Abstract
Introduction
Peripartum cardiomyopathy (PPCM) is a rare type of cardiomyopathy that manifests as acute heart failure associated with pregnancy. Delays in early identification result in poor recovery of left ventricular (LV) function; however, no risk prediction model exists. We sought to yield a scoring system known as the Padjadjaran Peripartum CardioMyopathy Recovery (PPCM recovery) score to predict the probability of poor LV function recovery in PPCM patients.
Methods
All baseline and clinical parameters were prospectively collected from a cohort of patients with PPCM admitted to Dr. Hasan Sadikin General Hospital in Bandung, Indonesia between January 2014 and December 2021. Logistic regression analyses were performed to investigate the relationship between each variable and the risk of poor LV function recovery in PPCM patients.
Results
This prospective cohort study included 113 patients with PPCM (84 recovered and 29 non-recovered patients). Significant mitral regurgitation (MR), left ventricular ejection fraction (LVEF) <30%, left ventricular end-diastolic diameter (LVEDD) ≥56 mm, and New York Heart Association functional class (NYHA FC) IV were all strong predictors of poor LV function recovery. These variables were integrated into the PPCM recovery score (AUC of 0.85). Patients with a score of ≥8 were nearly 18 times more likely to have poor LV function recovery (sensitivity 57%, specificity 93%).
Conclusion
PPCM recovery score is a convenient scoring system based on clinical and echocardiography assessment that may assist in distinguishing which patients are more likely to develop poor LV function recovery; therefore, these patients should be immediately referred to a tertiary referral hospital.
Introduction
Peripartum cardiomyopathy (PPCM) is an idiopathic cardiomyopathy presenting with heart failure due to reduced left ventricular (LV) ejection fraction (LVEF) below 45%. This type of cardiomyopathy occurs in women at the end of pregnancy or within months after delivery without other apparent causes of heart failure [Citation1,Citation2]. Patients with PPCM are mostly young, often desire to get pregnant again, and have just begun their families; therefore, the long-term consequences of PPCM are commonly questioned [Citation3]. Several etiology may harm cardiomyocytes and contribute to LV dysfunction in PPCM patients, including genetic and hemodynamic abnormalities, elevated prolactin levels, viral infection, autoimmune response, toxins, and a pro-inflammatory state [Citation4].
Recovery from PPCM has been defined as LVEF ≥50%, which seems to be more favorable than other forms of cardiomyopathy [Citation4–6]. Persistent cardiac dysfunction indicates irreversible damage and predicts poor survival [Citation7,Citation8]. Registry from The European Society of Cardiology - EURObservational Research Programme showed PPCM patients recover their cardiac anatomy and function completely within six months after initial diagnosis which occurred most commonly in the Asia-Pacific region (62%) [Citation9]. Despite more frequent reports on early recovery of LV function, the postulation of LV function recovery beyond six months has been increasingly acknowledged. A prospective multi-center American study of PPCM, Investigations of Pregnancy Associated Cardiomyopathy (IPAC) study, observed LV recovery in 72% of PPCM women at one-year follow-up [Citation10]. LV recovery in PPCM patients before a subsequent pregnancy is associated with better cardiac function and reduced mortality [Citation3]. Predicting LV recovery is therefore essential because women with impaired LV function at the onset of a subsequent pregnancy are at a greater risk of relapse, heart failure, and death. Therefore, this group of patients should avoid pregnancy [Citation3]. Early identification of patients at risk of poor LV recovery is essential because it can serve as a trigger for clinicians to provide appropriate, aggressive, and comprehensive treatment. Predictors of non-recovery have been previously reported and include baseline severely depressed LV function (LVEF <30%) and enlarged LV characterized by LV end-diastolic diameter (LVEDD) ≥56 cm [Citation4,Citation8–10]. These predictors are easily evaluated; however, a single predictor is less precise to foresee LV recovery. Combining baseline LVEF, LVEDD, and other parameters may enable more accurate LV recovery prediction. Thus far, no cohort study has generated a scoring system for independent predictors of poor LV recovery in patients with PPCM. Additionally, predicting the risk of poor recovery outcomes for PPCM patients becomes burdensome in remote areas with limited facilities. The present study aimed to evaluate the predictors of LV recovery and establish the Padjadjaran Peripartum CardioMyopathy Recovery (PPCM Recovery) score to predict the risk of poor LV recovery at one year after PPCM diagnosis.
