Initial Investigations Into Single Nucleotide Polymorphisms and Their Association with QTc and Left Ventricular Function in Peripartum Cardiomyopathy

Peripartum cardiomyopathy (PPCM) affects previously healthy women towards the end of pregnancy or first 6 months following delivery.1 The aetiology of PPCM is multifactorial and current understanding of the pathogenesis supports a ‘two-hit’ hypothesis: a predisposition followed by hormonal changes triggering cardiac dysfunction.2 PPCM outcomes vary, with some patients experiencing full recovery while others develop heart failure, thromboembolic events, arrhythmias or sudden cardiac death.3 Approximately 50% of PPCM patients present with a prolonged QTc at diagnosis,4 which has been shown to be a predictor of poor outcome.4, 5 The mechanism of QTc prolongation in PPCM remains poorly understood. This study investigated whether single nucleotide polymorphisms (SNPs) associated with long QT syndrome (LQTS) could be used as predictive biomarkers for outcomes in patients with PPCM. The investigation conforms with the principles outlined in the Declaration of Helsinki. The study protocol received ethical approval (HREC No. 517/2021) and informed consent was obtained from all participants recruited at Groote Schuur Hospital. Baseline data were collected, an echocardiogram performed with left ventricular dimensions and function assessed per contemporary guidelines; A baseline 12-lead electrocardiogram (MAC 5500 HD, GE Healthcare, Seattle, Washington, USA) was acquired, and heart rate-adjusted QT intervals (QTc) were calculated using Bazett's formula (QTcB).6 A QTc ≥460 ms was regarded as prolonged. DNA was extracted from blood, and candidate SNPs, known to be associated with QT interval duration, LQTS, and hypertrophic cardiomyopathy in prior studies, were analysed: AKAP9 rs7808587; AKAP9 rs2961024; AKAP9 rs11772585; AKAP9 rs2282972; NOS1AP rs16847548; NOS1AP rs4657139; KCNQ1 rs2074238 and KCNH2 rs1805123. Sanger sequencing was performed and visualized on an ABI 3130xl (Applied Biosystems™, Foster City, CA, USA). Poor outcome was defined as a composite endpoint comprising hospital readmission, heart transplantation or death, persistent New York Heart Association (NYHA) class III/IV symptoms, or severe left ventricular systolic dysfunction (left ventricular ejection fraction [LVEF] ≤35%) at 6-month follow-up following the index diagnosis.4 Given the established link between prolonged QTc and adverse cardiac events,7 we used these specific SNPs8-10 to determine if they were associated with prolonged QTc and/or could predict poor outcome in patients with PPCM. Data were collected on Research Electronic Data Capture (version 7.5.2) and analysed in Stata (version 17; Stata Corp., College Station, TX, USA) for statistical analysis. Continuous variables were summarized as means with standard deviations for parametric data or median with interquartile range for non-parametric data. Categorical variables were expressed as frequencies and percentages. Patients were classified as variant positive or negative for the eight SNPs. Univariable logistic regression analyses assessed the association between mutant SNPs and poor outcome. Kruskal–Wallis or Wilcoxon rank-sum test and chi-squared or Fisher's exact test compared patients that were SNP positive or negative. A p-value of <0.05 was regarded as statistically significant. This study examined the association of SNPs with QTc and outcomes in a cohort of 43 women with PPCM. The median age was 30.1 years, with 48.8% of mixed ancestry and 46.5% Black African. At presentation, 70% of the cohort were classified as NYHA class III/IV. Overall, patients had a median LVEDD of 58mm (IQR 53-63), LVEF of 31% (IQR 24-36), heart rate of 94bpm (IQR 86-116) and QTc of 465ms (IQR 436-481), with 57.1% of patients presenting with prolonged QTc (Tables 1 and 2). The AKAP9 rs2961024 mutant G allele was present in 27.9% of PPCM patients (Table 3) Although there was no significant difference in QTc measurements (470 ms [442–497] vs. 461 ms [430–477], p = 0.24) (Table 1), patients experienced poor outcomes (58.3% vs. 16.1%, p = 0.006) (odds ratio [OR] 7.3) (Tables 1 and 4); one fatality and two different readmissions were noted. Patients with the variant also had lower LVEF, larger left ventricular dimensions at follow-up (Figure 1) with less LVEF improvement compared to non-carriers (Figure 2). The AKAP9 rs7808587 mutant G allele was recorded in 46.5% of patients (Table 3) who tended to exhibit a prolonged QTc (470 ms [445–482] vs. 456 ms [429–481], p = 0.087), (Table 1). Carriers experienced a higher risk of adverse outcomes (45.0% vs. 13%) (OR 5.5), reduced LVEF (42% [35.0–50.0] vs. 54.5% [50–56], p = 0.016), larger LVEDD (59 mm vs. 52 mm) at follow-up (Figure 1) and less LVEF recovery (Figure 2); three readmissions and one fatality were recorded. The AKAP9 rs2282972 mutant T allele was identified in 67.4% of patients (Table 3) and was associated with a shorter QTc (455 ms [432–472] vs. 478 ms [460–497], p = 0.048; Figure 1) which appeared to be protective (p = 0.031) (Table 4). Carriers demonstrated significant LVEF improvement (p < 0.001) (Figure 2). The NOS1AP 16 847 548 mutant C allele was associated with a longer QTc but did not predict adverse outcome (OR 1.1) (Table 4). With few biomarkers available to identify women at risk of developing PPCM and to determine those who may face adverse outcomes, new strategies are under investigation to enhance prediction, assist with the diagnosis, and perform risk stratification for PPCM. As a prolonged QT interval is a known predictor of poor outcome,5, 7, 11, 12 we selected eight SNPs that were previously associated with prolonged QTc and/or outcome in LQTS. This exploratory study, which predominantly consists of patients of mixed ancestry or Black African cases, demonstrates that specific SNPs have the potential to be used as tools for risk stratification and predictors of poor outcomes in women with PPCM. A key finding was the association of specific AKAP9 SNPs with both QTc duration and clinical outcomes. The AKAP9 rs2961024 and rs7808587 were significantly associated with poor clinical outcomes, lower LVEF and larger left ventricular dimensions at 6-month follow-up. Conversely, the AKAP9 rs2282972 variant was linked to shorter QTc and improved outcomes, potentially offering a protective effect. While the former two SNPs were not directly linked to prolonged QTc in this cohort, their association with clinical outcomes highlights their potential as biomarkers for risk stratification in larger populations. These findings are not unexpected as it is known that AKAP9 significantly modifies QTc duration, cardiac event risk, and disease severity via the delayed rectifier potassium current (IKs).8 However, as our current understanding relies heavily on the IKs, extrapolating these findings directly to PPCM might be unreliable, and we prefer to avoid unsupported mechanistic speculations in this preliminary report due to the lack of specific data. Given the small sample size, limited ethnic diversity and higher prevalence of certain AKAP9 SNPs in the African population,8 the study findings regarding genetic associations with cardiac conditions and poor outcome should be interpreted with caution and are best viewed as hypothesis-generating. Validation in larger, ethnically diverse cohorts is crucial, especially considering the historical under-representation of African populations in genetic research. G.S. gratefully acknowledges support from the National Research Foundation (95627 and 105923) and the Medical Research Council of South Africa (SAMRC-Self Initiated Research (SAMRC-SIR) for 2019 and 2021. GS would also like to acknowledge each member of the Cardiovascular Genetics team. Conflict of interest: none declared.