Discussion

The principal findings of the present analysis can be summarised as follows: (1) The prevalence of clinically relevant cardiac disease among pregnant women amounted to 0.6%. The majority of pregnant women found to have heart disease were primigravidae. (2) The most common cardiac disorder was RHD, in particular rheumatic mitral stenosis. (3) NYHA functional class III or class IV, presence of pulmonary hypertension and severe mitral stenosis were identified as predictors of maternal or fetal/perinatal death in an univariate analysis.

The prevalence rate of cardiac disease among pregnant women presenting to a tertiary care university hospital in Nepal amounted to 6 per 1000 pregnancies. Consistent with previous reports from Southeast Asia RHD accounted for the majority of cardiac disorders complicating pregnancy.10 In turn, congenital heart disease was more common than acquired heart disease in a large cohort from Canada.2 Pregnant women were systematically screened by detailed history and physical examination; subsequent evaluation by ECG and echocardiography was performed only when indicated by detailed history and physical examination, thus underestimating the true prevalence of silent disease not manifesting during pregnancy. Two-thirds of the women with cardiac disease complicating pregnancy were primigravidae. This finding may be explained twofold. The haemodynamic burden of pregnancy may unmask early stages of congenital or acquired heart disease that are usually silent and become manifest normally only later in the course of the disease. Second, manifestation of cardiac complications during the first pregnancy may prevent women from subsequent pregnancies.

The majority of cardiac complications during pregnancy was attributable to RHD, specifically rheumatic mitral stenosis. The differential vulnerability to the haemodynamic changes of pregnancy may potentially introduce an observation bias. While mitral regurgitation is the predominant rheumatic valvular disease overall,8 haemodynamic changes induced by pregnancy such as an increase in cardiac output and a reduction in systemic vascular resistance are less well tolerated in patients with mitral stenosis. Eight among 14 women with severe mitral stenosis in our cohort suffered exercise intolerance NYHA functional class III or class IV. One woman underwent percutaneous mitral balloon valvuloplasty during pregnancy; the remaining were medically treated with diuretics and β blockers in order to decrease filling pressures and prolong the diastolic filling period, respectively. Mitral balloon valvuloplasty was not offered on a regular basis during the initial start-up phase of a new cardiac catheterisation laboratory in the hospital. Referral to a hospital offering mitral balloon valvuloplasty was often prevented by limited resources. A non-randomised comparison of percutaneous mitral balloon valvuloplasty with open mitral valve commissurotomy suggested a lower fetal and neonatal mortality in women with mitral balloon valvuloplasty.11

Maternal and fetal/neonatal mortality amounted to 20% and was considerably higher as compared with previous reports.10 ,12 These studies however included less patients with symptomatic cardiac disease reflected by a lower number of women in NYHA functional class III or class IV. In an univariate analysis NYHA functional class III or class IV, presence of pulmonary hypertension, and severe mitral stenosis were identified as predictors of maternal or fetal/perinatal death. An analysis from a prospective multicentre study identified NYHA functional class >II, maternal left heart obstruction, smoking, multiple gestations and the use of anticoagulants as predictors of neonatal events. Primary cardiac events in the mother were predicted by prior cardiac events or arrhythmia, NYHA functional class >II, left heart obstruction and systemic ventricular dysfunction.2 Similarly, Silversides et al12 documented an increased occurrence of adverse fetal/neonatal events with increasing severity of mitral stenosis.

The present analysis has several limitations. First, no systematic screening for heart disease in pregnant women using echocardiography or ECG was performed, inducing an observation bias for symptomatic and more advanced cardiac disease. Second, medical management prior to delivery was not documented in detail. More than 90% of women presented only at the time of labour and did not have regular visits with an obstetrician. Third, outcome data is limited to mortality rates. Information on other clinical outcomes was not collected systematically.

In conclusion, rheumatic mitral stenosis was the most frequent heart disease complicating pregnancy in a consecutive cohort from a teaching hospital in Nepal. Exercise intolerance, pulmonary hypertension and severe mitral stenosis were identified as predictors of maternal or fetal/perinatal mortality.