Cardiovascular diseases

In reply: In the interests of scientific exactitude, I am indebted to Marilyn Mann for corrections regarding the Simvastatin and Ezetimibe in Aortic Stenosis (SEAS) trial. However, as the Ezetimibe and Simvastatin in Hypercholesterolemia Enhances Atherosclerosis Regression (ENHANCE) trial did not include the control group ezetimibe versus placebo, the effects of ezetimibe on carotid intima media thickness remain somewhat speculative.1

Ian R Hamilton-Craig

What happened when a doctor on duty in the emergency department suddenly made the transition to “patient in Resus 2”? “I am not sure exactly why you would want to put your coronary arteries through a full day of paediatric anaesthetics ...” a colleague had written in an email I was perusing in the emergency department on the morning of New Years Day. Suddenly, and unrelated to the contents of the email, I developed extreme anxiety. Recalling that, years ago, frequent irregular ectopic heartbeats had created similar, less severe anxiety, I felt my pulse, expecting to find it irregular. My pulse was regular, but the rate was 120 beats/min. I walked to the tearoom, hoping that whatever it was would go away. Although I had no chest pain, the short walk convinced me I could not ignore the symptoms. I told a colleague I thought I had SVT. In seconds, I made the transition from “doctor on duty in the emergency department” to “patient in Resus 2”. Abbreviations Cath lab Catheterisation laboratory CCU Coronary care unit ECG Electrocardiogram IV Intravenous SVT Supraventricular tachycardia The anxiety worsened, which I attributed to my changed status. Unhelpfully, my ECG showed a left bundle branch block. I began to sweat, and developed mild chest discomfort. I also began to show signs of being a difficult patient. I insisted on having local anaesthetic for my IV line insertion (which is something I do for all my patients). My anxiety was far worse than the pain. I wanted midazolam, but received 5 mg morphine. We compromised about the benzodiazepines and I accepted 2.5 mg diazepam orally, though I knew that this would take 30 minutes to work. I took aspirin, glyceryl trinitrate and later enoxaparin without complaint. The oxygen mask caused claustrophobia and I insisted on nasal prongs. But as my discomfort and restlessness increased, the mask was put back in place — whether I liked it or not. I have seen many an anxious, restless, sweaty patient pull away at their oxygen mask shortly before their cardiac arrest and death, and it began to dawn on me that I might have something nasty going on. But the prospect of sudden death was not my main concern — what I wanted was treatment for my anxiety, which was far worse than my chest discomfort. Mobile phones are wonderful things. My brother chose this moment to phone from New Zealand to ask me if he should double his dose of sotalol to control his paroxysmal atrial fibrillation. (I’m not sure what advice I gave him, but I gather he is still alive.) Then my wife arrived, and best of all, she wheedled a further 5 mg of diazepam for me from some kind soul. The ambulance transfer to a city hospital was a bit of a blur — and that was just the way I wanted it. The city hospital was familiar, as 2 years earlier I had organised an attachment to see what went on in its cardiac catheter laboratory. My cardiologist wanted me in there again. I was not thrilled, but I felt I was on an unstoppable train. I was not particularly interested in hearing about the bad things that could happen there (which included death), and I signed the consent form without reading it. What I really wanted to know was how much midazolam I could have during the procedure. I was assured I could have “some”. A wardsman came to shave an area around the femoral artery and beyond. This also did not appeal. Being allowed midazolam before the shave seemed unlikely, so I closed my eyes, lay back and “thought of England”. Later, a sheepish and apologetic wardsman returned to tell me he had shaved the wrong side. (It was not as if he had chopped off the wrong leg, as some doctors have done.) More thinking of England. A nurse popped in during the procedure. I was to go to the cath lab immediately, and she had 5 mg diazepam for me. I tried to point out that, if I was going immediately, the diazepam wouldn’t have time to work. But I swallowed it anyway, bringing the total diazepam for the day to 12.5 mg. And I was counting. A small crowd had gathered at the cath lab, including a radiographer I had worked with elsewhere. Chlorhexidine in alcohol was used to paint my groin. The nurse doing the painting warned me that I would feel a burning pain if any of the solution reached sensitive parts of my nether regions. She was right! And she was ready! A syringe of sterile water was squirted on the affected area to relieve the burning. (I categorically deny that my continued complaints about the burning were because I liked having a woman squirt sterile water there — although I have considered having the procedure repeated, just in case.) After I had been draped — which restored some dignity to the proceedings — my cardiologist began to infiltrate local anaesthetic. I reminded him about the midazolam. He assumed that I was unhappy about the pain from the local, but I just wanted to be “out of it” while tubes were being put inside my heart. A milligram of midazolam and 25 μg of fentanyl were given intravenously. The fentanyl made me vomit. The benzodiazepines were working better than I thought, as the nausea and vomiting were not unpleasant — just embarrassing in front of an audience. And then I was dreaming. Not of angels greeting me at the pearly gates, not of a welcome from 30 naked virgins — my dream was all about shoes. As the beginnings of consciousness materialised, I seemed to be in two places at once. In one, there was this thing going on with the shoes; in the other, I was lying flat on my back surrounded by machines and people dressed in blue, one of whom was mucking around in my groin. It was a very strange dream indeed. I tried to will myself awake to get the business with the shoes sorted out. But as I awoke, it was the shoes that faded away and the people in blue who became more real. And I noticed that I had a fat lower lip. What had happened was this. The catheter placed in my femoral artery was fed into my left ventricle. Here, tickling an already irritable myocardium, it provoked ventricular fibrillation. During the 40 seconds it took to organise the defibrillator to shock me (Box 1), I presumably did what I have seen patients do as they have a cardiac arrest — I went into spasm similar to the tonic stage of an epileptic seizure. The seizure caused me to bite my lower lip. In 2004, in the columns of the MJA, I had indicated that if I had a cardiac arrest close to a defibrillator, I expected to survive.1 Nice to be proved right. The angiogram showed a single blockage of a coronary artery at a bifurcation (Box 2, A), which I was told might make stent placement difficult. I was also told I could expect to experience some (presumably ischaemic) pain during the procedure. My benzodiazepine level was such that I accepted this information with equanimity. I dozed as they fiddled; the procedure appeared to be a success (Box 2, B), and soon after nightfall I was back in the coronary care unit. I had received aspirin, clopidogrel, enoxaparin, heparin and abciximab, all to stop my blood clotting. A large hole had then been made in my femoral artery, from which I slowly and steadily bled. I was instructed to lie flat and not move my right leg. After replacing the saturated pressure dressing, the nurse applied pressure to the area for 10 minutes. Blood collected in various places. Those with knowledge of anatomy can imagine where these places were. For those without, they were around where the nurse had been squirting sterile water earlier in the evening. Cleaning up had to be done without my moving much. Periodically my bits and pieces got in the way and had to be flicked aside. (I closed my eyes, but I was no longer capable of thinking of England.) This cycle was repeated every hour for 8 hours. Finally, at 4 am, the bleeding stopped. I have never before had a nurse pay so much sustained attention to my groin — not even when I was a junior doctor. As day broke, I noted that I was alive. The previous day’s medication had worn off and I was fully awake. The groin pain was not sufficient to need treatment. All things considered, I was feeling pretty good. But that was easily fixed — I was given drugs: atenolol to slow the heart beat without actually stopping it; perindopril to lower the blood pressure while still keeping it measurable (a chronic cough allows the cardiologist to know that blood is still flowing); aspirin and clopidogrel (cardiologists feel sorry for gastroenterologists and try to add rare excitement to their lives — a patient with a dodgy heart and a stomach full of blood will do the trick); rosuvastatin (cardiologists favour functioning cardiac muscle over functioning skeletal muscle); and isosorbide mononitrate (the highlight of my hospital stay!). After swallowing the handful of pills, I was expecting indigestion, but what I got was a headache. I soon became a clock-watcher. I ensured I was given paracetamol every 4 hours. The headache sometimes became just bearable, but would not go away. I became aware of every beep, bang and noise in and around the CCU — and there were many. (Why is it that when I am giving anaesthetics, I can set the monitor alarms so that they go off only rarely, they mean something when they do, and I take notice of them — but when I’m in other hospital environments, alarms go off all the time, they mean nothing, and no one takes any notice?) The patient in the bed opposite commiserated — the racket coming from the nursing station at the change of shift reminded him of the noise made when he fed his raucous chooks. The next morning, full of nausea and headache, and with my heart being monitored by telemetry, I was moved from the CCU to an empty double room. Having a shower was my next big adventure. I had my left hand in a glove covering my IV line, I couldn’t flex my right leg at the hip, and I had ECG leads and wires going to a battery-operated telemetry box. I couldn’t find any soap. As I was not exactly dressed to go out looking for a nurse, I pressed a call button. I didn’t have my glasses on, and the button turned out to be for “emergency assistance”. Anyway, I was told that the only soap was in the liquid soap dispenser above the sink, some distance away from where I was supposed to be having a shower. A series of prolonged, complex, awkward and repetitive manoeuvres were required to get soap and water together where they were needed (which was pretty well everywhere). Sadly, there is no video clip of this dance. As the headache from my second isosorbide tablet was kicking in, a middle-aged man with chest pain was put in the next bed. I had to be careful what I said to him. He spent a good part of the time talking on his mobile phone. He was certain that his pain had nothing to do with his heart and said he was only there to keep his wife happy. I did not share his certainty as, through the curtains, I overheard him telling his story to his doctor. Later, he told me his pain was returning. I suggested that he press the call button, but it became evident that he was not going to do this. So I told him I needed some more paracetamol for my headache, and pressed my call button. Soon the nurses were taking ECGs and giving him glyceryl trinitrate. As night fell, I plotted how to organise sleep. Nothing seemed likely to happen in my room after 9.30 pm. After taking paracetamol and temazepam, I eventually drifted off. But before midnight, a nurse woke me. “Are you all right?” she asked. “No! I have a bloody headache and you have just woken me up”, I thought. “Yes”, I said. “Your heart rate is 140 on the monitor”, she said. The cause turned out to be a loose monitor lead. After this intrusion, I was unable to get back to sleep. I knew I could not have any more drugs until 1.30 am, so I asked if I could get out of bed. After wandering around in circles for a while, I came across a large bank of screens showing CCU and ward telemetry tracings. There were heart blocks, bradycardias, tachycardias, paced rhythms and evolving infarcts. I noted that I was alive and ticking, although that seemed a mixed blessing at the time. But what was my trace doing up there along with all the others? Were the gods punishing me for something I had written more than 20 years ago?2 So, what did I learn from being “on the other side”? Doctors are not immune from illness. Hospitals are dangerous places — avoid them if possible. Selected patients should be given benzodiazepines the moment they walk through the door. Those responsible for unnecessary hospital noise should be taken out and shot. And, after considering the alternatives, I can highly recommend ventricular fibrillation as a way of leaving this world. 1 Electrocardiogram showing cardioversion of ventricular fibrillation 2 Angiograms taken before and after placement of a stent A: Coronary arteries before the procedure, showing complete occlusion of the left anterior descending coronary artery. B: Coronary arteries after angioplasty and placement of a stent.

