people with type 1 diabetes has several disadvantages compared with circulating insulin in the non-diabetic: the onset of action of regular insulin is too slow, so that ideally it should be injected 30-60 minutes before meals (causing safety concerns if the meal is then delayed or forgotten), and this delayed peak of action makes it difficult to control the postprandial rise in blood glucose level. Attempts to reduce postprandial hyperglycaemia by increasing the insulin dose increase the risk of hypoglycaemia 4-6 hours later. Moreover, the action of regular insulin persists too long, increasing the risk of late hypoglycaemia, especially nocturnal hypoglycaemia. Finally, the absorption of regular insulin is very variable from day to day, and this contributes to unstable and unpredictable blood glucose levels. These disadvantages have led to the development of the insulin analogue insulin lispro (Box 1). With its more rapid action it should alleviate many of these concerns. Patients can eat immediately after injecting insulin lispro, which is more convenient, and postprandial hyperglycaemia is much reduced.5 Lispro is more predictably absorbed, which should give fewer day-to-day blood glucose fluctuations, and, in turn, should reduce the risk of unpredictable hypoglycaemia by day and night,6 giving greater patient confidence. Patients should be able to reduce or cease between-meal snacks without an increased risk of hypoglycaemia, and this could lead to weight loss. However, when compared with treatment with regular insulin, glycohaemoglobin values (expressed as %HbA1c) have been reported to change very little in unselected patient groups treated with lispro.5,7 In this study, the effectiveness of lispro was assessed in a group of patients selected because they were having difficulty achieving smooth, safe and satisfactory glycaemic control with conventional insulins. A change to lispro was not offered to patients whose diabetes was well controlled without significant hypoglycaemia (ie, in this group the change was not made primarily for reasons of personal convenience). The objective was to ascertain whether their glycaemic control could be improved (without increasing the risk of hypoglycaemia in these unstable patients), and whether their health and their tolerance of diabetes could be improved. Methods In one practice managing patients with diabetes, the first 150 patients who were changed to insulin lispro were studied. The reason(s) for the change are shown in Box 2 (in most patients there was more than one reason). Despite previous intensive insulin therapy, all except two patients had problems with their glycaemic control or with hypoglycaemia with conventional insulins; of the 17 patients listed as being changed for reasons of convenience, 15 had other significant problems as well (eg, unpredictable blood glucose level fluctuations). All patients injected insulin four times daily using a basal-bolus insulin regimen. Lispro was injected using a 1.5 mL injector pen (B-D Pen, Aza Research, Sydney). Most patients (81%) injected Protaphane at bedtime using a NovoPen 3 (Novo Nordisk, Sydney), the remainder using other long-acting insulins (eg, Monotard [Novo Nordisk], Ultratard [Novo Nordisk] or Humulin UL [Aza]). No patient injected a long-acting insulin before breakfast. All patients were asked to monitor their blood glucose level seven times daily initially, using their previous blood glucose monitor, before and two hours after meals and at bedtime, and were invited to telephone each morning before breakfast to report levels until their blood glucose and insulin doses were stable. Patients were instructed to use a sliding scale to adjust their lispro doses if their blood sugar levels were unexpectedly greater than 10 mmol/L or less than 5 mmol/L (Box 3). An attempt was made to encourage patients with HbA1cvalues below 7% to strive for higher blood glucose levels and to try to avoid their blood glucose level falling below 4 mmol/L, in the hope of decreasing lack of awareness of hypoglycaemia. All patients were asked to complete a questionnaire (Box 4). All responses are entirely subjective, and no statistical analysis of these was undertaken. Changes in HbA1c values and body weight were recorded, usually after 4-6 months. For purposes of classification, a variation in HbA1c between 2 0.2% and + 0.2% was regarded as no change, a rise or fall of 0.3%-0.9% as a minor increase/decrease, and a rise or fall of 1.0% or more as a major change. Similarly, a change in weight of 2 0.2 kg to + 0.2 kg was regarded as no change, a rise/fall of 0.3-1.9 kg as a minor increase/decrease, and a rise/fall of 2 kg or more as a major change. Changes in insulin doses and HbA1c are given as means and 95% confidence intervals (CIs) (Microsoft Excel). Correlations were determined by analysis of variance (ANOVA). Student's two-tailed t test was used to calculate the P values. Results The 150 patients were aged 13-86 years (mean, 46.6 years). There were 63 males and 87 females, who had been receiving treatment for diabetes for between one and 54 years (mean, 22.1 years). Eight of the 150 patients were lost to follow-up. Twelve patients had stopped injecting lispro (usually within one month) and reverted to conventional insulins: five reported that lispro made their diabetes more unstable, three that it produced more hypoglycaemia, and four that it made them feel unwell. Two patients had not started taking lispro, and three patients failed to have a repeat HbA1c measurement. Thus, there were 125 patients who completed a questionnaire and presented all the data necessary for analysis. Insulin doses: Nearly all patients needed to adjust their insulin doses, but there was wide individual variation. More patients needed to reduce than increase the daily dose of lispro (52 v. 39 patients), while more increased their bedtime insulin than reduced it (62 v. 19 patients). However, the mean dose adjustments were small: the total daily dose of insulin rose by 1.3 units (95% CI, 0.11, 2.49), the dose of lispro was reduced by 0.4 units (95% CI, 2 0.76, 1.56), and the bedtime insulin dose rose by 1.8 units (95% CI, 0.88, 2.72). Blood glucose fluctuation: Sixty-nine of 125 patients (55%) reported that their blood glucose level fluctuated less on lispro, 38 (30%) that there was no difference and 18 (14%) that their blood glucose levels fluctuated more. Fifty-one (41%) reported that their overall blood glucose levels were lower, 58 (46%) that there was no change, and 16 (13%) that their overall blood glucose levels were higher on lispro. Glycohaemoglobin (%HbA1c): In general, the changes in HbA1c values agreed with the patients' perceptions of their overall blood glucose control. Sixty-five patients (52%) had a fall in HbA1c of 0.3% or more, with 22 (18%) achieving an HbA1c value at least 1.0% lower with lispro. In 22 patients (18%) the HbA1c value rose by 0.3% or more, and in six patients (5%) it rose by 1.0% or more. From Box 5 it can be seen that the HbA1c value did not change significantly in patients whose initial level was less than 8%, but there was a highly significant reduction in those with initially higher levels, and in the group as a whole. The greatest improvement was seen in patients with initial HbA1c values of 8%-9%, of whom 51% (23/45) achieved HbA1c values less than 8% (Box 6). The group of patients whose HbA1c value fell took slightly more insulin (3.35 units/day; 95% CI, 1.81, 5.16), while those whose HbA1c value rose took slightly less (2 1.55 units/day; 95% CI, 2 4.17, 1.07). There was no correlation between individual changes in HbA1c and the total daily insulin dose (r = 0.152, P = 0.08), but an increased bedtime insulin dose was weakly correlated with a lower HbA1c value (r = 0.238, P < 0.02). Hypoglycaemia: The frequency of both daytime and nocturnal hypoglycaemia was reduced after the change to lispro, as shown in Box 7. Seventy-eight patients (62%) noted fewer daytime episodes, with 53 patients (42%) reporting that they had experienced much less hypoglycaemia or none at all. The frequency of nocturnal hypoglycaemia was even more dramatically reduced, with 103 patients (82%) noting fewer episodes. Forty-one patients (33% of the total) became completely free of nocturnal hypoglycaemia. Despite the reduction in episodes of hypoglycaemia, there was little change in hypoglycaemia awareness, 31 patients reporting that hypoglycaemia was harder to recognise, while 24 reported easier recognition. Only one patient complained of new hypoglycaemia unawareness. There was no apparent association between reduced awareness of hypoglycaemia and the HbA1c value achieved with lispro, nor with major reductions in HbA1c values. Thirty-nine patients (31%) experienced a fall in HbA1c value (mean, 1.0%; range, 0.3%-3.2%) associated with less daytime hypoglycaemia, and 52 patients (42%) achieved a lower HbA1c value (mean, 0.9%; range, 0.3%-3.2%) associated with a reduction in nocturnal hypoglycaemia. Weight: On changing to lispro, five patients were underweight, with body mass index (BMI) less than 20 kg/m2, 69 were normal (20-25 kg/m2), and 51 were by definition overweight (BMI >25 kg/m2). There was no systematic change in body weight: 55 patients lost weight, with 19 losing more than 2 kg, while 45 patients gained weight, of whom 25 gained more than 2 kg. There was no relationship between BMI at the commencement of the study and subsequent weight change, nor between weight change and changes in HbA1c values. Two-thirds of patients reduced their between-meal snacks, but only 18 (14%) were able to omit their mid-morning snack, while 33 (26%) and 26 (21%) were able to omit afternoon tea and bedtime supper altogether. Wellbeing and satisfaction: If the 12 patients who withdrew from the study are included (making a total of 137 patients), 80 patients (58%) reported feeling physically better when taking lispro, and 48 (35%) said they were much better. Ninety-eight patients (72%) reported greater satisfaction with the new insulin regimen, of whom 64 (47%) said they were much more satisfied. Of the 125 patients who completed the study, 10 (8%) reported that they were less satisfied with treatment, with three patients being much less satisfied -- and to these must be added the 12 patients who withdrew because of dissatisfaction with lispro, making a total of 16% (22/137). Discussion Many of the patients experienced significant clinical and biochemical benefits after changing to insulin lispro. This insulin analogue seems to be more physiological than regular insulin, but it is still not perfect, as indicated by the 16% of patients who were dissatisfied. Patients were selected because of difficulties with their glycaemic control or hypoglycaemia rather than for personal convenience, and this may partly explain the different results from those in earlier studies of lispro,5,7 in which no significant change in HbA1c was noted in cohorts of unselected patients. The aim was to provide an insulin regimen which would minimise the risks of both hypoglycaemia and long term complications. Results from the Diabetes Control and Complications Trial (DCCT) suggest that there is a curvilinear relationship between high HbA1c values and an increased risk of microvascular complications, and that a similar curvilinear relationship exists between lower HbA1c values and an increased risk of severe hypoglycaemia and hypoglycaemia unawareness.8 In practice, achieving HbA1c values between 7% and 8% seems to offer reasonable protection against both problems. During routine therapy with regular insulin the use of a sliding scale to correct unexpected high or low blood glucose levels is not logical, as this insulin works too slowly to achieve the desired result; however, with lispro, patients can add or subtract small doses, which helps in returning aberrant glucose levels to the desirable range more quickly. Most patients learn to use this method easily. More than half the group experienced a fall in HbA1c, which in 20% of patients was 1% or more. The greatest improvement occurred in patients whose initial HbA1c value was only moderately elevated (HbA1c, 8%-9%), and, of these subjects, more than 50% were able to achieve a value which, according to results from the DCCT,8 should offer better protection against microvascular complications. However, although patients with initially poor glycaemic control (HbA1c, >9%) often achieved HbA1c values less than 9%, only 4 of 24 (17%) achieved values below 8%; the remainder must be regarded as being still at increased risk of complications. Most patients in whom HbA1c values were initially less than 7% proved resistant to attempts to raise their blood glucose level, with only a third achieving HbA1c values of 7% or higher (Box 6). According to the DCCT, the risk of hypoglycaemia unawareness in this group is considerable,8 and it is pleasing to note that 10 of these 15 patients (Box 6) experienced less daytime hypoglycaemia and 13 had less nocturnal hypoglycaemia, with no significant change in hypoglycaemia awareness. Patients, their relatives and their physicians all welcome any reduction in risk of hypoglycaemia, and many will deliberately try to avoid hypoglycaemia by reducing insulin doses, even though this leads to poor overall glycaemic control and increases the risk of microvascular complications. It is particularly gratifying that 31% of the patients had less daytime hypoglycaemia and 42% had less nocturnal hypoglycaemia associated with a fall in HbA1c value, indicating greater safety together with better glycaemic control. A lower frequency of hypoglycaemia should be associated with improved hypoglycaemia awareness,9 but this did not eventuate. On the other hand, there was little evidence of any progression of hypoglycaemia unawareness over the short duration of the study. There was no correlation between individual changes in HbA1c and changes in either the daily dose of lispro or the total daily insulin requirement. Dose adjustments were small, but an inverse correlation was found with the bedtime insulin dose; this was sufficiently weak to indicate that other factors are likely to have contributed to this finding.10 Insulin lispro does not suit every patient with type 1 diabetes, but with careful education in its use many experience significant benefits, and achieve greater confidence and satisfaction. The advent of new long-acting and short-acting insulin analogues in the near future is awaited with interest; they will need careful clinical evaluation by both patients and their treating physicians. Conflict of interest: None. References Dimarchi RD, Chance RE, Long HB, et al. Preparation of an insulin with improved pharmacokinetics relative to human insulin through consideration of structural homology with insulin-like growth factor 1. Horm Res 1994; 41 (Suppl 2): 93-96. Ciszak E, Beals JM, Baker JC, et al. Role of C-terminal B-chain residues in insulin assembly; the structure of hexameric Lys B28 Pro B29-human insulin. Structure 1995; 3: 615-622. Howey DC, Bowsher RR, Brunelle RL, Woodworth JR. [Lys(B28), Pro(B29)]-human insulin. A rapidly absorbed analogue of human insulin. Diabetes 1994; 43: 396-402. Bhaskar R, Chou MCY, Field JB. Time action characteristics of regular and NPH insulin in insulin-treated diabetics. Clin Endocrinol Metab 1980; 50: 475-479. Anderson JH, Brunelle RL, Koivisto VA, et al, and the Multicenter Insulin Lispro Group. Reduction of postprandial hyperglycaemia and frequency of hypoglycaemia in IDDM patients on insulin-analog treatment. Diabetes 1997; 46: 265-270. Holleman F, Schmitt H, Rottiers R, et al, the Benelux-UK Insulin Lispro Study Group. Reduced frequency of severe hypoglycaemia and coma in well controlled IDDM patients treated with insulin lispro. Diabetes Care 1997; 20: 1827-1832. PfŸtzner A, KŸstner E, Forst T, et al, on behalf of the German Insulin Lispro/IDDM study group. Intensive insulin therapy with insulin lispro in patients with Type 1 diabetes reduces the frequency of hypoglycaemic episodes. Exp Clin Endocrinol 1996; 104: 25-30. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329: 977-986. Amiel SA. Limits of normality: the mechanisms of hypoglycaemia unawareness. Diabetic Med 1994; 11: 918-924. Sokal RR, Rohlf FJ. Biometry: the principles and practice of statistics in biological research. 2nd ed. New York: WH Freeman, 1981: 562-565. (Received 17 Aug 1998, accepted 18 Feb 1999) Authors' details The Brisbane Clinic, Brisbane, QLD. Alan E Stocks, AM, FRCP(Ed), FRACP, Physician. Reprints will not be available from the author. Correspondence: Dr A E Stocks, Brisbane Clinic, 79 Wickham Terrace, Brisbane, QLD 4000. Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/> 1: Insulin lispro Insulin lispro (Humalog, Aza Research) is an insulin analogue in which the amino acid sequence proline-lysine at positions 28 and 29 on the B chain of human insulin has been reversed.1 This molecule exists in solution as a monomer, unlike regular human insulin, which, in solution, is mainly in the dimeric or hexameric form.2 Monomeric insulin is more rapidly absorbed into the circulation from a subcutaneous injection site, and its action is much more rapid than conventional insulins. The onset of action is almost immediate, peak action occurs 60-90 minutes after injection and the total duration of action is 5-6 hours.3 Comparative figures for regular insulin are onset 30-45 minutes, peak 3-6 hours and total duration of action 8-16 hours.4 Back to text2: Reason(s) for change to insulin lispro (number of patients)Unstable/unpredictable bloodglucose levels139 Frequent hypoglycaemia(eg, more than twice a week)81Hypoglycaemia unawareness(needing assistance)53High glycohaemoglobin value(HbA1c >9.5%)14Weight gain/obesity37Convenience17Back to text 3: Suggested sliding scaleMealtime blood glucose level (mmol/L)Lispro adjustment (units)< 2.5-32.5-3.5-23.6-4.9-15-10010.1-12.5+ 112.6-15+ 215.1-17.5+ 317.6-20+ 4>20+ 5Back to text 4: QuestionnaireBlood sugar level overall?Higher -- LowerBlood sugar fluctuations?Greater -- LessDaytime hypoglycaemia? More -- Less -- NoneNocturnal hypoglycaemia?More -- Less -- NoneEase of recognition?Easier -- HarderBetween-meal snacks?Yes -- No/Same -- LessWeight change?Gain -- LossFeel different physically?Better -- WorseSatisfaction with treatment?Happier -- Less happyBack to text 5: Analysis of changes in glycohaemoglobin (HbA1c) valueInitialPatientsMean HbA1c value (95% CI)HbA1cn (%)Before lisproAfter lisproP<7%14 (11%)6.45%6.85%0.12(6.17%, 6.73%)(6.35%, 7.35%)7.0%-8.0%42 (34%)7.59% 7.58%0.88(7.51%, 7.67%)(7.38%, 7.78%)8.1%-9.0%45 (36%)8.50%8.01%<0.001(8.42%, 8.58%)(7.86%, 8.16%)>9%24 (19%)9.75%8.92%<0.001(9.49%, 10.39%)(8.41%, 9.43%)Total125 (100%)8.22%7.92%< 0.001group(8.03%, 8.41%)(7.75%, 8.09%)Back to text 6: Change in glycohaemoglobin (HbA1c) valuesInitial HbA1cnFinal HbA1cn<7%(15)<7%107%-8%2>8%3 7.0%-8.0%(41)<7%67%-8%28>8%7 8.1%-9%(45)<7%27%-8%21>8%22 >9%(24)<8%48%-9%9>9%11Back to text 7: Frequency of hypoglycaemiaDaytimeNocturnalMuch more3 (2%)1 (1%)More20 (16%)7 (6%)Unchanged24 (19%)14 (11%)Less25 (20%)19 (15%)Much less48 (38%)43 (34%)None at all5 (4%)41 (33%)Back to text

