Race, Hypertensive Disorders of Pregnancy and Outcomes in Peripartum Cardiomyopathy

2.1 Study population

The IPAC multi-center registry enrolled 100 women with newly diagnosed PPCM within 13 weeks PP. (14) Enrollment criteria included 18 years of age or older, no history of cardiac disease, left ventricular ejection fraction (LVEF) < 45% at the time of enrollment, and evaluation consistent with recent onset non-ischemic cardiomyopathy. Women with significant valvular disease, coronary disease, bacterial septicemia, ongoing drug or alcohol use disorders, history of chemotherapy, chest radiation within 5 years, or a history of previous cardiomyopathy were excluded.

2.2 Data collection

At the time of enrollment, demographic information including self-described race, pertinent delivery details, previous clinical evaluation, and current medical therapy were recorded. Medical therapy was recorded as drug class, dose was not recorded. Women were followed until 1 year PP. All hospitalizations and major cardiac events including implantation of an left ventricular assist device (LVAD), cardiac transplantation, or death were recorded. The study protocol was reviewed by the institutional review board of all participating centers, and informed consent was obtained from all subjects.

2.3 Echocardiographic assessment

Women enrolled in IPAC had an echocardiogram to assess LVEF at study entry, 6, and 12 months. Echocardiographic measurements including LVEF were assessed in a core laboratory (University of Pittsburgh). LVEF was calculated by biplane Simpson’s rule using a manual tracing of digital images. LV end-systolic and LV end-diastolic diameters (LVEDD) were assessed in the parasternal long-axis view and left atrial diameter (LAD) was assessed in the apical 4-chamber view.

2.4 Cohorts and subgroups

HDP were defined as having a diagnosis during the index pregnancy of pre-eclampsia, or gestational hypertension prior to the diagnosis of PPCM. Women with chronic hypertension (occurring prior to index pregnancy) were not included into the HDP subgroup, but were included in the no-HDP subgroup. Women who self-identified as Black or African American were included in the Black subgroup, and were compared with all other groups combined (women identifying as White, Asian or other).

2.5 Statistical Analysis

Differences in continuous variables between Black and HDP subgroups were assessed using ANOVA and two-way ANOVA. Bonferroni post-hoc testing was used to compare differences between subgroups. Differences in categorical variables between subgroups were calculated using chi-square. The Kaplan-Meier method was used to estimate time free from events, and the Log-Rank test was used to estimate differences between HDP and non-HDP subgroups. P-values <0.05 were considered statistically significant. Statistical analyses and graphics were completed using R, and Stata 12 (Stata Statistical Software: Release 12. College Station, TX: StataCorp LP.).

2.6 Outcomes

Persistent cardiomyopathy was defined as an LVEF ≤ 35% at 12 months post-partum, or last recorded LVEF. Survival free from major cardiovascular events (LVAD, transplantation, or death) was determined up to 12 months PP. Partial or full recovery was defined as LVEF>35%. A composite outcome of persistent cardiomyopathy, LVAD, transplantation, or death was used at 12 months.

3.1 Baseline Characteristics

Mean age in the overall PPCM cohort was 30 ± 6 years, with a baseline LVEF of 34% ± 10. Differences between HDP subgroups are outlined in Table 1. Among non-Black women, women with HDP delivered via caesarian more often than did non-Black women without HDP (non-Black: HDP: 15/21 vs non-HDP: 22/49; P=0.04). Among Black women with HDP, there were significantly higher rates of breastfeeding and twin-gestation compared with Black women without HDP (Breastfeeding: Black: HDP=4/16 vs non-HDP=0/14; P=0.04, Twin gestation: Black: HDP=4/16 vs non-HDP=0/14 ; P=0.04). Medication prescription including guideline directed medical therapy for heart failure was similar between the four subgroups.

3.2 Time to presentation

Differences in time to presentation post-partum are shown in Figure 1. Black women presented later than non-Black women (Black=43.0±25 days vs non-Black=26.6±23 days; P=0.0008), and women without HDP presented later than women with HDP (HDP=28.0±21 days vs no-HDP=33.5±26 days; P=0.02). Among women without HDP, Black women presented later than non-Black women (no-HDP: Black=53.6±27 days vs non-Black=27.8±24 days; P=0.02). However, among women with HDP, there was no difference between Black, and non-Black women (HDP: Black=33.6±21 days vs non-Black=23.8±21 days; P=1.00).

• Download figure
• Open in new tab

Figure 1:

Days PP at study entry by race and HDP -- Boxplots represent the interquartile range from the first to the third quartile. Median represented by a thick Black line across the interquartile range, and error bars represent 1.5 times the interquartile range. Days PP are shown between Black and non-Black women and by subgroups based on a history of HDP during index pregnancy. Significant differences between pairs of subgroups are shown using vertical hash (|) to represent comparison group, connected by a horizontal bar. HDP = hypertensive disorder of pregnancy; PP = post-partum

3.3 Remodeling, and myocardial recovery

Assessment of myocardial recovery, by measurement of LVEF at enrollment, 6, and 12 months in each subgroup is shown in Figure 2.

