Dr. Neville S. Wilson. 31 August, 2007.
The review and revision of Clinical Guidelines for purposes of risk stratifying and risk reduction in cardiovascular disease prevention is an ongoing challenge and current initiatives focus on their implications for primary care providers.
The AHA guidelines for 2007 offer a timely and broader spectrum of risk not previously captured by the traditional Framingham algorithm. (1)
In the UK (QRISK) and Scotland’s (ASSIGN) new scoring methods were recently advocated and tested against the Framingham equations and judged to be more accurate, by taking account of social deprivation as an additional factor of risk for CVD.
And the Fourth European Guidelines on Cardiovascular Disease Prevention in clinical practice is shortly to release its findings and recommendations.
The search for an updated and appropriate scoring tool justifies scepticism about the validity and limitations of traditional models and also the need for a critical appraisal of hypotheses underlying such models.
Furthermore, the central issue which demands attention in the clinical setting is the uniqueness of the individual in risk assessment. The clinical implication is for a focused and global assessment of the PERSON BEHIND THE NUMBERS with respect to a broader range of risk factors than previously considered.
THE PERSON BEHIND THE NUMBERS
Looking at the “Person behind the numbers” is a familiar slogan widely publicised by a well known manufacturer of a popular cholesterol lowering agent.
The slogan is arresting because it embodies a principle that is central to health care and serves as a precept that health professionals dare not ignore, and that is the uniqueness and importance of the individual. The Person being treated must remain ever central in healthcare delivery and individualising their wellness management must take precedence over blanket prescriptions dictated by statutory guidelines for risk management.
However, despite this worthy sentiment, numbers and statistics often predominate, and have increasingly become the focus of promotional guidelines designed to shape the prescribing habits of doctors.
Public health messages reflect a perception that morbidity and mortality reducing strategies are best served by aggressive reduction of patient plasma lipid levels with little reference to other vital parameters, and current guidelines are poised to perpetuate this perception and promote its application.
Opposing these initiatives are groups of independent researchers worldwide who question the validity of the evidence driving this programme, and warn of the consequences of aggressive and unwarranted lipid lowering strategies.
The concerned family doctor is caught between these polarised viewpoints and is burdened with the balancing act of providing evidence-based medical care while protecting his/her patients from the risks of unwanted consequences which can be debilitating and devastating, both for individuals and families, and which generally have been under-estimated and under-reported. (2)
It thus becomes an imperative for practitioners, against this background of rapidly changing targets, to make a critical appraisal of all recommendation regarding guidelines and to resist inclinations of blind submission to “recommended guidelines” without an individualised assessment of the patient’s peculiar and unique circumstances.
Looking at the “PERSON” behind the numbers, must take precedence over the figures on the laboratory report, and the age-old question must be asked, “Who is the PERSON having the disease, and not, “What disease does this person have?”
THE VULNERABLE PATIENT
Guidelines take account of potential risks to health and are driven by the desirable outcomes of reduced morbidity and improved mortality. They thus provide a working tool for evaluating and stratifying a wide range of risk factors, often modifiable, enabling physicians and patients to thereby engage in mutually agreed strategies aimed at patient wellness and longevity.
Guidelines for risk assessment have traditionally focused on lipid levels, hypertension and smoking habits with reference to age and gender, but a wider spread of factors that are potentially unfavourable to health have been ignored in such models, and these deserve appraisal for a realistic assessment of patient vulnerability.
Social and domestic circumstances, lifestyle habits, nutritional status, stress burdens, family and genetic histories, as well as age, gender, and ethnicity, in addition to traditional factors, demand assessment if we are to understand the Person before us in the clinical setting, since each of these has the potential for life-threatening consequences.
THE FRAMINGHAM MODEL – ITS WEAKNESSES
The Framingham studies serve as a reference base for scoring risk and produced figures that were in fact related to all cause mortality and not to cardiac mortality. (This fact is not widely recognised by practitioners ! )
Furthermore, the data for those who suffered heart attacks reveal that, contrary to expectation, almost half of these subjects had low plasma cholesterol levels. (3)
Also, 30-year follow up survival curves showed that those with higher cholesterol levels actually lived longer, and that mortality rates for males older than 48 years were no different for those with either high or low cholesterol levels.
