br Postoperative outcomes br Adverse events during the first
Adverse events during the first 30 days postoperatively were recorded and graded according to Clavien-Dindo . Grade three or more was regarded as severe, defined as AEs that EPZ-6438 resulted in surgical, endoscopic or radiological intervention (grade 3), IC/ICU management (grade 4) or death (grade 5). The highest graded se-vere postoperative AE for each study participant was used. Co-morbidities, American Society of Anesthesiologists (ASA) Physical Status score, disease- and treatment-related data were derived from medical reports.
Statistical analyses were performed with Stata version 14 (Sta-taCorp LP, College Station, TX). Categorical data were reported as frequency (percentage) and continuous data as median (range). Wilcoxon rank-sum or Fisher exact tests were used for cross-sectional group comparisons and Wilcoxon's signed-Rank test or McNamer's test for longitudinal comparisons. Longitudinal regression models based on generalized estimating equation (GEE) were applied to evaluate the effect of preoperative change in endocrine testicular function, TD, patient-, disease- and treatment-related factors on severe postoperative AE. The robust variance estimator was used for calculations of standard error. All models were adjusted for elapsed time between start of RT and surgery to account for the T decline within days and the consecutive physio-logical LH increase described previously . Final models were adjusted for age, BMI and ASA-score to account for the impact of age, obesity and comorbidity on the hypothalamic-pituitary-gonadal axis according to survey data from the general popula-tion . Smoking, tumor stage, distance from the anal verge and type of surgery did not change the point estimates by more than 10% and were omitted in the adjusted analysis.
Endocrine testicular function
Postoperative adverse events
In total 39 severe postoperative AE were recorded among the 26 study participants. One died from aspiration (grade 5: 1.0%), four had life-threatening complications requiring intensive care (grade
Please cite this article as: Tapper J et al., Acute primary testicular failure due to radiotherapy increases risk of severe postoperative adverse events in rectal cancer patients, European Journal of Surgical Oncology, https://doi.org/10.1016/j.ejso.2019.07.023
J. Tapper et al. / European Journal of Surgical Oncology xxx (xxxx) xxx 3
Fig. 1. Study participant flow chart.
4: 3.8%), and 21 participants required surgical, endoscopic or radiological intervention with (grade 3b: 16.3%) or without (grade 3a: 3.8%) general anesthesia. Types of severe postoperative AE are specified in appendix A.
Longitudinal regression analysis
The preoperative change in endocrine testicular function was not related significantly to severe postoperative AE in the univari-ate analysis, but change in T indicated an inverse association with AE (OR: 0.878, 95% CI 0.759e1.015, p ¼ 0.078) (Table 3). After adjustment for age, ASA-score and BMI, the preoperative change in T was inversely related to severe postoperative AE, OR 0.844 (95% CI 0.720e0.990, p ¼ 0.034), which equals an OR of 1.18 for severe postoperative AE per 1 nmol/l preoperative decrease in T. LH/T-ratio was also related to severe postoperative AE, OR 2.020 (95% CI 1.010e4.039, p ¼ 0.047). Fig. 2 illustrates the relationship between change in serum T and predicted probability of having a severe postoperative AE, where an individual study participant can change in either direction. The TD and the relative TD were not associated with severe postoperative AE. The remaining covariates, analyzed in this study, were not related to severe postoperative AE (appendix B).