Risk of Pregnancy in Moderate and Severe Aortic Stenosis From the Multinational ROPAC Registry

Fro Mu Ad ve Ka Co Me Ca (EO (20 (20 an No su rel Ma BACKGROUND Controversial results on maternal risk and fetal outcome have been reported in women with aortic

P regnancy carries a very low risk of death in developed countries, but overall, cardiac reasons remain the leading cause of maternal mortality (1).
Consistent with this, women with preexisting heart disease have 100 times greater mortality than normal (2). Pregnancy is associated with profound changes in hemodynamic parameters, perhaps explaining why pre-existing heart disease has such an adverse impact on morbidity and mortality in pregnant women (3). Although there is clear evidence that pregnancy is a high-risk endeavor in women with complex heart disease, and especially those with severe pulmonary hypertension (4, 5), available data are limited in women with less complex heart disease. Obstructive heart lesions will be aggravated by the increase in stroke volume occurring with pregnancy, and are therefore of particular concern. Aortic stenosis (AS) is one of these lesions, but it is relatively uncommon in women of childbearing age. However, when present, it has been reported to be associated with an increased risk of maternal cardiovascular events, including death, obstetric morbidity (such as pre-term birth), and fetal complications, including growth restriction, miscarriage, and stillbirth (6). The evidence in this setting is nevertheless limited, and the results of published reports are conflicting. Unfortunately, prior studies either encompassed all forms of heart disease (7,8) or included mild AS (9). In addition, some series report on historic patient cohorts (6). As a consequence, the reported maternal mortality rate ranges between 2% and 17.4%, and the risk in contemporary cohorts of women presenting with severe AS remains unclear (6,9).
The purpose of this study was therefore to investi- The present study retrospectively analyzed the outcome and complications in pregnant women with moderate or severe AS included in the registry up to April 2014. We focused exclusively on women with moderate or severe AS. Patients with additional congenital or acquired heart disease (with the exception of simple corrected pre-tricuspid shunts, aortic coarctation) were not included in the current study. The severity of AS was graded on the basis of available transthoracic echocardiographic data at baseline.
Moderate AS was defined as a peak transaortic gradient $36 mm Hg (corresponding to a peak velocity $3 m/s), whereas severe AS was defined as a peak aortic gradient $64 mm Hg (corresponding to a peak velocity $4 m/s) using the simplified Bernoulli equation (11,12). This is in agreement with current guidelines and general recommendations for assessing the severity of AS in the presence of normal flow rate (13).
Patients who had undergone aortic valve replacement before pregnancy were included if they fulfilled the hemodynamic criteria described in the preceding text.
However, further analyses were also performed and presented separately because prosthetic valve-related risks require additional consideration. Repeated pregnancies were excluded from the analysis.
Baseline characteristics included maternal age, general cardiovascular risk factors, major noncardiac disease, cardiac diagnosis, prior interventions, cardiac symptoms, medication, and obstetric history.
Heart failure before pregnancy was defined according to current guidelines clinically as a syndrome in which patients have typical symptoms (e.g., breathlessness, ankle swelling, and fatigue) and signs (e.g., elevated jugular venous pressure, pulmonary crackles, and displaced apex beat) (14). Maternal mortality was defined as death during pregnancy or up to 1 week after delivery. Miscarriage was defined as loss of pregnancy up to 24 weeks of gestation or a fetus weighting <500 g, whereas fetal mortality was defined as fetal loss beyond 24 weeks of pregnancy.

