Abstract
This cohort study examines the prevalence of hypertrophic cardiomyopathy in the UK Biobank population.
Hypertrophic cardiomyopathy (HCM) causes considerable morbidity and mortality, including sudden death and death from heart failure and stroke.1 Hypertrophic cardiomyopathy is defined by unexplained left ventricular hypertrophy, which is defined as a maximum left ventricular wall thickness (MLVWT) of 15 mm or more (or ≥13 mm in relatives), out of proportion to loading conditions.1 The cited prevalence in the general population is 0.2%, based on studies in the community, in the military, and among athletes2 that relied on echocardiography. One study using cardiac magnetic resonance (CMR) reported a prevalence of 1.4% in the Multi-Ethnic Study of Atherosclerosis population.3
Limitations of these estimates included a narrow age range for some of the cohorts and exclusion of individuals with hypertension, regardless of the severity. The second of these is important, because hypertension is highly prevalent, including in up to 25% of patients with HCM carrying disease-causing variants.4 Relatives fulfilling criteria for familial HCM (with MLVWTs of 13-15 mm) were also not captured.
Cardiac magnetic resonance produces images with optimal myocardial (endocardial and epicardial) blood definition; mass and wall-thickness measurements are more accurate and reproducible compared with echocardiography, and milder hypertrophy in some segments (basal anterior or lateral or apical) can be missed by echocardiography.1 The use of CMR in family screening increases the yield of phenotype detection.
Machine learning facilitates rapid and accurate analysis of large CMR data sets.5 The UK Biobank is a population-based, prospective cohort study that enrolled 500 000 individuals aged 40 to 69 years (2006-2010).6 It contains information on demographics, health and lifestyle data, biological samples, and outcomes through linkages to electronic health records or registries; the imaging substudy commenced in 2014. Compared with the general population, the UK Biobank participants were more likely to be older, be female, live in less deprived areas, and have fewer comorbidities.6
Methods
This study was covered by the general ethical approval for UK Biobank. The participants signed an electronic consent form at the time of their visit to the UK Biobank assessment center.
We adapted an automatic segmentation algorithm based on deep learning5 to measure LV wall thickness (maximum longitudinal distance between the epicardial and endocardial contours). We applied this method to calculate MLVWTs from short-axis measurements of 44 836 participants. Images with outlying MLVWT values and MLVWTs of 13 mm or more were manually validated by 2 European Association of Cardiovascular Imaging CMR level 3 experts (L.R.L. and N.A.). We excluded individuals with hypertension and aortic stenosis based on self-reported medical history and hospital episode statistics and those with phenocopies (Fabry disease, amyloidosis, glycogen storage diseases, and RASopathies). Data were collected from April 2014 to March 2020 and analyzed for this report from December 2020 to February 2021 with R version 3.6.1 (R Foundation for Statistical Computing).
Results
After excluding individuals with hypertension (n = 14 970; including 607 with MLVWTs ≥13 mm), aortic stenosis (n = 80; 21 with MLVWTs ≥13 mm), and phenocopies (n = 40; 3 with MLVWT ≥13 mm; 0 with MLVWTs ≥15 mm), 29 826 individuals remained (mean [SD] age, 62.8 [7.7] years; 12 805 men [42.9%]). A total of 34 individuals had MLVWTs of 15 mm or more, for a prevalence of 0.11% (95% CI, 0.08%-0.15%); the prevalence was 0.22% (95% CI, 0.14%-0.30%) in male participants and 0.04% (95% CI, 0.01%-0.06%) in female participants. With a cutoff value of 13 mm or more, the prevalence was 0.96% (95% CI, 0.85%-1.07%) overall, 2.07% (95% CI, 1.82%-2.32%) in male participants, and 0.12% (95% CI, 0.07%-0.18%) in female participants. The Figure shows the prevalence for cutoff values of 13, 14, and 15 mm. Participants with MLVWT values of 13 to 14 mm vs those with values less than 13 mm are compared in the Table.
Figure. Prevalence of Left Ventricular Hypertrophy and Distribution of Maximal Wall Thickness.
The value above each bar corresponds to the proportion of individuals with maximal wall thickness greater than the threshold value.
Table. Characteristics of Participants in the UK Biobank.