These investigators reported an 11% increase in cardiovascular mortality for every 1% mg/dl decrease in cholesterol levels. (4)
The Framingham data clearly reveal, contrary to the generally held view, that low cholesterol levels do not prolong life any longer than high levels.
Despite these clear revelations, the National Heart Lung and Blood Institute made the following statement, which clearly contradicts the findings of Framingham: “A 1% reduction of cholesterol corresponds to a 2% reduction in CHD risk.” (5)
It should be a source of great concern to practicing physicians, that not infrequently, professional bodies issue proclamations and subsequent press releases that distort, and thereby falsify facts, and project perceptions of benefit which have not been demonstrated. In so doing, they demolish their own credibility.
In the light of these undesirable trends it is quite legitimate to ask why guidelines continue to selectively focus on cholesterol levels as a major risk threat when the evidence is lacking.
EXAGGERATED TRIAL CLAIMS
it should also be noted that trial summaries usually report data for cholesterol reduction in relative rather than absolute risk percentages, thereby grossly exaggerating the perceived benefits and conveying to the public a distorted message about the dangers of cholesterol.
One of many examples is evident in announcements following the LIPID RESEARCH CLINICS CORONOARY PREVENTION TRIAL (LRC-CPPT) designed to test the efficacy of lipid lowering in reducing the rate of CHD. A 24% reduction in CHD and a 19% reduction in non-fatal MI was reported and hailed as “strong evidence for a causal role for these lipids in the pathogenesis of Coronary Heart Disease”. (6)
In reality no such “strong evidence” existed. This “benefit” of cholesterol-lowering was a relative % risk only, and when translated into absolute risk (AR) reduction reflected a mere 0.6% reduction for CHD death and a mere 1.1% reduction for non-fatal MI.
This gross manipulation of figures was later acknowledged by Dr. B. Rifkind who had been the study Director. (7)
Likewise, the figures for MRFIT trial purported to support the hypothesis that high cholesterol levels caused heart disease, but the data supporting this perception was shown to be inaccurate and carefully produced, and at times purposely falsified, according to Prof. Lars Wenko ( 8)
The Framingham revelation is not unique.
Numerous studies, including those from Canada, Britain, Finland, and France have produced similar conclusions. High levels of cholesterol do not decrease mortality, and yet traditional guidelines persist in targeting cholesterol as a major culprit and agent in the development of cardiovascular disease.
No evidence has ever been produced implicating cholesterol as a causative agent in cardiovascular or total mortality, and patients need to hear these honest facts from their physicians.
It is worth considering that to withhold this fact from patients is to unnecessarily burden them with anxiety and needless stress, and thereby increase their risk for morbidity . ( ? Negligence )
The WHO MONICA project tracked the rates of cardiovascular death compared to cholesterol levels in several European countries and revealed a clear pattern of inverse relationship between cholesterol levels and cardiovascular death.
Countries with the highest average cholesterol levels like Switzerland had the lowest cardiovascular death rate and those with the lowest cholesterol levels, like Russia, and the highest death rate. These were figures for males aged 35-74 years. (9)
The selective targeting of bio-chemical risk factors and exclusion of lifestyle, social and nutritional factors diminishes the power of risk stratification and ignores the real Person behind the numbers.
It is thus appropriate for researchers like Drs Abramson and Wright to ask whether lipid lowering guidelines are evidence based.
In their response to the 2005 meta-analysis reported by the cholesterol treatment trialists, (CTT) they present results from 8 randomised clinical trials in which no total mortality reduction by cholesterol lowering with statins was evident. (10)
With reference to cardiovascular events the “absolute risk reduction (ARR) of 1.5% was so small that 67 people would have to be treated for 5 years to prevent 1 event”, they reported.
They acknowledge that only in a select group of high risk males, between the ages of 30 and 69 years, there might be benefits. They furthermore contend that statins do not reduce CHD events in almost 11,000 women in pooled trials, nor do they reduce CHD events in men and women older than 69 years.
Wright states that physicians should be honest with their patients about the lack of evidence in low risk patients and the low absolute difference in relative risk even in high risk patients requiring primary prevention.
