Characterization and control of intraamniotic infection in an urban teaching hospital

OBJECTIVES: Our purpose was to determine (1) whether risk factors for intraamniotic infection were similar in women delivered of preterm infants versus term infants and (2) whether infection control techniques could decrease the incidence of intrapartum fever on a labor and delivery unit. STUDY DESIGN: A total of 5409 consecutive patients (group 1) admitted to the Medical College of Virginia's labor and delivery unit were followed up prospectively to determine the development of intraamniotic infection. Demographic and intrapartum data were collected by use of a standard data form by infection control practitioners. An additional 2549 consecutive patients (group 2) were followed up after institution of infection-control measures. RESULTS: Intraamniotic infection occurred in 416 of 5399 (7.7%) women (group 1) admitted to the labor and delivery suite. Odds ratios for term and preterm patients having intraamniotic infection with rupture of membranes ≤12 hours compared with <12 hours were 5.81 (95% confidence interval 512 to 6.59) and 2.49 (95% confidence interval 1.77 to 3.50), respectively. Odds ratios for term and preterm patients with internal monitors having intraamniotic infection compared with patients in whom internal monitors were not used were 2.01 (95% confidence interval 1.7 to 2.4) and 1.42 (95% confidence interval 0.99 to 2.04), respectively. Odds ratios for term and preterm patients having intraamniotic infection with more than four vaginal examinations compared with four or fewer vaginal examinations was 3.07 (95% confidence interval 2.53-3.73) and 1.59 (95% confidence interval 1.11-2.27), respectively. Intrapartum fever occurred in 475 (8.8%) women in group 1 and in 252 (9.8%) women in group 2 (not significant). CONCLUSIONS: Risk factors (duration of ruptured membranes, use of internal monitoring, number of vaginal examinations) were similar in both term and preterm women with intraamniotic infection. Infection control measures failed to decrease the incidence of intrapartum fever in our patient population.