Article Text
Abstract
Peripartum/postpartum cardiomyopathy (PPCM) is a potentially life-threatening disease of uncertain aetiology in previously healthy women. Clinical and experimental data suggested inflammation, autoimmune processes, apoptosis and endothelial dysfunction as typical pathophysiological features of PPCM. Recent data discovered that unbalanced peri/postpartum oxidative stress linked to proteolytic cleavage of the nursing hormone prolactin into a potent anti-angiogenic, pro-apoptotic and pro-inflammatory 16-kDa subform as a potential pathomechanism for the development of PPCM. Consistent with these idea, blockade of prolactin by bromocriptine, a dopamine D2 receptor agonist, prevented the onset of disease in an experimental model of PPCM and appeared successful in small pilot trials with respect to prevention or treatment of PPCM in patients. Here we highlight the current state of knowledge on diagnosis of PPCM, provide novel insights into the pathophysiology behind the disease and outline potential consequences for the clinical management and treatment options for patients at risk for or with PPCM.
- Cardiomyopathy dilated
- gender
- heart failure
- oxidative stress
- peripartum/postpartum cardiomyopathy
- prolactin
- STAT3
Statistics from Altmetric.com
- Cardiomyopathy dilated
- gender
- heart failure
- oxidative stress
- peripartum/postpartum cardiomyopathy
- prolactin
- STAT3
Peripartum cardiomyopathy (PPCM) is a disease of uncertain aetiology, characterised by left ventricular systolic dysfunction and symptoms of heart failure, which occur in previously healthy women mainly during the end of pregnancy and the first 5 months after delivery.1–3
As there are no extended prospective studies of PPCM to date, the current epidemiological profile of PPCM is unknown. The incidence varies with individual studies and has been estimated to range from one in 2300 to one in 4000 pregnancies in western countries.1 3 4 A geographical predilection is found in Haiti, with an estimated incidence of 1:299.4 In Africa, the reported incidence varies from one case per 100 deliveries in a small region in Nigeria to one case per 1000 deliveries in South Africa.2 3
The diagnosis of PPCM is based on exclusion criteria that include the absence of any identifiable cause for heart failure in women without known pre-existing cardiac disease, as defined by the National Heart, Lung and Blood Institute.1
At present, PPCM is listed as a form of dilated cardiomyopathy and is treated according to the guidelines for dilated cardiomyopathy, with no specific therapy targeting PPCM.3 Treatment includes standard pharmacotherapy for heart failure with ACE inhibitors, β-blockers, vasodilators and diuretics.3 Nevertheless, the prognosis of affected women is poor, with reported mortality rates averaged at 15% (<4% up to 30%) worldwide and full recovery in only 23–32% of PPCM patients, and continuous deterioration is reported in up to 50% of cases despite optimal medical treatment.3–9
Risk factors, such as age (teenager or older women), tocolytic therapy, or twin pregnancy may identify subgroups of women with substantially higher incidence rates. However, those risk factors have not been confirmed in recent prospective studies.3 Based on the higher incidence in certain geographical regions, genetic background, mainly African ancestry, may play a role but has not yet been proved.3
The suggested underlying pathomechanisms triggering PPCM include low selenium levels, various viral infections, stress-activated cytokines, an autoimmune reaction and a pathological response to haemodynamic stress.3 10 More recent experimental data raise the possibility of a common pathway, on which different aetiologies that induce PPCM may merge.11 This proposed pathway emphasises a coincidence of unbalanced oxidative stress, activation of the protease cathepsin D, which cleaves the nursing hormone prolactin into an angiostatic and pro-apoptotic 16 kDa form, which seems to drive the disease by impacting systemically on the endothelium as well as on the cardiac vasculature and cardiomyocyte function.11 The oxidative-stress–cathepsin D–16 kDa prolactin hypothesis represents a first potential disease-specific pathophysiological mechanism and is thus offering novel therapeutic options that include the blockade of prolactin with bromocriptine to treat PPCM. The first few studies assessing the use of bromocriptine alongside standard heart failure therapy for patients presenting with acute onset PPCM appear promising in that they suggest a benefit compared with patients who did not receive bromocriptine as part of their treatment. In this regard, a small proof-of-concept pilot study showed 40% mortality of patients receiving standard care and 10% mortality of patients additionally receiving bromocriptine. The survivors in the standard group showed a moderate recovery of left ventricular ejection fraction (LVEF) from 27% to 36%, while the LVEF of the bromocriptine-treated group improved from 27% to 58%.12 Meanwhile, patients with previous PPCM and subsequent pregnancy were offered 100% protection from relapse of PPCM when given bromocriptine in addition to standard care starting immediately after delivery for 8 weeks11; however, the number of patients is still very small.
