Screening for familial hypercholesterolaemia

Familial hypercholesterolaemia is a common disorder of lipid metabolism associated with a high risk of early mortality from coronary artery disease.¹ It is so common that general practitioners often have one or two families with the disorder in their practices, although they are frequently unaware of this.² People with familial hypercholesterolaemia often die from atherosclerotic heart disease before the age of 40; this is particularly true for men. These sudden deaths are tragic because they can easily be prevented once the condition has been recognised and treated properly.³ It has been well established that the clinical sequelae of familial hypercholesterolaemia are the consequence of the extremely high concentrations of low density lipoprotein cholesterol that these patients have been exposed to since early childhood.¹

The most effective and most widely prescribed class of cholesterol lowering drugs, the statins, was shown to be particularly effective in patients with this condition.⁴ While it would seem logical to assume that treatment with statins would reduce mortality if family physicians simply diagnosed and treated this disorder, this line of reasoning is fraught with assumptions. Firstly, we must ask whether familial hypercholesterolaemia is indeed as common as we believe, and secondly, are we as inept in recognising and diagnosing this disorder as most of us believe? But the most important question is: how can we improve our recognition of these patients in general practice and offer them appropriate care?

Fortunately, recent work from both the United Kingdom and the Netherlands has given clear answers to these questions. A recent survey of four general practices in the Netherlands indicated that familial hypercholesterolaemia occurs in 1 in 400 people; these numbers are likely to be similar in the United Kingdom.² A study by Neil et al, which recruited patients from the Simon Broome register for familial hypercholesterolaemia, the Oxford lipid clinic, and general practices throughout Oxfordshire, found that not even one quarter of patients with the condition are recognised in clinical practice, and most are not diagnosed until middle age, when atherosclerotic disease is already rampant.⁵

The weak points, if any, of this careful, prospective endeavour include the low response rate of the general practitioners, the possible errors associated with self reporting of family relationships, and the use of absolute cut-off points for cholesterol for diagnosis instead of DNA testing. The authors' findings are not jeopardised by these weaknesses, but their work may underestimate the true frequency of familial hypercholesterolaemia.

Now that it has been firmly established that familial hypercholesterolaemia is underdiagnosed it should be possible to remedy this situation. Familial hypercholesterolaemia is a genetic disorder that is autosomal dominant and fully penetrant in adolescence; so by definition one patient with the condition will lead you to many more just by examining the patient's family tree.⁶

Durrington and colleagues have used this old wisdom, and in this issue of the BMJ (p 1497) show that it is indeed possible to use this approach to find new patients.⁷ By testing all first degree relatives of 200 patients with familial hypercholesterolaemia they found another 121 patients. In the general population at least 60 000 tests would have been needed to identify this many people with the condition. With the aid of a nurse specialist, simple cholesterol testing, and the use of small pedigrees Durrington and colleagues convincingly show that adopting an active approach to case finding works for familial hypercholesterolaemia.

Other investigators, including our group in the Netherlands, have come to similar conclusions, with two modest differences in approach. Firstly, testing in the Netherlands was not restricted to first degree relatives but included everyone in the extended family. This obviously reduces the proportion of people identified as having the disorder. On average, over a four year period one index patient led us to 20 additional family members, and eight new patients were identified (unpublished data). The second and most profound difference, however, lies in the use of DNA diagnostics. If the most sensitive test is used—namely age specific and sex specific centiles for total and low density lipoprotein cholesterol—16.6% of cases would have been missed and 12.5% would have been diagnosed as having familial hypercholesterolaemia when they actually had polygenic hypercholesterolaemia. Hence, active screening for a disorder requires a diagnosis that is rock solid, and that can only be provided by using DNA testing to actually find the genetic mutation causing the disorder.

Durrington et al correctly point out that the screening criteria developed by Wilson and Jungner easily apply to familial hypercholesterolaemia,⁸ but it is unlikely that DNA testing for the disorder has harmful psychological consequences.^9–11

We know how to organise the screening, and we have the capacity for testing, be it for cholesterol concentrations or DNA mutations. We also have safe and effective treatment that can save lives and money. Our ministries of health should not hesitate but should support screening and treatment programmes; a few specialised nurses working in close collaboration with lipid clinics could work miracles.

Papers p 1497