Michael J Mackay MB ChB, MHA, FACRRM

There is still room for improvement, both in decreasing delays in, and deciding who is eligible for, reperfusion therapy In patients with acute ST-segment-elevation myocardial infarction (STEMI), early coronary reperfusion — within 1 to 2 hours of symptom onset — by either thrombolysis or primary percutaneous coronary intervention (PCI) reduces the mortality rate by half. However, this benefit quickly dissipates with further delay in treatment.1 As “time is muscle”, it is the time from symptom onset to reperfusion (or total ischaemic time), rather than the mode of reperfusion, that is the critical determinant of outcome. Hence the imperative to minimise: (i) delay by patients in recognising symptoms as possible myocardial infarction (MI) and seeking medical help; (ii) delay in ambulances responding to calls; (iii) delays in diagnosing STEMI on first medical contact; and (iv) omissions or delays in administering the most appropriate means of reperfusion in eligible patients. In this issue of the Journal, Huynh and colleagues, using data from a prospective Australian registry, report on processes of care and outcomes of 755 patients presenting with suspected STEMI.2 There is good and bad news in this report. The good news is that the median time from symptom onset to first medical contact in this cohort was 105 minutes (1.75 hours) compared with 3.2 hours for patients with undifferentiated chest pain, reported in 2005.3 If the sample in the study by Huynh and colleagues2 is representative of most patients with MI, this suggests that public recognition of warning symptoms and the need to seek medical help urgently has improved over the past 15 years in response to public education campaigns that target individuals at high risk and behavioural barriers to action.4 Reperfusion reduced mortality at 12 months (adjusted for baseline risk as calculated using the Global Registry of Acute Coronary Events [GRACE] risk score) by 65% (and by 78% if administered in a timely fashion), similar to results noted in recent overseas observational studies that used similar risk-adjustment methods.5 Finally, there was no difference between metropolitan and rural patients in the time to presentation or the proportions of patients who received reperfusion therapy, or who received it in a timely manner, and the same applied to inhospital and 12-month mortality rates. This suggests the “city–bush” gap in coronary care noted in past studies6 is being closed, although more rural patients (74%) received thrombolysis, while more metropolitan patients (68%) received primary PCI. The bad news is that one in three patients did not receive any form of reperfusion — a figure common to other countries and which has proven resistant to change.7 Unfortunately, contraindications to either form of reperfusion in individual patients were not reported, but contraindications and patient refusal have been reported to account for no more than 10% of all patients with STEMI.8 This means just over one in five patients were likely to have been eligible for reperfusion therapy but failed to receive it. Factors associated with not receiving reperfusion therapy on regression analysis included a past history of diabetes or documented coronary stenoses on angiography, acute pulmonary oedema on presentation, left bundle branch block on electrocardiogram (ECG), and a non-cardiologist as the treating doctor. In other studies of patients eligible for reperfusion therapy, additional factors have included older age, admission to a facility not capable of performing PCI, increasing time to presentation, renal insufficiency, prior stroke or coronary artery bypass grafting, being female, and presentation without chest pain or with an equivocal ECG.5,7 Some of these associations reflect diagnostic uncertainty in patients with atypical clinical presentations and non-diagnostic ECGs or clinician concern about the risk of bleeding in older patients (especially underweight women) and those with renal failure or prior stroke. However, registry data show that in this patient group at relatively high risk, early reperfusion therapy compared with no reperfusion reduces inhospital mortality by 38%, with primary PCI being more effective than thrombolysis.9 Clinicians may need to recalibrate their perceptions of benefit and risk in groups of patients who have often been excluded from clinical trials. The other bad news is that among patients receiving reperfusion therapy in the study by Huynh and colleagues (61%, primary PCI; 37%, thrombolysis), only one in three received it within an optimal time frame.2 The median door-to-needle time (D2N) for thrombolysis was 43 minutes (versus a 30-minute standard) and door-to-balloon time (D2B) for primary PCI was 102 minutes (versus a 90-minute standard). These times are longer than those reported in contemporary cohorts in other developed countries, such as 33 minutes D2N and 83 minutes D2B in a Canadian cohort,5 and 30 minutes D2N and 86 minutes D2B in the GRACE international registry.10 Attention has recently shifted to reducing total system delay, defined as the time from first contact with the health care system (ie, ambulance) to initiation of reperfusion therapy, which now appears to be more strongly associated with mortality than patient delay in seeking care.11 In reducing system delay, the timing of PCI (immediate v delayed v rescue) and its relation to thrombolysis in patients presenting to non-PCI-capable hospitals becomes a pivotal issue. Current Australian and New Zealand guidelines state that fibrinolysis is preferred to primary PCI in patients presenting within 1 hour of symptom onset unless balloon insufflation can occur within 60 minutes after first medical contact (in most cases, this is patient pick-up by ambulance).12 In patients presenting between 1 and 3 hours after symptom onset, fibrinolysis is preferred unless primary PCI can occur within 90 minutes of first medical contact. Studies show that in patients with symptom onset of less than 3 hours and for whom transfer to PCI-capable hospitals would delay primary PCI for more than 90 minutes, the combination of early lysis and aggressive use of rescue PCI (in the third of patients with persistent ST-segment elevation, cardiogenic shock, severe heart failure or serious ventricular arrhythmias) confers comparable outcomes with that achieved with primary PCI.13 In this regard, prehospital thrombolysis undertaken by ambulance paramedics, combined with early PCI where appropriate, seems to be an underused strategy in reducing system delay.14 Another issue is the role of risk stratification in deciding who should receive which form of reperfusion. A treatment-risk paradox is often seen whereby eligible patients at high absolute risk of death or recurrent MI are less likely to receive reperfusion therapy (for reasons already mentioned) than those at lower risk9 and in whom treatment delays attenuate the absolute benefit of reperfusion to a greater degree. In considering transferring patients presenting within 6 hours of symptom onset for primary PCI, the higher the risk profile, the larger the reduction in mortality benefit with primary PCI compared with thrombolysis for each 10-minute increase in PCI-related time delay.15 Delays must be minimised in high-risk patients, rather than simply working to a 60-minute or 90-minute D2B rule. The equipoint between primary PCI and fibrinolysis (the PCI-related time delay at which primary PCI loses its superiority in terms of mortality benefit compared with fibrinolysis) may be as little as a D2B time of 40 minutes in a high-risk situation (such as a young patient presenting early with a large anterior infarction) versus 179 minutes in lower risk situations (such as an older patient presenting late with a non-anterior infarction).16 Several strategies have been shown in both Australian and overseas studies to be effective in reducing total ischaemic time (Box),17-19 and these need to become mainstream care. This will require a multifaceted approach involving educating both patients and doctors; coordinating ambulance, emergency department and cardiac catheterisation laboratory components of care; establishing integrated networks of non-PCI and PCI-capable hospitals with decision support and transfer processes that take patient risk and time to presentation into account; and ongoing data collection and feedback within clinical registries. Strategies for decreasing delays in reperfusion therapy Hospital-based strategy Potential tools Prehospital ECG and field assessment by paramedics Prehospital ECG policy Guidelines for field assessment with electronic transmission to, and verification of ECG diagnosis by, emergency department staff Prehospital thrombolysis for patients who are within 1 hour of symptom onset Training of paramedics in ECG diagnosis and administration of thrombolytic agents Transfer of PCI-eligible patients direct to a PCI-capable facility Pre-destination protocol for paramedics Rapid assessment and ECG on patients presenting to emergency departments with chest pain Dedicated chest pain cubicles in emergency departments with ECGs taken within 10 minutes of arrival Rapid management of diagnostically uncertain cases Formal order sets for suspected myocardial infarction in cases of initially non-diagnostic ECG Rapid initiation of thrombolysis in eligible patients Formal thrombolysis protocols that can be initiated by emergency department nurses or physicians without consulting the cardiology department Emergency department bypass of PCI-eligible patients with direct transfer to a catheterisation laboratory Prehospital (or first hospital) assessment policy Guidelines for direct activation of the catheterisation laboratory by emergency department staff without review or approval by cardiologists Single-call activation of the catheterisation laboratory team Alert system with single person as contact (senior registrar or consultant) Catheterisation team fully operational within 30 minutes of activation Staff policy and roster Performance of PCI 7 days a week, 24 hours per day Clearance of elective cases; maintained availability of ready-to-go equipment and staff Prompt data feedback Time-entry forms for door-to-needle and door-to-balloon times, and these times notified to all team members after each procedure Team-based approach Team training program; limited handovers with single team approach Regionalised “hub-and-spoke” hospital networks which expedite patient transfer to PCI-capable facility Triage and expedited transfer guidelines for referring and receiving hospitals PCI = percutaneous coronary intervention. ECG=electrocardiogram.

