Viral myocarditis
The relationship between pregnancy and viral myocarditis was established in 1968 in pregnant mice (Farber & Glasgow, 1970). Myocarditis as a cause of PPCM in humans was first suggested by Gouley et al., who corroborated infection with enlarged hearts with focal areas of fibrosis and necrosis (Gouley et al., 1937). Since then, several investigators have suggested myocarditis as a cause of PPCM (Cenac, 2003 as cited in Ntusi, 2009; Melvin, 1982; O’Connell, 1986). The prevalence of viruses detected in endomyocardial biopsies varies considerably between the different studies, ranging from less than 10% (Rizeq et al., 1994) to 78% (Midei et al., 1990), with a similar incidence in controls, suggesting no specific role for viral infection in the etiology of PPCM. It is worth noting that the molecular pathological study of endomyocardial biopsies within a cohort with PPCM found a high prevalence of viral genomes (parvovirus B19, human cytomegalovirus and herpes virus 6, Epstein-Barr virus) as well as inflammatory changes consistent with myocarditis (30.7%) (Bultmann et al., 2005). Other investigation suggests that viral infection increases the severity of myocardial damage in postpartum mice in comparison with non-pregnant control subjects (Lyden & Huber, 1984, as cited in Ramaraj & Sorrell, 2009). It is possible that the postviral immune response to be directed inappropriately against native cardiac tissue proteins leading to LV systolic dysfunction in the presence of the characteristic hemodynamic changes during pregnancy. Given the imunosuppressed state of pregnancy, it is logical that pregnant women are more susceptible to infection or viral reactivation (Pearson et al., 2000). At the present time, the exact role of viral infection or reactivation is far from conclusive. No convincing data exist that myocarditis is the primary etiology of PPCM. Further studies using newer technologies such as PCR are needed for detecting actively replicating viruses and myocardial viral load in PPCM (Ntusi et al, 2009) and confirming a pathogenic role.
Other putative hypotheses
Abnormal immune response to pregnancy
Abnormal immune response to pregnancy is another potential mechanism, probably generated by the decreased immunity during pregnancy (Cruz et al., 2010). The abnormal immune response may be produced after previous exposure immunization from prior pregnancy, or previous exposure to paternal major histocompatibility antigens. A local tissue inflammatory response is induced, followed by releasing of cytokines and a nonspecific innocent bystander myotoxicity and myocarditis (Pearson et al., 2000). Circulating auto-antibodies to selected cardiac tissue proteins were reported by several studies in more than 50% of PPCM patients (Sliwa, 2000, 2006a; Sundstrom, 2002, as cited in Cruz, 2010). Auto-antibodies are associated with increased levels of cytokines (tumor necrosis factor-a, interleukin-6, soluble Fas receptors), and are correlated with dilation of LV and systolic dysfunction (Sliwa et al., 2006b). The circulating auto-antibodies are formed against proteins released after delivery (e.g. actin, myosin), when the degeneration of the uterus occurs, and may cross-react with "target-proteins" found in the maternal myocardium (Freedman, 2004; Jahns, R., 2004). It was reported that in all patients with PPCM, irrespective of geographic location, auto-antibodies against cardiac myosin are non-selectively increased immunoglobulins G (class G and subclasses G1, G2, G3) (Warraich et al., 2005). Other studies have reported the phenomenon called chimerism, when fetal cells of hematopoietic origin reside in maternal serum, but remain undetected because of the weak immunogenicity of paternal haploytpe or maternal altered immunity (Ansari, 2002, as cited in Ramaraj & Sorrell, 2009). If fetal cells lodge in maternal myocardium during pregnancy, it is possible to be recognized as non-self while postpartum immune recovery, and an abnormal immune response is triggered (Pearson et al., 2000). At the present time, it is unclear if all these data contribute directly to myocardial injury in PPCM, or should be considered as a consequence of the disease.
