Objective: To determine whether amnioinfusion is associated with labor and delivery complications, and whether complication type and reported incidence are related to infusion method. Methods: Questionnaires regarding amnioinfusion experience were sent to every academic obstetrics and gynecology department in the United States (78 maternal-fetal medicine fellowship directors or, if the department did not have a fellowship, 206 residency directors). A literature review on amnioinfusion was also performed. Results: Seventy-six percent of fellowship directors and 62% of residency directors responded to our survey, representing 644,910 deliveries per year and at least 22,833 amnioinfusions per year. A wide variety of infusion protocols were reported. Forty-nine centers reported at least one associated complication; none was significantly associated with any of the various aspects of the many protocols (P > .05). The mean number (± standard error of the mean) of amnioinfusions performed annually was similar between centers that did (261 ± 48) and did not (154 ± 29) report complications (P = .06). The literature review suggested that amnioinfusion is efficacious and relatively safe. Conclusion: Amnioinfusion is performed nationwide according to widely varying protocols with few associated complications. Neither the method employed nor the number of infusions performed appears to significantly increase the risk of having a complication. © 1995 The American College of Obstetricians and Gynecologists.