Methods
Study population
From January 2014 to December 2021, all PPCM patients aged ≥18 years admitted to Dr. Hasan Sadikin General Hospital in Bandung, Indonesia participated in this single-center, prospective cohort study. All included subjects provided informed consent at the beginning of the study and those who did not were omitted. This study was approved by the Medical Research Ethics Committee of Dr. Hasan Sadikin General Hospital, Bandung, Indonesia, in accordance with the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all the subjects. The exclusion criteria were as follows: 1. patients who died from a cause other than PPCM, and 2. patients who were lost to follow-up after the follow-up period.
Definition of variables
The diagnosis of PPCM was confirmed if the following criteria were fulfilled [2]: 1. patients presenting with congestive heart failure symptoms during the prepartum period or within six months after delivery, 2. absence of other possible etiology of heart failure discovered by clinical examination and echocardiography prior the last month of pregnancy; 3. LVEF <45%.
Data on numerous patients’ baseline characteristics were obtained when PPCM was initially diagnosed, including age, body mass index (BMI), comorbidity (chronic hypertension, hypertension in pregnancy, preeclampsia, and obesity (BMI ≥30 kg/m2), obstetric status (multiparity, twin pregnancy, and history of pregnancy resulting in fetal death), history of breastfeeding, PPCM onset, heart rate, and blood pressure.
The severity of heart failure symptoms in patients with PPCM was classified according to the New York Heart Association functional classification (NYHA FC): Class I, no limitation of activities; Class II, mild symptoms with ordinary physical activity; Class III, symptoms occurred with less than daily activity; and Class IV, symptoms at rest [Citation11,Citation12].
12-lead Electrocardiography (ECG) was performed to determine the rhythm, QRS duration, and QTc interval. A QTc interval ≥460 ms was considered prolonged QTc [Citation13].
Standard transthoracic echocardiographic examinations, including two-dimensional imaging, M-mode, and Doppler analysis, were conducted following the American Society of Echocardiography guidelines [Citation14]. LVEF was primarily measured using Simpson’s biplane. Teicholz was used as an alternative method to calculate the LVEF in the presence of suboptimal images to calculate the Simpson’s Biplane. The left ventricular end-systolic diameter (LVESD), LVEDD, interventricular septal end diastole (IVSD), and tricuspid annular plane systolic excursion (TAPSE) were obtained using M-mode echocardiography. Mitral regurgitation (MR) was analyzed using semi-quantitative methods, including proximal isovelocity surface area (PISA) radius and vena contracta width (VCW), which were classified as mild (VCW <3 mm or PISA radius <3 mm), moderate (VCW 3–7 mm or PISA radius 3–10 mm), or severe (VCW >7 mm or PISA radius >10 mm), according to the American Society of Echocardiography guidelines [Citation15]. Moderate and severe mitral regurgitation were categorized as having significant MR.
Study outcomes
All patients were reevaluated by echocardiography at six months and one year after PPCM diagnosis; LVEF <50% was regarded as poor LV function recovery [Citation4,Citation16]. Patients who died due to PPCM before one year yet their LVEF remained <50% during their last follow-up period were also classified into a non-recovery group.
Statistical analysis
All statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). The Kolmogorov-Smirnov test was used to evaluate the data distribution. Categorical variables are presented as numbers and percentages. Numerical variables are depicted as mean ± SD if the data distribution was normal, or median (interquartile range) was used if the data distribution was not normal. Chi-square and Fisher’s tests were used to assess the association between categorical variables and outcomes. The association between numerical variables and outcomes was analyzed using an independent t-test for normal data distribution, and the Wilcoxon test was used if the data distribution was not normal.
