Laparoscopic Treatment for Appendicitis During Pregnancy: Retrospective Cohort Study

Construction and analysis of the database and study protocol were authorized by the Institutional Evaluation Committee under N° B20-010 and P20-070, respectively.

Assuming a 3: 1 relationship between surgery within 48 hours of symptom onset and surgery after 48 hours of 3: 1, we have an 80% power to find a difference in postoperative morbidity between groups of 30% with error type I of .05.

Analysed variables include: demographic characteristics, gestational age, clinical presentation, symptom duration from onset until surgical resolution, complementary studies, ASA score (Classification of the American Society of Anaesthesiology), intraoperative findings, type of surgery performed, intra and postoperative complications, length of stay, readmissions, use of tocolytics, preterm delivery, birth weight, Apgar score, maternal and foetal mortality, and obstetric and perinatological complications in pregnant patients with a preoperative diagnosis of acute appendicitis.

Any surgery in which the exploratory laparoscopy did not reveal intra-abdominal pathology and the appendix was macroscopically and microscopically normal was classified as a "negative laparoscopy". An "intraoperative complication" was defined as any unexpected intraoperative event, excluding conversion to conventional surgery, which was analysed as an independent event. A "postoperative complication" was defined as any deviation from the usual postoperative course within 30 days of surgery. An "obstetric complication" was one that occurred from the appendectomy until the end of the pregnancy, including foetal death and excluding preterm delivery. Spontaneous abortion and foetal demise were included within the same "foetal loss" outcome. "Preterm labour" was considered to be deliveries or caesarean sections that occurred prior to the 37th week of gestation.

For continuous variables, mean, standard deviation and / or minimum and maximum, or median and interquartile interval (IQR) were used, according to distribution. For categorical variables the number and corresponding percentages were reported. Continuous parameter comparisons were made using the test for independent samples or Wilcoxon-rank test; and when there were more than two groups, Anova or Kruskal Wallis were applied. For the comparison of categorical variables, Chi-square or Fisher's exact test were used, as appropriate. A p <0.05 was considered statistically significant. Likewise, a multivariate analysis was performed for the presence of postoperative and obstetric complications, contemplating possible confounders.