Alan E Stocks

Position Statement New classification and criteria for diagnosis of diabetes mellitus Position Statement from the Australian Diabetes Society,* New Zealand Society for the Study of Diabetes, Royal College of Pathologists of Australasia and Australasian Association of Clinical Biochemists Peter G Colman,* David W Thomas, Paul Z Zimmet,* Timothy A Welborn, * Peter Garcia-Webb and M Peter Moore MJA 1999; 170: 375-378 Introduction - What are the new diagnostic criteria? - What about the oral glucose tolerance test? - Diabetes in pregnancy - How has the classification of diabetes changed? - Impaired glucose tolerance and impaired fasting glycaemia - References - Authors' details - - More articles on Endocrinology Introduction Recently, there has been major growth in knowledge about the aetiology and pathogenesis of different types of diabetes and about the predictive value of different blood glucose levels for development of complications. In response, both the American Diabetes Association (ADA) and the World Health Organization (WHO) have re-examined, redefined and updated the classification of and criteria for diabetes, which have been unchanged since 1985. While the two working parties had cross-representation, they met separately, and differences have emerged between their recommendations. The ADA published its final recommendations in 1997,1 while the WHO group published its provisional conclusions for consultation and comment in June 1998.2 The WHO process called for comments on the proposal by the end of September 1998, with the intention of finalising definitive classification and criteria by the end of December 1998 and of publishing these soon thereafter. However, WHO publications need to go through an internal approval process and it may be up to 12 months before the final WHO document appears. A combined working party of the Australian Diabetes Society, New Zealand Society for the Study of Diabetes, Royal College of Pathologists of Australasia and Australasian Association of Clinical Biochemists was formed to formulate an Australasian position on the two sets of recommendations and, in particular, on the differences between them. This is an interim statement pending the final WHO report, which will include recommendations on diabetes classification as well as criteria for diagnosis. We see it as very important to inform Australasian health professionals treating patients with diabetes about these changes. Position Statement key messages What are the new diagnostic criteria? The new WHO criteria for diagnosis of diabetes mellitus and hyperglycaemia are shown in Box 1. The major change from the previous WHO recommendation3 is the lowering of the diagnostic level of fasting plasma glucose to 7.0 mmol/L, from the former level of 7.8 mmol/L. For whole blood, the proposed new level is 6.1 mmol/L, from the former 6.7 mmol/L. This change is based primarily on cross-sectional studies demonstrating the presence of microvascular4 and macrovascular complications5 at these lower glucose concentrations. In addition, the 1985 WHO diagnostic criterion for diabetes based on fasting plasma glucose level ( 7.8 mmol/L) represents a greater degree of hyperglycaemia than the criterion based on plasma glucose level two hours after a 75 g glucose load ( 11.1 mmol/L).6 A fasting plasma glucose level of 7 mmol/L accords more closely with this 2 h post-glucose level. Recommendation: The ADA and the WHO committee are unanimous in adopting the changed diagnostic level, and the Australasian Working Party on Diagnostic Criteria recommends that healthcare providers in Australia and New Zealand should adopt it immediately. Clinicians should note that the diagnostic criteria differ between clinical and epidemiological settings. In clinical practice, when symptoms are typical of diabetes, a single fasting plasma glucose level of 7.0 mmol/L or 2 h post-glucose or casual postprandial plasma glucose level of 11.1 mmol/L suffices for diagnosis. If there are no symptoms, or symptoms are equivocal, at least one additional glucose measurement (preferably fasting) on a different day with a value in the diabetic range is necessary to confirm the diagnosis. Furthermore, severe hyperglycaemia detected under conditions of acute infective, traumatic, circulatory or other stress may be transitory and should not be regarded as diagnostic of diabetes. The situation should be reviewed when the primary condition has stabilised. In epidemiological settings, for study of high-prevalence populations or selective screening of high-risk individuals, a single measure -- the glucose-level 2 h post-glucose load -- will suffice to describe prevalence of impaired glucose tolerance (IGT). What about the oral glucose tolerance test? Previously, the oral glucose tolerance test (OGTT) was recommended in people with a fasting plasma glucose level of 5.5-7.7 mmol/L or random plasma glucose level of 7.8-11.0 mmol/L. After a 75 g glucose load, those with a 2 h plasma glucose level of < 7.8 mmol/L were classified as normoglycaemic, of 7.8-11.0 mmol/L as having IGT and of 11.1 mmol/L as having diabetes. The new diagnostic criteria proposed by the ADA and WHO differ in their recommendations on use of the OGTT. The ADA makes a strong recommendation that fasting plasma glucose level can be used on its own and that, in general, the OGTT need not be used.1 The WHO group2 argues strongly for the retention of the OGTT and suggests using fasting plasma glucose level alone only when circumstances prevent the performance of the OGTT. There are concerns that many people with a fasting plasma glucose level < 7.0 mmol/L will have manifestly abnormal results on the OGTT and are at risk of microvascular and macrovascular complications. This has major ramifications for the approach to diabetes screening, particularly when the Australian National Diabetes Strategy proposal,7 launched in June 1998 by Dr Michael Wooldridge, Federal Minister for Health and Aged Care, has early detection of type 2 diabetes as a key priority. Recommendation: The Australasian Working Party on Diagnostic Criteria has major concerns about discontinuing use of the OGTT and recommends that a formal recommendation on its use in diabetes screening be withheld until the final WHO recommendation is made. However, in the interim, the OGTT should continue to be used. Diabetes in pregnancy The ADA has retained its old criteria for diagnosis of gestational diabetes.1 These differ from those recommended by both WHO2 and the Australian Working Party on Diabetes in Pregnancy8 and are generally not recognised outside the United States. The new WHO statement retains the 1985 WHO recommendation that both IGT and diabetes should be classified as gestational diabetes. This is consistent with the recommendations of the Australasian Diabetes in Pregnancy Society, which recommended a diagnostic 2 h venous plasma glucose level on the OGTT of 8.0 mmol/L. In New Zealand, a cut-off level of 9.0 mmol/L has been applied.8 How has the classification of diabetes changed? The proposed new classification encompasses both clinical stages and aetiological types of hyperglycaemia and is supported by numerous epidemiological studies. The classification by aetiological type (Box 2) results from new knowledge of the causes of hyperglycaemia, including diabetes. The terms insulin-dependent and non-insulin-dependent diabetes (IDDM and NIDDM) are eliminated and the terms type 1 and type 2 diabetes retained. Other aetiological types, such as diabetes arising from genetic defects of -cell function or insulin action, are grouped as "other specific types", with gestational diabetes as a fourth category. The proposed staging (Box 3) reflects the fact that any aetiological type of diabetes can pass or progress through several clinical phases (both asymptomatic and symptomatic) during its natural history. Moreover, individuals may move in either direction between stages. Impaired glucose tolerance and impaired fasting glycaemia Impaired glucose tolerance (IGT), a discrete class in the previous classification, is now categorised as a stage in the natural history of disordered carbohydrate metabolism. Individuals with IGT are at increased risk of cardiovascular disease, and not all will be identified by fasting glucose level. In reducing the use of the OGTT, the ADA recommended a new category -- impaired fasting glycaemia (IFG) -- when fasting plasma glucose level is lower than that required to diagnose diabetes but higher than the reference range (< 7.0 mmol/L but 6.1 mmol/L). Limited data on this category show that it increases both risk of progressing to diabetes9 and cardiovascular risk.5 However, data are as yet insufficient to determine whether IFG has the same status as IGT as a risk factor for developing diabetes and cardiovascular disease and as strong an association with the metabolic syndrome (insulin resistance syndrome). IFG can be diagnosed by fasting glucose level alone, but if 2 h glucose level is also measured some individuals with IFG will have IGT and some may have diabetes. In addition, the number of people with OGTT results indicating diabetes but fasting plasma glucose level < 7.0 mmol/L is unknown, but early data suggest there may be major variation across different populations.10 A number of studies, including the DECODE initiative of the European Diabetes Epidemiology Group, have reported that individuals classified with IFG are not the same as the IGT group.11-15 The European Group believes that, on available European evidence, the ADA decision to rely solely on fasting glucose level would be unwise. Recommendation: The Australasian Working Party on Diagnostic Criteria recommends immediate adoption of the new classification. However, clinicians should be aware that some cases of diabetes will be missed unless an OGTT is performed. Thus, if there is any suspicion or other risk factor suggesting glucose intolerance, the working party continues to recommend use of an OGTT pending the final WHO recommendation. References Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997; 20: 1183-1197. Alberti KGMM, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Provisional Report of a WHO Consultation. Diabet Med 1998; 15: 539-553. World Health Organization. Diabetes mellitus. Report of a WHO study group. Technical report series 727. Geneva: WHO, 1985. McCance DR, Hanson RL, Charles MA, et al. Comparison of tests for glycated haemoglobin and fasting and two hour plasma glucose concentrations as diagnostic methods for diabetes. BMJ 1994; 308: 1323-1328. Charles MA, Balkau B, Vauzelle-Kervoeden F, et al. Revision of diagnostic criteria for diabetes [letter]. Lancet 1996; 348: 1657-1658. Finch CF, Zimmet PZ, Alberti KGMM. Determining diabetes prevalence: a rational basis for the use of fasting plasma glucose concentrations? Diabet Med 1990; 7: 603-610. Colagiuri S, Colagiuri R, Ward J. National diabetes strategy and implementation plan. Canberra: Diabetes Australia, 1998. Hoffman L, Nolan C, Wilson D, et al. Gestational diabetes mellitus -- management guidelines. The Australasian Diabetes in Pregnancy Society. Med J Aust 1998; 169: 93-97. Charles MA, Fontbonne A, Thibult N, et al. Risk factors for NIDDM in white population. Diabetes 1991; 40: 796-799. Keen H. Impact of new criteria for diabetes on pattern of disease. Lancet 1998; 352: 1000-1001. DECODE Study Group on behalf of the European Diabetes Epidemiology Study Group. Will new diagnostic criteria for diabetes mellitus change phenotype of patients with diabetes? Reanalysis of European epidemiological data. BMJ 1998; 317: 371-375. De Vegt F, Dekker JM, Stehouwer CDA, et al. The 1997 American Diabetes Association criteria versus the 1985 World Health Organization criteria for the diagnosis of abnormal glucose tolerance. Diabetes Care 1998; 21: 1686-1690. Harris MI, Eastman RC, Cowie CC, et al. Comparison of diabetes diagnostic categories in the US population according to 1997 American Diabetes Association and 1980-1985 World Health Organization diagnostic criteria. Diabetes Care 1997; 20: 1859-1862. Unwin N, Alberti KGMM, Bhopal R, et al. Comparison of the current WHO and new ADA criteria for the diagnosis of diabetes mellitus in three ethnic groups in the UK. Diabet Med 1998; 15: 554-557. Chang C-J, Wu J-S, Lu F-H, Lee H-L, et al. Fasting plasma glucose in screening for diabetes in the Taiwanese population. Diabetes Care 1998; 21: 1856-1860. Authors' details Department of Diabetes and Endocrinology, Royal Melbourne Hospital, Melbourne, VIC. Peter G Colman, FRACP, MD, Director. Chemical Pathology Services, Women's and Children's Hospital, Adelaide, SA. David W Thomas, FRACP, FRCPA, Head. International Diabetes Institute, Melbourne, VIC. Paul Z Zimmet, FRACP, MD, Director. Diabetes Centre, Sir Charles Gairdner Hospital, Perth, WA. Timothy A Welborn, FRACP, PhD, Head. St John of God Pathology, Perth, WA. Peter Garcia-Webb, MD, FRCPA, Chemical Pathologist. Diabetes Centre, Christchurch Hospital, Christchurch, NZ. M Peter Moore, FRACP, Clinical Director. Reprints will not be available from the authors. Correspondence: Dr P G Colman, Department of Diabetes and Endocrinology, Royal Melbourne Hospital, Parkville, VIC 3050. Email: peter.colmanATnwhcn.org.au Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/> Key messages Diagnosis of diabetes is not in doubt when there are classical symptoms of thirst and polyuria and a random venous plasma glucose level 11.1 mmol/L. The Australasian Working Party on Diagnostic Criteria for Diabetes Mellitus recommends: Immediate adoption of the new criterion for diagnosis of diabetes as proposed by the American Diabetes Association (ADA) and the World Health Organization (WHO) - fasting venous plasma glucose level 7.0 mmol/L; Immediate adoption of the new classification for diabetes mellitus proposed by the ADA and WHO, which comprises four aetiological types - type 1, type 2, other specific types, and gestational diabetes - with impaired glucose tolerance and impaired fasting glycaemia as stages in the natural history of disordered carbohydrate metabolism. Awareness that some cases of diabetes will be missed unless an oral glucose tolerance test (OGTT) is performed. If there is any suspicion or other risk factor suggesting glucose intolerance, the OGTT should continue to be used pending the final WHO recommendation. Back to text 1: Values for diagnosis of diabetes mellitus and other categories of hyperglycaemia2Glucose concentration (mmol/L [mg/dL]) Whole blood Venous Capillary Diabetes mellitusFasting 6.1 ( 110) 6.1 ( 110) or 2 h post-glucose load 10.0 ( 180) 11.1 ( 200) or bothImpaired glucose tolerance (IGT)Fasting (if measured)< 6.1 (< 110) < 6.1 (< 110) and 2 h post-glucose load 6.7 ( 120) 7.8 ( 140) and < 10.0 (< 180) and < 11.1 (< 200) Impaired fasting glycaemia (IFG)Fasting 5.6 ( 100) and 5.6 ( 100) and < 6.1 (< 110) < 6.1 (< 110) 2 h post-glucose load (if measured)< 6.7 (< 120) < 7.8 (< 140) Glucose concentration (mmol/L [mg/dL]) Plasma* Venous Capillary Diabetes mellitusFasting< 7.0 ( 126) 7.0 ( 126) or 2 h post-glucose load 11.1 ( 200) 12.2 ( 220) or bothImpaired glucose tolerance (IGT)Fasting (if measured)<7.0 (<126) < 7.0 (< 126) and 2 h post-glucose load 7.8 ( 140) 8.9 ( 160) and < 11.1 (< 200) and < 12.2 (< 220) Impaired fasting glycaemia (IFG)Fasting 6.1 ( 110) and 6.1 ( 110) and < 7.0 (< 126) < 7.0 (< 126) 2 h post-glucose load (if measured)< 7.8 (< 140) < 8.9 (< 160) For epidemiological or population screening purposes, the fasting or 2 h value after 75 g oral glucose may be used alone. For clinical purposes, the diagnosis of diabetes should always be confirmed by repeating the test on another day, unless there is unequivocal hyperglycaemia with acute metabolic decompensation or obvious symptoms. Glucose concentrations should not be determined on serum unless red cells are immediately removed, otherwise glycolysis will result in an unpredictable underestimation of the true concentrations. It should be stressed that glucose preservatives do not totally prevent glycolysis. If whole blood is used, the sample should be kept at 0-4oC or centrifuged immediately, or assayed immediately. Table reproduced with permission from Alberti KGMM, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Provisional Report of a WHO Consultation. Diabet Med 1998; 15: 539-553. Copyright John Wiley & Sons Limited. Back to text2: Aetiological classification of disorders of glycaemia* Type 1 (-cell destruction, usually leading to absolute insulin deficiency) Autoimmune Idiopathic Type 2 (may range from predominantly insulin resistance with relative insulin deficiency to a predominantly secretory defect with or without insulin resistance) Other specific types Genetic defects of -cell function Genetic defects in insulin action Diseases of the exocrine pancreas Endocrinopathies Drug or chemical induced Infections Uncommon forms of immune-mediated diabetes Other genetic syndromes sometimes associated with diabetes Gestational diabetes * As additional subtypes are discovered, it is anticipated they will be reclassified within their own specific category. Includes the former categories of gestational impaired glucose tolerance and gestational diabetes. Table reproduced with permission from Alberti KGMM, Zimmet PZ. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus. Provisional Report of a WHO Consultation. Diabet Med 1998; 15: 539-553. Copyright John Wiley & Sons Limited. Back to textBack to text

Peter G Colman · David W Thomas · Paul Z Zimmet · Timothy A Welborn · Peter Garcia-Webb

For Debate Polycystic ovary syndrome: a new direction in treatment Warren Kidson MJA 1998; 169: 537-540 Abstract - Introduction - Insulin resistance in the polycystic ovary syndrome - Polycystic ovaries and the "metabolic syndrome" - Diabetes therapies in the polycystic ovary syndrome - Managing polycystic ovary syndrome - Conclusion - References - Authors' details - - More articles on Endocrinology Abstract Polycystic ovary syndrome is a diagnosis made in 5%-10% of women between late adolescence and the menopause. Patients may present with oligomenorrhoea or amenorrhoea, anovulation or infertility, hirsutism or acne. Women with the syndrome have at least seven times the risk of myocardial infarction and ischaemic heart disease of other women, and by the age of 40 years up to 40% will have type 2 diabetes or impaired glucose tolerance. Polycystic ovary syndrome is associated with insulin resistance, with consequent hyperinsulinaemia and (frequently) hyperlipidaemia and obesity. Recent research has shown that the application of diabetes management techniques aimed at reducing insulin resistance and hyperinsulinaemia (such as weight reduction and the administration of oral hypoglycaemic agents) can not only reverse testosterone and luteinising hormone abnormalities and infertility, but can also improve glucose, insulin and lipid profiles. The management of polycystic ovary syndrome should now include patient education and attention to diabetes and cardiovascular risk factors such as hyperlipidaemia, obesity, physical exercise, glucose intolerance, hypertension and cigarette smoking. Introduction The finding of polycystic ovaries on ultrasound is not an unusual one, occurring in 21% of premenopausal women in population surveys.1 Polycystic ovaries are due to incomplete follicular development or to failure of ovulation, and are therefore often seen in women with bulimia, recovery from anorexia nervosa, conditions of increased adrenal androgen production and hyperprolactinaemia.2 In early to mid adolescence, the ultrasound finding of polycystic ovaries is common and not regarded as abnormal, presumably because of the high prevalence of non-ovulatory cycles at this age. Polycystic ovary syndrome, which affects 5% to 10% of premenopausal women,3,4 is one of the most common endocrinopathies of women. It consists of at least two of the following features: polycystic ovaries, hyperandrogenism and anovulation (see Box 1). The classic therapy for symptoms of hyperandrogenism (such as hirsutism and acne) was to suppress ovarian testosterone production with an oral oestrogen/progestogen contraceptive, often with the addition of the antiandrogenic progestogen cyproterone acetate. Infertility and anovulation are usually treated with clomiphene citrate, gonadotropins or laparoscopically applied physical therapies to the ovaries, such as laser or diathermy.12-14 Open wedge resection of ovaries is rarely used because of the risk of obstructive infertility from adhesions. Insulin resistance in the polycystic ovary syndrome In a study of women with polycystic ovary syndrome performed 18 years ago, most were found to be hyperinsulinaemic and to have a glucose metabolism that was resistant to the stimulatory effects of insulin.15The insulin resistance in type 2 diabetes and polycystic ovary syndrome occurs mainly in muscles,16 but also in the liver in obese women with polycystic ovary syndrome.11 Insulin resistance is aggravated by physical inactivity, upper abdominal obesity, hyperandrogenism, pregnancy, the ageing process and by medications such as thiazide diuretics, corticosteroids and certain hormonal steroid preparations. Insulin resistance in polycystic ovary syndrome is not due primarily to obesity (as lean women with polycystic ovary syndrome are insulin resistant) or to hyperandrogenism17 (as androgen blockade reduces insulin resistance by only 10%-15%).18 Insulin resistance leads to hyperinsulinaemia as pancreatic insulin secretion rises to maintain normoglycaemia. Hyperinsulinaemia can then stimulate lipid storage, altered lipoprotein and cholesterol metabolism and (possibly) altered steroid hormone metabolism. Hyperinsulinaemia increases ovarian androgen production19 by stimulating an ovarian enzyme complex cytochrome P450c17α, either directly and/or by stimulating pituitary luteinising hormone secretion. The accurate measurement of insulin resistance is an expensive, labour-intensive research technique. The easiest, but least sensitive, measure of insulin resistance is fasting serum insulin, with values between 10 and 14 mU/L (72-100 pmol/L) indicating mild insulin resistance and values above 14 mU/L indicating moderate or severe insulin resistance. As fasting serum insulin values lie in the least sensitive range of the immunoassay curve, fasting serum insulin is more accurate if a mean of three specimens taken over 10 minutes is used. The insulin assay should have no cross-reactivity with proinsulin. Insulin resistance can also be assessed by the serum insulin response to an oral glucose load during an oral glucose tolerance test, peak serum insulin levels above 100 mU/L (718 pmol/L) being highly suggestive of insulin resistance. As most women with polycystic ovary syndrome should have a glucose tolerance test, serum insulin can be measured on three fasting specimens, as well as at one and two hours, so that both parameters of insulin resistance can be assessed. Various measures of insulin resistance in polycystic ovary syndrome have recently been studied, and the ratio of fasting insulin (mU/L) to fasting glucose (mmol/L) has been found to be a simple and accurate indicator of insulin resistance (sensitivity 95%, specificity 84%, positive predictive value 87% and negative predictive value 94%) at values above 4mU/mmol.20 Polycystic ovaries and the "metabolic syndrome" Over the past decade, studies have shown that women with polycystic ovary syndrome have a high prevalence of hyperlipidaemia,5-7 hypertension,8 and progression to type 2 diabetes mellitus,11 similar to the features of the so-called "metabolic syndrome" or "syndrome X". Of greatest concern is an estimate that the relative risk of myocardial infarction in women with polycystic ovary syndrome is 7.4 times that of other women.9 An ovarian ultrasound study found that 46% of women undergoing coronary angiography for ischaemic heart disease before the age of 60 years had polycystic ovaries and that the degree of coronary narrowing was greater than in the remaining 54%.10 Assuming a prevalence rate of polycystic ovary syndrome of 10%, I calculate the relative risk of premature ischaemic heart disease in polycystic ovary syndrome in this study was 7.7. Diabetes therapies in the polycystic ovary syndrome The growing body of evidence linking polycystic ovary syndrome to an inherited resistance to insulin action, aggravated by lifestyle problems such as obesity, poor diet and physical inactivity (such as happens with type 2 diabetes mellitus), has led to trials of diabetic therapies in patients with the polycystic ovary syndrome. Many studies, not all referenced in this article, have demonstrated that the lifestyle measures of diet and weight reduction can not only lower insulin levels and reduce hyperlipidaemia, but can lower androgen and luteinising hormone levels, restore regular menstruation and ovulation and hence improve fertility.21-23 Regular physical exercise is often recommended,24 but has been poorly studied and documented, despite its apparent efficacy. Four trials (two controlled and two uncontrolled) of metformin, a diabetes medication that reduces insulin resistance, have demonstrated a fall in serum androgens, luteinising hormone and weight and an improvement in fertility and fibrinolysis in both obese and lean women with polycystic ovary syndrome.19,25-27 Two studies have shown no improvement with metformin.28,29 The women in the first of these two studies were Turkish, which may have influenced the result as it is known that many intracellular enzyme defects can lead to insulin resistance and that the nature of insulin resistance can vary between racial groups. In the second negative study, the diet of the subjects was modified to prevent weight loss during metformin therapy. A recent controlled trial was performed in the United States, Venezuela and Italy in which obese women with polycystic ovary syndrome were given either metformin or placebo.30 Women in both groups who had not ovulated by Day 35 were given clomiphene as additional medication. By Day 35, 34% of women taking metformin had ovulated, compared with only 4% of the placebo-treated women (P < 0.001). Within 18 days after the addition of clomiphene, 90% of the remaining women taking metformin ovulated, compared with 8% of the placebo-treated women (P < 0.001). In summary, within 53 days only 7% of women treated with metformin or metformin plus clomiphene had not ovulated, compared with 88% of women treated with clomiphene alone. Troglitazone, another diabetes medication, has been demonstrated to improve insulin sensitivity and to lower serum insulin, androgen and luteinising hormone levels without causing weight loss in two studies.31,32 A recent Lancet editorial on the use of insulin-sensitising agents in polycystic ovary syndrome concluded that "confirmation of the beneficial effects of metformin on hormonal and metabolic variables in women with polycystic ovary syndrome will have implications not only for the treatment of the common gynaecological presenting features, but also for the burden of vascular disease in women".33 Managing polycystic ovary syndrome Glucose tolerance testing and lipid measurements Most women with polycystic ovary syndrome should have an oral glucose tolerance test at diagnosis and at five-yearly intervals thereafter, and measurement of fasting lipids at diagnosis and at two- to three-yearly intervals. The exception would be a woman aged less than 20 who is not overweight and who does not have a family history of diabetes mellitus, gestational diabetes or large birth weight. A considerable number will be found to have impaired glucose tolerance or mild type 2 diabetes. The measurement of serum insulin in the fasting state and at one and two hours will detect most insulin-resistant women. The parents of the woman with polycystic ovary syndrome should also have glucose tolerance tests. Her siblings' glucose tolerance should be assessed if a parent is shown to be diabetic. Lifestyle changes Obesity should be treated with a structured diet and exercise program,34 with appropriate dietary modification for the woman with hyperlipidaemia. Simply telling a patient to "lose weight" or "eat less" is unlikely to result in a significant reduction in weight, judging from experience in the management of type 2 diabetes. Consultation with dietitians working in type 2 diabetes or commercial weight reduction programs may be useful in refractory obesity. Hyperlipidaemia persisting after dietary modification and weight reduction may require drug therapy. Cigarette smoking should be vigorously discouraged in all women with polycystic ovary syndrome as it will exacerbate the increased risk of atherosclerosis. Cigarette smoking has recently been shown to aggravate insulin resistance in type 2 diabetes mellitus.35 Attempts to reduce weight, increase exercise and stop smoking will fail if the woman with polycystic ovary syndrome is not educated about the long term adverse health implications of this condition. Too often these women are told only about the cosmetic nuisances of hirsutism or acne. Oral contraceptives Oral contraceptives reduce acne and hirsutism by inhibiting ovarian steroid production, including androgens, and by suppressing pituitary follicle stimulating hormone and luteinising hormone secretion. The effect is dependent on the oestrogen dose in the contraceptive. Oral contraceptives are also used to effect secretory change in the endometrium and reduce the suggested, but unproven, increased risk of endometrial carcinoma in polycystic ovary syndrome.36,37 Any therapy that results in regular ovulation, however, should reduce the risks of endometrial carcinoma. A recent uncontrolled study of 16 non-diabetic hyperandrogenic women treated with a combined oral contraceptive containing 150 µg of desogestrel and 30 µg of ethinyloestradiol demonstrated a significant deterioration in glucose tolerance over six months, with two women developing frank diabetes.38 This raises doubts about the short and long term safety of ovarian suppression in polycystic ovary syndrome with oral contraceptives. The effects of individual oral contraceptives on glucose tolerance will now need to be studied specifically in polycystic ovary syndrome before their use can be advocated. The use of a "triphasic" combined oral contraceptive should be avoided as the early cycle ethinyloestradiol dose is often too low to inhibit dominant follicle selection and hence follicular development,39 theoretically having the potential to increase, rather than reduce, the number of ovarian cysts. Androgen-blocking drugs Spironolactone and cyproterone acetate preparations are generally equally effective in the treatment of hirsutism and acne, with occasional patient differences in response. Spironolactone generally does not cause weight gain and slightly reduces insulin resistance,18 but can cause polymenorrhoea (which may respond to a reduction in dose from 100 mg daily to 50 mg). Serum potassium levels and renal function should be checked. Non-steroidal anti-inflammatory drugs can potentiate the effects of spironolactone on renal potassium retention. Ovulation is commonly restored by spironolactone therapy, an event which may or may not be desired by the patient. Cyproterone acetate not infrequently causes depression and, at a dose of 100 mg, weight gain,40 the latter aggravating insulin resistance. Cyproterone preparations are relatively expensive. Insulin-sensitising drugs Metformin is now being suggested as initial therapy for women with polycystic ovary syndrome whose condition does not respond to lifestyle measures, ahead of traditional hormonally active agents, for amelioration of hirsutism and for the restoration of regular menses and ovulation.41,42 The drug should be withdrawn after conception. My limited experience with metformin suggests that it is at least as effective as spironolactone or cyproterone in the treatment of hirsutism and acne. Further studies will undoubtedly define the indications for insulin-sensitising medications in polycystic ovary syndrome, but they are already indicated for women with established diabetes and may be useful for women with refractory obesity or hyperlipidaemia. Infertility Infertility in polycystic ovary syndrome is usually treated successfully by improving diet and exercise, weight reduction and spironolactone therapy. If these measures are not successful, conception can usually be achieved by one or more of clomiphene citrate, gonadotropins, gonadotropin-releasing hormone analogues, laparoscopically applied therapies to the ovaries and assisted reproductive techniques. Metformin therapy increases spontaneous ovulation and dramatically enhances the ovulatory response to clomiphene29 and will hopefully reduce the need for more expensive forms of ovulation induction. Pregnancy in polycystic ovary syndrome Women with polycystic ovary syndrome have recently been shown to have an increased prevalence of gestational diabetes or impaired glucose tolerance during pregnancy43,44 and should therefore have a glucose tolerance test early in pregnancy and again at 26 to 28 weeks. Conclusion Much epidemiological and basic endocrine research is yet required to answer questions about the proportion of women with type 2 diabetes who have had polycystic ovary syndrome and the role of other hormones such as androstenedione, dihydroepiandrosterone sulfate, oestrone, insulin-like growth factor and opioid receptors. However, we now have sufficient scientific information to recommend randomised controlled trials of new therapies that should not only make treatment of immediate problems more effective, but which should also reduce the long term prevalence of ischaemic heart disease and diabetes, with all of its attendant complications. Never before have we been able to identify such a large group of women at risk of diabetes and vascular disease at such an early age, when preventive measures are most effective. References Farquhar CM, Birdsall MA, Manning P, et al. The prevalence of polycystic ovaries on ultrasound scanning in a population of randomly selected women. Aust N Z J Obstet Gynaecol 1994; 34: 67-72. Isik AZ, Gulekli B, Zorlu CG, et al. Endocrinological and clinical analysis of hyperprolactinaemic patients with and without ultrasonically diagnosed polycystic ovarian changes. Gyn Obstet Invest 1997; 43: 183-185. Hull MGR. Epidemiology of infertility and polycystic ovarian disease: endocrinological and demographic studies. Gynecol Endocrinol 1987; 1: 235-245. Polson DW, Adams J, Wadsworth J, et al. Polycystic ovaries -- a common finding in normal women. Lancet 1988; 1: 870-872. Robinson S, Henderson AD, Gelding SV, et al. Dyslipidaemia is associated with insulin resistance in women with polycystic ovaries. Clin Endocrinol 1996; 44: 277-284. Birdsall MA, Farquhar CM. Polycystic ovaries in pre- and post-menopausal women. Clin Endocrinol 1996; 44: 269-276. Meirow D, Raz I, Yossepowitch O, et al. Dyslipidaemia in polycystic ovary syndrome: different groups, different aetiologies? Hum Reprod 1996; 11: 1848-1853. Wild RA. Obesity, lipids, cardiovascular risk and androgen excess. Am J Med 1995; 98 Suppl: 27S-32S. Dahlgren E, Jansen PO, Johansson S, et al. Polycystic ovary syndrome and risk for myocardial infaction. Evaluated from a risk factor model based on a prospective population study. Acta Obstet Gynaecol Scand 1992; 71: 599-604. Birdsall MA, Farquhar CM, White HD. Association between polycystic ovaries and extent of coronary artery disease in women having cardiac catherization. Ann Int Med 1997; 126: 32-35. Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocrine Rev 1997; 18: 774-800. Cohen J. Laparoscopic procedures for treatment of infertility related to polycystic ovary syndrome. Hum Reprod Update 1996; 2: 337-344. Pelosi MA, Pelosi III MA. Laparoscopic electrosurgical furrowing technique for the treatment of polycystic ovaries. J Am Assoc Gynecol Laparoscopists 1996; 4: 57-62. Pauthier S, Fernandez H, Lelaidier C, et al. Endoscopic laser co2 treatment of infertility due to polycystic ovary syndrome. Contraception Fertile Sexaulite 1997; 25: 147-151. Burghen GA, Givens JR, Kitabchi AE. Correlation of hyperandrogenism with hyperinsulinism in polycystic ovary disease. J Clin Endocrinol Metab 1980; 50: 113-116. DeFronzo RA. Lilly lecture 1987. The triumvirate: beta-cell, muscle, liver. A collusion responsible for NIDDM. Diabetes 1988; 37: 667-687. Barbieri RL, Hornstein MD. Hyperinsulinemia and ovarian hyperandrogenism: cause and effect. Endocrinol Metab Clin North Am 1988; 17: 685-703. Moghetti P, Tose F, Castello R, et al. The insulin resistance in women with hyperandrogenism is partially reversed by antiandrogen treatment: evidence that androgens impair insulin action in women. J Clin Endocrinol Metab 1996; 81: 952-960. Legro R, Finegood D, Dunaif A. A fasting glucose to insulin ratio is a useful measure of insulin sensitivity in women with polycystic ovary syndrome. J Clin Endocrinol Metabol 1998; 83: 2694-2698. Nestler JE, Jakubowicz DJ. Decreases in ovarian cytochrome P450c17a activity and serum free testosterone after reduction of insulin secretion in polycystic ovary syndrome. N Engl J Med 1996; 335: 617-623. Kiddy DS, Hamilton-Fairley D, Seppala M, et al. Diet-induced changes in sex hormone binding globulin and free testosterone in women with normal or polycystic ovaries: correlation with serum insulin and insulin-like growth factor-I. Clin Endocrinol 1989; 31: 757-763. Kiddy DS, Hamilton-Fairley D, Bush A, et al. Improvement in endocrine and ovarian function during dietary treatment of obese women with polycystic ovary syndrome. Clin Endocrinol 1992; 36: 105-111. Jakubowicz DJ, Nestler JE. 17 α-Hydroxyprogesterone responses to leuprolide and serum androgens in obese women with and without polycystic syndrome after dietary weight loss. J Clin Endocrinol Metab 1997; 82: 556-560. Derman RJ. Effects of sex steroids on womens' health: implications for practitioners. Am J Med 1995; 98 Suppl: 137S-143S. Velaquez EM, Mendoza S, Hamer T, et al. Metformin therapy in polycystic ovary syndrome reduces hyperinsulinaemia, insulin resistance, hyperandrogenaemia and systolic blood pressure whilst facilitating normal menses and pregnancy. Metab Clin Exper 1994; 43: 647-654. Velaquez EM, Mendoza S, Wang P, et al. Metformin therapy is associated with a decrease in plasma plasminogen activator inhibitor-1, lipoprotein(a) and immunoreactive insulin levels in patients with the polycystic ovary syndrome. Metab Clin Exper 1997; 46: 454-457. Nestler JE, Jakubowicz DJ. Lean women with polycystic ovary syndrome respond to insulin reduction with decreases in ovarian P450c170 α activity and serum androgens. J Clin Endocrinol Metab 1997; 82: 4075-4079. Acbay O, Gundogdu S. Can metformin reduce insulin resistance in polycystic ovary syndrome? Fertil Steril 1996; 65: 946-949. Ehrmann DA, Cavaghan MK, Imperial J, et al. Effects of metformin on insulin secretion, insulin action and ovarian steroidogenesis in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1997; 82: 524-530. Nestler J, Jakubowicz D, Evans W, et al. Effects of metformin on spontaneous and clomiphene-induced ovulation in the polycystic ovary syndrome. N Engl J Med 1998; 338: 1876-1880. Dunaif A, Scott D, Finegood D, et at. The insulin-sensitizing agent troglitazone improves metabolic and reproductive abnormalities in the polycystic ovary syndrome. J Clin Endocrinol Metab 1996; 81: 3299-3306. Ehrmann DA, Schneider DJ, Sobel BE, et al. Troglitazone improves defects in insulin action, insulin secretion, ovarian steroidogenesis and fibrinolysis in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1997; 82: 2108-2116. Sattar N, Hopkinson Z, Greer I. Insulin-sensitising agents in polycystic-ovary syndrome. Lancet 1998; 351: 305-307. Clark AM, Ledger W, Galletly C, et al. Weight loss results in significant improvement in pregnancy and ovulation rates in anovulatory obese women. Hum Reprod 1995; 10: 2705-2712. Targher G, Alberto M, Zenere M, et al. Cigarette smoking and insulin resistance in patients with noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab 1997; 82: 3619-3624. Dahlgren E, Friberg L, Johansson S, et al. Endometrial carcinoma; ovarian dysfunction -- a risk factor in young women. Eur J Obstet Gynaecol Reprod Biol 1991; 41: 143-150. Ho S, Tan K, Pang M, Ho T. Endometrial hyperplasia and the risk of endometrial carcinoma. Singapore Med J 1997; 38: 11-15. Nader S, Riad-Gabriel M, Saad M. The effect of a desogestrel-containing oral contraceptive on glucose tolerance and leptin concentrations in hyperandrogenic women. J Clin Endocrinol Metab 1997; 82: 3074-3077. Fauser BCJM, Van Heusden AM. Manipulation of human ovarian function: physiological concepts and clinical consequences. Endocrine Rev 1997; 18: 71-106. Belisle S, Love E. Clinical efficacy and safety of cyproterone acetate in severe hirsutism: results of a multicentered Canadian study. Fertil Steril 1986; 46: 1015-1020. Nestler JE. Role of hyperinsulinaemia in the pathogenesis of the polycystic ovary syndrome, and its clinical implications. Semin Reprod Endocrinol 1997; 15: 111-122. Utiger RD. Insulin and the polycystic ovary syndrome. N Engl J Med 1996; 335: 657-658. Paradisi G, Fulghesu A, Ferrazzani S, et al. Endocrino-metabolic features in women with polycystic ovary syndrome during pregnancy. Human Reprod 1998; 13: 542-546. Anttila L, Karjala K, Penttila R, et al. Polycystic ovaries in women with gestational diabetes. Obstet Gynecol 1998; 92: 13-16. (Received 27 Jan, accepted 21 Jul, 1998) Author's details Sydney, NSW. Warren Kidson, MB BS, FRACP, Visiting Endocrinologist, Prince of Wales Hospital Randwick and Visiting Physician, Royal Hospital for Women, Paddington, NSW. Reprints: Dr Warren Kidson, 24 Blenheim Street, Randwick, NSW 2031. E-mail: wkidsonATmedeserv.com.au Make a comment Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/>

