Introduction
Peripartum cardiomyopathy (PPCM) is a rare but potentially life-threatening condition that occurs in previously healthy women during the last month of pregnancy and up to 5-6 months postpartum. The etiology and pathophysiology remain uncertain, although recent observations strongly suggest the specific role of prolactin cleavage secondary to unbalanced peri/postpartum oxidative stress. PPCM is a diagnosis of exclusion, as it shares many clinical characteristics with other forms of systolic heart failure secondary to cardiomyopathy. The heart failure management requires a multidisciplinary approach during pregnancy, considering the possible adverse effects on the fetus. After delivery, the treatment is in accordance with the current guidelines of heart failure. Some novel therapies, such as prolactin blockade, are proposed to either prevent or treat the patients with PPCM. A critical individual counseling concerning the risks of subsequent pregnancy must be considered. Because of its rare incidence, geographical differences, and heterogeneous presentation, PPCM continues to be incompletely characterized and understood. For all these reasons, PPCM remains a challenge in clinical practice, so future epidemiological trials and national registries are needed to learn more about the disease.
Historical perspective, definition, nomenclature
Peripartum cardiomyopathy has been described since the 19th century. In 1849, Ritchie was the first to establish a relationship between heart failure and puerperium (Ritchie, 1849). After 20 years, Virchow and Porak reported autopsy evidence of myocardial degeneration in females who died in the puerperium (Porak, 1880).
However, PPCM was not recognized as a distinctive form of cardiomyopathy until 1937, when Gouley et al. described the clinical and pathological features of seven pregnant women. The patients had severe or fatal heart failure associated with a dilated cardiomyopathy in the later months of pregnancy, which persisted after delivery, and autopsy findings of enlarged hearts with focal areas of fibrosis and necrosis, but no ischemic lesions. The authors remarked these features as atypical compared with those of other forms of myocardial failure and proposed that this heart failure was related to pregnancy and puerperium, directly or indirectly (Gouley et al., 1937). Since then, there were many reports on this form of cardiomyopathy. In 1965, Walsh et al. was the first to propose the specific period for the diagnosis, and highlighted that other conditions, which may be revealed by pregnancy, labor or postpartum period, must be excluded (Walsh et al., 1965). In 1971, Demakis et al. described the natural history of 27 pregnant females who presented with cardiomegaly and congestive heart failure and defined the condition peripartum cardiomyopathy (Demakis et al., 1971). The investigators established 3 original diagnostic criteria, which were subsequently confirmed by the National Heart Lung and Blood Institute [NHLBI] and the Office of Rare Diseases of the National Institutes of Health [NIH] Workshop, and completed with an echocardiographic criterion (Pearson et al., 2000). The new definition based on the presence of 4 criteria is summarized in Table 1.
|
Classic criteria (Demakis et al., 1971) 1. The development of heart failure in the last month of pregnancy or within the first 5 months postpartum 2. The absence of an identifiable cause for heart failure |
|
3. The absence of recognizable heart disease prior to the last month of pregnancy Additional criterion (NHLBI & the Office of Rare Disease of NIH, 1997) |
|
4. Left ventricular systolic dysfunction demonstrated by classic echocardiographic criteria (depressed ejection fraction or shortening fraction) |
Table 1. Original definition of peripartum cardiomyopathy
In 1999, Hibbard et al. proposed a more precise echocardiographic criterion that parallels those for detecting idiopathic dilated cardiomyopathy (Hibbard et al., 1999) (Table 2). The new definition has been widely accepted and has improved the diagnosis of both ventricular dysfunction and PPCM. The original definition states that PPCM must develop during the last month of pregnancy or within 5 months after delivery. However, several reports described females who presented with clear PPCM symptoms earlier during pregnancy (Alvarez, 2001; Brown, 1992; Forssell, 1994; Rizeq, 1994; Yahagi, 1994). In 2005, Elkayam et al. provided the largest retrospective database, challenging the classic criteria when they found that clinical course and outcome of females with pregnancy-associated cardiomyopathy diagnosed earlier than the last gestational month are similar to those of females with traditional PPCM. The authors concluded that these two conditions might represent a continuum of a spectrum of the same disease (Elkayam et al., 2005). Since then, several definitions have been proposed (Table 2).