Ian A Scott FRACP, MHA, MEd

Objective: To describe the contemporary management and outcomes of patients presenting with ST-segment-elevation myocardial infarction (STEMI) in Australia.Design, participants and setting: Observational analysis of data for patients who presented with suspected STEMI and enrolled in the Australian Acute Coronary Syndrome Prospective Audit from 1 November 2005 to 31 July 2007.Main outcome measures: Factors associated with use of reperfusion therapy and timely use of reperfusion therapy, and the effects of reperfusion on mortality.Results: In total, 755 patients had suspected STEMI. Median time to presentation was 105 minutes (IQR, 60–235 minutes). Reperfusion therapy was used in 66.9% of patients (505/755), and timely reperfusion therapy in 23.1% (174/755). Thombolysis was administered in 39.2% of those who received reperfusion therapy (198/505), while 60.8% (307/505) received primary percutaneous intervention. Cardiac arrest (OR, 2.83; P = 0.001) and treatment under the auspices of a cardiology unit (OR, 2.14; P = 0.02) were associated with use of reperfusion therapy. A normal electrocardiogram on presentation (OR, 0.42; P = 0.01), left bundle branch block (OR, 0.18; P = 0.001), acute pulmonary oedema (OR, 0.34; P < 0.01), history of diabetes (OR, 0.54; P < 0.01), and previous lesion on angiogram of > 50% (OR, 0.51; P = 0.001) were associated with not using reperfusion. Inhospital mortality was 4.0% (30/755), mortality at 30 days was 4.8% (36/755), and mortality at 1 year was 7.8% (59/755). Receiving reperfusion therapy of any kind was associated with decreased 12-month mortality (hazard ratio [HR], 0.44; 95% CI, 0.25–0.78; P < 0.01). Timely reperfusion was associated with a reduction in mortality of 78% (HR, 0.22; P = 0.04). There were no significant differences in early and late mortality in rural patients compared with metropolitan patients (P = 0.66).Conclusion: Timely reperfusion, not the modality of reperfusion, was associated with significant outcome benefits. Australian use of timely or any reperfusion remains poor and incomplete.

Luan T Huynh MB BS, FRACP · Jamie M Rankin MB BS, FRACP · Phil Tideman FRACP · David B Brieger MB BS, PhD, FRACP · Matthew Erickson BM BS, FRACP · Andrew J Markwick MB BS · Carolyn Astley RB, BN · David J Kelaher BPharm, MSc · Derek P B Chew MB BS, MPH, FRACP

To the Editor: I thank Brieger and colleagues and Thompson for their comments on my article.1 Brieger et al2 refer to the United Kingdom National Institute for Health and Clinical Excellence Guideline Development Group that analysed the non-ST-elevation myocardial infarction (NSTEMI) trials to show an inverse relationship between rates of intervention and clinical outcomes.3 This is a post-hoc analysis and is therefore hypothesis generating rather than proof of concept; further, its results are confounded by the fact that intervention rates have increased in recent years, while medical treatment — particularly with clopidogrel and statins — has also improved. Brieger et al then contend that the Variations in the Application of Cardiac Care in Australia study4 provides justification for an interventional approach. This study demonstrated that patients treated in hospitals without percutaneous coronary intervention (PCI) facilities fare worse than patients at hospitals with PCI. The likely reason for this is that these patients are less likely to receive cardiac rehabilitation, aspirin, statins, β-blockers and timely thrombolysis. With respect to ST-elevation myocardial infarction (STEMI), Brieger and colleagues suggest that the evidence is even stronger because of CARESS-in-AMI (Combined Abciximab Reteplase Stent Study in Acute Myocardial Infarction).5 In this study, patients treated medically were administered half-dose reteplase and full-dose abciximab. This treatment is not currently recommended and the study therefore has no relevance to the current debate. Brieger et al state that I erroneously cited TRANSFER-AMI (Trial of Routine Angioplasty and Stenting after Fibrinolysis to Enhance Reperfusion in Acute Myocardial Infarction)6 as evidence against a strategy of routine urgent transfer. This is the largest contemporary trial that compares urgent transfer with routine selective transfer after thrombolysis in hospitals without PCI facilities. The trial is thus the most relevant to our discussion. The fact that clinicians chose to transfer 89% of patients for PCI after a mean time of 23 hours in the selective management group does not negate the lack of significant benefits of routine urgent transfer after thrombolysis. I agree that TRANSFER-AMI is limited by the high rate of intervention. The problem remains, however, that the guidelines mandate transfer of all STEMI patients and fail to provide evidence to justify it. In the first paragraph of his editorial, Thompson7 cites the American College of Cardiology/American Heart Association guidelines8 as evidence for early coronary intervention. However, these guidelines state that “in initially stabilized UA [unstable angina]/NSTEMI patients, an initial conservative (selective invasive) strategy may be considered as a treatment option”.8 This statement is not in the Australian guidelines. Thompson acknowledges that the ICTUS (Invasive versus Conservative Treatment in Unstable Coronary Syndromes) trial9 provides up-to-date medical treatment, but notes the lower mortality rates compared with patients in the ACACIA (Acute Coronary Syndrome Prospective Audit) registry.10 He therefore concludes that these patients were not high risk. Yet one of the entry criteria for the ICTUS trial was an elevated troponin level, which is defined as high risk in the Australian guidelines. I drew attention to the fact that the Australian guidelines seem to be over-inclusive in their definition of high risk. The lower mortality rate in the ICTUS study and in other routine-versus-selective-PCI trials in NSTEMI presumably reflects exclusion of patients who are older, have other comorbidities such as renal impairment, or whose conditions are clinically unstable. Physicians treating patients such as those entered in the ICTUS trial — with electrocardiogram changes and a raised troponin level without other comorbidities or shock — should have the option to manage these patients with a selective invasive approach. Thompson states “the conclusion that conservative management alone will remove the need to open a blocked artery is hardly justified based on the ICTUS study results, which comprise just over 10% of the evidence base”. I had included two meta-analyses of all of trials of routine intervention as the cornerstone of my argument and pointed out that they had not shown a reduction in mortality. The ICTUS trial confirmed the conclusions drawn from those trials and is the only trial in which medical treatment was adequate. Thompson stated that I suggested that modern medical management could render invasive treatment irrelevant. This statement was not in my article. In summary, nothing in the rebuttal by Brieger et al or Thompson’s editorial provided any real evidence to justify an expensive, invasive policy requiring routine urgent (as opposed to selective) transfer of all patients with ST-elevation acute coronary syndromes or NSTEMI for PCI, nor for the establishment of more PCI units throughout regional Australia.

Brett H Forge

In reply: We thank Forge for his continued and welcome contribution to this debate. He is critical of our reference to the National Institute for Health and Clinical Excellence (NICE) guidelines, which show an inverse relationship between rates of intervention and clinical outcomes. The NICE analysis is supported by others,1 and we stand by our argument that if studies with greater differences in the randomisation arms show a greater treatment effect, the strength of the evidence supporting the intervention is enhanced. Dismissing this type of analysis on scientific grounds is disingenuous and contrary to National Health and Medical Research Council recommendations regarding the assessment and application of scientific evidence.2 Data supporting the role of invasive therapy in management of acute coronary syndromes (ACS) continue to evolve. A recently published patient-level meta-analysis of three studies, comprising 5-year outcomes of 5467 patients, showed a significant reduction in myocardial infarction (hazard ratio [HR], 0.77; 95% CI, 0.65–0.90), with consistent trends towards reduction in cardiovascular death (HR, 0.83; 95% CI, 0.68–1.01) and all death (HR, 0.90; 95% CI, 0.77–1.05).3 This meta-analysis showed the largest absolute benefit among the highest-risk patients assessed by a global risk score, supporting evidence that risk stratification should involve more comprehensive clinical evaluation than that provided by a single biomarker.4 It is true that these studies have not shown a reduction in mortality. This is most likely to be an issue of power. Based on the number of patients studied in randomised trials so far, the power needed to detect a 10% difference in mortality between treatment strategies is only 27%.1 Regarding TRANSFER-AMI (Trial of Routine Angioplasty and Stenting after Fibrinolysis to Enhance Reperfusion in Acute Myocardial Infarction), Forge now agrees that there are limitations associated with the high transfer rates in the conservative arm, which are greater than those seen in contemporary Australian practice. When the conservative arm receives better-than-standard practice, the direct effect is a bias of the outcome towards the null, limiting the quantification of absolute benefit associated with routine urgent transfer. Forge dismisses CARESS-in-AMI (Combined Abciximab Reteplase Stent Study in Acute Myocardial Infarction) because of the management strategy in the conservative arm. The main limitation of this study is its small size; in an earlier study of 16 588 patients, combination half-dose reteplase and full-dose abciximab treatment had comparable mortality to full-dose reteplase alone.5 Finally, we would like to re-emphasise that National Heart Foundation of Australia (NHFA)/Cardiac Society of Australia and New Zealand (CSANZ) ACS guidelines are not prescriptive. They are designed to provide guidance to individual clinical practice, not to mandate one form of practice. Adherence to guidelines has been demonstrated to improve outcomes when instituted as part of a coordinated quality improvement strategy across a health care system. In addition to provision of optimal medical therapy and risk factor modification, approaches to ACS that incorporate timely access to revascularisation have made a significant contribution to this improvement in outcomes.

David B Brieger · Derek Chew · Constantine Aroney · Phil Aylward · Darren Walters · Anne-Maree Kelly · Andrew Boyden

In reply: Forge quotes the recommendation from the United States guidelines to consider conservative treatment after initial stabilisation of non-ST-elevation myocardial infarction.1 This may not the best option for patients in Australian regional hospital conditions. This pathway delivers less optimal outcomes1,2 and should only be considered if a seamless switch to an invasive approach can be made. The primary recommendation from the US guidelines for patients whose conditions are initially stabilised is that “an early invasive strategy . . . is indicated”.1 The benefit of an early invasive approach has been confirmed in the most recent meta-analysis comparing routine with selective invasive approaches.2 Over 5 years, the routine invasive approach achieved a relative reduction in deaths or myocardial infarctions of 19%, and an absolute reduction of 11.1% among highest-risk patients, 3.8% among intermediate-risk patients and 2% among lowest-risk patients. The results of the ICTUS (Invasive versus Conservative Treatment in Unstable Coronary Syndromes) trial3 showed no heterogeneity from the results of the other trials. Identifying patients at risk of complications and most likely to benefit from early intervention remains an ongoing challenge. Increasing experience has confirmed the value of an elevated troponin level as a predictor of increased risk, but also the need to apply some wisdom in interpreting it, with even more challenges with high-sensitivity troponin assays becoming available.4 For patients presenting to a regional hospital with non-ST-elevation acute coronary syndromes, the available evidence favours careful risk stratification and early transfer to an invasive facility for high-risk patients.

