Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
4S was a treat to target study
- Andrew D Martin (8 February 2008)
-
The CASE for Fire-and-forget.
- L Sam Lewis (9 February 2008)
-
Statins: “treating to target” benefits neither patient nor clinician
- D Graham Mackenzie, Philip Rutledge (13 February 2008)
-
Statins and Targets
- james alexander grant (28 February 2008)
-
Strategies for prescribing statins: evidence does NOT support prescribing a standard dose without further testing or dose adjustment
- Michael J Murphy, Li Wei, Alexander D. Watson, Thomas M. MacDonald (2 April 2008)
-
Can the role of statins be discussed without recognition of their effects on blood viscosity ?
- Les O. Simpson (3 April 2008)
-
Re: Strategies for prescribing statins: evidence does NOT support prescribing a standard dose without further testing or dose adjustment
- Norbert Donner-Banzhoff, Andreas Sönnichsen (11 April 2008)
|
|
|||
|
Andrew D Martin, Programme Director Mediicnes Management Bury Primary Care Trust, 3 Manchester Road, Bury BL9 0DR
Send response to journal:
|
Regarding the third from last paragraph, patients in 4S were treated to a target cholesterol level of 5.2 and the average dose of simvastatin used to achieve this was 27mg. Competing interests: None declared |
|||
|
|
|||
|
L Sam Lewis, GP Surgery, Newport, Pembrokeshire, SA42 0TJ
Send response to journal:
|
A practical ‘Fire-and-forget’ strategy is exemplified by Aspirin prophylaxis in IHD, or high-CVD-risk , patients. No blood parameters need be monitored, and a standard dose ( 75mg in UK practice ) is offered. Patients at particularly high risk, or those who suffer recurrence, can arguably be offered 150 or 300mg. Debate persists as to what might be the best dose, or what should be done about ‘non-responders’ or ‘resistance’, but the ‘fire-and-forget’ principle is widely accepted. Four factors ( cost, acceptability, safety, and effectiveness ) make the C.A.S.E for Shepherd’s original proposal to move the prescription of Statin onto the same ‘Fire-and-forget’ footing that Aspirin now occupies. 1: Cost. There exists a ‘diminishing returns’ dose-response curve, whereby the initial 10mg of Simvastatin , produces 75% of the maximum response, AND each dose increment produces a reducing incremental gain, at increasing cost per unit gain. For example, since 10mg of Atorvastatin produces half the cholesterol-lowering achieved by 80mg Atorvastatin, and drug-pricing is effectively linear, then offering eight people 10mg of Atorvastatin is 4 times as effective as offering one person 80mg. Atorvastatin is 20 times more costly than Simvastatin, but only twice as potent, dose for dose. 2. Acceptability. Avoidance of unnecessary repeated cholesterol tests and clinic visits is welcomed by patients, doctors, and health accountants. 3. Safety. The risk of side-effects increases with dose. 4: Effectiveness. All patients at CVD risk benefit from Statin treatment, regardless of their initial Cholesterol level. It is now abundantly clear that the benefits to the nation of generically-priced Simvastatin are 20 times that of patent-price Atorvastatin. It is also unarguable that the higher the CVD risk, the greater the gain, regardless of cholesterol. In sum, a threshold of risk exists ( presently a ten-year CVD risk > 20% ), above which the balance of harm is outweighed by the benefit of 10mg simvastatin, providing the greatest good to the greatest number, at an affordable cost to the nation.. Who needs polypills and polyclinics ? Your local pharmacist now offers a Simvastatin and an Aspirin a day, to keep the doctor away ! Competing interests: None declared |
|||
|
|
|||
|
D Graham Mackenzie, Consultant in Public Health NHS Lothian, Deaconess House, 148 Pleasance, Edinburgh, EH8 9RS, Philip Rutledge
Send response to journal:
|
Editor, Donner-Banzhoff and Sönnichsen should be commended for highlighting the lack of evidence for “treating to target” with statins.1 Treatment targets can lead to an escalation of treatment that is not evidence-based, drains clinical resources through repeat clinic visits and blood tests, and can lead to the unnecessary use of branded statins which bring no additional benefit to the patient and are at least thirteen times more expensive than generic statins.2 However Donner-Banzhoff and Sönnichsen have over-simplified the Scottish Intercollegiate Guidelines Network (SIGN) advice on statins. In fact the SIGN guideline distinguishes itself from other guidance in the United Kingdom by not recommending treatment targets for statins in primary prevention of CVD.3 The SIGN guideline also uses the findings of Hayward’s elegant exploration of statin treatment targets4 to note that “current clinical evidence does not demonstrate that lipid therapy should be titrated to achieve proposed LDL cholesterol