Volume 171 Issue 10

Clinical Practice Surgery in the Information Age Currently, surgical procedures involve direct data flows of sensation and mechanical output. As information technology progresses, surgery will change to a system of electronic data flows, with technical, ethical, and training implications. Patrick Cregan MJA 1999; 171: 514-516 Introduction - Electronic sensory input - Electronic mechanical output - The patient -- the data source - The surgeon - Conclusion - References - Authors' details - - More articles on Surgery Introduction Currently, a surgical procedure is a direct human-to-human process in which sensory input and mechanical output data flow directly between patient and surgeon (Box 1). Procedures are usually taught using human or animal models and the utility of the procedure and the surgeon's ability are measured predominantly by patient outcomes. However, new technologies in which computing, graphics, robotics, telecommunications and touch converge are changing all this. A surgical procedure will become a flow of electronic data: sensory input will be transmitted electronically to the surgeon, the surgeon's responses and actions will be converted to electronic data, and then translated back to mechanical intervention at the patient. Electronic sensory input The application of television to surgery was the basis for the revolution in laparoscopic cholecystectomy and similar advances in arthroscopic and other endoscopic surgical techniques (Box 2). With these techniques, the surgeon receives and responds to electronic data. Not only vision but other sensory input data -- force feedback sensation, touch and proprioception -- can be relayed locally or at a distance in real time.1,2 "Telesmell" is under development;3 currently about 30 smells can be rendered via an electronic interface. Implications Electronic sensory input can be manipulated. Images can be overlaid with other digital data -- a television monitor can display the patient's tumour, as imaged with spiral computed tomography (CT) or magnetic resonance imaging (MRI), superimposed on the live image to guide the surgeon (eg, Navitrack, Orthosoft Inc, Montreal4). Other data, such as history, vital signs or other physiological data, can be included in the image, or the image can be magnified, enhanced or rendered in three dimensions. The surgeon now has enhanced sensory input beyond that which is currently available in the traditional surgical scenario of the patient on the operating table. Additional sensory inputs can be added to further enhance skills; for example, proximity sensing with ultrasound can drive an audio feedback similar to that used in aircraft signalling ground approach.5 Electronic mechanical output Sensory input is only one half of the data loop in a surgical procedure. For electronic surgery the surgeon's actions must be translated reliably into electronic data which can be translated back to mechanical output. The means for mechanical-to-electronic translation is already in use in our daily lives -- the mouse and keyboard on our computer, or controllers for video games. Surgical control interfaces have been developed: these include hand-controlled adapted instruments (eg, Virtual Laparoscopic Interface, Immersion Corporation, San Jose, California6); gloves to measure and translate hand movement to data flows (eg, CyberGlove, Virtual Technologies Inc, Palo Alto, California7); and sensors fitted to ordinary instruments (eg, the 3D tool developed at Stanford University using a miniBIRD sensing system [Ascension Technology Corporation, Burlington, Vermont]8). Implications An electronic interface allows for other methods of data input. Voice control and activation are in routine use in operating rooms with the AESOP camera-holding robots (Computer Motion Inc, Santa Barbara, California)9 and the HERMES system of controlling many functions of the operating room (Computer Motion). Other possibilities for interaction being explored include electromyographically driven devices. Machines are good at things that humans are poor at. Machines can be immensely strong or gentle, they are perfectly still when required or can move very rapidly. Their motion can be scaled up and down and they do not become tired or bored. Electronic-to-mechanical translation has been slower to develop, but there are now robots (AESOP)9 holding and moving laparoscopic cameras in operating rooms, approved by the US Food and Drug Administration (FDA). The first fully robotic procedure, in which all the instruments were held and moved by a robot (ZEUS, Computer Motion),9 was performed in a real life tubal reanastomosis procedure at the Cleveland Clinic in June 1998. The FDA has now given approval for a trial in humans of ZEUS for microvascular cardiac and other minimally invasive procedures, and daVinci surgical robotic systems (Intuitive Surgical Inc, Mountain View, California)10 are being assessed for use in many common surgical procedures (Box 3). French surgeons have operated on human hearts using robots to hold and use the instruments,9 and in the United States several institutions are assessing these technologies.11,12 Implications A surgeon can be actively involved in mechanical movement in an operative procedure at a remote site (telesurgery). This has been performed at several places around the world, including a demonstration at the Royal Australasian College of Surgeons' Annual Scientific Congress in Sydney in 1998. Surgical skills can be improved and new procedures may be realised; for example, beating heart coronary bypass grafts with the effectors suturing under the surgeon's control while the heart and instruments appear still to the surgeon because the view and effector instruments are electronically synchronised with the heart's movements. The patient -- the data source In the electronic data flow model, the patient is the main source of data, encompassing anatomy, pathology, mechanical and physiological responses to tissue handling (local and general). These data change continuously in response to the surgeon's actions. There are now a number of virtual reality (VR) simulators where training can occur using a simulated system as the "patient", with excellent graphics and the ability to change anatomy, pathology and operative problems (eg, to simulate bleeding). For example, HT Medical Inc (Gaithersburg, Maryland) produces a realistic intravenous catheterisation simulator,13 and bronchoscopy, colonoscopy, gastroscopy and arthroscopy simulators are now appearing.5,14Implications The electronic data that represent the patient can be simulated, manipulated, stored and accessed at a distance in time and space from the surgeon. The need for live-patient or animal training is past; virtual surgery and VR surgical simulation have arrived. Preoperative simulation may raise ethical dilemmas as well as improve technique: do we believe a simulation that says that the patient is unsuitable for operation, or try anyway in a "hopeless" case? Patient privacy may be hard to guarantee when patients are reduced to "data pools". The surgeon Can the surgeon then be replaced in the data loop? Ultimately, the answer is "yes". However, there are technical limitations, which mean that this will happen slowly. In particular, the human ability to recognise complex patterns in real time, to develop and amend complicated strategies and to monitor the results, combined with machine limitations (insufficient computing power, lack of bandwidth to transfer the necessary data in real time and poor pattern recognition by machines), mean that surgeons will be needed at least for the foreseeable future. Implications Because the surgeon's actions are now a stream of data, they can be stored, measured, scored, reproduced and used as educational tools. One surgeon can teach procedures to many individuals simultaneously, assess their progress, and certify competence against a scale of objective measures. There is evidence that VR training using the MIST VR trainer15 or similar interface16 improves skill acquisition beyond conventional techniques. Skills decomposition studies, in which the various actions in a procedure are separated out and analysed, suggest that VR training may have a useful role in learning technique.17 Recertification or skills upgrading can be assessed through VR simulations. New techniques can be learned remotely and rehearsed prospectively. Ongoing monitoring of competence is available not only to the surgeon, but also to patients, hospital administrations, registration authorities, colleges and government. A glimpse of the future Dr Mary Jones wishes to perform a new technique for minimally invasive liver resection. She has studied the literature and videos, and attended a virtual workshop at the liver resection Internet site. The basic procedure was rehearsed at her surgical workstation, using a medical VR model developed by Dr Stephen Lucasberg. Her technical skills in the procedure have been assessed and accredited by the College of Surgeons' virtual certification process, and permission for the procedure was granted by the hospital's credentials committee. A virtual simulation of the procedure, using the actual patient's MRI- and CT-derived images, indicates a 92% likelihood of success, so, after discussion with the patient, the procedure is undertaken. As this is the first time the procedure has been done in Australia, the surgery is being telementored and assisted by Dr Gates in Seattle, using a telesurgery interface so that he will be able both to assist and complete the procedure if there are problems. The image of the patient's lesion is overlaid with data from the preoperative MRI scan, and CT scan data are available to overlay as well. After introduction of the instruments, the procedure commences. Despite the patient's respiratory excursion, and therefore movement of the liver, the working image and instruments remain still, as they are linked electronically to fiduciary points in the image. Magnification of the image is readily available as needed, and proximity sensors fitted to the tips of the instruments will help identify and avoid damage to the inferior vena cava, which lies close behind the lesion. As Dr Jones recently fractured her forearm while skiing, she will use electromyography to control her left-hand instruments and retinal tracking to control the picture. A voice interface controls the light and camera settings, cable position, and data display on the screen, and can call up laboratory results and other data. The procedure was completed successfully and Dr Jones's movements were all stored, analysed, and subsequently scored. Dr Gates reviewed the scores and noted some over-shooting of the liver resector posteriorly, and recommended that Dr Jones undertake additional training in depth perception analysis before her next procedure. Her registrar will replay the procedure a number of times, being guided by haptic feedback to the instruments she is holding so as to learn the tissue "feel" and proprioception in the procedure. Conclusion The convergence of Information Age technologies such as computing, robotics and telecommunications will radically alter the performance, teaching, recording and assessment of surgical procedures. The concept of surgery as a flow of data between surgeon and patient gives a framework for assessing these technologies and developing new techniques. References Satava RM, Jones SB. Human interface technology. In: Satava RM, editor. Cybersurgery: advanced technologies for surgical practice. In: Sackier JS, Series editor. Protocols in general surgery. New York: John Wiley & Sons, 1998; 24, 28, 30. Satava RM, Jones SB. Telepresence surgery. In: Satava RM, editor. Cybersurgery: advanced technologies for surgical practice. In: Sackier JS, Series editor. Protocols in general surgery. New York: John Wiley & Sons, 1998; 143-144. Krueger MW. Olfactory stimuli in virtual reality for medical applications. In: Medicine meets virtual reality. 