Pregnant women with underlying heart disease (HD) are at increased risk for adverse maternal and fetal outcomes. In this study, we sought to identify the risk and risk factors for adverse maternal and fetal events in pregnant women with underlying HD. Pregnant women referred for echocardiogram with known or suspected HD were categorized into those with (1) cardiomyopathy, (2) other HD (congenital, coronary, arrhythmia, or valvular), and (3) no HD. Primary outcome was major adverse cardiovascular events (MACE), defined as a composite of death, sustained arrhythmia, myocardial infarction, heart failure, and transient ischemic attack/stroke. Secondary outcome was fetal adverse clinical events (FACE), a composite of infant death, prematurity, underweight status, intracranial hemorrhage, and respiratory distress. Of the 173 pregnancies, 37 (21%) had cardiomyopathy, 65 (38%) had other HD, and 68 (39%) had no HD. MACE was higher in pregnancies with cardiomyopathy (p <0.001) because of higher rates of heart failure and cardiac arrest (up to 6 months postpartum, p <0.001 and 0.023, respectively). FACE rates were higher in cardiomyopathy pregnancies (p <0.001). In multivariate analysis, cardiomyopathy (odds ratio [OR] 11.5, 95% confidence interval [CI] 3.7 to 35.4), hypertension (OR 10.69, 95% CI 3.70 to 30.90), and arrhythmia (OR 7.6, 95% CI 2.1 to 27) were independently associated with higher MACE. Cardiomyopathy (OR 2.7, 95% CI 1.1 to 7.0) and hypertension (OR 3.6, 95% CI 1.4 to 9.0) were also independently predictive of higher FACE. In conclusion, pregnant women with cardiomyopathy had higher rates of adverse MACE and FACE rates. Cardiomyopathy, hypertension, and arrhythmia were independently associated with adverse cardiovascular and fetal clinical events, whereas other HD was not.
Little is known about the cardiac morbidity and risk stratification of pregnant patients with cardiomyopathy compared with other forms of heart disease (HD) and associated neonatal outcomes, particularly within the United States. Although based on limited data, existing guidelines generally advise against pregnancy in women with an ejection fraction of ≤30% or those with a history of peripartum cardiomyopathy with persistently abnormal left ventricular systolic function. Counseling women about the safety of pregnancy and the maternal and fetal risks can have a major impact on a woman’s life choices. It is, therefore, important that more objective data are available to appropriately provide prenatal counseling. The purpose of this study was (1) to examine the adverse maternal cardiac, obstetric, and fetal and/or neonatal events in women with significant HD other than cardiomyopathy, cardiomyopathy, and without any HD and (2) to determine the impact of cardiomyopathy on maternal and fetal adverse events during pregnancy relative to other forms of HD and the predictors of adverse maternal cardiovascular, clinical events, and adverse fetal events.
Methods
We examined the outcomes of 173 pregnancies in women with HD who received their obstetric care at the Stony Brook University Medical Center. Sixty-eight consecutive pregnant women referred for echocardiograms but were found to have normal studies without any evidence of HD by echocardiogram or by history were included as a comparison group. All women had echocardiograms performed from November 2004 to June 2012. Five patients with multiple pregnancies were included. Noncardiomyopathy HD (other HD) included patients with other congential, acquired, or arrhythmic diseases including previous corrective cardiac surgery and pulmonary hypertension. Those in whom cardiac arrhythmia was the primary diagnosis must have had symptomatic sustained tachyarrhythmias or bradyarrhythmias requiring treatment before pregnancy. Patients with isolated mitral valve prolapse without significant (≥3+) regurgitation were excluded as these patients are known to not be as high risk with pregnancy. Patients with valvular stenoses were included if it was at least moderate. Left-sided valvular regurgitation and pulmonary regurgitation was considered significant if ≥3+. Patients with isolated tricuspid regurgitation or isolated patent foramen ovale were excluded because of their benign course in pregnancy. Patients with a history of cardiomyopathy of any etiology including peripartum cardiomyopathy with recovery of left ventricular systolic function, hypertrophic, restrictive, or dilated cardiomyopathy were included and grouped together because of the uncommon nature of these disorders in pregnant women. Classification was performed independently by 2 investigators (PKS and CM). Patients who underwent therapeutic abortion or miscarriage before 20 weeks were excluded. Approval by the institutional review board was obtained before initiation of the study.
