Peripartum cardiomyopathy (PPCM) is a multifactorial disease with a wide geographical spread, and with significant morbidity and mortality [1‐3]. Ventricular reverse remodelling and deaths associated with the disease are more frequent in the first six months after delivery [1‐3]. Results from the peripartum cardiomyopathy in Nigeria (PEACE) registry showed significant variation in the burden of PPCM in the country [3]. Incidence rates as high as 1:96 and as low as 1:1350 deliveries were obtained in the Northern and Southern regions of the country respectively [4]. Although reasons for the regional disparity are not yet clear, reports from Niamey (southern Niger Republic) and Kano (northern Nigeria) suggest that selenium deficiency is common in PPCM patients in the regions, with critically low levels of serum selenium in up to 76.9% of the patients [5, 6]. Earlier epidemiological studies in the Keshan and neighbouring counties in northeastern China had established a causal relationship between selenium deficiency and a rapidly progressive and fatal type of cardiomyopathy, the Keshan disease [7]. Subsequently, treatment with sodium selenite showed efficacy in preventing and mitigating the clinical manifestations of the disease [8]. The most commonly used measures of selenium status are plasma and serum selenium concentrations. Quantification of one or more selenoproteins (such as glutathione peroxidase (GPX) and selenoprotein P) is also used as a functional measure of selenium status [9].

To the best of our knowledge, the effects of selenium supplementation on clinical outcomes had not been previously tested in PPCM. We designed this proof-of-concept trial to prospectively study the efficacy and safety of selenium supplementation on clinical outcomes in selenium deficient PPCM patients who had not recovered left ventricular (LV) systolic function at six months after delivery.

In this randomised, open-label proof-of-concept trial involving PPCM patients with LV systolic dysfunction and selenium deficiency, the risk of the primary composite outcome of persistence of HF symptoms and of LV systolic dysfunction, and death from any cause, was not significantly lowered by selenium supplementation. Of the three components of the composite outcome, selenium supplementation significantly reduced HF symptoms and there was a trend towards a reduction of all-cause mortality. There was no significant difference between the groups for the secondary outcome of all-cause rehospitalisations, although the trend was in favour of selenium supplementation. To the best of our knowledge, this is the first study to test the efficacy of selenium supplementation on clinical outcomes in PPCM patients.

Although selenium supplementation did not reduce the composite outcome at the end of the study, a reduction in favour of selenium supplementation was observed between the 9th and 18th month of follow-up, as shown in the Kaplan Meier curves (Fig. 2). This presumably occurred when the serum selenium levels of patients in the selenium group were significantly higher than in the control group, following the three months supplementation. Unfortunately we could not measure the serum levels of selenium group at the end of the three months treatment because of some prohibitive logistic challenges, to support our hypothesis. Our findings suggest that the three months treatment with selenium was probably too short, and prolonging the treatment until the selenium deficiency is corrected could have significantly lowered the composite outcome at the end of the study.

Studies have shown that in PPCM, adverse outcomes mostly occur in the first six months postpartum, and spontaneous ventricular reverse remodelling does occur even in sub-optimally treated patients [2, 3, 12‐14]. In the present study, we randomised PPCM patients who had survived the first six months after delivery, in addition to the other inclusion criteria, to test the effects of selenium supplementation on clinical outcomes, while minimising the impact of the spontaneity of recovery associated with the natural history of the disease. Thus we deliberately recruited the more stable patients to test the efficacy of selenium supplementation on LV functional recovery, though we appreciated that our patient selection could have significantly reduced the magnitude of the efficacy of selenium supplementation on survival and ventricular reverse remodelling. This finding will hopefully be taken into consideration in the design of our future trial on the subject.

Selenium has a wide spectrum of biological effects, and is a critical component of central antioxidant enzymes, including glutathione peroxidases [15]. Thus selenium deficiency could lead to decreased activity of GPX, excessive oxidative stress, myocardial damage and PPCM. Selenium deficiency can also impair many other biologic functions, including those of skeletal muscles, and endocrine and immune systems [15, 16]. Although the effect of selenium on skeletal muscle function was beyond the scope of our study, it is conceivable that it was the direct effect of the selenium supplementation on skeletal muscle function that led to the improvement of symptoms in the present study. We did not find significant correlation between serum selenium and GPX in the present study. Therefore our results support the observation that selenium supplementation may only affect plasma selenium levels and not GPX activity concentration, if study participants were selenium replete before they began taking the selenium supplements [17].

In our study, serum selenium levels increased by 1.71 fold in the selenium group, and by 1.55 fold in the control group, as compared with the baseline values. This was perhaps because at the end of the study, the effect of the selenium supplementation on the serum level had waned, and the general well-being, appetite and feeding of patients in both groups had improved. Serum levels of selenium are determined by many factors. The richest food sources of selenium are seafoods and organ meats [16]. Other sources include muscle meats, cereals and other grains, and dairy products. The concentration of selenium in plants is dependent on the concentration of selenium in the soil and its chemical properties [9]. Consequently, selenium concentrations in plants, as well as in animals that consume these plants, vary significantly by geographic location [9].

The selenium supplement that we used (L-selenomethionine) was well-tolerated without serious adverse effects or the need to withdraw the drug. The choice of the dose of 200 μg once daily of the selenium supplement used in the present study was informed by the findings of a previous study in Eastern United States, in which selenium deficient subjects took and tolerated the same dose well for a mean of eight years [18]. We are therefore confident that prolonging the duration of the treatment will also be well tolerated by patients.

Majority of patients in the present study were referred to the study centres while on treatment with loop diuretics, spironolactone and digoxin, but only 19.6% and 23.9% of the selenium group and 9.3% and 27.8% of the controls were on beta-blockers and ACEI or ARB respectively. The reasons for these poor prescriptions of guideline-directed HF therapies will include the fact that the referring hospitals and clinics lacked the equipment for echocardiography and the expertise to diagnose and start management of PPCM itself. However, prescriptions for these drugs improved significantly during follow-up. Beta-blockers, ACEI, ARB and other recommended HF therapies have been shown to reduce mortality and morbidity in symptomatic patients with HF and reduced ejection fraction, and are strongly recommended for treatment of PPCM patients except when contraindicated [19]. Bromocriptine could be beneficial to some PPCM patients in the acute phase of the disease and may be considered [20]. Implantable cardioverter defibrillator as well as cardiac resynchronisation therapy are recommended for PPCM patients presenting with severe LV dysfunction for more than six months despite optimal medical therapy [19]. Wearable and subcutaneous implantable cardioverter defibrillators represent an alternative to transvenous systems in PPCM patients because they can be more easily removed/extracted if cardiac function recovers.

In this randomised, open-label proof-of-concept trial involving PPCM patients with LV systolic dysfunction and selenium deficiency, the risk of the primary composite outcome of persistence of HF symptoms and of LV systolic dysfunction, and death from any cause, was not significantly lowered by selenium supplementation. However, selenium supplementation significantly reduced HF symptoms and there was a trend towards a reduction of all-cause mortality. There was no significant difference between the groups for the secondary outcome of all-cause rehospitalisations, although the trend was in favour of selenium supplementation.

This proof-of-concept trial has limitations that are inherent to its study design. We used an open-label study design that has potential for bias, and did not repeat measurement of the serum selenium levels at the end of treatment phase with selenium, but only at the end of the study. We also did not include a placebo-arm, partly because of the study’s proof-of-concept trial design, and partly because of limited funding. B-type natriuretic peptide measurement could have also been useful in the prognostication of the patients but it was not readily available at the study centres. We hope to overcome these limitations that were partly caused by our modest funding in our future studies on the subject matter.