Objectives: To analyze risks versus benefits of esophagogastroduodenoscopy (EGD) during pregnancy to the fetus and pregnant female.
Methods: Retrospective study of 83 consecutive pregnant patients who underwent EGD admitted to eight university teaching hospitals during a study period of up to 14 yr, with follow-up of fetal outcome. Controls included: 48 pregnant females matched for EGD indications who did not undergo EGD (pregnant controls), 83 nonpregnant females undergoing EGD matched for age and EGD indication (EGD controls), and national pregnancy outcome rates (national controls).
Results: EGD indications included acute GI bleeding in 37, vomiting and abdominal pain in 17, vomiting in 14, abdominal pain in 11, and other in four. The mean week of gestation was 19.8 +/ 8.9 EGD was diagnostic in 65 patients. The diagnostic yield of EGD was 95% for acute GI bleeding and ranged from 50 to 82% for the other indications (significantly different rates, odds ratio = 9.3, p < 0.001, odds ratio confidence interval = 2.22-45.5). Esophagitis was found in 62+ of patients with a diagnostic EGD. No significant endoscopic complications occurred. EGD did not induce labor. Excluding six voluntary abortions and three unknown pregnancy outcomes, 70 (95%) of 74 patients delivered healthy babies (pregnant control rate = 94%, national control rate = 98.4%, all not significant, Fisher's exact test). Moreover, the four poor outcomes (three stillbirths and one involuntary abortion) occurred in high risk pregnancies and were unrelated to EGD temporally or etiologically. No other infant had a congenital malformation noted in the neonatal nursery. Nine (12.7%) of the live born infants had a low birth weight (pregnant control rate = 8.5%, national control rate = 7.0%, all not significant, kappa 2). Mean Apgar scores of live born infants were 7.7 +/ 1.6 at 1 min and 9.0 +/ 0.6 at 5 min (pregnant control scores = 7.6 +/ 2.0 at 1 min and 8.5 +/ 1.1 at 5 min; national control mean scores = 8.0 +/ 1.4 at 1 min and 9.0 +/ 0.9 at 5 min; all not significant, Student's t test). In three cases with fetal cardiac monitoring, EGD did not induce abnormal fetal heart rates (preendoscopy rate = 143.3 +/ 5.8 beats/min, postendoscopy rate = 148.0 +/ 10.6 beats/min).
Conclusions: In this study, EGD did not induce labor or result in congenital malformations. EGD is not contraindicated during pregnancy. EGD is beneficial in pregnant patients with upper GI bleeding.