Preterm labor and late gestation fetal loss remain significant barriers to clinical fetal surgery. To investigate the response of the gravid uterus to anesthetic and tocolytic agents and surgical procedures, 27 chair-restrained pregnant rhesus monkeys from 123 to 152 days gestation (term 168 days) underwent implantation of electrodes to monitor uterine electromyographic (EMG) activity. Seven had electrodes placed at the time of hysterotomy for placement of intraamniotic pressure catheters, without disturbing the fetus; 12 at the time of hysterotomy for placement of fetal carotid and jugular catheters. Eight had electrodes placed as an initial procedure to study the uterine EMG response to one or more of the following procedures performed subsequently: amniocentesis, maternal laparotomy without uterine manipulations, hysterotomy without fetal surgery, and hysterotomy with fetal surgery. A total of 43 procedures was performed. Preterm labor and delivery were induced in 1 of 15 (6.7%) monkeys who underwent procedures with minimal uterine manipulation (electrode placement, amniocentesis, and maternal laparotomy), in 3 of 8 (38%) monkeys who had hysterotomies without fetal surgery, and in 11 of 20 (55%) monkeys who had hysterotomies with fetal manipulation. The difference between those undergoing minimal uterine manipulation and those undergoing hysterotomy (with and without fetal surgery; fetal loss in 14 of 28, 50%) is statistically significant (P = 0.01). In animals undergoing hysterotomy, frequent coordinated contractions (type I EMG pattern) emerged as the animal awoke. The uterine activity was inhibited by halothane anesthesia, but not by either preoperative indomethacin or postoperative ritodrine infusion.(ABSTRACT TRUNCATED AT 250 WORDS)