Warren Kidson

Gestational diabetes mellitus -- management guidelines The Australasian Diabetes in Pregnancy Society Linda Hoffman, Chris Nolan, J Dennis Wilson, Jeremy J N Oats and David Simmons MJA 1998; 169: 93-97 Synopsis - Screening - Diagnosis - Management of GDM - Patient education - Fetal surveillance - Timing of delivery - Delivery - Neonatal management - Maternal follow-up - Directions for future research - Acknowledgements - References - Authors' details - - ©MJA1998 Synopsis GDM is defined as carbohydrate intolerance of variable severity with onset or first recognition during pregnancy. Universal screening is recommended. If selective screening is considered more appropriate (because of limited resources or known low GDM incidence), screening may be reserved for those at higher risk. Risk factors include glycosuria, age over 30 years, obesity, family history of diabetes, past history of GDM or glucose intolerance, previous adverse pregnancy outcome and belonging to a high risk ethnic group. The recommended screening test for GDM is performed at 26-28 weeks' gestation and positive results are: 1 hour venous plasma glucose level ≥7.8 mmol/L after a 50 g glucose load (morning, non-fasting); or 1 hour venous plasma glucose level ≥8.0 mmol/L after a 75 g glucose load (morning, non-fasting). Confirmation of diagnosis after a positive screening test: a 75 g oral glucose tolerance test (fasting) with a venous plasma glucose level at 0 hours of ≥5.5 mmol/L and/or at 2 hours of ≥8.0 mmol/L.* Patient education is very important and a team approach, if available, is beneficial. Dietary therapy is the primary therapeutic strategy, with insulin added where required to achieve the minimum goals for glycaemic control: fasting blood glucose <5.5 mmol/L, 1 hour postprandial <8.0 mmol/L or 2 hour postprandial <7.0 mmol/L. Careful antepartum fetal surveillance is essential. Continuation of the pregnancy in uncomplicated GDM to 10 days beyond term is acceptable provided that indications from fetal monitoring are reassuring. Close neonatal monitoring is important, particularly for the detection of hypoglycaemia. Maternal follow-up, with an oral glucose tolerance test, should be performed 6-8 weeks postpartum, then at least every 2 years, because of the increased risk of developing permanent diabetes. Prospective trials are needed to clarify whether universal screening is justified, and to determine the degree of maternal hyperglycaemia that causes an adverse outcome for the offspring. The management strategies in this article have been the subject of widespread discussion with the ADIPS membership between 1991 and 1998 at annual scientific meetings, annual general meetings, and in ADIPS newsletters. They represent the majority opinion. * Cut-off point for Australia; cut-off point in New Zealand ≥9.0 mmol/L. Introduction Gestational diabetes mellitus (GDM) is defined as carbohydrate intolerance of variable severity with onset or first recognition during pregnancy.1 Women with GDM are a heterogeneous group and may include those with unrecognised pre-existing non-insulin-dependent diabetes (type 2) and also a small number with insulin-dependent diabetes. The presence of GDM has implications for both the baby and the mother. Although there is no evidence that perinatal mortality is increased in pregnancies with treated GDM, some studies have shown perinatal mortality to be increased in untreated GDM.2-4 GDM is associated with increased perinatal morbidity, the characteristics of which are the same as for infants of mothers with overt diabetes (eg, macrosomia, neonatal hypoglycaemia, hyperbilirubinaemia, respiratory distress syndrome).5 In considering longer term outcomes for the baby, evidence is gradually mounting that GDM adds an intrauterine environmental risk factor to an already increased genetic risk for the development of obesity and/or diabetes.6-8 In one follow-up study insulin therapy for GDM was associated with less adiposity in the offspring.9 For the mother, GDM is a very strong risk factor for the development of permanent diabetes later in life (49.9% with up to 28 years' follow-up).10 Screening There has been much debate about whether universal or selective screening of pregnant women for GDM is more appropriate.11-13 Moses and Colagiuri recently estimated that, between 1991 and 1994, 50% of pregnant women in New South Wales were not screened for gestational diabetes.14The Australasian Diabetes in Pregnancy Society (ADIPS) recommends that screening for GDM should be considered in all pregnant women. However, if resources are limited, screening may be reserved for those at highest risk. Risk factors include: Glycosuria; Age over 30 years; Obesity; Family history of diabetes; Past history of GDM or glucose intolerance; Previous adverse pregnancy outcome; and Belonging to an ethnic group with a high risk for GDM. Ethnicity is a particularly important factor determining incidence of GDM (eg, very high risk -- Australian Indigenous, Polynesian and South Asian [Indian] groups; moderate high risk -- Middle Eastern and other Asian groups).15 Most Australian centres report a GDM incidence of 5.5%-8.8%.16-19 When selective screening is deemed more appropriate because of known low GDM incidence, the ADIPS criteria are similar to those recommended by the American "Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus".20 A recent article by Naylor et al21 derived a risk factor scoring system that excluded the need for screening up to a third of pregnant women. However, complex criteria for selective screening may cause difficulties in busy clinical practice. A summary of the screening and diagnostic procedures recommended by ADIPS is given in the Table. Diagnosis The guidelines for diagnosing GDM in Australia are essentially unchanged from those recommended for use in Australasia in 1991.22 Although there are no uniform international criteria for the diagnosis of GDM, commonly used criteria are those of O'Sullivan and Mahan23 and the World Health Organization (WHO).24 One problem with the development of absolute diagnostic criteria is the lack of evidence that perinatal mortality is increased in pregnancies associated with mild degrees of hyperglycaemia. The commonly used diagnostic criteria were not formulated to assess the risk of adverse perinatal outcomes, although this was a factor taken into account in the diagnostic criteria at the Mercy Hospital for Women, Melbourne.4 The existence of different methods of performing glucose tolerance tests has also hindered the development of uniform diagnostic criteria for GDM. After consensus, ADIPS has endorsed the diagnostic criteria developed by the working party chaired by Dr F I R Martin in 1991, which are modified WHO criteria.22 In New Zealand, the 2 hour oral glucose tolerance test (OGTT) cut-off value for a positive diagnosis is a venous plasma glucose level of 9.0 mmol/L. This figure was chosen by a majority decision of specialists at the 1992 meeting of the New Zealand Society for the Study of Diabetes. They chose the higher figure to reduce the worry and inconvenience for women of being given a false positive diagnosis and to reduce the strain on stretched specialist resources in many centres. ADIPS recognises the importance of working towards an Australasian consensus on this issue. If the clinical suspicion of GDM is high, a diagnostic OGTT is indicated, irrespective of the stage of pregnancy. In such circumstances, if an OGTT gives normal results early in pregnancy the test should be repeated between 26 and 30 weeks' gestation. A 75 g OGTT should use 75 g of anhydrous glucose or the equivalent, and preferably should also be performed after a high carbohydrate diet of at least 150 g of carbohydrate for three days. Management of GDM A team approach is ideal for managing women with GDM and, if available, should be used. The team would usually comprise an obstetrician, diabetes physician, a diabetes educator (diabetes midwifery educator), dietitian, midwife and paediatrician. In practice, however, the team approach is not always possible due to limited resources. In such circumstances, management by an obstetrician or obstetric general practitioner knowledgeable in GDM management, often with the assistance of an appropriately skilled dietitian, diabetes educator or midwife, is acceptable. Patient education The importance of educating women with GDM (and their partners) about the condition and its management cannot be overemphasised. Compliance with the treatment plan depends on the patient's understanding of: The implications of GDM for her baby and herself; The dietary and exercise recommendations; and The how and when as well as the goals of self monitoring of blood glucose level. Care should be taken to minimise the anxiety of the women. Glycaemic control Dietary therapy: Dietary therapy is the primary therapeutic strategy for the achievement of acceptable glycaemic control in GDM. All women should receive nutritional advice, preferably from an appropriately skilled dietitian. However, it is important to avoid a severe calorie-restricted diet, as this can predispose to ketonuria, and also to infants that are small for their gestational age, which carries an increased risk of diabetes in later life.25 The diet needs to: Conform with the principles of dietary management of diabetes in general; Meet the nutritional requirements of pregnancy; Be individualised for each patient, depending on maternal weight and body mass index; and Be culturally appropriate. Moderate exercise has recently been recognised as an adjunct therapy, with potential benefits when used together with diet, or diet and insulin therapy, in the management of gestational diabetes in women without a medical or obstetric contraindication.26 Monitoring: Glycaemic control needs to be monitored. Self monitoring of blood glucose level is the optimal method and is well tolerated by most women. On commencement of self monitoring, at least one fasting and one 1 or 2 hour postprandial glucose level should be obtained daily. The frequency may be decreased or increased depending on the results of the blood glucose monitoring and the progress of the pregnancy. If self monitoring is not possible, fasting and 1 or 2 hour postprandial laboratory capillary blood or venous plasma glucose levels should be performed regularly (at 1 to 2 weekly intervals). In pregnancies complicated by GDM, the value of self monitoring of blood glucose and appropriate insulin therapy in the prevention of macrosomia and its associated perinatal complications has previously been demonstrated.27,28 The minimum goals for glycaemic control are: a fasting capillary (venous plasma) blood glucose level <5.5 mmol/L a 1 hour postprandial capillary (venous plasma) blood glucose level <8.0 mmol/L a 2 hour postprandial capillary (venous plasma) blood glucose level <7.0 mmol/L. These minimum goals have been set on the basis of informed consensus opinion in Australasia and vary little from those of the American Diabetes Association clinical practice recommendations on gestational diabetes (fasting glucose ≤5.8 mmol/L and 2 hour postprandial plasma glucose ≤6.7 mmol/L).29 The setting of minimal goals for glycaemic control is controversial, however, as some, but not all, studies show benefit from tight glycaemic control in women with GDM.27,28,30-32 The reasons for the variance in results between studies may relate to differences in the underlying rates of GDM complications from one study population to another. The recommended fasting glycaemia goal of <5.5 mmol/L is supported by Langer et al, who have shown that rates of large-for-gestational-age (LGA) infants are increased in diet-treated GDM pregnancies if the fasting glucose level is between 5.3 and 5.8 mmol/L (28.6% LGA) compared with ≤5.3 mmol/L (5.35% LGA).32 Insulin treatment was shown to reduce the rates of LGA infants to 10.3% in GDM pregnancies with fasting glucose levels between 5.3 and 5.8 mmol/L.32 In support of the 1 and 2 hour postprandial glycaemic goals of <8.0 and <7.0 mmol/L, respectively, it has been shown that glycohaemoglobin (HbA1c) levels, birth weight, and rates of macrosomia, neonatal hypoglycaemia and caesarean section (for cephalopelvic disproportion) can all be significantly reduced in insulin-treated GDM subjects if insulin therapy is adjusted according to 1 hour postprandial, rather than preprandial, glucose measurements, aiming for <7.8 mmol/L.27 HbA1c levels may be used as an ancillary test, as assurance that the self monitored blood glucose results are appropriate. Fructosamine levels are reduced during pregnancy because of the dilutional effect of pregnancy on plasma proteins. HbA1c and fructosamine are not reliable substitutes for self monitoring of blood glucose level. Insulin therapy: Insulin therapy should be considered if the blood glucose goals are exceeded on two or more occasions within a 1 to 2 week interval, particularly in association with clinical or investigational suspicion of macrosomia. However, the benefit of instituting insulin therapy after 38 weeks' gestation is unproven. Human insulin should be used. No insulin preparations have a pregnancy category listing, except for the new, rapidly acting insulin analogue lispro, which is Category B2 (Australian medicines in pregnancy category). Two cases of congenital malformations were recently noted in women with insulin-dependent diabetes treated in pregnancy with lispro.33 The number of women treated with lispro in pregnancy is small to date, but no causative relationship between lispro and teratogenicity has been documented. In general, the insulin preparations and dosage schedules should be tailored to the abnormalities present in the glycaemic profile (eg, postprandial and/or fasting hyperglycaemia) and patient acceptability. The doses may be higher than those required in non-pregnant subjects and should be reviewed frequently so that adequate glycaemic control is achieved rapidly. Care should be taken to minimise the risk of hypoglycaemia, especially nocturnal episodes. Oral hypoglycaemic agents have no place in treatment of GDM under normal circumstances. Fetal surveillance The timing of commencement and the frequency of fetal monitoring in pregnancies complicated by GDM depend on the presence of other pregnancy complications such as pre-eclampsia, hypertension, antepartum haemorrhage and intrauterine growth retardation. The regimen chosen should be dictated by the severity of the obstetric complication. Monitoring may be by either Doppler umbilical bloodflow measurement or cardiotocograph (CTG). Although CTG surveillance is commonly undertaken routinely from around 36 weeks' gestation, there is no objective evidence that fetal monitoring in uncomplicated GDM affects fetal outcome.34 Common practice in the United States is to commence CTG monitoring after 40 weeks' gestation, while awaiting spontaneous onset of labour in uncomplicated GDM pregnancies,35 but again there is no evidence-based medicine to support or refute this practice. Ultrasonography should be considered at around 34 weeks' gestation to detect abnormalities of fetal growth and polyhydramnios. It may be indicated earlier in some women, for example for women unsure of their dates, or those with morbid obesity or suspected undiagnosed non-insulin-dependent diabetes. Ultrasonography may need to be repeated if any abnormality is detected. Timing of delivery The possibility that diagnosis of GDM may lead to increased obstetric intervention, including induction of labour and caesarean section,36 is a concern. Delivery before full term is not indicated unless there is evidence of macrosomia, polyhydramnios, poor metabolic control or other obstetric indications (eg, pre-eclampsia or intrauterine growth retardation).37 Continuation of the pregnancy in uncomplicated GDM to 10 days beyond term is acceptable provided that indications from fetal monitoring are reassuring. Delivery During labour, good glycaemic control needs to be maintained while avoiding hypoglycaemia. Lower insulin requirements are common during labour (often no insulin is necessary). Fetal surveillance is needed, as it is for any high risk pregnancy. A paediatrician should be present at the delivery if significant neonatal morbidity is suspected. The maternal blood glucose level should be monitored for 24 hours postpartum and, if indicated, continued for longer. Neonatal management The neonates of mothers with GDM are at risk of all the complications of infants born to mothers with overt diabetes, particularly those infants born macrosomic (birth weight >4000 g).38 The neonates should be observed closely after delivery for respiratory distress. Capillary blood glucose should be monitored at 1 hour of age and before the first four feeds (and for up to 24 hours in high risk neonates). Currently, some amperometric blood glucose meters are acceptable for use in neonates, provided that suitable quality control procedures and operator training are in place. A neonatal blood glucose level <2.0 mmol/L needs to be verified by repeat testing (laboratory verification is preferred but should not delay the initiation of treatment). Levels <2.0 mmol/L should be considered abnormal and treated. If the baby is obviously macrosomic, calcium and magnesium levels should be checked on Day 2. Breastfeeding is actively encouraged. Maternal follow-up It is important that women with GDM be counselled with regard to their increased risk of developing permanent diabetes. They should be made aware of the symptoms of hyperglycaemia. Advice should be given about the importance of healthy eating and exercise patterns. Contraceptive advice should be given in the puerperium, and women should be advised to plan future pregnancies and be reviewed medically by their general practitioner before conception (a pre-conception OGTT should be considered). An OGTT, using WHO criteria for the non-pregnant population, should be performed at 6-8 weeks' postpartum to exclude permanent diabetes. Repeat OGTTs should be performed at least every two years (possibly at the same time as the cervical cancer screening). Impaired glucose tolerance merits careful follow-up, which should include at least twice-yearly checks for frank diabetes in addition to assessment of other risk factors for macrovascular disease. The rates of development of permanent diabetes are much higher in several non-European ethnic groups. For example, the prevalence of type 2 diabetes in Polynesian women having a postpartum OGTT has been reported to be 30%.39 Life-table analysis in a cohort of Latino women shown to have normal glucose tolerance in the postpartum period after pregnancy complicated by GDM revealed a 47% cumulative incidence of type 2 diabetes 5 years after delivery.40 Similarly, 62% of women in Trinidad have been reported to develop type 2 diabetes after 3.6-6.5 years of follow-up.41 Follow-up OGTTs, therefore, should be more frequent than every two years in those groups at highest risk. Directions for future research ADIPS emphasises that, due to a lack of good quality randomised controlled clinical trials in the area of GDM, these guidelines are based on what is a reasonable consensus of informed opinion in Australasia. They are designed as a guide to practical management rather than a strict protocol. It is expected that the guidelines will not be static but will evolve as the results of clinical trials become available. Carefully designed, randomised controlled clinical trials are needed in order to determine: Whether universal screening programs are warranted; The optimal criteria for diagnosis of GDM; The costs v. benefits of the team approach; Optimal management (eg, clarification of the indications for insulin therapy); The role of follow-up programs for affected mothers and babies; and Possible interventions to reduce the rates of development of permanent diabetes in the mother. One such trial is the prospective Australasian Carbohydrate Intolerance Study in Pregnancy (ACHOIS), which aims to clarify what degree of maternal hyperglycaemia results in specific adverse outcome. In the design of these trials consideration needs to be given not only to perinatal outcome, but also to the potential long term benefits of diagnosis and treatment for both the baby and the mother. Acknowledgements The assistance of all members of ADIPS who contributed to the consensus statement is gratefully acknowledged. Valerie Arnol's assistance is also gratefully acknowledged. References Metzger BE, editor. Proceedings of the third international workshop-conference on gestational diabetes mellitus. Diabetes 1991; 40 Suppl 2: 1-201. O'Sullivan JB, Charles D, Mahan CM, Dandrow RV. Gestational diabetes and perinatal mortality rate. Am J Obstet Gynecol 1973; 136: 901-904. Pettitt DJ, Knowler WC, Baird HR, Bennett PH. Gestational diabetes: infant and maternal complications of pregnancy in relation to third trimester glucose tolerance in Pima Indians. Diabetes Care 1980; 3: 458-464. Oats JN, Beischer NA. Gestational diabetes. Aust N Z J Obstet Gynaecol 1986; 26: 2-10. Hod M, Merlob P, Friedman S, et al. Gestational diabetes mellitus: a survey of perinatal complications in the 1980s. Diabetes 1991; 40 (Suppl 2): 74-78. Van Asche FA, Aerts L, Holemans K. The effects of maternal diabetes on the offspring. Baillieres Clin Obstet Gynaecol 1991; 5: 485-492. Silverman BL, Metzger BE, Cho NH, Loeb CA. Impaired glucose tolerance in adolescent offspring of diabetic mothers: relationship to fetal hyperinsulinism. Diabetes Care 1995; 18: 611-617. Pettitt DJ. Diabetes in subsequent generations. In: Dornhurst A, Hadden DR, editors. Diabetes and pregnancy: an international approach. Chichester: J Wiley and Sons, 1996: 367-376. Simmons D, Robertson S. Influence of maternal insulin treatment on the infants of women with gestational diabetes. Diabet Med 1997; 14: 762-765. O'Sullivan JB. The Boston Gestational Diabetes Studies: review and perspectives. In: Sutherland HW, Stowers JM, Pearson DWM, editors. Carbohydrate metabolism in pregnancy and the newborn. London: Springer-Verlag, 1989: 287-294. Greene MF. Screening for gestational diabetes. N Engl J Med 1997; 337: 1625-1626. Jarrett RJ. Should we screen for gestational diabetes? BMJ 1997; 315: 736-737. Soares J de AC, Dornhurst A, Beard RW. The case for screening for gestational diabetes. BMJ 1997; 315: 737-739. Moses RG, Colagiuri S. The extent of undiagnosed gestational diabetes mellitus in New South Wales. Med J Aust 1997; 167: 14-16. Beischer NA, Oats JN, Henry OA, et al. Incidence and severity of gestational diabetes mellitus according to country of birth in women living in Australia. Diabetes 1991; 40: 35-38. Beischer NA, Wein P, Sheedy MT, Steffen B. Identification and treatment of women with hyperglycaemia diagnosed during pregnancy can significantly reduce perinatal mortality rate. Aust N Z J Obstet Gynaecol 1996; 36: 239-247. Moses R, Griffiths R, McPherson S. The incidence of gestational diabetes in the Illawarra area of New South Wales. Aust N Z Obstet Gynaecol 1994; 34: 425-427. Martin FIR, Ratnaike S, Wootton A, et al. The 75 g oral glucose tolerance in pregnancy. Diabetes Res Clin Pract 1995; 27: 147-151. Yue DK, Molyneau LM, Ross GP, et al. Why does ethnicity affect prevalence of gestational diabetes? The underwater volcano theory. Diabet Med 1996; 13: 748-752. