Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Cascade testing from FH adult probands : the tried and tested and cost effective option.
- Steve E Humphries, Gaye Hadfield (14 September 2007)
-
First, do no harm
- Keith A Hopcroft (23 September 2007)
-
Screening for familial hypercholesterolaemia; the need for a fresh approach
- David S Wald, Jon P Bestwick, Nicholas J Wald (26 September 2007)
-
Screening for familial hypercholesterolaemia: any harms are small in relation to the benefits
- David S Wald, Jon P Bestwick, Nicholas J Wald (27 September 2007)
-
Familial hypercholesterolaemia: time for further action, not further debate
- Ian Hamilton-Craig (5 October 2007)
-
Screening for familial hypercholesterolemia
- Peter J. Lansberg, Maud N. Vissers, Hans J. Avis, Barbara A. Hutten, John J.P. Kastelein (5 October 2007)
|
|
|||
|
Steve E Humphries, BHF Professor of Cardiovascular Genetics Royal Free and University College London Medical School, University Street, WC1E 6JJ, Gaye Hadfield
Send response to journal:
|
Gaye Hadfield1, and Steve E. Humphries1,2, 1 London IDEAS Genetics Knowledge Park, London, UK, 2Centre for Cardiovascular Genetics, British Heart Foundation Laboratories, Royal Free and University College London Medical School, London, UK In today’s BMJ, Ward et al suggest that screening children for high cholesterol at the time of childhood immunisation, and then testing the parents of the identified children “elegantly screens for familial hypercholesterolaemia in two generations simultaneously…with the potential of preventing premature coronary heart disease in nearly everyone with the disorder.” This is an interesting model to find more patients with this life- threatening but treatable disorder, and NICE are preparing guidelines for the identification and management of patients with FH which are due to be published in August 2008. However, the paper does not address the important issue of acceptability of the test to parents (and therefore take-up rate), nor the cost of the programme (and therefore the cost- effectiveness). There is almost no data that addresses the issue of the likely acceptability of such a test, but a small study of testing for high cholesterol in newborns in London (1) concluded that their “pilot data raise questions about the extent to which assessing disease risks by (genetic) DNA analysis may result in a sense of fatalism, adversely affecting motivation to change behaviour and to reduce risks.” Clearly caution is needed. With regard to the cost, earlier data from our group (2), showed that “universal-screening” was not cost effective, and that “cascade-testing” the relatives from known FH patients (where 50% will be affected) is a better approach. With funding from the Department of Health (DH) www.fhcascade.org.uk we have recently completed a 5-site pilot of cascade testing. We found that cascade testing, using specially-trained nurses integrated in Lipid Clinics, was feasible, desirable, cost-effective and acceptable to patients and clinicians. The recommendations of the pilot, which have been sent to the DH, are that effective implementation will require the development of a suitable infrastructure to link clinics together (to ensure cascading can be effective nation-wide), that specialist services for children with FH will be required (currently very few are available), and that an improvement in the awareness of FH in Primary Care will be vital. Most importantly, the data suggest that the application of DNA testing to underpin the diagnosis of FH in index cases and relatives will be needed. Such a DNA-based approach to cascade testing for FH has been very successfully running in the Netherlands for more than 5 years (3) and has demonstrated excellent response rate and acceptability, a low cost per new patient found (4) and a cost effectiveness in terms of cost per life- year gained that is well within current NICE guidelines. Currently we estimate that roughly 15,000 of the predicted UK 110,000 FH patients are being treated in lipid clinics (5), so cascading from them should be the first approach. DNA testing will be a major driver for this process, and is becoming cheaper and faster. Currently a mutation can be detected in up to 80% of patients with the strongest clinical suspicion of the disease (6), a detection rate which compares very favourably with other diseases where genetic testing is now routine such as in familial Breast Cancer. DNA testing in adults with FH appears to have little adverse psychological impact (7) and may be associated with benefits if genetic testing motivates better compliance with medication use. Finding more new patients to cascade from will also be important, and using children may be feasible, but this could also be done by testing patients who have had a heart attack at a young age, or looking at GP records, the feasibility of which (although not the cost effectiveness) has recently been reported from a practice in London (8). Clearly the proposal to screen infants has clinical merit but would need to be piloted carefully. The data we have from the DH pilot suggests that, in reality, the overlap in cholesterol levels between FH and non-FH children is much higher than used in the Wald et al model. This will mean many “false-positive” children (and their parents) will end up being tested unnecessarily, or that many true-positives will be missed (because the cut-off is set too high). Importantly, the psychological impact on the family will need to be monitored at the time of testing and afterwards for some years. Currently the tried and tested approach of cascade testing from known FH patients appears to us to be the best option, and we estimate that with adequate funding this could find up to 50% of the expected cases in the UK within 7-10 years. 