Diagnosis

Echocardiography is key because it both identifies left ventricular dilatation and dysfunction, and excludes most other causes of this. Right ventricular dilatation and dysfunction may also be present. Secondary dilatation of the valve rings gives rise to varying degrees of mitral and/or tricuspid regurgitation, although the valves are structurally normal. Mural thrombus can form secondary to the low-flow cardiac output state and is most often seen in the left ventricular apex or atria. All patients require electrocardiography and basic laboratory investigations including renal and liver function tests, full blood count, iron studies, serum calcium and phosphate, thyroid stimulating hormone, creatine phosphokinase, and urine protein analyses.[4] The N-terminus of the prohormone brain natriuretic peptide (NT-proBNP) is a reliable biomarker for ventricular failure in pregnant and postpartum women,[5] but is nonspecific and does not identify the cause. Cardiac magnetic resonance imaging (MRI) is a useful addition to echocardiography because it can more accurately assess chamber volumes and has a higher sensitivity for the detection of left ventricular thrombus and myocardial infarction, fibrosis, or infiltration. Coronary angiography, noninvasive assessment of myocardial perfusion, and, rarely, endomyocardial biopsy may also be indicated.

Clinical features

Patients with DCM may be asymptomatic until they experience an additional hemodynamic stress such as pregnancy. Common symptoms include dyspnea on exertion or lying flat, effort intolerance, edema or ascites, and palpitations, which may represent atrial or ventricular arrhythmia.

Pregnancy outcome

A recent publication reported pregnancy outcomes in women with DCM,[6] and there are also publications that primarily compare PPCM or pregnancy-associated cardiomyopathy cases with small numbers of women with DCM of other etiologies (see below). The former study included 36 pregnancies in 32 women with idiopathic or postchemotherapy DCM from 1994 to 2008 as a substudy of a larger cohort of women with cardiac disease in pregnancy. In all, 84% of women were in NYHA functional class I or II at the time of the first antenatal visit. Fifty percent had a LVEF of 45–54%, 22% had a LVEF of 30–44%, and 28% had a severely impaired LVEF of <30%. Prepregnancy data were not included. There was no maternal mortality, but 14 (39%) pregnancies (including the postpartum period) were complicated by cardiac events including symptomatic cardiac failure, arrhythmia, and one transient ischemic attack. As expected, a LVEF of <45% and/or NYHA functional class III or IV were the main predictors of adverse cardiac events, together with a history of a prior event. All complications were successfully managed, however, and by 6 months postpartum 89% of women were in NYHA functional class I. It is worth noting that only 33% of women in this study continued taking beta-adrenergic receptor blockers throughout pregnancy. Treatment at the beginning of the study period did not reflect contemporary best practice, and this may have contributed to the complication rate.

There is preliminary evidence to support the use of serial measurement of serum NT-proBNP levels as an addition to clinical assessment.[7]

Only one study to date specifically investigated pregnancy outcome in women who are carriers of a genetic mutation known to cause DCM but have not yet manifested the condition. Palojoki described 11 pregnancies in 5 women with the lamin A/C (LMNA) gene mutation that affects the left ventricle and cardiac conduction system.[8] There were no adverse events and no worsening of the cardiac function, i.e. no progression to DCM.

Management in pregnancy

The data described above support the current practice in which women in NYHA functional class III or IV are counseled against pregnancy, as are those with a LVEF of <30%. In cases of moderate impairment of left ventricular function (LVEF of 31–44%, while recognizing that echocardiography assessment of LVEF is not absolute) with a good functional capacity reported by the patient and proven on exercise testing, there is still a risk of adverse events but the risk assessment should be individualized. Pregnant women with DCM should be managed by a combined cardiac obstetric service and have regular clinical and echocardiographic assessment throughout pregnancy.

The majority of patients will be taking a cardioselective or vasodilator beta-adrenergic blocking drug (such as bisoprolol or carvedilol), which should be continued throughout pregnancy. Angiotensin-converting enzyme (ACE) inhibitors should be stopped before conception. Alternative vasodilators such as hydralazine may be substituted if this becomes necessary. Loop diuretics may be used but spironolactone is contraindicated. Prophylactic dose low-molecular-weight heparin (LMWH) should be considered if there is significant left ventricular impairment, or a treatment dose given for atrial arrhythmia.