He concludes, “if physicians were truly honest with their patients, I think there probably would be very few people being treated for primary prevention with a statin.”
An honest appraisal of the trials that tout benefits for cholesterol lowering provides compelling evidence for a cautious approach to interpreting numbers and their significance until other factors are taken into account.
“Who is the Person, rather than “What are the numbers.”
LOWER TARGETS (TNT)
Directives from the NCEP are aimed at lowering lipid targets with aggressive statin therapy in accordance with TNT recommendations. (11)
Physicians would better serve the health interests of their patients by a careful appraisal of individualised circumstances and assessment of risk vs. benefit in the light of these proposals.
Proponents of the aggressive approach argue that while the cautious lobby disagrees and calls for further assurances of safety before implementing such targets. (12)
The conclusion of authors of a study which measured the degree of plaque regression when treated with statins contradicts this notion. They clearly state “lower is not better.” (13)
Dr. Bertram Pitt calls for a cautionary approach after noting that while in TNT 26 fewer CHD deaths were reported in the high dose group (atorvastation 80mg), 32 more deaths occurred in this group, again demonstrating that even with intensive lowering of cholesterol levels, mortality was not improved.
An important question raised by Dr. Pitt was whether LDL-C is the most effective target of lipid lowering strategies. He suggested that possibly the ratio of apolipoprotein (Apo)B to (Apo)A-1 might be a better indicator of cardiovascular risk. (14)
NO MORTALITY BENEFITS
A secondary analysis of TNT was undertaken by Wenger at al to assess the efficacy and safety of high dose atorvastation in patients 65 years of age or older. Her conclusions suggested a clinical benefit with aggressive reductions of LDL-C to less than 2.6 mM/L.
However the trend towards increased death from cancer in this group supports concerns about a possible dose related potential for cancer in the elderly.
Concerns for cancer related therapy were also expressed when reviewing the outcomes of PROSPER (24 deaths in the treatment group). (15), CARE (12 cases of breast cancer), J-LIT (12 cancer deaths).
Other trials which touted “benefits” from lipid-lowering, and in fact failed to demonstrate mortality benefits, were REVERSAL, EXCEL, ALLHAT, and ASCOT, the latter exaggerating benefits by reporting a 27% reduction in fatal and non-fatal stroke, when in fact the comparative figures were 1.7% (statins) and 2.4% (control) which equates to an absolute difference 0.7%, which is hardly a “significant benefit”. (16)
Thus again, there has been no evidence forthcoming for mortality benefits with cholesterol lowering, and the question, “Are the guidelines evidence based?” must again be asked.
The fact that nearly 50% of ACS events occur in patients with normal to below normal lipid levels is strong evidence for a non causal role for cholesterol in cardiovascular disease.
Its association with atherosclerosis may be linked to the repair process following intimal injury, together with several other factors which may also be present, thus serving as useful biomarkers.
The measurement of these “non-traditional” bio-markers may provide an alternative and superior model for risk assessment and their incorporation into guidelines makes good sense.
Circulating “non-traditional” bio-markers are currently being evaluated for their potential to indicate risk for cardiovascular disease.
Noteworthy amongst these are CRP, apolipoprotein B, homocysteine, fibrinogen, insulin and iron, although controversy surrounds their utility value.
More sophisticated emergent bio-markers for optimising risk stratification are myeloperoxidase, lipoprotein a, isoprosteines, small dense LDL, and oxidized LDL. (17)
The value of these bio-markers accrues from their capacity to indicate inflammation in the arterial tree which is the underlying mechanism implicated in atherogenesis.
The widely held view of CHD occurring as a consequence of vessel occluding deposits of cholesterol “sludge” is no longer scientifically tenable, and should not be propagated by health professionals or medical commentators, since it leads to false perceptions of “cholesterol plugging” and unwarranted public anxiety about cholesterol levels.
This false perception is widely propagated in the media and promoted by several lines of “health information”.
Cholesterol does not plug arteries, and the recent statement by a visiting academic about “plugging up the works” (18) is unscientific, inaccurate, and misleading, and clearly disappointing !
Nor do LDL particles “stick to the arterial wall,” as he suggested.
