Abstract
Background
LV recovery in Peripartum Cardiomyopathy (PPCM) depends on several baseline factors including left ventricular ejection fraction (LVEF), LV end-diastolic diameter (LVEDD), duration of symptoms. The role baseline RV function in LV recovery remains poorly described. This study sought to determine the relationship between baseline RV systolic function and LV recovery at six-months follow up among women with PPCM in Uganda.
Methods
Prospective cohort study of 80 PPCM cases and 80 healthy matched controls observed over a 6-month period while on goal-directed medical therapy (GDMT). All enrolled participants had a 12-lead electrocardiography, echocardiography at baseline and at 6-months follow-up for assessment of LV systolic function using LV global longitudinal strain (LV GLS) and LV EF whereas RV systolic function was assessed using; fractional area change (FAC), tricuspid annular plane excursion (TAPSE), RV lateral wall S’ (RV S’), fractional wall strain (FWS) and RV global longitudinal strain (RV GLS).
Results
The mean data of cases included age of 33.6 ± 6.6 years, LVEF 35.7 ± 11.0 %, LV GLS −11.9 ± 4.7 % and RV GLS −14.7 ± 10.9 %, RV FAC 32.9 ± 13.5 %, Lat S’ 10.6 ± 3.0 cm/s and RV FWS −17.1 ± 7.2 %. LV recovery occurred among 46.3 % cases. Factors which predicted LV recovery included heart rate and LVEDD.
Conclusion
Under half of PPCM cases had LV recovery at six-months on GDMT and Bromocriptine. Baseline heart rate and LVEDD predicted LV recovery.
Keywords: Peripartum cardiomyopathy, LV recovery, RV function, Africa
1. Introduction
Peripartum Cardiomyopathy (PPCM) emerges as a significant cause of heart failure (HF) during pregnancy that is characterized by a variable rapid clinical course particularly affecting previously healthy young women leading to substantial morbidity and mortality. A few cases have good probability for spontaneous recovery despite high rates of relapse with subsequent pregnancies [1], [2]. However, predictors of LV recovery remain poorly understood globally.
Over the past two decades, research findings to elucidate the pathophysiology of PPCM suggests a multi-hit model ranging from genetic predisposition, extremes of maternal age, multiple pregnancies, high parity and cleaved N-terminal 16-kDa prolactin fragment which is also a key predisposing factor and predictor of LV recovery [1], [3]. Treatment recommendations for PPCM patients primarily draw upon extrapolations from guidelines governing other types of systolic heart failure with an aim of controlling volume status, neutralizing maladaptive neurohormonal responses and preventing thromboembolic complications [4].
Retrospective studies of patients with PPCM in the United States have shown that mortality is as high as 18 %, LVEF returns to normal at 6 months to 1 year in only 40 % to 62 % of patients [5], [6], [7].
Contemporary studies from Nigeria and South Africa demonstrated variation in the frequencies of LV recovery with low frequencies of LV recovery associated with several predictors of poor outcomes while regular use of beta blockers correlated with late survival among the Nigerian cohort while data from the global EORP data showed regional variation in the clinical presentation and low recovery rates [8], [9].
Previous studies have identified numerous predictors of poor outcome among PPCM patients, and the most cited is the degree of LV systolic dysfunction and LV dilatation at diagnosis although none has focused on RV systolic function as a predictor of LV recovery [10].
This existing knowledge gap as evidenced by the recent publication under the PPCM prediction score [11] that did not include RV function parameters to predict LV recovery serves as an additional rationale for this study to explore the role of RV systolic function as a key prognostic indicator for LV recovery. However, RV function at diagnosis has been established as a predictor of outcomes in other acute cardiomyopathies [12] therefore it seems reasonable to expect that RV function at baseline will predict LV recovery among PPCM patients in Uganda.
2. Methods
This was a single center, prospective cohort study conducted from June 2022 to August 2023. Patient screening took place at the outpatient clinic of the Uganda Heart Institute (UHI), the antenatal and postnatal clinics of Kawempe Women’s National Referral Hospital. Recruitment and data collection took place at the Uganda Heart Institute. Participants were enrolled into the study after obtaining written informed consent where their identity remained anonymous. Ethical approval for the study was obtained from the respective Ethical Research Committees of Makerere University (School of Medicine Research Ethics Committee with IRB number Mak-SOMREC-2022–287) as the governing ethical board of UHI and University of Cape Town (Higher Research Ethics Committee with approval number HREC 093/2022) where my study is registered as a PhD research project. The study conformed to the ethical guidelines of the Declaration of Helsinki on the principles for the medical research involving human subjects [13].
2.1. Study population
The inclusion criteria for cases comprised; (i) peripartum state; (ii) manifestation of signs and/or symptoms of heart failure; (iii) left ventricular (LV) ejection fraction ≤ 45 %; and (iv) exclusion of alternative causes of heart failure. The exclusion criteria for cases included (i) Peripartum woman with heart failure confirmed with an identifiable cause of heart failure.
2.2. Sample size estimation
A study by Blauwet et al showed that using RV FAC at presentation predicted LV recovery and clinical events in patients with PPCM [14]. They further estimated that the standard deviation of the baseline RV FAC was 11.2, therefore based on these estimates we determined that at 5 % level of significance and 80 % power would require 80 participants to detect a 5-unit difference in RV FAC that ranges between 5 and 10 units. This sample size was obtained using formula of hypothesis testing for two population means.