For management of acute decompensation, see the section “Management of PPCM.” There is no indication for cesarean delivery or induction of labor for cardiac reasons for a woman with DCM who is stable and tolerating pregnancy. European Society of Cardiology guidelines suggest that cesarean section should be considered for a severely symptomatic woman in heart failure,[9] but opinion varies.

Definition

The Working Group on PPCM of the Heart Failure Association of the European Society of Cardiology defined PPCM as “an idiopathic cardiomyopathy presenting with heart failure secondary to left ventricular systolic dysfunction toward the end of pregnancy or in the months following delivery, where no other cause of heart failure is found.[10] It is a diagnosis of exclusion. The left ventricle may not be dilated but the ejection fraction is nearly always reduced below 45%.”

This has superseded older definitions based on left ventricular fractional shortening or diastolic dimension,[11,12] and has also relaxed the strict condition that PPCM must present in or after the last month of pregnancy. This recognizes the likelihood that PPCM and earlier presenting pregnancy-associated cardiomyopathy (PACM) are the same clinical entity.[13]

Incidence

Peripartum cardiomyopathy is rare but the true incidence is unknown because mild cases probably go unrecognized, and reported incidence rates are influenced by the inappropriate inclusion of women who are later found to have DCM from other causes or transient left ventricular dysfunction related to preeclampsia.[14] The reported incidence of PPCM varies between 1 per 300 and 1 per 15 000 live births, although the higher values relate to studies without echocardiographic confirmation of left ventricular dysfunction in which other causes of pulmonary edema such as fluid overload have not been excluded. There are marked geographical and ethnic variations in the reported incidence rates, with higher rates reported in Haiti and South Africa compared with USA and Singapore.[15,16] Within countries, higher rates are demonstrated in black African women compared with Hispanic and white European women, and there are higher rates in Malay vs Chinese women.[16] A study published in 2014 has demonstrated an increase in the rate of PPCM diagnosis in the USA. The overall PPCM rate was 10.3 per 10 000 (or 1 in 968) live births. PPCM incidence increased from 8.5 to 11.8 per 10 000 live births from 2004 to 2011.[17] A rate of 8.9 per 10 000 was reported from Singapore in 2013.[16] These documented rises in incidence may be attributable to a rise in average maternal age, an increase in multiple pregnancies secondary to increased use of assisted reproductive techniques, and improved recognition and diagnosis of the disease.[18] An international registry of patients with PPCM as part of the EURObservational Research Programme will allow a study into the differences in presentation and management around the world.[19]

Risk factors

Recognized associations of PPCM include multiple pregnancy, hypertension (pre-existing, pregnancy-induced, or preeclampsia), primiparity, increased maternal age, and black race.[16,18,20–22] The prevalence of preeclampsia and hypertension in women with PPCM was recently reported to be 40–60%.[16,21,22]

Pathophysiology

Different etiologies for PPCM have been suggested. These include:

Clustering of PPCM cases with familial DCM has been observed. Morales et al. searched a DCM database and identified 45 women with PPCM/PACM out of a total of 4110 women with DCM.[23] Nineteen cases were sequenced for known DCM genes, with mutations found in six. Familial clustering with DCM was found in 23 unrelated cases. A Dutch group reviewed the familial DCM database of the University Medical Centre in Groningen and identified 5 out of 90 (6%) families with PPCM cases.[24] They also identified ten independent cases of PPCM, screened first-degree relatives and found undiagnosed DCM in three families in which the proband with PPCM had failed to recover normal left ventricular function. A later study by the same group screened 18 families with PPCM and DCM for mutations with a panel of 48 genes implicated in inherited cardiomyopathy.[25] Pathogenic mutations were found in 4 out of 18 (22%) families and variants of as yet unknown significance in a further 6 out of 18 (33%). These studies support the hypothesis that in a proportion of cases the clinical syndrome of PPCM develops when the hemodynamic load of pregnancy unmasks previously clinically silent DCM, or a pregnancy-associated factor such as oxidative stress combines with genetic susceptibility.

