Familial hypercholesterolaemia (FH) is a disorder of lipoprotein metabolism characterised by raised circulating concentrations of low-density lipoprotein (LDL) cholesterol and an increased risk of premature atherosclerotic cardiovascular disease (Nordestgaard et al, 2013; Barkas et al, 2015; Navar-Boggan et al, 2015). It is caused by autosomal co-dominant mutations in genes encoding mostly for low-density lipoprotein receptors (LDLR), uncommonly for apolipoprotein B (APOB) and rarely for proprotein convertase subtilisin/kexin type 9 (PCSK9). Very rarely a recessive form of FH is caused by mutations in low-density lipoprotein receptor adaptor protein 1 (LDLRAP1). Other phenotypic FH cases are thought to be either due to polygenic mutations or unidentified monogenic mutations (Henderson et al, 2016).
Until the advent of genetic testing, diagnosis of FH was clinical. The integration of genetic testing into routine health practice has meant that these clinical diagnoses can now be confirmed or refuted by identifying reported pathogenic mutations.
To make a clinical diagnosis of FH, the National Institute for Health and Care Excellence (NICE) recommends using the Simon Broome Register Group (SBR) criteria, recently updated to include the option of Dutch Lipid Clinic Network (DLCN) criteria, and then, if available, to offer a DNA test to people with a clinical diagnosis of FH (NICE, 2019). The SBR criteria are widely used in the UK (Scientific Steering Committee on behalf of the Simon Broome Register Group, 1991); prior to 2019 this was the only NICE-recommended method of making a clinical diagnosis of FH.
Before 2017, patients with hypercholesterolaemia who were referred by their GPs to the adult lipid clinic at New Cross Hospital, Wolverhampton, were prospectively classified using SBR criteria as definite FH, or possible FH, or ‘non-FH’. Patients with SBR-defined definite and possible FH were managed as FH, according to NICE guidance CG71 (NICE, 2019). In March 2017, the West Midlands Regional Familial Hypercholesterolaemia Service, a region-wide genetic screening service for FH, was launched. Following a clinical diagnosis of possible or definite FH made in the lipid clinic, patients were referred to the nurse-led regional FH service for genetic testing.
The authors audited the impact of diagnosing FH using molecular genetics on a lipid clinic cohort of patients with SBR-defined definite and possible FH.
Methods
The West Midlands Regional FH Service
The West Midlands Regional FH service is a nurse-led service delivered in primary care serving a population of 5.6 million patients. The catchment area includes 20 clinical commissioning groups and 11 acute trusts. The service is clinically administered by a clinical manager, clinical adviser and five FH specialist nurses, each allocated to a geographical area serving a population of about 1.1 million patients. Referrals are accepted from primary and secondary care practitioners across all specialties for patients with primary hyperlipidaemia if they meet SBR clinical criteria for possible and definite FH and/or have serum total cholesterol over 9.0 mmol/litre and triglycerides under 5.0 mmol/litre.
Patients are invited to attend at any of 60 primary care locations of their choice. Specialist nurse assessment includes the patient's medical history, lifestyle, clinical examination (including blood pressure, pulse rate and rhythm, body mass index and stigmata of hyperlipidaemia) and taking a family history (including family pedigree drawing). Patients who have total cholesterol over 7.5 mmol/L but under 9 mmol/L are risk assessed using the Welsh Scoring criteria for genetic screening (Haralambos et al, 2015). If appropriate, after counselling and receiving written consent, samples for genetic testing are obtained during the consultation. If FH is genetically confirmed, the patients are counselled by an FH nurse specialists about FH, given lifestyle advice and, if not attending a lipid clinic, are referred directly to a local lipid clinic for cardiac assessment, further treatment and follow-up.
Cascade family screening is facilitated by the nurse-led service using an indirect contact method. If FH is not genetically confirmed and referral to a lipid clinic is not indicated, the patients are given lifestyle advice, advised to inform family members to have their cholesterol checked and discharged back to primary care or referred to a lipid clinic for cardiovascular risk management as advised by NICE CG181 (NICE, 2016).
The FH nurse team were recruited from a range of specialties across both primary and secondary care and required specialist training to undertake their role. This included understanding genetics, inheritance patterns and the implications for patients and families, obtaining clinical and family history, including drawing a family pedigree, basic genetic counselling, interpreting genetic reports and cascade family screening, including testing paediatric patients. The service is the largest of its kind in England and is continually evolving to meet the emerging needs of patients.
Service evaluation
Patients referred to the lipid clinic with hypercholesterolaemia after exclusion of secondary hyperlipidaemia were clinically classified as definite or possible FH using SBR criteria, which are based on clinical findings, personal and family history, and presenting fasting lipid profile (Table 1). The lipid profile consisted of total cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides and calculated LDL cholesterol. LDL cholesterol was calculated using Freidewald's equation, which is only valid if triglycerides are less than 4.5 mmol/L. Subjects with triglycerides over 4.5 mmol/L were therefore excluded, which is consistent with the notion that hypertriglyceridaemia mitigates against FH. Consequently, all patients had hypercholesterolaemia defined as a total serum cholesterol over 7.5 mmol/L and/or LDL cholesterol over 4.9 mmol/L, with triglycerides below 4.5 mmol/L.
| Criteria | Description |
|---|---|
| A | Total cholesterol >7.5 mmol/L or LDL cholesterol >4.9 mmol/L in adults |
| B | Tendinous xanthomata in the patient or a first-degree relative |
| C | DNA-based evidence of mutation in known causative gene |
| D | Family history of myocardial infarction: <50 years in second-degree relative or <60 years in a first-degree relative |
| E | Family history of total cholesterol >7.5 mmol/L in a first-degree relative or second-degree relative |
| Diagnosis | |
| Definite FH requires criteria A and B and/or C | |
| Possible FH requires criteria A and D and/or E | |
Patients who gave informed consent were referred to the nurse-led West Midlands Regional FH service and screened for pathogenic genetic variants encoding LDLR, APOB, PCSK9 and LDLRAP1. They were then given either a diagnosis of FH or non-FH hypercholesterolaemia. The authors then evaluated the impact of diagnosing FH using molecular genetics on a lipid clinic cohort of patients with SBR defined possible and definite FH.