Peter L Thompson

To the Editor: Recent articles in the Journal by Forge1 and Thompson2 expressing contradictory opinions on the management of patients with non-ST-elevation acute coronary syndromes (NSTEACS) highlight the difficulties in formulating guidelines for this condition. Thompson argues that the routine invasive approach for patients with high-risk NSTEACS recommended by the National Heart Foundation of Australia (NHFA) guidelines is preferable to the selective invasive approach advocated by Forge. In arguing his case, Forge relies on the results of the ICTUS (Invasive versus Conservative Treatment in Unstable Coronary Syndromes) trial,3 which showed no advantage for the routine invasive approach. Unlike Forge, Thompson believes the ICTUS trial is of limited relevance. In part, he bases this contention on the fact that the 1-year mortality in the ICTUS study was 2.5% compared with 10.5% among Australian patients with non-ST elevation myocardial infarction, indicating that participants in the ICTUS trial were not high-risk patients. This part of his argument is flawed. All patients who entered the ICTUS trial had elevated troponin levels, which according to NHFA guidelines automatically places them in a high-risk category that requires routine angiography. A more likely explanation for the mortality differences is that patients in the ICTUS trial either received better care, or were more compliant with their medical treatment than their Australian counterparts. Thompson rightly argues that the ICTUS trial is but one study comprising “just over 10% of the evidence base” and that the totality of the data favours a routine invasive approach. Conversely, Forge contends that the ICTUS trial is the most contemporary of the studies and the one in which medical treatment most closely concurs with current best practice. Therefore, he argues, the results of this study should take precedence. Faced with these conflicting viewpoints, writers of guidelines have an onerous task, particularly when compliance with guidelines may be used by bureaucrats and administrators to judge physician performance. With respect to the NHFA guidelines for NSTEACS, a case can be made for more flexibility. For example, is it necessary to transfer all stable patients with NSTEACS and borderline troponin elevations from rural hospitals to metropolitan hospitals for angiography even when they may wait many days for such a transfer to occur? Ensuring that such patients receive, and are compliant with, optimal cardioprotective medications would be a greater gain. Guidelines are a valuable aid to clinical practice; they are not necessarily the overriding factor.

Richard W Harper

To the Editor: The recent discussion in the Journal on acute coronary syndromes is welcome. Forge1 questions the validity of the National Heart Foundation of Australia (NHFA) guidelines,2 which recommend universal routine early invasive management for patients with high-risk non-ST-elevation myocardial infarction (NSTEMI), interpreted by Forge as “virtually all patients with objective evidence of ischaemia”. As older patients usually have comorbidities, often major, the influence of NSTEMI on the clinical picture can be uncertain. Transfer from local surroundings and family support is a major decision. It is critical to establish which subgroups gain from early percutaneous coronary intervention, at present not clearly defined. The number needed to treat for substantial benefit is an essential guide to management policy. The logistics of transfer are complex, involving time and distance factors, availability of ambulances and cardiology beds — after acceptance by patient and relatives. Forge validly emphasises the potential loss of intensive cardiac care skills, staff morale and recruitment from the early transfer policy and large costs to the country hospitals from ambulance fees. It is irrefutable that general implementation of the NHFA guidelines would be a major budget item, involving specialised laboratory facilities with nursing, resident and interventional cardiologist availability, 24 hours a day, 7 days a week. Potentially, there are many competing medical financial needs, which a better targeted coronary intervention program could facilitate. Forge raises the issue of conflict of interest in relation to the development of the guidelines. Breiger and colleagues read this as an accusation that they had a pecuniary interest as authors.3 Their stated competing interests list membership of advisory boards, payment for presentations and expert testimony, receipt of research grants, and travel and accommodation expenses. This reflects their role as medical advisors to pharmaceutical and biotechnology companies. What is the aim of publishing competing interests except to provide a guide to significant potential bias affecting the validity of the presentation? Does the editorial staff have criteria for rejection of submissions for publication, or are readers left to make their own judgements? It is undoubted that a major expansion in expensive medical procedures would increase cardiologists’ income, but this is only one component of the program budget. The broader issue is that conflict of interest remains an integral part of personal and professional life, not easily controllable by governance.

John F Niall

To the Editor: We report the use of nebulised frusemide for the symptomatic treatment of end-stage congestive heart failure (CHF). An 84-year-old man with New York Heart Association class IV CHF was referred to the Community Heart Failure Team at St Vincent’s Hospital, Sydney, for ongoing management after a hospital admission for acute pulmonary oedema. His medical history included aortic stenosis, pulmonary hypertension, type 2 diabetes, chronic renal failure, atrial fibrillation, hypertension, chronic obstructive pulmonary disease (COPD) and hypercholesterolaemia. The patient’s medications were home oxygen via a concentrator at 2–4 L/minute; digoxin 62.5 μg three times a week; warfarin 5 mg daily; glyceryl trinitrate 25 mg daily (delivered via a patch); simvastatin 40 mg nightly; spironolactone 12.5 mg daily; frusemide 80 mg orally twice daily (flexible regimen); insulin/isophane (Protaphane; Novo Nordisk) variable dose twice daily; fluticasone 250 μg/salmeterol 50 μg (Seretide; GlaxoSmithKline) one dose twice daily; and omeprazole 20 mg daily. Previous trials of a β-blocker and angiotensin-converting enzyme inhibitor were not tolerated. Two days after the patient was discharged, a home visit by the clinical nurse consultant (CNC) found him with grossly oedematous legs, jugular venous pressure (JVP) elevated above his ears, and crepitations from the bases to the upper mid zones of his lungs. On Day 1 and 2 of CNC care at home, the patient received intravenous bolus doses of frusemide 80 mg, resulting in good diuresis. On Day 3, the CNC was unable to gain intravenous access and, after consulting the Community Heart Failure Team cardiologist and pharmacist, administered frusemide 80 mg via a nebuliser. The patient reported immediate improvement. Oxygen saturation increased from 88% to 97% on room air and his chest was clearer on auscultation. Increased diuresis occurred, with weight loss of 1 kg. Because the patient’s JVP and leg oedema were unchanged, the dose was repeated daily for 5 days until respite admission (for social reasons and intravenous frusemide administration). Nebulised frusemide had provided symptomatic relief from dyspnoea for about 5 hours with no adverse effects for the patient, but did not provide sufficient diuresis to reduce his fluid overload symptoms. Ultimately, a central catheter (“long line”) was inserted to enable the CNC to administer frusemide intravenously at the patient’s home. CHF-associated dyspnoea causes significant morbidity and distress for patients and carers. Nebulised frusemide has been used for relief of dyspnoea associated with asthma, COPD and malignancy.1 Its precise mechanism of action is unknown, but is believed to be through local lung rather than renal effects.1 Our searches of MEDLINE, EMBASE, CINAHL and the internet found no reports of the use of nebulised frusemide for dyspnoea resulting from pulmonary oedema or CHF. Patients with CHF receiving palliative care have limited options for diuresis when oral administration is ineffective and intravenous access is unavailable. The use of nebulised frusemide could have potential in this setting, but requires further research.

Kate A Towers · Kimberley A Bardsley · Peter S Macdonald

To the Editor: We report a case of ischaemic stroke in a 34-year-old man with severe vitamin C deficiency caused by poor nutrition. The patient was a lifelong non-smoker with no history of hypertension or hypercholesterolaemia, and no family history of stroke, although he had recently been diagnosed with type 2 diabetes mellitus. At presentation, neurological examination showed profound left-sided hemiparesis, with normal sensory examination and visual fields. Cardiovascular examination was normal, and there were no carotid bruits. The patient’s body mass index was 25.5 kg/m2. Magnetic resonance imaging of of his brain showed acute infarction in the right posterior corona radiata (Box, A). Coagulation and lipid profiles were normal. Glycosylated haemoglobin was 7.1%. Comprehensive testing for underlying thrombophilia, vasculitides and Fabry disease all returned negative results. Computed tomography angiography and carotid ultrasonography confirmed normal carotid and vertebral arteries. Transoesophageal echocardiography showed a structurally normal heart without a source of embolus. The patient had poor dentition, with calculus deposition, scorbutic gums and gingival inflammation (Box, B), and reported easy bruising in recent months. Suspecting a diagnosis of scurvy, we conducted a nutritional assessment of the patient. His diet consisted mainly of fast food, with negligible vegetable and fruit intake, and no vitamin supplementation. For the week before admission, we determined that his average vitamin C intake was 4 mg/day. This corresponded to a > 99% probability of inadequate intake when compared with the estimated average requirement of 30 mg/day for adults1 (z = − 4.33; P = 0.0015). Laboratory testing confirmed the presence of severe vitamin C deficiency (< 5 μmol/L; reference range, 40–100 µmol/L). The patient was admitted to a stroke unit, commenced on aspirin, ramipril and atorvastatin, and received dietary counselling. Vitamin C 1000 mg daily was prescribed for one month. Subsequent testing confirmed normalisation of his plasma vitamin C. Following inpatient rehabilitation, he regained motor function and returned to independent living. There is growing evidence that vitamin C deficiency is an important, but largely unrecognised, risk factor for modification in patients with cerebrovascular disease.2 Vitamin C is a water-soluble antioxidant that inhibits oxidation of low-density lipoprotein and protects against endothelial dysfunction. Primate models have confirmed that cerebral infarct size is inversely related to cerebral vitamin C content.3 Although scurvy is now relatively rare, subclinical vitamin C deficiency is not uncommon, being present in about 10% of the general population.4 Alcoholics, institutionalised and elderly people are particularly at risk. In this case, we hypothesise that an unhealthy diet resulted in deficiencies in antioxidants (including vitamin C), and that this contributed to stroke pathogenesis. The marked prematurity of disease onset may have resulted from effect modification of antioxidant deficiency on conventional atherosclerotic risk factors (such as diabetes). A cohort study previously observed the modifying effect of vitamin C deficiency on the association between stroke and hypertension.5 However, it is unlikely that a direct causal link will ever be established. Since malnutrition and unhealthy eating practices continue to be serious public health problems, we suggest attention to nutritional status should be incorporated into the new standard of stroke care. Perhaps a new adage should be considered: an orange a day keeps stroke away? A: Diffusion-weighted magnetic resonance image of the patient’s brain showing an acute infarction in the posterior limb of the right corona radiata. B: The patient’s mouth showing scorbutic gums consistent with scurvy.