Various studies have suggested a role for infection in the etiology of PPCM. Endomyocardial biopsy has been used for more specific diagnosis and prognosis. There is often nonspecific evidence of myofiber hypertrophy, degeneration, and fibrosis.[11] In one study, 14 out of 18 women (78%) with PPCM showed evidence of myocarditis.[26] Of these, ten were treated with immunosuppressive therapy, and nine of these had both subjective and objective improvement. Follow-up endomyocardial biopsies showed resolution or substantial improvement in the myocardial inflammation. However, four women with myocarditis who were not treated with immunosuppressives also improved. In another American study of 42 cases of women with PPCM who all had endomyocardial biopsy, 26 (62%) had myocarditis.[27] Molecular pathological investigation of biopsy specimens from 26 women with PPCM in Germany revealed viral genomes (parvovirus B19, human herpesvirus 6, Epstein–Barr virus, and human cytomegalovirus) in eight women (31%) that were immunohistologically associated with interstitial inflammation.[28] The authors concluded that their findings supported a high prevalence of virus-associated inflammatory changes in PPCM.

In a study of 100 African women with PPCM, levels of C-reactive protein (CRP) at presentation correlated positively with baseline left ventricular end-diastolic and end-systolic diameters and inversely with LVEF.[29] Women who died presented with significantly lower mean ejection fractions and higher plasma levels of Fas/Apo-1 (a cell surface molecule that induces apoptosis in susceptible cells). Similarly, Chinese researchers found that levels of high-sensitivity CRP were significantly higher in women with PPCM than in control women.[30] Cardiac myocyte apoptosis played a causal role in the pathogenesis of cardiomyopathy in a murine model of PPCM.[29]

In DCM, there is selective upregulation of immunoglobulins (Igs) against cardiac myosin of the G3 subclass (IgG3s are immunoglobulins with proinflammatory characteristics). However, in women with PPCM all class G subclass (G1, G2, and G3) immunoglobulins were raised.[31] IgG3-positive women were in a higher NYHA class at initial presentation. The authors concluded that, unlike DCM, the impact of PPCM on humoral immunity is not subclass restricted. However, raised levels of IgG3 may be of prognostic value in clinical PPCM.

Workers have also proposed fetal microchimerism as a cause of PPCM. During pregnancy fetal cells released into the maternal circulation are tolerated by the mother because of the natural immunosuppression of pregnancy. However, after delivery, this increased immunity is lost, and it is postulated that fetal cells persisting in cardiac tissue may cause a pathological autoimmune response leading to PPCM. Peripartum cardiomyopathy is associated with unique sets of autoantibodies and autoantigens and some workers have suggested a model of PPCM that depends on the interaction of several factors including hemodynamic stress, genetics, immune dysregulation, and fetal microchimerism.[31]

Hilfiker-Kleiner and Sliwa demonstrated a role for the 16 kDa fragment of prolactin, which is angiostatic and proapoptotic to the endothelium and impairs the metabolic activity of cardiomyocytes.[32] 16-kDa prolactin activates nuclear factor κ-B (NF-κB) signaling in endothelial cells and thereby upregulates microRNA146a (miR146a), which mediates most of the adverse effects of 16 kDa prolactin in endothelial cells. Proteolytic cleavage of 23-kDa prolactin released from the pituitary gland to the 16 kDa fragment by proteases such as cathepsin D or matrix metalloproteinases takes place under conditions of oxidative stress in the myocardium. The level of oxidative stress rises during pregnancy, and late pregnancy is associated with the formation of particles that are susceptible to oxidation (high level of low-density lipoprotein cholesterol) and an increase in oxidative damage. In normal pregnancy, increased production of reactive oxidative species is matched by an increase in antioxidant capacity, with an early postpartum peak in healthy women. A compromised antioxidant defense system (such as in preeclampsia) results in a shift toward increased oxidative stress, thus predisposing to PPCM. Oxidative stress also results from downregulation of the transcription factor, signal transducer and activator of transcription 3 (STAT3). This protein plays a key role in many cellular processes such as cell growth and apoptosis.