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 1997; 20: 1183-1197. Naylor CD, Sermer M, Chen E, Farine D. Selective screening for gestational diabetes mellitus. N Engl J Med 1997; 337: 1591-1596. Martin FIR. The diagnosis of gestational diabetes. Med J Aust 1991; 155: 112. O'Sullivan JB, Mahan CM. Criteria for the oral glucose tolerance test in pregnancy. Diabetes 1964; 13: 278-285. World Health Organization Study Group, Diabetes Mellitus. World Health Organ Tech Rep Ser 1985; 727: 13-14. McCance DR, Pettitt DJ, Hanson RL, et al. Birth weight in non insulin dependent diabetes. Thrifty genotype, thrifty phenotype or surviving small baby genotype? BMJ 1994; 308: 942-945. Jovenovic-Peterson L, Peterson CM. Is exercise safe or useful for gestational diabetic women? Diabetes 1991; 40 Suppl 2: 179-181. De Veciana M, Major CA, Morgan MA, et al. Postprandial versus preprandial blood glucose monitoring in women with gestational diabetes mellitus requiring insulin therapy. N Engl J Med 1995; 333: 1237-1241. Langer O, Rodriguez DA, Xenakis EMJ, et al. Intensified versus conventional management of gestational diabetes. Am J Obstet Gynecol 1994; 170: 1036-1047. American Diabetes Association: clinical practice recommendations 1997. Diabetes Care 1997; 20 Suppl 1: S1-S70. Hare JW. Gestational diabetes mellitus. Levels of glycemia as management goals. Diabetes 1991; 40 Suppl 2: 193-196. Garner P, Okun N, Keely E, et al. A randomized controlled trial of strict glycemic control and tertiary level obstetric care versus routine obstetric care in the management of gestational diabetes: a pilot study. Am J Obstet Gynecol 1997; 177: 190-195. Langer O, Berkus M, Brustman L, et al. Rationale for insulin management in gestational diabetes mellitus. Diabetes 1991; 40 Suppl 2: 186-190. Diamond T, Kormas N. Possible adverse effect of insulin lispro. N Engl J Med 1997; 337: 1009-1010. Landon MB, Langer O, Gabbe SG, et al. Fetal surveillance in pregnancies complicated by insulin-dependent diabetes mellitus. Am J Obstet Gynecol 1992; 167: 617-621. Carr DB, Gabbe S. Gestational diabetes: detection, management and implications. Clin Diabetes 1998; 16: 4-11. Hunter DJS, Keirse MJNC. Gestational diabetes. In: Chalmers I, Enkin M, Keirse MJ NC, editors. Effective care in pregnancy and childbirth. Vol 1. Oxford: Oxford University Press, 1989: 403-410. Rasmussen MJ, Firth R, Roley M, Stronge JM. The timing of delivery in diabetic pregnancy: a 10-year review. Aust N Z J Obstet Gynaecol 1992; 32: 313-317. Maresh M, Beard RW, Bray CS, et al. Factors predisposing to and outcome of gestational diabetes. Obstet Gynaecol 1989; 74: 542-546. Simmons D, Conroy C, Thompson C. Diabetes in pregnancy in South Auckland. Proceedings of the Australasian Diabetes in Pregnancy Society Annual Scientific Meeting, Melbourne, 1995. Melbourne: ADIPS, 1995: 55. Kjos SL, Peters RK, Xiang A, et al. Predicting future diabetes in Latino women with gestational diabetes. Utility of early postpartum glucose tolerance testing. Diabetes 1995; 44: 586-591. Ali Z, Alexis SD. Occurrence of diabetes mellitus after gestational diabetes mellitus in Trinidad. Diabetes Care 1990; 13: 527-529. Authors' details Department of Diabetes and Endocrinology, Royal Hobart Hospital, Hobart, TAS. Linda Hoffman, MD, FRACP, Visiting Specialist. School of Nutrition and Public Health, Deakin University, and Geelong Hospital, Geelong, VIC. Chris Nolan, PhD, FRACP, Postdoctoral Research Fellow, and Visiting Specialist. Department of Endocrinology, The Canberra Hospital, Canberra, ACT. J Dennis Wilson, MD, FRACP, Director of Endocrinology. Master Misericordiae Mothers' Hospital, South Brisbane, QLD. Jeremy J N Oats, DM, FRACOG, Director, and Clinical Professor of Obstetrics and Gynaecology. Department of Medicine, Middlemore Hospital, University of Auckland, Auckland, New Zealand. David Simmons, MD, FRACP, Senior Lecturer in Medicine. Reprints: Associate Professor L Hoffman, Department of Diabetes and Endocrinology, Royal Hobart Hospital, 48 Liverpool Street, Hobart, TAS 7001. - Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/>

Linda Hoffman · Chris Nolan · David Simmons

Prostate-specific antigen testing in Australia and association with prostate cancer incidence in New South Wales David P Smith and Bruce K Armstrong MJA 1998; 169: 17-20 For editorial comment, see McCredie & Cox Abstract - Introduction - Methods - Results - Discussion - Acknowledgements - References - Authors' details - - ©MJA1998 Abstract Objective: To describe patterns and trends in prostate-specific antigen (PSA) testing in Australia and assess its role in the increasing incidence of prostate cancer. Design: Descriptive analysis of (i) Medicare records of PSA testing in Australia, and (ii) prostate cancer recorded incidence in New South Wales. Data: (i) Medicare data for all males who received a Medicare-reimbursed PSA test between August 1989 and December 1996. (ii) NSW Central Cancer Registry data for all males in NSW with prostate cancer diagnosed between 1988 and 1995. Main outcome measures: (i) Number of PSA tests, age-standardised rates of PSA tests by State and Territory, and proportions of males who had a PSA test. (ii) Recorded incidence of prostate cancer in NSW. Results: (i) More than 2.2 million PSA tests were done on more than 1.1 million Australians between 1989 and 1996. The annual number of males tested increased fivefold in this period and peaked in 1995. Twenty-seven per cent of Australian men aged 50 years or over had at least one PSA test in 1995 or 1996; 33% of men aged 60-69 years had a test in this period. (ii) In NSW the number of PSA tests per quarter was highly correlated with the number of new cases of prostate cancer (R2 = 0.92). Conclusions: Although no organised program for prostate cancer screening exists, and despite repeated advice against it, opportunistic screening has been occurring at high rates. There was a high correlation between PSA testing and prostate cancer incidence between 1990 and 1995 in NSW. Introduction Prostate cancer is the most common cancer in Australians after non-melanocytic skin cancers, and accounts for about a third of all newly diagnosed internal cancers in males.1,2 It is the second most common cause of cancer death among Australian males.3 During the late 1980s and early 1990s, recorded prostate cancer incidences increased substantially in Australia, while mortality from prostate cancer remained stable.1,2,4-6 Much of the increase in incidence has been attributed to detection of latent cancers by increased use of the prostate-specific antigen (PSA) test, transurethral resection of the prostate, prostatic biopsy and radical prostatectomy.7There are considerable arguments for8,9 and against10,11 screening for prostate cancer. In 1996 the Australian Health Technology Advisory Committee reviewed the evidence and recommended against screening. However, the committee recognised that de facto screening occurs in the community and stated a need to monitor and review the evidence when important developments occur.12 Data collected by the Health Insurance Commission provide the best available means to monitor trends and patterns in PSA testing. We analysed these data to determine the association between PSA testing and reported incidence of prostate cancer in Australia and in New South Wales. Methods Data PSA tests: The Commonwealth Department of Health and Family Services Medicare Estimates and Statistics Unit provided de-identified data, extracted from the national dataset of all services rendered on a fee-for-service basis for which a Medicare benefit has been paid. The data included all tests itemised under the Medicare Benefits Schedule codes that included PSA tests. PSA tests first appeared in the Schedule in August 1989, but were included with 20 "other" biochemical tests until November 1993. Since then they have been itemised together with prostatic acid phosphatase (PAP) tests. PSA tests could not be differentiated from PAP tests with the available information. To estimate the proportion of non-PSA tests in the dataset, we investigated the patterns of use of the other tests, using data provided by the Commonwealth Department of Health and Family Services. We calculated retrospective projections using exponential regression on the number of tests by age group and annual quarter to give an adjusted number of PSA and PAP tests for the period between August 1989 and November 1993. This study included data for all males who had at least one PSA or PAP test between 1989 and 1996 reimbursed by Medicare. The data included a unique identification number for each person, age, date of service, postcode, and fee charged for the service. The identification number allowed individuals to be linked over time to quantify those who had multiple tests, but was not linkable to any other identifying data. Prostate cancer: Data on prostate cancer incidence in NSW were obtained from the NSW Central Cancer Registry, a population-based register which began collecting data in 1972. Notification of malignant neoplasms has been a statutory requirement for all NSW public and private hospitals, radiation oncology departments and nursing homes since 1972, and for all pathology and outpatient departments since 1985.1 Population: The Australian Bureau of Statistics provided the estimated resident populations of Australia by five-year age group and State or Territory for the years 1989 to 1996.13 Analysis We calculated the rate of testing and the number of males tested by age group, State or Territory, and year from August 1989 to December 1996. We calculated the number of males having multiple tests from January 1995 to December 1996 and the proportion of males tested in each age group in this period. We used a two-year period for these calculations to reduce errors introduced by men moving up age groups. We compared the number of tests carried out in NSW by quarter with the number of prostate cancers diagnosed between 1990 and 1995 (the most recent year for which prostate cancer data were available). All rates, standardised to the total male and female Australian 1991 population, are expressed per 100 000 males. We used SAS software for statistical analysis.14 Results PSA testing in Australia From August 1989 to December 1996 more than 2.2 million PSA or PAP tests were reimbursed by Medicare in Australia. About 1.1 million males were tested during this period. Eighty-eight per cent of the tests were for men aged 50 years or over, with the largest proportion (34%) for men aged 60-69 years (Table 1). Age-standardised rates for males having one or more PSA/PAP tests per year increased fivefold between 1990 and 1996, and the greatest increases occurred between 1993 and 1995. There was substantial variation in the rates of testing by State and Territory (Figure 1). Except in the Australian Capital Territory, the rates peaked in 1995, when the highest rates were in Western Australia (8668 tests per 100 000) and the Australian Capital Territory (8284 tests per 100 000) and the lowest in the Northern Territory (3270 per 100 000). From January 1995 to December 1996, 709 523 Australian males had at least one PSA/PAP test reimbursed by Medicare. Most of those tested (73%) had one test, 17% had two tests, 5% had three tests and 5% had four or more tests. Older men were more likely to have had multiple tests (Table 2). In this period, 27% of Australian men aged 50 years and over had one or more PSA/PAP tests reimbursed by Medicare. This figure peaked at 33% in men aged 60-69 years (Table 3). In 1996 the Medicare schedule fee for a single PSA/PAP test was $19.90 or, where two or more tests were undertaken at the same consultation, $36.65. The overall amount reimbursed by Medicare for PSA/PAP tests in 1996 was $10 675 880 (mean per test, $20.73; mode, $19.90; range, $11.00 to $36.65), not including the fee that may have been charged for the accompanying consultation. PSA/PAP tests and prostate cancer in NSW Between 1990 and 1995, more than 625 000 PSA/PAP tests were carried out in NSW, and 20 120 prostate cancers were diagnosed. The number of tests was highly correlated with the number of prostate cancers diagnosed (R2 = 0.92) (Figure 2). The ratio of number of tests performed to number of new cases of prostate cancer diagnosed increased from about 19 in 1990 to 45 in 1995. In 1995, in NSW, 150 479 males had one or more PSA/PAP tests. Figure 3 shows the age-specific rates of testing and reported incidence of prostate cancer. The number of PSA/PAP tests per prostate cancer detected in 1995 varied between age groups from a high of 579 in men aged 40-49 years to fewer than 24 in men aged 70 years or over. Discussion More than 2.2 million PSA tests were carried out in Australia from 1989 to 1996. More than 1.1 million males were tested in this period, and the annual number of males tested peaked in 1995. Data from this study support the hypothesis that the rising incidence of prostate cancer is associated with increased PSA testing. In NSW, the number of PSA tests was highly correlated with the number of new cases of prostate cancer. The PSA test is a blood test used in diagnosis and monitoring of prostate disease. First used in Australia in the late 1980s to monitor clinically identified disease, it has since been used in the diagnosis of relevant symptoms and as a screening test for asymptomatic men. It was not possible to identify from Medicare data the reasons why the tests were ordered. However, recent research found that 67 of 118 PSA tests (57%) were ordered for screening.15 Although we adjusted for other tests included under the same Medicare Benefits Schedule item from 1989 to 1993, we could not adjust for PAP tests, which are used to monitor the clinical progress of prostate cancer. However, in a continuous six-month period the ratio of PAP tests to PSA tests processed by a large, representative private pathology laboratory in NSW (covering city, suburban and regional centres) was less than 2% (Dr G Caldwell, Pathologist, Douglass Hanly Moir Pathology, personal communication). Data from a large public pathology laboratory in South Australia indicate that the proportion of PAP tests to the total PSA and PAP tests fell from 50% in 1991 to 6% in 1996 (Dr H A Morris, Manager, Endocrine Unit, Institute of Medical and Veterinary Science, personal communication). At their peak in 1995, the rates of PSA/PAP testing in Australian males ranged from 3270 per 100 000 in the Northern Territory to 8668 per 100 000 in Western Australia. These are probably underestimates because Medicare data do not include services provided free to public patients in public hospitals, to Veterans' Affairs patients and to men offered screening under the research activities of centres such as the Perth-based Urological Research Centre. In the one-year period April 1993 to March 1994, 39 626 PSA tests were done on 30 739 veterans.16 Data from the Department of Veterans' Affairs show about 50 000 tests were done each year in Australia between 1994 and 1996, which would have contributed a further 10% to the number of Medicare-reimbursed PSA tests. In South Australia 72 000 PSA tests that would not appear in Medicare statistics were undertaken between 1990 and 1996 by a public laboratory (Dr H A Morris, personal communication). These and the tests done on veterans would have accounted for an approximate under-enumeration of 33% annually in South Australia. Nationally, considering all these extra sources of PSA tests, we estimate that Medicare data underenumerate PSA tests by 14%. In a recent study of self-reported rates of prostate cancer screening in the Central Sydney Area Health Service, about one in five men aged 50 years or over reported being screened in the previous 12 months.17 This agrees quite closely with our results, which show that during the two years 1995 to 1996 about one in four Australian men aged 50 years or over had a PSA test, and in 1995 one in six (17%) had a test. More prostate cancers would result in more PSA tests used for monitoring. However, the overall effect of this on PSA test-ordering is thought to be small. More than 70% of males tested in 1995 and 1996 had only one test, suggesting that most tests were for screening rather than monitoring disease activity. A further possible indication that most tests were undertaken for screening rather than for monitoring or diagnosis is the increase in the ratio of tests to newly diagnosed prostate cancers in New South Wales. This ratio continued to increase in 1995, when the reported incidence of prostate cancer had begun to fall. Increasing recorded incidences of prostate cancer have been reported from the United States,18-22 France23 and elsewhere in Australia.1,2,4-6 Incidence figures began rising earlier in the United States than in Australia, and appeared to peak in 1992 and 1993.20 South Australian and Western Australian age-standardised recorded incidences peaked in 1994 and fell by 22% and 13%, respectively, between 1995 and 1996.2,4,5 The rate of PSA testing peaked in 1995 in both States and fell 10% and 16%, respectively, in 1996. These data and the high correlation between PSA tests and newly diagnosed prostate cancers in NSW support the hypothesis that the rising incidence figures for prostate cancer in the early 1990s were a direct result of screening.7 Based on these trends and correlations, it is likely that the recorded incidence for prostate cancer in most Australian States and Territories will continue to fall after 1995. Given that screening for prostate cancer has never been recommended in Australia, the rates of de facto screening in men aged over 50 years, and especially those aged between 60 and 69 years, are quite remarkable. These findings have important implications for public health policy and for patient and practitioner education aimed at reducing prostate cancer screening. Acknowledgements We would like to thank Mr Ross Saunders, Director of the Medicare Statistics Section of the Department of Health and Family Services, for supplying the data. References Coates M, Armstrong B. Cancer in New South Wales. Incidence and mortality 1994. Sydney: NSW Cancer Council, 1997. South Australian Cancer Registry. Epidemiology of cancer in South Australia. Incidence, mortality and survival 1977 to 1996, incidence and mortality 1996 analysed by type and geographical location. Twenty years of data. Adelaide: South Australian Health Commission, 1997. Australian Bureau of Statistics. Causes of death, Australia, 1995. Canberra: AGPS, 1996. (Catalogue No. 3303.0.) Threlfall T, Whitford M, Thompson J. Cancer incidence and mortality in Western Australia 1992 to 1994. A report of the Western Australian Cancer Registry. Perth: Health Department of Western Australia, 1996. Threlfall T, Thompson J. Cancer incidence and mortality in Western Australia, 1995. A report of the Western Australian Cancer Registry. Perth: Health Department of Western Australia, 1997. Shugg D, Dwyer T, Blizzard L. Cancer in Tasmania. Incidence and mortality 1994. Hobart: Menzies Centre for Population Health Research, 1997. McCredie M, Coates M, Churches T, Rogers J. Rising incidence of prostate cancer in Australia: a result of 'screening'? J Epidemiol Biostatistics 1996; 1: 99-105. Lange PH. Is screening for prostate cancer the current gold standard? -- "Yes". Eur J Cancer 1997; 33: 354-356. Kaye KW. Prostate cancer: enthusiasm for screening. Med J Aust 1995; 162: 540-541. Kramer BS, Gohagan JK, Prorok PC. Is screening for prostate cancer the current gold standard? -- "No". Eur J Cancer 1997; 33: 348-353. Hirst GHL, Ward JE, Del Mar CB. Screening for prostate cancer: the case against. Med J Aust 1996; 164: 285-288. Australian Health Technology Advisory Committee. Prostate cancer screening. Canberra: AGPS, 1996. Australian Bureau of Statistics. Estimated resident population by sex and age: States and Territories of Australia 1996. Canberra: Australian Bureau of Statistics, 1997. (Catalogue no. 3201.0.) SAS [computer program]. Version 6.12. Cary, North Carolina: SAS Institute, 1996. Ward JE, Gupta L, Taylor NJ. Do general practitioners use prostate-specific antigen as a screening test for early prostate cancer? Med J Aust 1998: 169; 29-31. Parkes AJ, Killer GT. Prostate-specific antigen -- is it already being used as a screening test? [letter]. Med J Aust 1994; 161: 722-723. Ward JE, Hughes A-M, Hirst GHL, Winchester L. Men's estimates of prostate cancer risk and self-reported rates of screening. Med J Aust 1997; 167: 250-253. Stephenson RA, Smart CR, Mineau GP, et al. The fall in incidence of prostate carcinoma. On the down side of a prostate specific antigen induced peak in incidence -- data from the Utah Cancer Registry. Cancer 1996; 77: 1342-1348. Jacobsen SJ, Katusic SK, Bergstralh EJ, et al. Incidence of prostate cancer diagnosis in the eras before and after serum prostate-specific antigen testing. JAMA 1995; 274: 1445-1449. Merrill RM, Potosky AL, Feuer EJ. Changing trends in U. S. prostate cancer incidence rates. J Natl Cancer Inst 1996; 88: 1683-1685. Gann PH. Interpreting recent trends in prostate cancer incidence and mortality. Epidemiology 1997; 8: 117-120. Lu-Yao GL, Greenberg R. Changes in prostate cancer incidence and treatment in USA. Lancet 1994; 343: 251-254. Menegoz F, Colonna M, Exbrayat C, et al. A recent increase in the incidence of prostatic carcinoma in a French population: role of ultrasonography and prostatic specific antigen. Eur J Cancer 1995; 31A: 55-58. (Received 28 Nov 1997, accepted 24 Apr 1998) Authors' details Cancer Control Information Centre, New South Wales Cancer Council, Sydney, NSW. David P Smith, BA, MPH, Research Coordinator, Cancer Epidemiology Research Unit; Bruce K Armstrong, DPhil, FRACP, Director. Reprints will not be available from the authors. Correspondence: Mr D P Smith, Cancer Epidemiology Research Unit, NSW Cancer Council, PO Box 572, Woolloomooloo, NSW 2011. E-mail: dsmithATnswcc.org.au Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/>

