3.3 One-year outcome
Primary outcomes of cardiac death occurred in 31 patients, all-cause mortality in 72 patients, and MACE in 65 patients within 1 year. There were 4 patients in Traditional group and 1 patient in Late group who underwent cardiac transplantation. Overall, patients in Late group had worse prognosis compared with pregnancy-associated HF in Early group and previously defined PPCM in Traditional group (Fig. 3, lower panel), with higher rate of cardiac death (9.5% vs 1.1% and 2.8%), all-cause mortality (15.8% vs 3.4% and 7.3%), and MACE (15.8% vs 6.8% and 5.9%) (Table 3).
Cumulative incidence plots showed significant differences among 3 groups for all primary outcomes: cardiovascular death (P = .001), all-cause mortality (P = .003), and MACE (P = .001). In terms of cardiac death, patients in Late group had significantly worse prognosis compared with Early group (P = .012) and Traditional group (P < .001). In terms of all-cause mortality, patients in Late group also had significantly worse prognosis compared with Early group (P = .005) and Traditional group (P = .004). In terms of MACE, patients in Late group again had significantly worse prognosis compared with Early group (P = .047) and Traditional group (P < .001). There was no difference in cumulative events between Early and Traditional groups for all primary outcomes (Fig. 4).
In multivariate Cox proportional hazards models, clinical variables of age, previous delivery, delivery type at PPCM, multiparity type at PPCM, preeclampsia/eclampsia or hypertension, and gestational diabetes or DM were adjusted (Table 3). With regard to cardiac death, patients in Late group had significantly worse outcome compared with Early group (hazards ratio [HR] = 9.09, CI: 1.11–74.12, P = .039) and Traditional group (HR = 3.21, CI: 1.43–7.19, P = .005). With regard to all-cause mortality, patients in Late group also had significantly worse outcome compared with Early group (HR = 5.31, CI: 1.50–18.78, P = .010) and Traditional group (HR = 2.24, CI: 1.24–4.07, P = .008). With regard to MACE, patients in Late group again had significantly worse outcome compared with Early group (HR = 2.70, CI: 1.01–7.19, P = .048) and Traditional group (HR = 2.75, CI: 1.50–5.03, P = .001). There was no difference between Early and Traditional groups with regard to cardiac death, all-cause mortality, or MACE.
Our study had several findings. This is the largest national cohort study of PPCM in Asia. The incidence of PPCM in Taiwan in recent years was comparable to Asian Americans in the United States, suggesting genetic underpinnings have an important role in PPCM. A majority of the PPCM occurred within the first month following delivery, suggesting peripartum stress is the most likely underlying mechanism for causing PPCM. A new category, the Late group, defined as PPCM diagnosed sixth to twelfth months post-delivery, showed significantly worse clinical outcome compared with both Early group and Traditional group.
4.1 Epidemiology of PPCM
Currently, published studies of PPCM were from South Africa, Haiti, Brazil, Germany, Japan, and United States with prevalence from 1:300 in Haiti to 1:20,000 in Japan[9–11,19–23] and data in other countries have been lacking. Our study used nationwide health insurance program in Taiwan, allowing the study of information on incidence, maternal cardiovascular and delivery history, associations, interventions, medications, and 1-year outcome of PPCM without selection and participation biases. The epidemiology of PPCM in Taiwan during the 15-year was 1 in 3,790, which was higher than previously reported incidence in Japan but similar to Asian Americans with 1 in 2,675.
Most previous studies enrolled patients with pregnancy-associated HF from last trimester prior to delivery till 5 months post-delivery. Our study is unique in that patients with myocardial failures occurred from early pregnancy till extended months post-delivery were included for analysis, and there were quite a number of patients in the Early and Late groups with unexplained HF. With most of patients diagnosed of PPCM in the first month following delivery at 65.9%, peripartum stress due to altered physiological conditions was the most important cause of the condition.
4.2 Associated conditions, treatment, and outcome of PPCM
From the clinical characteristics of study patients, a higher incidence of previous delivery was found in Late group (44.2%) compared with Early (29.5%) and Traditional (31.9%) groups. This suggests previous births may delay the onset, presentation, and diagnosis of PPCM. There was also higher percentages of vaginal delivery in Late group (52.6%), compared with Early (20.5%) and Traditional (30.3%) groups, suggesting higher stability during delivery and possibly more indolent presentation of the PPCM in the Late group. On the other hand, a higher rate of Cesarean sections higher in Early (72.7%) and Traditional (67.9%) groups compared with the Late group (45.3%) may suggest increased maternal risks requiring immediate delivery. Risk factors speculated for the development of PPCM are advanced maternal age, high number of parity, high number of gravidity, twin pregnancy, use of tocolytic therapy, African descent, non-Caucasian ethnicity, and poverty. Multiparity has been traditionally considered a risk factor for PPCM, however most studies in the United States have reported the development of PPCM in conjunction with the first or second pregnancy in 50% of patients. Previous study noted rate of multiple birth in PPCM was 9%, whereas the rate in average estimate was 3%.[25,26] In our study, the rate of multigestations in Traditional group was 7.4%.