Baseline data were collected through chart review, including maternal age, race, body mass index, marital status, gravida and parity status, hypertension, history of transient ischemic attack or cerebrovascular event, diabetes mellitus, and smoking status. Clinical data were recorded from last clinic visit before delivery or at the start of the hospitalization for delivery. Baseline echocardiographic data included left ventricular ejection fraction and cavity size, left atrial diameter and volume, valvular stenosis and regurgitation grade, and right ventricular systolic pressure. Chamber quantification and valvular assessment were performed according to the American Society of Echocardiography guidelines. The biplane method of discs (modified Simpson’s rule) was used to estimate left ventricular ejection fraction; visual estimation was used if endocardial definition was suboptimal. For patients with multiple echocardiograms during their pregnancy, an antepartum echocardiogram performed closest to the date of delivery was used for analysis.
Adverse maternal events were recorded during the antepartum, peripartum, and postpartum period up to 6 months. The primary outcomes of interest were maternal cardiac and neonatal events. Maternal major adverse cardiovascular events (MACE) were defined as the presence of heart failure (signs and symptoms of dyspnea, pulmonary edema by clinical assessment or chest roentgenography, and/or elevated brain natriuretic peptide >100 pg/ml), myocardial infarction, arrhythmia (significant tachyarrhythmia or bradyarrhythmia documented by electrocardiography and requiring therapy), cardiac arrest, transient ischemic attack or cerebrovascular event, and cardiac death. Maternal adverse clinical events (maternal clinical events) recorded included preterm labor, preterm delivery, intensive care unit stay, and pregnancy-related hypertensive disorders (increase in blood pressure >140/90 after 20-week gestation) including eclampsia and preeclampsia. Hospital length of stay, mode of delivery (vaginal vs cesarean section), induced delivery, urgent/emergent cesarean section, and peripartum blood loss were recorded. Adverse maternal events were limited to inpatient events, only admitted to Stony Brook. Definitions of predictors and outcomes were determined by consensus before collection of the data (PKS, CM, PBP, and KS). Data on outcomes were obtained from independent review of hospital and clinical records by 2 investigators (PKS and CM). Discrepancies in event characterization were resolved by consensus.
Neonatal outcomes were defined as prematurity (birth <37-week gestation), small for gestational age birth weight (birth weight less than the tenth percentile for gestational age or weight <2,500 g), respiratory distress syndrome, neonatal intensive care unit stay, intracranial hemorrhage, low Appearance/complexion, Pulse rate, Reflex irritability grimace, Activity, Respiratory effort (APGAR) scores (<7), and fetal death (defined death after 22 weeks of pregnancy or birth weight <500 g). Neonatal and fetal adverse events were recorded for inpatient only from the birth hospitalization.
Data were summarized by descriptive statistics. Continuous variables were expressed as mean (±SD) and categorical as frequency (%). Univariate analyses were performed to compare demographics, medical history, and clinical maternal and fetal outcomes of patients with cardiomyopathy versus other HD. The chi-square test was used when applicable to compare differences in categorical variables, and Student’s t test was used for continuous variables. Multivariable logistic regression was used to evaluate the influence of cardiomyopathy on MACE while controlling for demographic characteristics and medical history. Predictors for the logistic regression were selected based on statistical significance in the odds ratios by univariate analysis (p <0.2) and included age, race, body mass index, diabetes mellitus, and history of arrhythmia. STATA 12 (StataCorp LP, College Station, Texas) was used for data analysis, and a 2-tailed p value of 0.05 was regarded as statistically significant.