7: The convergence of physical and informational technologies: options for a new era in healthcare. Proceedings of the Seventh Medicine Meets Virtual Reality Conference. 1999 Jan 20-23, San Francisco. 61. Navitrack. Orthosoft.<http://www.orthosoft.ca/navitrack.html>. Accessed 12 October 1999. Karron DB, Bucholz RD, Wegner K, Zicarelli D. Tactical audio for neurosurgical navigation: first clinical experience. In: Medicine meets virtual reality. 7: The convergence of physical and informational technologies: options for a new era in healthcare. Proceedings of the seventh Medicine Meets Virtual Reality Conference. 1999 Jan 20-23, San Francisco. 58. Immersion Corporation. <http://www.immerse.com/>. Accessed 12 October 1999. Virtual Technologies, Inc. <http://www.virtex.com/>. Accessed 12 October 1999. Ascension Technology Corporation. <http://www.ascension-tech.com/>. Accessed 12 October 1999. Computer Motion, Inc. <http://www.computermotion.com/>. Accessed 12 October 1999. Intuitive Surgical, Inc. <http://www.intusurg.com/>. Accessed 12 October 1999. Department of Surgery. Uniformed Services University of the Health Sciences.<http://surgery.usuhs.mil/>. Accessed 12 October 1999. Department of Surgery. Yale University. <http://yalesurgery.med.yale.edu/>. Accessed 6 October 1999. HT Medical Systems, Inc. <http://www.ht.com>. Accessed 12 October 1999. Englmeier K, Haubner M, Krapichler C, Reiser M. A new hybrid renderer for virtual bronchoscopy. In: Westwood JD, Hoffman HM, Robb RA, Stredney D, editors. Medicine meets virtual reality. In: studies in health technology and informatics technology. Vol. 62. Ohmsha: IOS Press, 1999; 109-115. Chaudhry A, Irvine V, Sutton C, McCloy R. Quality of human-computer interaction, learning rate, fixed and variable factors affecting performance on a laparoscopic simulator, MIST VR. In: Medicine meets virtual reality. 7: The convergence of physical and informational technologies: options for a new era in healthcare. Proceedings of the seventh Medicine Meets Virtual Reality Conference. 1999 Jan 20-23, San Francisco. 43-44. Gorman PJ, Lieser JD, Murray WB, Haluck RS, Hummel TM. Evaluation of skill acquisition using a force feedback, virtual reality based surgical trainer. In: Westwood JD, Hoffman HM, Robb RA, Stredney D, editors. Medicine meets virtual reality. In: studies in health technology and informatics technology. Vol. 62. Ohmsha: IOS Press, 1999; 121-123. Cao CGL, MacKenzie ML, Ibbotson JA, et al. Hierarchical decomposition of laparoscopic procedures. In: Medicine meets virtual reality. 7: The convergence of physical and informational technologies: options for a new era in healthcare. Proceedings of the seventh Medicine Meets Virtual Reality Conference. 1999 Jan 20-23, San Francisco. 43. Authors' details Nepean Hospital, Penrith, NSW. Patrick Cregan, FRACS, Surgeon, Department of Surgery. Reprints: Dr P Cregan, PO Box 1124, Penrith, NSW 2751. Patrick_CreganATonaustralia.com.au Back to text Back to text Back to text

Patrick Cregan

Clinical Practice The potential impact of home telecare on clinical practice Home telecare, in which the health status of patients at home is monitored remotely, has the potential to improve care and reduce costs. Its widespread implementation would require fundamental changes in the healthcare system. Branko G Celler, Nigel H Lovell and Daniel K Y Chan MJA 1999; 171: 518-521 Introduction - Telecare in clinical practice - Evidence for cost effectiveness of home telecare - Data security - Conclusions - References - Authors' details - - More articles on Administration and health services Introduction Home telecare is the use of information, communications, measurement and monitoring technologies to evaluate health status and deliver healthcare from a distance to patients at home (Box, Figure 1). In the United States, home healthcare is the fastest-growing healthcare delivery sector1,8 -- more than 5.9 million Americans received home healthcare services valued at more than $US25 billion in 1996. Only about 50 of almost 1800 US home care agencies are currently active in home telecare, but, driven by changes in healthcare provision and reimbursement policies, many more are participating in trials of cost effectiveness; home telecare is expected to grow dramatically. Outside the US, the move towards telecare is being driven by the acceptance that national health services have a responsibility to manage the needs of an ageing population.2,9,10 People aged 65 years and over now represent 12% of the Australian population,11 a figure which will increase to 25% by 2051. Furthermore, average healthcare expenditure per person is currently $2536 per year, but increases almost tenfold for those aged 75 years and over. The increasing cost of providing healthcare services to an ageing population and changing patterns of use of hospital resources (a rise in admissions but a fall in the average length of stay) are powerful forces for shifting the focus of care from the hospital to the home. Telecare in clinical practice The recently completed Australian Coordinated Care Trials12 identified home telecare as having significant potential for contributing to the management of patients with acute exacerbation of chronic conditions as well as at-risk elderly people living alone at home. As home telecare may become increasingly relevant as the population ages, it is useful to consider management and clinical service delivery requirements in the context of the need for services. An individual older person may suffer from multiple chronic medical problems, each requiring multiple treatments ranging from medical services to simple functional assistance. For convenience, we chose to describe the need for aged care services in three broad and overlapping age-related categories: preventive healthcare and education for self-management, health maintenance, and health support (Figure 2). Preventive healthcare: In the first phase, a rapid increase in healthcare costs begins between the ages of 55 and 65 years. This age group coincides with retirement, when many people have the resources, skills and time required to prepare for their old age through participation in preventive healthcare programs and education for self-management. For this group, key questions include: What will be the effect of using the Internet to create an information-rich environment in the home on the demand for healthcare services? Can the Internet be used to create a virtual community of the aged where participants are linked through a network of information and communication resources for education, entertainment and access to healthcare? Can such a virtual community be used to increase social interaction, improve the quality of life and reduce age-related morbidity? As the ageing community becomes better educated and increasingly able to access high quality healthcare information (through sites such as Healthfinder14 in the US and HealthInsite15 in Australia), a reduced demand for primary healthcare services may occur, but little research has been reported; the possibility that the demand on health resources may increase cannot be discounted. Health maintenance: In the second phase, pre-existing medical problems will require more assiduous management, and new morbidity associated with the chronic and degenerative diseases of old age will emerge. Established general practice and community health services are the cornerstone of primary care, and the development of innovative information technology solutions to support GPs may shift the point of care from the hospital to the home and may provide continuity of care and transparency of data flow across healthcare sectors. Innovative devices for physiological monitoring in the home of cardiac rhythms, lung function, hypertension and risk of falls may give users greater confidence and security, and could provide healthcare services with 24-hour continuous diagnostic data whenever necessary. We can expect that these healthcare services and facilities will become fully integrated with emerging "smart home" technology (which allows intelligent monitoring and control of the home environment and security requirements) in new housing estates and large scale retirement village developments. Patient data collected in the home could be automatically collected, analysed and summarised, and inserted directly into an electronic patient record to be reviewed at the next consultation. In the absence of electronic record systems in general practice, data could be collected at a dedicated central server, where they could be viewed over the Web or transmitted to the GP via facsimile. Email reminders of the existence of new data could also be generated automatically. It is not yet clear who will take responsibility for managing the collection and analysis of data, and coordinating the delivery of healthcare services. GPs, in their gatekeeper role of determining access to specialist services, are well placed to assume overall clinical responsibility, but may need to delegate operational responsibility to community-based service organisations, local community hospitals or commercial service providers. Health support: The third phase is characterised by increasing frailty, and the increasing intervention of GP and community services to support elderly people in their homes. Unobtrusive monitoring of changes in mobility or patterns of use of selected domestic services in the home (Figure 1) may provide sensitive indicators of