In PPCM, oxidative stress plays a central part in disease pathogenesis. The vasculo-hormonal hypothesis was tested in experimental study, and STAT3 was shown to play a role in cardiomyocyte protection from reactive oxygen species (ROS). Loss of STAT3 in murine model leads to increased ROS, triggering secretion of cathepsin D that in turn cleaves prolactin into 16-kDa fragment promoting cell death in PPCM. Bromocriptine blocks prolactin secretion and was effective in the treatment of PPCM in mice. Prospective observation registry in Germany has shown promise of 4-week bromocriptine therapy on top of standard HF medications including beta-blockers and ACEi/ARB, with higher recovery rate. In our study, bromocriptine was used in 7 patients (0.9%) in Traditional group but none in Early and Late groups.
Patients with PPCM at times present with rapid progression leading to critical LV failure and acute pulmonary edema, requiring use of inotropic agents and mechanical assist device such as IABP and ECMO. Inotropes was used more often in Traditional and Late groups, possibly reflecting the more severely depressed LV function in these patients than Early group. Cardiac rehabilitation has been shown to improve clinical status, HF readmission, and outcome. This exercise training was prescribed with higher rates in Early and Late groups compared with Traditional group, reflecting the higher exercise capacity in the early pregnancy or extended months post-delivery.
A national inpatient database revealed in-hospital mortality in United States during 2004 to 2011 was 1.3% for patients with PPCM. We reported an overall in-hospital death of 9.2% in these women with PPCM, and 1.0%, 10.6%, and 6.9% for Early, Traditional, and Late groups, respectively. Recently, IPAC study reported that the mortality rate was 4% in the 100 women with PPCM that were followed up 1-year post-partum. In our 1-year follow-up, cardiac death was found in 1.1%, 2.8%, and 9.5% whereas all-cause mortality was 3.4%, 7.3%, and 15.8% in Early, Traditional, and Late groups, respectively.
4.3 The late group
In the study by Elkayam et al, the authors noted classic criteria for the diagnosis of PPCM as established by Demakis et al[1,2] limited the diagnosis to the last gestational month and first 5 months after delivery. However, several reports published later described women presented with cardiomyopathy earlier in the pregnancy.[30–34] In addition, although PPCM is usually diagnosed within the first 5 months postpartum, it is often missed or delayed because most of the signs and symptoms of normal pregnancy are similar to those of HF. In addition, pregnancy-associated cardiovascular death can occur up to 1 year following delivery. Furthermore, PPCM and pregnancy-associated cardiomyopathy had been described to be part of the same clinical spectrum.
We included the patients with HF occurring >5 months post-delivery in the Late group since hormonal imbalance in postpartum women can persist as long as 12 months after delivery with some experts considering postpartum care necessary up to 1 year after giving birth.[13,14] Importantly, there were noticeable differences in outcomes of PPCM between the study groups. Both cumulative incidence and multivariate Cox proportional hazards model showed significantly worse prognosis in all primary outcomes in Late group, whereas no difference existed between Traditional and Early group. The 1-year rate of cardiac death, all-cause mortality and MACE in Late group was more than 2 to 3 times compared with Early and Traditional groups.
As stated earlier, the genetic evidence in the development of PPCM was demonstrated in the animal model with loss of STAT3, leading to increased ROS, cleaving of prolactin into 16-kDa fragment, and eventual cardiomyocyte apoptosis. Through negative-feedback loop, prolactin secretion is typically regulated and inhibited by dopamine. With secretion of placental lactogen, relatively low levels of prolactin is maintained during early and mid-pregnancy. In combination with reduction of dopamine secretion and insensitivity to the negative feedback mechanism, the effects of placental lactogen is overcome with the large nocturnal surge of prolactin prior to parturition. The prolactin level then remains elevated while breastfeeding continues during postpartum.[37,38] This period typically lasts 1 year as guideline recommendations and may coincide with the late onset of PPCM. On the other hand, a study also reported that breastfeeding had no adverse effects to the mother in 67% of patients with PPCM, and instead was associated with recovery of LV systolic function.
Hypertensive pregnancy disorders complicated 5% to 7% of all pregnancies, and been linked to later cardiovascular events. Preeclampsia has also been shown to be associated with persistent postpartum cardiovascular impairment and abnormal LV function in prospective longitudinal case-control study. Our study patients had a higher percentage of preeclampsia/eclampsia or hypertension in Traditional group (25.5%) and Late group (25.2%) compared with Early group (18.2%). The combination of aforementioned higher incidence of prior births, genetic mutation, postpartum hormonal imbalance, especially elevated prolactin level, and preeclampsia/eclampsia or hypertension in the Late group may explain the worst outcome among all patients. In summary, our findings showed that late presentation and diagnosis of PPCM had distinctly higher cardiac death, all-cause mortality, and MACE.