Results
Classification of HD and cardiomyopathy groups are summarized in Table 1 and Figure 1 . Of the 173 pregnancies, 37 (21%) had cardiomyopathy, 65 (38%) other HD, and 68 (40%) had no HD. Forty-three percent of the other HD group had history of congenital HD and 38% had previous corrective cardiac surgery; 26% had a history of arrhythmias, 6% had pulmonary hypertension, and 18% had a history of valvular HD. Of the cardiomyopathy group, 30% had history of dilated cardiomyopathy, 8% had a history of peripartum cardiomyopathy, 5% had hypertrophic cardiomyopathy, and 3% had restrictive cardiomyopathy. Demographic and baseline medical history data are described in Table 2 . Cardiomyopathy patients were significantly older than other HD and no HD groups. Patients with cardiomyopathy were more likely to be black. Maternal weight was significantly different among the 3 groups, with patients with cardiomyopathy having highest weight compared with other HD and no HD. Differences were present in the presence of baseline hypertension (most in cardiomyopathy, fewest in other HD; Table 2 ), and history of arrhythmia was highest in the other HD group. The no HD group had a history of hypertension in 25% of cases.
| Other Heart Disease (n=65) | Cardiomyopathy (n=37) | |
|---|---|---|
| Congenital Heart Disease | 28 (43%) | 1 (3%) |
| Atrial septal defect | 7 (11%) | 0 |
| Coarctation of aorta | 5 (8%) | 0 |
| Congenital pulmonic stenosis | 3 (5%) | 0 |
| Dilated Aortic Root | 3 (5%) | 0 |
| Tetralogy of Fallot | 1 (2%) | 0 |
| Transposition of great arteries | 0 (0%) | 1 (3%) |
| Ventricular septal defect | 10 (15%) | 0 |
| Truncus arteriosus | 1 (2%) | 0 |
| Patent ductus arteriosus | 2 (3%) | 0 |
| Persistent superior vena cava | 1 (2%) | 0 |
| Dilated coronary sinus | 1 (2%) | 0 |
| Prior Corrective Cardiac Surgery | 25 (38%) | 3 (8%) |
| Aortoplasty | 2 (3%) | 0 |
| Aortic valvuloplasty | 2 (3%) | 0 |
| Atrial septal defect repair | 6 (9%) | 0 |
| AV canal defect repair | 2 (3%) | 0 |
| Konno Procedure | 2 (3%) | 0 |
| Mitral Valvuloplasty | 2 (3%) | 0 |
| Mustard procedure | 0 | 1 (3%) |
| Heart transplant | 0 | 1 (3%) |
| Pulmonic valvuloplasty | 4 (6%) | 0 |
| Tetralogy of Fallot repair | 1 (2%) | 0 |
| Ventricular septal defect repair | 8 (12%) | 0 |
| Myomectomy | 0 | 1 (3%) |
| Ischemic Heart Disease | ||
| Prior percutaneous coronary intervention with stent | 1 (2%) | 3 (8%) |
| Prior myocardial infarction | 1 (2%) | 3 (8%) |
| Implantable Cardioverter defibrillator | 0 | 3 (8%) |
| Arrhythmia (By history) | 17 (26%) | 4 (11%) |
| Ventricular tachycardia | 2 (3%) | 1 (3%) |
| Supraventricular tachycardia | 10 (15%) | 1 (3%) |
| Prior cardiac ablation | 4 (6%) | 0 |
| Cardiomyopathy | ||
| History of cardiomyopathy ∗ | 0 | 18 (49%) |
| Dilated cardiomyopathy | 0 | 11 (30%) |
| Restrictive cardiomyopathy | 0 | 1 (3%) |
| Hypertrophic cardiomyopathy | 0 | 2 (5%) |
| Peripartum cardiomyopathy | 0 | 3 (8%) |
| History of heart failure | 0 | 9 (24%) |
| Other | 0 | 4 (11 %) |
| Pulmonary hypertension | 4 (6%) | 1 (3%) |
| Valvular Heart Disease | 12 (18%) | 1 (3%) |
| Bioprosthetic valve | 2 (3%) | 1 (3%) |
| Mechanical valve | 0 | 0 |
| Mitral stenosis | 2 (3%) | 0 |
| Aortic stenosis | 3 (5%) | 0 |
| Pulmonic stenosis | 6 (9%) | 0 |
| Mitral regurgitation | 1 (2%) | 0 |
| Aortic regurgitation | 1 (2%) | 0 |
| Tricuspid regurgitation | 2 (3%) | 0 |
| Pulmonic regurgitation | 2 (3%) | 0 |
∗ History of cardiomyopathy indicates a patient with a known previous diagnosis of cardiomyopathy, even if ejection fraction is normal and clinical symptoms have resolved.
