Sex-disaggregated data in hypertrophic cardiomyopathy: a targeted literature review, meta-analysis and primary cohort study

Abstract

Objective

Hypertrophic cardiomyopathy (HCM) is the most common inherited cardiomyopathy characterised by pathological left ventricular hypertrophy and categorised into obstructive and non-obstructive subtypes. While its autosomal dominant inheritance pattern would suggest similar prevalence, emerging evidence suggests sex-based physiological differences in presentation, progression and outcomes. We sought to better understand these differences in the literature and among our own cohort of patients with HCM.

Methods

A targeted literature review (TLR) of MEDLINE and Web of Science, and subsequent meta-analysis of NT-proBNP captured sex-disaggregated echocardiographic and cardiac MRI values, genetic mutation effects, biomarker levels and outcomes. A retrospective study of patients diagnosed with HCM at University Hospitals Coventry and Warwickshire NHS Trust, UK between January 2007 and January 2025 examined sex-based differences in presentation, diagnosis and outcomes in HCM.

Results

The TLR found that females were older at diagnosis and had significantly lower septal and left ventricular wall thickness and left ventricular mass. The meta-analysis showed males had lower NT-proBNP levels than females, potentially reflecting sex-based differences in cardiovascular stress. In our cohort (n=207) comprising 78 (38%) females, age at diagnosis was higher (59.4±20.5 vs 53.5±16.1 years; p=0.026) among females than males, with no significant baseline differences in ethnicity, HCM subtype or social deprivation. Maximal wall thickness was higher in males (21.7±6.8 vs 18.2±2.9 mm; p=0.007). Females had higher rates of septal reduction therapy (44.1% vs 19.6%; p=0.018). Mortality was higher among females (20.5% vs 9.3%; p=0.022; 28 deaths at end of follow-up) though adjusted analyses did not reach significance (HR 2.09, 95% CI 0.998, 4.45; p=0.057).

Conclusions

The findings from our cohort support the results of the TLR that females tend to be diagnosed with HCM later than males and experience worse symptoms and outcomes. Further research should clarify the implications of sex-based disparities.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Emerging evidence suggests there are differences in presentation, progression and outcomes associated with sex in patients with hypertrophic cardiomyopathy (HCM).

WHAT THIS STUDY ADDS

• A targeted literature review of prior data found differences in clinical presentation and outcomes depending on patients’ sex.

• A retrospective study of patients at a single UK centre found that female patients were significantly older at HCM diagnosis and underwent more invasive septal reduction therapy procedures. This may reflect a range of biological, clinical and healthcare-system factors, including differences in underlying disease markers, disease progression, symptom burden and clinical decision-making.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• Recognising and addressing sex-based differences may facilitate efforts to address disparities in diagnosis, management and outcomes between male and female patients with HCM.

Introduction

Hypertrophic cardiomyopathy (HCM) is the most prevalent form of genetic cardiomyopathy, affecting 0.5% of the adult population.¹ HCM is characterised by pathological left ventricular hypertrophy (LVH). Given the autosomal dominant inheritance pattern of HCM, the prevalence of HCM in the general population is expected to be similar between males and females.² However, published studies frequently report a higher incidence of HCM among males than females, with males accounting for approximately 55%–70% of patient cohorts.^3–10 Recent investigations have highlighted sex-based differences in the clinical manifestation, progression and prognosis of HCM, which may be linked to this apparent disparity between expectation surrounding HCM prevalence and that which is reported in the literature. Notably, evidence shows that females tend to receive a HCM diagnosis later in life (≥65 years) than males,^{2 11} with one multicentre study of 969 patients conducted in Italy reporting that female patients diagnosed with HCM were up to 14 years older than their male counterparts.⁴ Observed sex-based differences in HCM diagnosis and outcomes may also be linked to potential differences in disease manifestation. Females have been documented to display a higher symptom burden compared with males, including exertional dyspnoea, fatigue, palpitations and chest pain and New York Heart Association (NYHA) functional class III to IV.^3–10 In addition, females with HCM tend to present with higher rates of left ventricular outflow tract (LVOT) obstruction than males and are less likely to be prescribed common medicines such as beta blockers, angiotensin-converting enzyme inhibitors or anticoagulants.^{2 12} Females also display an increased risk of heart failure and mortality due to HCM compared with males.²