The scoring system of poor LV recovery in PPCM was designed using the univariate and multivariate analysis of the baseline characteristics of our cohort. These analyses were performed using logistic regression analysis with a stepwise backward method. All variables with a p-value <0.25 in univariate analysis were included in multivariate analysis to discover independent factors of related outcomes along with the coefficient (B) and standard error (SE) of each factor. Statistical significance was set at p-value <0.05. Each independent factor’s score was calculated using the formula (B/SE). Receiver operating characteristic (ROC) analysis was used to estimate the area under the curve (AUC) for predicting the risk of related outcomes. The de Long approach was used to select the best cutoff in terms of sensitivity and specificity.
Results
Baseline characteristics
Initially, there were a total of 134 PPCM patients admitted to Dr. Hasan Sadikin General Hospital Bandung. We excluded one patient who died after one year of follow-up due to renal disease, and 20 patients were lost to follow-up. A total of 113 patients were enrolled in the study. The mean age of all participants was 30 ± 6.5 years and the mean LVEF was 33 ± 6%. The flow of the participant selection process is shown in .
Figure 1. Study flowchart. LVEF: left ventricular ejection fraction; PPCM: peripartum cardiomyopathy.
As shown in , baseline characteristics were analyzed according to the two patient groups based on their LV function recovery (n = 84) and non-recovery (n = 29). Age, comorbidities, obstetric status, breastfeeding history, PPCM onset, BMI, heart rate, systolic blood pressure, diastolic blood pressure, ECG markers, and medications did not differ significantly across the groups (p > 0.05). NYHA FC IV was more prevalent in the non-recovery group (p = 0.015). In terms of echocardiographic variables, the non-recovery group had a lower mean LVEF than the recovery group (30 ± 6% vs. 34 ± 6%, p = 0.003). Accordingly, patients in the non-recovery group had a higher risk of having severely reduced LVEF (59% vs 31%, p = 0.008). The non-recovery group exhibited larger LVEDD (60 ± 4 mm vs 54 ± 8 mm, p < 0.001), LVESD (52 ± 5 mm vs. 45 ± 8 mm, p < 0.001), LA (41 ± 6 mm vs. 38 ± 6mm, p = 0.028), a higher incidence of LV dilatation (89% vs. 43%, p < 0.001), LVESD >42 mm (96% vs. 75%, p < 0.001), and LA dilatation (66% vs. 40%, p = 0.02) compared with the recovery group. Significant MR was more frequent in the non-recovery group than in the recovery group (38% vs. 7%, p = 0.03). However, reduced RV function was comparable between the two groups (19% vs. 28%, p = 0.332).
Table 1. Baseline characteristics of study participants.
Predictors of poor LV function recovery in PPCM
shows univariate and multivariate analyses of poor LV function recovery predictors in PPCM. Univariate analysis showed that NYHA FC IV (OR = 3.32 [95% CI 1.23–8.99]; p = 0.018), LVEF <30% (OR = 3.16 [95% CI 1.32–7.56]; p = 0.01), LVEDD ≥56 mm (OR = 11.11 [95% CI 3.11–39.69]; p < 0.001), LVESD >42 mm (OR 9.15 [95% CI 1.17–71.49]; p = 0.035), LA >40 mm (OR = 2.8 [95% CI 1.16–6.74]; p = 0.02) and significant MR (OR = 7.94 [95% CI = 2.60–24.32]; p < 0.001) were significantly associated with a higher risk of poor LV function recovery within 12 months following PPCM diagnosis.
Table 2. Univariate and multivariate analysis of predictors of poor LV function recovery following PPCM diagnosis.
Several possible confounders were adjusted in the multivariate analysis, including obesity, preeclampsia, heart rate, systolic blood pressure, and prolonged QTc. This finding revealed that NYHA FC IV (OR = 17 [95% CI 2.92–98.81]; p < 0.01), LVEF <30% (OR = 3.54 [95% CI 1.06–11.82]; p < 0.05), LVEDD ≥56 mm (OR = 7.24 [95% CI 1.73–30.27]; p < 0.01), and significant MR (OR = 17.12 [95% CI 2.75–106.5]; p < 0.01) were significantly associated with poor LV function recovery within 12 months of follow-up.
