Devil in Disguise

• Editorials
• ethnic groups
• heart failure
• incidence
• peripartum cardiomyopathy
• pregnancy

See Article by S. Lee et al

Pregnancy-associated heart diseases substantially contribute to maternal morbidity and mortality. Among these, peripartum cardiomyopathy (PPCM), an idiopathic cardiomyopathy that causes heart failure with reduced left ventricular ejection fraction, represents one of the major life-threatening diseases in previously healthy women. The clinical course ranges from milder forms with slightly depressed left ventricular ejection fraction to severe forms with cardiogenic shock.^1,2 Although greater awareness and understanding of PPCM have developed over recent years, major gaps remain about epidemiology, risk factors, pathophysiology, and targeted therapy. As such, the exact diagnosis of PPCM remains a fundamental challenge in both clinical practice and scientific analysis.

In this issue of Circulation: Heart Failure, Lee et al³ present important data on the incidence and risk factors of PPCM in South Korea. The authors retrospectively analyzed a nationwide database that covers a total of 97% of the Korean population, hence expanding the knowledge of PPCM in Asian countries. The estimated incidence of PPCM in South Korea is 1:1741 (795 cases in 1 384 449 pregnancies; Figure). This compares well to the incidences reported in the United States and Germany but is markedly higher than those previously described in an analysis from Japan.^4–6 The incidence of PPCM differs widely depending on the ethnic/racial and regional background of women. Interestingly, Africans and African Americans are at a higher risk for developing PPCM, with an estimated incidence of 1:100 in Nigeria and 1:299 in Haiti,^1,3,7–9 whereas incidences in Caucasian populations range from 1:1000 in Germany to 1:10149 in Denmark.^1,5,6,10 In a Japanese cohort, the incidence was as low as 1:20 000¹¹; however, these results should be interpreted with caution because of methodological aspects, and are not confidently representative of the Asian population. Thus, additional data from Asian countries are urgently needed to clarify the incidence of PPCM in this ethnic group. A summary of the latest data is depicted in the Figure.

• Download figure
• Open in new tab
• Download powerpoint

Figure.

Incidences of PPCM. Data are derived from several studies.^1,3–6,10 Attention should be paid to the methodological differences of the cohorts. PPCM indicates peripartum cardiomyopathy.

It is of utmost importance to emphasize the different methodologies of data acquisition (prospective versus retrospective, registries versus health insurance records) when comparing PPCM incidence around the globe. Lee et al³ use a Korean nationwide database; however, PPCM diagnoses are not validated by echocardiographic data nor patient records review. A recently published Danish study⁵ also extracted data from a well-structured nationwide registry. Using International Classification of Diseases, Tenth Revision codes, these researchers identified women obviously meeting PPCM criteria. Surprisingly, ≈70% of these women had to be excluded from analysis after chart review reclassified them as not having PPCM. This limitation of nationwide registries should be taken into account when interpreting and comparing PPCM incidences.

Beyond reporting incidences of PPCM in various countries and ethnic groups, it is important to further describe these cohorts in terms of disease characteristics and risk factors to optimize timely diagnosis, therapeutic approaches, and potential prevention of PPCM. Rather than causing PPCM, pregnancy-related complications including bleeding and cesarean section are instead caused by PPCM and, therefore, do not represent classical risk factors. In contrast, older maternal age, preeclampsia, multifetal pregnancies, and higher parity (among others) have been identified as risk factors.^1,5,6,9,11 In a conclusive overview of the worldwide EURObservational Research Programme registry of the European Society of Cardiology, characteristics and risk factors of PPCM patients with different ethnicities (including Asians) were reported.¹² Despite the huge differences in ethnic and socioeconomic backgrounds among regions of the world, the baseline characteristics and mode of presentation of PPCM patients were remarkably similar. In the current analysis from Korea by Lee et al,³ several risk factors are reported: Older maternal age (≥35 years), preeclampsia, primiparity, multifetal pregnancy, gestational diabetes mellitus, peripartum hysterectomy, cesarean section, and uterine arterial embolization—each with an odds ratio and confidence interval >1 in the multivariate analysis. The risk factors in the Korean cohort are mostly consistent with those previously reported in Caucasian and African/African American women.^1,5,6,11 Data from the EURObservational Research Programme registry and the IPAC study (Investigation of Pregnancy-Associated Cardiomyopathy) from North America also support most of the Korean results.^9,12 Contrary to the data of Lee et al,³ the mean parity in the overall group in the EURObservational Research Programme registry was 3.1 (median 3.0), and the number of women with primiparity was only 15%. This difference may be explained by the very low fertility rate in South Korea, with only a minority of women entering a second pregnancy.

