Prediction of foetal atrioventricular conduction using maternal disease, treatment and anti-Ro/La autoantibody levels.

Abstract

To investigate the relationship between maternal disease, anti-rheumatic treatment, autoantibody levels and foetal atrioventricular (AV) conduction. A total of 324 pregnancies from two groups of anti-Ro/La autoantibody-positive women were included. The Surveillance group consisted of pregnancies with known anti-Ro positivity followed from early pregnancy; n = 302, including nine with AV block (AVB), and the Bradycardia group with no prior knowledge of anti-Ro/La autoantibodies and referred upon detection of foetal bradycardia, with a subsequent AVB II-III diagnosis (n = 22). Foetal AV conduction and routine anti-Ro52, anti-Ro60 and anti-La autoantibody levels were analysed. Women with primary Sjögren disease (SjD) had longer foetal AV time intervals (130.5 ± 11.9 vs 125.8 ± 8.7 ms, P = 0.002) compared with autoantibody-positive women with neither SjD nor SLE. Notably, AZA, mostly used in women with SLE, was associated with shorter foetal AV time intervals compared with autoantibody-positive pregnancies without AZA treatment (122.5 ± 8.9 vs 128.7 ± 11.0 ms, P = 0.038). For antibody-based prediction, anti-Ro52, anti-Ro60 and anti-La antibody levels obtained from routine assays could identify nearly 20% of pregnancies as low risk, and high levels (≥8 U/ml) of anti-La antibodies were associated with an increased risk of AVB I-III (OR = 6.0, 1.26-28.5). However, the positive predictive value for AVB II-III of the three autoantibodies did not reach 5% for a sensitivity of 100%. This report, based on over 20 years of surveillance of anti-Ro positive pregnancies, identifies associations between maternal disease, their treatments and foetal AV conduction. Routine detection assays for anti-Ro/La antibodies had low performance in predicting AVB among an anti-Ro/La positive population.