targets”. The only treatment target mentioned in the SIGN guideline relates to secondary prevention of CVD and this appears to have been a compromise in response to the targets set in the Quality and Outcomes Framework (QOF). Accordingly, QOF should move away from setting treatment targets for statins. One alternative would be a QOF target for compliance with statin treatment. In conclusion, the main point arising from this discussion of statin treatment targets is that the great majority of patients eligible for a statin should receive generic treatment, an argument with clear echoes of Moon and Bogle’s editorial on “switching statins”.5 Yours faithfully, D Graham Mackenzie, Consultant in Public Health Medicine Philip Rutledge, Consultant in Medicines Management NHS Lothian, Deaconess House, 148 Pleasance, Edinburgh, EH8 9RS References 1) Donner-Banzhoff N, Sönnichsen A. Strategies for prescribing statins. Evidence supports prescribing a standard dose without further testing or dose adjustment. BMJ 2008;336:288-9 2) Scottish drug tariff. February 2008. www.isdscotland.org/isd/2246.html 3) Risk estimation and the prevention of cardiovascular disease. A national clinical guideline. Scottish Intercollegiate Guidelines Network 2007. SIGN 97. 4) Hayward RA, Hofer TP, Vijan S. Lack of Evidence for Recommended Low-Density Lipoprotein Treatment Targets: A Solvable Problem. Ann Intern Med. 2006;145:520-530. 5) Moon JC, Bogle RG. Switching statins. BMJ 2006;332:1344–5 Competing interests: None declared |
|||
|
|
|||
|
james alexander grant, General Practitioner Chairman SIGN 97 PH3 1JJ
Send response to journal:
|
Dear Sir In their editorial of the 9th of February, Donner–Banzhoff and Sonnichsen1 cited SIGN 972,3,4 as emphasising the importance of measuring cholesterol and having targets for cholesterol reduction. They are mistaken – as Mackenzie and Rutledge have already pointed out.5 I would wish to emphasise again that SIGN 97 did not recommend any specific targets for either total cholesterol or LDL cholesterol reduction. For secondary prevention of cardiovascular disease reference was made to present QOF target levels.2,5 Reduction of both total and LDL cholesterol is now universally acknowledged as very important for both primary and secondary prevention of cardiovascular disease. However, as in all SIGN guidelines the integrity of the development process requires that the evidence is carefully assessed and evaluated. Only after that process has been rigorously carried out can any recommendations be made. Following an exhaustive review of the literature it was clear to the SIGN 97 Development Group that there was no published evidence for recommending any cholesterol target. SIGN 97 clearly states that for all high risk individuals, statin therapy should be introduced following an informed discussion with the patient. In those with established symptomatic cardiovascular disease intensive statin therapy should be considered. Other guidelines have extrapolated evidence and thereby made assumptions that there are desirable target figures. SIGN 97 indicates that the published evidence does not support such assumptions. Yours faithfully Dr James A Grant
1 Donner-Banhoff N, Sönnichsen A. Strategies for prescribing statins. BMJ 2008;336:288-9. 2 Scottish Intercollegiate Guidelines Network. Risk estimation and the prevention of cardiovascular disease. SIGN 97. www.sign- ac.uk/pdf/sign97.pdf 3 Begg AG. The new SIGN Guidance on CHD and its implications for primary care. Br J Cardiol 2007;14:66-7. 4 Jennings K, Ritchie LD. The new SIGN Guidance on CHD and its implications for secondary care, Br J Cardiol 2007;14:125-6. 5 Mackenzie DG, Rutledge P. Treat to target benefits no one. BMJ 2008;336:406. Competing interests: None declared |
|||
|
|
|||
|
Michael J Murphy, Senior Lecturer in Biochemical Medicine University of Dundee, Ninewells Hospital & Medical School, DD1 9SY, Li Wei, Alexander D. Watson, Thomas M. MacDonald
Send response to journal:
|
In their recent editorial on statin prescribing strategies (1), Donner-Banzhoff and Sönnichsen rightly point out the mismatch between the evidence from statin trials and what is recommended for everyday practice. However, they nevertheless conclude that “the fire and forget strategy is supported by many high quality clinical trials” simply on the basis that these trials did not make treat to target adjustments. We believe that this conclusion is not justified on the basis of the available evidence, which (contrary to the sub-heading for this editorial) does not support prescribing a standard dose of statin without further testing or dose adjustment. The major statin trials were designed to establish the efficacy of statins in various well-defined – and controlled – clinical contexts. None of them was designed to address the