DISCUSSION
The current registry-based study suggests that mortality risk in pregnant women with AS, including  Forrest plots illustrating the results of the univariate logistic regression analysis for adverse fetal outcome (low birth weight, defined as birth weight <2,500 g). Severe aortic stenosis, peak aortic gradient, and NYHA class >1 at baseline were predictive of worse fetal outcome. An asterisk denotes significant results. Abbreviations as in Figure 1.    * * Forrest plots illustrating the results of the univariate logistic regression analysis for adverse fetal outcome (small for gestational age). Severe aortic stenosis and peak aortic gradient were predictive of worse fetal outcome. An asterisk denotes significant results. Abbreviations as in Although historic series report maternal mortality rates ranging from 11% to 20% (6,9,(15)(16)(17), in more recent series, cardiac event rates were lower, with approximately 10% of patients experiencing a cardiac event, and mortality being rare (15,17,18). However, some of these studies also included women with mild AS, diluting the clinically interesting group of patients with moderate/severe AS. The current multinational study of 96 pregnant women followed between 2007 and 2014, and focusing specifically on moderate or severe AS, showed no maternal mortality, thus supporting the notion that maternal morbidity, rather than mortality, is the main clinical problem in this setting. Previous studies suggest a risk of heart failure during pregnancy of approximately 10% (1,15). Our data are consistent with this estimate, showing that heart failure occurred in 11.5% of patients with moderate or severe AS. However, in patients with pre-existing symptoms, pregnancy was complicated by heart failure in 26.3%. Arrhythmias are also a recognized, common complication during pregnancy in this setting, with a reported incidence of 3% to 25% in previous reports (1). In the current series, arrhythmias were also reported as a complication, but with a relatively low rate of around 3%.
Overall, 57% of women included in the current study delivered by Cesarean section. This rate is comparable to that in a recent AS study (18), but higher than the overall Cesarean section rate of 42% reported in the ROPAC registry (19) and in comparison to previous AS studies (17,20). This difference may be related to the current study including a larger proportion of patients with severe AS and a considerable number on anticoagulation. Although current recommendations support vaginal delivery with an assisted second stage of labor in the majority of patients, Cesarean section continues to be advocated by some physicians in women with severe AS and in those requiring anticoagulation (1). Furthermore, country-specific preferences may account for some of the differences, as discussed in detail previously (21).
Severe fetal complications were rare in the current study, but pre-term birth and low birth weight were observed in one-third of patients with severe AS. In addition, newborns of women with severe AS were more likely to have a low Apgar score, to present with low birth weight, and to be small for gestational age.  (15,17).
Two women included in this study underwent an aortic valve intervention during pregnancy. Both procedures were successful, and the patients continued to have uneventful pregnancies with vaginal deliveries of healthy babies. This illustrates that in experienced hands, complications can be managed successfully in the current era; however, the risks associated with these procedures (especially of fetal loss) should not be underestimated (23). Therefore, appropriate preconceptional assessment and counseling are paramount to avoid pregnancy complications, and to allow for elective procedures to be performed before pregnancy, especially in symptomatic women with severe AS. Nevertheless, the low complication rate in asymptomatic patients, even those with hemodynamically severe AS, supports a conservative approach and avoidance of prophylactic surgery in this group, considering the dilemma of managing pregnancy in the presence of a prosthetic valve (1). STUDY LIMITATIONS. Due to the multinational nature of the registry, we included pregnancies in developed and emerging countries. We accept that outcomes may be different in the latter setting.
However, the vast majority of patients included (74%) were from resource-rich countries of the European Union or from the United States. In addition, obstetric outcomes were not statistically different between the European Union/United States and the remaining countries. On the basis of the hemodynamic inclusion criteria, we also enrolled 13 patients with a mechanical heart valve. It is well appreciated that mechanical heart valves are associated with increased morbidity during pregnancy and worse fetal outcomes (24).  sults. In addition, the causality between pregnancy and later mortality may be difficult to establish in AS.
The current study did not investigate the problem of ascending aortopathy, which is associated especially with bicuspid aortic valve disease. Nevertheless, we did not observe pregnancy-associated aortic dissection in our cohort. Awareness of this potential complication, and appropriate preconception and pregnancy assessment of aortic dimensions should be part of routine care for patients with known bicuspid aortic valve.
Not all of the patients had a pre-pregnancy echocardiographic assessment. We are aware that gradients are expected to increase with increasing cardiac output. Therefore, the classification of AS severity before pregnancy may have been different in some patients with borderline measurements and first presentation during advanced pregnancy (i.e., some of the patients classified to have severe AS during pregnancy may have had moderate AS if prepregnancy data had been available).

CONCLUSIONS
The current multinational study, including 96 women with at least moderate AS managed in the current era during pregnancy, highlights the ongoing challenge of pregnancy in this setting. Although, the maternal mortality rate was zero in this cohort, hospitalizations for cardiac reasons were frequent, and both pre-term birth and low birth weight were observed in one-third of patients with severe AS. The fact that heart failure, the most frequent maternal complication, occurred predominantly in patients with severe AS who had at least mild symptoms, indicating intervention already before pregnancy, emphasizes the importance of appropriate pre-conception patient evaluation and counseling.