A comprehensive search was conducted in PubMed, Cochrane Library and Embase databases by using the keywords: “hypertrophic cardiomyopathy”, “HCM”, “mavacamten”, “aficamten”, “cardiac myosin inhibitor”, “CMI”, “echocardiography”, “echocardiogram”, and “echocardiographic” for studies available up to September 14, 2025. The comprehensive search strategy for each database is available in the Supplementary Materials. Subsequently, the eligibility criteria were applied for screening.

We included human studies in this study if they met the following inclusion criteria: (1) participants must be adults aged 18 years and older, with a confirmed diagnosis of hypertrophic cardiomyopathy; (2) outcomes must include echocardiographic measures of cardiac structure and function, as well as any reported adverse effects; (3) studies must report outcomes via standardized echocardiographic techniques and measurements; and (4) no language restrictions will be applied, but only studies with full-text availability will be included. The exclusion criteria were as follows: (1) case reports or reviews; (2) studies that did not report relevant echocardiographic measures; and (3) studies that did not provide sufficient data for meta-analysis or lacked clarity in methodology.

Data were systematically gathered by two authors into a spreadsheet designated for analytical purposes. Any inconsistencies were addressed through collaborative discussion. A third researcher was involved to resolve any discrepancies. The information retrieved from the qualifying studies comprised the study titles, dates of publication, sample sizes, baseline characteristics of participants, types of treatment administered, duration of the treatment, and baseline echocardiographic features. We extracted data from studies on arrhythmia events, including AF and AFL. We also extracted the mean difference (MD) and 95% confidence interval (CI) or standard deviation in the change from baseline (CFB) of echocardiographic parameters between the treatment and placebo groups. These parameters include interventricular septum thickness (IVST) or maximal left ventricular wall thickness (MLVWT), left ventricular mass index (LVMI), left ventricular ejection fraction (LVEF), left atrial volume index (LAVI), lateral e^′velocity, septal e^′ velocity, lateral E/e^′ ratio, and septal E/e^′ ratio. For the single-arm cohort study, which aimed to observe the long-term efficacy of CMIs without a control group, we extracted the CFB for the above echocardiographic parameters.

Two authors independently evaluated the risk of bias in the included studies using the Cochrane Risk of Bias Tool (RoB2) [14] for RCTs and the Newcastle-Ottawa scale (NOS) [15] for the cohort study. A third researcher was involved to resolve any discrepancies.

This meta-analysis was conducted using the “meta” package in R software version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria; https://www.r-project.org/). The results are expressed as risk ratio (RR) and MD along with a 95% CI. Heterogeneity among the included studies was evaluated using the I² statistic, where an I² value exceeding 50% suggests the presence of at least moderate heterogeneity. Given the premise of considerable clinical heterogeneity, a random-effects model was utilized for this meta-analysis. For RCTs, we chose the MD in echocardiographic parameters within the CFB between the treatment and placebo groups as the measure of treatment effect. For single-arm cohort studies, we used the CFB of echocardiographic parameters for the meta-analysis. For effects with substantial heterogeneity, we conducted sensitivity and subgroup analyses to explore the sources of heterogeneity. Since most methods of detecting publication bias are only applicable to more than 10 studies included, we have only created the funnel plots.

A total of 457 records were obtained from Pubmed, Cochrane Library and Embase databases. Among them, 111 records were deleted because of duplication, and 220 records were excluded on the basis of titles and abstracts. Among them, there are 108 conference abstracts, 66 reviews or meta-analyses, 26 expert commentaries, and 20 letters. Next, 116 studies that did not report echocardiographic outcomes or adverse events suitable for analysis were excluded. Finally, 10 studies [4, 5, 6, 7, 8, 16, 17, 18, 19, 20] met the eligibility criteria (Fig. 1). These 10 studies include 1 phase II RCT (MAVERICK-HCM [4]), 4 phase III RCTs (EXPL-ORER-HCM [5], EXPLORER-CN [6], SEQUOIA-HCM [7] and VALOR-HCM [8]), their corresponding 3 substudies of echocardiographic features [16, 17, 18], and 2 long-term cohort studies [19, 20].

Fig. 1.

Preferred reporting items for systematic review and meta-analysis (PRISMA) flowchart of the study screen.