changes in health status. This unobtrusive monitoring, together with appropriate emergency alarm services and continuous ambulatory monitoring of selected physiological parameters, will provide the GP with advance warning of deterioration in health status, and thus allow optimum coordination of preventive health services. Again, GPs are well placed to assess the level of frailty and risk at which telecare monitoring and support can be recommended to patients and their carers. If the objective is the prevention of acute episodes leading to hospitalisation and high cost care, careful management of the evolving risk will become essential so that appropriate clinical and community resources can be mobilised in a timely fashion. These alarm processes can be partly automated through use of medical expert systems to identify and communicate to the clinician significant changes in health status. Home telecare technologies Home telecare technologies have been reviewed by several authors,1-4 and fall broadly into three generations. First-generation systems are designed to reduce anxiety among elderly and high-risk patients and reduce their use of primary healthcare services. Typical technologies include personal alarm systems and emergency response telephones that make a voice connection between the patient and the response centre whenever a pendant alarm button is pressed. Second-generation systems can generate alarms without the intervention of the patient, on the suspicion that something may be wrong. These systems can continuously monitor a large number of variables sensitive to changes in functional health status (Figure 1), and generate an alarm when significant changes are observed.5 With an intelligent decision-support system using robust algorithms, false alarms are unlikely. These second-generation systems are unobtrusive, do not require direct patient participation and can be integrated with evolving "smart home" technology for home automation, security and environmental control. New developments include sensor arrays worn by the patient and capable of measuring factors such as temperature, respiration, electrocardiogram and skin blood flow.6 Ambulatory data can be transmitted to a local computer or specialised controller via low cost telemetry before transmission to a central computer. Local intelligence can be used to detect emergencies and long term trends in health status can be identified and acted upon at the response centre. Third-generation systems attempt to deal with issues of loneliness and quality of life of patients by creating a virtual community of clients, carers, healthcare providers and other community services, connected via the telephone, interactive television, and the Internet. In our laboratory, we are investigating the integration of third-generation systems with Internet and Web technology,7 both for communication and for management and control of monitoring services. Back to text Evidence for cost effectiveness of home telecare Very few studies to evaluate the cost effectiveness of home telecare have been published, and fewer still relate to Australia. According to a recent report by the United Kingdom Audit Commission,16 about 40% of total hospital and community health service expenditure is on people aged over 65 years, and those aged over 75 years occupy more than 50% of all available hospital beds. The Royal Commission on Long Term Care reported that it costs £454 a week on average for full-time residential care and £250 a week for private home care.9 Therefore, there is strong economic justification for transferring resources from residential to home-based care. There is also substantial evidence that heathcare outcomes and quality of life improve when healthcare services are home based.9,16A UK report, Technologies for telecare in the home,17 concluded that, for a typical Community Health National Health Service Trust (similar to a Division of General Practice and associated Area Health Service), 15% of home visits could be replaced with telecare, saving £1.26 million per annum in the first year, after accounting for establishment and operating costs. A retrospective review of home nursing visits in the UK similarly suggested that 14%-16% of these visits could be replaced by telecare services;18 a similar and very comprehensive study in the US concluded that 46% of all activities carried out by on-site nursing could reasonably be replaced by telenursing.19 Studies of the cost-effectiveness of home telecare are most compelling for chronic conditions. In 1988, a trial of telephonic cardiac surveillance of post-infarct patients20 found that cardiac death or arrest was decreased by 29% in the monitored group when compared with the control group. In addition, control subjects were 2.4 times more likely to be clinically depressed, and they returned to work less quickly. A US study of patients with chronic disease demonstrated savings of over $8000 per patient, arising from a reduction of costs from $100 for conventional visits to $15-$40 for telecare services.21 In another study, on cardiac rehabilitation for congestive heart failure in the home, a 74% reduction in readmission rates was demonstrated at 90 days.22 Detecting delirium Mrs P is 85 years old and has mild dementia. She lives alone at home and receives home care and Meals on Wheels. In the past six months, Mrs P has twice been admitted to hospital because of increasing confusion (delirium). The first episode was found to be caused by a newly introduced medication, and the second was due to urinary sepsis. After the second admission, continuous home monitoring of behavioural patterns was introduced. One day, the monitoring system recognised that Mrs P's pattern of behaviour was significantly different than usual, and so it alerted her GP. The GP recognised that this could be a new episode of delirium and arranged an urgent geriatric review. The geriatrician found that Mrs P was constipated and had urinary retention. Mrs P was admitted to hospital and given an enema, and she was able to void. Her mid-stream urine microscopy result was clear. She was sent home on the same day, and her behaviour pattern returned to baseline. A possible long hospital admission was avoided by early identification and intervention in a medical emergency. Data security Ethical issues arising from home telecare and the storing and accessing of clinical data by multiple providers are complex, but are reviewed (at least in the European context) by Stanberry.23,24 Implementation of home telecare services requires informed consent, must be voluntary and must complement, not exclude, traditional methods of healthcare delivery. Moreover, individuals are entitled to assurance that personal information will not be subject to unauthorised access, and will be used only for the purposes it has been collected for. This requires procedures and processes to ensure that personal data can only be accessed by those authorised to do so. Automatic encryption of data should be mandatory for any transmission of identified patient data. Web-based security, fuelled by the explosive growth in e-commerce, is now dependent on two standards, Secure Sockets Layer and Secure Electronic Transaction, which must be implemented if patient data are to be transmitted via the Web. These security systems depend on the exchanging or sharing of keys, which must be protected from access by others. The strength of any cryptographic system depends on key- distribution techniques and the existence of a trusted third party to manage the process. Who will become the trusted third party in Australia is yet to be resolved. Home heart monitoring Mr D (aged 75 years) lives alone on a farm outside a small country town, and has known ischaemic heart disease. Because of his remoteness, it is inconvenient for him to have regular medical consultations. He has had several episodes of angina but has never had an infarct. Recently, Mr D suffered a prolonged episode of chest pain, and attended his nearest hospital clinic. This is manned by a nurse after hours, and is some distance away from the nearest doctor. The nurse recorded an electrocardiogram (ECG) and transferred the information via the Internet to a cardiologist, who made a diagnosis of unstable angina. Mr D was admitted overnight to a local community hospital, given the appropriate management, and reviewed by his own doctor the next day. He was instructed on the use of a simple ECG monitoring device capable of transmitting ECG recordings over the telephone. Mr D returned home secure in the knowledge that his condition would be monitored, and that medical services would receive and act upon subsequent anginal attacks. Conclusions The interaction and evolution of home telecare technologies with smart home technology, and advanced communications (Internet and Web) services into the home may lead to an increased capacity for self-management through improved education and enhanced perceptions of personal security and safety. As home telecare technology evolves, data collected in the home, in the surgery, or even in the routine pattern of daily life, may be analysed and used to help better coordinate the delivery of healthcare services. Implementation of home telecare will require fundamental changes in every sector of the healthcare services as GPs, hospitals and specialists adjust to different modes of service delivery, often based on the transfer of data and information and telemedicine diagnosis and consultation, and driven by patient demands for a greater role in shared decision making. For clinicians to become significant participants and leaders in these changes, they will need to become fully integrated in a communications network based on Web and Internet technologies. The effective incorporation of such networks in clinical practice, however, will require a major paradigm shift among clinicians and the active involvement of the medical colleges, academic institutions and government. Widespread implementation of home telecare, however, may not be limited by the availability of technology, which is becoming less costly and more effective every year, but rather by inertia and resistance to change within the healthcare system. Large-scale trials to compare quality of life and healthcare outcomes of a matched cohort of elderly subjects, some supported through home telecare and others supported through conventional healthcare services, are required to provide evidence that home telecare leads to a clear diminution of acute-care episodes, improved quality of life and reduced age-related morbidity at reduced cost. References Kinsella A. Home telecare in the United States. J Telemed Telecare 1998; 4: 195-200. Doughty K, Cameron K, Garner