Atheroma occurs not in the vessel lumen, but in the vessel wall, and involves a complex mechanism of inflammatory response to intimal injury, possibly as a consequence of micro-environmental changes linked to factors such as ageing, sepsis, and/or overproduction of reactive oxygen species (ROS).
No evidence exists for the causative involvement of cholesterol in atheroma formation, and its presence in calcified plaque is thought to represent a protective activity. Reducing cholesterol levels has not been shown to reduce plaque.
LOW DENSITY LIPO-PROTEIN ( LDL-C)
It has long been known that LDL itself is not a reliable independent risk factor for CHD and that oxidized LDL, which represents altered LDL in the arterial wall, has a stronger association with atherogenesis.
Furthermore, “lower is not better”, is the conclusion of the authors in their study on plaque regression, (19) supported by other investigators, declaring that LDL lowering is not beneficial. (20)
Neither has intensive LDL lowering conferred benefits for patients with calcified aortic stenosis. The SALTIRE trial demonstrated a failure of statins to halt the progression of calcified aortic stenosis, and the authors state, “we do not recommend statin therapy for patients with calcified aortic stenosis in the absence of coexisting vascular disease.” (21)
COEXISTING VASCULAR DISEASE
A meta-analysis in 2004 by Pierre Amarenco demonstrated the benefits of statins in stroke reduction in patients with documented CHD, and young men with pre-existing cardiovascular disease and diabetes have demonstrated benefit from satin therapy (CARDS).
Likewise, patients with previous ischaemic stroke were shown to have a 16% reduction benefit for recurrent stroke and a 35% reduction for a major cardiovascular event with high dose atorvastatin (SPARCL). The risk/benefit ratio, however, for haemorragic stroke remains a question of debate. (22)
Also, in a recent study, statin withdrawal after vascular surgery appeared to be associated with an increased risk for peri-operative cardiac events, suggesting a benefit for extended release fluvastatin in these patients. The benefits of using statins in these groups of patients has been clearly demonstrated. (23), but these benefits are likely to be due to pleiotrophic effects (anti-inflammatory ) and not to cholesterol lowering .
THE ROLE OF HDL
The beneficial effect of intensive statin therapy on the rate of plaque regression was shown by Nicholls et al in an analysis from 4 previous trials, but no significant differences were found regarding clinical outcome with either the higher or lower dose statin.
The likelihood of atheroma regression resulting from a small increase in HDL (7.5%) is supported by the GREASE Study which demonstrated the clinical benefits of small increases in HDL. (24)
THE ROLE OF OXIDIZED LDL
Unlike plasma cholesterol and circulating LDL, oxidized LDL (OX-LDL) correlates well with lipoprotein (a) in the presence of acute coronary syndromes (ACS), suggesting thereby a potential role as a plasma marker for risk.
Also, in patients with metabolic syndrome, high levels of OX- LDL were associated with a greater predisposition to atherothrombotic coronary disease.
There are numerous research reports demonstrating the harmful effect of oxidized LDL (OX-LDL) on vascular endothelium (eg. Navab, Holvoet), and growing evidence for a causal role in atherosclerosis for reactive oxygen species (ROS).
The protective role of anti-oxidants in ROS reduction, despite the disappointing outcome of recent antioxidant trials, is considered by some researchers to provide avenues for future research. (25)
The positive effects of N3 – polyunsaturated oils (Omega –3 ) in this regard have been noted.
A recent Canadian study evaluated the effect of nutritional intervention on circulating OX-LDL showing a distinct beneficial reduction of OX-LDL without any changes in plasma LDL-C or LDL levels. (26)
These findings support earlier observations in the Lyon Diet Heart Study that LDL-C levels were not altered in the reduction in myocardial infarction ( heart attack ) risk in response to a Mediterranean diet, as did the ATTICA study (2004) which demonstrated reductions in OX-LDL but not LDL-C in response to a Mediterranean diet. (27)
These findings have implications for non-pharmacological intervention by physicians when reviewing the nutritional status of their patients.
A broader view of individualised risk factors in respect of personal and lifestyle habits, social deprivation, nutritional vulnerability, and emotional stress deserves attention and intervention, before resorting to pharmacological manipulation of biochemical numbers.