• Download figure
• Open in new tab

Figure 2:

LVEF (%) by race and HDP at baseline, 6, and 12 months PP -- Boxplots represent the interquartile range from the first to the third quartile. Median represented by a thick Black line across the interquartile range, and error bars represent 1.5 times the interquartile range. LVEF is shown between Black and non-Black women and by subgroups based on a history of HDP during index pregnancy. (A) Baseline. (B) Six-months (C) Twelve-months. Significant differences between pairs of subgroups are shown using vertical hash (|) to represent comparison group, connected by a horizontal bar. LVEF = left ventricular ejection fraction; HDP = hypertensive disorder of pregnancy; PP = post-partum

Remodeling

Differences in cardiac remodeling including LVEDD and LAD in each subgroup are shown in Figure 3. Among patients who did not have HDP, patients who were Black had worse LVEDD compared to non-Black (no-HDP: Black=60±7mm vs non-Black=54±6mm; P=0.029). Among patients who were Black, patients without HDP had worse LAD compared to patients who did not have HDP (Black: HDP=37±7mm vs no-HDP=42±6mm; P=0.033). Overall, women who were Black and without HDP had worse LVEDD and LAD.

• Download figure
• Open in new tab

Figure 3:

LVEDD (A) and LAD (B) (mm) by race and HDP at baseline -- Boxplots represent the interquartile range from the first to the third quartile. Median represented by a thick Black line across the interquartile range, and error bars represent 1.5 times the interquartile range. LVEDD and LAD are shown between Black and non-Black women and by subgroups based on a history of HDP during index pregnancy. Significant differences between pairs of subgroups are shown using vertical hash (|) to represent comparison group, connected by a horizontal bar. LVEDD = left ventricular end-diastolic diameter; LAD = left atrial diameter; HDP = hypertensive disorder of pregnancy.

3.4 Survival and outcomes

At the end of 12 months post-partum, 6 women had either required an LVAD or cardiac transplantation or died. Another 6 women had no recovery of LV function (LVEF ≤ 35% at 12 months). The number of women who met the composite end-point of persistent cardiomyopathy, LVAD, cardiac transplantation, or death, at the end of the 12 month follow-up period are shown in Figure 4. More Black women met the composite end-point than non-Black women (P=0.0008), and more women without HDP met the composite outcome than women with HDP (P=0.001) (Figure 4). There was no difference in event-free survival in women with versus without HDP, however no women with HDP experienced an event of LVAD or cardiac transplantation, or death (Log-Rank; P=0.06).

• Download figure
• Open in new tab

Figure 4.

Number of women with full or partial recovery compared to, LVAD, transplant, death or persistent cardiomyopathy at 12 months PP by Race and HDP– Panel A bar graph represents the proportion of women in each subgroup at 12-months, and panel B represents the number of women in each subgroup at 12-months of follow up. The outcomes at 12-months are illustrated using colored bars, no recovery/event is defined as: persistent cardiomyopathy (LVEF ≤ 35%), LVAD, cardiac transplantation, or death. Full/partial recovery is defined as those with LVEF those who experienced a full or partial recovery (LVEF > 35%) at 12 months or last follow-up. Chi-squared between groups was used to calculate statistical significance. Post-hoc testing of adjusted residuals showed the subgroup non-HDP/Black was significantly different than other subgroups (P=0.00013). Statistical significance of all groups is shown in figure (P=0.001). LVAD = left ventricular assist device; PP = post-partum; LVEF = left ventricular ejection fraction; HDP = hypertensive disorder of pregnancy.

4.1 Myocardial recovery and outcomes

These data are concordant with several studies which demonstrate worse myocardial recovery and outcomes among Black women with PPCM. (16,20,21) Observations of racial differences in this cohort have been discussed previously.(14) Reports from Germany and Japan from mostly non-Black cohorts have observed better recovery and LVEF among women with HDP. (22,23) Our study found the presence of HDP among Black women was associated with earlier presentation and better outcomes compared to Black women without HDP. Similar trends were observed from two studies of Black women from South Africa which showed an association of higher baseline systolic blood pressure in survivors compared to non-survivors. However, these studies excluded women with hypertension, and speculated lower baseline blood pressure may be related to cardiogenic shock rather than an absence of HDP. (24,25)

There are several possible explanations for our findings. The earlier presentation of patients with HDP may be explained by the observation that developing HDP during pregnancy may facilitate overall closer follow up care including into the postpartum setting, thus potentially attenuating any barriers to accessing medical care. Lewey et. al. showed delayed time to diagnosis was associated with worse outcomes among women with PPCM, a finding that was maintained in Black and non-Black subgroups irrespective of HDP status. (20) It is well known that earlier initiation of evidence-based therapies for heart failure is associated with improved cardiac remodeling and longer survival. Although prompt diagnosis is necessary before intentional prescription of evidence-based therapies such as beta-blockers or angiotensin-converting enzyme inhibitors, pharmacologic management of HDP in the peripartum period includes medicines from the same drug classes. It is possible that women with HDP may have been initiated on the same or similar medicines that would be used for guideline-directed medical therapy in heart failure, for management of HDP. Our study found no difference in heart failure medications, however patients’ medications prior to diagnosis and titration during the study period were not assessed.

4.2 Strengths and Limitations

There are several strengths and limitations that must be discussed. A dedicated echo core lab which was used to adjudicate all echo measurements for the study. In addition, this study represents the largest US registry of patients with PPCM, of which risk factors and general demographics were well representative of the general US population (30% Black individuals, 37% with HDP). The IPAC study was conducted across 30 different centers throughout the United States. This imposes heterogeneity in practice patterns and patient management which was left to the discretion of the managing physician. Detailed information regarding medical therapy was not available including doses of HF therapies. Further, although this is the largest US registry of patients with PPCM, it is still a small sample size and lacks statistical power.