Emily Y-J He · Louis W Wang · Matthew C Kiernan

Carotid artery stenting has a place in managing symptomatic stenosis, but only sometimes Carotid stenosis remains one of the most readily treatable causes of ischaemic stroke. Its treatment has undergone many changes in recent years, with the advent of endovascular techniques seeming to offer great promise of a much less invasive procedure. In the late 1990s, the Australian Association of Neurologists published guidelines for the use of carotid balloon angioplasty alone (ie, without stenting) as a treatment for carotid stenosis, recommending cautious use in the absence of data from randomised controlled trials (RCTs).1 However, advances in vascular stent technology have resulted in a greatly increased use of carotid artery stenting (stenting), often with specialised filter devices deployed distal to the carotid stenosis to catch embolic debris that may arise from catheter or stent manipulation. Recent evidence from RCTs has cast doubt on the safety of widespread use of stenting for the treatment of patients with symptomatic or asymptomatic carotid stenosis. The overall results from these RCTs indicate that carotid endarterectomy (endarterectomy) is still the preferred treatment option for symptomatic carotid stenosis. The results of three major European studies into the treatment of symptomatic stenosis — Endarterectomy versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S); Stent-Protected Angioplasty versus Carotid Endarterectomy (SPACE); International Carotid Stenting Study (ICSS) — showed that stenting was more hazardous than endarterectomy for the outcomes of stroke and death during the periprocedural period (30 days), and on longer-term follow-up.2-4 Perhaps in contrast, the recent North American Carotid Revascularization Endarterectomy vs Stenting Trial (CREST) demonstrated equivalent (non-significant) rates of stroke, myocardial infarction and death in its stenting and endarterectomy groups in the periprocedural period, and at 4 years.5 However, in CREST there was a significantly reduced rate of stroke and death in the endarterectomy group, but this was offset by an increased risk of myocardial infarction, in part due to the definition of myocardial infarction used, and a greater number of study patients with comorbid ischaemic heart disease.5 Use of cerebral protection devices during stenting was mandated in CREST but not in the European studies. The value of these devices is in question, with evidence indicating that they were no more effective in reducing the clinical risk of stroke than unprotected stenting.6,7 Moreover, data from a magnetic resonance imaging substudy undertaken as part of the ICSS indicated that, compared with endarterectomy, there was a threefold increase in silent brain infarctions in stenting with use of cerebral protection devices (adjusted odds ratio, 3.28 [95% CI, 1.5–7.2]).7 A meta-analysis of 11 RCTs (including EVA-3S, SPACE and ICSS, but not CREST) showed that endarterectomy was superior to stenting in short-term but possibly not longer-term outcomes, a difference largely driven by non-disabling stroke.8 The Australasian guidelines9,10 have now been upgraded following additional published data from CREST5 and a meta-analysis of the three large European trials (EVA-3S, SPACE, ICSS)11 indicating that stenting is at least as safe as endarterectomy in patients under 70 years of age, while it presents a greater risk of stroke for those older than 70 years (Box). The RCTs indicate that there is currently no clear evidence to support either endarterectomy or stenting as a treatment for asymptomatic carotid stenosis. Indeed, current medical therapies have reduced the risk of stroke in asymptomatic stenosis to as low as 0.5% per year.12 Ongoing RCTs continue to address this question (eg, the Asymptomatic Carotid Surgery Trial 2, comparing stenting and endarterectomy in the treatment of asymptomatic carotid stenosis).13 Until recently, in Australia and New Zealand, there were no specific published guidelines for carotid artery stenting. To redress this, an intercollegiate working group was formed — the Carotid Stenting Guidelines Committee — comprising expert representatives of the Royal Australasian College of Physicians, the Royal Australasian College of Surgeons, and the Royal Australian and New Zealand College of Radiologists. Consensus for guideline parameters was reached using the modified Delphi consensus method of iterative consultation. The committee’s guidelines recommend clinical selection criteria for carotid artery stenting, as well as cognitive and technical requirements that clinicians should meet before performing stenting.9,10 The guidelines do not deal with training criteria and procedural accreditation as these are determined by the Conjoint Committee for Recognition of Training in Peripheral Endovascular Therapy of the abovementioned colleges. On current evidence, the guidelines recommend that carotid artery stenting may be considered a treatment option in specific, high-risk patients with symptomatic severe stenosis who are considered unsuitable for endarterectomy (Box).10 It is important to note that these are relative contraindications to endarterectomy as there is no evidence to support stenting in these patients (indeed, they are often excluded from clinical trials). Finally, it is important that all patients being considered for a carotid intervention have preprocedural neuroimaging and independent neurological assessment before and after the procedure. This allows for an audit comparison with the results of RCTs where neurological evaluation of all patients is routine, and serves as a benchmark for best clinical practice. In summary, the evidence from RCTs indicates that, at present, a cautious approach should be taken to recommending carotid artery stenting — festina lente (hasten slowly). Stenting should not be performed in most patients with symptomatic severe carotid stenosis, and there is currently no evidence to support stenting as a treatment for asymptomatic carotid stenosis. Stenting warrants consideration in younger patients (< 70 years of age) and those with symptomatic severe carotid stenosis unsuitable for endarterectomy. These standards should apply in all health care settings, public and private. Advances in endovascular technologies, and evidence from future RCTs and meta-analyses, will guide revisions of the guidelines. Carotid Stenting Guidelines Committee: recommended indications and contraindications for carotid artery stenting (CAS)10 Indications Symptomatic carotid disease in the following conditions may be assessed at high surgical risk for carotid endarterectomy (CEA) by an appropriate clinician experienced in the management of carotid stenosis: post-radiation therapy block dissection of the neck in-situ tracheostomy recurrent stenosis following previous CEA severe cervical spine arthritis surgically inaccessible carotid stenosis (eg, obesity, high carotid bifurcation) contralateral recurrent laryngeal nerve injury contralateral internal carotid artery occlusion Symptomatic severe carotid stenosis* in patients under 70 years of age, where carotid revascularisation is considered appropriate Symptomatic or asymptomatic carotid stenosis where carotid revascularisation is considered appropriate, and the patient is randomised to CAS in a clinical trial Contraindications Absolute Carotid stenosis in a patient with significant contraindications to angiography Carotid stenosis with angiographically visible intraluminal thrombus Carotid occlusion Relative Carotid stenosis associated with an intracranial vascular malformation Contraindications related to vascular anatomy and atherosclerosis (eg, type 2–3 arch; bovine arch; severe aortic arch or ipsilateral common carotid atherosclerosis; severe proximal common carotid artery tortuosity; severe distal internal carotid artery tortuosity (possibly compromising embolic protection devices); sharply angulated internal carotid artery; carotid string sign; circumferential calcification of carotid plaque; loose thrombus associated with carotid plaque) * Severe carotoid stenosis is defined as ≥ 70% using North American Symptomatic Cartoid Endarterectomy Trial criteria.5

on behalf of the Carotid Stenting Guidelines Committee (Australia and New Zealand)

Routine monitoring of performance in the provision of cardiac services aids quality assurance and enables comparisons of performance to national and international standards. The Australasian Society of Cardiac and Thoracic Surgeons conducts a surgical registry that has grown from six hospitals participating in 2001 to 21 contributing in 2010. Variation in performance is monitored on a quarterly basis through the use of control chart methodology, and a peer-review mechanism and governance process for reporting have been established. Proposed future developments of the registry include its expansion to include interventional cardiology procedures, such as implantation of stents and cardiac devices, and a modular format, with the patient rather than the procedure being the key element of the system. An Australian Cardiac Procedures Registry will provide information to stakeholders, including consumers, clinicians, health funders and policymakers, on performance standards and quality of care of medical services affecting an ever-increasing number of Australians.

Christopher M Reid DipEd, MSc, PhD · Angela L Brennan RN, CCRN · Diem T Dinh BSc(Hons), PhD · Baki Billah PhD, MAS, MSc · Carl B Costolloe BSc(Photonics), MEng · Gilbert C Shardey MD, FRACS · Andrew E Ajani MB BS, MD