| No Heart Disease (n=68) | Other Heart Disease (n=65) | Cardiomyopathy (n=37) | p value | |
|---|---|---|---|---|
| Age (years±Standard Deviation) | 30 ± 6 | 29 ± 6 | 32 ± 5 | 0.017 |
| Weight (kg) | 86 ± 24 | 78 ± 18 | 88 ± 24 | 0.023 |
| Height (m) | 1.6 ± 0.1 | 1.6 ± 0.1 | 1.6 ± 0.1 | 0.614 |
| Body Mass Index (kg/m 2 ) | 31.9 ± 8.9 | 30.1 ± 6.1 | 33.4 ± 8.0 | 0.102 |
| Hispanic | 14 (21%) | 16 (25%) | 6 (16%) | 0.600 |
| Race | 0.002 | |||
| White | 47 (70%) | 48 (74%) | 23 (62%) | |
| Black | 9 (13%) | 2 (3%) | 12 (32%) | |
| Asian | 1 (2%) | 0 | 0 | |
| Other | 11 (16%) | 15 (23%) | 2 (5%) | |
| English as first Language | 43 (63%) | 37 (57%) | 25 (68%) | 0.539 |
| Married | 30 (44%) | 27 (42%) | 18 (49%) | 0.785 |
| Gravida | 2.8 ± 2.0 | 3.0 ± 1.9 | 2.8 ± 2.0 | 0.827 |
| Para | 1.0 ± 1.3 | 1.2 ± 1.3 | 1.4 ± 1.3 | 0.293 |
| Hypertension (baseline or gestational) | 17 (25%) | 8 (12%) | 14 (38%) | 0.011 |
| Transient Ischemic Attack | 0 | 0 | 1 (3%) | 0.164 |
| Stroke | 0 | 1 (2%) | 0 | 0.444 |
| Diabetes Mellitus (baseline or gestational) | 6 (9%) | 8 (12%) | 7 (19%) | 0.324 |
| Active Smoker | 11 (16%) | 14 (22%) | 10 (27%) | 0.410 |
| History of Arrhythmia | 0 | 17 (26%) | 4 (11%) | <0.001 |
| Ventricular Tachycardia | 0 | 2 (3%) | 1 (4%) | 0.358 |
| Supraventricular Tachycardia | 0 | 10 (15%) | 1 (3%) | 0.001 |
Cardiomyopathy patients were more frequently observed to have jugular venous distention at clinical presentation ( Table 3 ) and the other HD group most frequently had an abnormal cardiac murmur noted. There were no other differences in cardiac physical findings or hemodynamic parameters measured. More patients with cardiomyopathy took medications during pregnancy compared with other HD and no HD, including β blockers, diuretics, and digoxin ( Table 3 ). No patients took warfarin, angiotensin-converting enzyme inhibitors, or statin medications during the pregnancy.