Citokine-mediated inflammation
Citokine-mediated inflammation is a basic pathophysiological mechanism in heart failure. The vasodepressor pro-inflammatory cytokines, like tumor necrosis factor-a, interleukin-6 and 1, interpheron-y, expressed at high concentrations result in LV systolic dysfunction and remodeling, fetal gene expression, and cardiomyopathy. Increased levels of the same cytokines, and of hs-C-reactive protein have been reported in the serum of patients with PPCM (Fett, 2004; Sliwa, 2006b). It is still unclear if a true causal link between cytokines and PPCM does exist. If cytokines are involved in the pathogenesis of PPCM, these would prove useful targets for immunomodulatory therapy.
Increased myocyte apoptosis
Increased myocyte apoptosis represents an imbalance between cellular elimination and cellular regeneration. Experimental data suggest that terminally differentiated cardiac myocytes undergo apoptosis as the final common pathway in many cardiomyopathies (Narula, 2000; Wencker, 2003). Transgenic mice develop PPCM when cardiac-specific a-subunit of Gq is over-expressed (Hayakawa, 2003, as cited in Hilfiker-Kleiner, 2008). The Gq subunit is discussed to be responsible for coupling several cell surface receptors to intracellular signaling pathways involved in cardiomyocyte hypertrophy and apoptosis.
The inhibition of caspases, the proteases that mediate apoptosis, has been demonstrated to improve LV systolic function and reduce the mortality in pregnant Gaq mice. Recently, the proapoptotic gene Nix or Bnip3 have been demonstrated to play a key role in peripartum cardiac apoptosis and heart failure (Diwan, 2008, as cited in Hilfiker-Kleiner, 2008). Thus, experimental models, as well as indirect evidence in humans (increased plasma levels of key-proteins like Fas and Fas ligand), provides evidence for a role of apoptosis (Sliwa et al., 2006a). On the other side, the role of cardiomyocyte loss as a general key mechanism seems unlikely, since complete recovery of cardiac function has been observed in PPCM patients. Further studies are needed to evaluate the prevalence and exact role of apoptosis in PPCM patients, as well as the therapeutic value to prevent cardiomyopathy decompensation.
Abnormal response to hemodynamic stress
Abnormal response to hemodynamic stressis a hypothesis that suggests the exaggerated decrease in systolic function of LV in the presence of the cardiovascular changes in pregnancy (Ntusi et al., 2009). The normal hemodynamic changes during pregnancy result in a physiological transient and reversible hypertrophy and enlargement of the LV to meet the needs of the fetus and mother (Geva, 1997, as cited in Ntusi, 2009). These changes normally maintain up to 2-3 weeks postpartum and may persist until the 12th week after delivery. In patients with PPCM, LV anatomy may return to normal, but the contractile reserve is persistently decreased when assessed by dobutamine stress echocardiography (Lampert et al., 1997). Until now, there are no convincing data to support this hypothesis.
Genetic susceptibility
Genetic susceptibilitywas first suggested in the 1960s (Pierce at al., 1963). Since that time, several other documented cases with familial clustering of PPCM, as well as familial reports with familial PPCM and idiopathic dilated cardiomyopathy have been published, suggesting the contribution of genetics (Sliwa et al., 2010a). It is not clearly documented whether these cases meet the criterion of absence of an identifiable cause of heart failure, or whether an inherited idiopathic dilated cardiomyopathy becomes symptomatic because of the hemodynamic changes during pregnancy and after delivery. Also, the very high incidence in certain geographic regions or communities is strongly suggestive for environmental factors role. A genetic mutation cannot be excluded, but genetic testing is not usually performed in PPCM. Secondly, experimental studies have reported a genetic susceptibility to viral myocarditis in animals deficient in transforming growth factor-P, as well as the potential role of the defective STATE3 gene, or gene polymorphism of MnSOD (Hilfiker-Kleiner, 2008; Horwitz, 2007; Kim, 2005; Kuhl 2005b; Lang 2008). Recently, van Spaendonck-Zwarts et al. investigated the occurrence of PPCM in 90 families with idiopathic dilated cardiomyopathy. The authors suggested that a subset of PPCM could be an initial manifestation of the disease, when a mutation in the gene encoding cardiac troponin C was identified (van Spaendonck-Zwarts et al., 2010). In another study from the USA, Morales et al. confirmed PPCM in 5 cases with gene mutations. The involved genes encoded myosin heavy chain 7 (MYH7), sodium channel, voltage-gated, type V, a-subunit (SCN5A), and presenilin 2 (PSEN2) in 3 cases with familial disease and myosin heavy chain 6 (MYH6), cardiac troponin T2 (TNNT2) in 2 with sporadic disease (Morales et al., 2010). Both reports have important implications, suggesting the necessity for the cardiologic screening in first-degree family members of PPCM patients without recovery of LV function and dimensions. In addition, reproductive risk counseling about PPCM or pregnancy-associated cardiomyopathy is appropriate for first-degree family members of patients with idiopathic dilated cardiomyopathy in the context of a genetic evaluation (Hershberger et al., 2009).
