Risk Factors for Adhesion-Related Readmission and Abdominal Reoperation after Gynecological Surgery: A Nationwide Cohort Study
Masja K. Toneman, Tjitske D Groenveld, Pepijn Krielen, Angelo B. Hooker, Rudy Leon De Wilde, Luz Angela Torres-de la Roche, Atillio Di Spiezio Sardo, Philippe R. Koninckx, Ying Cheong, Annemiek Nap, Harry van Goor, Pille Pargmae, Richard P. G. ten Broek
Abstract
More than half of women in developed countries undergo surgery during their lifetime, putting them at risk of adhesion-related complications. Adhesion-related complications include small bowel obstruction, chronic (pelvic) pain, subfertility, and complications associated with adhesiolysis during reoperation. The aim of this study is to predict the risk for adhesion-related readmission and reoperation after gynecological surgery. A Scottish nationwide retrospective cohort study was conducted including all women undergoing a gynecological procedure as their initial abdominal or pelvic operation between 1 June 2009 and 30 June 2011, with a five-year follow-up. Prediction models for two- and five-year risk of adhesion-related readmission and reoperation were constructed and visualized using nomograms. To evaluate the reliability of the created prediction model, internal cross-validation was performed using bootstrap methods. During the study period, 18,452 women were operated on, and 2719 (14.7%) of them were readmitted for reasons possibly related to adhesions. A total of 2679 (14.5%) women underwent reoperation. Risk factors for adhesion-related readmission were younger age, malignancy as indication, intra-abdominal infection, previous radiotherapy, application of a mesh, and concomitant inflammatory bowel disease. Transvaginal surgery was associated with a lower risk of adhesion-related complications as compared to laparoscopic or open surgeries. The prediction model for both readmissions and reoperations had moderate predictive reliability (c-statistics 0.711 and 0.651). This study identified risk factors for adhesion-related morbidity. The constructed prediction models can guide the targeted use of adhesion prevention methods and preoperative patient information in decision-making.