Cimicifuga racemosa, better known as black cohosh, has been widely used in Western cultures as a herbal treatment for relieving symptoms of menopause. It has previously been linked to cases of liver toxicity. We report a case of reversible complete heart block in a woman who had recently begun taking a herbal supplement containing black cohosh. We review the known side effect profile of black cohosh and its relationship to our case. Clinical recordIn April 2009, a 59-year-old woman presented to a hospital emergency department after experiencing three episodes of syncope. She had never experienced cardiac ischaemic symptoms and there was no identifiable precipitant for a vasovagal event. She had no history of thyroid disease, hypertension, hyperlipidaemia or diabetes. Her personal and family medical histories were unremarkable, and she was a lifelong non-smoker and non-drinker. She had no recent febrile illness. She took no regular medications, but 2 weeks earlier had commenced taking one tablet daily of Remifemin (Schaper & Brümmer, Salzgitter, Germany; distributed by SciNat Australia, Gold Coast, Qld), a herbal preparation for the alleviation of menopausal symptoms. While undergoing cardiac monitoring in the emergency department, the patient experienced a further episode of syncope. Telemetry (Box 1, A) and an electrocardiogram (ECG) (Box 1, B) demonstrated complete heart block. An atropine bolus was administered and an isoprenaline infusion commenced. The initial ECG performed after commencement of isoprenaline demonstrated 2:1 heart block. The patient’s serum electrolyte levels were normal. Further symptomatic complete heart block occurred while the patient was receiving the isoprenaline infusion (10 μg/min). A temporary pacing wire was placed, with reliable capture at a rate of 60 beats/min, an output of 0.5 V and reliable sensing at greater than 10 mV. Remifemin was ceased. Measures of thyroid function and serial serum cardiac troponin I levels were within normal ranges. Echocardiography demonstrated a structurally and functionally normal heart. The patient required intermittent pacing for heart block throughout the next 2 days, after which pacing was no longer required, and she was successfully discharged on Day 5 with normal sinus rhythm. She underwent 24-hour ECG Holter monitoring 1 week after discharge, which confirmed no further episodes of heart block. She did not recommence Remifemin treatment and, 12 months later, reported no further episodes of syncope. DiscussionThe only listed active ingredient of Remifemin is isopropanolic Cimicifuga racemosa root extract, also known as Actaea racemosa and most commonly known as black cohosh (BC). Remifemin contains the most thoroughly researched formulation of BC.1 BC was traditionally used by Native Americans of Canada and the eastern United States to treat malaria, impaired kidney function, sore throat, rheumatism, menstrual irregularities, and pain during childbirth.2 Recently, there has been interest in its use in the treatment of menopausal symptoms. A recent systematic review identified over 72 studies of BC,3 but only 13 of these were clinical studies involving BC-only preparations published since an earlier review in 2003.4 Findings regarding adverse events were consistent with those of another earlier review, which had found that in more than 2800 patients, the rate of adverse events was about 5.4%, and over 97% of events were minor.3 Most adverse events identified by the more recent review were gastrointestinal symptoms and musculoskeletal and connective tissue disorders.3 Three recently published reviews have examined hepatotoxicity3,5,6 — the most commonly reported serious adverse event associated with BC. They described seven, 42 and 31 cases of hepatotoxicity, respectively, but all three concluded that, in general, data supporting definite causality are lacking. The US Pharmacopeia Dietary Supplements Information Expert Committee, the European Medicines Agency, and the Australian Therapeutic Goods Administration (TGA) recommend that preparations containing BC should carry a warning of possible hepatotoxicity.5-7 Other serious adverse events reported include anaphylaxis, cutaneous vasculitis and myotoxicity.8-10 Studies of BC for mutagenicity, teratogenicity and carcinogenicity have produced negative findings.11 A search of MEDLINE identified no reports of bradycardia due to BC in the literature. However, a Google internet search using the terms “black cohosh” and “heart rate” yielded numerous natural therapy websites describing “slow heart rate” as a side effect of BC. Slow heart rate is also described as a side effect of BC in the Micromedex AltMedDex System database (version 5.1; Thomson Reuters [Healthcare] Inc, Denver, Colo, USA). We notified the TGA of this adverse event. The TGA has received 33 previous reports of suspected adverse events involving BC; none have involved bradycardia or syncope, and one involved hypotension (Rob Crowdy, Adverse Drug Reactions System Database Manager, TGA, personal communication, 3 June 2010). The mechanism by which BC exerts its effects is uncertain. The rhizome of BC contains a number of biologically active constituents, including the triterpene glycosides actein, 27-deoxyactein and cimicifugoside, as well as long-chain fatty acids, resins, caffeic acids, isoferulic acids, phytosterin, fukinolic acid, salicylic acid, sugars and tannins.12 To date, over 50 compounds derived from BC have been described.13 Serotonergic effects not due to serotonin selective reuptake inhibition have been demonstrated with BC preparations.14,15 BC exhibits competitive binding to the 5-HT1A, 5-HT1D and 5-HT7 receptors14-16 and is a partial agonist at serotonin receptors.17 This is noteworthy, as studies show that activation of 5-HT1A receptors in the hypothalamus inhibit hypothalamus-mediated increases in heart rate and blood pressure.18,19 One study investigating the vasoactive effects of BC demonstrated that BC-derived cimicifugic acids inhibit noradrenaline-mediated contraction in rat aortas by inhibition of calcium influx.20 In their 1993 review,4 Borelli and colleagues described a 1935 study in which four glycosidic fractions obtained from the rhizome of BC were administered to dogs; the fraction insoluble in water was found to induce strong arterial hypotension, a decrease in cardiac contraction, and bradycardia to the point of death. Based on the published pharmacology of the components of BC, it is difficult to provide a clear explanation as to how it mediates complete heart block. It is noteworthy that bradycardia is a widely listed side effect of BC in non-academic literature and that profound bradycardia has been documented in animal studies following administration of its extracts.4 Applying the Naranjo algorithm to this case shows that BC was probably responsible for the presentation of our patient (Box 2).21 Given the severity of the adverse reaction and the 2-week delay until its onset, reintroduction of BC while appropriately monitoring the patient to strengthen the argument for causality is not feasible. Although BC is potentially useful in the treatment of menopausal symptoms, it has not been subjected to the extensive postmarketing surveillance that conventional pharmacological agents receive and has potential for numerous adverse effects. It should be considered as a potential cause of unexplained signs or symptoms of cardiac conduction disturbance. 1 Patient’s telemetry and electrocardiogram (ECG) traces Cardiac monitoring telemetry trace (A) and ECG (B) showing complete heart block. 2 Application of the Naranjo algorithm21 to this case Criteria (score applied) Score 1. Are there previous conclusive reports on this reaction? Yes (+1) No (0) Do not know or not done (0) 0 2. Did the adverse event appear after the suspected drug was given? Yes (+2) No (- 1) Do not know or not done (0) 2 3. Did the adverse reaction improve when the drug was discontinued or a specific antagonist was given? Yes (+1) No (0) Do not know or not done (0) 1 4. Did the adverse reaction appear when the drug was readministered? Yes (+2) No (- 2) Do not know or not done (0) 0 5. Are there alternative causes that could have caused the reaction? Yes (- 1) No (+2) Do not know or not done (0) 2 6. Did the reaction reappear when a placebo was given? Yes (- 1) No (+1) Do not know or not done (0) 0 7. Was the drug detected in any body fluid in toxic concentrations? Yes (+1) No (0) Do not know or not done (0) 0 8. Was the reaction more severe when the dose was increased, or less severe when the dose was decreased? Yes (+1) No (0) Do not know or not done (0) 0 9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure? Yes (+1) No (0) Do not know or not done (0) 0 10. Was the adverse event confirmed by any objective evidence? Yes (+1) No (0) Do not know or not done (0) 1 Total score* 6 * ≥ 9 = definite adverse drug reaction (ADR); 5–8 = probable ADR; 1–4 = possible ADR; 0 = doubtful ADR.

Scott C McKenzie MB BS, BSc, FRACP · Atifur Rahman FRACP

Objective: To determine the cost-effectiveness of routine administration, irrespective of blood pressure (BP), of a fixed-dose combination of perindopril and indapamide to patients with type 2 diabetes mellitus.Design, setting and participants: Prospective cost-effectiveness analysis within the Action in Diabetes and Vascular Disease: Preterax and Diamicron-MR Controlled Evaluation (ADVANCE) trial, an international, multicentre, randomised controlled trial of 11 140 participants with type 2 diabetes randomly allocated to receive perindopril plus indapamide (4 mg–1.25 mg/day) or placebo.Main outcome measures: Health-related quality-of-life measured by the EuroQol-5D, resource utilisation, and cost-effectiveness (cost per death averted at 4.3 years’ average follow-up, and estimated cost per life-year gained, by extrapolation).Results: The mean health-related quality-of-life score of survivors was 0.80 (on a 0–1 scale [death to full health]), with no difference between treatment groups. Active treatment reduced hospital admissions for coronary heart disease and coronary revascularisation by 5%. For the Australian participants, perindopril–indapamide cost A$1368 per patient during the trial period, but reduced total hospitalisation costs by A$410 and other medication costs (mainly other BP-lowering drugs) by A$332. The absolute reduction in all-cause mortality for the active treatment group was 1.1%, giving a cost per life saved of A$49 200. Lifetime extrapolation gave an estimated cost per life-year saved of A$10 040 (discounted at 5% per year).Conclusion: The combination of perindopril and indapamide in patients with type 2 diabetes appears to be cost-effective.Trial registration: United States National Library of Medicine NCT00145925.

Paul P Glasziou MB BS, PhD · Philip M Clarke MEc, PhD · Jan Alexander · Mohana Rajmokan MSc · Elaine Beller BSc, MAppStat · Mark Woodward PhD · John Chalmers MD, PhD, FRACP · Neil Poulter MSc, FRCP · Anushka A Patel FRACP, PhD

To the Editor: Evidence is beginning to accumulate on the effectiveness of the low-density lipoprotein (LDL) cholesterol-lowering medicine ezetimibe. While there are no completed trials investigating ezetimibe’s effect on clinically important end points, two recent trials investigating its effect on carotid intima media thickness (CIMT) have both reported disappointing results.1,2 After each of these trials, the Journal has published editorials by Hamilton-Craig, who offers reassurance about ezetimibe and encourages ongoing prescription of this drug to patients who have elevated LDL levels despite maximum-tolerated statin therapy.3,4 Such a sanguine opinion seems at odds with the negative trial evidence, and therefore worthy of debate. Briefly, the ENHANCE (Ezetimibe and Simvastatin in Hypercholesterolemia Enhances Atherosclerosis Regression) trial, which compared ezetimibe plus simvastatin with simvastatin treatment alone in 720 patients with familial hypercholesterolaemia, found no significant difference (and a trend in the direction of harm) with respect to the primary end point of CIMT.1 The ARBITER 6-HALTS (Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol 6 — HDL and LDL Treatment Strategies in Atherosclerosis) trial compared ezetimibe with extended-release niacin in statin-treated patients with coronary heart disease.2 Among 315 patients with available results, the group taking niacin showed a statistically significant reduction in CIMT, but the group taking ezetimibe showed no such reduction. Of concern, increased cumulative exposure to ezetimibe was associated with progression of CIMT (P = 0.05). Although far from definitive, the results of these two trials offer no reassurance of benefit from ezetimibe and, in my view, may portend harm. It may seem counterintuitive that ezetimibe, which significantly lowers LDL cholesterol levels,1,2 could be ineffective or harmful. However, the history of medicine is replete with examples of interventions that improve numerical disease measures without benefit to patients. One recent example was torcetrapib, which, despite increasing high-density lipoprotein cholesterol and reducing LDL cholesterol levels in a promising manner, was found to cause serious adverse events, including death.5 I agree with Hamilton-Craig that we require trials measuring major cardiovascular events to really understand the effects of ezetimibe. Where we disagree is how to manage our patients during the period of uncertainty until publication of the results of these trials. While he argues for continued prescribing of ezetimibe, I suggest we should explicitly share our uncertainty about the safety and efficacy of this drug with our patients by discussing the existing research. Some patients will, like Hamilton-Craig, place their faith in the cholesterol hypothesis and be reassured by an assumption of cardiovascular protection as their LDL falls. Others will choose to wait until we have more robust evidence that ezetimibe is safe and effective. I would wait.