In 2010, the experts considered the modification of the first criterion might be necessary. This definition specifically excludes females who develop cardiomyopathy early in their pregnancy and emphasizes that not all cases of PPCM present with LV dilation. In addition, it is recommended that other conditions which may be exacerbated and associated with heart failure in the puerperium, are excluded before the diagnosis of PPCM is considered. However, in clinical practice, it remains difficult to distinguish females with preexisting asymptomatic cardiomyopathy, progressing during pregnancy and labor, from actual PPCM females (Sliwa et al., 2010a).
Ever since the early descriptions of PPCM, the condition has been defined by several confusing names, such as post-partum heart failure, post-partum myocarditis, Meadow’s syndrome, idiopathic myocardial degeneration associated with pregnancy, Zaria syndrome, toxic post-partum heart disease, or recently, postpartal heart disease, post-partum cardiomyopathy or peripartum cardiomyopathy.
|
Hibbard et al., 1999 |
NHLBI definition and a strict ecocardiographic criterion of left ventricular (LV) dysfunction: 1. ejection fraction < 45% or fractional shortening < 30% 2. end-diastolic dimension > 2.7cm/m2 |
|
American Heart Association [AHA] Scientific Statement on contemporary definitions and classifications of the cardiomyopathies (Maron et al., 2006) |
A rare and dilated acquired primary cardiomyopathy associated with LV dysfunction and heart failure |
|
European Society of Cardiology [ESC] on the classification of cardiomyopathies (Dickstein et al., 2008) |
A non-familial, non-genetic form of dilated cardiomyopathy associated with pregnancy |
|
Heart Failure Association of the ESC Working Group on PPCM (Sliwa et al., 2010a) |
An idiopathic cardiomyopathy presenting with heart failure secondary to LV systolic dysfunction towards the end of pregnancy or in the months following delivery, where no other cause of heart failure is found. It is a diagnosis of exclusion. The LV may not be dilated but the ejection fraction is nearly always reduced below 45% |
Table 2. Definitions of peripartum cardiomyopathy
Peripartum cardiomyopathy is the preferred term because it highlights the overall chronological spectrum of the disease (Abboud et al., 2007). Another accepted term is pregnancy-associated cardiomyopathy or early peripartum cardiomyopathy, used for those females with cardiomyopathy developing heart failure before the last month of pregnancy or at least five months after delivery (Ntobeko et al., 2009). These cases may be subclinical dilated cardiomyopathies presenting the first symptoms in early pregnancy, or viral myocarditis, both distinct entities from PPCM (Pyatt & Dubey, 2011).
Epidemiology
Good data about incidence are unavailable because so few population-based registries exist. Most studies have been performed in South Africa, Haiti, and USA, but PPCM was also reported in Caucasian, Japanese, Chinese, Indian, and Korean women. Until recently, only small prospective studies reporting the experience of single centers were available to estimate the incidence of the disease (Desai et al., 1995; Fett et al., 2002, 2005a; Pyatt & Dubey, 2011). Only two large retrospective population-based studies have been conducted in USA to identify cases of PPCM. Mielniczuk et al. reported an estimated incidence of 1:3189 live births, with a trend toward an increase over the study period (1 case/2289 live births for the years 2000-2002), probably related to increasing maternal age and rates of multiple births or to increasing recognition and diagnosis of the disease (Mielniczuk et al., 2006). The second study was performed by Brar et al., who reported an incidence of 1:4025 live births (Brar et al., 2007). The estimates are almost similar for Japan and Australia. PPCM is sporadic in Europe in the white women (Bahloul et al., 2009; Ramaraj & Sorell, 2009). The estimated in-hospital mortality due to PPCM in USA is 1.36% in more recent reports, less than in older series, perhaps due in part to high utilization of modern heart failure therapy (Mielniczuk et al., 2006). These more recent data from the United States suggest a significant difference in the incidence between certain ethnic groups. The lowest observed incidence is reported in Hispanics and the highest in African-Americans (Brar et al., 2007). Outside the United States, the most comprehensive data come from the Peripartum Cardiomyopathy Project in Haiti, which estimates the incidence of PPCM as high as 1case/299 live births (Fett et al., 2005a). The data have been confirmed by Gentry et al., who noted an incidence of 1 case/1000 live births in South Africa (Gentry et al, 2010). In fact, in the absence of a multicentric trial, the incidence varies widely between African countries. For example, in Tunisia the reported incidence is very low unlike Nigeria where older studies have reported 1 case/100 live births (Bahloul et al., 2009).