Malnutrition
Malnutrition was thought to be involved because of increased incidence of PPCM in communities with low socio-economic level (Hull, 1937, as cited in Ntusi, 2009; Walsh, 1965). For example, selenium deficiency has been reported in Sahel region of Africa (Cenac et al., 1992) but not in Haiti, and excessive consumption of salt in Nigeria (Ntusi et al., 2009). However, malnutrition it is not a key factor because many cases of PPCM are reported in well-nourished cohorts.
Abnormal hormonal regulation
Abnormal hormonal regulation although proposed in the 1930s, cannot be affirmed (Musser, 1938, as cited in Ntusi, 2009). Estrogens and relaxin were believed to play a role in PPCM, due to the cardiovascular effects, but no convincing evidence has been documented.
Increased adrenergic tone
Increased adrenergic tone secondary to physical or emotional stress has been proposed to cause "myocardial stunning" and transient cardiac dysfunction, fluid overload, decreased colloid osmotic pressure (Wittstein et al., 2005). Considering the evidence for the role of P1- adrenergic receptor antibodies, it is possible to contribute to cardiac muscle dysfunction (Jahns R., 2004; Freedman, 2004).
Vascular disease
Vascular disease with subsequent myocardial ischemia has also been suggested, but morphology and function of coronary arteries were unaffected in PPCM patients (Koide, 1972; Lampert, 1995, as cited in Cruz, 2010).
Other possible mechanisms
Other possible mechanisms postulate the role of cardiac nitric oxide synthase, cardiac dystrophin, immature dendritic cells, toll-like receptors etc (Ramaraj & Sorell, 2009).
Main concern
What causes PPCM? Contributing factors and specific mechanisms remain unclear. Although various hypotheses have been proposed, so far no cause has been clearly identified. It is likely that PPCM is a heterogenous disorder, with a multifactorial etiology and complex biopathological processes.
Implications for research
Further studies are needed to elucidate this difficult condition. "The challenge will be to devise a study with sufficient power to give valid results" (Fett, 2010).
Diagnosis
Clinical presentation
Patients with PPCM present with classical signs and symptoms of systolic heart failure due to other cardiomyopathies. The most common symptoms are dyspnea and fatigue (90%), tachycardia (62%), and peripheral edema (60%) (Elkayam et al., 2005). Other symptoms like persistent nocturnal dry cough, orthopnea, paroxysmal nocturnal dyspnea are frequently reported (Moioli et al., 2010). NYHA class III or IV functional status seem to be the most common initial presentation (Desai et al., 1995). Other non-specific signs and symptoms include dizziness, non-specific praecordial pain (50%), abdominal discomfort, palpitations, most frequently due to tachycardia or supraventricular tachiarryhtmias (Bertrand, 1977; de Beus, 2003; Weinblatt, 1995). Complex ventricular arrhythmias and cardiac arrest have also been reported (Diao et al., 2004). Some case series describe unusual presentations such as acute cyanosis (Cole et al., 2001), multiple thromboembolic events (Carlson et al., 2000) or liver failure (Fussell et al., 2005). Systemic and pulmonary embolic episodes are found during the clinical course of PPCM more frequently than in patients with other forms of cardiomyopathy (Bennani, 2003; Box, 2004; Helms & Kittner, 2005; Jha, 2005; Lasinska-Kowara, 2001). Sudden dyspnea, pleuritic pain, and hemoptysis suggest an episode of pulmonary embolism.