Brett D Montgomery

In reply: I agree with Montgomery that cardiovascular disease (CVD) outcomes are required to determine the role of ezetimibe. The Simvastatin and Ezetimibe in Aortic Stenosis (SEAS) trial (in which patients with aortic stenosis were treated for 52.2 months with statin plus ezetimibe or statin plus placebo) showed a 4.7% reduction in ischaemic CVD events in the ezetimibe group (P = 0.02; number needed to treat, 23), driven by a reduced need for coronary artery bypass grafting.1 In contrast to previous trials showing regression of atherosclerosis in response to statin therapy, baseline carotid intima media thickness (CIMT) levels in the ENHANCE (Ezetimibe and Simvastatin in Hypercholesterolemia Enhances Atherosclerosis Regression) trial were normal, due to previous statin therapy. This is likely to account for the lack of change in CIMT with ezetimibe treatment in the ENHANCE trial.2 As no placebo group was included, neither lack of benefit nor harm from ezetimibe therapy can be inferred.2 Data from animal studies have shown atherosclerosis regression after ezetimibe treatment through multiple mechanisms.3-5 Prospective randomised controlled trials with statins, resins or surgery have independently shown an approximate 1% reduction in CVD per 1% reduction in low-density lipoprotein cholesterol (LDL-C) level. Evidence for the benefits of lowering LDL-C is among the most robust in medicine. Pending the outcomes of IMPROVE-IT (the Improved Reduction of Outcomes: Vytorin Efficacy International Trial, a multicentre study of ezetimibe plus simvastatin versus simvastatin treatment of patients with acute coronary syndrome [http://clinicaltrials.gov/ct2/show/NCT00202878]), or clinical outcome data confirming those of the ARBITER 6-HALTS (Arterial Biology for the Investigation of the Treatment Effects of Reducing Cholesterol 6 — HDL and LDL Treatment Strategies in Atherosclerosis) trial,6 it seems reasonable to continue to use ezetimibe to lower LDL-C levels in patients who are not achieving LDL-C targets despite statin therapy or who are intolerant to statins. Extended-release nicotinic acid (Niaspan [Abbott Laboratories, Chicago, Ill, USA]) may be an appropriate alternative to statins as second-line therapy, and should be made available under the Pharmaceutical Benefits Scheme for treating patients with dyslipidaemia. (Niaspan has approval from the Therapeutic Goods Administration for marketing in Australia, but is not being imported into Australia at this stage.)

Ian R Hamilton-Craig

To the Editor: In his letter1 regarding Heeley et al’s article on cardiovascular risk perception and evidence–practice gaps in Australian general practice,2 Radford quotes, “no [automated blood pressure-measuring machines] were accurate enough to ... replac[e] a manual sphygmomanometer”. The citation dates from 1973, when oscillometric blood pressure monitors such as the Omron HEM-907 — distributed by the High Blood Pressure Research Council of Australia (HBPRCA) with support from the Servier Foundation — did not exist. In defence of digital blood pressure devices, a cluster randomised controlled trial conducted in Australian general practice has demonstrated their superiority compared with existing manual devices.3

Mark R Nelson

Objectives: To assess whether a collaborative interdepartmental pathway involving emergency department (ED) physicians activating the cardiac catheterisation laboratory (CCL) with immediate patient transfer to the CCL reduces door-to-balloon (DTB) times for patients with suspected ST-elevation myocardial infarction (STEMI).Design, setting and participants: A quasi-experimental before-and-after observational study using a prospective database, supplemented by chart review, of consecutive patients transferred from the ED to the CCL for suspected STEMI, from January 2007 to October 2009, at Sir Charles Gairdner Hospital, an adult tertiary-care hospital, Western Australia.Main outcomes measures: Median DTB time and proportion of patients with DTB time of < 90 minutes. Secondary outcomes, based on analysis of predefined subgroups, included door-to-activation time, activation-to-balloon time and false-positive activations of the CCL.Results: Two hundred and thirty-four patients underwent emergency coronary angiography for suspected STEMI, with 188 (80%) undergoing percutaneous coronary intervention (118 before and 70 after implementation of the new pathway). Following implementation of the new pathway, median DTB time reduced from 97 to 77 minutes (P < 0.001), median door-to-activation time from 28 to 15 minutes (P = 0.002) and median activation-to-balloon time from 66 to 53 minutes (P < 0.001). The proportion of patients with recommended DTB time of < 90 minutes increased from 41% to 77% (P < 0.001) with no change in false positive CCL activation rates (12% v 11%; P = 0.38).Conclusion: ED physician activation of CCL with immediate patient transfer is associated with highly significant improvements in DTB time without increased false positive rates.

Alexander B Willson MB BS(Hons), MPH, FRACP · David Mountain MB BS, FACEM · Joanne M Jeffers MB BS · Cheryl G Blanton MSc · Brendan M McQuillan MB BS, PhD, FRACP · Joseph Hung MB BS, FRACP, FCSANZ · Michael H Muhlmann MB BS, FRACP · Michael C Nguyen MB BS, FRACP

Prompt myocardial reperfusion, particularly if achieved within 2 hours of the onset of symptoms, improves outcomes in patients with ST-elevation myocardial infarction (STEMI). Recent data suggest that ambulance-administered prehospital thrombolysis, if given within 2 hours of the onset of STEMI, produces superior outcomes to primary percutaneous coronary intervention (PCI); if given within 4 hours, the outcomes are similar. For optimal results after thrombolysis, patients require angiography (and PCI where appropriate) within 24 hours of the event. These developments have major implications for the practice of cardiology and for the organisation of health services in Australia.

Richard W Harper MB BS, FRACP, FACC · Jeffrey Lefkovits MB BS, FRACP, FCSANZ

Evidence concerning the safety of rosiglitazone continues to evolve In February 2010, the United States Senate Committee on Finance released a report on the safety of rosiglitazone.1 The report concluded that there were possible cardiac risks associated with rosiglitazone and that the manufacturer, GlaxoSmithKline (GSK), was aware of this well before it became public. The authors further stated that, rather than warn patients and regulatory authorities promptly, GSK executives chose to intimidate independent physicians who publicised the possible risks, minimise the impact of adverse findings, and downplay the possibly beneficial cardiovascular effects of the other available drug in the class, pioglitazone.1 A week before the Finance Committee report was released, an editorial by Steve Nissen, lead author of the meta-analysis that first raised cardiovascular concerns regarding rosiglitazone in 2007,2 was published online.3 The editorial related to an article on the increased risk of heart failure, a recognised adverse effect of glitazones, found in the Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycaemia in Diabetes (RECORD) trial.4 However, it was more a detailed account of the sequence of events surrounding rosiglitazone’s possible adverse cardiovascular effects, starting with its US approval in 1999. The editorial viewed the RECORD trial (the main results of which were published mid 20095 but not considered in the Finance Committee report) as underpowered, despite demonstrating that rosiglitazone was non-inferior to metformin or sulfonylurea for the primary endpoint of cardiovascular hospitalisation or death. Although there were no new safety concerns or efficacy data in either the Finance Committee report or Nissen’s editorial, there was a prompt media response. The New York Times, for example, highlighted a suggestion by the Finance Committee that if every diabetic patient in the US taking rosiglitazone was given pioglitazone instead, 500 heart attacks and 300 cases of heart failure would be averted every month.6 There was also a quick response from GSK, which categorically rejected the findings of the Finance Committee and the assertions of the New York Times in separate media statements, and published a point-by-point response to Nissen’s editorial.7 More recently, the results of a retrospective analysis of US Medicare data for older patients8 and an expanded meta-analysis from Nissen’s group,9 both suggesting adverse cardiovascular effects of rosiglitazone, have contrasted with a post-hoc analysis of data from the Bypass Angioplasty Revascularization Investigation in Type 2 Diabetes study, in which rosiglitazone was found to be of significant benefit in patients with established coronary artery disease.10 At present, rosiglitazone remains approved for use in Australia as monotherapy or as part of dual oral combination therapy with metformin or a sulfonylurea,11 although only the combination therapy is subsidised by the Pharmaceutical Benefits Scheme. The product information contains a boxed warning for patients with known ischaemic heart disease, particularly those taking nitrates, and highlights the increased risk of myocardial ischaemia found in pooled short-term clinical studies.11 In the US and Europe, rosiglitazone remains available despite the recent media reports. The US product information has similar warnings to those of the Australian version, while the European version has the general recommendation that the drug not be used by patients with ischaemic heart disease and/or peripheral arterial disease. One possible reason why rosiglitazone was not withdrawn in 2007 is that the statistical methods used in the original meta-analysis2 were questionable. Alternative reasonable approaches can yield increased or decreased risks that are either statistically significant or not significant for both myocardial infarction and cardiovascular death.12 As there are no trials with cardiovascular events as the primary endpoint showing benefit of pioglitazone over other therapies, the most compelling evidence for its apparently better cardiovascular profile comes from a similar meta-analysis to that for rosiglitazone.3 A cardiovascular disease outcome study of rosiglitazone versus pioglitazone is, therefore, justifiable and in progress (Thiazolidinedione Intervention with Vitamin D Evaluation [TIDE]; ClinicalTrials.gov NCT00879970). Nevertheless, TIDE might become a casualty of the recently reactivated controversy before it reports in 2015, as the Endocrinologic and Metabolic Drugs Advisory Committee of the US Food and Drug Administration (FDA) continues to review the ethical and clinical implications of the available rosiglitazone safety data. Although the debate about the safety of rosiglitazone has centred on cardiovascular risk, a further potential concern is fracture.13 The deleterious effects of glitazones on bone emerged in animal studies dating back to 1996. Unfortunately, despite knowledge of these data, neither glitazone manufacturer included prespecified bone loss parameters and endpoints in any clinical trial. However, retrospective analyses of data from blood glucose-lowering efficacy trials involving rosiglitazone and pioglitazone, reported in 2006 and 2007, respectively, confirmed an increased fracture risk in humans.13 Given that these drugs have been available in Australia and most other countries for only 10 years, their long-term effect on fracture rates is worrying, especially in postmenopausal women. Glitazone therapy can improve glycaemic control in patients with type 2 diabetes, but patients should be selected according to drug-specific contraindications and warnings, the glycaemic effect should be reviewed after at least 3 months to confirm response, and adverse effects including weight gain, fluid retention and reduced bone density should be monitored during continued use. The recent adverse publicity regarding rosiglitazone highlights issues that can arise when drugs are approved and marketed without definitive efficacy and safety data. There is a need for pharmaceutical companies, academia and regulatory authorities to use preclinical and early phase clinical data to identify, through careful phenotyping, the patient population with the most potential for benefit and the least potential for harm when new drugs are being evaluated for registration. One important consequence of the rosiglitazone controversy is that adequately powered Phase IV cardiovascular safety studies are now required by the FDA when new therapies for diabetes are registered.3 The promise of the glitazones was that they targeted one of the central pathophysiological defects in type 2 diabetes, namely insulin resistance, and improved markers of cardiovascular risk including serum C-reactive protein and microalbuminuria. Unfortunately, based on a variety of clinical trials and observational studies, they do not appear to have a consistent cardiovascular advantage over established blood glucose-lowering agents, including metformin and sulfonylureas.

