Peripartum cardiomyopathy and pre‐eclampsia: two tips of the same iceberg

This article refers to ‘Hypertensive disorders in women with peripartum cardiomyopathy: insights from the ESC EORP PPCM Registry’ by A.M. Jackson et al., published in this issue on pages 2058–2069. Our understanding of the pathophysiological basis of peripartum cardiomyopathy (PPCM) has evolved rapidly over the last 20 years.1 Convincing evidence suggests that increased levels of oxidative stress as a result of proteolytic cleavage of prolactin (PRL) plays an important role, at least in a proportion of women.2 However, increased production of a 16 kDa PRL fragment as a result of imbalances in oxidative stress is not a specific feature of PPCM, and is also recognised in pre-eclampsia (PE) and non-gestational cardiovascular disease.3, 4 Increased circulating levels of 16 kDa PRL and soluble fms-like tyrosine kinase (sFlt-1) mediate endothelial dysfunction as well as vascular and myocardial injury.1 These shared disease mechanisms play an important role in the development of both PPCM and PE, with systolic myocardial dysfunction defining the former and afterload-driven diastolic myocardial dysfunction typical of the latter. The co-existence of hypertensive disorders in up to 40% of women with PPCM supports the presence of a common pathophysiological basis converging around the biological mechanisms of oxidative stress and endothelial dysfunction.5 There is also a growing recognition that some women with PPCM have an underlying genetic susceptibility to developing myocardial dysfunction. Whether such presentations represent a distinct disease from gene negative cases, or a spectrum of the same disease is debatable. Around 10–15% of women with PPCM carry truncating variants in TTN (TTNtv), the most common genetic cause of dilated cardiomyopathy.6, 7 A commonly held hypothesis is that such variants create susceptibility to developing contractile impairment but require a ‘second hit’ to unmask a phenotype. Indeed, in a recent multicentre study in PPCM, the proportion of patients with TTNtv was similar in women with gestational hypertension compared to those who presented with normotensive PPCM.6 This supports the concept that gestational vascular dysfunction may represent the ‘second hit’ in genetically susceptible women and that vascular and myocardial susceptibility co-exist in many women rather than representing separate phenotypes. Furthermore, outcomes in women with TTNtv were similar to genotype negative patients with similar rates of recovery at 1 year.6 Interestingly, a smaller number of women carried variants in FLNC, DSP and BAG3.6 Such variants typically have more complete penetrance and are associated with more aggressive phenotypes and worse outcomes. They may be best considered separately. In this issue of the Journal, Jackson and colleagues present more fascinating data on the phenotypic spectrum of women with PPCM and the overlap with hypertensive disorders of pregnancy from a large global registry.8 The investigators are to be congratulated for the impressive feat of collecting data on 752 women from across 51 countries, allowing them to provide novel insights into a rare and understudied condition. The results reiterate the frequent co-existence of hypertensive disorders with PPCM; almost 40% of women with PPCM were diagnosed with either hypertension or PE. Those with PPCM-PE had more severe symptoms despite having better markers of left ventricular systolic function and developed these symptoms later in pregnancy. It is certainly possible that women with PPCM-PE and a predominant vascular phenotype had greater diastolic dysfunction than those without hypertension explaining the discordance in symptoms. As expected, women without hypertension had features more suggestive of intrinsic myocardial disease, with more left bundle branch block, longer QRS duration and more biventricular involvement. Interestingly though, a family history of dilated cardiomyopathy was most common in women with PPCM and hypertension (9%), followed by those with PPCM without hypertension (4%). Although numbers are small, this supports the findings from the recent genetic study and the concept of hypertension during pregnancy acting as a second hit amongst susceptible individuals carrying TTNtv. As we begin to understand how best to apply this increasing volume of genetic data to clinical practice, it may be best to avoid restricting clinical screening of first-degree relatives and genetic testing to only those without hypertensive disorders. Much focus in clinical care is correctly placed on the completeness of left ventricular reverse remodelling following a diagnosis of PPCM. This helps inform the risks of future pregnancy and is closely associated with long-term outcome. A reassuringly high prevalence of recovery in left ventricular ejection fraction (LVEF) was observed, with around 50% of women having a LVEF > 50% within 6 months. We hope that this proportion would increase even further over time. Interestingly, women with PE were twice as likely to reach this ‘recovery’ threshold compared to women without hypertension. As with dilated cardiomyopathy, higher systolic blood pressure was associated with greater chances of improvement in LVEF.9 Perhaps this is related to greater capacity to introduce heart failure therapy or more rapid resolution of an afterload-driven pathology following delivery. However, it is also important to bear in mind that we should not determine recovery solely on the basis of LVEF, but also include assessment of symptoms, diastolic function and neurohormonal activation. It is possible that fewer women with PPCM-PE would have met recovery thresholds if these additional prognostic markers, that are more relevant in an afterload-driven pathology, had been assessed. This emphasises the importance of tailored, multidisciplinary follow-up informed by the phenotypic presentation (Figure 1). What is also evident is that the women diagnosed with PPCM in these expert centres are likely to represent only the tip of an iceberg. With a mean LVEF of just over 30% and the majority reporting New York Heart Association class III or IV symptoms at presentation, there are likely to be many more women with more subtle forms of cardiovascular dysfunction towards the end of pregnancy that go undiagnosed. Whether this is a vascular phenomenon driven by imbalances in oxidative stress, a myocardial susceptibility or a combination of both, the development of a gestational phenotype is likely to identify women with an increased risk of cardiovascular disease in the future.10 Increased recognition and diagnosis, perhaps through judicious use of circulating biomarkers, presents an opportunity to identify an at-risk group of women who may benefit from cardiac imaging and targeted intervention to improve long-term cardiovascular health. Collaboration between obstetricians and cardiologists will be key to this goal. Incorporating the results from this study, in-depth characterisation of the vascular and myocardial phenotype may better inform management around future pregnancies, guide medical therapy and enable the identification of at-risk family members. Conflict of interest: none declared.