Scoring system for prediction of poor left ventricular function recovery in PPCM
All independent predictors were enrolled in the scoring system with a score of 3 for significant MR, LVEDD ≥56 mm, and NYHA FC IV and a score of 2 for LVEF <30% (). The overall PPCM recovery score was calculated by summing the combined scores (0–11 points). The probabilities of PPCM recovery scores are presented in . Accordingly, in the present study, we discovered the higher the PPCM recovery scores, the higher percentage of non- recovered PPCM patients [0 points, 0% (0/14); 2 points, 0% (0/5); 3 points, 10% (3/32); 5 points, 17% (3/18); 6 points, 27% (6/22); 8 points, 54% (7/13); 9 points, 100% (2/2); 11 points, 100% (7/7). The cutoff value for PPCM recovery score was ≥8 in ROC analysis (AUC, 0.85 [95% CI 0.77–0.94]; p < 0.001; sensitivity: 57%; specificity: 93%) (). In this study, there were 22 and 91 patients with scores of ≥8 and <8, respectively. In addition, there were 16 non-recovered patients who had scores ≥8 (57%). Patients with PPCM recovery score ≥8 had an almost eighteen times greater risk of poor LV function recovery in distinctive to those with PPCM recovery score <8 (OR = 18.4 [95% CI 6.06–55.97]; p < 0.001).
Figure 2. The probabilities of non-recovery LV function within one year following PPCM diagnosis by PPCM recovery scoring system. LV: left ventricle; LVEDD: left ventricular end diastolic diameter; LVEF: left ventricular ejection fraction; MR: mitral regurgitation; NYHA FC IV: New York Heart Association Functional Class IV; PPCM: peripartum cardiomyopathy; PPCM Recovery score: padjadjaran peripartum cardiomyopathy recovery score.
Figure 3. The area under the curve of PPCM recovery score. AUC: Area under the curve; PPCM Recovery score: padjadjaran peripartum cardiomyopathy recovery score.
Table 3. Scoring system of predictors of poor left ventricular function in PPCM.
Discussion
To the best of our knowledge, this is the first prospective cohort study that develops a scoring system to predict poor LV recovery within one year of follow-up in patients with PPCM. Additionally, this is the first PPCM cohort study to originate in Southeast Asia. The primary findings of this study are as follows: 1. PPCM Recovery score comprised of four independent predictors of poor LV function recovery in PPCM patients: NYHA FC IV at the time of the initial PPCM presentation, baseline echocardiography of LVEF <30%, LVEDd ≥56 mm, and significant MR, 2. The AUC of the scoring system demonstrated high discriminatory ability (AUC of 0.85), and by using a cutoff of ≥8, the scoring system predicted poor LV function recovery with a sensitivity of 57% and a specificity of 93%.
LV recovery due to PPCM is more often achieved compared to other types of cardiomyopathy, with a variety of LV recovery rates ranging from 24 to 72% [Citation10,Citation17]. Our study discovered that non-recovery of LV function was noted in 26% of PPCM patients within one year of follow-up, which is comparable to the non-recovery observed in 28% of PPCM patients in the IPAC study [Citation10].
LVEF reflects the LV systolic function, in which lower LVEF was linearly associated with a lower LV contractility capacity [Citation18]. In our study, the non-recovery group had significantly lower baseline LVEF than the recovery group. Similarly, a meta-analysis by Hosseinpour et al. demonstrated that the baseline LVEF in the non-recovery group was significantly lower than that in the recovery group [Citation19]. However, instead of using numerical variables, we employed categorical variables in multivariate analysis by categorizing LVEF into two groups to maximize its relevance in clinical practice. Our findings were consistent with the multivariate analysis of a retrospective study by Goland et al. and an IPAC study, which found that PPCM patients with baseline LVEF <30% were nearly four times more likely to have poor LV function recovery [Citation4,Citation9,Citation10,Citation16].
