Cardiac and obstetric outcomes of pregnancies for women after cardiotoxic therapy in childhood: a single center observational study

Abstract

Background: Childhood cancer survivors (CCS) are at increased risk of cardiomyopathy during pregnancy if they have prior cardiotoxic exposure. Currently, there is no consensus on the necessity, timing and modality of cardiac monitoring during and after pregnancy. Therefore, we examined cardiac function using contemporary echocardiographic parameters during pregnancy in CCS with cardiotoxic treatment exposure, and we observed obstetric outcomes in CCS, including in women without previous cardiotoxic treatment exposure.

Method: A single-center retrospective cohort study was conducted among 39 women enrolled in our institution's cancer survivorship outpatient clinic. Information on potential cardiotoxic exposure in childhood, cancer diagnosis and outcomes of all pregnancies were collected through interviews and review of health records. Echocardiographic exams before and during pregnancy were retrospectively analyzed for left ventricular ejection fraction (LVEF) and global longitudinal strain (GLS) if available. The primary outcomes were (i) left ventricular dysfunction (LVD) during pregnancy, defined as LVEF < 50% or a decline of ≥ 10% in LVEF below normal (< 54%), and (ii) symptomatic heart failure (HF). Rate of obstetric and fetal complications was compared to the general population through the national perinatal registry (PERINED).

Results: All pregnancies (91) of 39 women were included in this study. The most common malignancy was leukemia (N = 17, 43.6%). In 22 patients, echocardiograms were retrospectively analyzed. LVEF_baseline was 55.4 ± 1.2% and pre-existing subnormal LVEF was common (7/22, 31.8/%). The minimum value of LVEF during pregnancy was 3.8% lower than baseline (p = 0.002). LVD occurred in 9/22 (40.9%) patients and HF was not observed. When GLS was normal at baseline (< -18.0%; N = 12), none of the women developed LVD. Nine of out ten women with abnormal GLS at baseline developed LVD later in pregnancy. In our cohort, the obstetric outcomes seemed comparable with the general population unless patients underwent abdominal irradiation (N = 5), where high rates of preterm birth (only 5/18 born at term) and miscarriage (6/18 pregnancies) were observed.

Conclusion: Our study suggests that women with prior cardiotoxic treatment have a low risk of LVD during pregnancy if GLS at baseline was normal. Pregnancy outcomes are similar to the healthy population except when patients underwent abdominal irradiation.

Keywords: Anthracyclines; Cardiotoxicity; Echocardiography; Heart failure; Pregnancy; Surveillance; Survivorship.