Physiological differences, such as those that arise as a result of the size disparity between male and female cardiac structures, may contribute to sex-based differences in HCM manifestations.¹⁰ Predictors of clinical outcomes such as biomarkers may also offer insights that explain sex-based differences. One such biomarker widely used in clinical practice is NT-proBNP, which is an indicator of heart failure. It has been found that levels of NT-proBNP in patients with heart failure differ according to sex, with one study indicating that the sex-specific NT-proBNP threshold for risk reduction is lower in females than in males. Interventions that target this biomarker or its underlying cause may therefore have a protective effect.¹³

In addition to physiological factors, external factors have been found to impact development and outcomes of cardiovascular disease, such as socioeconomic status.¹⁴ It has been found that individuals living in less deprived areas, as measured by Indices of Multiple Deprivation (IMD), were more likely to attend CVD screening than those with lower IMD.¹⁵ For HCM specifically, one US-based single centre study found that socioeconomically vulnerable HCM patients had higher mortality when not referred to specialty care than individuals in higher socioeconomic status groups.¹⁶

In this study, a targeted literature review (TLR) was undertaken to identify sex-disaggregated clinical outcomes and events in patients with HCM. This was supported by a meta-analysis that sought to explore sex-specific differences in the levels of the prognostic biomarker, NT-proBNP. Separately, a retrospective cohort study was conducted using primary data from University Hospitals Coventry and Warwickshire NHS Trust, UK, to investigate the prevalence of sex-based differences in clinical presentation, diagnosis, patient outcomes and healthcare resource use among those diagnosed with HCM.

Methods

Targeted literature review

The TLR searches were conducted using MEDLINE and Web of Science. The search strategy used a combination of selected keywords, which included: “hypertrophic cardiomyopathy” OR “HCM”, “gender” OR “sex”, “females OR “women” OR “males” OR “men”, “wall thickness”, “left ventricular mass”, “sex differences in hypertrophic cardiomyopathy” OR “gender differences in hypertrophic cardiomyopathy”. Searching of these databases was supplemented by targeted web searching using these keywords.

These searches aimed to capture sex-disaggregated echocardiographic and cardiac MRI values (wall thickness (WT) and LV mass (LVM)), genetic mutation effects, biomarker levels and outcomes related to HCM. The mean and SD for WT and LVM were collected from each study in which these were reported, with the units of measurement for WT indicated in either mm or mm/m² [(indexed for body surface area (BSA)], and those for LVM in g/m². Outcomes reported included atrial fibrillation (AF), sudden cardiac death (SCD), coronary artery disease (CAD), hypertension, systemic hypertension, aortic stenosis (AS) and hypercholesterolaemia.

Inclusion criteria specified studies in adult HCM patients; studies performed in humans; those that were published in English; and studies that referred to sex-disaggregated clinical outcomes and events, clinical features, transthoracic echocardiography (TTE) or cardiac magnetic resonance (CMR) relevant measurements (WT and/or LVM), gene mutations, LVOT obstruction or biomarker levels in males and females that were specified separately.

Studies were excluded if they were performed in animals, purely statistical datasets, investigations involving athletes and paediatric populations, as well as studies reporting follow-up outcomes after septal ablation or surgical myectomy procedures. Additionally, studies published in languages other than English were omitted. Studies focusing exclusively on ECG criteria were also excluded, as ECG-based parameters are primarily used for initial identification, rather than diagnostic confirmation, of HCM. Moreover, investigations that mentioned HCM phenocopies (such as cardiac amyloidosis, Anderson-Fabry disease, Pompe disease, Danon disease, etc) or HCM in hypertensive patients were excluded due to their association with distinct prognostic pathways. Finally, studies that failed to report sex-specific categorisation in their abstracts were not included. The methodology and findings were guided by the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) 2020 checklist.⁵

Meta-analysis

NT-proBNP level was chosen as an outcome based on the availability of data noted from the prior TLR. To perform a meta-analysis of the mean NT-proBNP differences between males and females, studies identified in the TLR were re-screened, alongside newly identified studies via supplemental search terms (“NT-proBNP”, “women” OR “females”, “males” OR “men”, “hypertrophic cardiomyopathy” OR “HCM”, “gender differences” OR “sex differences”, “biomarkers”) in MEDLINE and Web of Science.