David P Smith · Bruce K Armstrong

Hyperparathyroidism: what does surgery have to offer? Even asymptomatic patients should be considered for surgery MJA 1998; 168: 148-149 The report by Delbridge et al in this issue of the Journal1 describes the changing clinical picture of primary hyperparathyroidism (PHPT) in patients referred for surgical treatment at Royal North Shore Hospital, Sydney, over the past 30 years. The most common presentation in the 1990s is in women over the age of 50 years with bone loss indistinguishable from that of postmenopausal or senile osteoporosis (31%). Patients with kidney stones still constitute 15%, and those with subtle but distressing symptoms of fatigue, depression or altered cognitive function constitute 20%. Delbridge et al have also observed an "exponential" increase in the number of cases referred for surgery over the past decade, reflecting contemporary epidemiological studies which have unearthed a virtual epidemic of PHPT among postmenopausal women -- it affects 2% to 3% of women in this age group.2,3 Even so, the absolute number of patients undergoing operation at Royal North Shore Hospital each year remains small when compared with the number expected by extrapolation from this prevalence. Presumably, this is due in part to underdiagnosis of PHPT, as well as continuing reluctance among physicians to refer patients for surgery unless they have marked hypercalcaemia or overt "bone or stone" disease. Modern surgical treatment of PHPT is very safe, with a hospital stay of two days or less. As documented by Delbridge et al, a primary neck exploration for PHPT by experienced surgeons is successful in 95% to 98% of cases. Most authorities would agree with these authors that preoperative tumour-localising studies are not cost effective because of the significant number of both false positive and false negative results.4 The limitations of these tests are not universally appreciated by physicians, many of whom still consider a positive tumour-localising study to be a prerequisite for surgical referral. In an attempt to discourage this, John Doppman (Chief, Department of Diagnostic Radiology, United States National Institutes of Health [NIH]) stated at the NIH Consensus Development Conference on PHPT that "the only localisation study needed by a patient undergoing initial parathyroid surgery is to locate an experienced parathyroid surgeon".5 A parathyroid neck exploration is often so easy that the uninitiated surgeon may wonder what all the fuss is about. However, the operation has many potential pitfalls which can lead to failure, with the result that the patient then has to undergo further irksome and costly investigations and a potentially hazardous reoperation. There is thus a high premium on succeeding at the first try. In this regard there is no substitute for experience on the part of the surgeon, who must be able to recognise a parathyroid gland, know the distribution of the glands, where they can be hidden, and how to distinguish between normal and abnormal glands with the naked eye. To acquire this expertise requires a dedicated training in a specialised centre by an experienced preceptor.4 Which patients with PHPT should have surgery? The NIH Consensus Development Conference agreed on criteria for surgery in patients with PHPT.5 These include any one of the following: Serum calcium above 2.99 mmol/L; Marked hypercalciuria (> 9.98 mmol/day); Any overt manifestation of PHPT (nephrolithiasis, osteitis fibrosa cystica, or classic neuromuscular disease); Markedly reduced cortical bone density; Reduced creatinine clearance in the absence of another cause; and Age less than 50 years. The recent demonstration by Silverberg et al6 that parathyroidectomy markedly improves cancellous (ie, lumbar spine) bone density in patients with PHPT has added vertebral osteopenia to the list of indications for surgery. There is also increasing evidence, including a recent prospective case-control study,7 attesting to the beneficial effect of parathyroidectomy on symptoms of hypercalcaemia, such as muscle weakness, fatigue, lethargy, depression, and memory loss. Patients with such symptoms now comprise 20% of those undergoing surgery for PHPT at Royal North Shore Hospital. As for patients who are truly asymptomatic with uncomplicated (or "biochemical") PHPT, the NIH Consensus Development Conference concluded that such patients should also be considered for surgery.5 The natural history of biochemical PHPT is unpredictable, and the costs of long term surveillance are not inconsiderable.8 Given the safety and efficacy of modern parathyroid surgery, it presents an attractive alternative to indefinite follow-up. The trend toward liberalising the indications for surgical treatment of PHPT has been given added impetus recently by several Scandinavian population-based studies indicating that patients with untreated PHPT have an increased risk of dying from cardiovascular disease and malignancy when compared with age- and sex-matched controls in the normal population.9 This increased risk of death can be reduced, if not eliminated, by parathyroid surgery.10 In this context, it is particularly relevant that the earlier the disease is recognised the more rapidly the mortality risk returns to normal after surgery. What is the future of parathyroid surgery? Delbridge et al have described the latest enthusiasm for endoscopic parathyroidectomy as "a passing interest", but this may be prematurely dismissive. Rightly or wrongly, endoscopic parathyroidectomy is already being done, and workshops teaching the technique are now conducted regularly at more than one European centre. In my opinion, it would be unfortunate if this procedure were to become the province of self-proclaimed "experts" in endosurgery who have little experience of the vagaries and nuances of parathyroid anatomy and pathology -- they could have difficulty in exploring the neck if the parathyroid tumour is not found endoscopically and conversion to open operation is required. Theoretically, endoscopic parathyroid surgery may offer advantages to patients through less postoperative pain, shorter hospital stay and smaller, more aesthetically pleasing scars. It is therefore appropriate to evaluate critically the efficacy, safety and cost of this procedure relative to those of a competently performed open parathyroidectomy, by way of a prospective controlled trial.11 Anthony J Edis Surgeon, Mount Hospital, Perth, WA Delbridge L, Younes N, Guinea A, et al. Surgery for primary hyperparathyroidism 1962-1996: indications and outcomes. Med J Aust 1998; 168: 153-156. Palmer M, Jakobson S, Akerstrom G, et al. Prevalence of hypercalcemia in a health survey: a 14-year follow-up study of serum calcium values. Eur J Clin Invest 1988; 18: 39-46. Lundgren E, Ridefelt P, Akerstrom G, et al. Parathyroid tissue in normocalcemic and hypercalcemic primary hyperparathyroidism recruited by health screening. World J Surg 1996; 20: 727-735. Edis A. Primary hyperparathyroidism in 1992: questions and answers. RACS Bull Nov 1992: 2-4. NIH Consensus Development Conference Panel. Diagnosis and management of asymptomatic primary hyperparathyroidism. Consensus development statement. Ann Int Med 1991; 114: 593-597. Silverberg S, Locker F, Bilezikian J. Vertebral osteopenia: a new indication for surgery in primary hyperparathyroidism. J Clin Endocrinol Metab 1996; 81: 4007-4012. Chan A, Duh Q, Catz M, et al. Clinical manifestations of primary hyperparathyroidism before and after parathyroidectomy: a case-control study. Ann Surg 1995; 222: 402-414. Heath H III, Hodgson S, Kennedy M. Primary hyperparathyroidism, incidence, morbidity and potential impact in a community. N Engl J Med 1980; 302: 189-193. Palmer M, Adami H-O, Bergstrom R, et al. Survival and renal function in persons with untreated hypercalcemia: a population-based cohort study with 14 years of follow-up. Lancet 1987; 1: 59-62. Hedback G, Oden A, Tisell L-E. The influence of surgery on the risk of death in patients with primary hyperparathyroidism. World J Surg 1991; 15: 399-407. Edis A, Sheiner H. Laparoscopic surgery: an ethical dilemma. Aust N Z J Surg 1996; 66: 201. Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/>

Anthony J Edis

Surgery for primary hyperparathyroidism 1962-1996: indications and outcomes Leigh W Delbridge, Nidal A Younes, Ana I Guinea, Thomas S Reeve, Phillip Clifton-Bligh and Bruce G Robinson MJA 1998; 168: 153-156 For editorial comment see Edis Abstract - Introduction - Methods - Results - Discussion - Acknowledgements - References - Authors' details - - ©MJA1998 Abstract Objective: To examine changes over the past three decades in the indications for, and outcomes of, surgery for primary hyperparathyroidism. Design: Survey of a prospective hospital database. Setting: Royal North Shore Hospital (a tertiary referral and university teaching hospital), Sydney, New South Wales, January 1962 to December 1996. Patients: All 733 patients who underwent neck exploration for primary hyperparathyroidism. Results: The annual number of parathyroidectomies increased virtually exponentially, from a mean of two in 1962-1969 to 73 in 1996. In the 1960s and 1970s, the most common indication for surgery was the presence of renal calculi (58% and 43%, respectively), but in the 1980s there was a marked increase in presentation of asymptomatic disease after biochemical screening (19%). In the 1990s, low bone mineral density detected by osteodensitometry has become the most common indication for surgery (31%). After initial operation, 11 patients (2%) had persistent hypercalcaemia, with five of these cured by reoperation -- an overall failure rate of 1%. Conclusions: Surgery for primary hyperparathyroidism has become increasingly common, with low bone mineral density replacing renal calculi as the most common indication for surgery. Neck exploration in experienced hands results in an overall cure rate of 99%. Introduction The past few decades have seen dramatic changes in the apparent incidence, presentation and management of primary hyperparathyroidism. With advances such as ready access to serum calcium and parathyroid hormone measurements, the condition is being increasingly recognised and treated, and new patterns of presentation are being seen. The classic bone disease osteitis fibrosa cystica, the principal manifestation early this century, is now rare.1 A clinical picture has emerged characterised either by an absence of symptoms or by subtle and vague symptoms,2-4 such as fatigue, weakness, and variable aches and pains, with the condition often suspected only because of an incidental finding of an elevated serum calcium level on biochemical testing. The recent introduction of bone mineral density screening by osteodensitometry seems also to have increased detection of cases of primary hyperparathyroidism.5 Our aim was to examine the changes over the past three decades in presentation and management of primary hyperparathyroidism at a single large referral centre, and to analyse the indications for, and outcomes of, surgery for this condition. Methods Subjects were all patients who underwent surgery for primary hyperparathyroidism in the Endocrine Surgical Unit at Royal North Shore Hospital, Sydney, New South Wales, from January 1962 (date of the first parathyroidectomy at the unit) until December 1996. Information was obtained from the prospective database of all endocrine surgical procedures maintained at that hospital since January 1957. The database was searched for any follow-up to December 1997. The diagnosis of primary hyperparathyroidism was based on the finding of an elevated serum calcium concentration and, when available, an inappropriately normal or elevated parathyroid hormone concentration. Before 1972, parathyroid hormone assays were not available, and the diagnosis was based on a combination of biochemical, radiological and clinical changes (eg, renal stones, urinary tract infection, abdominal pain and neuropsychological disturbances). Patients with secondary or tertiary hyperparathyroidism were excluded from the study. Information was obtained on the presentation, indications for surgery, operative details, postoperative complications, histopathological results and surgical outcomes. Persistent hyperparathyroidism was defined as hyper calcaemia continuing after surgery. Recurrent hyperparathyroidism was defined as hypercalcaemia returning after a minimum of six months of postoperative normocalcaemia. Results Between January 1962 and December 1996, 733 patients underwent neck exploration for primary hyperparathyroidism at Royal North Shore Hospital. They comprised 161 males (22%) and 572 females (78%), with an age range of 9-96 years (median age, 53 years for males and 60 years for females). Annual numbers of operations for primary hyperparathyroidism are shown in Figure 1. The number has risen virtually exponentially, from one to four annually in the 1960s, to 73 in 1996. Indications for surgery: These are shown in Box 1. The presence of renal calculi was the principal indication for surgery between 1962 and 1969, accounting for 58% of operations, but has progressively decreased in importance, accounting for only 15% of operations between 1990 and 1996. Numbers of asymptomatic patients (in whom none of the recognised symptoms of primary hyperparathyroidism could be identified preoperatively) increased in the 1970s and 1980s, but remained low overall (105 patients, 14%). Between 1990 and 1996, the most marked change was the increase in patients with low bone mineral density detected on screening for osteoporosis (by osteodensitometry or quantitative computed tomography) as the principal indication for surgery. In the 1990s, it was the most common indication (31% of cases), followed by neuromuscular or neuropsychiatric symptoms (20%) and renal calculi (15%), with only 14% of cases considered truly asymptomatic. Surgical and pathological findings: At surgery, single-gland disease (presence of only one enlarged gland) was found in 556 patients (76%), two-gland disease in 40 (5%), three-gland disease in two (0.3%), four-or-more-gland disease in 121 (17%), carcinoma in two (0.3%), cyst in seven (1%), and no parathyroid abnormality in five (0.7%). Complications of surgery: Postoperative hypocalcaemia requiring calcium supplementation was seen in 81 patients (11%), but only two of these had permanent hypoparathyroidism. Eight patients developed a wound infection (1%), eight required reoperation for haemorrhage (1%), six had a permanent vocal cord palsy (1%), and one required a temporary tracheostomy because of intraoral haemorrhage caused by injury to the tongue. One patient with pre-existing ischaemic heart disease died in the immediate postoperative period. Outcomes of surgery: Hypercalcaemia was cured immediately in 716 of the 733 patients. In six of the remaining 17, initial neck exploration showed no abnormalities, and they were subsequently shown to have conditions other than hyperparathyroidism, for which surgery was not indicated: sarcoidosis (three), familial hypocalciuric hypercalcaemia (two), and persistent hypercalcaemia with no apparent cause (one). Eleven patients had persistent hypercalcaemia caused by primary hyperparathyroidism after initial surgery, giving an initial cure rate of 98%. Details of initial and subsequent surgery for these 11 are shown in Box 2. Five were cured by re-exploration of the neck, two remained hypercalcaemic after a further unsuccessful re-exploration, and four had not undergone further surgery (two refused and two were awaiting further assessment). Thus, the overall cure rate for surgery was 99%. Hypercalcaemia was known to have recurred in two patients, at a mean of 10.5 years after successful surgery. Recurrence was associated with multiple endocrine neoplasia Type 1 syndrome (MEN1) in one of these patients and was cured by removal of the remnant and autotransplantation. The second patient had mild asymptomatic hyperparathyroidism for which reoperation was not indicated. Discussion We found that, as expected, surgery for primary hyperparathyroidism became increasingly common at Royal North Shore Hospital over the past three decades. In the 1990s, low bone mineral density replaced renal calculi as the most common indication for surgery. Initial neck exploration resulted in cure in 98% of cases. Primary hyperparathyroidism occurs relatively frequently in the community, with an incidence of at least 1 in 1000 individuals,6 and may be as frequent as 1 in every 500 women over the age of 50 years.7 The very small numbers of patients being diagnosed and treated as recently as two decades ago related to lack of ready access to serum calcium and parathyroid hormone measurements, as well as lack of awareness of the disease. Surgery is indicated in patients with symptoms or a high serum calcium level, and in asymptomatic patients who are not suitable for conservative management.8 The commonest indication for surgery now is the presence of low bone mineral density (36% of operations in 1996). Recovery of bone mass has been documented after successful parathyroid surgery in many series, even in patients with mild or asymptomatic hyperparathyroidism.5 The presence of renal calculi remains a major indication for surgery. Although the percentage of patients undergoing neck exploration for this indication has declined significantly each decade, the actual numbers have, in fact, steadily increased. Renal calculus formation is reduced after successful parathyroid surgery,9 although preformed stones or those associated with idiopathic hypercalciuria may continue to be passed.10 Marked muscular atrophy is rarely seen nowadays, but muscular weakness contributing to a general feeling of tiredness and malaise is noticed in most patients with primary hyperparathyroidism.11 We found neuromuscular disease was the primary indication in 11% of our patients. Abnormalities included muscular atrophy, generalised weakness and fatigue, which are thought to be related to low plasma phosphate level and possibly hypokalaemia. Many patients with these symptoms have reported improvement after parathyroidectomy.12,13 Neuropsychiatric symptoms were the primary indication for surgery in only 9% of patients in this series, but may be found (if sought) in significant numbers of patients with primary hyperparathyroidism (reported incidence, 30%- 100%).14 Symptoms include depression, anxiety, fatigue, lassitude, concentration difficulties, and failing memory. They have been reported to improve or disappear in most patients after surgery.12,15 A recent study showed that the most dramatic changes are reductions in body pain and improvements in vitality and emotional function.16 Similarly, although abdominal symptoms (which may be related to peptic ulcer disease, pancreatitis or constipation) were the primary indication for surgery in only 5% of our patients, they may be seen in up to 20% of patients with primary hyperparathyroidism.17 In our series, 14% of patients appeared asymptomatic. The proportion of patients with primary hyperparathyroidism reported to be asymptomatic varies greatly, from 2% to 80%,3,17-19 possibly depending on the care with which they are evaluated. Indeed, vague psychiatric and neuromuscular symptoms and generalised weakness may be fully appreciated only in retrospect, once normocalcaemia has been achieved by surgery.20 It is important to consider surgery even in the asymptomatic, as there is increasing evidence that primary hyperparathyroidism affects longevity. Several studies have shown that untreated individuals with mild hypercalcaemia have a reduced survival rate.21 A study of 441 patients followed up for a mean of eight years showed that successful parathyroid surgery reduced the risk of dying,21 while a more recent study of 896 patients confirmed this result and showed that the duration of hyperparathyroidism is also a factor, with early surgery reducing the risk of dying.22 A National Institutes of Health consensus statement from 1990 addressing the management of asymptomatic primary hyperparathyroidism recommends that "all patients with primary hyperparathyroidism should be considered to be candidates for surgery".8 The aim of surgery in primary hyperparathyroidism is to identify and remove all abnormal parathyroid tissue. As multiple-gland disease is common (22% of patients in this series), the mainstay of good surgical technique is to identify all (four or more) parathyroid glands in order to differentiate normal from abnormal glands (see Figure 2). We believe that current passing interest in "minimal access" parathyroid surgery (endoscopic or unilateral minimal incisions based on preoperative localisation) is misguided. Such techniques should be avoided as they will inevitably increase failure rates from unsuspected multiple-gland disease for, at best, a very marginal cosmetic advantage. Localisation techniques such as ultrasonography, computed tomography and scintigraphy with sestamibi have not shown sufficient sensitivity and specificity to justify routine use before initial operation and are certainly not cost-effective.23,24 False positive and false negative results from preoperative localisation tests may add confusion, especially for the inexperienced surgeon. Indeed, surgery undertaken by those not experienced in the procedure has been shown to be associated with a high failure rate and need for reoperation, as well as increased complications.25 For example, a Scandinavian study showed that surgery performed in units doing fewer than 10 parathyroidectomies per year resulted in only 70% of patients achieving long-term normocalcaemia,25 whereas in experienced units a success rate of 98% should be achieved. Acknowledgements We wish to acknowledge the following additional physicians and endocrinologists who have contributed at least several patients each to this study: Dr J Beattie, Dr D Darnell, Dr T Diamond, Dr G Fulcher, Dr S Grant, Dr I Hales, Dr A Jameson, Dr A Joasoo, Dr F Lomas, Assoc Prof JD Wilson, Dr A McElduff, Dr J Miller, Professor S Posen, Dr M Prowse, Dr P Rohl, Dr M Rosman, Dr J Stiel, Dr R Slobodniuk, Dr C White and Dr E Wilmshurst. We also thank the many other physicians and endocrinologists who have each contributed one or two patients. References Welbourn RB. The history of endocrine surgery. New York: Praeger, 1990. Heath H. Clinical spectrum of primary hyperparathyroidism: Evolution with changes in medical practice and technology. J Bone Miner Res 1991; 6: S63-S70. Heath H, Hodgson SE, Kennedy MA. Primary hyperparathyroidism: incidence, morbidity and potential economic impact in a community. N Engl J Med 1980; 302: 189-193. Chan AK, Duh Q-Y, Katz MH, et al. Clinical manifestations of primary hyperparathyroidism before and after parathyroidectomy. Ann Surg 1995; 222: 402-414. Warner J, Clifton-Bligh P, MacElduff A, et al. Longitudinal changes in forearm bone mineral content in primary hyperparathyroidism. J Bone Miner Res 1991; 6 Suppl 2: 91-95. Christenson T, Hellstrom K, Wengle R, et al. Prevalence of hypercalcemia in a health screening in Stockholm. Acta Med Scan 1976; 200: 131-137. Boonstra CE, Jackson JE. Serum calcium survey for hyperparathyoidism: results in 5000 clinical patients. Am J Clin Pathol 1971; 55: 523-526. Consensus Development Conference Panel. Diagnosis and management of asymptomatic primary hyperparathyroidism: consensus development conference statement. Ann Int Med 1991; 114: 593-597. Deaconson TF, Wilson SD, Lemann J Jr. The effect of parathyroidectomy on the recurrence of nepherolithiasis. Surgery 1987; 102: 910-913. Posen S, Clifton-Bligh P, Reeve TS, et al. Is parathyroidectomy of benefit in primary hyperparathyroidism? QJM 1985; 54: 241-251. Turken SA, Cafferty M, Silverberg SJ, et al. Neuromuscular involvement in mild asymptomatic primary hyperparathyroidism. Am J Med 1989; 87: 553-557. Delbridge LW, Marshman D, Reeve TS. Neuromuscular symptoms in elderly patients with hyperparathyroidism: improvement with parathyroid surgery. Med J Aust 1988; 149: 74-76. Kristoffersson A, Bostrom A, Soderberg T. Muscle strength is improved after parathyroidectomy in patients with hyperparathyroidism. Br J Surg 1992; 79: 165-168. Joborn C, Hetta J, Palmer M, et al. Psychiatric symptomatology in patients with primary hyperparathyroidism. Ups J Med Sci 1986; 91: 77-87. Joborn C, Hetta J, Lind L, et al. Self rated psychiatric symptoms in patients operated on because of primary hyperparathyroidism and in patients with longstanding mild hypercalcemia. Surgery 1989; 105: 72-78. Burney R, Jones K, Coon J, et al. Assessment of patient outcomes after operation for primary hyperparathyroidism. Surgery 1996; 120: 1013-1019. Kaplan EL, Yashiro T, Salti G. Primary hyperparathyroidism in the 90s. Ann Surg 1991; 215: 300-317. Ljunghall S, Hellman P, Rasted J, Akersorm G. Primary hyperparathyroidism: epidemiology, diagnosis, and clinical picture. World J Surg 1991; 15: 681-687. Van Heerden JA, Grant CS. Surgical treatment of primary hyperparathyroidism: an institutional perspective. World J Surg 1991; 15: 688-692. Harrison BJ, Wheeler MH. Asymptomatic primary hyperparathyroidism. World J Surg 1991; 15: 724-729. Palmer M, Adami H-O, Bergstrom R, et al. Mortality after operation for primary hyperparathyroidism. A follow-up of 441 patients operated on during 1956-1979. Surgery 1987; 102: 1-7. Hedback G, Oden A, Tisell L. The influence of surgery on the risk of death in patients with primary hyperparathyroidism. World J Surg 1991; 15: 399-407. Miller DC. Preoperative localisation and interventional treatment of parathyroid tumours: when and how. World J Surg 1992; 15: 706-715. Serpell JW, Cambell PR, Young AE. Pre-operative localisation of parathyroid tumours does not reduce operating time. Br J Surg 1991; 78: 589-590. Malmaaeus J, Granberg PO, Halvorsen J, et al. Parathyroid surgery in Scandinavia. Acta Chir Scand 1988; 154: 409-413. (Received 1 Jul, accepted 26 Nov, 1997) Authors' details Department of Surgery, University of Sydney, and Royal North Shore Hospital, Sydney, NSW. Leigh W Delbridge, MD, FRACS, Professor of Surgery; Nidal A Younes, MD, Fellow in Endocrine Surgery; currently, Surgeon, University of Jordan Hospital, Amman, Jordan; Ana I Guinea, BSc(Psych)(Hons), Psychologist; Thomas S Reeve, MD, FRACS, Emeritus Professor. Department of Endocrinology, University of Sydney, and Royal North Shore Hospital, Sydney, NSW. Phillip Clifton-Bligh, FRACP, Clinical Associate Professor in Medicine; Bruce G Robinson, MD, FRACP, Professor of Medicine (Endocrinology), University of Sydney, and Kolling Institute of Medical Research, Sydney, NSW. Reprints will not be available from the authors. Correspondence: Professor L W Delbridge, Department of Surgery, Royal North Shore Hospital, St Leonards, NSW 2065. E-mail: leighd AT med.su.oz.au - Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/>