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 [Citation2]. Consistent with our results, a meta-analysis by Hosseinpour et al. revealed that the recovery group had a substantially lower baseline LVEDD value than the non-recovery group [Citation19]. Similar to LVEF, the multivariate analysis in our study was performed by including LV dilatation as a categorical variable of LVEDD. In accordance with multivariate analysis of a retrospective study by Goland et al. and Amos et al., the present study reported that LVEDD ≥56 mm had an approximately seven times greater risk of poor LV function recovery in PPCM patients [Citation16,Citation20].
LV dilatation can induce mitral annular dilatation and eventually lead to functional MR. In our study, significant MR independently increased the risk of poor LV function recovery by approximately seventeen times higher compared to those who did not have significant MR. Duran et al. found that MR was not significantly associated with poor LV function recovery; however, severe MR was more prevalent in deceased PPCM patients [Citation17]. Similar to LVEF and LV dilatation, multivariate analysis in our study was performed by including MR as a categorical variable of significant MR and mild MR [Citation15]. Our study found that PPCM patients with significant MR had a 17-fold increased probability of poor LV function recovery. In the present study, we classified MR according to PISA Radius and VCW because these methods were deemed less subjective, independent of hemodynamic status, and highly apparent in distinguishing mild and significant MR [Citation15].
The last component of this scoring system was NYHA FC IV, which was independently linked with a 17-fold higher risk of poor LV function recovery within one year after PPCM diagnosis. NYHA proposes a basic functional categorization of patients with HF, which has been used clinically for over a century as a core method for HF risk stratification. Our findings suggested that the more severe functional capacity leads to higher rates of non-recovery in patients with PPCM. Consistent with the results of Hoevelmann et al., our findings showed that patients with non-recovered PPCM were more likely to present with poorer NYHA FC [Citation21].
This study presented that preeclampsia and prescribed medications to PPCM patients may not be regarded as independent predictors of poor LV recovery in PPCM. Preeclampsia has been identified as one of the strong risk factors for the development of PPCM [Citation22]. Despite the postulation that preeclampsia and PPCM have a common pathophysiology [Citation22], our results supported the meta-analysis which showed preeclampsia fails to become an independent predictor of LV recovery in PPCM [Citation19]. In terms of therapy, both recovery and non-recovery groups in PPCM patients obtained similar heart failure medications, including angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, beta-blocker, mineralocorticoid receptor antagonist, and diuretics. In our cohort, however, it is difficult to conclude whether these drugs, despite having anti-remodeling and neurohumoral modulation, could play a role in improving LV recovery in PPCM patients. Along with heart failure therapies, bromocriptine has been considered beneficial in reducing mortality in severe PPCM patients [Citation2]. In this study, only a few patients received this drug because it was not covered by national health coverage in Indonesia; thus, it was only prescribed for PPCM patients with severely poor LV function (LVEF ≤35%) who could afford the drug. Due to the inadequate number of subjects, we hardly can conclude whether bromocriptine utterly contributes to LV recovery or not in PPCM.
Since an echocardiogram is a noninvasive tool that is available in most health care facilities and NYHA FC is the most utilized clinical parameter in HF patients, clinicians can easily and quickly use PPCM recovery score to evaluate the prognosis of PPCM patients at first medical contact. Prior research has shown that non-recovered PPCM patients are more likely to experience major adverse cardiac events, such as thromboembolic events, pulmonary edema, arrhythmias, the need for implantable cardioverter-defibrillators implantation, or cardiac transplantation [Citation4,Citation17,Citation20,Citation23,Citation24]. Despite its low sensitivity, the PPCM recovery score is highly specific in identifying patients who are more likely to recover poorly at the time of PPCM diagnosis. Thus, this scoring system may determine which patients should be immediately referred to heart failure specialists. This should be a main priority since they may require more extensive therapy and closer monitoring to achieve a better outcome.