Regarding the physical signs in PPCM, a high incidence of the third heart sound (92%) and displaced apical impulse (72%) are reported (Desai et al., 1996). New murmurs consistent with mitral and tricuspid regurgitation are present in almost 50% of PPCM patients (Fadouach et al., 1994). Sinus tachycardia is the rule of cardiac exam. In the later stages, signs of pulmonary hypertension, including a loud or split second heart sound and pulmonary crackles, are common. Elevated jugular venous pressure and hepatomegaly associated with edema are present as signs of congestive heart failure. Blood pressure may be normal or increased (when gestational hypertension is associated). In the later stages, postural hypotension can occur (Sliwa et al., 2010a). A latent form of PPCM without overt clinical symptoms has been reported (Elkayam et al., 2005).
The clinical diagnosis still represents a challenge because symptoms of early heart failure such as dyspnea, fatigue, palpitations, pedal edema, can appear in normal late pregnancy and after delivery. Therefore, in many cases, patients and their physicians may consider the symptoms to be normal.
There are some important clues for making the diagnosis. Clinical exam remains essential because a persistent sinus tachycardia, third heart sound, basal pulmonary crackles, and elevated jugular venous pressure are abnormal for pregnancy state and heart failure may be considered. Secondly, the diagnosis should be considered whenever women experience unexplained heart failure symptoms and signs during the last month of pregnancy or within 5 months following delivery, in accordance with PPCM definition. It is important to note that 78% of PPCM cases develop heart failure symptoms in the first 4 months after delivery, and only 9% of patients present in the last month of pregnancy (Lampert et al., 1995). It is possible that some patients to present later in postpartum because their symptoms are not initially recognized as heart failure (Sliwa et al., 2010a). Interestingly, Fett et al. reported clinically normal postpartum in Haitian women with asymptomatic echocardiographic systolic dysfunction, who either developed dilated cardiomyopathy or completely recovered LV function (Fett et al., 2005b). These cases may represent a latent phase of PPCM before the development of dilated cardiomyopathy later in life or subclinical dilated cardiomyopathy presenting in early pregnancy or a viral myocarditis, distinct conditions from true PPCM (Fett, 2008; Pyatt &Dubey, 2011). Thirdly, the rapid onset of heart failure symptoms in the peripartum period may also distinguish this clinical entity and requires further investigations.
In conclusion, there are no specific criteria for differentiating symptoms of early heart failure from normal late pregnancy, so it is imperative to maintain a high index of suspicion in conjunction with timing of symptoms to identify the patients with PPCM.
Investigation of peripartum cardiomyopathy
Blood tests should be done in all patients, although none of these can help in screening or positive diagnosis of PPCM. Initial laboratory assessment should include complete blood count and biochemical parameters. The thyroid function, a septic screen, and viral serology should also be performed in order to exclude other causes of cardiomyopathy and heart failure (Pyatt &Dubey, 2011). Molecular markers of an inflammatory process are found in most of the patients. It was reported that 90% of the patients with PPCM had high levels of plasma C-reactive protein, positively related with LV dimensions and inversely with LV ejection fraction (Fett, 2005a; Sliwa, 2006b). Cardiac markers, such as troponin T determined early after the onset of PPCM, are suggested to have prognostic significance. Only B-type natriuretic peptide (BNP) and N-terminal pro-BNP (NT-proBNP), commonly increased in patients with PPCM, are recommended by the Heart Failure Association of the ESC Working Group on PPCM to be determined (Sliwa et al., 2010a). Measurement of natriuretic peptides can be also helpful for risk stratification and volume status assessment. Increased levels in pregnancy have been related with systolic dysfunction, increased LV filling pressures and LV hypertrophy, acute myocardial infarction (Hameed, 2009; Garrison, 2005).
Genetic testing is not recommended as a routine, but only for research purposes (Sliwa et al., 2010a).