3. Study procedure
3.1. Echocardiography protocol and equipment
A Vivid E95® (GE Healthcare) transthoracic echocardiographic machine was used to acquire the cardiac images by two cardiologists with a 1.5–4.6 MHz transducer (M5Sc) according to either ASE or ESC echo guidelines. All images were stored and analyzed by a single observer. LVEF was assessed by 2D method from parasternal long axis view using Teicholz method to assess LVEF. MAPSE was measured by M−Mode at the apical 4-chamber view. Strain analysis with STE was achieved by using LV global and regional longitudinal systolic strain measurement through 3 different views (apical 3-, 4-, and 2-chamber views). Manual adjustment was attempted, in the case of unsatisfied endocardial tracking. All the regional longitudinal strains were demonstrated on a Bull’s eye diagram, while GLS was calculated as the mean of the regional longitudinal strains. Right ventricular systolic function assessment was performed using the following modalities; Right Ventricular end-diastolic area (RV EDA) and RV end-systolic area (RV ESA) were measured with manual tracing of digital images from the apical 4-chamber view (Fig. 4). RV fractional area change (RV FAC) was calculated as follows: RV FAC (%) = (RV EDA −RV ESA)/RV EDA X 100 [15].
Fig. 4.
Kaplan Meier survival curve showing survival probability of 92.4%.
RV longitudinal free wall and global wall strain were assessed with speckle-tracking echocardiographic analysis. Measurements were performed online with dedicated software (2D Cardiac Performance Analysis) peak longitudinal systolic strain was obtained for the basal, middle and apical segments of the RV free wall and interventricular septum from software generated curves and averaged for mean RV free wall and RV global strain values, respectively in Fig. 1.
Fig. 1.
Examples of imaging modalities for assessing right ventricular systolic function in a PPCM patient with figures A showing reduced TAPSE at 1.2 cm; B shows reduced tricuspid annular S’ at 7 cm; C shows RVA (d) 20.2 cm2D shows RVA (s) 18.6 cm2 hence a reduced RV FAC 8 %; E shows reduced RV strain; F shows reduced LV strain.
3.2. Echocardiographic categorization of parameters at baseline
Using ASE guideline thresholds [16], the categorical distribution of the left ventricular systolic function parameters included; (i) Left ventricular ejection fraction categories with mild to moderate abnormal LVEF ranging between 45 % to 35 % and severely abnormal whereby LVEF was below 35 %. (ii) Global Longitudinal Strain (GLS) normal parameters > -18 % to −26 %, mildly reduced −15 % to −12.5 %, moderately reduced ranging −8.1 % to −12.5 % and severe < -8.0 %.
Categorical distribution of right ventricular systolic function parameters included (i) Tricuspid Annular Plane Systolic excursion (TAPSE) whereby the normal TAPSE was 17 mm to 25 mm, mildly reduced TAPSE was 16 mm to 15 mm, moderately reduced was 15 mm to 10 mm and severely reduced TAPSE was below 10 mm. (ii) Lateral S’ wave categorical distribution included mildly impaired with measurements ranging from 9 cm/s to 12 cm/s, moderately impaired ranging 10 cm/s to 15 cm/s and severely reduced Lat S’ ranging from 6 cm/s to 9 cm/s. (iii) Right ventricular GLS was categorized as follows; normal ranging from −35 % to −17 %, mildly reduced ranging from −16 % to −12 %, moderately reduced ranging from −11 % to −8% and severely reduced was defined as RV GLS < −7%. Categorical distribution is demonstrated in Fig. 2.
Fig. 2.
Frequency distribution of echocardiographic parameters at baseline and six-months follow up. PPCM, peripartum cardiomyopathy; LVEF, left ventricular ejection fraction; RV, right ventricle; left ventricle; GLS, global longitudinal strain.
4. Categorization of left ventricular recovery parameters at six months
The definition of the left ventricular systolic function recovery was based on ASE guidelines using Left ventricular ejection fraction ≥ 50 % at six months follow up on heart failure GDMT.
4.1. Pharmacological treatment
The utilization of optimal GDMT for heart failure included; two thirds were on beta blockers and almost quarter of the cases (n = 22, 27.5 %) were prescribed angiotensin receptor blockers, a third (n = 33, 41.3 %) were prescribed spironolactone. 58 (72.5 %) were prescribed beta blockers. Digoxin was used in 41 (51.3 %) and a diuretic in 60 (75.0 %). A total of 68.7 % were prescribed bromocriptine at a dose of 2.5 mg twice daily for 14 days and 2.5 mg once daily for 6 weeks. At six-months follow up, 12.5 % participants were prescribed Warfarin due to its availability without cost implications compared to the none-vitamin K antagonist oral anticoagulants (NOACs).
4.2. Statistical analysis
Normally distributed continuous data was summarized as mean and standard deviation (SD) while skewed data was summarized as median with corresponding interquartile range (IQR). Categorical data was summarized as frequencies and percentages and were compared using the χ2 test. For categorical variables with more than two categories, proportions were presented in tables and bar graphs. We determined the relationship between Right Ventricular systolic function parameters and LV recovery at bivariate analysis using a Pearson Chi-square test or Fisher’s exact test for categorical variables, a two-sample t-test or Wilcoxon rank sum for continuous variable. P value < 0.05 was considered statistically significant. Kaplan-Meier survival curves were traced according to baseline RV function and compared with the log-rank test. RV function parameters were included in a Cox proportional hazards regression model to correct for the baseline variables. All statistical analysis was carried out using SPSS version 17.0 software.