1. Senior V, Marteau TM, Peters TJ.Will genetic testing for predisposition for disease result in fatalism? A qualitative study of parents responses to neonatal screening for familial hypercholesterolaemia. Soc Sci Med. 1999 Jun;48(12):1857-60. 2. Marks D, Wonderling D, Thorogood M, Lambert H, Humphries SE, Neil AW. Cost-effectiveness analysis of different approaches of screening for familial hypercholesterolaemia. BMJ 2002;324:1303-1306 3. Umans-Eckenhausen, M. A., J. C. Defesche, E. J. Sijbrands, R. L. Scheerder, and J. J. Kastelein, 2001 Review of first 5 years of screening for familial hypercholesterolaemia in the Netherlands. Lancet 357: 165- 168. for familial hypercholesterolemia. Clinical Genetics 66: 483-487. 4. Wonderling D, Umans-Eckenhausen MA, Marks D, Defesche JC, Kastelein JJ, Thorogood M.Cost-effectiveness analysis of the genetic screening program for familial hypercholesterolemia in The Netherlands. Semin Vasc Med. 2004 Feb;4(1):97-104. 5. Marks D, Thorogood M, Farrer M, Humphries SE. Census of clinics providing specialist lipid services in the United Kingdom. J Public Health Med 2004;26(4):353-4 6. Humphries SE, Whittall RA, Hubbart CS, Cooper JA, Soutar AK, Naoumova R, Thompson GR, Seed M, Durrington PN, Miller JP, Betteridge DJB, Neil HAW. Genetic causes of familial hypercholesterolaemia in UK patients: relation to plasma lipid levels and coronary heart disease risk. J Med Genet 2006;43:943-949 7. Marteau, T., V. Senior, S. E. Humphries, M. Bobrow, T. Cranston et al. 2004 Psychological impact of genetic testing for familial hypercholesterolemia within a previously aware population: a randomized controlled trial. Am.J.Med.Genet.A 128: 285-293. 8. Gray J, Jaiyeola A, Whiting M, Model M, Wierzbicki A.Identifying patients with Familial Hypercholestrolaemia in primary care: an informatics-based approach in one primary care centre. Heart. 2007 Jun 17; [Epub ahead of print] Competing interests: None declared |
|||
|
|
|||
|
Keith A Hopcroft, GP Laindon Health Centre, Laindon, Basildon, Essex SS15 5TR
Send response to journal:
|
Sir Wald et al's suggestions for child-parent screening for familial hypercholesterolaemia are interesting and innovative. But, in their statement that their proposals fulfil eight out of the 10 requirements for a worthwhile screening programme, they're also somewhat glib. Using whatever criteria for 'worthwhile screening' you care to choose, there is clearly an implication that you should be doing more good than harm. Indeed, 'First do no harm' is one of the basic ethical principles of medicine. However, the potential for screening of this sort to have an adverse psychological impact isn't mentioned - but is relevant in two ways. First - no matter how well these issues are communicated to the public - the concept of checking cholesterol levels in babies will exacerbate the cholesterol neurosis which much of the population already seems to suffer. Worthy and well intentioned pronouncements from on high are all well and good, but GPs in particular are on the sharp end of this type of initiative - and are only too well aware that a screening programme which sounds fine in theory can translate into an impenetrable and unquantifiable mass of anxiety in patients when viewed from the grass roots level. Throw into the mix the fact that children are involved and the emotional charge and the anxieties raised will be that much higher than usual. Second, the specific potential psychological impact of this novel type of programme on parents surely warrant caution. For a parent to be told that their child has a high cholesterol will inevitably cause a great deal of worry, and being paralysed by fear renders putting things into perspective very difficult. There is a real danger that a shadow will be cast over childhood because parents suddenly view their progeny as threatened or fragile on the basis of a screening test. This is likely to be compounded by the fact that no immediate therapeutic action can be taken. The parent will be left with the feeling that their child is vulnerable and that nothing can done about this in the immediate term. At best this will make them uncomfortable, at worst it could blight their years of parenthood. The authors suggest that treatment to lower cholesterol would be delayed in the child until adulthood. At which point - after a childhood in which they were perhaps aware of being perceived as less 'well' than their peers and of being over-protected by their parents - they may or may not decide to take treatment and may or may not be happy that this test was ever imposed on them. Perhaps their subsequent thwarted efforts to obtain life insurance and the like will help them make up their mind. Keith Hopcroft FRCGP Competing interests: None declared |