Primary cohort study

The retrospective cohort study was conducted utilising patient records from all adults (≥18 years) diagnosed with HCM between 1962 and January 2025 at University Hospitals Coventry and Warwickshire NHS Trust, UK: a large, high-volume centre with ~1000 beds, serving a population of over 1 million people. The study plan was reviewed and granted a waiver by the UHCW Research Ethics Committee. Patients were diagnosed with HCM on the basis of clinical history and unexplained LVH on cardiac imaging, in accordance with ESC guidance for HCM.¹⁷ All patients for whom complete patient records were accessible via the electronic patient record (EPR) system were screened. Patients were excluded if complete data were unavailable owing to moving out of the catchment of the hospital or incomplete recording of clinical information in the EPR to allow accurate data collection. Demographics, diagnostic data and patient outcomes were sex-disaggregated, and the prognostic significance of diagnostic parameters at a differential threshold for adults diagnosed with HCM were explored. Baseline characteristics including HCM subtype (obstructive or non-obstructive), age at diagnosis, ethnicity, body mass index (BMI) (kg/m²) and IMD were captured. IMD incorporated seven standalone indexes which are combined and weighted together: income decile, employment decile, education and skills decile, health and disability decile, crime decile, barriers to housing and services decile and living environment decile.

Records indicated whether patients were in receipt of septal reduction therapy (SRT) and cardiac electronic implantable devices (CIED), encompassing permanent pacemaker, implantable cardioverter-defibrillator (ICD), cardiac resynchronisation therapy pacemaker (CRTP), CRTP with defibrillator (CRTD) and subcutaneous ICD (sICD). Patient comorbidities were also captured, including AF, ventricular arrhythmias (NSVT, VT/VF) and out-of-hospital/in-hospital cardiac arrest, heart failure states (by LVEF category), hypertension, ischaemic heart disease and prior myocardial infarction (MI) and stroke/transient ischaemic attack. The full list of patient comorbidities captured is presented in online supplemental table 1. Patient mortality censored on 16 February 2025 was also recorded and stratified by sex and HCM subtype.

Statistical methods

RStudio software (V.4.4.1) was used to develop a random-effects meta-analysis model. Where NT-proBNP level was reported as median with range or IQR, values were converted to mean (SD), using the formula developed by Hozo et al, to enable inclusion in the meta-analysis.¹⁸ Heterogeneity was assessed using the I-squared (I²) statistic, with values exceeding 50% considered to be indicative of substantial heterogeneity, alongside the Cochrane Q test.¹⁹ The significance level was set to p <0.05.

Characteristics and outcomes associations with sex were analysed using χ2 tests or Fisher’s exact tests, as appropriate. For continuous and ordinal variables, the normality assumption was tested visually by plotting the data on a histogram, which revealed the data to be not normally distributed. Continuous and ordinal variables were therefore compared using Mann-Whitney U tests. Survival analyses for time-to-mortality were conducted via Kaplan-Meier plots, with comparisons made using the log-rank test, and Cox proportional hazards models were fitted to assess differences in mortality risk. To identify predictors of CIED implantation, univariate logistic regression analyses were initially performed, followed by multivariable logistic regression models to obtain adjusted estimates accounting for potential confounders including age at diagnosis, sex, deprivation index and presence of LV outflow tract obstruction (LVOTO). Statistical significance was set at p <0.05.

Results

Targeted literature review

Of 3253 records screened, 225 were sought for retrieval, of which 88 could not be retrieved. On assessment for eligibility (n=137), 94 reports were excluded based on no sex comparison; reports on phenocopy conditions; other types of HCM that were not obstructive or non-obstructive; or were surgical follow-up studies. Forty-three unique studies were included overall, of which 38 were identified for inclusion in the TLR. A flow diagram of all included and excluded studies is shown in online supplemental figure 1.

The results of the TLR indicated that, overall, there may be sex-specific differences in the diagnosis and outcomes of patients with HCM. Studies that reported on TTE or CMR data demonstrated that females with HCM tend to exhibit lower septal, posterior, LV posterior WT (mm/m²) and LVM (g/m²) compared with males with HCM.^{7 20–28} However, in those studies that adjusted WT values for BSA (mm/m²), higher indexed values were reported for females than for males.^{25–27 29} Of those studies that adjusted WT for BSA, one study did not report a sex-specific difference in indexed values for maximal wall thickness (MWT), but adjustment notably brought the values into closer range than the absolute MWT values for males and females (6.8±2.1 mm/m² vs 6.6±2.4 mm/m², compared with absolute values of 14.0±3.9 mm vs 11.5±3.8 mm, respectively).²³