P. Three generations of telecare of the elderly. J Telemed Telecare 1996; 2: 71-80. Ruggiero C, Sacile R, Giacomini M. Home telecare. J Telemed Telecare 1999; 5: 11-17. Stoecke JD, Lorch S. Why go see the doctor? -- Care goes from office to home as technology divorces function from geography. Int J Technol Assess Health Care 1997; 13: 537-546. Celler BG, Earnshaw W, Ilsar ED, et al. Remote monitoring of health status of the elderly at home. A multidisciplinary project on aging at the University of New South Wales. Int J Biomed Computing 1995; 40: 147-155. Johnson P, Andrews DC. Remote continuous physiological monitoring in the home. J Telemed Telecare 1996; 2: 107-113. Magrabi F, Lovell NH, Celler BG. A Web-based approach for ECG monitoring in the home. Int J Med Inf 1999; 54: 145-153. National Home Care Association. Basic statistics about home care. Washington: NHCA, 1995. Sutherland S, chairman. With respect to old age: long term care -- rights and responsibilities. A report by the Royal Commission on Long Term Care. March 1999. <http://www.official-documents.co.uk/document/cm41/4192/4192.htm>. Accessed 19 October 1999. Hokenstad MC, Johansson MC. Eldercare in Sweden: issues in service provision and case management. J Case Management 1996; 5: 137-141. Australian Institute of Health and Welfare. Australia's health services expenditure to 1997-98. Health Expenditure Bulletin Series, No. 15. Canberra: AIHW, 1999. Available at <http://www.aihw.gov.au/publications/health/heb/heb15.pdf>. Commonwealth Department of Health and Aged Care. The Australian Coordinated Care Trials: background and trial descriptions. Canberra: Commonwealth of Australia, 1999. Available at <http://www.health.gov.au/hsdd/cocare/pdf/cctbook1.pdf>. Lapsley HM, Grant C. The Australian health care system 1992. Australian Studies in Health Services Administration No. 75, 1993. Healthfinder. United States Department of Health and Human Services. <http://www.healthfinder.gov/>. Accessed 19 October 1999. HealthInsite. Commonwealth Department of Health and Aged Care. <http://www.healthinsite.gov.au/>. Accessed 19 October 1999. UK Audit Commission. The coming of age: improving care services for older people. London: Audit Commission Publications, 1997. Gann D, Tang P, Curry R. Feasibility study: technologies for telecare in the home. SPRU, December 1998. Wooton RM, Loane M, Mair F, et al. A joint US-UK study of home telenursing. TeleMed97. Heathrow, 26-27 November 1997. J Telemed Telecare 1998; 4 Suppl 1: 83-84. Allen A, Doolittle GC, Boysen CD, et al. An analysis of the suitability of home health visits for telemedicine. J Telemed Telecare 1999; 5: 90-96. Capone RJ, Stablein D, Visco J, et al. The effects of transtelephonic surveillance and prehospital emergency intervention system on the 1 year course following acute myocardial infarction. Am Heart J 1988; 116: 1606-1615. Schiller AE, Bondmass M, Avitall B. Technology based home care for disease management. Remington Report 1997; September/October: 10-12. Roglieri JL, Futterman R, McDonough KL, et al. Disease management interventions to improve outcomes in congestive heart failure. Am J Manag Care 1997; 3: 1831-1839. Stanberry B. The legal and ethical aspects of telemedicine. 1: Confidentiality and the patient's rights of access. J Telemed Telecare 1997; 3: 179-187. Stanberry B. The legal and ethical aspects of telemedicine. 2: Data protection, security and European law. J Telemed Telecare 1998; 4: 18-24. Authors' details Centre for Health Telematics, University of New South Wales, Sydney, NSW. Branko G Celler, BSc, BE, PhD, Professor, and Co-Director; Nigel H Lovell, BE, PhD, Senior Lecturer, and Deputy Director. Prince of Wales Hospital, University of New South Wales, Sydney, NSW. Daniel K Y Chan, MB BS, FRACP, Senior Staff Specialist Geriatrician. Reprints will not be available from the authors. Correspondence: Professor B G Celler, Centre for Health Telematics, University of New South Wales, Sydney, NSW 2052. B. CellerATunsw.edu.au Back to textBack to text

Branko G Celler · Nigel H Lovell

Clinical Practice Health online: the future isn't what it used to be Over the next 10 years, the healthcare system will change to focus more on preventive medicine and healthcare in the home, with fewer doctors and a new class of home healthcare providers. Healthcare professionals need to debate how best to manage these changes. Peter M Yellowlees and Peter M Brooks MJA 1999; 171: 522-525 Introduction - Changing information presentation - The effects - The solutions - References - Authors' details - - More articles on Informatics and computers Introduction Over the past 30 years the framework in which doctors and other healthcare professionals practise has changed relatively little in comparison with the enormous changes seen in transport, manufacturing and telecommunications. While many doctors and health service managers prefer to ignore the extraordinary changes outside of the health system, they do so at their peril. Healthcare will be very different by 2010; the focus will be on the patient at home rather than the provider in the institution. There are three major drivers for this change.1 The first is the economic imperative to restrain healthcare costs in a setting of an ageing community and escalating costs of institutional care and technology. Our present model of care primarily focused on institutions, be these hospitals or related step-down facilities and nursing homes, is not sustainable. In Australia, we already spend more than $21 billion per year on institutional (hospital and nursing home) care. We have to explore ways of reducing this cost. The second is increasing consumerism, and the evolution of the "informed patient". As the "baby boomer" generation ages it will be increasingly concerned about its own welfare and will focus more and more on health. Every social issue that this generation has touched has changed radically, and there is no reason why healthcare should be exempt. The dynamic, yet often self-centred, approach by the baby-boomers is likely to be translated into a much stronger push for home healthcare. The third is the extraordinary changes in communication technology, and the evolution of the Internet. Knowledge has never been as important and as accessible as it is today -- it is now one of the economic cornerstones of our society. The distribution of knowledge is occurring at a remarkable pace via the Internet, as well as through multiple other media outlets. For some years, clinical care has been increasingly delivered electronically via telemedicine, as well as telephonically. Australia is at the leading edge of these developments.2,3 Changing information presentation Before we can benefit from the new technologies (see Box), we have to solve the problems of information quality and information overload, especially on the Internet.9 A variety of sites are being developed as quality health information portals, such as the National Library of Medicine in the United States,10 HealthInsite in Australia,11 and Omni in the United Kingdom.12 In addition, approaches are being developed to allow clinicians and patients to better assess the reliability and validity of health information.13 A more comprehensive Internet classification and coding system using metadata, as well as the development of sophisticated search engines, needs to emerge as a long term solution for this important problem.9 Only then will doctors be able to effectively obtain good quality decision support information within the time and process of a typical consultation. Within the health industry there have been enormous strides in the past five years in the development of electronic patient records, many using Internet protocols. The health system is, unfortunately and inevitably, still replete with many different types of information systems, most of which have been focused on financial and administrative applications. The challenge in the future will be to get all of these "legacy" systems to talk to the new Internet-based systems. Fortunately, Australia is well placed in this respect with the recent funding by the Federal Government of the Cooperative Research Centre for Distributed Systems Technology,14 which has a long term research program with the Centre for Online Health at the University of Queensland.15 There are many related activities at Monash University,16 the University of New South Wales,17 and in private industry. The national approach taken by the Collaborative Health Informatics Centre18 is greatly assisting the integration of the health and information technology industries. The effects We have a good idea of the illnesses and diseases that will be most prevalent and will cause the greatest disease burden by 2020.19 These are chronic cardiac, respiratory and psychiatric diseases, as well as road traffic accidents. The cardiac, respiratory and psychiatric diseases are all highly amenable to the provision of long term home care, while clearly a much more active approach to prevention and education is required to reduce the impact of road traffic accidents. A changing paradigm If we assume that by 2010 health information will be available in the homes of most Australians, and certainly on every health professional's desktop, then what will be the effects? The health system is already moving away from supporting episodic care to supporting continuity of care, and from a service-provider focus to an informed-patient focus. Increasingly, our past individual approach to treatment will be overtaken by the need for team approaches, underpinned by evidence and outcomes, clinical pathways, and guidelines. "Wellness promotion" will be seen as being more important than illness treatment. There will be a move away from institutional care to community care and to the development of the shared, distributed electronic patient record. Hospitals and health departments will make the shift from being autonomous, slowly-growing empires to becoming fluid and rapidly changing enterprises, as has occurred in industry. Therefore, there will be a need, quite simply, for fewer hospitals and more home care and community support services. Figure 1 shows a mock-up of how a patient's Web browser might look during an online consultation with a general practitioner. This patient-accessible electronic home care record system would allow instant contact with a range of healthcare professionals, information sources and other health services in an electronic distributed environment. The system will involve telemetry, monitoring, video links and automatic ordering systems, all delivered via Internet2. The entire health system will be focused on the patient at home (Figure 2) rather than on providers and institutions. If