Limitations
This study had several limitations. First, although being conducted at a referral centre center for PPCM patients in West Java, Indonesia, this single-center study cannot explain the disparities in racial outcomes, which are likely to be multifactorial, including potential genetic, environmental, and possible health-care system differences. Therefore, the application of this scoring system in other race needs to be interpreted with caution. Second, biomarkers are not widely available in every medical center. This scoring system, however, highlighted the ease of implementation within the vast majority of health care services; therefore, the present study did not include cardiac biomarkers that have previously been recognized as the parameters of LV function recovery in PPCM patients, including B-type natriuretic peptide (BNP) and N-terminal (NT)-proBNP [Citation4,Citation21]. Likewise, genetic sequencing analysis is limited in Indonesia; thus, our study was not able to present which genotypes are more susceptible to poor LV recovery in PPCM patients. We acknowledged that the disproportionate and small numbers of non-recovery cases might limit this research’s statistical power. As an observational cohort study of a relatively rare disease, the same patient cohort was used to define predictors of poor LV recovery and to evaluate the PPCM recovery score. Therefore, we encourage the validation of our findings in a larger multi-center cohort to expand the usefulness and applicability of this scoring system for patients with PPCM worldwide.
Conclusions
This study yielded several clinical independent predictors of poor LV recovery within one year of follow-up in PPCM patients, including NYHA FC IV, LVEF <30%, LVEDD ≥56 mm, and significant MR at first PPCM diagnosis. Using these independent parameters, the PPCM recovery score could indicate the risk of poor LV recovery, with a cutoff score of ≥8 as a predictor of non-recovery and AUC of 0.85. The score is potentially helpful in determining which PPCM patients should be immediately transferred to a tertiary care center for more extensive therapy by heart failure specialists to improve LV recovery.
Ethics approval and consent to participate
This study was approved by the Medical Research Ethics Committee of Dr. Hasan Sadikin General Hospital, Bandung, Indonesia according to the principles outlined in the Declaration of Helsinki. Written informed consent was acquired from all subjects.
Acknowledgements
Prof. Karen Sliwa for her valuable input as a PPCM expert. We would also like to thank the European Society of Cardiology/EurObservational Research Programme (ESC/EORP) and the chairs of the PPCM registry (Prof. Karen Sliwa and Prof. Johann Bauersachs) for using the PPCM registry CRF for collecting our data of PPCM patients.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
The data used and analyzed during the current study were available from the prospective PPCM study at Dr. Hasan Sadikin General Hospital, Bandung, Indonesia.
Additional information
Funding
References
- Ricke-Hoch M, Pfeffer TJ, Hilfiker-Kleiner D. Peripartum cardiomyopathy: basic mechanisms and hope for new therapies. Cardiovasc Res. 2020;116(3):520–531. doi:10.1093/cvr/cvz252.
- Bauersachs J, König T, van der Meer P, et al. Pathophysiology, diagnosis and management of peripartum cardiomyopathy: a position statement from the heart failure association of the European Society of cardiology study group on peripartum cardiomyopathy. Eur J Heart Fail. 2019;21(7):827–843. doi:10.1002/ejhf.1493.
- Sliwa K, Petrie MC, Hilfiker‐Kleiner D, et al. Long‐term prognosis, subsequent pregnancy, contraception and overall management of peripartum cardiomyopathy: practical guidance paper from the heart failure association of the European society of cardiology study group on peripartum cardiomyopathy. Eur J Heart Fail. 2018;20(6):951–962. doi:10.1002/ejhf.1178.
- Sliwa K. Peripartum cardiomyopathy: from pathophysiology to management. Elsevier Science; 2021. doi:10.1016/C2018-0-02473-1.
- Bhattacharyya A, Basra SS, Sen P, et al. Peripartum cardiomyopathy: a review. Texas Heart Inst J. 2012;39(1):8.
- Biteker M, Ilhan E, Biteker G, et al. Delayed recovery in peripartum cardiomyopathy: an indication for long‐term follow‐up and sustained therapy. Eur J Heart Fail. 2012;14(8):895–901. doi:10.1093/eurjhf/hfs070.
- Pearson GD, Veille J-C, Rahimtoola S, 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.