5. Results
5.1. Description of study participants
The baseline characteristics of the study participants enrolled are presented in Table 1. The mean age of the cases was 33.6 ± 6.6 years, mean heart rate was 90.4 ± 23.6 bpm, mean blood pressure was 114.3 ± 11.7 mmHg. Baseline echocardiographic parameters are listed in Table 2, Table 3. The mean LVEF at baseline was 35.7 ± 11.0 % with 27.5 % of the cases having LVEF < 35 %. The mean LV GLS was −11.9 ± 4.7 %, mean RV FAC was 32.9 ± 13.5 % and the mean RV GLS was −14.7 ± 10.9 %. Categorical distribution of the baseline right ventricular parameters was as follows; Using RV GLS, 34 of 80 (42.5 %) had normal RV GLS, 18 of 80 participants (22.5 %) had mild RV GLS dysfunction, 17 of 80 (21.2 %) had moderate LV systolic dysfunction and 11 of 80 participants (13.8 %) had severe systolic dysfunction using RV GLS. Table 4.
Table 1.
Baseline socio-demographic and obstetric characteristics of the study participants.
| Variables | Cases(N = 80) | Controls (N = 80) | p value |
|---|---|---|---|
| Demographic characteristics | |||
| Age in years, mean ± SD | 33.6 ± 6.6 | 30.2 ± 5.9 | 0.353 |
| Body Mass Index | 27 ± 7.0 | 25.4 ± 6.6 | 0.581 |
| Body Surface Area | 1.8 ± 0.4 | 1.8 ± 0.3 | 0.012 |
| Weeks postpostartum | 9.5 ± 6.5 | 9.2 ± 5.9 | 0.023 |
| Occupation status 0.507 | |||
| Employed | 50(48.1) | 54(51.9) | |
| Education status 0.582 | |||
| No Education | 7(58.3) | 5(41.7) | |
| Primary Education | 22(57.9) | 16(42.1) | |
| Secondary Education | 21(44.7) | 26(55.3) | |
| Tertiary Education | 30(47.6) | 33(52.4) | |
| Living location 0.011 | |||
| Urban | 67(55.8) | 53(44.2) | |
| Rural | 13(32.5) | 27(67.5) | |
| Marriage status 0.004 | |||
| Never married | 4(23.5) | 13(76.5) | |
| Married | 75(55.6) | 60(44.4) | |
| Widowed or divorced | 1(12.5) | 7(87.5) | |
| Obstetric History | |||
| Parity | 3.6 ± 2.1 | 2.7 ± 1.6 | 0.021 |
|
Gravidity 1 2–4 ≥5 |
19 (54.3)50 (54.9)11 (32.3) |
16 (45.7)41 (45.6)23 (67.6) |
0.068 |
| Number of twin pregnancies | 0.1 ± 0.3 | 0.8 ± 0.3 | 0.045 |
| Weeks at onset of heart failure | 9.5 ± 6.5 | 0.076 | |
Table 2.
Categorical distribution of echocardiographic parameters at baseline and six-months follow up.
| Echocardiogram |
P value |
||
|---|---|---|---|
| Baseline n = 80 | 6 months n = 74 | ||
| LV function parameters | |||
| LVEF | <0.001 | ||
| Recovered | 0 | 37 (100) | |
| Mild to moderate | 58 (100) | 14(31.8) | |
| Severely abnormal | 22 (48.9) | 23 (51.1) | |
| LV GLS | 0.017 | ||
| Normal | 13 (46.4) | 15 (53.6) | |
| Mildly reduced | 17 (36.2) | 30 (63.8) | |
| Moderate | 32 (68.1) | 15 (31.9) | |
| Severe | 18 (56.2) | 14 (43.8) | |
| Right ventricular function | |||
| TAPSE | 0.006 | ||
| Normal | 21 (37.5) | 35 (62.5) | |
| Mild | 0 | 3 (100) | |
| Moderate | 14 (60.9) | 9 (39.1) | |
| Severe | 45 (62.5) | 27 (37.5) | |
| RV Lat | 0.155 | ||
| Normal | 20 (43.5) | 26 (56.5) | |
| Moderate | 39 (51.3) | 37 (48.7) | |
| Severe | 21 (65.6) | 11 (34.4) | |
| RV GLS | 0.012 | ||
| Normal | 34 (43.6) | 44 (56.4) | |
| Mildly reduced | 18 (47.4) | 20 (52.6) | |
| Moderate | 17 (80.9) | 4 (19.1) | |
| Severe | 11 (64.7) | 6 (35.3) | |
Table 3.
Echocardiographic parameters of cases at baseline and six-months.
| Parameter | Baseline | 6 months | P Value |
|---|---|---|---|
| Left Ventricle Function | |||
| LVD | 6.1 ± 0.7 | 5.5 ± 1.0 | <0.001 |
| LVS | 4.9 ± 0.8 | 4.2 ± 1.2 | <0.001 |
| LVEDV | 171.8 ± 57.8 | 150.7 ± 70.3 | 0.044 |
| LVEDVI | 97.6 ± 35.3 | 83.8 ± 38.5 | 0.023 |
| LVESV | 116.3 ± 49.0 | 87.9 ± 60.6 | 0.002 |
| LVDSVI | 65.8 ± 29.2 | 46.1 ± 30.2 | <0.001 |
| LVPWD | 0.8 ± 0.6 | 0.9 ± 0.3 | 0.529 |
| LVOT_VTI | 19.6 ± 8.0 | 21.9 ± 19.0 | 0.321 |
| LV_GLS | −12.0 ± 4.7 | −12.3 ± 7.8 | 0.786 |
| Right Ventricle Function | |||
| RVD | 3.4 ± 0.8 | 2.9 ± 0.6 | <0.001 |
| RVEF | 41.2 ± 12.8 | 50.1 ± 12.3 | <0.001 |
| RV_FAC | 32.9 ± 13.5 | 39.9 ± 15.3 | 0.004 |
| RV_LAT_S | 10.6 ± 3.0 | 11.6 ± 3.0 | 0.029 |
| RV_FREEWALL_GLS | −15.7 ± 9.3 | −17.5 ± 5.1 | 0.154 |
| RV_FWS | −17.1 ± 7.2 | −20.1 ± 5.0 | 0.003 |
Table 4.