Leigh W Delbridge · Nidal A Younes · Ana I Guinea · Thomas S Reeve · Phillip Clifton-Bligh · Bruce G Robinson

Hip fracture in elderly men: prognostic factors and outcomes Terrence H Diamond, Stephen W Thornley, Ronald Sekel and Peter Smerdely For editorial comment, see Seeman Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au/>". Abstract - Introduction - Methods - Statistical analysis - Results - Discussion - References - Authors' details - ©MJA1997 Abstract Objective: To examine prognostic factors and outcomes after hip fracture in men aged 60 years and older. Design and setting: Cohort study of all men presenting to St George Hospital (a 650-bed tertiary care centre) with hip fractures in 1995, recruited retrospectively from medical records and evaluated prospectively at six and 12 months after fracture. Patients: 51 men aged 60 years or more (and, for comparison, 51 age-matched women) who presented with hip fracture not caused by high impact injuries or local bone disease. Main outcome measures: Prognostic factors (such as pre-existing illness and osteoporotic risk factors) and outcome data (such as fracture-related complications, mortality, and level of function as measured by the Barthel index of activities of daily living at six and 12 months postfracture). Results: Median age of the 51 men was 80 years (interquartile range, 74-86 years); four were aged under 70 years. Outcome assessment was possible for 41 men (80%). Similar proportions of men and women came from institutions (32% v. 28%), and similar additional proportions required institutionalisation after discharge (18% v. 14%). Fracture-related complications affected similar proportions of men and women (30% v. 32%), and mean length of hospital stay was similar. Fourteen per cent of men died in hospital compared with only 6% of women (P = 0.06). Men had more risk factors for osteoporosis (P < 0.01). Physical functioning (measured by the Barthel index) deteriorated significantly in men from 14.9 at baseline to 13.4 at six months (P < 0.05) and 12.4 at 12 months (P < 0.05) after fracture. Conclusion: Compared with women, elderly men presenting with hip fracture have higher mortality and have more risk factors for osteoporosis. Like women with hip fracture, men are usually fragile, with pre-existing medical illness and fracture-related complications contributing to their overall poor outcomes. MJA 1997; 167: 412-415 Introduction The incidence of hip fracture in elderly men is approximately one-third of that reported in elderly women.1-5 In a recent Australian study, the incidence of hip fracture in men was calculated as 19.4 per 1000 population per year, with the highest incidence in those aged 80 years and older.3 The economic and social implications of hip fracture in the Australian community are enormous, with the overall cost approximating $420 million annually.6 Previous studies have shown an increased morbidity and mortality associated with hip fracture in elderly women, in whom outcomes are usually poor and partly related to age and medical conditions.7-12 Until recently, there have been comparatively few data on hip fracture in elderly men.12-19 Given the importance of hip fracture, we analysed the mortality and functional outcome of hip fracture in elderly men (aged 60 years and over) who presented to our hospital. Methods St George Hospital is a 650-bed tertiary care referral centre serving a population of 195 000 in the southern metropolitan area of Sydney. Patients admitted to this hospital with hip fractures are usually treated with internal fixation of the fractured hip within 24-48 hours of admission. We retrospectively audited the medical records of all men and women with hip fractures presenting to St George Hospital between 1 January and 31 December 1995. Men were eligible for the study if they were aged 60 years or over, and if their hip fracture was not the result of high impact injuries or local bone disease. The prognostic factors and outcomes following hip fracture of eligible men were compared with those of an equal number of age-matched women who presented with hip fractures during the same 12-month period. Ethical approval for this study was granted by the St George Hospital Ethics Committee. From the hospital medical records, we recorded: Patient's age; Prognostic factors, such as pre-existing illness, osteoporotic risk factors14,17-19 and type of fracture; and Outcome data, such as fracture-related complications, length of hospital stay, mortality, and level of function (as measured by the Barthel index of activities).20 These data are routinely recorded by the orthopaedic intern or registrar, appropriate consultative services, occupational therapist, and/or aged care and rehabilitative services. The Barthel index consists of a questionnaire containing 10 questions pertaining to activities of daily living, such as mobility, bathing, dressing, and toilet use, and is scored out of a total of 20 points -- the lower the score, the worse the disability. Any additional data that were required were obtained by telephone interviews with the patients and/or their family members. Subsequently, patients were followed up prospectively six and 12 months after hip fracture by telephone interview. We obtained data pertaining to whether they were living at home or in an institution, and then used the original 10 questions on the Barthel index questionnaire again to assess activities of daily living, self-care ability and mobility. Statistical analysis Results were analysed with StatCalc21 statistics package. Data for men and women were compared by Student's t test or analysis of variance, where applicable. The main predictors of death and institutionalisation were determined by stepwise regression analysis; the variables entered into the equation included the patient's age, smoking history, alcohol intake, pre-existing medical illness, prefracture Barthel score, length of hospital stay, and fracture-related complications. Results One hundred and eighty-nine people had presented with hip fracture during 1995, comprising 57 men (30%) and 132 women (70%). Six of the men were excluded because their fractures were related to high impact injuries or local bone disease. Hence, we compared the prognostic factors and outcomes following hip fracture of the remaining 51 men with those of 51 age-matched women of the 132 who presented with hip fractures during the same period. Forty-one men (80%) were contactable for assessment of outcomes at six and 12 months. We were unable to contact 10 men, either by a mailed questionnaire or through the telephone directory services. Their names had not been recorded in the death registry of the New South Wales Bureau of Births, Deaths and Marriages. They were considered lost to follow-up. Box 1 (below) compares the clinical data of the men and women with hip fracture. The median age of the 51 eligible men was 80 years; four were aged less than 70 years. Their mean length of hospital stay was 13 days (range, 3-55 days). Sixteen men (32%) came from hostels or nursing homes before admission. Thirty men (58%) were classified radiologically as having trochanteric fractures and 21 (42%) as having cervical fractures; they did not differ with respect to clinical presentation or postfracture outcomes (data not shown). Compared with women, men had a higher prevalence of excessive alcohol consumption (chi-squared = 13.95; P = 0.004) and current smoking (chi-squared = 14.96; P = 0.0004). Forty-eight men (95%) had at least one medical problem before admission; the mean number of medical problems per patient was three (range, 0-4). Forty-seven women (92%) had at least one medical problem, with a mean number of medical problems per patient of two (range, 0-4) (Box 2). Forty men (78%) had at least one risk factor for osteoporosis; the mean number of risk factors per patient was one (range, 0-3). Similarly, 36 women (72%) had at least one risk factor for osteoporosis (not including menopausal status), with a mean number per patient of one (range, 0-4) (Box 2). Ten men (20%) with hip fractures died: seven during hospital admission and another three during the first six months after fracture. More men than women with hip fractures died during their acute hospital admission, but this difference was not significant (P = 0.06). Fifteen men (30%) developed fracture-related complications, five of whom died in hospital. Those who had complications developed an average of two complications each (range, 0-4). This was similar for women; 16 (32%) developed fracture-related complications, three of whom died in hospital, and the average number of complications was one (range, 1-4). The occurrence of individual fracture-related complications did not differ significantly between men and women with hip fracture. Fracture-related complications were the single most important predictor of death in men (odds ratio [OR], 13.5; 95% CI, 1.74-132; P = 0.06). By contrast, the most important predictor in women was the prefracture Barthel index score. In men, age, smoking history, alcohol intake, pre-existing medical illness, prefracture Barthel index score, and length of hospital stay did not contribute significantly to the fracture-related mortality. The Barthel index score (mean, 14.9 at baseline) deteriorated significantly by six months (mean, 13.4; P < 0.05) and 12 months (mean, 12.4; P < 0.05). The baseline Barthel score did not differ significantly between men and women. After fracture, an additional nine men required hostel or nursing home accommodation. This correlated significantly with both the patient's age and the per cent decline in the Barthel index score (r = 0.41; P = 0.0002). Discussion Our findings support the limited published data on hip fracture in men.12-19 As in other studies,3-5 we found that men represented 30% of all hip fractures, and that men who sustained hip fractures were elderly, had pre-existing medical conditions5,13-16 and at least one risk factor for osteoporosis.14,18,19 Compared with the men in our study, women with hip fractures had fewer risk factors for osteoporosis (P < 0.01). Although our study has the major limitation of small sample size, it is the first Australian study to show a significant decline in physical functioning in men after hip fracture. A decline in physical functioning has been noted in both men and women after hip fracture, with many survivors requiring institutionalisation. Almost one-third (32%) of our men originated from institutionalised care, and an additional 18% were subsequently discharged to institutionalised care. Age and percent decline in Barthel index score were the most important criteria leading to their institutionalisation. While the use of the Barthel index may potentially identify individuals who will need long term institutionalisation, this is only a gross assessment of activities of daily living. Thirty per cent of the men in our study returned home to their previous level of function as measured by the Barthel index, but anecdotally many reported a decline in more subtle activities not measured by this index. In a study by Marotolli et al., 29% of all hip fracture patients were institutionalised at six months postfracture.13 Another study reported that 79% of the patients surviving at one year were residing in nursing homes or intermediate care facilities, while those who returned home had significant functional decline, with almost 60% limping or requiring a cane or walker.15 Reported rates of mortality and morbidity in men with hip fracture vary from 13%-44%,1,4,5,9,10,11,16,22 with the likelihood of a man dying after hip fracture increased by 83% and the likelihood of subsequent hospital admission after hip fracture increased by 231%.5 In our study, 20% of men with hip fractures died, either during the initial hospital admission or within the first six months after fracture. Although not statistically significant, we found that, compared with women, twice as many men with hip fractures died during the acute hospital admission (P = 0.06). These data are consistent with those of many other studies which have shown higher postfracture mortality rates in men.1,10,22 For example, Holt et al. recorded 17% mortality in men compared with 11.5% in women,22 while Jacobsen et al. reported mortality rates per 1000 person-months postfracture of 33.7 in white men compared with 17.2 in white women.10 In our study, death occurred predominantly within the first two months postfracture, and fracture-related complications was the strongest predictor of death. In a longitudinal study of ageing, Wolinsky and colleagues found a one-year postfracture mortality among 7527 members of approximately 24%, with the greatest risk of dying in the first six months after hip fracture (hazard-risk ratio, 57.4; 95% CI, 43.7-75.3); survival rates estimated at six months postfracture returned to a trend similar to that of control subjects.5 Despite the limitations of small sample size and a lack of control subjects who had not had hip fracture, this study shows men who present with hip fracture are usually elderly and fragile and have numerous risk factors for osteoporosis, pre-existing medical illnesses and fracture-related complications. In men, this results in higher postfracture mortality compared with age-matched women, as well as in significant functional decline. This study highlights the need to identify men with osteoporosis in the community, and the need to find effective strategies for preventing hip fracture. References Cummings SE, Kelsey JL, Nevitt MC, O'Dowd KJ. Epidemiology of osteoporosis and osteoporotic hip fractures. Epidemiol Rev 1985; 7: 178-208. Lord SR. Hip fractures: changing patterns in hospital bed use in NSW between 1979 and 1990. Aust N Z J Surg 1993; 63: 352-355. Jones G, Nguyen T, Sambrook PN, et al. Symptomatic fracture incidence in elderly men and women: the Dubbo Osteoporosis Epidemiology Study (DOES). Osteoporosis Int 1994; 4: 277-282. Cumming RG, Klineberg RJ. Case-control study of risk factors for hip fractures in the elderly. Am J Epidemiol 1994; 139: 493-503. Wolinsky FD, Fitzgerald JF, Stump TE. The effect of hip fracture on mortality, hospitalization and functional status: a prospective study. Am J Public Health 1997; 87: 398-403. Wark JD. Osteoporosis: the emerging epidemic. Med J Aust 1996; 164: 327-328. Aitken JM. Relevance of osteoporosis in women with fracture of the femoral neck. BMJ 1984; 288: 597-601. Pettiti DB, Sidney S. Hip fracture in women. Clin Orthop 1989; 246: 150-155. Myers AM, Robinson EG, Van Natta ML, et al. Hip fractures among the elderly: factors associated with in-hospital mortality. Am J Epidemiol 1991; 134: 1128-1137. Jacobsen SJ, Goldberg J, Miles TP, et al. Race and sex differences in mortality following fracture of the hip. Am J Public Health 1992; 82: 1147-1150. Cooper C, Atkinson EJ, Jacobsen SJ, et al. Population-based study of survival after osteoporotic fractures. Am J Epidemiol 1993; 137: 1001-1005. Marotolli RA, Berkman LF, Cooney LM. Decline in physical function following hip fracture. J Am Geriatr Soc 1992; 40: 861-866. Marotolli RA, Berkman LF, Leo-Summers L, Cooney LM. Predictors of mortality and institutionalisation after hip fracture: The New Haven EPESE Cohort. Am J Public Health 1994; 84: 1807-1812. Seeman E. Osteoporosis in men: epidemiology, pathophysiology and treatment possibilities. Am J Med 1993; 95 (Suppl 5A): 23S-28S. Poor G, Atkinson EJ, Lewallen DJ, et al. Age-related hip fractures in men: clinical spectrum and short-term outcomes. Osteoporosis Int 1995; 5: 419-426. Poor G, Atkinson EJ, O'Fallon WM, Melton LJ. Determinants of reduced survival following hip fractures in men. Clin Orthop 1995; 319: 260-265. Ringe JD. Hip fractures in men. Osteoporosis Int 1996; 6 (Suppl 3): 48-51. Looker AC, Mussolino ME, Madans JH, Orwoll ES. Risk factors for hip fractures in white men: The NHANES I Epidemiologic Follow up Study [abstract]. J Bone Miner Res 1996; 11: 233. Smerdely P, Thornley S, Sekel R, Diamond T. Subclinical vitamin D deficiency is the major biochemical risk factor associated with hip fracture in elderly men [abstract]. J Bone Miner Res 1996; 11: 233. Mahoney F, Barthel D. The Barthel index. Maryland State Med J 1965; 14: 61-65. StatCalc (Epi Info) [computer program]. Version 6.046. Geneva: WHO, 1997. Holt EM, Evans RA, Hindley CJ, Metcalfe JW. 1000 femoral neck fractures: the effect of pre-injury mobility and surgical experience on outcome. Injury 1994; 25: 91-95. (Received 25 Mar, accepted 10 Jul, 1997) Authors' details St George Hospital, Sydney, NSW. Terrence H Diamond, FRACP, Senior Endocrinologist, Department of Endocrinology; Stephen W Thornley, FRACP, Endocrine Registrar, Department of Endocrinology; Ronald Sekel, FRCS, Senior Orthopaedic Surgeon, Department of Orthopaedic Surgery; Peter Smerdely, PhD, FRACP, Endocrinologist, Department of Aged Care. Reprints: Dr T Diamond, Department of Endocrinology, St George Hospital, 32 Belgrave Street, Kogarah, NSW 2217. ©MJA 1997 <URL: http://www.mja.com.au/> © 1997 Medical Journal of Australia.