| No Heart Disease (n=68) | Other Heart Disease (n=65) | Cardiomyopathy (n=37) | p value | |
|---|---|---|---|---|
| Clinical Presentation | ||||
| Systolic Blood Pressure (mmHg) | 132 ± 22 | 125 ± 16 | 125 ± 21 | 0.086 |
| Diastolic Blood Pressure (mmHg) | 77 ± 18 | 71 ± 10 | 72 ± 12 | 0.088 |
| Heart Rate (beats per minute) | 88 ± 13 | 84 ± 12 | 89 ± 15 | 0.167 |
| Jugular Venous Distention | 1 (2%) | 0 | 3 (8%) | 0.028 |
| Abnormal Cardiac Murmur | 0 | 18 (28%) | 6 (16%) | <0.001 |
| Pregnancy Medications | ||||
| Beta Blocker | 11 (16%) | 13 (20%) | 15 (41%) | 0.014 |
| Calcium Channel Blocker | 4 (6%) | 2 (3%) | 5 (14%) | 0.116 |
| Digoxin | 0 | 0 | 4 (11%) | 0.001 |
| Aspirin | 1 (2%) | 4 (6%) | 5 (14%) | 0.043 |
| Clopidogrel | 0 | 0 | 1 (3%) | 0.164 |
| Unfractionated Heparin | 2 (3%) | 4 (6%) | 1 (3%) | 0.574 |
| Enoxaparin | 3 (4%) | 1 (2%) | 0 | 0.311 |
| Warfarin | 0 | 0 | 0 | –– |
| ACE-Inhibitor | 0 | 0 | 0 | –– |
| Nitrates | 0 | 0 | 2 (5 %) | 0.026 |
| Furosemide | 0 | 1 (2%) | 7 (19%) | <0.001 |
| Statin | 0 | 0 | 0 | –– |
| Echocardiographic Data | ||||
| Left Ventricular Ejection Fraction | 62 ± 5 | 62 ± 6 | 44 ± 14 | <0.001 |
| Left Ventricular Ejection Fraction < 45% | 0 | 0 | 18 (49%) | <0.001 |
| Wall motion abnormality (%) | 0 | 0 | 3 (8%) | 0.004 |
| Left Ventricular Cavity Size (cm) | ||||
| Diastole | 4.6 ± 0.5 | 4.5 ± 0.7 | 5.2 ± 0.9 | <0.001 |
| Systole | 3.1 ± 0.5 | 3.0 ± 0.6 | 4.0 ± 1.0 | <0.001 |
| Left ventricular end diastolic dimension/body surface area (cm/m 2 ) | 2.4 ± 0.34 | 2.5 ± 0.5 | 2.7 ± 0.5 | 0.01 |
| Left Atrium Diameter | 3.4 ± 0.4 | 3.4 ± 0.9 | 3.8 ± 0.6 | 0.018 |
| Left Atrium Volume (ml) | 42 ± 21 | 45 ± 22 | 50 ± 23 | 0.278 |
| Aortic Stenosis | 0 | 1 (2%) | 1 (3%) | 0.444 |
| Pulmonic Stenosis | 0 | 3 (5%) | 0 | 0.085 |
| Aortic Regurgitation (> 3+) | 0 | 1 (2%) | 0 | 0.444 |
| Mitral Stenosis | 0 | 0 | 0 | –– |
| Mitral Regurgitation (> 3+) | 0 | 4 (6%) | 4 (11%) | 0.034 |
| Tricuspid Regurgitation (> 3+) | 0 | 1 (2%) | 1 (3%) | 0.444 |
| Pulmonic Regurgitation (> 3+) | 0 | 2 (3%) | 0 | 0.195 |
| Right Ventricular Systolic Pressure (mmHg) | 29 ± 6 | 31 ± 8 | 29 ± 10 | 0.622 |
Compared with other HD and no HD, patients with cardiomyopathy had lower ejection fractions ( Table 3 ), larger left ventricular end-diastolic and end-systolic dimensions, and larger left atrial diameters. Of the patients with cardiomyopathy, half of the patients had an ejection fraction ≤45%. There was no significant difference in right ventricular systolic pressure noted.
Maternal clinical adverse events and MACE are summarized in Table 4 . Compared with women with other HD and no HD, women with cardiomyopathy had significantly higher rates of maternal adverse clinical events, including preterm labor, preterm delivery, and admission to an intensive care unit. MACE rates were higher in women with cardiomyopathy compared with women with other HD and no HD, driven primarily by higher rates of heart failure. Two of the patients with cardiomyopathy had cardiac arrest, one of which died. Arrhythmia was significantly higher in the cardiomyopathy group compared with the other groups. Hospital length of stay for delivery was not significantly different among the 3 groups; length of stay was significantly longer in the cardiomyopathy group compared with other HD alone. Mode of delivery, induced labor, urgent/emergent cesarean section, peripartum maternal blood loss, preterm labor, and pregnancy-related hypertensive disorders were not statistically significant among the 3 groups. Cesarean delivery was more common in the cardiomyopathy group compared with other HD alone (p = 0.01, data not shown).