Bivariate and Multivariate analysis for predictors of LV recovery.
| Characteristic | Unadjusted OR | P-value | Adjusted OR | P-value |
|---|---|---|---|---|
| Age | 1.01 [0.94, 1.08] | 0.777 | ||
| BMI | 0.97 [0.90, 1.04] | 0.430 | ||
| Systolic BP | 1.03 [0.99, 1.07] | 0.158 | 1.01 [0.92, 1.10] | 0.893 |
| Diastolic BP | 1.01 [0.96, 1.06] | 0.748 | ||
| Heart Rate | 0.98 [0.96, 1.00] | 0.072 | 0.95 [0.91–0.99] | 0.015 |
| Creatinine | 0.17 [0.02, 0.93] | 0.062 | 0.17 [0.00, 30.96] | 0.506 |
| Hemoglobin (HB) | 1.20 [0.91, 1.60] | 0.209 | ||
| White Blood Cells (WBCs) | 0.77 [0.59, 0.96] | 0.032 | 0.90 [0.58, 1.36] | 0.603 |
| Platelets | 1.00 [1.00, 1.01] | 0.476 | ||
| Sodium | 1.13 [0.94, 1.39] | 0.210 | ||
| Potassium | 1.47 [0.78, 3.20] | 0.273 | ||
| eGFR | 1.01 [1.00, 1.03] | 0.134 | 1.02 [1.00–1.04] | 0.051 |
| NT-proBNP | 1.00 [1.00, 1.00] | 0.274 | ||
| Prolactin | 1.01 [1.00, 1.02] | 0.324 | ||
| QTc Interval | 1.00 [1.00, 1.01] | 0.475 | ||
| QRS Duration | 0.98 [0.95, 1.00] | 0.091 | 0.99 [0.96, 1.03] | 0.701 |
| LVEF | 1.08 [1.02, 1.14] | 0.006 | 1.05 [0.95, 1.18] | 0.354 |
| LVEDD | 0.17 [0.05, 0.44] | 0.001 | 0.19 [0.04–0.65] | 0.020 |
| TAPSE | 1.01 [0.95, 1.07] | 0.747 | ||
| Atrial Fibrillation | ||||
| No | 1.00 | |||
| Yes | 0.57 [0.05, 6.55] | 0.651 | ||
| LBBB | ||||
| No | 1.00 | |||
| Yes | 0.84 [0.22, 3.17] | 0.799 | ||
| Parity | ||||
| <5 | 1.00 | |||
| ≥5 | 1.13 [0.38, 3.42] | 0.823 | ||
| Symptoms of heart failure | ||||
| Class I | 1.00 | |||
| Class II | 0.85 [0.27, 2.63] | 0.773 | 4.81 [0.30, 134.58] | 0.298 |
| Class III | 1.38 [0.38, 5.07] | 0.623 | 4.84 [0.19, 303.90] | 0.393 |
| Class IV | 0.28 [0.07, 1.12] | 0.071 | 1.29 [0.10, 23.27] | 0.852 |
At six-months follow up, the RV GLS distribution was as follows; 47(59.5 %) had normal RV GLS, 22 (27.0 %) had mild RV systolic dysfunction, 17 (21.2 %) participants had moderate RV systolic dysfunction and 11 of participants (13.8 %) still had severe RV systolic dysfunction.
At baseline using LV GLS, 13 of the 80 participants (16.2 %) had normal LV GLS despite low LVEF (<45 %) at diagnosis suggestive of subclinical systolic dysfunction, 17 (21.2 %) had mild LV systolic dysfunction while the largest percentage of participants 32 (40.0 %) had moderate LV GLS and 18 of the 80 participants (22.5 %) had severe LV GLS dysfunction in Fig. 2.
5.2. Echocardiographic outcomes at six-months follow-up (Table 5)
Table 5.
Showing echocardiographic parameters at baseline, six-months follow up and post hoc analysis.