Terrence H Diamond · Stephen W Thornley · Ronald Sekel · Peter Smerdely

Use of inpatient hospital services by people aged 90-99 years Josephine H Harris, Paul M Finucane, Denise C Healy and Anthony C Bakarich Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au/>". Abstract - Introduction - Methods - Results - Discussion - Acknowledgements - References - Authors' details - ©MJA1997 Abstract Objective: To examine the use of inpatient hospital services by people aged 90-99 years. Design: Retrospective case note review. Setting: Flinders Medical Centre, a 516-bed university teaching hospital in Adelaide, South Australia. Patients: All patients aged 90-99 years on the separation register for 1995. Main outcome measures: Patient demographic characteristics, principal diagnosis, length of hospital stay and outcome, including destination at discharge. Results: In 1995, 317 separations involved 214 patients aged 90-99 years; 148 patients (69%) were admitted to hospital once, 43 (20%) twice and 23 (11%) three times or more. In 54% of separations, patients came from the community, and these were less likely to be emergency admissions (72%) than were admissions from hostels (87%) and nursing homes (93%). Patients had a wide range of acute medical and surgical problems and a median of five documented comorbidities. Patients survived to leave hospital in 290 separations (91%) and returned directly to their previous living circumstances in 212 (67%). Median hospital stay was 5.0 days, and in 25% of separations stay was one day or less. Patients admitted under the care of geriatricians had more emergency admissions (98%) and longer mean hospital stays (8.9 days) than those admitted under surgeons (69%; 5.9 days) or other physicians (66%; 5.0 days). Conclusion: Despite the acute nature of their illnesses and their multiple medical problems, most hospitalised nonagenarians in this study returned directly to their previous living circumstances after short hospital stays. MJA 1997; 167: 417-420 Introduction Australia's rapidly ageing population and changing patterns of health care delivery are combining to increase the number of old people using acute hospital services.1 The greatest proportional population increase is among the very old; between 1990 and 1995 the estimated number of people aged 85 years and over in Adelaide grew by 27.3%, while the general population grew by only 3%.2 The impact on hospital services is substantial. For example, inpatient separations for people aged 90 years and over at Flinders Medical Centre, Adelaide, increased by 66% between the 1989-90 and 1994-95 financial years, from 274 to 454 separations annually (unpublished data). Little is known about the characteristics of hospitalised nonagenarians. Existing studies either have few subjects3 or include only medical4 or only surgical patients.5-7 To our knowledge, no study has focused on hospitalised nonagenarians in Australia. Our lack of knowledge about this group may hinder development of appropriate clinical services to meet their needs and allow prejudices and unfounded negative stereotypes to proliferate. For example, it has been implied that some patients present to hospital with acute social rather than medical crises,8 and that elderly patients become "bed blockers" who are difficult to discharge from hospital.9 Elderly patients are satirised and derided in fiction,10 while their right to access expensive medical technology is debated in scientific publications.11 To learn more about the use of acute hospital services by very elderly people, we reviewed the case notes of all people aged 90 years and over who were admitted to our hospital in 1995. In particular, we focused on the demographic characteristics of this group, the problems for which they were hospitalised and the outcomes of hospitalisation. Methods Setting and subjects We examined retrospectively the case notes of all people aged 90-99 years on the separation register for Flinders Medical Centre during 1995. The Centre, with 516 beds, is the largest hospital in the southern metropolitan region of Adelaide and the principal teaching hospital of Flinders University. Relative to other Australian public hospitals of a similar size, a large proportion of its caseload is non-elective (Dr C Baggoley, Director of Emergency Department, Flinders Medical Centre, personal communication). The emphasis is on acute care; hospitalised patients needing rehabilitation are generally transferred to other public and private facilities in the region. Some acute surgical specialties (e.g., urology and vascular surgery) are largely provided at a sister institution. Flinders Medical Centre has an age-related admission policy whereby elderly patients with complex medical problems are admitted under the care of a geriatrician, while those with more specific problems are admitted under the appropriate specialist physician or surgeon. The study group was identified from a computerised age and sex register of all separations. This included day-only patients (length of stay, 0-1 days), but excluded those attending the emergency department who were subsequently not admitted. Data collection and analysis The principal diagnosis for each separ ation was obtained from the hospital's Australian national diagnosis-related group (AN-DRG) coding system and verified by review of all case notes. The diagnosis was further classified according to the major body system affected. Comorbidities and cognitive impairment were identified from the discharge letter. Data were analysed with the Statview statistical package.12 Other data were derived from the case notes. Differences between patient groups were assessed by chi-squared tests for all characteristics except length of hospital stay, which was assessed by the Kruskal-Wallis one-way analysis of variance by ranks, a non-parametric test. The association between patients' living circumstances before and after hospitalisation was determined by Cohen's k test.13 The study received ethical approval from Flinders Medical Centre's Committee on Clinical Invest igation. Results During 1995, 317 of the hospital's 27 833 inpatient separations (1.1%) involved 214 patients aged 90-99 years. Patient characteristics Of the 214 patients, 157 (73%) were women and 57 (27%) were men, with median age, 92 years. Age distribution was 57%, 90-93 years; 27%, 94-95; and 16%, over 95. Before initial hospitalisation, 111 subjects (52%) lived in the community, 58 (27%) in hostels and 45 (21%) in nursing homes. Most subjects (148; 69%) were admitted to hospital once during 1995, 43 (20%) twice and 23 (11%) three times or more. The maximum number of separations per patient in the year was six. Separation characteristics The major reason for each hospitalisation (identified as the principal diagnosis at separation) is shown in Box 1. Orthopaedic problems (especially fractured neck of femur) and cardiovascular disorders (especially myocardial ischaemic syndromes and congestive cardiac failure) were most common. Most separations (80%) had been classed as emergency admissions. Median number of comorbidities was five (range, 0-14), with fewer than four comorbidities recorded for 31% of separations, and eight or more for 14%. Cognitive impairment, either acute or chronic was documented for 94 separ ations (30%). Length of hospital stay is shown in Figure 1 (below). Median stay was 5.0 days (range, 1-74 days) and the mean was 6.9 days (SD, 8.3 days), compared with 3.7 days for the total inpatient population. Seventy-eight separations (25%) were day-only; in 10% of cases this was because of early death. Outcomes Destinations at separation are shown in Figure 2. Overall, patients returned directly to their previous living circumstances in 212 separations (67%), were transferred to other hospitals for continuing rehabilitation or "step down" care in 56 (18%) or to a more supportive residential environment in 22 (7%), and died in 27 (9%). The percentage who returned directly to their previous living circumstances rose to 91% when deaths and hospital transfers were excluded, and the trend for this return was statistically significant (Cohen's k = 0.85, P < 0.001). Hospital stay was shorter for patients who returned directly to their previous living circumstances (mean [SD], 5.2 [5.7] days) than for those who did not (mean [SD], 10.5 [11.2] days). Comparison of patients from the community and from residential care Characteristics of patients admitted from different living circumstances are compared in Box 2. Women predomin ated in all categories, but the proportion of women was higher among those admitted from residential care, especially nursing homes, than among those from the community. Patients from residential care were more likely to have emergency admissions and be under the care of surgeons than those from the community, but less likely to be under the care of physicians (other than a geriatrician). The proportions under the care of geriatricians were similar in each category. Hospital stay did not differ significantly between patients from different living circumstances (mean in days [SD]: community, 7.4 [10.0]; hostels, 6.9 [6.1]; and nursing homes, 5.6 [5.4]; Kruskal-Wallis test statistic, T = 3.46; P = 0.18). However, mortality rose progressively in patients from the community (6%), hostels (9%) and nursing homes (16%), in that order. Comparison of surgical and medical patients Patients admitted under the care of different specialists are compared in Box 3. Most patients were admitted under the care of surgeons (133, 42%), while 126 (40%) were admitted under the care of geriatricians and 58 (18%) under the care of other physicians. Patients admitted under the care of physicians other than geriatricians were most likely to have day-only separations, to be living in the community and to return directly to the community. In contrast, admissions under geriatricians were almost all emergencies and were least likely to be day-only. Hospital stay was significantly longer for these patients (mean in days [SD]: geriatricians, 8.9 [9.6]; surgeons, 5.9 [6.6]; other physicians 5.0 [8.2]; Kruskal-Wallis T = 28.98; P < 0.001). However, when day-only separations were excluded, the difference lost significance (mean in days: geriatricians, 10.0; surgeons, 7.9; other physicians, 8.3; Kruskal-Wallis T = 5.74; P = 0.06). Mortality was highest for patients admitted under geriatricians (12%) and lowest for those admitted under surgeons (5%), but this difference was not statistically significant. Discussion This study challenges some negative stereotypes about use of acute hospital services by very elderly people. We found that people aged 90-99 years accounted for 1.1% of all separations, with most (54%) coming from the community. Over 90% survived to leave hospital and most returned to their previous living circumstances after a median hospital stay of just under a week. They presented with a wide range of acute problems, predominantly orthopaedic and cardiovascular problems. Only a small proportion were frequent users of inpatient beds. The limitations of this study need to be recognised. Its retrospective nature and reliance on case notes mean that some information, particularly about comorbidities and presence of cognitive impairment, may be inaccurate and may underestimate their true extent. However, validity was enhanced by the use of defined objective measures, not subject to observer bias. The extent to which our results may be generalised is uncertain. Although Flinders Medical Centre is mostly typical of large university teaching hospitals, it has a higher proportion of non-elective cases. Differences in its clientele, range of services and service delivery are also possible. There are no data on nonagenarians admitted to other Australian hospitals for comparison. Nor can we readily compare our study results with those from other countries, as these have excluded particular patient groups, such as medical6,7,14 or surgical4 patients or those living in residential care.3 Nevertheless, others have found similarly that hospitalised nonagenarians present with a wide range of medical and surgical problems. The longer mean hospital stays in other studies may reflect different patient profiles or management practices. We found that more nonagenarians were admitted under the care of surgeons than under geriatricians or other physicians. Those admitted under surgeons had the lowest mortality, with 95% surviving to leave hospital, even though almost 70% were admitted as emergencies. It is recognised that surgical patients in general have lower mortality rates than medical patients,15 and our study suggests that this holds true for the very old. We did not determine the number who actually underwent a surgical procedure, so cannot estimate perioperative mortality. Others have estimated it to be 10%-30%,5-7,14 with a higher mortality rate for emergency compared with elective procedures.15 However, recent advances in anaesthetic and surgical techniques17 have probably improved survival prospects for very elderly surgical patients. Our finding that nonagenarians admitted under the care of geriatricians were most likely to be admitted as emergencies, to have documented cognitive impairment and to come from residential care, probably reflects the hospital's policy of assigning this type of specialist to very elderly patients with complex medical conditions. This may also explain why patients admitted under geriatricians were less likely to return directly to the community and had longer hospital stays. Alternatively, the longer hospital stays after admission under geriatricians can be explained by casemix factors as, for example, the difference was not statistically significant when day-only patients were excluded. It was notable that 25% of all separ ations were day only, suggesting that strategies to minimise hospital stay with day surgery and other day procedures are being applied successfully to the very old as well as to younger age groups. Further, hospital stay was twice as long in people who needed transfer to a rehabilitation facility or more supportive level of residential care. While it is widely recognised that inability to readily access rehabilitation and residential care facilities prolongs stay in acute hospitals, our study provides quantitative evidence for this. We believe that many people can be reassured by this study. Firstly, hospital administrators and health planners can be reassured that very elderly people seem to make appropriate use of acute hospital services. Secondly, health professionals can take a positive approach to treating acute illness in their elderly patients. Finally, and perhaps most importantly, very elderly people needing acute hospitalisation can be optimistic in the knowledge that most will survive and return home after a short hospital stay. Acknowledgements We gratefully acknowledge Dr Michael Clark (Department of Rehabilitation and Aged Care) for his assistance with statistical analyses. This survey was supported by a research grant from Flinders 2000, a research foundation based at Flinders Medical Centre. References Lipski P. Optimum care of the elderly in an acute general hospital. Med J Aust 1996; 164: 5-6. Australian Bureau of Statistics. Estimated resident population by age and sex in statistical local areas of South Australia. Canberra: ABS, 1991 and 1996. (Catalogue no. 3204.4). Patterson C, Crescenzi C, Steel K. Hospital use by the extremely elderly (nonagenarians): a two-year study. J Am Geriatr Soc 1984; 32: 350-352. Saint Jean O, Thibert JB, Holstein J, et al. Hospitalisation en medecine interne des nonagenaires. Etude de 150 sejours. Rev Med Interne 1993; 14: 825-831. Ackermann RJ, Vogel RL, Johnson LA, et al. Surgery in nonagenarians: morbidity, mortality and functional outcome. J Fam Pract 1995; 40: 129-135. Hosking MP, Warner MA, Lobdell CM, et al. Outcomes of surgery in patients 90 years of age and older. JAMA 1989; 261: 1909-1915. Cohen JR, Johnson H, Eaton S, et al. Surgical procedures in patients during the tenth decade of life. Surgery 1988; 104: 646-651. Hobbs R. Rising emergency admissions. BMJ 1995; 310: 207-209. Lewis H, Purdie G. The blocked bed: a prospective study. N Z Med J 1988; 101: 575-577. Shem S. The house of God. London: Bodley Head, 1978. Callahan D. Controlling the costs of health care for the elderly -- fair means and foul. N Engl J Med 1996; 335: 744-746. Apple Macintosh statview statistical package. Version 512+. Calabasas, Cal: Brain Power Inc, 1986. Cohen JA. A coefficient of agreement of nominal scales. Educ Psychol Meas 1960; 20: 37-46. Denney JL, Denson JS. Risk of surgery in patients over 90. Geriatrics 1972; 27: 115-118. Green J, Passman LJ, Wintfeld N. Analysing hospital mortality: the consequences of diversity in patient mix. JAMA 1991; 265: 1849-1853. Adkins RB, Scott HW. Surgical procedures in patients aged 90 years and older. South Med J 1984; 77: 1357-1364. Finucane P, Phillips G. Preoperative assessment and postoperative management of the elderly surgical patient. Med J Aust 1995; 163: 328-330. (Received 2 Jun, accepted 27 Aug 1997) Authors' details Department of Rehabilitation and Aged Care, Flinders University of South Australia, Adelaide, SA. Josephine H Harris, BM BS(Hons), Medical Registrar; Paul M Finucane, FRACP, FRCPI, Professor; Denise C Healy, RN, RM, Research Assistant. Flinders Medical Centre, Adelaide, SA. Anthony C Bakarich, RN, BN, Assistant Director of Nursing. Reprints will not be available from the authors. Correspondence: Dr J H Harris, Flinders Medical Centre, Bedford Park, SA 5042. ©MJA 1997 <URL: http://www.mja.com.au/> © 1997 Medical Journal of Australia.

Josephine H Harris · Paul M Finucane · Denise C Healy · Anthony C Bakarich

Consensus Statement Glycohaemoglobin: a crucial measurement in modern diabetes management Progress towards standardisation and improved precision of measurement* Peter G Colman, G Ian Goodall, Peter Garcia-Webb, Paul F Williams and Marjorie E Dunlop MJA 1997; 167: 96-98 Introduction - What is glycohaemoglobin and why should we measure it? - Types of assays available - Importance of reproducible measurement - How reliable are assays in Australia? - Progress towards standardisation - References - Authors' details - - More articles on Endocrinology Abstract Synopsis There are currently four principal glycohaemoglobin assay techniques (ion-exchange chromatography, electrophoresis, affinity chromatography and immunoassay) and about 20 different methods that measure different glycated products and report different units. Standardisation will lead to all assays reporting results in a standard unit, the HbA1c percentage of total serum haemoglobin, and should be in place within the next one to three years. In the interim, clinicians using glycohaemoglobin assays should be aware that the ranges indicating good and poor glycaemic control can vary markedly between different assays. The reproducibility of some assays may be insufficient to provide definitive evidence of changes in glycaemic control. Some assays may be so imprecise that they are unable to separate patients with good and poor control. Interim recommendations The terminology to be used for the assay is glycohaemoglobin (GHb) assay (recommendation from the combined meetings of the International Federation of Clinical Chemistry [IFCC] Working Group on HbA1c standardisation and the American Association of Clinical Chemistry [AACC] Subcommittee on Glycohemoglobin). The unit of measurement for GHb assays should be reported as %HbA1c (Diabetes Control and Complications Trial equivalent). Other units, such as % total GHb or %HbA1, should not be used. Assays producing these units should be converted to %HbA1c reporting units. Assays with high precision are highly desirable. The IFCC/AACC are currently recommending between-run coefficients of variation of less than 5% for manufacturers of kits and instruments. However, between-run coefficients of variation of less than 3% are far more clinically useful and therefore desirable. Introduction The landmark Diabetes Control and Complications Trial (DCCT)1 has focused increased attention on the importance of glycaemic control in preventing or retarding the progression of complications in patients with diabetes.2 Regular measurement of glycohaemoglobin is now recognised as an essential adjunct to self-measurement of blood glucose in achieving the best possible glycaemic control. However, clinicians using glycohaemoglobin assays should be aware of several potential problems which can confound the interpretation of the glycohaemoglobin result. What is glycohaemoglobin and why should we measure it? Glycohaemoglobin (GHb) is formed by a non-enzymatic interaction between glucose and the amino groups of the valine and lysine residues in haemoglobin. Formation of glycohaemoglobin is irreversible and the level in the red blood cell depends on the blood glucose concentration. Thus, measuring glycohaemoglobin provides a measurement of glycaemic control over time, and its use has been proven to evoke changes in diabetes treatment, resulting in improved metabolic control.3 First introduced in the 1970s, it is now accepted as a unique and important index of metabolic control and was a major outcome measure in the DCCT.1In the DCCT, 1441 patients with insulin-dependent diabetes were randomly allocated to intensive treatment and monitoring (usually with four insulin injections a day or pump treatment) with the aim of achieving normoglycaemia or to conventional treatment (usually with one or two injections a day). The effectiveness of intensive therapy was reflected in clear differences in mean blood glucose and glycohaemoglobin levels between the two groups. The intensive treatment group achieved a mean daily blood glucose level of 8.6 mmol/L and a median HbA1c value of 7.2% compared with the conventional treatment group, which achieved a mean blood glucose level of 12.8 mmol/L and a median HbA1c of 8.9%. These differences in glycaemic control were maintained over a mean period of 6.5 years and were associated with a 35%-76% reduction in retinopathy, nephropathy and neuropathy. Using the knowledge gained in the DCCT, doctors caring for patients with diabetes can now establish targets for glycaemic control that are based on observed outcomes, and which, if met, should minimise the development of complications. Inevitably, because glycohaemoglobin measurements reflect an integrated view of glycaemic control over time, the patients and their carers will place increasing reliance on the glycohaemoglobin result. So it is timely to evaluate the types of assays available, the moves toward standardisation of the reporting units and the precision and reproducibility of current assays. In the DCCT all glycohaemoglobin measurements were performed using the same closely standardised method. Unfortunately, in Australia there are currently four principal glycohaemoglobin assay techniques and about 20 different specific methods, most of which are not standardised between laboratories. Types of assays available The four principal techniques used to measure glycohaemoglobin are ion-exchange chromatography, electrophoresis, affinity chromatography and immunoassay. The techniques measure slightly different glycated products and use at least three different units for reporting the results (%HbA1c, %HbA1 and % total GHb). They can produce different values for the same patient specimen. This was demonstrated in a recent study in which four whole blood samples with HbA1c levels of 5.1% (representing non-diabetes), 6.7% (representing excellent glycaemic control), 8.5% (representing moderate glycaemic control) and 11.4% (representing poor glycaemic control) were distributed to 29 laboratories in Victoria for glycohaemoglobin determinations.4 The range of values obtained for the non-diabetic (4.1%-6.8%), good control (5.1%-9.3%), moderate control (6.7%- 11.9%) and poor control (10.1%-17.3%) specimens demonstrated extensive overlap between measurements of samples from patients with markedly different degrees of glycaemic control (Box 1). At present it is impossible to compare the results from two different laboratories; this can be confusing not only for patients but also for their carers. Laboratory- specific reference ranges are a means by which results from different laboratories can be compared, but the data used to derive such ranges are arbitrary and the categories into which different glycohaemoglobin levels are divided may be misleading. Importance of reproducible measurement A major use of the glycohaemoglobin assay is to assess changes in metabolic control that follow an alteration in treatment. The ability of any assay to reliably detect a change depends on its reproducibility (the ability of the assay and laboratory to get the same answer for the same sample each time). Reproducibility is normally expressed as the coefficient of variation (CV) of an assay. The CV is obtained by measuring the same sample at least 20 times in different assay runs and calculating the mean and standard deviation (SD) of the measurements; the CV is calculated by dividing the SD by the mean and expressing the result as a percentage. An assay with a high CV suffers from poor reproducibility and cannot demonstrate whether glycohaemoglobin levels have changed in different samples. Laboratories normally accept an assay for reporting purposes if the result for quality control samples falls within three SDs of the mean (3SD range). The imprecision of measurement of patient samples will be similar to that of the quality control samples. For example, if the result of an HbA1c assay with good precision (3% CV) was 7%, the 3SD range would be 6.37%-7.63%; for a result of 9%, the 3SD range would be 8.19%-9.81%. These two results can clearly be separated. In contrast, the same results of an assay with poor precision (6% CV) would have 3SD ranges of 5.74%-8.26% (for the 7% level) and 7.38%-10.62% (for the 9% level), and could not be differentiated. How reliable are assays in Australia? The Royal College of Pathologists of Australasia/Australasian Association of Clinical Biochemists Chemical Pathology Quality Assurance Programme provides external quality control samples for Australian laboratories that report glycohaemoglobin levels.5 The program runs on a six-monthly cycle, in which participating laboratories analyse two random samples per month, drawn from lyophilised whole blood samples representing six levels of glycohaemoglobin. The use of lyophilised samples can lead to minor variations in assay values for some methods. However, a recent study has excluded this as a complicating factor.6When measuring control samples with the value of 7.2% HbA1c (the mean outcome of intensive treatment in the DCCT), Australian laboratories reported HbA1c assay results between 6% and 9% HbA1c, while the range of values reported for all units (percentage of HbA1c, HbA1 and total GHb) was between 6% and 12.6%. When measuring control samples with the value of 8.9% HbA1c (the mean outcome level for conventional treatment in the DCCT trial), laboratories reported HbA1c values between 7.4% and 11%, while the range of values for all glycohaemoglobin units was 7.4% to 16.4%. The overlap between values obtained for these samples epitomises the problems currently facing clinicians in interpreting glycohaemoglobin levels and changes in levels reported by different laboratories. The interlaboratory CV obtained varied between 1.6% and 8.9% for the most common assays. To critically evaluate changes in HbA1c, the precision of individual laboratory assays for glycohaemoglobin must be known. For example, the difference in mean HbA1c value between the intensive and the conventional treatment groups in the DCCT was only 1.7%, and any assay used should at least be able to detect a difference of this order. With most laboratories using the 3SD range to accept or reject assay runs, glycohaemoglobin assays with CVs close to 3% are necessary to differentiate the two DCCT group means (Box 2). At 3% CV, the 3SD range of values for a patient with a true HbA1c level of 8.05 %HbA1c would be 7.33 to 8.77 %HbA1c. This range is less than ideal, but, realistically, only high pressure liquid chromatography assays currently achieve such precision. We recommend that the CV of the assay currently being used by the reporting laboratory be made available to carers who use glycohaemoglobin measurements. This will allow them to determine if the assay has the ability to differentiate between reported levels. Reference laboratories in the International Federation of Clinical Chemistry (IFCC)/American Association of Clinical Chemistry (AACC) International Standardization Programme must be able to achieve a CV below 3% at HbA1c levels of 6% and 9%.7 Manufacturers' assays should be able to achieve a CV below 5%. Currently, some GHb assays are unable to achieve these limits. Progress towards standardisation Standardisation is crucial to allow comparison of results obtained in different laboratories. A working party of the IFCC and AACC is coordinating an international effort by which all methods will be standardised to a designated method. This will be performed at the manufacturer level. Glycohaemoglobin analyser and kit manufacturers will have their assays standardised by reference laboratories established and monitored monthly by the IFCC/AACC working party. Thus, ultimately all laboratory methods will report their results in %HbA1c units which have been standardised against the DCCT method.8 Patients and carers will then be able to directly compare their level of glycaemic control against the enormous amount of data obtained by the DCCT trial on the onset and incidence of diabetes-related complications. References Larsen ML, Horder M, Mogensen EF. Effect of long-term monitoring of glycosylated hemoglobin levels in insulin-dependent diabetes mellitus. N Engl J Med 1990; 323: 1021-1025. Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long term complications in insulin dependent diabetes mellitus. N Engl J Med 1993; 329: 977-986. Yue DK, Colagiuri S, McElduff A, Silink M. Diabetes Control and Complications Trial. Position Statement of the Australian Diabetes Society. Med J Aust 1993; 159: 803-804. Gilbert RE, Goodall I, Young V, Jerums G. Interlaboratory variation of GHb assays in Victoria, Australia. Diabetes Care 1996; 19: 730-734. Goodall I, Gill J, Penberthy L, Gilbert R. Interlaboratory variability of glycohaemoglobin. The Australian experience. In: Proceedings of the International Congress of Clinical Chemistry, 8-12 July, 1996 (editors: Martin SM, Halloran SP). Association of Clinical Biochemists, London, UK. July C 493 (ISSN 0959-9029), London, UK. Weykamp CW, Penders TJ, Muskiet FAJ, van der Slik W. Evaluation of reference material for glycated haemoglobin. Eur J Clin Chem Clin Biochem 1996; 34: 67-72. National Glycohemoglobin Standardization Program (NGSP) (Web site) http://www. missouri.edu/,diabetes/ngsp.html Hoelzel W, Miedema K. Development of a reference system for the international standardisation of HbA1c/glycohemoglobin determinations. J Int Fed Clin Chem 1996; 9: 62-67. * Consensus statement from the Australian Diabetes Society, the Royal College of Pathologists of Australasia and the Australasian Association of Clinical Biochemists Authors' details Department of Diabetes and Endocrinology, Royal Melbourne Hospital, Melbourne, VIC. Peter G Colman, FRACP, MD, Director. Special Chemistry Unit, Austin and Repatriation Medical Centre, Melbourne, VIC. G Ian Goodall, BSc, FAACB, Unit Manager. St John of God Pathology, Perth, WA. Peter Garcia-Webb, MD, FRCPA, Clinical Pathologist. Royal Prince Alfred Hospital, Sydney, NSW. Paul F Williams, MSc, PhD, Principal Hospital Scientist. University of Melbourne Department of Medicine, Melbourne, VIC. Marjorie E Dunlop, MSc, PhD, Principal Research Fellow. Reprints: Dr P G Colman, Department of Diabetes and Endocrinology, Royal Melbourne Hospital, PO Box 3050, Parkville, VIC 3050. E-mail: petercATnursing.medrmh.unimelb.edu.au Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au>". <URL: http://www.mja.com.au/> 1: Results of glycohaemoglobin assays of four samples in 29 Victorian laboratories4 The four samples were from patients with differing degrees of diabetes control. The closed circles represent individual laboratory results for each sample and the open circles represent the notional target value. The notional target value was set by the Biorad Diamat (Biorad Laboratories, Hercules, California) in a laboratory where the assay was referenced against the DCCT method.2 Six methods of measuring glycohaemoglobin were used by the laboratories: High pressure liquid chromatography (cation exchange, measuring HbA1c)Immunoassay (measuring HbA1c)Ion exchange chromatography (manual assay, measuring HbA1c or HbA1)Affinity chromatography (measuring total GHb but expressed as either total GHb or %HbA1c)Electrophoresis (measuring HbA1c or HbA1)Low pressure liquid chromatography (measuring HbA1c and including HbF). Back to text Back to text