|||
|
|
|||
|
David S Wald, Consultant Cardiologist and Senior Lecturer Wolfson Institute of Preventive Medicine, Barts and the London School of Medicine, London, EC1M6BQ, Jon P Bestwick, Nicholas J Wald
Send response to journal:
|
Hadfield and Humphries favour cascade testing for familial hypercholesterolaemia (identifying relatives of affected individuals) over our proposed child-parent screening approach 1 but concede that at present cascade testing may only identify 50% of affected cases in the population. Cascade testing is useful once a family is identified for testing.2 Identifying all, or most of the families first is a pre-requisite for cascade testing to be a practical proposition.3 Our child-parent screening proposal would identify most families so that the programme could switch to cascade testing in the long-term. Without a general screening approach to find index cases in the population, cascade testing, or other so-called “case finding” approaches (such as identifying families from a general practice register with a family history of a young myocardial infarction) is piecemeal and has limited overall benefit. Expert groups have not recommended a general screening approach because of uncertainty regarding the performance of cholesterol measurement as a screening test for familial hypercholesterolaemia and how well such tests discriminate between affected and unaffected individuals. Our study provides a reasonable answer based to this and shows that nearly all cases (about 90%) can be identified with a very low false-positive rate (about 0.1%) in the age group 1-9 years. This finding underpins our proposal and provides a basis for general screening. Hadfield and Humphries state that they have data from a Department of Health funded pilot study that suggests that the overlap in cholesterol levels between children with and without familial hypercholesterolaemia is higher than that presented in our meta-analysis. They do not show the data or quantify the effect this has on screening performance. It would be useful to see any new data they have on this and assess the extent to which it may alter our estimate. Hadfield and Humphries advocate DNA testing to confirm cases of familial hypercholesterolaemia. Whilst this may be sensible once an affected family is identified (such that only the mutation causing the disorder need be tested for within families), this would miss about 20% of affected children and add considerable cost to a population screening approach (DNA tests that identify about 80% of mutation cost about £400- £500 per person). For this reason cholesterol measurement alone may be a preferable option in the child-parent screening we propose, recognising that a pilot study would be needed to determine the cost and acceptability of the two approaches. Cascade testing may have been “tried and tested” and found to be useful in identifying cases of familial hypercholesterolaemia within families, but it is not “tried and tested” as a general screening policy. A large proportion of affected cases would be missed, probably more than half. References 1. Wald DS, Bestwick JP, Wald NJ. Child-parent screening for familial hypercholesterolaemia: screening strategy based on a meta-analysis. BMJ 2007 335:599 2. Is cascade testing a sensible method of population screening? J Med Screening 2004;11(2):57-58 3. Morris JK, law MR, Wald NJ. is cascade testing a sensible method of screening a population for autosomal recessive disorders? Am J Med Genet 2004;128A:271-5 Competing interests: None declared |
|||
|
|
|||
|
David S Wald, Consultant cardiologist and Senior Lecturer Wolfson Institute of Preventive Medicine, Barts and the London School of Medicine, EC1M 6BQ, Jon P Bestwick, Nicholas J Wald
Send response to journal:
|
Hopcroft’s allegation of glibness is unjustified. The child-parent screening strategy for familial hypercholesterolaemia we propose1 does satisfy the main criteria for a worthwhile screening test.2 Cost and acceptability remain to be established, and this, as we suggest, should be done as part of a pilot implementation project that would also be useful for clarifying the choice of cholesterol cut-off used in screening to ensure a low false-positive rate and a high detection rate. What is surprising is Hopcroft’s suggestion that the death of a parent from the complications of familial hypercholesterolaemia is preferable to the possible anxiety associated with screening for the disorder in childhood. Hopcroft believes that screening confers little benefit to the child until such time as treatment is started. But the detection and treatment of the affected parent, benefits the child in preventing the premature death of parent. The essence of our screening proposal is that child and parent both benefit, and therefore the family as a whole. All screening involves a degree of anxiety. Indeed it is the anxiety of having a serious a medical disorder and the desire to avoid its harmful consequences that forms the rational basis for being screened. Anxiety per se is not harmful; it is excess or inappropriate anxiety that is harmful and with care much can be done to minimise this in medical screening. There is no reason why “neurosis” needs to develop in a child or their parents found to have familial hypercholesterolaemia, provided the nature of the condition is described properly and the rationale for the use of effective treatment is explained. There is also no reason why screening for familial hypercholesterolaemia should “cast a shadow on childhood” and every reason why it would save the life of a parent and later the life of the child when an adult. Hopcroft has a well deserved reputation for communicating medicine and medical advances to the profession and the public. We hope that he may take a fresh look at our proposal and support its practical assessment in a feasibility project. 