Between January and May 2016, 160 women were randomly assigned to the interventiongroup (cooling the lower abdomen, n = 81) or the control group (n = 79). Baseline characteristics were similar between groups, with the exception of mean blood loss during the thirdstage of labor. The primary outcome was not reduced by cooling, compared with no intervention (mean blood loss, 513.3 vs. 478.1 g, respectively; MD = 35.2 g; 95% confidenceinterval = −65.3–135.7). No adverse events occurred; however, seven (8.7%) women in theintervention group declined to continue cooling the lower abdomen because of discomfo