Peter G Colman · Peter Garcia-Webb · Paul F Williams · Marjorie E Dunlop

Abstract Objectives: To estimate the number of pregnant women in New South Wales who had not been tested for gestational diabetes mellitus (GDM) during the period 1991-1994. Design: The number of women not tested for GDM was estimated from the recorded data available in the NSW Midwives Data Collection (MDC) annual reports and compared with three incidence surveys. Main outcome measure: The number of pregnant women not tested for GDM. Results: Over the four-year period, the average annual recorded MDC incidence of GDM was 2.2%. This percentage was adjusted up to 3.3% after review of the MDC validation study and an incidence study. However, the expected incidence of GDM from three incidence surveys was 6.6%. Thus, half of the pregnant women in NSW do not appear to have been tested for GDM. Conclusion: For the four-year period 1991-1994, an estimated 50% of women in NSW were probably not tested for GDM. MJA 1997; 167: 14-16 Introduction Gestational diabetes mellitus (GDM) is carbohydrate intolerance of variable severity with onset or first recognition during the current pregnancy.1 Undiagnosed, and therefore untreated, diabetes is a serious disorder of pregnancy with an increased perinatal mortality rate.2,3,4 Following the Australasian Diabetes in Pregnancy Society's (ADIPS) 1991 recommendation that all pregnant women should be tested for GDM in every pregnancy,5 the NSW Midwives Data Collection (MDC) recorded a gradual increase in the number of women with GDM from 1.5% in 1991 to 2.7% in 1994. In contrast, from surveys in other Australian centres since 1991, the incidence of GDM ranges from 5.5%-8.8%.4,6-8 The difference between these rates and those recorded by the MDC suggests that many women are still not being tested for GDM in NSW and are at risk of complications which could either be prevented or substantially reduced. The purpose of our study was to estimate the extent of undiagnosed GDM in NSW for the period 1991-1994 by estimating the number of pregnant women who had not been tested for GDM. Methods We obtained data from the NSW MDC annual reports for 1991-1994. It is a statutory requirement under the Public Health Act 1991 (NSW) that all births (including home births) are reported to the MDC, which records information on maternal medical and obstetric conditions, delivery and infant characteristics. Incidence of GDM recorded on the MDC was compared with the incidence of GDM found in three different Australian surveys. The sources of these data are shown in Box 1. The three surveys used the 75 g oral glucose tolerance test for the diagnosis of GDM based on the ADIPS criteria.5 According to these criteria, GDM is diagnosed if the fasting plasma glucose level is ≥ 5.5 mmol/L and/or the two-hour plasma glucose level is ≥ 8.0 mmol/L. By combining these data, we obtain an average incidence of GDM, which we refer to as the expected incidence. At the Mercy Hospital for Women in Melbourne, a 50 g glucose load is used and capillary blood samples taken for plasma glucose measurement; a diagnosis of GDM is made if the glucose level at one hour is ≥ 9.0 mmol/L and the glucose level at two hours is ≥ 7.0 mmol/L. The incidence of GDM during the period 1991-1994 was 8.8%.4 While the glucose load and diagnostic criteria are different, a comparison of the Mercy Hospital and ADIPS criteria found that they diagnosed similar women and in the same proportion.9 While the incidence of GDM will vary depending on the ethnic composition of the population being examined, these surveys all included a majority of people of European extraction. Validation of the NSW Midwives Data Collection A 1990 validation study of the MDC, using a 1% sample, found that the recording of data relating to GDM had a sensitivity of 66.7% and a specificity of 99.6%.10 That is, when a case of GDM was recorded it was accurate, but only two-thirds of the known cases were recorded. Therefore, by increasing the recorded MDC incidence by 50% (representing the underestimation of GDM according to the sensitivity data), we are likely to obtain the actual MDC incidence, which we refer to as the estimated incidence. A 1993 survey carried out in the Wollongong area to determine the incidence of GDM was compared with the recorded MDC incidence of GDM in the area.6 The MDC recorded a rate of 5.0%, while the survey found a rate of 7.2%. A 50% addition to the recorded MDC incidence (as indicated by the validation study) gave a figure of 7.5%, which is similar to the 7.2% found in the survey. Therefore, the addition of a 50% correction factor to the recorded MDC incidence of GDM is a reasonable approximation. Results The proportion of women with a diagnosis of GDM recorded by the MDC for each different Area Health Service in NSW between 1991 and 1994 is shown in Box 2. It was assumed that the accuracy of data recording in all Area Health Services was equal and had not changed over the four-year period. For the whole of NSW, the percentage of women diagnosed with GDM increased from 1.5% in 1991 to 1.9% in 1992 (P < 0.001) and from 1.9% in 1992 to 2.6% in 1993 (P < 0.001). There was no significant change between 1993 and 1994. For the four-year period, the overall recorded MDC incidence of GDM was 2.2%. In the Sydney metropolitan area, the incidence of GDM ranged from a high of 3.1% in the Central Sydney Area Health Service to a low of 1.5% in the Northern Sydney Area Health Service. The incidence of GDM in the main metropolitan areas of Sydney, Newcastle and Wollongong was 2.4%, which was significantly higher than the country areas (1.6%) (P < 0.001). Some country areas also showed marked changes over the four-year period. For example, the proportion of women diagnosed with GDM increased from 0.8% to 2.7% in the Mid West Area Health Service and declined from 3.4% to 0.8% in the Far West Area Health Service. By applying the 50% correction factor to the recorded MDC incidence of 2.2%, the estimated incidence of GDM is 3.3%. However, the expected incidence of GDM is 6.6% (Box 1). Therefore, half of the pregnant women in NSW during the period 1991-1994, or 173 534 women, were probably not tested for GDM. Discussion The expected incidence of GDM was derived from pooled incidence data from the three surveys where the diagnosis had been based on the ADIPS criteria. The lowest incidence was reported in the two surveys which used data gathered from prenatal clinics.7,8 A higher incidence was reported in the survey which included private patients in the data collection.6 The higher rate found in the survey which included private patients could be anticipated, as private patients are older than clinic patients and increasing age is a high risk factor for developing GDM.4 Thus, the expected incidence of GDM of 6.6% is likely to be conservative. The recorded MDC incidence of GDM for 1991-1994 was lower than the expected incidence. With the application of the 50% correction factor to the recorded MDC incidence (as suggested by the MDC validation study and by data collection in the Wollongong area6 ) and a conservative estimate for the expected incidence of GDM, it is possible that half of all the women who delivered in NSW between 1991 and 1994 were not tested for GDM. Some women may have decided not to be tested for GDM. Others may not have been able to tolerate the glucose solution. In most cases, however, the decision not to test would have undoubtedly been made by the clinician responsible. Box 2 shows that testing for GDM is not evenly distributed, with a recorded MDC incidence range of 0.7%-3.2%. Of particular concern is that lower than average rates of testing were found in areas where a higher than average rate of GDM could be anticipated. In metropolitan Sydney, the lowest number of recorded cases of GDM were from the Northern Sydney Area Health Service. The MDC has shown that this is also the Area Health Service with the highest proportion of women aged 35 years or over at the time of delivery, a risk factor for GDM. An incidence of GDM of at least 10%4 could be anticipated in this age group. Women with an Aboriginal background have a high incidence of GDM.8 The New England and Macquarie Area Health Service, with a high proportion of Aboriginal Australians, had a rate of GDM less than the non-metropolitan average. The Far West Area Health Service also has a high proportion of Aboriginal Australians, and the recorded incidence of GDM there fell over the four-year period from 3.4% to 0.8%. Observational studies of women with untreated GDM have found a higher perinatal mortality rate than that found in glucose-tolerant women.2,3 At the Mercy Hospital for Women in Melbourne, Beischer et al. showed that, despite a steady reduction in the overall perinatal mortality rate, women who were not tested for GDM (and therefore not treated) had a significantly higher perinatal mortality rate than women who were tested.4 An estimate can be made of the excess perinatal mortality rate for women in NSW who have not been tested for GDM ( Box 3). Although the recorded incidence of GDM has increased (particularly in the two years after the ADIPS recommendation for universal testing was made), our results confirmed our belief that not all pregnant women in NSW are being tested for GDM. This may have implications for perinatal mortality rates. References Metzger BE. Summary and recommendations of the Third International Workshop-Conference on Gestational Diabetes Mellitus; 1991 Dec. Diabetes 1991; 40 Suppl 2: 197-201. O'Sullivan JB, Charles D, Mahan CM, et al. Gestational diabetes and perinatal mortality rate. Am J Obstet Gynecol 1973; 116: 901-904. Pettitt DJ, Knowler WC, Baird R, et al. Gestational diabetes: infant and maternal complications of pregnancy in relation to third-trimester glucose tolerance in the Pima Indians. Diabetes Care 1980; 3: 458-464. Beischer NA, Wein P, Sheedy MT, et al. Identification and treatment of women with hyperglycaemia diagnosed during pregnancy can significantly reduce perinatal mortality rates. Aust N Z J Obstet Gynaecol 1996; 36: 239-247. Martin FIR. The diagnosis of gestational diabetes. Med J Aust 1991; 155: 112. Moses R, Griffiths R, McPherson S. The incidence of gestational diabetes in the Illawarra area of New South Wales. Aust N Z J Obstet Gynaecol 1994; 34: 425-427. Martin FIR, Ratnaike S, Wootton A, et al. The 75 g oral glucose tolerance in pregnancy. Diabetes Res Clin Pract 1995; 27: 147-151. Yue DK, Molyneaux LM, Ross GP, et al. Why does ethnicity affect prevalence of gestational diabetes? The underwater volcano theory. Diabet Med 1996; 13: 748-752. Henry OA, Shelley-Jones DC, Oats JN, et al. Comparison of the 50 g capillary plasma glucose tolerance test with the 75 g venous plasma glucose tolerance test in pregnancy. J Obstet Gynecol Res 1996; 22: 215-219. Epidemiology and Health Services Evaluation Branch, Public Health Division, NSW Health Department. Validation Study of the New South Wales Midwives Data Collection 1990. Public Health Bull 1993; 4 Suppl 8: S5. (Received 15 Jan, accepted 8 May, 1997) Authors' details Illawarra Area Health Service, NSW. Robert G Moses, FRACP, Clinical Director of Diabetes Services. Prince of Wales Hospital, Sydney, NSW. Stephen Colagiuri, FRACP, Director of Diabetes Services.

Robert G Moses · Stephen Colagiuri

Editorial Of mice and (wo)men: the obesity (ob) gene, its product, leptin, and obesity Ground-breaking discoveries in obesity demonstrate that research can pay big dividends for society MJA 1996; 164: 393-394 Scientists are a unique breed whose research may seem to have little relevance to human health. Often, they are unable to communicate effectively what they are actually doing (particularly to economic rationalists intent on slashing research budgets), and especially when, as with obesity, the research appeared to have "hit the wall", with little hope of an imminent breakthrough. Then, along comes a discovery in obese mice1 so stunning that it opens up a new era in obesity research, with possible spin-offs for management of obesity and other associated disorders such as non-insulin-dependent diabetes mellitus (NIDDM). Obesity is epidemic in developed nations, including Australia2 and the United States,3 and is rapidly becoming so in many developing countries (particularly Pacific Island nations), as a penalty of modernisation,2 and in disadvantaged communities in developed countries (e.g., Afro-Americans and Mexican Americans).3 The annual cost of obesity to the United States is close to US$69 billion4 and this includes the cost of morbidity and mortality from cardiovascular disease, gallbladder disease, NIDDM, cancer and musculoskeletal disorders. Who can guess the personal cost to millions of obese people who splurge US$33 billion annually on new diet books or new "fad" diet programs?5 There has been no lack of effort or interest in obesity research, but the tangible results for clinical practice have been disappointing. This explains the community focus on each new miracle diet. We know about the importance of nutrition, exercise, community lifestyle interventions and pharmacotherapy for obesity, and the role of surgery for morbid obesity. Also, the genetic, sociocultural and behavioural risk determinants of obesity are well understood,6 but the basic physiological mechanisms that regulate body weight and adipose tissue have largely remained a mystery. Then, in late 1994 came the cloning of the mouse ob gene and its human homologue,1 followed within months by reports that injections of the ob protein/hormone expressed by the ob gene (named leptin, from the Greek root leptos, meaning thin) make obese mice thin.7-9 Based on research involving parabiosis (joining of the mice by anastomosis of the skin, which allows cross-circulation experiments) of obese (ob/ob) and diabetes mutant (db/db) mice, Coleman suggested over 20 years ago that a satiety factor produced in adipose tissue circulated in plasma and affected appetite through interaction with the hypothalamus.10 He further suggested that ob/ob mice lacked the satiety factor that could regulate adiposity by modulation of appetite and metabolism. There the suggestion remained until Friedman and his colleagues cloned the ob gene.1 They and other researchers subsequently prepared the recombinant ob protein, leptin; injecting ob/ob mice with leptin resulted in diminished food intake, increased energy expenditure, and dramatic weight reduction.7-9 After two weeks of treatment, there was a reduction of body fat from 12.2% to 0.7%!8 The ob gene is overexpressed in adipose tissue of obese human subjects,11,12 and overexpression and hyperleptinaemia have now been demonstrated in the best animal model of human NIDDM, Psammomys obesus.13 We are currently exploring the role of leptin in the high frequency of hyperglycaemia, hyperinsulinaemia and obesity which occurs in this rodent model. Thus, research begun over 20 years ago has culminated in findings that have set the obesity field alight and opened up new possibilities in pharmacotherapy of obesity. A look into the crystal ball reveals a vista to be explored in intermediary metabolism. It could revolutionise our knowledge of appetite control and energy regulation, and the interaction of leptin with other key hormones, such as insulin and glucagon in insulin sensitivity and resistance, remains to be explored. How many other unknown hormones are being produced by adipose tissue? Already we know that leptin administration to mice lowers blood glucose and insulin levels in obese mice.7 While assays for leptin are still in the early stages of development, high blood leptin concentrations (four to five times higher than in non-obese persons) have been demonstrated in obese subjects,14,15 and we have recently confirmed this and demonstrated a highly significant direct correlation between leptin, body mass index and serum insulin (Zimmet et al., unpublished data). Whether leptin itself is the "magic bullet" to cure obesity remains to be established, as studies in humans suggest that the problem in obese subjects may be decreased sensitivity to leptin (i.e., leptin resistance). The significance of this will become apparent as research moves to the next phase: the search for and study of the hypothalamic leptin receptor, and human clinical trials. The pace at which new developments are emerging is breathtaking, and in the space of a few weeks publications have appeared on the identification and cloning of the leptin receptor in mice,16 and a mutation has been identified in the leptin receptor of the db/db mouse.17 There will undoubtedly be concern about misuse of therapeutic agents with so much promise -- either leptin itself or drugs directed at the hypothalamic leptin receptor -- particularly with the possibility of a person gorging and then having an injection or taking a tablet to undo the consequences of the indulgence! While these concerns are important, this discovery provides a quantum leap in our understanding of the pathophysiological mechanisms leading to obesity and has clearly defined an avenue for its prevention. This then leads to exciting possibilities for understanding the aetiology and reducing the morbidity and mortality of a host of chronic conditions associated with obesity, including coronary artery disease, the insulin resistance metabolic syndrome (or syndrome X) and NIDDM. Coleman's elegant parabiosis experiments and Friedman's relentless search for the ob gene and leptin bring hope to hundreds of millions of obese people around the world. Debate about the appropriateness of animal experimentation will continue forever, but here is one classic example where such research may pay huge human dividends. This discovery may help the community understand how medical research works for society's ultimate benefit, and gives researchers a tangible result to convince politicians that funds applied to long term basic medical research can be an excellent investment! Paul Zimmet Chief Executive Officer, International Diabetes Institute Melbourne, VIC Greg R Collier Senior Lecturer, School of Nutrition and Public Health Deakin University, Geelong, VIC Zhang Y, Proenca R, Maffei M, et al. Positional cloning of the mouse obese gene and its human homologue. Nature 1994; 372: 425-432. Segal L, Carter R, Zimmet P. The cost of obesity. The Australian perspective. PharmacoEconom 1994; 5(Suppl 1): 45-52. VanItallie TB. Worldwide epidemiology of obesity. PharmacoEconom 1994; 5(Suppl 1): 1-7. Wolf AM, Colditz GA. The cost of obesity: the US perspective. PharmacoEconom 1994; 5(Suppl 1): 34-37. Berg FM. Diet industry hard hit since 1990, hopes for recovery. Healthy Weight Journal 1994; 8: 67-68. Lissner L. Causes, diagnosis and risks of obesity. PharmacoEconom 1994; 5 (Suppl 1): 8-17. Pelleymounter MA, Cullen MJ, Baker MB, et al. Effects of the obese gene product on body weight regulation in ob/ob mice. Science 1995; 269: 540-543. Halaas JL, Gajiwala KS, Maffei M, et al. Weight-reducing effects of the plasma protein encoded by the obese gene. Science 1995; 269: 543-546. Campfield LA, Smith FJ, Guisez Y, et al. Recombinant mouse OB protein: evidence for a peripheral signal linking adiposity and central neural networks. Science 1995; 269: 546-549. Coleman DL. Effects of parabiosis of obese with diabetic and normal mice. Diabetologia 1973; 9: 294-298. Masuzaki H, Ogawa Y, Isse N, et al. Human obese gene expression: adipocyte- specific expression and regional differences in the adipose tissue. Diabetes 1995; 44: 855-858. Lšnnquist F, Arner P, Nordfors L, et al. Overexpression of the obese (ob) gene in adipose tissue of human obese subjects. Nat Med 1995; 1: 950-953. Walder K, Zimmet P, Collier GR. Expression of the ob (obese) gene in Psammomys obesus, an animal model of obesity and non-insulin dependent diabetes mellitus (NIDDM). Proceedings of the 3rd Scientific Meeting of the Australasian Association for the Study of Obesity [abstract]. Melbourne: Australasian Association for the Study of Obesity, 1995: 42. Maffei M, Halaas J, Ravussin E, et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nat Med 1995; 1: 1155-1161. Ionsidine RV, Sinha MK, Heiman ML, et al. Serum immunoreactive leptin concentrations in normal-weight and obese humans. N Engl J Med 1996; 334: 292-295. Tartaglia LA, Dembski M, Weng X, et al. Identification and expression cloning of a leptin receptor, OB-R. Cell 1995; 83: 1263-1271. Lee G-W, Proenca R, Montez JM, et al. Abnormal splicing of the leptin receptor in diabetic mice. Nature 1996. In press. Reprints: Professor P Zimmet, Chief Executive Officer, International Diabetes Institute, 260 Kooyong Road, Caulfield, VIC 3162.

Paul Zimmet · Greg R Collier