This review summarises briefly the current state of knowledge on the symptoms and diagnosis of PPCM, provides novel insights into the pathophysiology behind PPCM, concerning the prolactin–pathogenetic mechanisms, and the potential consequences of this for the clinical management of PPCM patients.
Diagnosis of PPCM is challenged by the physiology of peripartum discomfort
Symptoms
PPCM becomes manifest in the last weeks of pregnancy and up to 5 months post-delivery in previously healthy women mainly through typical symptoms of cardiac failure as dyspnoea on exertion, cough, orthopnoea and paroxysmal nocturnal dyspnoea,3 13 In addition, more non-specific symptoms of cardiac congestion such as abdominal discomfort, pleuritic chest pain and palpitations can occur.
Early signs and symptoms of heart failure can be mistaken for pregnancy/peripartum-associated physiological discomfort leading to delayed diagnosis.2–4 Therefore, most frequently, initial presentation is with New York Heart Association (NYHA) grades III and IV, a feature that is associated with increased morbidity and mortality, which may at least partly be caused by late diagnosis. The experience of the authors of the present paper suggest that in Germany the delay in reaching a correct diagnosis ranged from weeks to months in approximately 30% of cases,11 14 which agrees with the findings of other authors.3 15 16
Diagnosis
As defined by the National Heart, Blood, and Lung Institute,1 the diagnosis of PPCM is based on exclusion criteria, mainly on the absence of any identifiable cause for heart failure in peripartum women without known pre-existing cardiac disease. The strict time limit used in the diagnostic criteria was intended to exclude congenital and acquired causes of heart failure, which usually manifest by the second trimester (20–28 weeks) due to physiological volume expansion.17 Physical examination should focus on signs of heart failure, such as hypoxia, S3 gallop, jugular venous distension, hepatomegaly and rales.13 The ECG presentation is quite variable: it may show anything from a normal ECG as in the majority of PPCM cases to prolongation of the PR or QRS intervals and evidence of left ventricular hypertrophy and dysrhythmias.6
Classic transthoracic echocardiography is the key tool for accurate diagnosis and is defined as an impairment of systolic function manifest by a decrease in LVEF (<45%), fractional shortening (<30%) or both.6 Cardiac enlargement with a left ventricular end-diastolic dimension greater than 2.7 cm/m2 is also frequently evident, particularly in those women presenting late. However, some PPCM patients display normal ventricular dimensions suggesting that dilatation may not be specific for PPCM (Sliwa and Hilfiker-Kleiner, personal observations).
MRI has recently been used for the detection of myocardial damage in PPCM disease.18 So far only a few disease-specific cardiac MRI patterns have been reported in PPCM patients, such as the detection of inflammation.18 However, MRI might be useful to detect disease-specific myocardial alterations, that is perfusion problems and it may help for risk stratification by monitoring inflammation and fibrosis during follow-up analysis.18 19
Furthermore, general laboratory tests, including full blood count, urea and electrolytes, C-reactive protein, blood glucose, D-dimer, creatine kinase-MB, and cardiac troponin T, should be performed in patients with suspected PPCM.20 However, many of these parameters are non-specific and are frequently not altered (ie, normal creatine kinase-MB, cardiac troponin T and C-reactive protein even in patients presenting with severe heart failure) in PPCM patients.14 The tests thus assist in excluding further complications or coincident diseases such as myocarditis, pulmonary embolism and myocardial infarction. In addition, it checks for metabolic derailment, which might cause additive stress on the heart. In turn, a more specific marker profile including elevated NT-proBNP, oxidised low-density lipoprotein, prolactin and interferon-γ emerged in a recently performed study in 43 PPCM patients.21 Furthermore, serum markers involved in the prolactin-cleaving pathway, such as activated cathepsin D and 16 kDa prolactin11 12 may be used as more specific diagnostic tools for PPCM.
