Vaginally obtained amniotic fluid samples in the diagnosis of subclinical chorioamnionitis.

Abstract

Preterm birth is often caused by infection or inflammation. High concentration of lactate dehydrogenase and low concentration of glucose in amniotic fluid obtained by amniocentesis are associated with subclinical chorioamnionitis. We evaluated amniotic fluid lactate dehydrogenase and glucose concentrations in relation to histologic chorioamnionitis in vaginally obtained samples. In a prospective study, vaginally obtained amniotic fluid samples were collected from 53 women with preterm prelabor rupture of membranes at 23(+4) to 34(+5) weeks of gestation at University Hospital, Helsinki, Finland. Amniotic fluid lactate dehydrogenase and amniotic fluid glucose were measured by immunochemiluminometric assays. Histopathologic examination of placenta was performed. The main outcome measure was histologic chorioamnionitis. Median concentration of vaginally obtained amniotic fluid lactate dehydrogenase was higher in women with histologic chorioamnionitis than in women without (1400 IU/L vs. 784.5 IU/L, p = 0.005). By receiver operating characteristics curve the optimal cut-off for amniotic fluid lactate dehydrogenase in relation to histologic chorioamnionitis was 1029 IU/L (sensitivity 65%, specificity 69%, positive predictive value 83% and negative predictive value 46%). Amniotic fluid lactate dehydrogenase concentrations showed striking fluctuation in repeat samples. Amniotic fluid glucose concentrations did not differ among women with or without histologic chorioamnionitis (0 mmol/L vs. 0.65 mmol/L, p = 0.20). Elevated amniotic fluid lactate dehydrogenase was associated with histologic chorioamnionitis, but decreased amniotic fluid glucose was not. However, the clinical value of vaginally obtained amniotic fluid lactate dehydrogenase is limited because of high sample-to-sample variability. Better biomarkers for optimal timing of delivery in women with preterm prelabor rupture of membranes are needed.