On the basis of several reports series of PPCM, varying genetic pools and diverse environmental factors have been proposed as risk factors in different areas. Although not clearly delineated, there are several suggested risk factors for development and recurrence of PPCM (Bahloul et al., 2009; Demakis et al., 1971; Fett et al., 2005a; Fisher et al., 2008; Murali & Baldisseri, 2005; Moioli et al., 2010; Nkoua et al., 1991; Ntusi & Mayosi, 2009; Pearson et al., 2000; Sliwa 2006a, 2006b):
- African race – appears to be the strongest risk factor, possibly due to a greater incidence of arterial hypertension in this group. Brar et al. reported the incidence of PPCM in African-American women to be 2.9-fold higher than in whites, and 7-fold than in Hispanics (Brar et al., 2007). Recently, Elkayam has shown that PPCM in USA is not limited to African women (Elkayam et al., 2005). It remains unclear whether race represents an independent risk factor.
- advanced maternal age – the disease generally occurs over the age of 30 years;
- multiparity - 71% of cases occur after > 3 pregnancies compared with 8% in primigravidas (Demakis, 1971, as cited in Ntusi, 2009);
- twin pregnancies - which are observed in 8-13% of cases compared with 1-2% rate noted among healthy women;
- gestational hypertension - with an incidence of approximately 43%, substantially higher than the 8% to 10% incidence in the overall pregnant population. It is important to note that pregnancy-related hypertensive disorders should be considered as distinct entities from PPCM, and not included in the spectrum of PPCM. The complete recovery of LV function in pregnancy-related hypertensive disorders is the rule, whereas persistent cardiac dysfunction is frequent in PPCM patients.
- prolonged use of tocolytics refers to the use of terbutaline, salbutamol, ritodrine, isoxsuprine, magnesium sulfate etc for a period of at least four weeks (Bassett, 1985 as cited in Ntusi, 2009). The association with left ventricular dysfunction seems to be unique to pregnancy, as the same drugs do not determine similar complications in non-pregnant patients, even at high doses.
- certain cultural practices performed during the puerperium which are frequently related with high incidence of PPCM, such as consuming lake salt or rock salt known as "kanwa" (to promote the flow of breast milk), or heating of the body on a clay bed with a fire beneath to keep warm (Moioli et al., 2010; Murali & Baldisseri, 2005);
- socio-economic level is discussed as a risk factor, and can be summarized in a stereotyped profile: "poor African female, with malnutrition and multiparity, making strenuous and sustained physical effort during pregnancy" (Bahloul et al., 2009).
Main concerns
What is the true incidence? Physicians still do not know how often PPCM occurs. Despite being a rare disease in many geographic areas of the world, PPCM remains an important cause of morbidity and mortality in pregnant females.
Who is at risk? There are several cardiac factors that may play a causative role. Regardless of the documented risk factor, the association with PPCM is not clearly explained.
Implications for research
Collaborative, multicenter, prospective, population-based, well-conducted trials are required for adequate diagnosis of this condition.
Etiology and pathogenesis
Despite extensive research into its underlying etiology and pathogenesis, it is not clear exactly how PPCM occurs (Ntusi et al., 2009).
Previously, PPCM was generally considered a form of idiopathic dilated cardiomyopathy that was unmasked by the hemodynamic stress of pregnancy (Cunningham et al., 1986). In this case, one would expect PPCM to present during the second trimester coincident with the maximum hemodynamic load of pregnancy. However, it more commonly presents later in pregnancy or postpartum. Moreover, 30% of patients with PPCM experience complete recovery, with partial recovery in many cases, in contrast to rare recovery in idiopathic dilated cardiomyopathy (Fett et al., 2002). Finally, epidemiological data show that PPCM is diagnosed in young women during the peripartum period, whereas idiopathic dilated cardiomyopathy is more common in older patients (Pearson et al., 2000). Although the two conditions have similar clinical presentations and hemodynamic features, there are also significant differences in histological characteristics.
It is now accepted that PPCM is a distinct entity, rather than a clinically silent underlying cardiomyopathy exacerbated by the hemodynamic changes during pregnancy (Robson et al., 1989).