1. Wald DS, Bestwick JP, Wald NJ. Child-parent screening for familial hypercholesterolaemia: screening strategy based on a meta-analysis. BMJ 2007 335:599 2. Wald NJ. The Epidemiological Approach 4th Edition. London 2004 Wolfson Institute/RSM Press. ISBN 1-85315-584-5 Competing interests: None declared |
|||
|
|
|||
|
Ian Hamilton-Craig, Professor of Preventive Cardiology Griffith University School of Medicine, Gold Coast, Queensland 4215, Australia
Send response to journal:
|
In spite of major advances in molecular diagnosis and therapy, the detection rate of familial hypercholesterolaemia (FH) remains unacceptably low in most countries where diagnostic criteria for FH are confusing and poorly understood by general practitioners, and public and political awareness of the disorder also almost non-existent (1). Up to 75% of patients may be unrecognized prior to their first coronary event, which is often sudden death. Improving the diagnosis rate at an early stage of disease is therefore critical to improve the outcome of FH management. Wald et al’s recent article proposing population-wide screening of 1- 9 year old children at the time of immunization is therefore timely, appealing and challenging (2). It is appealing firstly because early detection offers the best hope of controlling otherwise rapidly progressive atherosclerosis by therapy to lower LDL cholesterol levels, as shown for carotid intima-medial thickening in children treated from the age of 10 years (3). Secondly, children frequently return to their GPs for further immunizations, thereby offering the chance to have a ‘second bite of the cherry’. Thirdly, detection of FH in childhood offers the opportunity to counsel against smoking and other adverse lifestyle habits that usually start in childhood. Finally, screening of parents of affected children will make the programme more cost-effective. The challenge comes in determining whether to choose a community-wide versus a selective screening approach, the age to perform screening, and the diagnostic criteria to use. Diagnosis of FH can be made at any phase of life: in utero (for homozygous FH), at birth (shown to be unreliable because of birthing factors affecting cord blood lipid levels), in the neonatal period (at age 5 days, heel-prick blood is used to diagnose the rarer inherited metabolic disorders), during childhood, adolescence and in adulthood. A variety of parameters can be used for diagnosis, including clinical stigmata of tissue cholesterol deposition (in adults), levels of total cholesterol, LDL cholesterol, ApoB and/or ApoB/A-1 ratio, Lp(a) and apoB, or the detection of an LDL receptor mutation. In adults, cascade family screening has been performed most successfully in Iceland and the Netherlands, where most cases of FH have already been detected and LDL receptor assays used for diagnosis (4). These countries are small, have a single coordinating centre, are heavily funded by government, and the Netherlands has an effective patient support organization similar to HEART-UK. In these countries, approximately 8 near relatives are detected with FH for each index case (1). The ratio of cases detected/index case indicates the effectiveness of cascade family screening. In most countries, however, this ratio is considerably lower, partly because of logistical difficulties in performing cascade screening, lack of funding, lack of community and professional awareness, and privacy issues (1). In Australia the Privacy Act prevents doctors from approaching near relatives without their permission – a ‘catch-22’ situation. Index FH patients are left to contact relatives, an unsatisfactory situation with poor results for several reasons. Many relatives are non-contactable, while others refuse screening because they are ignorant of the nature of FH and the benefits of early treatment. The paper by Wald et al should return screening of parents and children back to the mainstream of preventive strategies for management of FH. It is time for further action, not further debate. Community-wide screening programmes in children at the time of immunization should be initiated now. References 1. Hamilton-Craig I. Case-finding for familial hypercholesterolemia in the Asia-Pacific region. Seminars in Vascular Med. 2004; 4:87-92. 2. Wald DS, Bestwick JP, Wald NJ. Child-parent screening for familial hypercholesterolaemia: screening strategy based on a meta-analysis. BMJ. 2007; 335:599 3. Rodenburg J, Vissers MN, Wiegman A et al. Statin treatment in children with familial hypercholesterolemia: the younger, the better. Circulation. 2007 Aug 7;116(6):664-8. 4. Fouchier SW, Defesche JC, Umans-Eckenhausen MW, Kastelein JP. The molecular basis of familial hypercholesterolemia in The Netherlands. Hum Genet. 2001 Dec;109(6):602-15. Competing interests: None declared |