|
Parameters (mean±SD) |
Baseline |
6 Month Follow UP |
P value |
Post-hoc analysis |
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Controls |
Baseline vs Baseline |
Baseline vs 6-Months |
Baseline vs 6-Months |
6-Months vs 6Months |
6-Months vs Baseline |
6-Months vs Baseline |
|||||||||||
|
Case |
Control |
Case |
Case |
Case |
Control |
Case |
Control |
Case |
Control |
Control |
Control |
||||||
|
Case |
Control |
Case |
Controls |
Mean |
p value |
Mean |
p value |
Mean |
p value |
Mean |
p value |
Mean |
p value |
Mean |
p value |
||
| (n=80) | (n=80) | (n=74) | (n=69) | difference | difference | difference | difference | difference | difference | ||||||||
| Left Ventricular Function | |||||||||||||||||
| LVD, mm | 6.1±0.7 | 4.6±4 | 5.5±1 | 5.2±4.9 | <0.001 | -0.6 | 0.022 | -0.6 | 1 | -0.9 | 0.687 | -0.9 | 1 | -1.5 | 0.504 | -0.3 | 1 |
| LV, GLS, -% | -11.9±4.7 | -17.1±4.9 | -12.3±7.8 | -16.2±9.5 | <0.001 | -0.4 | <0.001 | -0.4 | 1 | -4.8 | 0.00115 | -4.3 | 0.00484 | -5.2 | <0.001 | -3.9 | 1 |
| LVEF, % | 35.7±11 | 67.2±4.5 | 45.9±15.7 | 65.9±9.9 | <0.001 | 10.2 | <0.001 | 10.2 | <0.001 | 21.3 | <0.001 | 30.2 | <0.001 | 31.5 | <0.001 | 20 | 1 |
| Right Ventricular Function | |||||||||||||||||
| RVEF, % | 41.2±12.8 | 53.3±7.9 | 50.1±12.3 | 56±9.5 | <0.001 | 12.1 | <0.001 | 8.9 | <0.001 | 14.8 | <0.001 | 5.9 | 0.0071 | 3.2 | 0.4473 | 2.7 | 0.7113 |
| RV FAC, % | 32.9±13.5 | 38.4±7.7 | 39.9±15.3 | 43.7±12.1 | <0.001 | 5.5 | 0.0346 | 7 | 0.0038 | 10.8 | <0.001 | 3.8 | 0.3972 | -1.5 | 1 | 5.3 | 0.0598 |
| TAPSE, cm | 8.7±7.6 | 8.8±8.6 | 12.5±8.2 | 11.6±8.7 | 0.636 | 0.1 | 1 | 3.8 | 0.028 | 2.9 | 0.198 | -0.9 | 0.198 | -3.7 | 0.036 | 2.8 | 0.24 |
| RV Lats', cm/s | 10.6±3 | 12.8±6.9 | 11.6±3 | 11.9±2.3 | <0.001 | 2.2 | 0.0085 | 1 | 0.7349 | 1.3 | 0.3729 | 0.3 | 1 | 1.2 | 0.6509 | -0.9 | 1 |
| RV Free wall GLS, -% | -14.7±10.9 | -20.4±4.7 | -17.5±5.1 | -20.2±4.8 | <0.001 | -5.7 | <0.001 | -2.8 | 0.084 | -5.5 | <0.001 | -2.7 | 0.109 | -2.9 | 0.053 | 0.2 | 1 |
| RV FWS, -% | -17.1±7.2 | -21±8.3 | -20.1±4.9 | -21.3±4.2 | <0.001 | -3.9 | 0.00109 | -3 | 0.02679 | -4.2 | <0.001 | -1.2 | 1 | -0.9 | 1 | -0.3 | 1 |
LV recovery (LV EF > 50 %) occurred in 37 (46.3 %) participants, mean LVEF was 45.9 ± 15.7 %, there was an absolute increase in LVEF from 35.7 ± 11.0 % to 45.9 ± 15.7 % (p < 0.001) which was accompanied by a reduction in LV end-systolic diameter (LVESD) from 6.1 ± 0.7 cm to 5.5 ± 1.0 cm (p < 0.001).
At the end of the six-months, 14 (31.8 %) participants had moderate LV systolic dysfunction and 23 (51.1 %) had severe LV systolic dysfunction.
At six-months follow up, there was documented improvement with the RV GLS systolic function. A total of 48 (59.5 %) participants had normal RV GLS, and a substantial number of 4(5.4 %) participants had moderate RV GLS dysfunction which indicated an improvement with RV systolic function on goal directed medical therapy at six-months follow up.
There was a documented improvement in LV systolic function using LV GLS at six-month follow up with 32 (40.5 %) participants documented to have mild LV GLS dysfunction and (16) 20.3 % participants with moderate LV GLS dysfunction (Fig. 3).
Fig. 3.
Percentage frequencies of PPCM cases and changes in patients’ RV GLS and LV GLS at baseline and at 6 months follow-up.
The six-month Kaplan Meier survival curve probabilities of the 80 cases is shown in Fig. 4. (2.5 %) participants were lost to follow-up and 6 (7.5 %) participants died. A total of 10 (12.5 %) of the participants received anticoagulation for atrial fibrillation (n = 2, 2.5 %) or intracardiac thrombus (n = 8, 10 %), 2 (2.5 %) participants developed stroke. 1 (1.25 %) participants had an ICD inserted for intractable ventricular tachycardia, 52 (65 %) participants were in NYHA Class 1. Fig. 5.
Fig. 5.
Calibration plot of the model.
The evaluated right and left ventricular functional parameters demonstrated limited discriminatory ability for the outcome in Table 6, with AUC values ranging from 0.35 to 0.55. Among these, LV GLS showed the highest AUC (0.55) with high specificity (84 %) but poor sensitivity (27 %), while RV EF (AUC 0.51) and RV FAC (AUC 0.50) performed at chance level with modest sensitivity but low specificity. Other indices, including RV GLS, TAPSE and RV Lat S’ which yielded AUC values below 0.5 which reflected weak classification power. Although LVEF (AUC 0.40, p = 0.038) and LVED (AUC 0.35, p < 0.001) were statistically significant, their poor predictive performance suggested that these associations may lack clinical utility. Overall, none of the tested parameters provided meaningful balance between sensitivity and specificity, highlighting their limited value as standalone predictors.
Table 6.