Female patients tended to be diagnosed with HCM later than their male counterparts,^{23 26–29} with included studies reporting a mean age difference ranging from 1 year up to 11.2 years [mean (SD) 53 (12) years vs 52 (14) years and 66 years vs 54.8 years in females vs males, respectively].^{27 28} It was also observed that females tended to experience worse symptoms than males, with more females being classed as NYHA classes III/IV, and exhibiting SCD, nonfatal embolic stroke, systemic hypertension and death from all causes (including heart failure and stroke) compared with males with HCM.2 4 6 7 12 22 24 26 29–32 One study reported that the likelihood of developing heart failure was 87% higher in females with HCM than in males after controlling for obstruction, systolic dysfunction, hypertension and age (HR 1.87, 95% CI 1.48 to 2.32; p<0.001).⁹ By contrast, males tended to demonstrate a higher prevalence of dyslipidaemia and CAD than females with HCM.^{12 22} Another of the included studies indicated that female sex may be a predictor of AF in HCM, which is in turn associated with increased morbidity and mortality.³³ Overall, the results of the TLR support the hypothesis that females tend to be diagnosed with HCM later in life and may experience worse symptoms and outcomes than males.

Meta-analysis

Seven studies that were identified through the initial screening and supplementary searches were included in the meta-analysis. In six of these studies, males with obstructive HCM exhibited lower NT-proBNP levels than females (standardised mean difference 0.35, 95% CI −0.62 to −0.08, p<0.001; figure 1), though substantial heterogeneity was noted (I²=89.92%).

Figure 1.

Forest plot of NT-proBNP levels. SMD, standardised mean difference.

Primary cohort study

A total of 213 patients were enrolled in the retrospective cohort study, of which six were excluded due to insufficient data (such as date of symptoms that triggered diagnosis); 207 patients were included in the final analysis (table 1). The median (Q1, Q3) follow-up for the overall patient cohort, excluding patients with a missing time of diagnosis (n=14), was 127.6 months (84.1, 192.2). In the analysis excluding deceased patients, median (Q1, Q3) follow-up was 133.0 months (91.1, 203.0).

Table 1.

Baseline characteristics and outcomes of the primary cohort

	Female (n=78)	Male (n=129)	P value
Obstructive HCM n (%)	34 (43.6)	46 (35.7)	0.256
Age at diagnosis (years)			0.003
Mean (SD)	59.4 (20.47)	53.5 (16.08)
Median [lower quartile (LQ), upper quartile (UQ)]	62.5 (47.3, 71.0)	53.0 (45.0, 63.0)
Minimum-maximum	1 – 115	17 – 109
Ethnicity, n (%)			0.291
White	66 (85.7)	99 (79.8)
Any Black	1 (1.3)	9 (7.3)
Any Asian	7 (9.1)	12 (9.7)
Other	3 (3.9)	4 (3.2)
IMD decile			0.581
Median (Q1, Q3)	6.00 (3.00, 8.00)	6.00 (3.00, 8.00)
Minimum–maximum	1–10	1–10
BMI (kg/m2)			0.093
Mean (SD)	32.7 (7.62)	29.9 (5.02)
Median (Q1, Q3)	31.6 (26.6, 35.9)	29.5 (26.9, 32.6)
Minimum-maximum	22.3 – 50.9	21.0 – 47.2
Maximum LV wall thickness (mm)	18.2±2.9	21.7±6.8	0.007
Septal reduction therapy n (%)	34 (44.1)	25 (19.6)	0.018
Cardiac transplantation n (%)	1 (1.3)	0 (0)	–
Cardiac implantable electronic device* n (%)	19 (30.6)	40 (39.2)	0.267
Permanent pacemaker	6 (10.3)	11 (11.2)	0.865
Implantable cardioverter defibrillator	11 (19.0)	27 (27.3)	0.241
Cardiac resynchronisation therapy pacemaker	0 (0.0)	1 (1.1)	1.000
Cardiac resynchronisation therapy with defibrillator	2 (3.7)	3 (3.3)	1.000
sICD	2 (3.7)	3 (3.3)	1.000
Device extraction n (%)	0 (0)	1 (0.8)	–
Mortality n (%)	16 (20.5)	12 (9.3)	0.022
Age at death, years (mean±SD)	71.3±14.6	77.0±15.3	0.33
Time from diagnosis to all-cause mortality, years (mean±SD)	7.3±4.3	7.8±4.6	0.77
All-cause mortality by HCM subtype	–	–	0.82
Obstructive	6 (37.5)	4 (33.3)	–
Non-obstructive	10 (62.5)	8 (66.7)	–

P value for ethnicity compares proportions of white ethnicity.