Table 1 (Ref. [4, 5, 6, 7, 8, 16, 17, 18, 19, 20]) summarizes the main characteristics of the ten studies. A total of 938 patients were enrolled in these studies, with the majority being male. The studies reported a mean patient age between 51.0 and 60.9 years. With the exception of Ho et al., 2020 [4], who enrolled patients with nHCM, all other studies focused on patients with oHCM. Olivotto et al., 2020 [5], Hegde et al., 2021 [16], Tian et al., 2023 [6] and Tian et al., 2025 [17] administered mavacamten with dose titration guided by echocardiographic measurements of LVEF and LVOTO, with a treatment duration of 30 weeks. Maron et al., 2024 [7] and Hegde et al., 2024 [18] used aficamten, with dose adjustments based on LVEF and LVOTO, and a treatment period of 24 weeks. Desai et al., 2022 [8] was a phase 3 RCT named the VALOR-HCM trial, which enrolled oHCM patients who had indications for septal reduction therapy (SRT). Desai et al., 2022 [8] was designed with a relatively short primary endpoint timeline of 16 weeks, which may be particularly relevant for patients considering SRT [21]. Desai et al., 2025 [19] is a long-term follow-up study of the VALOR-HCM trial. After 16 weeks, patients in the placebo group crossed over to receive mavacamten treatment and were followed up to 128 weeks, with corresponding data collected. Saberi et al., 2025 [20] is also a long-term cohort study, which enrolled 46 patients and continued aficamten treatment for up to 48 weeks. Table 2 (Ref. [8, 16, 17, 18, 19, 20]) presents the echocardiographic characteristics of the included studies. All patients exhibited ventricular wall thickness and LVMI above the normal upper limits, indicating severe LV structural abnormalities in oHCM patients. Elevated LAVI and E/e^′ ratios suggest significant LV diastolic dysfunction in this population. LVEF ranged from 67.9% to 77.8%, slightly higher than in healthy individuals, indicating a compensatory increase in LV systolic function.

Table 3 (Ref. [4, 5, 6, 7, 8, 16, 17, 18]) and Table 4 (Ref. [19, 20]) present the risk of bias of the studies [4, 5, 6, 7, 8, 16, 17, 18, 19, 20] included in this meta-analysis. All of included RCTs [4, 5, 6, 7, 8, 16, 17, 18] were judged as having a low risk of bias.

Hegde et al., 2021 [16], Hegde et al., 2024 [18] and Tian et al., 2025 [17] reported changes in IVST, whereas Hegde et al., 2021 [16], Desai et al., 2022 [8], Hegde et al., 2024 [18] and Tian et al., 2025 [17] reported changes in LVMI to reflect structural remodelling of the left ventricle after treatment. The pooled analysis reveals that CMIs significantly decreased IVST levels compared with those in the placebo group (MD: –1.77, 95% CI: –3.30 to –0.23, p = 0.0240; Fig. 2A). The aggregated results reveals that CMIs also significantly reduced the LVMI (MD: –18.15, 95% CI: –32.65 to –3.65, p = 0.0141; Fig. 2B). The two single-arm cohort studies by Desai et al., 2025 [19] and Saberi et al., 2025 [20] both reported changes in LV wall thickness and LVMI. Analysis results show that CMIs significantly decreased MLVWT compared to baseline (MD: –1.16, 95% CI: –1.51 to –0.81, p 0.0001; Supplementary Fig. 1A). The findings also suggest that CMIs reduced LVMI (MD: –12.05, 95% CI: –16.46 to –7.63, p 0.0001; Supplementary Fig. 1B). Supplementary Fig. 2 presents the sensitivity analysis for IVST. It was found that heterogeneity significantly decreased when Tian et al., 2025 [17] was excluded, whereas heterogeneity remained high when either of the other two studies was removed. Supplementary Fig. 3 shows the subgroup analysis for LVMI. Four studies were divided into two groups based on sample size. The results indicate that the subgroup with larger sample sizes exhibited lower heterogeneity. Since the treatment duration of CMIs in Desai et al., 2022 [8] was shorter than that in other RCTs, a subgroup analysis was conducted to explore the impact of different treatment durations on the efficacy of CMIs. Supplementary Fig. 4A shows that there is no significant difference in the improvement of LVMI between subgroups with different treatment durations (p = 0.1170).

Fig. 2.

Forest plot for improvement of left ventricular structure, including two echocardiographic parameters: (A) interventricular septum thickness, and (B) left ventricular mass index. CI, confidence interval; MD, mean difference.

