Intended for healthcare professionals

Letters

Cholesterol: how low is low enough?

BMJ 1999; 318 doi: https://doi.org/10.1136/bmj.318.7182.538b (Published 20 February 1999) Cite this as: BMJ 1999;318:538

Effect of a given concentration depends on several factors

  1. A R P Walker, Head (alexw{at}mail.saimr.wits.ac.za)
  1. Human Biochemistry Research Unit, Department of Tropical Diseases, School of Pathology of the University of the Witwatersrand and the South African Institute for Medical Research, Johannesburg, South Africa
  2. King's College, St Thomas's Hospital Campus, London SE1 7EH
  3. Department of Chemical Pathology, Burton Hospital, Burton on Trent DE13 0RB

    EDITOR—In his editorial Rosengren considers that reaching target cholesterol concentrations may be better than relative reductions.1 He stated that “in observational studies a prolonged difference in usual serum cholesterol value of 0.06 mmol/l is associated with an almost 30% reduction in risk of coronary disease.”

    No one would question the need for reducing serum cholesterol concentration in large proportions of adults in Western populations. However, the subject of cholesterol concentration and its pathological importance is complex.

    Known risk factors for coronary heart disease, of which serum cholesterol concentration is one,explain only half of the variance in the occurrence of the disease.2 There are also numerous contextual problems. For example, in the Sheffield risk table, cholesterol reduction may be called for at 5.5 mmol/l in those at high risk, whereas intervention may not be needed until 9.0 mmol/l in those at low risk.3 The experience of coronary heart disease also differs between Belfast and Toulouse. Although mean cholesterol concentrations in the two cities are similar, 6.19 and 5.94 mmol/l, mortality from the disease is 3-4 times higher in Belfast than in Toulouse.4

    Perplexities are also common in populations in developing countries, as in South Africa.5 In early studies mean cholesterol concentration in middle aged African men was about 4.0 mmol/l, with observations at necropsy showing a low rate of atherosclerotic lesions of the aorta and a negligible number of deaths from the disease. Currently, in Soweto (3-4 million inhabitants) the mean cholesterol concentration is about 5 mmol/l. Yet coronary heart disease remains uncommon, being responsible for less than 0.5% of total deaths. Similar reports of the disease's comparative rarity have emanated from big cities in other countries in Africa. In comparison, in the seven countries study the same mean serum concentration of cholesterol (about 5.15 mmol/l) prevails in Mediterranean countries, but coronary heart disease is responsible for 4.7% of total deaths.

    An additional complicating factor is the wide range of cholesterol concentrations in a community. In African village schoolchildren, almost all of the same poor socioeconomic state and accustomed to the same low atherogenic diet, cholesterol concentration varied from 2.5 to 4.2 mmol/l.

    To reiterate, although cholesterol concentration has a role in coronary heart disease, a given concentration has widely different connotations for ill, being affected by familiality, ethnic group, sex, and environmental factors.

    References

    Doctors have been slow in getting evidence on lowering cholesterol into practice

    1. A S Wierzbicki, Senior lecturer in chemical pathology (p.lumb{at}umds.ac.uk),
    2. T M Reynolds, Professor
    1. Human Biochemistry Research Unit, Department of Tropical Diseases, School of Pathology of the University of the Witwatersrand and the South African Institute for Medical Research, Johannesburg, South Africa
    2. King's College, St Thomas's Hospital Campus, London SE1 7EH
    3. Department of Chemical Pathology, Burton Hospital, Burton on Trent DE13 0RB

      EDITOR—Rosengren rightly states that treatment of patients with established coronary heart disease is mandatory.1 Prolonged statin treatment reduces events by more than 30%: the seven year results of the Scandinavian simvastatin survival study show a 45% reduction (T Pedersen, drugs and lipid metabolism symposium, Florence, June 1998). However, the suggested target concentration for low density lipoprotein cholesterol of 2.6 mmol/l differs from the recommendation of the Standing Medical Advisory Committee of <3.2 mmol/l after infarction. There is little evidence for which target is correct, except in patients with coronary bypass grafts (<2.6 mmol/l).

      The first infarction is fatal for many patients. One study in both sexes showed a 34% reductionin events starting at a risk of 1.2 % per year,2 and the west of Scotland coronary prevention study in men showed a 33% reduction in mortality at a risk of 1.5% per year.3 Yet, Ramsay et al still suggest using a threshold of 3% per year rather than the internationally recognised 2% per year.4 The Sheffield tables are an attempt to ration statin treatment despite the evidence not because of it.

      Patients with familial hyperlipidaemias are common (1 in 250) and have a large excess mortality and morbidity at an early age when calculated risk does not approach 1% per year. The Sheffield tables fail to define patients with a familial hyperlipidaemia accurately as physical signs are often absent and the penetrance of coronary heart disease may be poor. In the case of a euthyroid patient with a total cholesterol concentration of 10 mmol/l (after exclusion of renal disease, which carries a fourfold increased risk of coronary disease) the chances of this being due to non-familial causes are <1 in 10 000.

      We agree with Ramsay et al that general practitioners ought to decide. Yet it seems strange that opinion from lipid specialists on the utility of the Sheffield tables is not worthy of note by others concerned with the production of guidelines. The tables cited by McLeod and Armitage or the European societies allow general practitioners to calculate an approximate risk and then decide inthe light of international recommendations.5 In contrast, a prescriptive approach is adopted in the Sheffield tables. Guidelines excepted, it is a depressing fact that most patients with coronary artery disease are still not receiving statin treatment fouryears after the publication of the Scandinavian simvastatin survival study and that most of those treated are treated inadequately.

      References

      View Abstract

      Log in

      Log in through your institution

      Subscribe

      * For online subscription