P values for HCM subtype and ethnicity obtained using chi-squared tests and rest by Mann-Whitney U test.

BMI, body mass index; HCM, hypertrophic cardiomyopathy; IMD, index of multiple deprivation; LV, left ventricular; sICD, subcutaneous implantable cardioverter-defibrillator.

Among the 207 patients included in the study, non-obstructive HCM was more common (61%) than obstructive HCM (39%); 38% of patients were female (n=78), while 62% were male (n=129). Obstructive HCM was more prevalent in females (43.6%, n=34) than in males (35.7%, n=46), although not to a significant degree (p=0.256).

The study population mainly comprised White patients (85.7% of females and 79.8% of males), with no significant differences between females and males (p=0.291). Females were significantly older than males at HCM diagnosis, with a mean (SD) age of 59.4 (20.47) years versus 53.5 (16.08) years, respectively (p=0.003). There were no significant differences in BMI (p=0.093). Absolute MWT was significantly higher in males, with females demonstrating a mean (SD) MWT of 18.2 (2.9) mm and males an MWT of 21.7 (6.8) mm (p=0.007). There was no significant sex-based difference in the index of multiple deprivation (IMD) decile, with a median of 6.00 (Q1 3.00, Q3 8.00) for both females and males with HCM (p=0.581). Comorbidities were also statistically balanced between the sexes, except for incidence of thyroid dysfunction which was significantly higher in females at 16.7%, vs males at 3.9% (p=0.002) (online supplemental table 1).

Females with HCM were shown to have a significantly higher rate of SRT, with 44.1% of females having undergone SRT versus 19.6% of males (p=0.018; table 1). CIED implantation was similar between sexes (online supplemental table 2). Adjusted and unadjusted analyses did not demonstrate a significant association between CIED implantation and sex, age at diagnosis, deprivation index and presence of LVOTO (online supplemental table 3 and online supplemental figure 2). That said, females were less likely to have implantation of any device (with the largest discrepancy between the sexes observed in ICD implantation) in both the unadjusted and adjusted analyses. Potential associations between ICD implantation and sex, age at diagnosis, deprivation index and presence of LVOTO were similarly analysed. While unadjusted results showed a significant association between ICD implantation and age at diagnosis, adjusted analyses did not (online supplemental table 4 and online supplemental figure 3). Multivariable logistic regression analyses demonstrated an association between female sex and decreased likelihood of ICD implantation, though this did not reach statistical significance (online supplemental figure 3).

At the end of follow-up, 28 deaths (13.5%) were recorded (table 1). The cause of mortality was documented in just 11 patients, comprising cardiac (four females, two males) and non-cardiac causes (four females, one male). The Kaplan-Meier survival plot is presented in figure 2. Adjusted analysis demonstrated that age at diagnosis was positively and significantly correlated with mortality (HR 1.15, 95% CI 1.06 to 1.26; p=0.001 per 5 year age difference, table 2), though female sex was not (HR 2.09, 95% CI 0.998 to 4.45; p=0.057, table 2).

Figure 2.

Kaplan-Meier survival plot for the primary cohort.

Table 2.

Assessment of potential predictors of time to death

P	Hazard ratio (95% CI), p value
	Unadjusted analysis	Adjusted analysis
Age at diagnosis (per 5 years age difference)	1.17 (1.07 to 1.27), <0.001	1.15 (1.06 to 1.26), 0.001
Female gender (reference is male)	2.34 (1.11 to 4.94), 0.026	2.09 (0.98 to 4.45), 0.057

All-cause mortality was not significantly different based on obstructive or non-obstructive HCM subtype, nor were there sex-based differences in mortality in obstructive and non-obstructive HCM. There were no sex-based differences in age at time of death, or time from diagnosis to mortality.

Discussion

Sex-disaggregated cohort data and our literature review demonstrated that females were diagnosed with HCM later in life and experience worse symptoms and clinical outcomes relative to males.