we can reduce institutional care by just 10%, over $2 billion per year will be made available for redistribution. The traditional doctor-patient relationship will alter, being driven much more intensively by patients. The doctor's role will become more advisory, analytic and interventionist. Doctors will need to be experts in assessing information from many different sources and in clinical reasoning, particularly for patients requiring more than a guidelines-and-pathways approach to care. Healthcare education will also be radically different. Medical schools and other health education institutions need to be thinking today about educating clinicians to work within a distributed, primary care focused environment.20,21 If we are correct in our predictions, there will be a need for fewer doctors beyond 2010. Not only will so much of our present-day medical content, knowledge and expertise be less important, but a group of highly skilled home healthcare professionals will exist, probably evolved from today's nurse practitioners. These healthcare providers will have prescribing and other treatment roles for patients being treated within pathway and guideline protocols. The changes will not stop there. There will also be massive opportunities, particularly for Australia, as it may be possible for Australian physicians to provide very much cheaper electronic healthcare into the US than is available locally within that country, if only because of differences in the cost of living and the strength of the dollar. It is more likely, however, that electronic healthcare will be provided in three main time zones (Figure 3), as it is highly unlikely that a doctor in, say, Australia will be prepared to consistently get up in the middle of the night to treat a patient in, for instance, Saudi Arabia or Brazil. The introduction of global electronic physicians and virtual healthcare systems will raise many important cultural, ethical, legal and legislative issues. These include, for example, the need for international medical registration and medical defence systems, the development of global information and security standards for the Internet, more flexibility in drug licensing across countries, and the need to integrate Eastern and Western styles of medicine when working across cultural boundaries. The solutions What are the implications of these changes for the present Australian healthcare sector? What should be done to prepare for this scenario? The following are suggestions. Cultural and political understanding and attitudes: There is a need for increased awareness of the importance of communications technologies in healthcare. Most global companies assume that 5%-10% of their budget will be spent on communication and information technologies. Research into distributed healthcare, both clinical models and technical solutions, is likely to be just as important as biotechnology in improving our national health profile. The clinical and information management issues are more important than, and have to drive, the technological changes. Once government makes a commitment to the changes looming in the near future, there will be the opportunity to create the necessary cultural and social changes required nationally to enable us to move to a future where information technology underpins healthcare delivery. The healthcare environment: There is a need for urgent, widespread debate about the future of healthcare, about the respective roles of doctors, patients, and other healthcare professionals, and about how best to transform a hospital-focused health system to one centred on patients and home care. If change is not guided from within the health system, it will certainly be enforced by external global and national factors. Technological requirements: There is a need to link the many existing computing systems into an Internet-based future. The necessary technological and information-based research and development programs must be focused on the development of user-friendly interfaces for patients of all ages, as well as for clinicians, and will involve the development of electronic clinical care protocols, whether these be delivered in real time, or by "store-and-forward" email, video mail, video conferencing, telephony or other methods. Specific projects need to be developed in home care, in wireless and collaborative environments, and in the development of improved electronic records. Australia needs closer links to the Internet2 consortium and to the exciting opportunities occurring in other countries, such as the Multi Media Super Corridor in Malaysia.22 Henry Ford, around the start of the 20th century, was quoted as saying that "history is bunk". While we believe strongly that history is of great importance, we also have to be well aware that the range and variety of changes confronting the world at present are greater, and are occurring more rapidly, than has ever been the case in the history of mankind. To quote Dr Rick Satava, an eminent surgeon with NASA: "The future isn't what it used to be."23 References Yellowlees P. Therapy online. Kansas: Telemedicine Today, 1999. Queensland Telemedicine Network. Queensland Health. <http://www.health. qld.gov.au/qtn/home.htm>. Accessed 18 October 1999. Yellowlees PM, Kennedy C. Telemedicine: here to stay. Med J Aust 1997; 166: 262-265. University Corporation for Advanced Internet Development. The Internet2 Project. <http://www.internet2.edu/>. Accessed 18 October 1999. Cairncross F. The death of distance. How the communications revolution will change our lives. Boston: Harvard Business School Publishing, 1997. Virtual Collaborative Clinic. <http://www.nren.nasa.gov/vdoc.html>. Accessed 5 October 1999. Graphics Visualisation and Usability Center, College of Computing, Georgia Tech. Virtual Reality Exposure Therapy. <http://www.cc.gatech.edu/gvu/virtual/Phobia/>. Accessed 5 October 1999. Van Houweling D. Distributed Education. 1998. Telecon '98 Conference, Anaheim, California. Appleyard R. Enhancing internet medical document retrieval with 'medical core metadata'. Health Information on the Internet 1999; 10: 6-8. United States National Library of Medicine. <http://www.nlm.nih.gov/>. Accessed 18 October 1999. HealthInsite. Commonwealth Department of Health and Aged Care. <http://www.healthinsite.gov.au/>. Accessed 18 October 1999. OMNI: Organising Medical Networked Information. <http://omni.ac.uk/>. Accessed 5 October 1999. Discern Online. <http://www.discern.org.uk/>. Accessed 12 October 1999. Distributed Systems Technology Centre. <http://www.dstc.edu.au/>. Accessed 5 October 1999. Centre for Online Health. University of Queensland. <http://www.coh.uq.edu.au/>. Accessed 5 October 1999. Centre of Medical Informatics. Monash University. <http://www.monash.edu.au/informatics/>. Accessed 5 October 1999. Biomedical Systems Laboratory. University of New South Wales. <http://www.bsl.unsw.edu.au/>. Accessed 5 October 1999. Collaborative Health Informatics Centre. <http://www.chic.org.au/main.html>. Accessed 5 October 1999. Murray CJ, Lopez AD, editors. The global burden of disease: a comprehensive assessment of mortality and disability from diseases, injuries, and risk factors in 1990 and projected to 2020. Harvard School of Public Health, 1996. Carlile S, Sefton AJ. Healthcare and the information age: implications for medical education. Med J Aust 1998; 168: 340-343. Coiera E. Medical informatics meets medical education. Med J Aust 1998; 168: 319-320. Mohan J. Malaysia's Telemedicine Vision and Initiatives. 1997. Telemed Asia '97 Conference, Kuala Lumpur, Malaysia. Satava RM. Telemedicine and virtual reality. American Telemedicine Association Annual Meeting, Salt Lake City. 17-21 April 1999. Authors' details Faculty of Health Sciences, University of Queensland, Brisbane, QLD. Peter M Yellowlees, MD, FRANZCP, Professor of Psychiatry, and Director, Centre for Online Health; Peter M Brooks, MD (Monash), FRACP, Professor, and Executive Dean. Reprints will not be available from the authors. Correspondence: Professor P M Yellowlees, Department of Psychiatry, University of Queensland, K Floor, Mental Health Centre, Royal Brisbane Hospital, Brisbane, QLD. P. YellowleesATmailbox.uq.edu.au http://www.coh.uq.edu.au Changing information technologies The technical ability to obtain high quality health information in the home or on the doctor's desktop depends on two things: bandwidth and accessibility. Both are on the point of being transformed to make massive amounts of information easily available to the clinician. The bandwidth issue relates to the development of Internet2 by a consortium of about 200 partners, mainly in North America, and including over 130 universities and more than 40 commercial concerns. Internet2 is expected to be between 100 and 1000 times more powerful than the present Internet. It will use much more efficient methods of information packaging to send more information down an equivalent-sized channel in a given time. This will enable a whole new generation of applications and has the potential to transform our lives in ways we cannot yet imagine.4 The issue of accessibility to information is also being resolved within First World countries, although it is crucial to note that in 1999 two-thirds of the world's population still do not even have access to a telephone.5 There are already over half a million kilometres of fibreoptic cable connecting cities and countries around the world. This will double within the next five years. By the end of 2000, it is expected that the latest of many intercontinental data links, a massively powerful fibreoptic cable weaving from Germany through the Mediterranean, across south-east Asia and on to Japan and Korea, will be installed. Simultaneously, greatly improved interactive satellites are being launched. There are now more than 200 such satellites in low earth orbit, acting like mobile phone towers or repeaters above the earth. Within five years, it is likely there will be more than 1000 such satellites, providing accessible global coverage. Improved bandwidth and accessibility will provide the opportunity to radically change the way we work and conduct business. We will be able to develop fully digitised libraries that include comprehensive video and audio collections, as well as develop cyberclinics such as the NASA-sponsored Virtual Collaborative Clinic.6 There will also be collaborative virtual research laboratories enabling "tele-immersion" - the ability to move inside space, inside the human body and into virtual reality situations. Virtual reality scenarios, where the patients move into a virtual world as part of their treatment process, are already being used to treat patients with specific phobias of heights and spiders.7 Scenarios also exist to allow surgeons to immerse themselves within a virtual middle ear, and teach the anatomy, pathology and surgery of the ear from within that organ.8 Glossary Bandwidth: The data transfer rate of an electronic communications system. Internet: An electronic communications network that connects computer networks and organisational computer facilities around the world. Internet protocol: The communications methods used for the Internet. Metadata: Data about data, such as what field the data relate to, who compiled the dataset, or how the data are formatted. Telemetry: Measurement of data and transmission to another site for storage or analysis. Virtual reality: An artificial environment which is experienced through sensory stimuli (as sights and sounds) provided by a computer and in which one's actions partially determine what happens in the environment. Back to textBack to textBack to textBack to text

