Route of hysterectomy: a retrospective, cohort study in English NHS Hospitals from 2011 to 2017

Objective To assess variation in the route of hysterectomy over 7 years and to assess regional variation in practice. Design Retrospective cohort study. Setting English NHS Hospitals 2011-2017. Population 230 876 patients having a hysterectomy for six diagnostic categories (endometrial cancer, endometriosis and pain, menstrual disorders, fibroids, benign adnexal masses, and 'other') identified from Hospital Episode Statistics. Methods The proportion of hysterectomies carried out by each route was calculated for each year overall and for each primary diagnosis by year. Comparisons between 2011 and 2017 were via chi-square test. Rank correlation coefficients were calculated to assess trends over the study period. Analysis of regional variation in practice was restricted to 2017. A multivariable logistic regression was performed to obtain crude and adjusted odds of having a minimal access hysterectomy. Main outcome measures The proportion of abdominal, vaginal, laparoscopic, and failed laparoscopic procedures for each primary diagnosis by study year. Odds of a minimal access hysterectomy in 2017. Results The proportion of hysterectomies performed laparoscopically increased from 20.2% in 2011 to 47.2% in 2017, as did the proportion of failed laparoscopic procedures; 1.7% in 2011 to 2.8% in 2017. The proportion of abdominal hysterectomies decreased from 70.4% in 2011 to 46.5% in 2017. There was a smaller decrease in vaginal procedures from 7.8% in 2011 to 3.5% in 2017. Regional variation in the route of hysterectomy was demonstrated in 2017, which persisted when adjusted for confounding factors. Conclusions The proportion of laparoscopic procedures has increased, and it was the commonest route of hysterectomy for this cohort in 2017. There were significant regional differences in route of hysterectomy in 2017. Tweetable abstract Increasing laparoscopic hysterectomy and decreasing abdominal hysterectomy rates from 2011 to 2017 with regional variation in practice.