The higher proportion of males presenting in the primary cohort study aligns with the literature, wherein males represent ~60% of HCM study cohorts.^3–10 A recent retrospective cohort of patients diagnosed with asymptomatic HCM aged ≤16 years reported that just 33.0% of those diagnosed were female.³⁴ This contrasts with the expectation that males and females are equally likely to develop HCM, given its autosomal dominant inheritance pattern,² and indicates that males are more likely to receive an HCM diagnosis. It is possible that this reflects a combination of biological, clinical and healthcare system factors. A plausible healthcare system-related explanation for this is that males may be more likely to be referred to a specialist and receive a HCM diagnosis when presenting with a cardiology-related condition, as they tend to have an overall higher incidence and earlier onset of cardiac symptoms/diagnosis than females.³⁵ This may result in a higher likelihood of males undergoing diagnostic tests. The diagnostic criteria for HCM may also favour the diagnosis of males given the size disparity between male and female cardiac structures, which is not accounted for in the standard diagnostic cut-off of an LV thickness ≥15 mm. This also raises the possibility that females may receive a HCM diagnosis later in the disease course, as they may be less likely to meet the current diagnostic threshold. This highlights the need for further research to account for sex-based differences in disease presentation, to ensure that both male and female patients receive timely diagnoses.³⁶

HCM diagnoses among females took place at an older age relative to males, aligning with the literature. Studies included in the TLR reported a mean age difference between females and males ranging from one to 11.2 years; it is notable that the largest range was reported for a study published in 1999, while the smallest range was reported in 2021. It is possible that this reflects improvements in the timing of diagnosis for females with HCM, or earlier presentation among females in more recent years. The difference in the timing of diagnosis between males and females is supported by the findings of the primary cohort data, with a significant difference in mean age at diagnosis of 5.9 years between the sexes.

Almost all other baseline characteristics and comorbidities were well-balanced in our cohort, suggesting that measured baseline differences alone are unlikely to fully explain the observed sex-based difference in outcomes. A notable exception in comorbidities was underlying thyroid dysfunction, which was found to be significantly higher in females, a trend also noted in the general population.³⁷ In the context of HCM, thyroid dysfunction may be associated with more severe outcomes, with experimental studies demonstrating an association between thyroid hormones and cardiovascular homeostasis.³⁸ More broadly, biological factors, such as sex-specific genetic expression, hormonal influences and differential ventricular remodelling may underpin sex-specific disparities observed in HCM patients. For instance, oestrogen’s cardioprotective effects may modulate the hypertrophic response, potentially leading to less hypertrophy but differing remodelling patterns in females vs males.³⁹

Considering the role of biomarkers, our meta-analysis demonstrated that males had lower NT-proBNP levels than females. NT-proBNP has been shown to be a marker of cardiovascular stress and is increased in patients with heart failure; therefore, lower NT-proBNP levels may be suggestive of lower cardiovascular stress in males than in females.⁴⁰ This may support a biologically plausible mechanism contributing to sex-based differences in HCM. While some studies have previously reported sex-based differences in NT-proBNP in patients with heart failure,¹³ the level of heterogeneity observed in the current meta-analysis emphasises that measurement techniques, patient demographics and disease heterogeneity can influence these biomarker levels, complicating their prognostic utility across different patient populations. NT-proBNP was not collected as part of our cohort study, so it was not possible to gauge whether there is any correlation with patient characteristics or outcomes.

We observed that males displayed a significantly higher maximum LV WT than females. This was similarly noted in our literature review, although when studies adjusted WT for BSA, females tended to display higher index values. In a large international study of 1645 patients—529 females and 1116 males—it was found that while female patients had a statistically significantly lower MWT than their male counterparts (18 mm vs 19 mm, respectively; p <0.001), they had a significantly higher mean z-score for MWT (5.1 vs 4.5, respectively; p=0.05). This indicates that although females had lower absolute MWT than males, their MWT deviated further from age-, sex- and BSA-adjusted reference values, suggesting a greater degree of hypertrophy.⁴¹ Inherent size disparities in cardiac structures between males and females may therefore play a role in diagnosis and progression of HCM in females.¹⁰ Despite this, current guidelines for diagnosis of HCM [European Society of Cardiology (ESC), American College of Cardiology (ACC)/American Heart Association (AHA)] stipulate an absolute value of ≥15 mm LV WT as the primary criterion for an HCM diagnosis, irrespective of sex or body size.^{17 42} Females may therefore reach a more advanced stage of hypertrophy before they receive a formal diagnosis, which could, in part, explain why females in our cohort were more likely to undergo SRT than males. This highlights the potential relevance of sex- and BSA-adjusted diagnostic criteria that account for inherent physiological differences between males and females. Indeed, one study found that demographic-based adjustment of LV WT thresholds using age, sex and BSA reduced the male predominance among the patient cohort defined as hypertrophic from 89% to 56%,⁴¹ closer to the expected prevalence of HCM in the general population. Additionally, the adjusted threshold in that study was lower than 15 mm in 48% of individuals, particularly females, younger individuals and those with lower BSA.⁴¹ Taken together, these findings suggest that demographic-adjusted LV WT thresholds may support more timely diagnosis, potentially resulting in improved outcomes for patients with HCM.