effectiveness of statin prescribing strategies in routine clinical practice, and it is simply erroneous to suggest that they provide evidence for one or the other. Indeed it would be well-nigh impossible to perform the kind of randomised clinical trial (RCT) that might help to compare treat to target and fire and forget strategies. First, patients enrolled into RCTs are monitored carefully and regularly, a situation that is far removed from fire and forget in practice. This makes it difficult to design a RCT that would properly address the effectiveness of treatment strategies in the setting of normal care. Second, there may be ethical issues surrounding a RCT of fire and forget prescribing given the evidence of added benefit from intensive statin treatment (2,3). Finally, as the authors themselves point out, it is by no means clear where the funding for such a trial would come from. Given the methodological issues with extrapolating statin trial data, and the difficulties in performing the kind of RCT that is needed to examine this issue, it is not surprising that the evidence base on prescribing strategies is comparatively slim. We compared treat to target and fire and forget strategies with respect to adherence and cardiovascular outcomes (4). Adherence to statins was significantly better in patients treated to target (adjusted odds ratio 2.51, 95%CI 2.26-2.78), and the cardiovascular disease (CVD) event rate lower (hazard ratio of CVD or cardiovascular death 0.41 (0.35-0.48) after adjustment for adherence and baseline CVD risk). The problem with such a retrospective observational approach is that the characteristics used to define fire and forget (patients prescribed a statin, no further cholesterol measurement observed) may reflect patient behaviour as much as a prescribing strategy. Nevertheless, adherence to statins might reasonably be expected to be a marker of such behaviour, and the CVD event rate was substantially lower even after taking this (and baseline risk) into account (4). In short, the evidence such as it is does not support prescribing a standard dose without further testing or dose adjustment – very much the opposite. The authors claim that it is unclear whether possible benefit (of statins) really translates into clinical practice. Evidence is emerging that it does; the impact of statins on cholesterol concentrations in the setting of normal care appears to be comparable with the reductions seen in the major statin trials (5). (1) Donner-Banzhoff N, Sönnichsen A. Strategies for prescribing statins. BMJ 2008;336:288-9. (2) Cannon CP, Braunwald E, McCabe CH, Rader DJ, Rouleau JL, Belder R, et al. Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 2004;350:1495-505. (3) LaRosa J, Grundy SM, Waters DD, Shear C, Barter P, Fruchart J-C, et al. Intensive lipid lowering with atorvastatin in patients with stable coronary disease. N Engl J Med 2005;352:1425-35. (4) Wei L, MacDonald TM, Watson AD, Murphy MJ. Effectiveness of two statin prescribing strategies with respect to adherence and cardiovascular outcomes: observational study. Pharmacoepidemiol Drug Saf 2007;16:385-92. (5) Murphy MJ, Wei L, Watson, AD, MacDonald TM. ‘Real-life’ reduction in cholesterol with statins, 1993 to 2002. Br J Clin Pharmacol 2008;65:587 -92. Competing interests: Competing interests: MJ Murphy and TM MacDonald have received honorariums for lectures and advisory boards from several statin manufacturers. AD Watson has recruited patients for a Pfizer study. |
|||
|
|
|||
|
Les O. Simpson, retired medical research worker Dunedin, New Zealand 9077
Send response to journal:
|
In the two months since the publication of this editorial, not one of the rapid responses has mentioned the fact that hypercholesterolemic states are associated with an increase in blood viscosity. Furthermore, several studies have reported that when plasma cholesterol levels are raised, there is an increase in the cholesterol content of the red cell membrane. This change is accompanied by a reduction in red cell deformability which will have an adverse influence on capillary blood flow, as well as contributing to blood viscosity. In such situations, blood flow problems will be exacerbated when atherosclerotic changes are present and they will have an adverse effect on cardiac function. Appropriate dose levels of statins lower blood viscosity and restore red cell deformability, but as each case will have idiosyncratic features, it would be prudent to monitor lipid levels. However it is highly unlikely that the basis of the beneficial effects of statins will be understood without recognition of their effect on blood rheology. Competing interests: None declared |
|||
|
|
|||
|
Norbert Donner-Banzhoff, Professor of General Practice Department of General Practice, University of Marburg, D-35032 Marburg, Germany, Andreas Sönnichsen