|||
|
|
|||
|
Peter J. Lansberg, Staff member Dept of Vascular Medicine Academic Medical Center Amsterdam 1100 DD Amsterdam; StOEH Amsterdam 1105, Maud N. Vissers, Hans J. Avis, Barbara A. Hutten, John J.P. Kastelein
Send response to journal:
|
Since 1994, cascade screening is the preferred strategy in the Netherlands for the active identification of familial hypercholesterolemia (FH) patients. After an initial pilot phase, the programme was expanded and accelerated in 2003 and is now one of the four national screening programmes that include breast cancer, cervical cancer, and the screening of inborn errors of metabolism at birth. Once an FH mutation is identified in a clinically diagnosed patient, relatives are subsequently tested through cascade screening for the same mutation (1). The success of this programme is reflected by the fact that more than 13.500, over a third, of the estimated 40.000 FH carriers in the Netherlands have already been identified and it is expected that by 2014 the vast majority of FH patients will be known. Also, our screening programme has already been proven to be cost-effective (2). We were very surprised that Wald and colleagues (3) seemed to be unaware of this established programme and made no reference to it in their manuscript. These colleagues, in fact, proposed an intriguing strategy that would provide a proactive approach to identify FH patients at an early age. They suggest screening all infants for FH at approximately 15 months of age, based on their low-density lipoprotein cholesterol (LDL-c) level. If the clinical diagnosis is established in the child, the parents are then to be screened. Simply, the parent with the highest LDL-c is determined to be the one affected. They hypothesize that it will take approximately 30 years to identify all affected families in the UK. We do not think that this strategy is superior to the one that has been implemented in the Netherlands and we feel that some important caveats of the Wald approach merit discussion. Tracing index patients solely through infant screening may result in missing a large number of affected individuals. FH carriers with one or two children have a 50% and 25% chance, respectively, of giving birth to non-affected offspring and, like FH carriers without children, might therefore never be identified. Also, Wald et al. suggest that non paternity could only be found in approximately 4% the tested children, but this is probably a very conservative estimate since incidence rates as high as 30% have been reported (4). Secondly, Wald et al. place very little emphasis on false negative results. In the Dutch screening programme, 18% of the mutation positive FH patients were found to have LDL-c levels below the 90th percentile at the moment of screening; the majority of these patients would not be diagnosed to have FH based on clinical criteria alone. A similar situation will occur when screening infants. Hence, to establish the diagnosis of FH solely based on clinical parameters is, in our opinion, inadequate for national screening. Unfortunately, genetic confirmation is also not always possible because of putative mutations in novel genes associated with FH such as apolipoprotein B, PCSK9 and likely others. Therefore, clinical and genetic methods complement each other and, in our opinion, reduce errors that would limit finding affected family members. Finally, we would like to underline that the identification of an index patient, by whatever strategy, is only the start of a far more complex process: finding affected family members. To set up and implement an organization that can effectively and efficiently perform this complicated task at a national level is pivotal for the success of an FH screening programme and should not be underestimated. We have shown, however, that with appropriate government support such a programme can be successfully implemented. We believe that the suggested strategy by Wald et al., although commendable for its proactive approach, has several important flaws that would impede effective screening of FH families. 1. Umans-Eckenhausen MA, Defesche JC, Sijbrands EJ, Scheerder RL, Kastelein JJ. Review of first 5 years of screening for familial hypercholesterolaemia in the Netherlands. Lancet 2001;357(9251):165-8. 2. Wonderling D, Umans-Eckenhausen MA, Marks D, Defesche JC, Kastelein JJ, Thorogood M.Cost-effectiveness analysis of the genetic screening program for familial hypercholesterolemia in The Netherlands. Semin Vasc Med 2004;4(1):97-104. 3. Wald DS, Bestwick JP, Wald NJ. Child-parent screening for familial hypercholesterolaemia: screening strategy based on a meta-analysis. BMJ 2007;335(7620):599. Epub 2007 Sep 13. 4. Anneke Lucassen, Michael Parker. Revealing false paternity: some ethical considerations. Lancet 2001; 357: 1033-1036 Competing interests: None declared |
|||