- Elkayam U, Akhter MW, Singh H, 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.
- Sliwa K, Bauersachs J, Arany Z, et al. Peripartum cardiomyopathy: from genetics to management. Eur Heart J. 2021;42(32):3094–3102. doi:10.1093/eurheartj/ehab458.
- McNamara DM, Elkayam U, Alharethi R, 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.
- Dolgin M, Association NYH, Fox AC, , et al. Nomenclature and criteria for diagnosis of diseases of the heart and great vessels. Boston, MA: Lippincott Williams and Wilkins; 1994.
- McDonagh TA, Metra M, Adamo M, et al. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: developed by the task force for the diagnosis and treatment of acute and chronic heart failure of the European society of cardiology (ESC) with the special contribution of the heart failure association (HFA) of the ESC. Eur Heart J. 2021;42(36):3599–3726. doi:10.1093/eurheartj/ehab368.
- Khatib R, Sabir FR, Omari C, et al. Managing drug-induced QT prolongation in clinical practice. Postgrad Med J. 2021;97(1149):452–458. doi:10.1136/postgradmedj-2020-138661.
- Lang RM, Badano LP, Mor-Avi V, 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. Eur Heart J Cardiovasc Imaging. 2015;16(3):233–270. doi:10.1093/ehjci/jev014.
- Zoghbi WA, Adams D, Bonow RO, et al. Recommendations for noninvasive evaluation of native valvular regurgitation: a report from the American society of echocardiography developed in collaboration with the society for cardiovascular magnetic resonance. J Am Soc Echocardiogr. 2017;30(4):303–371. doi:10.1016/j.echo.2017.01.007.
- Goland S, Bitar F, Modi K, et al. Evaluation of the clinical relevance of baseline left ventricular ejection fraction as a predictor of recovery or persistence of severe dysfunction in women in the United States with peripartum cardiomyopathy. J Card Fail. 2011;17(5):426–430. doi:10.1016/j.cardfail.2011.01.007.
- Duran N, Günes H, Duran I, et al. Predictors of prognosis in patients with peripartum cardiomyopathy. Int J Gynaecol Obstet. 2008;101(2):137–140. doi:10.1016/j.ijgo.2007.11.007.
- Zipes DP, Libby P, Bonow RO, et al. Braunwald’s heart disease e-book: a textbook of cardiovascular medicine. Elsevier Health Sciences; 2018. https://books.google.co.id/books?id=LwBGDwAAQBAJ.
- Hosseinpour A, Hosseinpour H, Kheshti F, et al. Prognostic value of various markers in recovery from peripartum cardiomyopathy: a systematic review and meta‐analysis. ESC Heart Fail. 2022;9(5):3483–3495. doi:10.1002/ehf2.14085.
- Amos AM, Jaber WA, Russell SD. Improved outcomes in peripartum cardiomyopathy with contemporary. Am Heart J. 2006;152(3):509–513. doi:10.1016/j.ahj.2006.02.008.
- Hoevelmann J, Muller E, Azibani F, et al. Prognostic value of NT-proBNP for myocardial recovery in peripartum cardiomyopathy (PPCM). Clin Res Cardiol. 2021;110(8):1259–1269. doi:10.1007/s00392-021-01808-z.
- Bello N, Rendon ISH, Arany Z. The relationship between pre-eclampsia and peripartum cardiomyopathy: a systematic review and meta-analysis. J Am Coll Cardiol. 2013;62(18):1715–1723. doi:10.1016/j.jacc.2013.08.717.
- Ekizler FA, Cay S. A novel marker of persistent left ventricular systolic dysfunction in patients with peripartum cardiomyopathy: monocyte count-to-HDL cholesterol ratio. BMC Cardiovasc Disord. 2019;19(1):114. doi:10.1186/s12872-019-1100-9.
- Li W, Li H, Long Y. Clinical characteristics and long-term predictors of persistent left ventricular systolic dysfunction in peripartum cardiomyopathy. Can J Cardiol. 2016;32(3):362–368. doi:10.1016/j.cjca.2015.07.733.