Showing parameter associated with persistent severe left ventricular systolic dysfunction.
| PARAMETER | Cutoff | Sensitivity % | Specificity % | AUC | P Value |
|---|---|---|---|---|---|
| RVEF | ≤45 | 59 | 42 | 0.51 | 0.910 |
| RV_FAC | ≤36 | 59 | 40 | 0.50 | 0.932 |
| LVGLS | >-18 | 27 | 84 | 0.55 | 0.247 |
| RVGLS | >-13.6 | 32 | 63 | 0.48 | 0.661 |
| TAPSE | ≤16 | 70 | 23 | 0.47 | 0.518 |
| RV LAT S’ | <10 | 22 | 70 | 0.46 | 0.389 |
| LVEF | ≤28 | 16 | 63 | 0.40 | 0.038 |
| LVEDD | >57 | 62 | 70 | 0.35 | <0.001 |
6. Discussion
The major findings of our study are as follows: (1) A substantial proportion of patients with PPCM recovered based on an absolute threshold of the LVEF ≥ 50 % at six-months of follow up which was accompanied with a reduction in LV end-systolic diameter. (2) There was improvement of RV systolic function parameters specifically RV TAPSE and RV GLS from baseline to six-months follow up however, none of the RV systolic function parameters predicted LV recovery. (3) The present study demonstrates heart rate and LVEDD predicted LV recovery in this cohort of PPCM cases. To the best of our knowledge, this prospective study represents one of the first studies in an African population to address the prognostic predictors of LV recovery and the value of RV systolic function parameters as independent predictors of LV recovery in women with PPCM specifically in the Ugandan context.
PPCM is distinct from other types of cardiomyopathies since it demonstrates a more favorable recovery rate ranging from 45 % to 78 % within six months post diagnosis although the symptoms are similar to those of dilated cardiomyopathy [17], [18] Oxidative stress, genetic susceptibility, autoimmunity and myocarditis have all been implicated in the pathogenesis of PPCM and other types of cardiomyopathy [19], [20].
Our findings indicate a higher burden of RV systolic dysfunction compared to most PPCM cohort studies [21]. Findings from the EuroObservational Research Programme registry demonstrated a less burden of RV systolic dysfunction which was reported among 47 % of women [22]. These differences may be attributed to factors such as difference in the study methodology, classification of RV size and function and potential regional phenotypical differences.
TAPSE is an easily obtainable parameter that is commonly used to assess RV function in clinical practice, however our study consistent with findings from Blauwet L et al revealed TAPSE is not a reliable predictor of PPCM recovery [14]. TAPSE assumes that the displacement of basal and adjacent segments of RV represents the entire RV function which in reality contributes 60–70 % of RV function therefore giving a limitation to assess the contractility of the interventricular septum. We however documented improvement of TAPSE from baseline to 6 months follow up. TAPSE has been found to be significantly lower in PPCM cases, compared to the level in those diagnosed with dilated nonischemic cardiomyopathies [23].
Other parameters such as RV lateral S’ is not a reliable RV function parameter as it relies on tissue doppler that is subject to accuracy in the angle of alignment and volume overload in the left ventricle. In other studies, FAC has been found to have a significant role in predicting LV recovery[14]. RV FAC has the ability to detect RV alterations especially those affecting the interventricular septum.
Previous findings from the Investigations of Pregnancy-Associated Cardiomyopathy (IPAC) study showed a high incidence of LV recovery among this cohort of PPCM participants. Furthermore, the study showed that RV function per echocardiographic fractional area change at presentation was correlated with both subsequent LV recovery and clinical outcomes emphasizing its prognostic significance [14].
Studies using strain methodology to assess RV function in PPCM remain limited, however we documented a significant change in the RV GLS from baseline to six months without statistical significance in predicting LV recovery.
Complete LV recovery was documented among 46.3 % of PPCM cases in the present study, other studies have reported complete recovery rates between 20 % and 60 % [24], [25]. In previous studies, factors that predicted LV recovery included baseline ejection fraction ≥ 30 % and LVEDD < 5.5 cm [26], [27]. All these studies were from single centers and involved fewer than 55 PPCM cases with limited follow up. A recent prospective study by McNamara et al followed up 100 patients in USA where they documented LV recovery in 65 % of women at 6 months and demonstrated baseline EF and race were the predictors for LV recovery [28].
A recent report of 176 South African patients with PPCM followed up for only six months concluded that older age and lower LV end-systolic diameter predicted LV recovery [25]. This variation in predictors of LV recovery highlight the difference in genetic and geographical findings in PPCM cohorts.
LV dilatation does not occur in all patients with PPCM. As a response to LV dysfunction caused by volume overload issues, the cardiac remodeling process may generate LV dilatation therefore, our results highlight the importance of integrating LVEDD and heart rate at diagnosis, providing valuable prognostic insights that could inform patient management strategies for PPCM both in Uganda and globally.
Further still, supporting findings from other studies that RV systolic dysfunction in acute nonischemic dilated cardiomyopathy including PPCM is an independent predictor of morbidity and mortality. The exact mechanisms of the underlying mechanisms of RV systolic dysfunction in these patients is unclear. However, it seems likely that RV function is linked to LV function in at least some patients and potential mechanisms by which LV systolic dysfunction leads to RV systolic dysfunction include involvement of RV by the same cardiomyopathic process involving the LV and secondary pulmonary hypertension resulting in increased RV afterload [29].
Similarly, the recently published ESC EORP predictive score variables [11], results from our study demonstrated LVEDD as a common predictor for LV recovery in PPCM patients in Uganda.
Since an echocardiogram is a noninvasive tool that is not available in most health care facilities in Uganda, clinicians can easily and quickly use heart rate to evaluate the prognosis of PPCM patients at first medical contact in low income countries.
6.1. Limitations
This study had several limitations. We had a small sample size of 80 cases given the rarity of PPCM which limits the generalizability of our study findings majorly to the interest of local readers. Additionally, the deaths of 6 participants further compromised the robustness of the outcome assessment and our conclusions therefore further studies with larger sample sizes are warranted.
7. Conclusion
Findings in this prospective study potentially contribute valuable insights with LVEDD and heart rate at presentation are strong independent prognostic predictors of LV recovery in patients with PPCM in Uganda and sub-Saharan Africa.
Funding
The work was supported by Uganda Heart Institute and Cape Heart Institute.