Electrocardiogram is seldom normal in patients with heart failure caused by PPCM, but it is mostly non-specific. Sinus rhythm or sinus tachycardia are usually present, atrial fibrillation or ventricular tachycardia may occur, particularly if LV systolic dysfunction becomes chronic (Diao, 2004; Duran, 2008). An intraventricular block pattern or prolonged PR and QRS are seldom reported. LV hypertrophy pattern, Q waves in the anteroseptal leads, ST-T abnormalities largely vary in incidence between studies (Brown, 1998; O’Connell, 1986, as cited in Moioli, 2010). Negative T waves can be of ischemic origin in 50% of cases (Bertrand, 1975, as cited in Bahloul, 2009).
Chest X-ray should be part of the initial assessment of all patients with PPCM and clinical heart failure. Radiological findings can be cardiomegaly, pulmonary congestion/edema, and pleural effusion.
Echocardiography is the most widely used imaging method, which provides valuable, reproducible diagnostic and prognostic information. The technique is not diagnostic for PPCM, but is important to exclude other causes of heart failure. Hibbard et al. proposed precise echocardiographic criteria that should be applied (Table 2). Several studies highlighted the strong relation between LV end-diastolic diameter > 60 mm, or ejection fraction < 30% and the recovery of LV function (Duran, 2008; Elkayam, 2005). Another important finding may be the presence of LV thrombus, particularly when LV function is severely depressed. It is important to note that LV dilatation is not always present (Kane, 2001; Sliwa, 2000). It is strongly recommended to monitor the evolution under treatment before patient’s discharge, at 6 weeks, 6 months, and annually (Sliwa et al., 2010a). Cardiac magnetic resonance imaging (MRI) is widely used in other forms of cardiomyopathy for assessment of cardiac structure and function as a reference technique. Also it has a high ability to detect myocardial fibrosis as a consequence of myocarditis, using delayed contrast enhancement technique with gadolinium. In PPCM, cardiac MRI provides more accurate quantification of chamber volumes and ventricular function, and is more sensitive in detecting LV thrombus than echocardiography (Mouquet, 2008; Srichai, 2006). At the present time, there are four case series with PPCM assessed by cardiac MRI (Caballero-Borrego, 2008; Kawano, 2008; Leurent, 2009; Mouquet, 2008). In only two of these studies the technique revealed myocardial inflammation. Baruteau et al. consider that all these MRI results are not in contradiction, but underline the complex pathogenesis of PPCM. Cardiac MRI can distinguish two forms of PPCM, inflammatory and non-inflammatory, according to the presence or absence of late gadolinium enhancement. Therefore, cardiac MRI can be helpful at initial presentation to conduct further etiologic investigations (Baruteau et al., 2010). The interest for the technique is also suggested by the ability to differentiate PPCM from other forms of cardiomyopathy, like Tako-Tsubo or ischemic cardiomyopathy. The technique might be a useful method for guiding biopsy to the abnormal area (Leurent et al., 2009), and for prognostic stratification (Kawano et al., 2008). In his comment, Fett supports cardiac MRI for PPCM which is not responding to conventional therapy as long as late gadolinium enhancement is more likely to be present in these cases (Fett, 2009). In other words, cardiac MRI could guide the immunosuppressive therapy in the inflammatory forms of PPCM, as this option of treatment has successfully been tested in "myocarditis-like" PPCM (Ntusi et al., 2009). Further larger prospective studies are needed to evaluate these findings, and the real diagnostic contribution of MRI in PPCM. The Heart Failure Association of the ESSC Working Group on PPCM recommends cardiac MRI to be performed at 6 and 12 months for a better assessment of cardiac functional changes (Sliwa et al., 2010a). It remains the problem of using gadolinium during pregnancy, not recommended by the European Society of Radiology until after delivery, unless absolutely necessary (Webb et al., 2005).
Invasive evaluation including cardiac catheterization and coronary angiography are not routinely indicated, as no specific findings are present, and coronary arteries are usually normal in PPCM.