PPCM develops in mice bred to have a cardiomyocyte-specific deletion of Stat3. Deletion of Stat3 led to enhanced expression of cardiac cathepsin D, thus promoting formation of 16-kDa prolactin.[33] In women with PPCM, STAT3 protein levels are low in the heart, and serum levels of activated cathepsin D and 16-kDa prolactin are elevated. The STAT3 transcription factor also increases cardiac expression of vascular endothelial growth factor (VEGF), which binds to the antiangiogenic soluble Fms-like tyrosine kinase 1 (Flt-1; also known as vascular endothelial growth factor receptor 1 [VEGFR-1]). Oxidative stress downregulates STAT3. The high levels of soluble Flt-1 and low levels of VEGF seen in preeclampsia may explain the increased risk of PPCM in women with preeclampsia.[32]

In the mouse model, blocking prolactin with bromocriptine completely prevented the onset of PPCM,[34] and administration of recombinant VEGF improved heart failure.[32] Small clinical studies suggest that in women with PPCM the addition of bromocriptine to standard heart failure treatment leads to an improvement in LVEF and NYHA functional class,[34] and a randomized controlled trial is currently in progress in Germany. The nonrandomized single-center pilot study of 20 patients showed that women receiving bromocriptine had a better recovery of LVEF (from 27% to 58%; p=0.012) compared with women receiving standard heart failure therapy without bromocriptine (from 27% to 36%) at 6 months. One patient in the bromocriptine group and four patients in the nonbromocriptine group died, and eight women given standard therapy alone experienced the composite end point of death, NYHA functional class III–IV, or LVEF of <35% at 6 months compared with only one in the bromocriptine group (p=0.006).[34]

Clinical features

The severity of PPCM varies from catastrophic to subclinical (when it may be discovered only fortuitously through echocardiography). The presenting features are those of DCM of any cause. The symptoms and signs are usually those of left ventricular failure, but less acute cases may be heralded by features of right heart failure. The diagnosis should be suspected in any woman in the third trimester of pregnancy, peripartum, or postpartum who presents with breathlessness, tachycardia, orthopnea, paroxysmal nocturnal dyspnea, arrhythmia, or signs of heart failure. Of particular concern are women who complain of cough and breathlessness and who may also have audible wheeze, where there is a risk of misdiagnosing asthma.

Presentation can be with collapse, either because of severe pulmonary edema, ventricular arrhythmias, or pulmonary embolism or because of systemic emboli from mural thrombus. Pulmonary edema precipitated by fluid overload is a common form of presentation and is often caused by the use of oxytocin (Syntocinon®; Alliance, Chippenham, UK) or by fluids given to maintain cardiac output during spinal anesthesia for delivery. The combination of a twin pregnancy, preeclampsia, steroids to induce fetal lung maturation, and then Syntocinon postpartum is not uncommonly encountered in women who subsequently present with PPCM. In the past, tocolytic therapy with beta-sympathomimetics was a common association.

The chest X-ray often shows an enlarged heart with pulmonary congestion or edema and bilateral pleural effusion.

Diagnosis

Diagnosis of PPCM involves the exclusion of pre-existing DCM or an underlying cardiac pathology such as mitral stenosis, congenital, ischemic, or hypertensive heart disease, and other causes of high output heart failure such as sepsis, severe anemia, or thyrotoxicosis.

The differential diagnosis includes other causes of postpartum pulmonary edema including preeclampsia, iatrogenic fluid overload, and nonsteroidal anti-inflammatory drugs, and other causes of collapse including pulmonary thromboembolism, amniotic fluid embolism, concealed hemorrhage, and sepsis.

Transthoracic echocardiography is the key initial investigation, and will show left ventricular systolic dysfunction. This is usually global but may be more marked in a particular territory of the ventricle. The LVEF at diagnosis is fairly consistent across studies, at 26–31%. All four cardiac chambers may be enlarged. Peripartum cardiomyopathy does not differ phenotypically from DCM of other etiologies, except in its temporal relationship to pregnancy. Cardiac MRI may be a useful additional imaging modality.[5]

High-sensitivity CRP is significantly elevated in PPCM.[30] The electrocardiogram may show a sinus tachycardia, nonspecific ST/T wave abnormality, and widened QRS complexes/left bundle branch block (LBBB), or may be normal.

Management of PPCM

Women with proven or suspected PPCM should be managed by cardiologists with expertise in heart failure as part of a multidisciplinary team. In pregnant and puerperal women, this should be in close collaboration with obstetricians and obstetric anesthetists with expertise in cardiac disease in pregnancy. For undelivered women, the high-dependency area or unit of the delivery suite is the most appropriate place to care for the mother and the fetus. Following delivery, the coronary or intensive care unit may be more appropriate for all but the mildest cases.