Peter M Yellowlees · Peter M Brooks

Diagnosis The stage is set for the diffusion of positron emission tomography (PET) in oncology A large body of evidence now attests to the diagnostic accuracy and cost-effectiveness of PET in oncology MJA 1999; 171: 527-528 Positron emission tomography (PET) is an exact, non-invasive technique for studying the body's biochemistry. The patient is injected with a positron-emitting radioisotope of a biologically active substance -- for oncological investigations, fluorodeoxyglucose (FDG) (2-deoxyglucose labelled with the positron emitter fluorine 18) is used.1,2 FDG is actively concentrated in cancer cells, and the PET camera detects the location of the FDG by registering the ejection of positrons from the nuclei of fluorine 18 atoms. As the positrons are ejected they collide with electrons; both particles annihilate and emit two 511 keV gamma rays at 180º to each other. Rings of detectors in PET cameras register these signals and the resulting images of particular organs, or the whole body, are displayed in three dimensions. Thus, PET can reveal the presence of cancer by recording an increased rate of glucose metabolism before any of the structural changes detectable by ultrasound, radiography, computed tomography (CT) and magnetic resonance imaging (MRI) have occurred. The history of PET in Australia is given in Box 1. Recognition of the utility of PET, particularly in cancer management, is reflected in its expanding applications around the world, and in the proportion of papers on PET (rising from 10% to 36%) presented at annual meetings of the US Society of Nuclear Medicine.3 The majority of these papers relate to its applications in oncology. Similarly, most clinical PET studies performed in Australia (over 80%) have been for cancer management.4 Box 2 gives a summary (adapted from Valk5) of the current role of PET in cancer management. Although these overseas studies demonstrate the superior diagnostic accuracy of PET in a wide range of applications in oncology, the article in this issue of the Journal by Hicks et al6 is the first extensive Australian report of the use of PET. The contributions of PET to patient management in oncology reported by Hicks and colleagues are similar to those recorded in the international literature. Hicks and colleagues did not assess cost-effectiveness as part of their audit of PET studies, but studies in other countries have provided a large body of evidence of the cost-effectiveness of PET in oncology. However, Valk emphasises that, while there are adequate cost-effectiveness data on diagnosis of pulmonary nodules and mediastinal staging of lung cancer, cost-effectiveness data for the use of PET in other conditions are incomplete.7 Cost-effectiveness studies have demonstrated that the benefits of PET include avoidance of unnecessary imaging procedures (radiography, CT, and MRI) and biopsies, as well as prevention of unnecessary surgery and hospitalisation.7-9 Influenced by the findings of diagnostic accuracy and cost-effectiveness, the US government has now included the oncology applications for PET studies detailed below among its Medicare reimbursement categories: Characterisation of solitary pulmonary nodule; and initial staging of non-small-cell carcinoma of the lung (since January 1998); and Colorectal cancer recurrence or metastasis; lymphoma staging and characterisation; melanoma recurrence or metastasis (since July 1999).10 In Australia, it is time to consider making PET available at additional sites, both to improve medical outcomes for a greater number of patients and to obtain our own cost-effectiveness data. Essential to this process are: A reliable supply of FDG; Acceptable instrumentation; Accredited PET staff (physicians, scientists and technologists); Appropriate locations for PET services; and Provision for ongoing evaluation. The current situation in Australia in relation to these critical factors is as follows. Fluorine-18 deoxyglucose (FDG) supply: In the past decade, Australia has made a multimillion dollar investment in cyclotrons -- the National Medical Cyclotron (NMC) in Sydney, operated by the Australian Nuclear Science and Technology Organisation (ANSTO), and two small cyclotrons in Melbourne. These cyclotrons can supply enough FDG to meet the present and immediate future needs of all capital cities except Perth and Darwin. If PET continues to expand, it may be necessary to install further regional cyclotrons. Acceptable instrumentation: The bulk of the evidence used by expert committees in the United States and Australia to determine existing reimbursement policies in PET in oncology came from PET studies using scanners equipped with bismuth germanate crystals (BGO). PET centres at Royal Prince Alfred Hospital and the Austin and Repatriation Medical Centre are equipped with BGO cameras, which remain the reference standard for FDG PET oncology studies. Further studies on cost-effectiveness need to be based on data obtained using comparable instruments. A range of instruments claiming similar performance exists, and there is an ongoing need for these to be evaluated. Accredited PET staff (physicians, scientists,11,12 technologists): Australia has a number of PET-trained physicians, scientists and technologists. This pool of expertise will need to be expanded and appropriate accreditation guidelines defined and implemented. Appropriate location of PET services: Delivery of advanced oncology therapy is centred predominantly in major hospitals which have comprehensive diagnostic services (eg, radiography, CT, MRI, nuclear medicine), a range of other specialties and radiation oncology planning and treatment services. The oncological dominance of clinical PET usage patterns automatically proposes centres such as these as logical sites for the diffusion of PET. Ongoing evaluation: Diffusion of PET services should be carried out in conjunction with the established, transparent evaluation processes. The oncology stakeholders, including patient advocacy groups, should be involved in the formulation and implementation of protocols. Liaison with the Medical Services Advisory Committee, the federal Department of Health and Aged Care and State and Territory health departments is essential. The Federal Department of Health and Aged Care is currently conducting a review of PET with input from a Medical Services Advisory Committee PET working party and involving existing PET providers. In addition, NSW and Victoria are conducting their own reviews of PET services. As well as demonstrating clinical utility, the study of Hicks et al shows the advantages that flow from locating PET facilities in major oncology referral centres. The model for the role of PET suggested here proposes the collocation of PET services in the nuclear medicine departments of comprehensive refer ral hospitals which have existing regional oncology services. The advantages of this proposal are that PET is located where the greatest number of oncology patients can benefit from its clinical accuracy, and further large-scale evaluations of its cost effectiveness can be undertaken. John G Morris, AO Professor of Clinical Medicine Australian Nuclear Medicine and PET Consultants, Sydney, NSW Sokoloff L, Reivich M, Kennedy C, et al. The [14C] deoxyglucose method of local cerebral glucose utilisation: Theory, procedure, and normal values in the conscious and anaethestised albino rat. J Neurochem 1977; 28: 897-916. Som P, Atkins HL, Bandoypadhyay D, et al. A fluorinated glucose analog, 2-fluoro-2-deoxy-D-glucose (F-18): non-toxic tracer for rapid tumor detection. J Nucl Med 1980; 21: 670-675. Proceedings of the 46th Annual Meeting of the Society of Nuclear Medicine, Los Angeles, California, 1999. J Nucl Med 1999; 40 (Suppl): 5. Commonwealth Department of Health and Aged Care (Diagnostics and Technology Branch). Review of positron emission tomography (PET). Canberra: The Department, July 1999. Valk PE. Effect of FDG-PET on patient management and cost. Handout book. Reston, Va (USA): Society of Nuclear Medicine, 1999: 200-204. Hicks RJ, Binns DS, Fawcett ME, et al. Positron emission tomography (PET): experience with a large-field-of-view three-dimensional PET scanner. Med J Aust 1999; 171: 529-532. Valk PE, Pounds TR, Tesar TD, et al. Cost-effectiveness of PET in clinical oncology. Nucl Med Biol 1996; 23: 737-743. Lowe VJ, Fletcher JW, Gobar L, et al. Prospective evaluation of positron emission tomography in lung nodules. J Clin Oncol 1998; 16: 1075-1084. Gambhir SS, Hoh CK, et al. Decision tree sensitivity analysis for cost-effectiveness of FDG-PET in the staging and management of non-small-cell lung carcinoma. J Nucl Med 1996; 37: 1428-1436. HCFA expands Medicare coverage of PET. J Nucl Med 1999; 50: 23N. Bailey DL, Miller MP, Spinks TJ, et al. Experience with fully 3D PET and implications for future high-resolution 3D tomographs. Phys Med Biol 1998; 43: 777-786. Hutton B. Emerging clinical applications of quantitative emission computed tomography. In: Pham B, Braun M, Maeder AJ, Eckert MP, editors. New approaches in medical image analysis, 1999. Proceedings of SPIE (International Society of Optical Engineering) 1999; 3747: 57-76. (ISBN 0-8194-3229-6.) 1: History of positron emission tomography in Australia In Australia, PET was first used in 1992 at the Royal Prince Alfred Hospital (RPAH), Sydney, and the Austin Hospital, in Melbourne (now the Austin and Repatriation Medical Centre). In 1993, the Federal Government, in conjunction with the New South Wales and Victorian governments, funded a five-year evaluation project involving PET units at these two sites. The intention was to undertake a rigorous evaluation of the effectiveness of PET as a diagnostic tool in Australian clinical practice and provide information on which to base decisions about future resource allocation. This project had difficulties in establishing appropriate protocols, and in October 1997 a modified evaluation strategy was adopted. This involved limited Medical Benefits Schedule (MBS) funding of the RPAH and Austin PET centres to develop evidence on the clinical role, value and cost effectiveness of PET. MBS reimbursement for oncology covered breast, gastrointestinal, genitourinary, head and neck, haematological, hepatobiliary, soft tissue and thoracic cancer. In 1996 and 1998, respectively, unfunded PET centres began operating at the Peter MacCallum Cancer Institute in Melbourne and the Wesley Hospital in Brisbane. Back to text2: Uses of positron emission tomography (PET) in oncology4 1. Primary tumour diagnosis Solitary pulmonary nodule Unknown primary tumour 2. Primary tumour staging Non-small-cell lung cancer Hodgkin's disease Breast cancer Oesophageal cancer 3. Recurrent tumour - diagnosis and staging Recurrent colorectal cancer Metastatic melanoma Recurrent head and neck cancer Other tumours: lymphoma, ovarian cancer, breast cancer, non-small-cell lung cancer 4. Treatment evaluation (surgical, chemotherapy, radiation therapy) Non-small-cell lung cancer Non-Hodgkin's lymphoma Recurrent head and neck cancer Hepatic recurrence of colorectal cancer Metastatic breast cancer Adapted from Society of Nuclear Medicine 1999 Handout Book, June 1999. Back to text