| Characteristic | Maximal wall thickness, No. (%) | P value | ||
|---|---|---|---|---|
| <13 mm | 13-14 mm | ≥15 mm | ||
| No. | 29 540 | 252 | 34 | NA |
| Age, mean (SD), y | 62.8 (7.7) | 65.9 (8.3) | 66.4 (7.4) | <.001 |
| Male | 12 540 (42.5) | 237 (94.0) | 28 (82.4) | <.001 |
| Race/ethnicity | ||||
| White | 28 620 (96.9) | 244 (96.8) | 33 (97.1) | .68 |
| Asian | 298 (1.0) | 1 (0.4) | 0 | |
| Black | 165 (0.6) | 3 (1.2) | 0 | |
| Chinese | 95 (0.3) | 0 | 0 | |
| Mixed | 148 (0.5) | 1 (0.4) | 1 (2.9) | |
| Other | 155 (0.5) | 2 (0.8) | 0 | |
| Unknown | 59 (0.2) | 1 (0.4) | 0 | |
| BMI, mean (SD) | 25.7 (4.0) | 28.6 (4.2) | 28.5 (5.1) | <.001 |
| Degree-level education | 20 222 (68.5) | 175 (69.4) | 20 (58.8) | .46 |
| Smoking status | ||||
| Never | 18 999 (64.9) | 142 (57.3) | 15 (45.5) | .001 |
| Previous | 9250 (31.6) | 88 (35.5) | 16 (48.5) | |
| Current | 1024 (3.5) | 18 (7.3) | 2 (6.1) | |
| Regular alcohol intake | 13 124 (44.7) | 137 (54.8) | 15 (45.5) | .006 |
| Dyslipidemia | 7084 (24.0) | 80 (31.7) | 9 (26.5) | .02 |
| Diabetes | 809 (2.7) | 13 (5.2) | 0 | .04 |
| Left ventricle, mean (SD) | ||||
| End-diastolic volume, mL | 144.7 (32.5) | 170.6 (42.2) | 170.3 (36.9) | <.001 |
| End-systolic volume, mL | 59.0 (18.1) | 72.0 (24.6) | 72.1 (16.8) | <.001 |
| Stroke volume, mL | 85.7 (18.2) | 98.6 (24.2) | 98.2 (29.6) | <.001 |
| Ejection fraction, % | 59.6 (5.8) | 58.2 (7.2) | 57.2 (7.8) | <.001 |
| Mass, g | 82.0 (20.6) | 124.7 (24.1) | 142.9 (30.3) | <.001 |
| Maximal wall thickness, mean (SD), mm | 9.0 (1.4) | 13.6 (0.5) | 16.0 (1.0) | NA |
Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); NA, not applicable.
Guidelines1 consider that MLVWT values of 15 mm or more in White individuals and 20 mm or more in Black individuals are in favor of HCM vs hypertensive heart disease. If we included these participants regardless of hypertension status, prevalence of HCM was 0.22% (95% CI, 0.18%-0.27%). The presence of MLVWT values of 15 mm or more was mostly located in the basal anterior and anteroseptal segments (Figure).
Discussion
To our knowledge, this is the largest HCM prevalence study based on imaging. Our estimates of 0.11% to 0.22% are consistent with previous reports; the sex-based difference was present in previous studies to a similar degree2,3 and might be partially explained by wall thickness not being adjusted to sex or body size. It is not entirely clear what the causative mechanism of left ventricular hypertrophy for the cases within 13 to 14 mm is. Environmental factors might be a relevant contributor. The higher volumes might be partially explained by the same reason, associated with larger body sizes; left ventricular dimensions and derived volumes are known to be smaller in echocardiography compared with CMR, because the contrast between the blood pool and endocardial border is worse. Older ages may reflect the age-associated penetrance of HCM. There is an increased prevalence of women and healthy volunteer selection bias in the UK Biobank, which might limit generalizability of our findings to more diverse populations. However, the use of a more accurate imaging technique, the very large cohort (>40 000), the inclusion of individuals with MLVWT values of 13 mm or more, and a more nuanced approach regarding the presence of hypertension, likely compensated for these limitations.
References
- 1.Elliott PM, Anastasakis A, Borger MA, et al. ; Authors/Task Force members . 2014 ESC guidelines on diagnosis and management of hypertrophic cardiomyopathy: the Task Force for the Diagnosis and Management of Hypertrophic Cardiomyopathy of the European Society of Cardiology (ESC). Eur Heart J. 2014;35(39):2733-2779. doi: 10.1093/eurheartj/ehu284 [DOI] [PubMed] [Google Scholar]
- 2.McKenna WJ, Judge DP. Epidemiology of the inherited cardiomyopathies. Nat Rev Cardiol. 2021;18(1):22-36. [DOI] [PubMed] [Google Scholar]
- 3.Massera D, McClelland RL, Ambale-Venkatesh B, et al. Prevalence of unexplained left ventricular hypertrophy by cardiac magnetic resonance imaging in MESA. J Am Heart Assoc. 2019;8(8):e012250. doi: 10.1161/JAHA.119.012250 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Heliö T, Elliott P, Koskenvuo JW, et al. ; EORP Cardiomyopathy Registry Investigators Group . ESC EORP Cardiomyopathy Registry: real-life practice of genetic counselling and testing in adult cardiomyopathy patients. ESC Heart Fail. 2020;7(5):3013-3021. doi: 10.1002/ehf2.12925 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bai W, Sinclair M, Tarroni G, et al. Automated cardiovascular magnetic resonance image analysis with fully convolutional networks. J Cardiovasc Magn Reson. 2018;20(1):65. doi: 10.1186/s12968-018-0471-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Fry A, Littlejohns TJ, Sudlow C, et al. Comparison of sociodemographic and health-related characteristics of UK Biobank participants with those of the general population. Am J Epidemiol. 2017;186(9):1026-1034. doi: 10.1093/aje/kwx246 [DOI] [PMC free article] [PubMed] [Google Scholar]