To determine if latency following preterm premature rupture of membranes can be predicted using the amniotic fluid index, a retrospective observational study was performed using patient records to identify those admitted with a diagnosis of preterm premature rupture of membranes. Fifty-one patients with initial evaluation within 48 hours of rupture and gestational age 26 to 34 weeks were identified. Amniotic fluid index was evaluated along with a biophysical profile; these were followed serially until delivery. Delivery was accomplished because of spontaneous labor or chorioamnionitis. Amniotic fluid index at initial evaluation was stratified into three groups: Low (less than 5.0), reduced (4.0 to 7.9), and normal (8.0 or higher). Latency in days from preterm premature rupture of membranes to delivery was evaluated for these groups. A difference was noted in that latency was significantly longer in the group with normal versus low or reduced amniotic fluid index groups. The amount of residual amniotic fluid, as measured by the amniotic fluid index, following preterm premature rupture of membranes, is predictive of latency.