Association between endometriosis and obstetric complications: Insight from the National Inpatient Sample

OBJECTIVE: This study aimed to investigate the relationship between endometriosis and adverse obstetric outcomes using data from the National Inpatient Sample (NIS) database. METHODS: The ICD-10 coding system was used to identify codes for endometriosis and obstetric outcomes, and data from the NIS (2016-2019) were analyzed. Descriptive statistics were used to summarize variables, while the chi-square test was used to detect significant differences for categorical variables. Univariate and multivariate regression analyses were conducted to assess the association between endometriosis and obstetric outcomes. On multivariate analysis, adjustment was done for age, race, hospital region, smoking status, and alcohol misuse. Forest plots were used to visualize odds ratios and their 95% confidence intervals. RESULTS: Overall, 2,854,149 women were included in this analysis, of whom 4,006 women had endometriosis. The post-hoc Bonferroni correction was applied to account for multiple comparisons, and our analyses revealed several statistically significant associations (p < 0.004). Specifically, on univariate analysis, significant associations with endometriosis were identified for ruptured uterus, placenta previa, placental abruption, postpartum hemorrhage, preeclampsia, amniotic fluid abnormality, gestational diabetes, preterm labor, and multiple gestation. On multivariate analysis, significant associations with endometriosis were observed for placenta previa, placental abruption, postpartum hemorrhage, preeclampsia, amniotic fluid abnormality, preterm labor, premature rupture of membranes, and multiple gestation. CONCLUSION: The present findings provide important insights into the potential relationship between endometriosis and various adverse obstetric outcomes and may help inform clinical practice and future research. Further studies that use more detailed clinical data and longitudinal designs are needed to solidify the presented conclusions.