In summary, due to the observations that early signs and symptoms of heart failure can be mistaken for pregnancy/peripartum-associated physiological discomfort, disease-specific biomarkers would be helpful for the early diagnosis of PPCM. It is important to note that the diagnosis of PPCM, defined as a systolic dysfunction, can only be objectified by echocardiography or MRI.
Physiological and pathophysiological mechanisms in the maternal heart peripartum
Pregnancy-associated alterations in cardiac signal transduction
In order to compensate for increased pregnancy-related haemodynamic volume load the maternal heart undergoes eccentric hypertrophic growth associated with cardiomyocyte growth, a proportional growth of the capillary network and a slight reduction in systolic function.22 Pregnancy-induced hypertrophy is reversible and is not associated with fibrosis or alterations in classic markers of pathological hypertrophy, that is the upregulation of β-myosin heavy chain and atrial natriuretic peptide and the downregulation of α-myosin heavy chain and sarcoplasmic reticulum Ca(2+)-ATPase.22 It is assumed that hormonal changes during pregnancy such as upregulated oestrogen levels may exert protection to the maternal heart. Indeed, associated with increased oestrogen levels the upregulation of cardioprotective c-Src-MAPK (ERK) and Akt signalling have been reported in the hearts of pregnant mice and rats.11 22 It has been shown that oestrogen induces the activation of Akt signalling in cardiomyocytes and by this signalling molecule promotes cardiomyocyte survival and function.23 24 Moreover, oestrogen has been shown to enhance coronary angiogenesis in vivo.25 Therefore, it is likely that oestrogens confer cardioprotection to the maternal heart during pregnancy. The placenta appears to be the major source of oestrogen production during pregnancy explaining the sudden drop in oestrogen with the expulsion of the placenta during delivery, a feature that is associated with a decrease in cardiac Akt activation.11 22
As pregnancy in patients with a previous episode of PPCM is often quite well tolerated but the disease emerges again after delivery,3 one could speculate that oestrogens protect these patients from pregnancy-related cardiac stress and this protection is lost after delivery. Such a notion is supported by observations in our mouse model for PPCM. In this model female mice carry a cardiomyocyte-restricted knockout of the signal transducer and activator of transcription factor-3 (STAT3, STAT3-KO). Similar to the patient collective mentioned above, STAT3-KO mice never showed symptoms during pregnancy, when haemodynamic load accumulates, but always developed a postpartum cardiomyopathy with a high mortality rate.11 These mice showed normal pregnancy-induced cardiac hypertrophy and vessel growth, but displayed a rapid loss of myocardial capillaries, increased cardiac apoptosis, extensive fibrosis, ventricular dilatation with frequent ventricular thrombi and heart failure postpartum.11 This observation suggests that in contrast to Akt signalling, STAT3 is optional during pregnancy, but is necessary for cardioprotection postpartum. Indeed, myocardial activation of STAT3 was noted in wild-type mice late in pregnancy and postpartum.11 The nursing hormone prolactin, which is a dominant hormone postpartum, is known to activate STAT3 by means of its specific receptors in various cell types, including cardiomyocytes in vitro and the heart in vivo.11 26 Therefore, prolactin might be at least partly responsible for the postpartum activation of cardioprotective STAT3 signalling.
The role of prolactin in the development of PPCM
In contrast to these potential cardioprotective effects of full-length prolactin to the maternal heart, prolactin has been hypothesised as a potential factor in the pathogenesis of PPCM.10 Interestingly, recent work showed that prolactin is a hormone that can either stimulate or inhibit various stages of vessel formation and remodelling.2 27 This potential to exert opposing effects on angiogenesis depends on the proteolytic processing of the pro-angiogenic full-length 23 kDa prolactin by the protease cathepsin D or by various metalloproteinases into an anti-angiogenic 16 kDa form, which is known to induce endothelial cell dissociation and apoptosis.2 27
Indeed, it is unlikely that the full-length 23 kDa prolactin, which induces lactation and promotes reverse remodelling of the uterus, is responsible for PPCM. In fact, systemic infusion of 23 kDa prolactin in wild-type and STAT3-KO mice had no adverse effects on the heart.11 However, there is experimental evidence that 23 kDa prolactin is processed in its 16 kDa form in postpartum STAT3-KO hearts, while this is not the case in wild-type hearts.11 High expression of 16 kDa prolactin, even in the absence of the postpartum physiology, destroyed the cardiac microvasculature, lowered cardiac function, and promoted ventricular dilatation.11 Furthermore, it affected cardiomyocyte metabolism and contractility in vitro.11 This detrimental effect of 16 kDa prolactin on the cardiac microvasculature is consistent with observations in tumour biology, in which 16 kDa prolactin induces apoptosis and the dissociation of endothelial cells and prevents their proliferation and migration.27 28 Moreover, 16 kDa prolactin promotes vasoconstriction10 and enhances adhesion of inflammatory cells to the endothelium.10 Interestingly, 16 kDa prolactin does not act by means of the known prolactin receptors,10 indicating that its biological role is completely different from full-length prolactin.