John G Morris

Diagnosis Positron emission tomography (PET): experience with a large-field-of-view three-dimensional PET scanner Positron emission tomography (PET) using fluorine-18 fluorodeoxyglucose (FDG) is an accurate technique for staging and therapeutic monitoring in oncology. We evaluated our use of FDG PET in an oncology centre after our first 2500 studies, and summarise our experience of PET for the major referral indications. Optimised for clinical service, PET offers lower scanning costs and therefore improved cost-effectiveness. Rodney J Hicks, David S Binns, Meagan E Fawcett, Robert E Ware, Victor Kalff, Allan F McKenzie, John P Zalcberg and Lester J Peters MJA 1999; 171: 529-532 For related articles see Morris Introduction - Major referral indications - Providing cost-effective PET in oncology - Conclusion - Acknowledgements - References - Authors' details - - More articles on Oncology Introduction There is increasing recognition of the role of positron emission tomography (PET) in oncology, supplementing its established roles in the evaluation of myocardial viability1 and epilepsy.2 The radiopharmaceutical fluorine-18 fluorodeoxyglucose (FDG), an analogue of glucose, has high uptake in a wide range of tumours. FDG PET has been shown to be an accurate technique for tumour staging3 and for therapeutic monitoring.4However, the high establishment and operating costs of conventional PET facilities make economic justification more difficult than for cheaper imaging methods. Before funding new technologies, government and third-party payers increasingly require evidence of cost-effectiveness as well as diagnostic accuracy. High unit scanning costs demand substantially greater effectiveness. The development of lower-cost positron imaging systems over the past 10 years offers a realistic opportunity to expand the clinical availability of PET by improving this balance. Our PET facility, which uses such a system and is optimised for clinical service provision, was commissioned in September 1996. Our aim was to summarise our experience of PET scanning after the first 2500 FDG PET studies performed at our centre (Figure 1). Major referral indications Lung cancer The most common referral indication was known or suspected lung cancer -- 522 studies (20%). Our preliminary, prospective evaluation of the impact of PET scans on the management of 47 consecutive lung cancer patients found that, in over 60% of cases, management was significantly influenced by the scan result (Figure 2).16 In particular, in 32% of patients being considered for treatment with curative intent, management was changed to palliative therapy after documentation of previously unrecognised extrathoracic disease. In a more recent review limited to 140 consecutive patients being staged before definitive treatment of non-small-cell lung cancer, we found that PET changed treatment intent or modality in 47 patients (33%), altered delivery of the intended therapy in 39 patients (28%), and confirmed that the intended therapy was appropriate in only 39 patients (28%). In 15 patients (11%), an abnormality on PET, subsequently confirmed at follow-up, was ignored, with adverse outcome in all but two patients. Melanoma Assessing resectability of clinically localised metastatic disease in melanoma and staging of high-risk primary malignant melanoma have also been frequent indications -- 386 studies (14%) (possibly reflecting the presence of a dedicated melanoma unit at our institution). Based on data suggesting excellent accuracy of FDG PET for staging malignant melanoma,8 we have recently performed a comparison of this technique with high-dose gallium-67 (Ga-67) scanning, previously the standard functional imaging technique for high risk melanoma at our institution. This prospective comparison in 108 patients demonstrated concordance between these investigations in over 80% of cases.17 However, in a limited number of patients, absence of gallium avidity in metastatic melanoma deposits led to striking discordance between FDG PET and Ga-67 results (Figure 3). As FDG PET is a much more convenient study for patients, being completed in less than three hours (whereas Ga-67 requires scanning up to a week after injection), we have largely replaced Ga-67 scanning with PET for this indication at our institution. Head and neck cancer After surgery or radiotherapy, normal tissue planes can be scarred and disrupted. This complicates structural imaging or clinical examination for evaluation of residual or recurrent head and neck cancer. Thus, restaging was the most frequent indication for PET among the 318 studies (13%) in this subgroup. A preliminary review of our experience with FDG PET in 72 patients with head and neck cancer showed a positive predictive accuracy of 92% and a negative predictive accuracy of 100% for restaging.18Gastrointestinal malignancy Referrals of patients with gastrointestinal malignancy (251 studies --10%) have mainly been after primary management. Confirmation of resectability of apparently localised metastatic disease or suitability for local radiotherapy have been the major clinical indications. In a preliminary review of our experience in 41 patients,19 PET altered clinical management in 21 patients (52%), including 14 patients whose management was converted from aggressive locoregional therapy to palliative treatment based on demonstration of previously occult metastases (Figure 4). Breast cancer Evaluation of suspected recurrent or residual disease after treatment of breast cancer (190 studies -- 8%) has been the most common reason for referral of patients with breast cancer. Epilepsy Of the non-oncological indications, localisation of epileptogenic foci of complex partial seizures (88 studies -- 4%) was the most common reason for FDG PET studies. A recent review of our experience in epilepsy involving 52 patients demonstrated a sensitivity for localisation of a seizure focus of 83% versus only 49% by volumetric magnetic resonance imaging (MRI) in the same patient cohort.20 Of 20 patients with localising PET studies who have undergone surgery, 18 are currently seizure free and the other two have had a single seizure associated with drug withdrawal (unpublished data). Providing cost-effective PET in oncology Cost-effectiveness Despite an increasing body of evidence supporting the accuracy of FDG PET in oncology,5-7,9-15,21 its high cost and limited cost-effectiveness data have militated against funding for routine clinical use. In the United States, FDG PET scanning has been shown to be a cost-effective alternative to conventional diagnostic methods of assessing solitary pulmonary nodules22 and staging non-small-cell lung cancer,23 and now attracts reimbursement for these indications. US government funding of PET scans has recently been extended (on the basis of as yet unpublished cost-effectiveness analyses) to evaluation of suspected recurrent colorectal cancer and staging of melanoma and lymphoma (in place of high-dose gallium-67 scanning). In the United Kingdom, because of a reduction in surgical procedures, cost-effectiveness of PET for lung cancer staging has been reported.24 Our own preliminary data suggest a significant management impact of PET on lung cancer.16 Cost-benefit analyses to justify the use of FDG PET have shown significant savings even when based on costs derived from conventional PET facilities (quoted at US$1200, which includes technical and reporting costs).23 The ultimate cost of clinical PET scans depends on throughput of patients, availability and cost of radiopharmaceutical supplies, and the case mix of PET studies. Further evolution of lower-cost positron imaging devices and the development of production and distribution facilities to supply FDG to sites remote from a cyclotron have the potential to further reduce costs. If the cost of PET scans becomes more competitive, the merit of funding of PET for clinical use could be argued not on the basis of cost, but on its proven diagnostic and prognostic accuracy compared with standard investigations. A clinical service model As clinical service provision has been the major focus at our facility, our equipment and staffing reflect this orientation. Most other PET centres have focused on performing basic research as well as clinical studies. The capital and human establishment costs needed to perform the complex investigative studies that advance and validate PET technology increase the overall operational costs of such centres. The PET scanner at Peter McCallum Cancer Institute (GE Quest-300H, UGM Medical Systems Inc, Philadelphia, Pennsylvania, USA) uses scintillation crystals similar to those used in standard nuclear medicine gamma cameras. This significantly reduces the purchase price compared with conventional PET scanners. However, the documented spatial resolution and sensitivity are similar to current generation three-dimensional PET scanners.25 The larger axial field-of-view (25 cm v 16 cm) allows higher patient throughput. For example, whole-body imaging studies can be completed in less than an hour. The scanner characteristics limit administered radioacitivity to around 111 MBq (3 mCi) of F-18 FDG, compared with the typical dose of 300-555 MBq (8-15 mCi) with conventional PET scanners. This reduces operating costs, but limits the potential use