Timothy M E Davis MB BS, DPhil, FRACP · Johannes B Prins MB BS, PhD, FRACP

To the Editor: As a long-time resident of the “swamp of uncertainty” which is general practice — where specialists dare not go — I am accustomed to receiving guidelines and consensus statements from esteemed colleagues and friends who have taken to studying some specific part of the human condition. But these statements inevitably involve primary care, where general practitioners try to achieve balance. Balance is hard to achieve. The “gastro-partialist” thinks that everyone should be on a proton-pump inhibitor. On the other hand, the “osteo-partialist” thinks no-one should be on a proton-pump inhibitor because they stop calcium absorption and double the risk of hip fracture after 10 years. I could go on. However, here I wish to comment on Heeley and colleagues’ article on the perception and management of cardiovascular disease (CVD) risk in Australian primary care.1 Two of the authors are paid by Servier — the company which provided GPs with free automated blood pressure-measuring machines (ABPMs). A review in the American Journal of Cardiology concluded that no ABPMs were accurate enough to be recommended for replacing a manual sphygmomanometer,2 particularly in older people, in whom vessel stiffness leads to overestimation.3 In the study by Heeley et al, ABPMs were used by 90% of the GPs. Australia’s Therapeutic guidelines: cardiovascular say that the blood pressure treatment target should be 140/90 mmHg for all patients.4 It quotes the Cochrane review, which states: A sensitivity analysis in diabetic patients and in patients with chronic renal disease ... did not show a reduction in ... mortality and morbidity outcomes with lower targets as compared to standard targets.5 Heeley et al have used other guidelines. Based on the data provided by Heeley et al, about 17% of the patients in their study were aged over 75 years. Risk calculators do not usually go beyond 74 years, but it appears that with all risk factors at optimum levels, a 75-year-old man still has greater than 15% 5-year risk of a cardiovascular event (ie, a high risk). Being 75 years old is risky! Trying to treat a 75-year-old man’s blood pressure is particularly risky because evidence suggests that there is a paradoxical increase in cardiovascular mortality in men aged over 75 years whose blood pressure is lowered by treatment.6 Heeley and colleagues also stated that “two thirds of patients at high risk of a first CVD event were not prescribed a combination of a [blood pressure]-lowering medication and a statin”. Considering the 17% of patients who were older than 75 years, although the PROSPER (Prospective Study of Pravastatin in the Elderly at Risk) trial showed a reduction in cardiovascular mortality within this age group, there was no reduction in overall mortality.7 So why would a GP treat such patients when overall mortality is not reduced? I imagine that some GPs undertreat because they anticipate poor compliance, because uncorrectable factors are so great that pharmacological interventions will have a miniscule effect, or because the side effects of drug doses needed to achieve guideline targets will be problematic. Maybe they also feel that their ABPMs read a bit high! Balance is an unconscious compromise between real evidence and what is achievable. We need to consider the whole as well as the part if guidelines are to be clinically relevant.

Peter J Radford

In reply: We thank Radford for his insightful comments, seasoned with spice from the frontline of primary care. Although guidelines are an accepted part of clinical practice, they are just recommendations and are not without their limitations. Treatments need to be individually tailored according to many factors. Our aim, therefore, was to provide a current snapshot of adherence to cardiovascular guidelines in primary care in Australia.1 It would be naive of us to think that there would be complete adherence to the guidelines in the “real world”. We wished to obtain an overall benchmark figure and, more importantly, identify treatment gaps or disparities in care across important patient subgroups defined by risk of cardiovascular event. We recognise that digital blood pressure (BP) monitors are no better than mercury sphygmomanometers and require frequent (6-monthly) calibration. However, BP measurement technique is a far more important issue. The use of digital BP monitors in our study provided a standardised measurement machine, but we had no influence on their use or on BP measurement technique. Radford makes a good point regarding the absence of direct randomised trial evidence for the benefit of more intensive BP lowering. The BP target of less than 140/90 mmHg in individuals who are at high risk of a cardiovascular event, such as those with diabetes or chronic renal failure, is a sensible extrapolation from consistent observational epidemiological data. There is, however, convincing evidence of the benefits of multifactorial cardiovascular risk intervention among individuals at high risk.2 We also believe that the available data from randomised trials3,4 provide support for the efficacy of lowering BP in the elderly, and are more robust than the observational data cited by Radford. Similarly, systematic analysis of randomised trials of statin therapy supports efficacy of treatment in older patients.5 Despite their limitations, we believe that current evidence-based recommendations regarding preventive therapies are appropriate guidelines for managing patients who are at risk of a cardiovascular event. As with any guidance for clinical decision making, the application of these recommendations needs to take into account individual patient characteristics and circumstances. Finally, we believe that Servier’s support for this study demonstrates a successful academia–industry partnership. The academic partners had full control of the study design, analyses and publications.

Emma L Heeley · Craig S Anderson · Anushka A Patel · Alan Cass · David P Peiris · John P Chalmers

To the Editor: It seems that the perceived value of invasive therapy for different acute coronary syndromes varies according to how one chooses and values the various trials.1-3 The differing opinions may have been enriched had they not only considered short-term cardiovascular outcome but also weighed quantitatively — in absolute terms — the purported benefits (or lack thereof) of invasive therapy against the immediate complications and the long-term risks and burden of the combined aspirin and clopidogrel antiplatelet therapy usually necessary after stent placement. It may be that with all this information, patients also have a range of opinions regarding benefits and risks.4 It is difficult to deny that cardiologists in general and the industry sponsors of academic cardiology have an enormous vested interest in complicated, expensive therapies. It does not necessarily follow that the current guidelines and their defenders are unduly biased, but nevertheless the conflict of interest should be confronted and addressed, not taken personally, defensively and indignantly.2 Ideally, the guidelines would both be genuine and seen to be so. For conservative physicians, there is also a conflict of interest of sorts in that a conservative interpretation of the literature supports their practice. Forge’s suggestions regarding this issue should be taken more seriously.1 This problem of vested interest is, however, not confined to cardiology. Recently, while highlighting vested interest in mammography, Quanstrum and Hayward broadened their scope and suggested that “independent panels of generalists . . . [be] responsible for objectively synthesizing the medical evidence around a given question or process of care”.5

Stephen P Fitzgerald

To the Editor: In her editorial, Christiansen states that the 2010 edition of Therapeutic guidelines: antibiotic (version 14) is very likely to recommend 24 hours of antibiotic prophylaxis for cardiac surgery,1 rather than the present regimen, which recommends that patients having routine cardiac surgery be given a large dose of cephazolin at induction, with a second dose if the operation is prolonged for more than 3 hours, and no doses after surgery.2 The only evidence presented in support of this proposed change is a single randomised controlled trial that purported to show a higher rate of surgical site infections after a single dose of cephazolin, than after a prolonged, multidose regimen.3 The study was flawed, for two reasons. First, cephazolin was not given until 20–30 minutes after induction of anaesthesia, which is likely to have been too late, because, as Christiansen points out, β-lactams should be given 30–60 minutes before incision.1 Second, the trial was analysed on a per-protocol, rather than an intention-to-treat basis, and 189 of the 1027 participants (18%) were excluded, so the findings may be seriously biased.4 Three other trials have compared one or two doses of a cephalosporin with multiple doses of the same antibiotic in patients having cardiac surgery; none found that multiple doses were superior, although all three were small studies with faults in their design.5-7 In 1998, McDonald and colleagues published a detailed review of single versus multiple doses of antimicrobial prophylaxis for major surgery. The analysis was in response to a suggestion by Christiansen and others that single-dose antibiotic prophylaxis may be inadequate for patients undergoing vascular surgery.8 McDonald and colleagues pointed out that the recommendation for single-dose surgical prophylaxis in Therapeutic guidelines: antibiotic (version 13),2 is based on microbiological first principles, published studies reporting efficacy, convenience of administration, reduced antibiotic resistance and toxicity, and relatively low cost. Their careful analysis of 28 randomised trials, in which the same antimicrobial was used in each arm, showed no advantage from the administration of multiple doses; the odds ratio for infection was 1.06 (95% CI, 0.89–1.25). There is no microbiological reason to suppose that the crucial interaction between contaminating bacteria and the prophylactic antibiotic in the heart is any different from that in the lung, biliary tree, uterus, bowel, prostate or bone.8 In the absence of such evidence, there is no sound reason to change the current, long-standing Therapeutic guidelines: antibiotic2 recommendation.

Frank Shann