Results
The demographics and biochemical characteristics of patients classified with SBR definite FH (n=6) and possible FH (n=134) are shown in Table 2. In summary, a pathogenic FH mutation (pathogenic variant) was detected in all six patients with a clinical diagnosis of definite hypercholesterolaemia due to the presence of tendon xanthomata. A monogenic FH mutation (including two variants of unknown significance) was present in 32 (24%) of patients with possible FH.
| Characteristic | Definite FH | Possible FH |
|---|---|---|
| Number | 6 | 134 |
| Mean (SD) age (years) | 55 (17.9) | 53.4 (12.4) |
| Male, n (%) | 2 (33.3) | 40 (29.9) |
| Mean (SD) total cholesterol (mmol/L) | 10.75 (1.92) | 8.69 (1.38) |
| Mean (SD) HDL cholesterol (mmol/L) | 1.43 (0.32) | 1.50 (0.40) |
| Mean (SD) LDL cholesterol (mmol/L) | 6.35 (2.05) | 6.07 (1.16) |
| Mean (SD) triglycerides (mmol/L) | 1.65 (0.77) | 1.86 (1.05) |
| Tendon xanthomata in index or 1st/2nd degree relative, n (%) | 6 (100) | 0 (0) |
| Personal history of premature MI, n (%) | 0 (0) | 11 (8.2) |
| Premature MI (<60 years <50 years) in 1st/2nd degree relative, n (%) | 5 (83.3) | 75 (56.0) |
| Total cholesterol >7.5 mmol/L in adult 1st/2nd degree relative, n (%) | 5 (83.3) | 118 (88.1) |
| Monogenic FH mutation, n (%) | 6 (100) | 32 (23.9) |
FH=familial hypercholesterolaemia; HDL=high-density lipoprotein; LDL=low-density lipoprotein; MI=myocardial infarction; SD=standard deviation
Discussion
All six patients in the cohort with tendon xanthomata had definite FH due to a pathogenic FH mutation. Tendon xanthomata in hypercholesterolaemic patients are highly predictive of FH (Scientific Steering Committee on behalf of the Simon Broome Register Group, 1991), but not pathognomonic because sitosterolaemia, a rare autosomal recessive disorder, may present with a similar clinical picture (Tada et al, 2018). Therefore xanthomata in patients with hypercholesterolaemia who are FH-gene negative should be evaluated for sitosterolaemia.
Patients identified as gene-positive FH are managed according to NICE guidance (CG71) for FH (NICE, 2019). Our local protocol is to manage patients with a genetic variant of uncertain significance (VUS) as FH because they have significant hypercholesterolaemia. Cascade screening (segregation analysis) is not undertaken in patients with VUS because of lack of funding. Patients, however, are asked by the FH nurses to inform their relatives to have fasting lipids checked and, if found to be hypercholesterolaemic, they are referred to the FH service. If the VUS is again identified following full FH genetic analysis, it may be reclassified as pathogenic.
This study indicates that genetic screening is invaluable in excluding FH, since the management of FH resides in secondary care (NICE, 2019) and is different to that of non-FH lipid management, which lies predominantly in primary care (NICE, 2016). The major difference in lipid management is that all FH patients are treated with statins with or without ezetimibe and have access, if eligible, to injectable PCSK9 inhibitors for primary and secondary prevention. Whereas patients without FH are treated with statins and/or ezetimibe based on their absolute cardiovascular risk and have access, if eligible, to PCSK9 inhibitors, but only for secondary prevention. In addition, FH and non-FH patients are treated to different lipid targets. In our cohort, over 70% of patients with SBR-defined possible and definite FH did not have genetically confirmed FH, and the management and follow-up of these patients should have largely been within primary care. Genetic testing, therefore, offers a rational and cost-effective approach to the management of hypercholesterolaemia and eases the burden on hospital lipid services.
Another major advantage of genetic screening is focused on cascade screening for the known pathogenic mutation. In the absence of genetic screening, NICE CG71 (NICE, 2019) recommends the use the gender- and age-specific criteria for LDL cholesterol cut-offs in identifying relatives likely, uncertain or unlikely to have FH. Our personal experience was that a number of, usually young, individuals identified using NICE-recommended tables as having clinically likely FH did not have FH, mostly because the index case did not have genetically confirmed FH or, less commonly, they had not inherited the pathogenic mutation from an affected family member. We thus had in the past initiated inappropriate treatment and caused unnecessary anxiety, which would have been avoided had we had access to genetic screening.
In conclusion, a nurse-led FH service using molecular genetics will greatly facilitate the appropriate and cost-effective management of hyperlipidaemia, including FH, in primary and secondary care.