of more short-lived PET tracers. Unlike amortisation costs, which fall, radioisotope costs increase as the number of patients studied per day increases. Because of radioactive decay, patients studied late in the day require far more isotope to be dispensed at the time of production. Decay also occurs during transport, and therefore proximity of the end-user to the cyclotron also influences daily isotope requirements and costs. A more sensitive scanner has particular advantages when used at a site remote from the production cyclotron. We believe that the relatively low start-up costs, high throughput and reduced operating expenses enable our facility to offer clinical PET studies at a cost that is significantly less than that generally quoted in the literature. Even with the lower cost of our model of practice, PET is likely to remain more expensive than other tomographic diagnostic procedures commonly used in cancer staging and therapeutic monitoring. However, the relatively poor diagnostic accuracy of these tests, when used alone, means that multiple investigations are often used or tests are supplemented by invasive staging procedures, making overall costs considerably higher. The advantages to patient quality of life of more accurate staging, particularly that which spares futile surgical intervention or reduces patient anxiety by timely assessment of therapeutic response, although more difficult to express in economic terms, can not be underestimated. Conclusion PET scanning has been available at our institution for three years. During this time it has been readily adopted by clinicians for planning of cancer management and for therapeutic monitoring. The high proportion of referrals from outside our institution suggests that there is growing awareness and high clinical acceptance of this technology among the medical community. Our own preliminary data support its utility in a wide range of oncological settings. More detailed prospective evaluation of the diagnostic accuracy and impact of PET in our institution is in progress and will help to further define the role of F-18 FDG PET in clinical oncology. By reducing scanning costs, the model of practice proposed offers the potential for PET to be become more widely available to the Australian community as a clinical rather than a research investigation. Acknowledgements Thanks to the staff of the cyclotron facilities at the Austin and Repatriation Medical Centre and the National Medical Cyclotron for providing timely supply of isotopes for clinical studies. We also thank the staff of our Department of Nuclear Medicine for taking on significantly increased work-loads with only a minimal increase in staffing levels, and Dr John Morris, the Chief Executive Officer of Peter McCallum Cancer Institute, for his ongoing support of the PET Program. References Tillisch J, Brunken R, Marshall R, et al. Reversibility of cardiac wall-motion abnormalities predicted by positron tomography. N Engl J Med 1986; 314: 884-888. Engel J, Henry TR, Risinger MW, et al. Presurgical evaluation for partial epilepsy: relative contributions of chronic depth-electrode recordings versus FDG-PET and scalp-sphenoidal ictal EEG. Neurology 1990; 40: 1670-1677. Rigo P, Paulus P, Kaschten BJ, et al. Oncological applications of positron emission tomography with fluorine-18 fluorodeoxyglucose. Eur J Nucl Med 1996; 23: 1641-1674. Price P, Jones T. Can positron emission tomography (PET) be used to detect subclinical response to cancer therapy? Eur J Cancer 1995; 31A: 1924-1927. Valk P, Pounds TR, Hopkins DM, et al. Staging non-small cell lung cancer by whole-body positron emission tomographic imaging. Ann Thorac Surg 1995; 60: 1573-1582. Coleman RE. PET in lung cancer. J Nucl Med 1999; 40: 814-820. Steinert HC, Hauser M, Allemann F, et al. Non-small cell lung cancer: nodal staging with FDG PET versus CT with correlative lymph node sampling. Radiology 1997; 202: 441-446. Steinert HC, Huch BÖni RA, Buck A, et al. Malignant melanoma: staging with whole-body positron emission tomography and 2-[F-18]-fluoro-2-deoxy-D-glucose. Radiology 1995; 195: 705-709. Delbeke D. Oncological application of FDG PET imaging; brain tumors, colorectal cancer, lymphoma and melanoma. J Nucl Med 1999; 40: 591-603. Benchaou M, Lehman W, Slosman DO, et al. The role of FDG-PET in the preoperative assessment of N-staging in head and neck cancer. Acta Otolaryngol 1996; 116: 332-335. Newman JS, Francis IR, Kaminski MS, Wahl RL. Imaging of lymphoma with PET with 2-[F-18]-fluoro-2-deoxy-D-glucose: correlation with CT. Radiology 1994; 190: 111-116. Hoh CK, Glaspy J, Rosen P, et al. Whole-body FDG-PET imaging for staging of Hodgkin's disease and lymphoma. J Nucl Med 1997; 38: 343-348. Tse NY, Hoh CK, Hawkins RA, et al. The application of positron emission tomographic imaging with fluorodeoxyglucose to the evaluation of breast disease. Ann Surg 1992; 216: 27-34. Utech CI, Young CS, Winter PF. Prospective evaluation of fluorine-18 fluorodeoxyglucose positron emission tomography in breast cancer for staging of the axilla related to surgery and immunocytochemistry. Eur J Nucl Med 1996; 23: 1588-1593. Avril N, Bense S, Ziegler SI, et al. Breast imaging with fluorine-18-FDG PET: quantitative image analysis. J Nucl Med 1997; 38: 1186-1191. Kalff V, Hicks RJ, MacManus M, et al. The clinical impact of PET scanning in patients with lung cancer: a prospective study [Abstract]. J Nucl Med 1998; 39: 249. Hicks RJ, Kalff V, Binns DS, et al. Are high dose Ga-67 scans as good as FDG PET in staging melanoma? [Abstract] J Nucl Med 1998; 39: 11. Porceddu S, Hicks RJ, Rischin D, Peters L. Impact of PET on head and neck cancer. Proceedings of the 49th Annual Scientific Meeting of the Royal Australasian College of Radiology, 1998 Oct; Brisbane, Australia. [Abstract]: 290-291. Kalff V, Binns DS, Fawcett ME, Hicks RJ. Clinical impact of FDG PET scanning in colon cancer: a prospective study. Nucl Med Commun 1999; 20: 381. Murphy M, O'Brien TJ, Hicks RJ, et al. Experience of a 3-D, large-field-of-view PET scanner for the localisation of partial epilepsy. Annual Meeting of the American Epilepsy Society, 1999. Epilepsia (Suppl). In press. Strauss LG, Conti PS. The applications of PET in oncology. J Nucl Med 1991; 32: 623-648. Dewan NA, Reeb SD, Gupta N, et al. PET-FDG imaging and transthoracic needle lung aspiration biopsy in evaluation of pulmonary lesions: a comparative risk- benefit analysis. Chest 1995; 108: 441-446. Gambhir SS, Hoh CK, Phelps ME, et al. Decision tree sensitivity analysis for cost-effectiveness of FDG-PET in the staging and management of non-small-cell lung carcinoma. J Nucl Med 1996; 37: 1428-1436. Lewis P, Griffin S, Marsden P, et al. Whole body 18F-fluorodeoxyglucose positron emission tomography in preoperative evaluation of lung cancer. Lancet 1994; 344: 1265-1266. Karp JS, Muehllehner G, Mankoff DA, et al. Continuous-slice PENN-PET: a positron tomograph with volume imaging capability. J Nucl Med 1990; 31: 617-627. (Received 1 Jul, accepted 5 Oct, 1999) Authors' details The Peter MacCallum Cancer Institute, Melbourne, VIC. Rodney J Hicks, MD, FRACP, Director of Diagnostic Imaging. David S Binns, DipAppSci, ANMT, Chief Nuclear Medicine Technologist. Meagan E Fawcett, BAppSci, ANMT, Nuclear Medicine Technologist. Robert E Ware, MB BS, FCP(South Africa), Honorary Physician in Nuclear Medicine; currently, Director, Hobart Isotope Imaging, Hobart, TAS. Victor Kalff, MB BS, FRACP, Honorary Physician in Nuclear Medicine; currently, Deputy Director, Alfred Hospital, Melbourne, VIC. Allan F McKenzie, MB BS, FRACR, Director of Radiology. John R Zalcberg, PhD, FRACP, Professor, and Director of Medical Oncology. Lester J Peters, MD, FRACR, Professor, and Director of Radiation Oncology. Reprints: Dr R J Hicks, Director of Diagnostic Imaging, Peter MacCallum Cancer Institute, Locked Bag 1, A'Beckett Street, Melbourne, VIC 3000. rhicksATpetermac.unimelb.edu.au The primary referral diagnosis was that prospectively assigned at the time of the PET study. Oncological indications were grouped by system (eg, "lung cancer" primarily comprised patients with non-small-cell tumours, but also included small-cell lung cancer and solitary pulmonary nodule). All imaging performed on a given day was counted as a single study. Almost all oncology patients had screening of areas remote from the known or suspected site of the primary tumour. The referral indication was known or suspected cancer in 2390/2500 cases (95.6%); 1881 individual patients had scans, with 619 follow-up studies in 379 patients for therapeutic monitoring or restaging after treatment. More than a third of our referrals were from clinicians without formal affiliation with our institution, including referrals from all States and Territories. The main referral indications at our facility reflect those cancers which have been shown to be accurately evaluated by PET - including lung cancer,5-7 melanoma,8,9 head and neck cancer,10 colorectal cancer,9 lymphoma,9,11,12 and breast cancer.13-15Back to text Figure 2. Non-small-cell lung cancer was diagnosed at bronchoscopy. A computed tomography scan was equivocal for mediastinal disease. FDG PET scanning of the thorax and abdomen was performed to assess suitability for surgical resection. The primary tumour (large arrows) and mediastinal nodal metastasis (small arrows) are clearly identified in transaxial (upper panel), sagittal (middle panel) and coronal (lower panel) projections. The patient was offered chemoradiotherapy rather than surgery. Back to text Figure 3: After resection of a melanoma of the left cheek, a palpable lymph node was shown (by fine-needle aspiration biopsy) to be a metastasis. A high-dose gallium-67 scan (left panel), which included tomographic imaging, did not demonstrate significant abnormality, so radical neck dissection was considered. Whole body FDG PET scanning (right panel) demonstrated disseminated metastases (arrows). The patient was spared unnecessary neck dissection and offered systemic therapy. Back to text Figure 4. After resection of a Duke C rectal carcinoma, this patient had rising carcinoembryonic antigen (CEA) levels and an apparently solitary hepatic metastasis (found at laparotomy - large arrow in left panel). At operation, a small bowel obstruction was thought to be related to postsurgical adhesions. FDG PET, performed to evaluate the patient's suitability for subsequent hepatic resection (right panel), demonstrated previously unrecognised widespread metastases (small arrows) and the patient was spared further futile surgery. Chemotherapy was commenced. Back to text

Rodney J Hicks · David S Binns · Meagan E Fawcett · Robert E Ware · Victor Kalff · Allan F McKenzie · John P Zalcberg · Lester J Peters