Oxidative stress, which is clearly increased in hearts of postpartum STAT3-KO mice,11 is a potent stimulus for the activation of cathepsin D as it triggers its own release from lysosomes in cardiomyocytes,11 27 Furthermore, oxidative stress also promotes the activation of MMP-2,29 another enzyme able to generate the 16 kDa prolactin. While cathepsin D works best under acidic conditions,27 active cathepsin D released from cardiomyocytes into the cell culture supernatant under physiological conditions in vitro, is able to generate 16 kDa prolactin from recombinant 23 kDa prolactin.11
Imbalanced oxidative and anti-oxidative capacity, a key feature triggering prolactin cleavage
Oxidative stress is caused by an imbalance between the production of reactive oxygen species (ROS; reactive molecules that contain oxygen ions and peroxides are highly reactive due to the presence of unpaired valence shell electrons) and a biological system's ability to detoxify ROS readily or easily repair the resulting damage. It should be noted that oxidative stress rises during pregnancy and culminates in the last trimester as part of normal physiological alterations during pregnancy.10 Its function might increase the maternal defence potential in a state of suppressed immunity and increased risk of infections caused around term and during delivery. However, it is parallelled by an increase in total antioxidant capacity with a peak early postpartum in healthy women.10 This suggests a need for an efficient antioxidant defence mechanism late in pregnancy and postpartum, and indicates an important role for a precise balance between oxidative and anti-oxidative capacity late in pregnancy and early postpartum.
In postpartum STAT3-KO hearts a deficit in anti-oxidative enzymes was identified, especially the upregulation of manganese sodium dismutase, a powerful ROS scavenging enzyme located in the mitochondria. This manganese sodium dismutase enzyme was found to be blunted in these hearts; thereby resulting in increased levels of ROS.11
Enhanced unbalanced oxidative stress may indeed be a common feature in the aetiology of many PPCM patients because various risk factors associated with PPCM are known to promote oxidative stress: infections or autoimmune disease; haemodynamic stress (high salt diet and heating, hypertension); diabetes; adiposity and smoking.2
The simultaneous occurrence of unbalanced oxidative stress, subsequent activation of prolactin-cleaving enzymes (cathepsin D and/or metalloproteinases) and high prolactin levels (pituitary and cardiac) thus appear to be causative for PPCM in mice (figure 1).
How do findings from experimental models relate to human PPCM?
Impaired cardiac signalling pathways in PPCM patients
In terms of deregulated cardioprotective signalling pathways, that is STAT3, in the postpartum heart, explanted terminally failing hearts from PPCM patients display reduced STAT3 protein levels.11 However, the downregulation of STAT3 appears to be a general feature of end-stage failing hearts.30 As biopsies from patients with acute PPCM are scarce and so far no information on the status of signalling pathways in the early stage of the disease exist, it is impossible to distinguish whether the downregulation of cardiac STAT3 in end-stage PPCM is a prerequisite or whether it is a secondary event in response to heart failure.
Presence of an oxidative stress, cathepsin D–16 kDa prolactin cascade in PPCM patients
There is evidence that patients with acute PPCM experience an imbalance in oxidative and anti-oxidative capacity as increased serum levels of oxidised low-density lipoprotein, indicative of enhanced oxidative stress, were detected in PPCM patients.11 21 Furthermore, levels of activated cathepsin D and 16 kDa prolactin are significantly higher in PPCM patients compared with pregnancy-matched healthy controls, suggesting the presence of a systemically activated oxidative stress–cathepsin D–16 kDa prolactin cascade in human PPCM.11 12 It is therefore conceivable that activation of this cascade is a key feature of PPCM in humans (figure 2).