There are several limitations in epidemiologic data from NHIRD. First, using ICD-9-CM codes for patient screening may miss some cases for conditions not coded correctly. Second, the main criteria used in diagnosis of PPCM using LV ejection fraction was not available. However as mentioned in the Methods section, the diagnosis of HF by NHIRD has high accuracy against the gold standard EMR. Third, the diagnosis of HF associated with pregnancy, delivery, and postpartum period required the patients to have delivery at hospital inpatient services. In rare situations patients may still give birth at clinics not using ICD codes. Last, since our study consisted of uniform ethnic background, and application of the results to other populations awaits further studies.
Our study of PPCM was the largest nationwide population-based cohort in Asia that showed timing of diagnosis of PPCM had different outcomes. Late group of patients with PPCM had significantly worse outcome compared with both Early and Traditional groups, even after adjusted for clinical variables.
We would like to thank Hsing-Fen Lin, BA, for the statistical assistance during the completion of this manuscript.
. Demakis JG, Rahimotoola SH. Peripartum cardiomyopathy
. Circulation 1971;44:964–8.
. Demakis JG, Rahimtoola SH, Sutton GC, et al. Natural course of peripartum cardiomyopathy
. Circulation 1971;44:1053–61.
. Manolio TA, Baughman KL, Rodeheffer R, et al. Prevalence and etiology of idiopathic dilated cardiomyopathy (summary of a National Heart, Lung, and Blood Institute workshop. Am J Cardiol 1992;69:1458–66.
. Hibbard JU, Lindheimer M, Lang RM. A modified definition for peripartum cardiomyopathy
and prognosis based on echocardiography. Obstet Gynecol 1999;94:311–6.
. Silwa K, Hilfiker-Kleiner D, Petrie MC, et al. Heart Failure Association of the European Society of Cardiology Working Group on Peripartum Cardiomyopathy
. 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;12:767–78.
. Elkayam U. Clinical characteristics of peripartum cardiomyopathy
in the United States: diagnosis, prognosis, and management. J Am Coll Cardiol 2011;58:659–70.
. Hilfiker-Kleiner D, Sliwa K. Pathophysiology and epidemiology
of peripartum cardiomyopathy
. Nat Rev Cardiol 2014;11:364–70.
. Lu CH, Lee WC, Wu M, et al. Comparison of clinical outcomes in peripartum cardiomyopathy
and age-matched dilated cardiomyopathy: A 15-year nationwide population-based study in Asia. Medicine 2017;96:e6898.
. Fett JD, Christie LG, Carraway RD, et al. Unrecognized peripartum cardiomyopathy
in Haitian women. Int J Gynaecol Obstet 2005;90:161–6.
. Mielniczuk LM, Williams K, Davis DR, et al. Frequency of peripartum cardiomyopathy
. Am J Cardiol 2006;97:1765–8.
. Brar SS, Khan SS, Sandhu GK, et al. Incidence, mortality, and racial differences in peripartum cardiomyopathy
. Am J Cardiol 2007;100:302–4.
. Sliwa K, Böhm M. Incidence and prevalence of pregnancy-related heart disease. Cardiovasc Res 2014;101:554–60.
. Sliwa K, Hilfiker-Kleiner D, Mebazaa A, et al. EURObservational Research Programme: a worldwide registry on peripartum cardiomyopathy
(PPCM) in conjunction with the Heart Failure Association of the European Society of Cardiology Working Group on PPCM. Eur J Heart Fail 2014;16:583–91.
. Gjerdingen DK, Froberg DG, Chaloner KM, et al. Changes in women's physical health during the first postpartum year. Arch Fam Med 1993;2:277–83.
. Banti S, Mauri M, Oppo A, et al. From the third month of pregnancy to 1 year postpartum. Prevalence, incidence, recurrence, and new onset of depression. Results from the perinatal depression-research & screening unit study. Compr Psychiatry 2011;52:343–51.
. Hameeed AB, Lawton ES, McCain CL, et al. Pregnancy-related cardiovascular deaths in California: beyond peripartum cardiomyopathy
. Am J Obstet Gynecol 2015;213:379.e1-10.
. Hsing AW, Ioannidis JP. Nationwide population science: lessons from the Taiwan National Health Insurance Research Database. JAMA Intern Med 2015;175:1527–9.
. Cheng CL, Chien HC, Lee CH, et al. Validity of in-hospital mortality data among patients with acute myocardial infarction or stroke in National Health Insurance Research Database in Taiwan. Int J Cardiol 2015;201:96–101.