Management is the same as for DCM of other causes [35] and should include oxygen, diuretics, and vasodilators in the first instance.[5] ACE inhibitors are used only in the postpartum period and are safe in breastfeeding. A cardioselective or vasodilator beta-adrenergic blocking drug (such as bisoprolol or carvedilol) should be added once tolerated hemodynamically.

Thromboprophylaxis is imperative because of the increased risk of thromboembolism.[36] A high prophylactic dose of LMWH (e.g. 40 mg enoxaparin subcutaneously twice daily or 5000 IU dalteparin subcutaneously twice daily) is appropriate, unless there has been suspected or proven associated thromboembolic events or documented mural thrombus, in which case a full treatment dose of LMWH is indicated.

The most gravely ill women will need intubation, ventilation, and monitoring with the use of inotropes and sometimes temporary support from an intra-aortic balloon pump or ventricular assist device.[5] Heart transplantation may be the only chance of survival in severe cases, and early liaison with the regional cardiac transplant center is advisable for women with severe PPCM. Maximally effective contraception is imperative.[37]

Breastfeeding and bromocriptine

Given the role of the antiangiogenic 16-kDa prolactin fragment in the pathogenesis of PPCM, studies are under way into the potential beneficial therapeutic effects of bromocriptine.[34] This raises the issue of whether breastfeeding may be harmful by prolonging the period of raised prolactin levels and whether or not women with PPCM should be advised not to breast feed.[5] However, a study of 55 patients with PPCM reported that breastfeeding was significantly associated with LVEF recovery (39.5% vs 8.3%, p=0.04).[38] Given the current evidence, we would advise that clinically stable women with PPCM should not be discouraged from breastfeeding.[35]

Long-term management

As discussed below, women with severe myocardial dysfunction are unlikely to regain normal cardiac function on follow-up.[39] Those with persisting left ventricular dysfunction should be managed by a heart failure service according to current therapeutic guidelines [40] with an ACE inhibitor, beta-adrenergic receptor blocker, and consideration of cardiac resynchronization therapy.[41] The sudden cardiac death risk should be assessed and an implantable cardioverter defibrillator considered for those meeting criteria.[42]

Mortality and prognosis

Peripartum cardiomyopathy is one of the commonest causes of maternal death in the UK and the USA.[34,43] In the UK, cardiomyopathy, of which the vast majority in pregnancy is PPCM, caused 19% of the maternal deaths due to cardiac disease in the period 2006–2008.[43]

However, the mortality rate from PPCM is decreasing,[44] and series from South Africa (2013) [45] and the USA (2005) [13] quote figures of 13% and 11%, respectively. A 2011 review of US data found mortality rates of 0–19%.[35] In a study published in 2000 and comparing PPCM with other causes of cardiomyopathy, 5-year survival was 94%.[27] This reduction in mortality might reflect both improvements in therapy, and the fact that milder cases of PPCM are now being recognized.

The majority of women with PPCM present within the first month postpartum and most deaths occur from 6 weeks to 5 months after delivery.[35] Women who die usually do so soon after presentation from intractable heart failure, arrhythmia, or pulmonary or systemic thromboembolism. Predictors of death and poor outcome/ongoing poor ejection fraction include:

A literature review suggests normalization of left ventricular function (LVEF ≥50%) in >50% of women with PPCM, although there may be regional and racial variations.[35] Women who recover usually begin to do so early (within 1 month) and recover completely within 6 months; however, beyond this time, recovery is still possible.[35] The rate of left ventricular recovery is reported to be significantly lower in African-American patients compared with white patients.[35,47] The percentage recovering to normal was 55% in a population in Chicago [48] and 54–64% in a population in California [13,49] but only 21–23% in a population in South Africa.[45,48] Published US series report recovery of left ventricular function (LVEF ≥50%) at 6 months in 45–78% of patients, with a mean of 54%.[35]

Most studies have shown that long-term prognosis correlates with the degree of left ventricular dysfunction at diagnosis. Women who survive PPCM have higher ejection fractions and fractional shortening and smaller left ventricular end-diastolic dimensions (LVEDDs) at presentation than those who die, and poor prognosis is related to greater left ventricular chamber dimensions. However, Witlin et al. noted no association between fractional shortening and long-term survival and documented a trend toward increased left ventricular diameters in women with progressive disease.[50] Witlin et al. also reported that women with an LVEDD >60 mm and fractional shortening of <21% were unlikely to regain normal cardiac function on follow-up echocardiography.[39] It has been shown that inotropic contractile reserve during dobutamine stress echocardiography accurately correlates with subsequent recovery of left ventricular function and indicates a benign prognosis.[51] More recently, an LVEF of >30% and a left ventricular end-diastolic dimension of <55 mm have been shown to be significantly related to left ventricular recovery.[52]

