Preterm deliveries in women with systemic lupus erythematosus

To compare the clinical, laboratory, and demographic variables of women in our clinic with systemic lupus erythematosus (SLE) who have had a pregnancy resulting in a live birth and identify any correlations with either term or preterm delivery.

Pregnancies in women with SLE from 1999 to 2001 were retrospectively reviewed. We recorded demographic data, disease activity (SLE Disease Activity Index, SLEDAI), obstetric history, prednisone dosage, other medications taken during pregnancy, history of renal disease, and autoantibody status [including antinuclear antibody, anti-DNA, anticardiolipin IgG (aCL), and lupus anticoagulant (LAC)]. Preterm delivery was defined as gestational age at delivery < 37 weeks. We performed a literature survey using PubMed and the key words SLE, pregnancy, and outcome.

Of the 72 pregnancies, 28 (38.9%) resulted in preterm deliveries. There were no significant differences in any demographic or disease variables measured comparing term versus preterm delivery groups. More women in the preterm group were taking > or = 10 mg/day prednisone during their pregnancy (50.0% vs 22.2%; p = 0.028), and the mean dose was significantly higher than the term group taking > or = 10 mg/day (24.8 vs 16.7 mg/day; p = 0.047). There was a higher prevalence of women with aCL IgG in the preterm group (p = 0.023). The mean weeks gestation was shorter for women positive for aCL IgG compared to the group negative for aCL (34.9 +/- 4.4 vs 37.5 +/- 3.2 weeks, respectively; p = 0.032). There was no difference in second trimester disease activity between the term and preterm groups (33.3% and 36.4% of each group had a SLEDAI of 0). However, significantly more women in the term group received no medication during their pregnancies compared to women in the preterm group (20.0% vs 0.0%; p = 0.031).

The rates of preterm deliveries, premature rupture of membranes, intrauterine growth restriction, and aPL in SLE pregnancies vary considerably in published reports, most of which are retrospective analyses. Our rates closely approximate the median values for all measures. We found preterm deliveries to be associated with disease activity (as determined by the use of any medication throughout pregnancy vs no medication, and prednisone dose > or = 10 mg/day) and the presence of aCL IgG but not LAC. Our results suggest that inactive disease rather than controlled disease at the onset of pregnancy may be the determining factor in extending SLE pregnancies to full term, thereby decreasing maternal and fetal morbidity.