The exact diagnosis of PPCM and the exclusion of differential diagnoses are crucial in scientific analysis of PPCM cohorts and in clinical practice. The Heart Failure Association Study Group on PPCM of the European Society of Cardiology defines PPCM as de novo heart failure occurring toward the end of pregnancy or the following months postpartum with a left ventricular ejection fraction <45% and no other identifiable cause.¹ Despite the essential role of echocardiography in diagnosing PPCM, left ventricular ejection fraction is not available in the current Korean study nor in most of other retrospective analyses. This limitation should be kept in mind when interpreting study results.

At first glance, the definition of PPCM seems straightforward, but is it really that simple in daily practice? Previously healthy women suddenly fall ill with symptoms ranging from mild dyspnea and peripheral edema to acute severe heart failure with cardiogenic shock. Treating physicians often face diagnostic and therapeutic challenges because signs and symptoms can either mimic unspecific pregnancy-associated or postpartum transitory complaints or can be related to other serious diseases such as pulmonary embolism or acute myocardial infarction.¹³ First, 2 crucial diagnostic tests should be performed: transthoracic echocardiography can rapidly strengthen or rule out the diagnosis of PPCM. Natriuretic peptides (BNP [brain natriuretic peptide] and NT-proBNP [N-terminal pro-B-type natriuretic peptide]) remain the most well-established commercially available screening biomarker for heart failure. Although elevated levels are not specific for PPCM, normal values rule out heart failure with high certainty.

The next step in diagnosing PPCM is more difficult—exclusion of other causes of heart failure. Previously healthy young women and future mothers have often never been attended by their general physician, and only very occasionally by a cardiologist. Thus, echocardiographic data are rarely available, and physicians often have to rely on medical history and preexisting heart failure symptoms. Preexisting heart diseases (eg, congenital or radio-/chemotherapy-induced heart disease) are incompatible with the diagnosis of PPCM and should identified.

The third and final step can be the most challenging piece in the puzzle—defining time cut offs. The National Heart, Lung, and Blood Institute and the Office of Rare Diseases initially defined in 2000 that PPCM occurs during the last month of pregnancy or within the first 5-month-postpartum.¹⁴ Subsequently, it became obvious that this time frame could not be maintained, and an extension was required so as to avoid underestimation and missed diagnoses. Sliwa et al¹ redefined the PPCM diagnosis in 2010 and extended the time period to “towards the end of pregnancy or in the months following delivery.” This modification raised another question: How should we handle women with new onset of heart failure symptoms beyond the previously defined arbitrary boundary of 5 months? The ongoing discussion is whether these women should really be classified as having PPCM or another heart disease (like dilated cardiomyopathy or myocarditis) unmasked or occurring de novo in this period. Some cases can be solved by an extensive workup including cardiac magnetic resonance imaging or endomyocardial biopsy in selected cases, but some cases remain unclear despite all efforts.⁴

Lee et al³ initially used the traditional PPCM criteria (disease occurrence in the last month of pregnancy or within the following 5-month-postdelivery) for including patients in their analysis. Interestingly, the number of PPCM cases increased by 8.1% when they added data up to 12 months postpartum. This strengthens the importance of the revised PPCM definition and the need to provide women postpartum care for at least 1 year after giving birth.^1,2

Irrespective of the cause of heart failure with reduced ejection fraction, the importance of an evidence-based oral heart failure medication at evidence-based dosages to improve outcomes should be emphasized. Unfortunately, most of the published epidemiological data on PPCM cohorts do not include information on drug therapy. On top of standard heart failure medication in PPCM patients, data have accumulated to strengthen the benefit of additional treatment with the dopamine agonist bromocriptine.¹⁵

In summary, PPCM still remains the devil in disguise given that the current definition PPCM refers to a clinical syndrome rather than a distinct entity, and symptoms frequently overlap with the common discomforts that many women experience during or after childbirth. However, when women toward the end of their pregnancy or in months after delivery (up to 1 year) complain about dyspnea, peripheral edema, dizziness, chest pain, abdominal discomfort, etc, a high suspicion for PPCM is necessary. As PPCM patients can deteriorate relatively quickly with serious complications, exclusion of PPCM via measurement of BNP or NT-proBNP and transthoracic echocardiography is recommended. In general, characterization of PPCM patients and determination of risk factors in different patient cohorts from different countries and regions is desirable. Robust and convincing data play a key role, and therefore prospective data collection is strongly recommended. Close follow-up of all PPCM cases is necessary to report mid- and particularly long-term results since data on late maternal mortality (at least up to 1-year-postpartum) are lacking.²

• © 2018 American Heart Association, Inc.