Prolactin, a novel therapeutic target in PPCM
Based on experimental observations that prolactin blockade with bromocriptine prevented the onset of PPCM in mice with a predisposition for this condition (STAT3-KO), first healing attempts were initiated in 18 patients. These patients had PPCM in a previous pregnancy and presented with a subsequent pregnancy, which placed them at a high risk of a relapse of the disease.2 10 Concerning the safety of bromocriptine in pregnancy, a survey of more than 1400 pregnant women taking bromocriptine primarily during the first few weeks of pregnancy found no evidence of increased rates of abortion or congenital malformations.31 Twelve patients obtained bromocriptine in addition to standard therapy for heart failure and all of them had an uneventful post-pregnancy follow-up, whereas all patients (n=6) in the group, obtaining only standard therapy, had a recurrence of PPCM.11 Furthermore, three patients in the non-bromocriptine group subsequently died.11
Next, several healing attempts were started in patients with acute onset of PPCM, who obtained bromocriptine in addition to the standard therapy for heart failure (summarised in Selle et al).2 Meanwhile, a first small, randomised trial in 20 patients with acute PPCM, diagnosed within the first postpartal months with ejection fraction less than 35%, was carried out in South Africa.12 In this trial PPCM patients obtaining bromocriptine had a significantly better recovery of cardiac function and lower mortality rate compared with patients receiving standard therapy only.12 Based on these positive observations, a larger randomised trial to test the efficacy of bromocriptine has currently been initiated in Germany.
Complications in pregnancy and postpartum
Preeclampsia
Preeclampsia is a characteristic hypertensive disorder of human pregnancy and a leading cause of maternal and fetal mortality and morbidity worldwide. As reviewed by Gilbert et al32 the potential mechanisms underlying the pathogenesis of preeclampsia are generally regarded to be placental ischaemia/hypoxia, which in turn results in the elaboration of a variety of factors from the placenta, that is soluble fms-like tyrosine kinase-1, the angiotensin II type 1 receptor autoantibody, and cytokines such as tumour necrosis factor α, which generate widespread dysfunction of the maternal vascular endothelium. This dysfunction manifests as enhanced formation of factors such as endothelin, ROS and augmented vascular sensitivity to angiotensin II. Alternatively, the preeclampsia syndrome may also be evidenced as the decreased formation of vasodilators such as nitric oxide and prostacyclin. Taken together, these alterations cause hypertension by impairing renal pressure natriuresis and increasing total peripheral resistance.
A study by Leanos-Miranda et al33 has demonstrated that urine prolactin excretion is markedly elevated in preeclampsia at diagnosis. Furthermore, anti-angiogenic 14–16 kDa prolactin fragments were only detected in urine from preeclamptic women and both total urine prolactin concentrations and the presence of 14–16 kDa prolactin fragments were closely associated with the severity of preeclampsia and the occurrence of adverse pregnancy outcomes, including placental abruption, acute renal failure and pulmonary oedema (figure 2).33 An earlier study by Gonzalez et al34 reported a marked elevation of angiostatic 9–16 kDa prolactin forms in the amniotic fluid of preeclamptic women, which correlated with low birth weight. Prolactin cleavage was mainly performed by activated cathepsin D in placental trophoblasts, indicating a local generation of 16 kDa and subsequent release into the placental and potentially systemic vasculature of mother and fetus.34
Preeclampsia does not necessary cause heart failure but might be important in women with underlying vascular disease or when multiple compounding factors are present.35 Indeed, preeclampsia during pregnancy is frequently reported in some patient collectives who develop PPCM in the peripartal period,35 while it seems to be less common in PPCM collectives from South Africa and Haiti.3 4
An oxidative stress-triggered prolactin cleavage mechanism may thus be part of the pathophysiology of preeclampsia, a feature that could increase the risk of developing PPCM later on, and may therefore raise the question of whether bromocriptine may also be beneficial in preeclamptic women. However, preeclampsia is not a necessary prerequisite for PPCM.
