Clinical course of pregnancies in patients with cardiac sarcoidosis

Introduction

Cardiac sarcoidosis (CS) requently affects patients of reproductive age (20–40 years). While associated with arrhythmia risks during pregnancy, data is limited and conflicting. This is the largest study to date evaluating maternal and fetal outcomes to identify key predictors of maternal adverse cardiovascular events in this population.

Methods

This multicentre retrospective study included female patients with histologically-proven sarcoidosis and positive cardiac involvement based on imaging from the Registry on Cardiac Sarcoidosis. Patients with manifest CS who experienced pregnancy (CS-pregnancy group, n=15, median age 38 [32-40] years, 73% on corticosteroids, 27% on TNFa-inhibitors, 27% on methotrexate, 20% on azathioprine) were matched for age and immunosuppressive therapy to 15 female CS patients without pregnancy (CS-group, median age 43 [35-48] years, 80% on corticosteroids, 33% on TNFa-inhibitors, 20% on methotrexate, 7% on azathioprine). Maternal clinical and imaging data at baseline, during pregnancy and 12 months postpartum, as well as fetal outcomes were collected. Primary outcomes included maternal (major adverse clinical events (MACE): life-threatening sustained ventricular arrhythmia (VA), hospitalisation for heart failure (HF), cardiac death, or all-cause mortality) and obstetric/fetal complications.

Results

Pregnancy complications occurred in 73% (n=11/15). C-sections were performed in 60% of cases, with two preterm births. Immunosuppression was maintained (corticosteroids and TNFalfa-inhibitors remained constant, methotrexate was stopped). Neonatal complications included intrauterine growth restriction in 13 (2/15) and small for gestational age in 33% (5/15). Postpartum maternal complications occurred in 27% (4/15, infections in 20%, haemorrhage in 7%).

MACE occurred in 33% (5/15) during pregnancy vs. 13% (2/15) in controls at 1-year follow-up (table 1). No patient with CS died during FU. VA predominantly occurred during the second and third trimester. Elevated serum TNFalfa values during pregnancy (measured at 5.6months [4.7-7.4]) were independent predictors for VA (HR 7, 95%CI 1.2-40, p=0.029, table 2).

Left ventricular ejection fraction (LVEF) worsened in 31% of patients during pregnancy (4/13, LVEF 55% [49-59] vs 55% [39-58], p=0.153, 21 months [13-30]) and remained reduced at 1-year follow-up (53% [38-57], p=0.007, 12months [8-12]). TNFalfa-inhibitor use both before and during pregnancy was a significant predictor for stable LVEF during pregnancy and at follow-up (B=0.639, 95%CI 2.8-24.9, p=0.019; B=0.63, 95% CI 3.5-22.9, p=0.011, respectively).

Conclusions

Pregnancy in patients with CS confers an increased risk of malignant arrhythmia for the mother, but fetal outcomes are good. Risk factors for MACE during pregnancy are previous sustained VA and elevated serum TNFalfa during pregnancy. Thorough patient counseling and a multidisciplinary cardio-obstetric approach is advised.Adverse events in CS patientsPredictors of MACE in CS