. Haghikia A, Podewski E, Libhaber E, et al. Phenotyping and outcome
on contemporary management in a German cohort of patients with peripartum cardiomyopathy
. Basic Res Cardiol 2013;108:366.
. Kamiya CA, Kitakaze M, Ishibashi-Ueda H, et al. Different characteristics of peripartum cardiomyopathy
between patients complicated with and without hypertensive disorders. Results from the Japanese Nationwide survey of peripartum cardiomyopathy
. Circ J 2011;75:1975–81.
. Desai D, Moodley J, Naidoo D. Peripartum cardiomyopathy
: experiences at King Edward VIII Hospital, Durban, South Africa and a review of the literature. Trop Doct 1995;25:118–23.
. Carvalho A, Brandao A, Martinez EE, et al. Prognosis in peripartum cardiomyopathy
. Am J Cardiol 1989;64:540–2.
. Kolte D, Khera S, Aronow WS, et al. Temporal trends in incidence and outcomes of peripartum cardiomyopathy
in the United States: a nationwide population-based study. J Am Heart Assoc 2014;3:e001056.
. Ntusi NB, Mayosi BM. Aetiology and risk factors of peripartum cardiomyopathy
: a systemic review. Int J Cardiol 2009;131:168–79.
. Bello N, Rendon IS, Arany Z. The relationship between pre-eclampsia and peripartum cardiomyopathy
: a systemic review and meta-analysis. J Am Coll Cardiol 2013;62:1715–23.
. Russell RB, Petrini JR, Damus K, et al. The changing epidemiology
of multiple births in the United States. Obstet Gynecol 2003;101:129–35.
. Hilfiker-Kleiner D, Kaminski K, Podewski E, et al. A cathepsin D-cleaved 16 kDa form of prolactin mediates postpartum cardiomyopathy. Cell 2007;128:589–600.
. McNamara DM, Elkayam U, Alharethi R, et al. Clinical outcomes for peripartum cardiomyopathy
in North America: results of the IPAC study (investigators of pregnancy-associated cardiomyopathy). J Am Coll Cardiol 2015;66:905–14.
. Elkayam U, Akhter MW, Singh H, et al. Pregnancy-associated cardiomyopathy: clinical characteristics and a comparison between early and late presentation. Circulation 2005;111:2050–5.
. Brown G, O’Leary M, Douglas I, et al. Perioperative management of a case of severe peripartum cardiomyopathy
. Anaesth Intensive Care 1992;20:80–3.
. Forssell G, laska J, Olofsson C, et al. Peripartum cardiomyopathy
: three cases. J Intern Med 1995;235:493–6.
. Yahagi N, Kumon K, Nakatani T, et al. Peripartum cardiomyopathy
and tachycardia followed by multiple organ failure. Anesth Analg 1994;79:581–2.
. Rizeq MN, Rickenbacher PR, Fowler MB, et al. Incidence of myocarditis in peripartum cardiomyopathy
. Am J Cardiol 1994;74:474–7.
. Abboud J, Murad Y, Chen-Scarabelli C, et al. Peripartum cardiomyopathy
: a comprehensive review. Int J Cardiol 2007;118:295–303.
. Biteker M, Kayatas K, Duman D, et al. Peripartum cardiomyopathy
: current state of knowledge, new developments and future directions. Curr Cardiol Rev 2014;10:317–26.
. Grattan DR, Steyn FJ, Kokay IC, et al. Pregnancy-induced adaptation in the neuroendocrine control of prolactin secretion. J Neuroendocrinol 2008;20:497–507.
. Noel GL, Suh HK, Frantz AG. Prolactin release during nursing and breast stimulation in postpartum and nonpostpartum subjects. J Clin Endocrinol Metab 1974;38:413–23.
. Bollen IA, Van Deel ED, Kuster DW, et al. Peripartum cardiomyopathy
and dilated cardiomyopathy: different at heart. Front Physiol 2015;5:531.
. Johnston M, Landers S, Nobel L, et al. Breastfeeding and the use of human milk. Pediatrics 2012;129:e827–41.
. Safirstein JG, Ro AS, Grandhi S, et al. Predictors of left ventricular recovery in a cohort of peripartum cardiomyopathy
patients recruited via the internet. Int J Cardiol 2012;154:27–31.
. Lykke JA, Langhoff-Roos J, Sibai BM, et al. Hypertensive pregnancy disorders and subsequent cardiovascular morbidity and type 2 diabetes in the mother. Hypertension 2009;53:944–51.
. Melchiorre K, Sutherland GR, Liberati M, et al. Preeclampsia is associated with persistent postpartum cardiovascular impairment. Hypertension 2011;58:709–15.
epidemiology; outcome; peripartum cardiomyopathy
Supplemental Digital Content
Copyright © 2017 The Authors. Published by Wolters Kluwer Health, Inc. All rights reserved.