Thromboembolism
Ventricular thrombi are frequently observed in PPCM patients with a low ejection fraction.2 3 It should be noted that patients with PPCM may face in general an increased risk of thromboembolism as the blood stasis postpartum naturally displays a marked increase in procoagulant activity in maternal blood, characterised by the elevation of factors VII, X, VIII, fibrinogen and von Willebrand factor.36 Further interaction between pregnancy-dependent hormonal changes and the coagulation system refer to thrombin, a serine protease predominantly located on endothelial cells. Thrombin plays a central role in the coagulation cascade and endothelial cell biology by cleaving soluble fibrinogen into insoluble fibrin, with subsequent activation of coagulation factors V, VII, VIII, XIII. In addition, various members of the growth hormone family including prolactin are also processed by thrombin. The recognition site for thrombin on prolactin is different from the site recognised by cathepsin D, and leads to the generation of a C-terminal 16 kDa fragment. For this C-terminal fragment so far no biological activity has been found.27 However, prolactin may act as a scavenger for thrombin and thereby modulate the coagulation state postpartum. The elimination of prolactin, for example by bromocriptine, may therefore increase the risk of thromboembolism, a feature that has been suggested in some case reports, in which bromocriptine was associated with thrombotic events, that is myocardial infarction or stroke in postpartum women.37 However, transient ischaemic attack, hemiplegia, pulmonary embolism, acute myocardial infarction, mesenteric artery occlusion presenting as acute abdomen, have been reported in postpartum women independent of bromocriptine, prevalently diagnosed in patients with an ejection fraction less than 35%.38 In addition, a recent case report demonstrated that ventricular thrombi can resolve under bromocriptine combined with adequate anticoagulation therapy.39
The current consensus in the management of PPCM is therefore to initiate anticoagulation therapy in the presence of severe left ventricular function (ejection fraction ≤35%).1 In addition, due to the potential prothrombotic effects of bromocriptine, we recommend that patients who are taking bromocriptine should obtain a heparin-based anticoagulation therapy (low molecular weight heparin) even in the absence of visible thrombi.
Systemic lupus erythematosus
Systemic lupus erythematosus (SLE) occurs more frequently in women than men by a 9:1 ratio. This female gender bias suggests that female hormones (eg, prolactin, oestrogen) may play a role in the pathogenesis of this autoimmune disease.40 Furthermore, women with SLE frequently have an acute episode post-delivery, proposing that the changing hormone status in the direction of higher prolactin levels might be the triggering factor.41 This assumption was strengthened by the observation that the blockage of prolactin release by bromocriptine in patients with high prolactin levels suppressed SLE in a subgroup and reduced the number of lupus flares.40
SLE is a systemic disease parallelled by increased oxidative stress and decreased anti-oxidative capacity,42 as well as disease-associated vasculitis and systemic endothelial dysfunction indicative of a vascular involvement with endothelial cell damage.42 This scenario is quite similar to patients with PPCM, proposing that prolactin cleavage generating 16 kDa prolactin also takes place in SLE patients, a notion that should be evaluated (figure 2).
Indeed, a recent report discovered a high frequency of systolic heart failure in young patients with SLE43 and case reports mention PPCM in SLE patients.44
However, the prolactin-associated pathophysiology observed in SLE might not represent a general feature of autoimmune diseases, because in other forms of autoimmunity no consistent correlation between prolactin levels and disease activity has been observed.45
Conclusion
With the recent discovery of an oxidative stress–cathepsin D–16 kDa prolactin cascade in experimental and human PPCM, a disease-specific pathophysiological mechanism for PPCM has emerged that may provide a rational basis for a specific therapeutic intervention (figure 1). Bromocriptine, a drug blocking the release of prolactin, systemically and locally, which has been used for many years in women to stop lactation, is currently being tested in a larger randomised trial for its efficacy in PPCM in Germany.
Moreover, as there are no extended prospective studies, the current epidemiological profile of PPCM is unknown. In fact, based on the difficulties in diagnosis it is estimated that many milder cases are not detected, a feature that is worrisome because the risk and severity of PPCM in a subsequent pregnancy is increased by 20–50%.3 Therefore, biomarkers specific for PPCM patients in relation to normal physiological conditions in peripartum women are urgently needed for diagnosis and risk stratification. In this regard, the high prevalence of elevated NT-proBNP, activated cathepsin D and 16 kDa prolactin in the serum of PPCM patients compared with healthy nursing women21 46 may be a first hint towards a disease-specific biomarker profile.
The systematic collection of data prospectively is thus required as well as international cardiac registries to study the incidence, the aetiology and different pathogenic mechanisms of PPCM, including potential genetic and lifestyle aspects.
References
Footnotes
Funding Original work from our group in this review was supported by the Foundation Leducq and the German Research Foundation.
Competing interests None.
Provenance and peer review Commissioned; externally peer reviewed.