Endomyocardial biopsy is not routinely recommended in PPCM for multiple reasons. Its role is controversial because a specific microscopic pattern for PPCM does not exist, even though a "myocarditis-like" form is frequently found (Fett, 2006a; Zimmermann, 2005). In addition, the technique is not widely available, is invasive, and has a relatively high complication rate.
Considering all these data, PPCM should be suspected whenever the patient experiences symptoms and signs of heart failure during peripartum period. A careful history and physical exam should be performed to identify heart failure due to other cardiac or non-cardiac entities. The differential diagnosis of PPCM should include all the pre-existing clinical conditions, either unrecognized, such as congenital heart disease, or unmasked by pregnancy, such as sporadic and familial idiopathic dilated cardiomyopathies, HIV/ AIDS cardiomyopathy, valvular heart disease, particularly rheumatic mitral valve disease. Other non-cardiac conditions (collagen vascular disease, sexually transmitted disease, thyroid disorders) and precipitating factors (current use of alcohol, tobacco, illicit drugs, sodium intake, other therapies) may be also considered. A useful clue for diagnosis is the onset of symptoms, most frequently in postpartum for PPCM unlike the other clinical conditions, which usually present by the 2nd trimester. Pregnancy-associated myocardial infarction, venous thromboembolism, hypertensive heart disease must be also included in the diagnostic approach. Timely diagnosis of PPCM is critical for best outcomes of survival and recovery. Very recently, Fett proposed a screening tool for early diagnosis of PPCM. The test is a focused medical history for PPCM screening, looking for the most common early signs and symptoms of heart failure during last month of pregnancy (Fett, 2011). The author proposes 6 clinical categories, easy to quantify, which are included in a self-scoring system (Table 3). A score > 5 has always been associated with LV systolic dysfunction. A score > 4 suggests the need for further investigation. In this case, a blood BNP test and an echocardiography are recommended. If the score is < 4 the patient should be monitored for BNP and C-reactive protein levels. If increased levels, echocardiography should be performed. The author emphasizes that this test is not diagnostic for PPCM, but encourages an expanded use, because it may be a useful tool for early recognition of the new onset heart failure.
In conclusion, the diagnostic work-up should focus on precise echocardiographic identification of new LV systolic dysfunction, peptide natriuretic measurement (Murali, 2005; Pearson, 2000), and ruling out other causes of heart failure. Additional investigations should be based on clinical suspicion. PPCM remains a diagnosis of exclusion. Early detection is critically important to the patient with PPCM, because delayed diagnosis may be associated with increased morbidity and mortality (Fett, 2008; Fussell, 2005; Pearson, 2000; Sliwa, 2006a).
|
Sign/symptom |
Characterstics |
Scoring |
|
Orthopnea (difficulty breathing when lying flat) |
None |
0 |
|
Need to elevate head |
1 |
|
|
Need to elevate > 450 |
2 |
|
|
Dyspnea (shortness of breath on exertion) |
None Climbing 8 or more steps Walking on level |
0 1 2 |
|
Unexplained cough |
None |
0 |
|
At night Day and night |
1 2 |
|
|
Swelling lower extremities |
None |
0 |
|
Below knee |
1 |
|
|
Above and below knee |
2 |
|
|
Excessive weight gain (during last month of pregnancy) |
< 2 pounds/week |
0 |
|
2-4 pounds/week |
1 |
|
|
> 4 pounds per week |
2 |
|
|
Palpitations (sensation of irregular heart beats) |
None |
0 |
|
When lying down at night |
1 |
|
|
Day and night, any position |
2 |
Table 3. Self-test for early diagnosis of heart failure in PPCM
Main concerns
How to optimize the diagnosis? Early involvement of a cardiologist is needed for a timely diagnosis. The rapid onset of heart failure symptoms in the peripartum period may distinguish this difficult entity, only if other causes of cardiomyopathy are excluded. A screening clinical self-test for early recognition of PPCM is now proposed. Cardiac MRI is also suggested to have a great diagnostic and prognostic potential.
What’s next in PPCM investigation?
A multicentre registry systematically using these tools may be considered for a better diagnostic approach.