Females underwent SRT significantly more often than males. This may be a consequence of later diagnosis and hypertrophy being at a more advanced stage. Clinical decision-making, among other plausible explanations, may also contribute to sex-based differences in SRT rates; for example, differences in patient presentation and how symptoms are interpreted in males and females may influence timing of initiating pharmaceutical intervention for HCM. This may result in males receiving newer, effective HCM therapies earlier in their disease course, such as cardiac myosin inhibitors, delaying the need for surgical intervention as the result of disease progression and worsening heart failure. Nevertheless, other biological or clinical factors, including sex-based differences in disease prognosis and progression, may also contribute to these observed differences.

With respect to mortality, age at diagnosis was significantly associated with increased mortality in adjusted analyses. Considering the later diagnosis of females, this emphasises the importance of early detection to ensure appropriate diagnosis and treatment. However, while there was a trend towards numerically higher all-cause mortality among females than males, there were relatively few deaths within our cohort during the follow-up period and adjusted analysis did not reach statistical significance.

Our study has several limitations. The retrospective nature of the primary cohort study may be subject to selection bias, particularly as patients with incomplete EPRs were excluded. If missing records were associated with patient or disease characteristics, this may have impacted the representativeness of our cohort and the observed sex-based differences. In addition, the retrospective design meant that the study was limited by the availability of data collected in a real-world setting. There is also the potential for certain variables such as age at diagnosis or other unobserved variables to confound analyses. Notably, the predominant ethnicity in both male and female groups was White, limiting the overall generalisability of the findings to regions outside of the UK and Europe. The relatively small sample size and single-centre nature of the data may also limit generalisability. Given the retrospective study design, formal power analysis was not conducted, and therefore, the study may have been underpowered to detect statistically significant differences. The mortality data are limited by a relatively low event rate (just 28 deaths at the end of follow-up), so limited conclusions can be drawn and, as above, may be confounded by other variables, especially given the reason for mortality was not documented in several cases. The literature review was undertaken for hypothesis generation purposes and was not conducted systematically; rather, it was targeted in nature, only searching two major databases, with no grey literature searches or risk of bias assessment conducted. Relevant records identified by the TLR were also screened by a single reviewer. The TLR incorporated studies from a range of geographic regions, which while more representative than our cohort, may impact generalisability of findings given disparities between demographics and differences in diagnostic criteria. Furthermore, the meta-analysis of NT-proBNP levels was limited by the availability of evidence, with the I² statistic indicating substantial heterogeneity across included studies. This may reflect differences in NT-proBNP measurement techniques, patient demographics and heterogeneity in disease profile across the included studies. As such, the pooled estimate may not be generalisable across different settings and should be interpreted with caution.

Conclusion

Future research should aim to elucidate the mechanisms underlying sex-based disparities in HCM, including prospective longitudinal studies evaluating hormonal and genetic factors. Additionally, interventional studies that tailor management strategies based on sex distinctions could determine whether outcomes can be optimised through personalised approaches. Our study highlights significant sex-based differences in the presentation, management and outcomes of adults with HCM. Recognising sex-based differences may facilitate more individualised care and support efforts to address differences in HCM management and outcomes between male and female patients.

Data availability statement

Data are available upon reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

The study plan was reviewed and granted a waiver by the Research & Development Department, University Hospitals Coventry & Warwickshire NHS Trust Research Ethics Committee, with reference GF0447, on the basis that it made use of previously collected, non-identifiable information undertaken by staff within a care team using previously collected information during the course of care of their own patients or clients.