Send response to journal:
|
Sir, We would like to comment on some of the responses to our editorial “Strategies for prescribing statins” • SIGN recommendations The SIGN guideline no. 97 specifies a total cholesterol target of 5 mmol/L. Anion exchange resins and even ezitimibe (!) are explicitly recommended if targets are not reached (p.35). Patients with cardiovascular disease should be treated “more aggressively”. The report titled Management of Coronary Heart Disease by the CHD Steering Group provides tables with drugs in increasing dose to reach targets in secondary care patients (p. 58). Although this approach differs from what lipid enthusiasts in the US tend to recommend, this is certainly not the “Fire-and-Forget-Approach” as suggested by Shepherd. • Real-life studies as opposed to RCTs To decide which approach is more effective, Dr. Murphy mentions two studies of his based on Scottish administrative data. Based on the results, he concludes that patients with repeated cholesterol measurements and dose increase (his definition of Treat-to-Target [TT]) fare better with regard to adherence to therapy and outcomes. Unfortunately these studies are fraught with several shortcomings and do not justify these conclusions. Even the authors of these studies mention the problem of imbalances between the two groups, a problem typically seen in retrospective observational studies. Moreover, there is the possibility of confounding by indication. Practitioners may have chosen the Fire-and-Forget-Approach [FF] in a particular patient because they expect difficulties with regard to adherence. In fact the FF group in their analysis was quite small. Since against the background of local recommendations (see above) this was “substandard” care, one can expect several patient and doctor characteristics to be associated with FF that have a negative impact on prognosis. On closer scrutiny the primary outcome, i.e. statin adherence, is no more than a measure of statin consumption. That in patients treated by the TT–strategy more statin is prescribed should not come as a surprise. The large proportion of small doses of simvastatin in their FF-group (simvastatin 10 mg – 41%; simvastatin 20 mg – 23%) cast doubt on their definition of the FF-arm of their analysis. Even the most sophisticated statistical models cannot account for confounding in such a retrospective observational approach. Possible confounders have to be identified beforehand, measured precisely and accounted for in a mathematical model. This can almost never be achieved retrospectively, there is always the possibility of unknown confounders and residual bias. Analyses of administrative data are even less reliable in this regard. However, the randomized controlled design - as chosen in the large statin trials - in its simplicity controls even for confounders that are completely unknown to the investigators. The advantages of this goldstandard should not be sacrificed for designs that promise increased external validity at a price of questionable internal validity. Instead we should focus on simple and large trials that preserve internal validity by randomization. Simple and pragmatic study procedures can help to include a broad spectrum of centres and/or patients without interfering with routine care. We admit that a randomized controlled trial investigating the issue we are discussing here would pose considerable difficulties. However, the results of such a trial would be far more reliable than the studies mentioned above. • Is the 4S study TT or FF? Due to limitations in space we could not provide detailed information on individual studies. There are few studies with minor dose adjustment early in the study (4S, AFCaps/TexCaps), and a much larger number with fixed dose (WOSCOPS, LIPID, HPS, PROSPER, ASCOT-LLA, the recent high dose trials). ALLHAT switched from dose adjustment to fixed dose during the study. Only CARE can be regarded as a titration study, however, with targets much higher than in current recommendations. If one compares inclusion criteria of early with later studies, criteria have become narrower with subsequent loss of external validity. Norbert Donner-Banzhoff, Andreas Sönnichsen 1) http://www.sign.ac.uk/pdf/sign97.pdf 2) http://www.sign.ac.uk/pdf/chdaudit.pdf 3) Wei L, MacDonald TM, Watson AD, Murphy MJ. Effectiveness of two statin prescribing strategies with respect to adherence and cardiovascular outcomes: observational study. Pharmacoepidemiol Drug Saf 2007;16:385-92. 4) Murphy MJ, Wei L, Watson, AD, MacDonald TM. ‘Real-life’ reduction in cholesterol with statins, 1993 to 2002. Br J Clin Pharmacol 2008;65:587 -92. 5) Yusuf S, Collins R, Peto R. Why do we need some large, simple randomized trials? Stat med 1984;3:409-20. Competing interests: None declared |
|||