Amnioinfusion in preterm premature rupture of membranes (AMIPROM): a randomised controlled trial of amnioinfusion versus expectant management in very early preterm premature rupture of membranes--a pilot study.

Background: Fetal survival is severely compromised when the amniotic membrane ruptures between 16 and 24 weeks of pregnancy. Reduced amniotic fluid levels are associated with poor lung development, whereas adequate levels lead to better perinatal outcomes. Restoring amniotic fluid by means of ultrasound-guided amnioinfusion (AI) may be of benefit in improving perinatal and long-term outcomes in children of pregnancies with this condition.

Objective: The AI in preterm premature rupture of membranes (AMIPROM) pilot study was conducted to assess the feasibility of recruitment, the methods for conduct and the retention through to long-term follow-up of participants with very early rupture of amniotic membranes (between 16 and 24 weeks of pregnancy). It was also performed to assess outcomes and collect data to inform a larger, more definitive, clinical trial.

Design: A prospective, non-blinded randomised controlled trial. A computer-generated random sequence using a 1 : 1 ratio was used. Randomisation was stratified for pregnancies in which the amniotic membrane ruptured between 16(+0) and 19(+6) weeks' gestation and 20(+0) and 24(+0) weeks' gestation. The randomisation sequence was generated in blocks of four. Telephone randomisation and intention-to-treat analysis were used.

Setting: Four UK hospital-based fetal medicine units - Liverpool Women's NHS Trust, St. Mary's Hospital, Manchester, Birmingham Women's NHS Foundation Trust and Wirral University Hospitals Trust.

Participants: Women with confirmed preterm prelabour rupture of membranes between 16(+0) and 24(+0) weeks' gestation. Women with multiple pregnancies, resultant fetal abnormalities or obstetric indication for immediate delivery were excluded.

Interventions: Participants were randomly allocated to either serial weekly transabdominal AI or expectant management (Exp) until 37 weeks of pregnancy, if the deepest pool of amniotic fluid was < 2 cm.

Main Outcome Measure: Short-term maternal, pregnancy and neonatal outcomes and long-term outcomes for the child were studied. Long-term respiratory morbidity was assessed using validated respiratory questionnaires at 6, 12 and 18 months of age and infant lung function was assessed at approximately 12 months of age. Neurodevelopment was assessed using Bayley's Scale of Infant Development II at a corrected age of 2 years.

Results: Fifty-eight women were randomised and two were excluded from the analysis owing to termination of pregnancy for lethal anomaly, leaving 56 participants (28 serial AI, 28 Exp) recruited between 2002 and 2009, with annual recruitment rates varying between 2 and 14. Recruitment to the study improved significantly from 2007 with National Institute for Health Research (NIHR) funding. There was no significant difference in perinatal mortality [19/28 vs. 19/28; relative risk (RR) 1.0; 95% confidence interval (CI) 0.70 to 1.43], maternal morbidity or neonatal morbidity. The overall chance of surviving without long-term respiratory or neurodevelopmental disability is 4/56 (7.1%): 4/28 (14.3%) in the AI arm and 0/28 in the expectant arm (0%) (RR 9.0; 95% CI 0.51 to 159.70).

Conclusions: This pilot study found no major differences in maternal, perinatal or pregnancy outcomes. The study was not designed to show a difference between the arms and the number of survivors was too small to draw any conclusions about long-term outcomes. It does signal, however, that a larger, definitive, study to evaluate AI for improvement in healthy survival is indicated. The results suggest that, with appropriate funding, such a study is feasible. A larger, definitive, study with full health economic analysis and patient perspective assessment is required to show whether AI can improve the healthy survivor rate.