The pathogenetic mechanisms of PPCM have been difficult to study as its incidence is too low to allow meaningful evaluations, and the suitable animal models to study the disease are rare. Several hypotheses have been proposed (Figure 1), but at the present time, two hypotheses are foremost: pregnancy associated hormonal changes, specifically the role of prolactin, and viral infection.
Excessive prolactin production
Pregnancy is a physiological state associated with enhanced oxidative stress related to high metabolic turnover and elevated tissue oxygen requirements. In order to protect the heart, an efficient antioxidant defense mechanism counteracts the oxidative stress. The total antioxidant capacity increases in the last trimester with a peak early postpartum (Toescu et al., 2002).
Prolactin has been suggested as a potential mechanism in the development of PPCM (Kothari, 1997). Experimental data in a mouse model of PPCM demonstrates the activation of STAT3 pathway by 23-kDa prolactin to be necessary (Hilfiker-Kleiner et al., 2007a). STAT3 is a cardiac tissue-specific DNA-binding protein, activator of transcription-3 that promotes myocardial angiogenesis and cardiomyocyte hypertrophy. In addition,
STAT3 protects the heart from pregnancy-induced oxidative stress in part by up-regulation of a powerful reactive oxygen species, scavenging mitochondrial enzyme named manganese superoxide dismutase (MnSOD) (Negoro et al., 2001). Reduced levels of STAT3 lead to an unbalanced peri/postpartum oxidative stress, a potent stimulus for the activation of prolactin-cleaving protease catehpsin D in cardiomyocytes. The result is cleavage of the nursing hormone prolactin into an antiangiogenic, proapoptotic, and proinflammatory 16-kDa subfragment (Roberg & Ollinger, 1998). Interestingly, prolactin is a hormone with opposing cardiovascular effects, depending on the circulating form. The full-length 23-kDa prolactin had no adverse effects on the heart (Hilfiker-Kleiner et al., 2007a). In contrast, high expression of 16-kDa fragment destroys the cardiac microvasculature, reduces in vivo cardiac function, promotes ventricular dilatation. The same fragment inhibits vascular endothelial growth factor-induced proliferation of endothelial cells and migration, induces apoptosis, dissociation of capillary structures, impairs nitric oxide-mediated vasorelaxation, and cardiomyocyte function (Hilfiker-Kleiner et al., 2008). Prolactin production is not limited to pituitary gland, various other cell types, such as fibroblasts, being able to produce it (Nagafuchi et al., 1999). PPCM is often associated with a high degree of cardiac fibrosis mediated by locally produced prolactin, which enhances the circulating pituitary 16-kDa prolactin damaging cardiac effects.
Fig. 1. Summary of proposed pathogenic mechanisms for PPCM.
There is more evidence linking findings from experimental models to human PPCM. Patients with acute PPCM have increased serum levels of oxidized low-density lipoprotein indicative for enhanced oxidative stress, activated cathepsin D, and 16-kDa prolactin compared with pregnancy matched healthy controls (Hilfiker-Kleiner et al., 2007a). It is therefore likely that activation of this cascade plays a key functional role in human PPCM. PPCM patients have also significantly elevated pro-apoptotic serum markers (e.g. soluble death receptor sFas/Apo-1) with predictive power of impaired functional status and mortality (Sliwa et al., 2006b). In explanted terminally failing hearts from PPCM patients, low STAT3 protein levels are displayed, suggesting the role of this signaling pathway in the pathogenesis (Hilfiker-Kleiner et al., 2007a) (Figure 2).
Fig. 2. Schematic mechanism for the development of PPCM
Consistent with the idea of prolactin involvement, blockade by bromocriptine, a dopamine D2 receptor agonist, was tested. Bromocriptine eliminates the substrate for the generation of 16-kDa prolactin, and prevents the onset of disease in the mouse model of PPCM (Hilfiker-Kleiner et al., 2007a) (Figure 2). Several reports suggest that bromocriptine may have beneficial effects when added to the standard therapy of heart failure in women with acute onset of PPCM (Habedank et al., 2008; Hilfiker-Kleiner et al., 2007b; Sliwa et al., 2010b). However, at present, bromocriptine is not recommended until results of ongoing controlled randomized trials will provide information for the actual benefit of this therapy concept in patients with PPCM.