The under-utilization of N-3 polyunsaturated fatty acids (Omega 3) as a powerful anti-inflammatory nutrient is sadly evident in clinical practice, and could provide a clinically proven strategy for modulating the progression of atherosclerosis and reducing MI mortality by 20%. (29) as well as reducing elevated plasma triglyceride levels.
BENEFITS OF CHOLESTEROL
Managing risk involves being honest with patients, and the unfounded fear of cholesterol needs to be addressed by physicians when counselling patients, particularly the elderly, where its protective role has been repeatedly demonstrated in trials.
Apart from the normal physiological benefits of cholesterol in respect of hormone synthesis (testosterone, oestrogen, cortisone), the synthesis of Vitamin D, the production of digestive bile acids, immune system enhancement, maintenance of healthy nerve myelin sheaths and stabilization of cellular membranes, several trials have demonstrated a strong association between high levels of cholesterol and reduced mortality from bacterial and viral infections.
The ability of lipo proteins, especially LDL, to neutralise harmful bacterial endotoxins implicated in a wide range of critical infections is indicative of the immunoprotective effects of cholesterol and contradicts the prevailing paradigm that cholesterol is harmful and is to be feared. (30)
LOW CHOLESTEROL LEVELS
Low blood levels of cholesterol, on the other hand, have been shown to correlate closely with neurological effects such as violent and aggressive behaviour, mood disturbance, and even depression and suicide.
The unnecessary reduction of plasma cholesterol levels has the potential to compromise health and longevity, while higher untreated levels, especially in women and elderly, appear to prolong life with less risk of drug induced adverse events.
The beneficial effects of higher cholesterol levels in heart failure patients has been demonstrated and explained in terms of lipoprotein modulation of inflammation, while low levels of circulating lipoprotein and cholesterol have been shown to be independent predictors of impaired outcome in heart failure patients, but the association cannot reliably be shown to be causal.
The Fourth European Guidelines on Cardiovascular Disease Prevention in Clinical Practice represents an update version of the 2003 Guidelines and will attempt to broaden the base for risk assessment by including lifestyle parameters hitherto ignored.
In so doing it is not likely to be “simpler” than previous guidelines (as requested) but more appropriate, and serve better as a means of understanding the “Person behind the numbers.”
The Guidelines, therefore, call for more than a mere manipulation of lipid numbers, since they will challenge us to evaluate individual vulnerability, with considered use of appropriate, non-traditional biomarkers, and options for non-pharmacological interventions at appropriate levels of risk.
Dr. Neville Wilson
The Leinster Clinic
August 31, 2007
1. Circ. 2007; 115—1481-1501
2. Statin Drugs – Side Effects Duane Graveline M.D. 3. Annals of Int. Med. 90; 85-91, 1979
4. JAMA 1987, April 24, 257, 2176-2180
5. Circ. 81, 1721-33, 1990
6. JAMA 251 (3) 351-64, Jan. 20, 1984
7. BMJ 301, 815, 1990
8. J. of Int. Med. 237, 507-578, 1995
10. Lancet 2007, 369: 168-169
11. NEJM Med 2005, 351
12. BMJ 206; 332: 1330-1332
13. AMJ Cardiol. 2003: 92: 334-336
14. NEJ Med 2005 (www.nejm.org)
15. Lancet 2002; 360; 1623-30
16. Lancet 2003; 361 (9364): 1149-58
17. J. Am College of Cardiol. 2006; 47: 19-31
18. IMT 20/4/07
19. AMAJ Cardiol. 2003: 92: 334-336
20. NEJM 2005, 325; 2889
21. NEJM 2005, 352: 2389-2397, 2440-2443
22. STROKE 2004; 35: 2902-2909
23. AMJ Cardiology 2007; July 15: 100(2) 316-321
24. JAMA 2007; 297; 499-508
25. Athero. Thromb. & Vasc. Biol. 2005; 25-29
26. J. Nutr. 135; 410-415
27. Atherosclerosis 173; 353-367
28. Eur. J. Med. Res. 2003 aug. 20 (8) 337
29. Lancet 354; 447-455
30. QJMed 2003; 90; 927-934