LVEF is the echocardiographic parameter reflecting LV systolic function and was reported in Hegde et al., 2021 [16], Desai et al., 2022 [8], Hegde et al., 2024 [18] and Tian et al., 2025 [17]. However, the reported changes in LVEF across the four studies [8, 16, 18, 20] were inconsistent. Compared with those in the placebo group, Hegde et al., 2021 [16], Desai et al., 2022 [8] and Hegde et al., 2024 [18] reported a statistically significant reduction in LVEF after treatment with CMIs. Tian et al., 2025 [17] reported a slight, nonsignificant increase in LVEF compared with placebo. The pooled analysis demonstrates an overall statistically significant reduction in LVEF (MD: –3.22, 95% CI: –5.60 to –0.85, p = 0.0078; Fig. 3A). The combined results of the two long-term cohort studies also demonstrate a decrease in LVEF following treatment (MD: –4.88, 95% CI: –5.96 to –3.79, p 0.0001; Supplementary Fig. 1C). Supplementary Fig. 4B shows that treatment of different duration results in a decrease in LVEF, with no significant difference between groups (p = 0.5530).

Fig. 3.

Forest plot for improvement of left ventricular function, including six echocardiographic parameters: (A) left ventricular ejection fraction, (B) left atrial volume index, (C) lateral e^′ velocity, (D) septal e^′ velocity, (E) lateral E/e^′ ratio, and (F) septal E/e^′ ratio.

Hegde et al., 2021 [16], Hegde et al., 2024 [18] and Tian et al., 2025 [17] reported five echocardiographic parameters reflecting LV diastolic function, including LAVI, lateral e^′, septal e^′, lateral E/e^′, and septal E/e^′. Desai et al., 2022 [8] reported only LAVI and septal E/e^′. Significant improvement in LAVI (MD: –5.75, 95% CI: –7.87 to –3.64, p 0.0001; Fig. 3B), lateral e^′ (MD: 1.56, 95% CI: 0.85 to 2.26, p 0.0001; Fig. 3C), septal e^′ (MD: 0.81, 95% CI: 0.31 to 1.31, p = 0.0016; Fig. 3D), lateral E/e^′ (MD: –4.23, 95% CI: –5.27 to –3.19, p 0.0001; Fig. 3E), and septal E/e^′ (MD: –3.80, 95% CI: –4.74 to –2.87, p 0.0001; Fig. 3F) were observed in the aggregated results. The combined results of the two long-term cohort studies also demonstrate improvement in LAVI (MD: –3.65, 95% CI: –5.20 to –2.10, p 0.0001; Supplementary Fig. 1D) and septal E/e^′ (MD: –3.75, 95% CI: –6.00 to –1.50, p = 0.0011; Supplementary Fig. 1E). Supplementary Fig. 4C,D demonstrate that diastolic function improves across subgroups with different treatment duration, with no significant differences between them (p $>$ 0.05).

Ho et al., 2020 [4], Olivotto et al., 2020 [5], Desai et al., 2022 [8], Tian et al., 2023 [6] and Maron et al., 2024 [7] reported the number of atrial arrhythmia events in the treatment group and the placebo group. The results of these five RCTs did not show that receiving CMIs treatment increased the risk of atrial arrhythmias. No significant association with the risk of atrial arrhythmias was observed in the meta-analysis (RR: 0.98, 95% CI: 0.33 to 2.94, p = 0.9689; Fig. 4).

Fig. 4.

Forest plot for the risk of atrial arrhythmias. RR, risk ratio.

Our meta-analysis has several limitations. First, although this meta-analysis included a total of 10 studies, the primary results regarding echocardiographic characteristics were derived mainly from 4 RCTs. The small number of eligible RCTs precluded a methodologically rigorous evaluation of publication bias using quantitative tests (e.g., Egger’s test), which typically require $\ge$10 studies to achieve sufficient statistical power. While funnel plots (Supplementary Fig. 5) were generated for visual inspection, their interpretability remains constrained under such limited conditions. Furthermore, the included studies presented a short mean follow-up period. Therefore, additional studies with extended follow-up durations are essential to verify the long-term effects. Since this meta-analysis only included RCTs involving oHCM, its conclusions cannot be applied to nHCM. More randomized controlled trials are needed in the future to validate the effects of CMIs in nHCM. Additionally, this meta-analysis combined the effects of two types of CMIs on oHCM without differentiating between their individual impacts. Existing literature has indicated that CMIs can improve mitral regurgitation (MR) in patients with oHCM. However, the studies included in our meta-analysis did not report on changes in MR. Therefore, this meta-analysis is unable to assess the impact of CMIs on MR. This meta-analysis specifically focused on the effects of CMIs on echocardiographic characteristics in HCM patients and did not address cardiac magnetic resonance or cardiac biomarker.