CRediT authorship contribution statement
Juliet Nabbaale: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Emmy Okello: Writing – review & editing, Visualization, Validation, Supervision, Resources, Methodology, Funding acquisition, Formal analysis, Conceptualization. Annettee Nakimuli: Writing – review & editing, Writing – original draft, Validation, Supervision, Project administration, Methodology, Funding acquisition, Data curation, Conceptualization. Graham Chakafana: Writing – review & editing, Visualization, Validation, Supervision, Software, Resources, Methodology, Investigation, Formal analysis, Conceptualization. Karen Sliwa: Writing – review & editing, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Funding acquisition, Formal analysis, Data curation, Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
“This author takes responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation”.
References
- 1.Sliwa K., Bauersachs J., Arany Z., Spracklen T.F., Hilfiker-Kleiner D. Peripartum cardiomyopathy: from genetics to management. Eur. Heart J. 2021;42(32):3094–3102. doi: 10.1093/eurheartj/ehab458. [DOI] [PubMed] [Google Scholar]
- 2.Bhattacharyya A., Basra S.S., Sen P., Kar B. Peripartum cardiomyopathy: a review. Tex. Heart Inst. J. 2012;39(1):8–16. [PMC free article] [PubMed] [Google Scholar]
- 3.Hoes M.F., Arany Z., Bauersachs J., Hilfiker-Kleiner D., Petrie M.C., Sliwa K., et al. Pathophysiology and risk factors of peripartum cardiomyopathy. Nat. Rev. Cardiol. 2022;19(8):555–565. doi: 10.1038/s41569-021-00664-8. [DOI] [PubMed] [Google Scholar]
- 4.Bauersachs J., Arrigo M., Hilfiker-Kleiner D., Veltmann C., Coats A.J., Crespo-Leiro M.G., et al. Current management of patients with severe acute peripartum cardiomyopathy: practical guidance from the Heart failure association of the european society of cardiology study group on peripartum cardiomyopathy. Eur. J. Heart Fail. 2016;18(9):1096–1105. doi: 10.1002/ejhf.586. [DOI] [PubMed] [Google Scholar]
- 5.Chapa J.B., Heiberger H.B., Weinert L., Decara J., Lang R.M., Hibbard J.U. Prognostic value of echocardiography in peripartum cardiomyopathy. Obstet. Gynecol. 2005;105(6):1303–1308. doi: 10.1097/01.AOG.0000161382.30233.ba. [DOI] [PubMed] [Google Scholar]
- 6.Amos A.M., Jaber W.A., Russell S.D. Improved outcomes in peripartum cardiomyopathy with contemporary. Am. Heart J. 2006;152(3):509–513. doi: 10.1016/j.ahj.2006.02.008. [DOI] [PubMed] [Google Scholar]
- 7.Mielniczuk L.M., Williams K., Davis D.R., Tang A.S., Lemery R., Green M.S., et al. Frequency of peripartum cardiomyopathy. Am. J. Cardiol. 2006;97(12):1765–1768. doi: 10.1016/j.amjcard.2006.01.039. [DOI] [PubMed] [Google Scholar]
- 8.Sliwa K., Petrie M.C., van der Meer P., Mebazaa A., Hilfiker-Kleiner D., Jackson A.M., et al. Clinical presentation, management, and 6-month outcomes in women with peripartum cardiomyopathy: an ESC EORP registry. Eur. Heart J. 2020;41(39):3787–3797. doi: 10.1093/eurheartj/ehaa455. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Karaye K.M., Sa'idu H., Balarabe S.A., Ishaq N.A., Adamu U.G., Mohammed I.Y., et al. Clinical features and outcomes of peripartum cardiomyopathy in nigeria. J. Am. Coll. Cardiol. 2020;76(20):2352–2364. doi: 10.1016/j.jacc.2020.09.540. [DOI] [PubMed] [Google Scholar]
- 10.McNamara D.M., Starling R.C., Cooper L.T., Boehmer J.P., Mather P.J., Janosko K.M., et al. Clinical and demographic predictors of outcomes in recent onset dilated cardiomyopathy: results of the IMAC (intervention in Myocarditis and Acute Cardiomyopathy)-2 study. J. Am. Coll. Cardiol. 2011;58(11):1112–1118. doi: 10.1016/j.jacc.2011.05.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jackson A.M., Goland S., Farhan H.A., Yaseen I.F., Prameswari H.S., Böhm M., et al. A novel score to predict left ventricular recovery in peripartum cardiomyopathy derived from the ESC EORP Peripartum Cardiomyopathy Registry. Eur. Heart J. 2024;45(16):1430–1439. doi: 10.1093/eurheartj/ehad888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gulati A., Ismail T.F., Jabbour A., Alpendurada F., Guha K., Ismail N.A., et al. The prevalence and prognostic significance of right ventricular systolic dysfunction in nonischemic dilated cardiomyopathy. Circulation. 2013;128(15):1623–1633. doi: 10.1161/CIRCULATIONAHA.113.002518. [DOI] [PubMed] [Google Scholar]
- 13.World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. Jama. 2013;310(20):2191-4. [DOI] [PubMed]
- 14.Blauwet L.A., Delgado-Montero A., Ryo K., Marek J.J., Alharethi R., Mather P.J., et al. Right Ventricular Function in Peripartum Cardiomyopathy at Presentation is Associated with subsequent Left Ventricular Recovery and Clinical Outcomes. Circ. Heart Fail. 2016;9(5) doi: 10.1161/CIRCHEARTFAILURE.115.002756. [DOI] [PubMed] [Google Scholar]