Three retrospective cohort studies have compared the short- and long-term prognoses of PPCM and DCM due to other causes and reported worse pregnancy outcomes in women with PPCM. A 10-year study of 23 women with PPCM and 8 with DCM diagnosed prior to pregnancy found that maternal outcomes in the PPCM group were significantly worse, with three maternal deaths and four women undergoing heart transplants (p=0.05).[53] In the DCM group, one woman with a prepregnancy LVEF of 16% underwent transplantation after termination of pregnancy for genetic indications. None of the other women in the DCM group had a significant decline in cardiac status.

A Japanese group reported 29 pregnancies in 29 women with pre-existing DCM or DCM diagnosed during the first 7 months of pregnancy.[54] The study was designed to exclude PPCM by the use of previous diagnostic criteria, and compared early PACM with pre-existing DCM. This study therefore also compared acutely presenting with pre-existing cardiomyopathy and found that those presenting in pregnancy had more severely impaired left ventricular function and worse outcomes (three deaths in six women).

A study from India of 36 women with DCM (8 idiopathic, 28 PPCM) in pregnancy reported 6 maternal deaths, all in the PPCM group.[55] All of these women presented in NYHA class IV with severely impaired left ventricular systolic function.

The evidence therefore suggests that the acute maternal outcome is worse for women who present with heart failure during pregnancy. A major contributory factor may be that women with known DCM are preselected, with appropriate prepregnancy counseling and receive close supervision and medical therapy throughout pregnancy, while those with the milder forms of PPCM may go undiagnosed.

In the long term, however, the prognosis of PPCM compares favorably with that of DCM due to other causes. A US study followed 1230 patients with left ventricular impairment of various causes, including half with idiopathic cardiomyopathy (50%) and 51 with PPCM.[27] The nonpregnancy causes included myocarditis, ischemic heart disease, infiltrative myocardial disease, hypertension, HIV infection, connective tissue disease, substance abuse, and therapy with doxorubicin chemotherapy. After a mean follow-up of 4.4 years, 417 patients had died and 57 had undergone cardiac transplantation. Compared with the patients with idiopathic cardiomyopathy, women with PPCM had better survival even when adjusted for age, sex, and race (adjusted hazard ratio for death 0.31 [95% confidence interval (CI) 0.09, 0.98]).

The IMAC-2 (Intervention in Myocarditis and Acute Cardiomyopathy 2) study, a prospective US multicenter investigation of myocardial recovery in recent-onset DCM and myocarditis, enrolled 373 subjects from 2002–2008: 230 men, 104 nonperipartum women, and 39 women with PPCM.[47] All had symptom duration <6 months, a LVEF of <40%, and hypertension, valve disease, and coronary disease excluded to meet the diagnostic criteria for DCM. They were evaluated at 1 and 6 months, and then 6-monthly for up to 4 years. Medical therapy and NYHA class were comparable across the groups at enrollment. LVEF was highest at enrollment and 6-month follow-up in the PPCM group, but only by a few percent. The change in LVEF (+19%) tended to be higher in the PPCM group but failed to reach significance; however, the proportion of subjects achieving a final LVEF of >50% was significantly higher in the PPCM group (19% vs 34% vs 48%). A racial difference was apparent, with myocardial recovery to a LVEF of >50% occurring in 56% of white women and 30% of black women with PPCM. Approximately 20% of each group was left with significant left ventricular impairment (LVEF <30%). Three-year survival was 100% in women (both groups) vs 92% in men.

A study of women referred for a wearable cardioverter defibrillator reported no ventricular tachycardia/fibrillation and no mortality in 107 women with PPCM wearing the defibrillator for a mean of 124 days compared with 2 appropriate shocks (ventricular tachycardia/VF), and 11 deaths in 159 women with DCM for other causes wearing the defibrillator for a mean of 96 days.[56] Total 3-year mortality was 2.8% in the PPCM group and 16% in the DCM group.