- 15.Lang R.M., Badano L.P., Mor-Avi V., Afilalo J., Armstrong A., Ernande L., et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J. Am. Soc. Echocardiogr. 2015;28(1):1–39.e14. doi: 10.1016/j.echo.2014.10.003. [DOI] [PubMed] [Google Scholar]
- 16.Rudski L.G., Lai W.W., Afilalo J., Hua L., Handschumacher M.D., Chandrasekaran K., et al. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American Society of Echocardiography: endorsed by the European Association of Echocardiography, a registered branch of the European Society of Cardiology, and the Canadian Society of Echocardiography. J. Am. Soc. Echocardiogr. 2010;23(7):685–713. doi: 10.1016/j.echo.2010.05.010. [DOI] [PubMed] [Google Scholar]
- 17.McNamara D.M., Elkayam U., Alharethi R., Damp J., Hsich E., Ewald G., et al. Clinical Outcomes for Peripartum Cardiomyopathy in North America: results of the IPAC Study (Investigations of Pregnancy-Associated Cardiomyopathy) J. Am. Coll. Cardiol. 2015;66(8):905–914. doi: 10.1016/j.jacc.2015.06.1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Davis M.B., Arany Z., McNamara D.M., Goland S., Elkayam U. Peripartum Cardiomyopathy: JACC State-of-the-Art Review. J. Am. Coll. Cardiol. 2020;75(2):207–221. doi: 10.1016/j.jacc.2019.11.014. [DOI] [PubMed] [Google Scholar]
- 19.Sliwa K., Hilfiker-Kleiner D., Petrie M.C., Mebazaa A., Pieske B., Buchmann E., et al. Current state of knowledge on aetiology, diagnosis, management, and therapy of peripartum cardiomyopathy: a position statement from the Heart failure Association of the European Society of Cardiology Working Group on peripartum cardiomyopathy. Eur. J. Heart Fail. 2010;12(8):767–778. doi: 10.1093/eurjhf/hfq120. [DOI] [PubMed] [Google Scholar]
- 20.Pearson G.D., Veille J.C., Rahimtoola S., Hsia J., Oakley C.M., Hosenpud J.D., et al. Peripartum cardiomyopathy: National Heart, Lung, and Blood Institute and Office of Rare Diseases (National Institutes of Health) workshop recommendations and review. JAMA. 2000;283(9):1183–1188. doi: 10.1001/jama.283.9.1183. [DOI] [PubMed] [Google Scholar]
- 21.Haghikia A., Röntgen P., Vogel-Claussen J., Schwab J., Westenfeld R., Ehlermann P., et al. Prognostic implication of right ventricular involvement in peripartum cardiomyopathy: a cardiovascular magnetic resonance study. ESC Heart Failure. 2015;2(4):139–149. doi: 10.1002/ehf2.12059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sliwa K., Mebazaa A., Hilfiker-Kleiner D., Petrie M.C., Maggioni A.P., Laroche C., et al. Clinical characteristics of patients from the worldwide registry on peripartum cardiomyopathy (PPCM) EURObservational Research Programme in conjunction with the Heart failure Association of the European Society of Cardiology Study Group on PPCM. Eur. J. Heart Fail. 2017;19(9):1131–1141. doi: 10.1002/ejhf.780. [DOI] [PubMed] [Google Scholar]
- 23.Karaye K.M. Right ventricular systolic function in peripartum and dilated cardiomyopathies. Eur. J. Echocardiogr. 2011;12(5):372–374. doi: 10.1093/ejechocard/jer024. [DOI] [PubMed] [Google Scholar]
- 24.Elkayam U., Akhter M.W., Singh H., Khan S., Bitar F., Hameed A., et al. Pregnancy-associated cardiomyopathy: clinical characteristics and a comparison between early and late presentation. Circulation. 2005;111(16):2050–2055. doi: 10.1161/01.CIR.0000162478.36652.7E. [DOI] [PubMed] [Google Scholar]
- 25.Blauwet L.A., Libhaber E., Forster O., Tibazarwa K., Mebazaa A., Hilfiker-Kleiner D., et al. Predictors of outcome in 176 South african patients with peripartum cardiomyopathy. Heart. 2013;99(5):308–313. doi: 10.1136/heartjnl-2012-302760. [DOI] [PubMed] [Google Scholar]
- 26.Safirstein J.G., Ro A.S., Grandhi S., Wang L., Fett J.D., Staniloae C. Predictors of left ventricular recovery in a cohort of peripartum cardiomyopathy patients recruited via the internet. Int. J. Cardiol. 2012;154(1):27–31. doi: 10.1016/j.ijcard.2010.08.065. [DOI] [PubMed] [Google Scholar]
- 27.Duran N., Günes H., Duran I., Biteker M., Özkan M. Predictors of prognosis in patients with peripartum cardiomyopathy. Int. J. Gynecol. Obstet. 2008;101(2):137–140. doi: 10.1016/j.ijgo.2007.11.007. [DOI] [PubMed] [Google Scholar]
- 28.McNamara D., Damp J., Elkayam U., Hsich E., Ewald G., Cooper L., et al. Lippincott Williams & Wilkins; 2013. Myocardial recovery at six months in peripartum cardiomyopathy: results of the NHLBI Multicenter IPAC study. [Google Scholar]
- 29.Voelkel N.F., Quaife R.A., Leinwand L.A., Barst R.J., McGoon M.D., Meldrum D.R., et al. Right ventricular function and failure: report of a National Heart, Lung, and Blood Institute working group on cellular and molecular mechanisms of right heart failure. Circulation. 2006;114(17):1883–1891. doi: 10.1161/CIRCULATIONAHA.106.632208. [DOI] [PubMed] [Google Scholar]