Moderated Poster
Presentation Authors: Scott Lundy*, Neel Parekh, Tianming Gao, Edmund Sabanegh, Sarah Vij, Cleveland, OH
Introduction: Microscopic testicular sperm extraction (mTESE) represents a viable option for sperm retrieval in men with nonobstructive azoospermia (NOA), and histopathology is the only known variable that consistently predicts surgical sperm retrieval rate (SRR). We examined our institutional experience with mTESE sperm retrieval rates and performed multivariable regression to develop a predictive nomogram for successful sperm retrieval.
Methods: A retrospective chart review was performed for all patients undergoing mTESE at our institution from 2012 to 2018. Patient demographics, preoperative laboratory workup, semen analysis, operative findings, pathology, and success rates were reviewed. Only patients with NOA were included in the final analysis. Univariate analysis was performed using chi-square for categorical variables and t-tests for continuous variables. A multivariable logistic regression was then fitted and backward variable selection performed based upon concordance index. The final model was internally validated via bootstrapping and calibration of the prediction model was visually confirmed.
Results: Between 2012 and 2018, 118 patients underwent a total of 129 mTESE procedures with an overall SRR of 57%. 95 patients with NOA of varying etiologies were included in the final analysis. Semen pH ranged from 6.8 to 8. Final pathology demonstrated Sertoli only pattern in 57%, maturation arrest in 14%, hypospermatogenesis in 19%, and mixed or unavailable pathology in 11% of patients. Univariate analysis demonstrated an association between semen pH and successful sperm retrieval (p=0.035). Multivariable logistic regression identified semen volume, semen pH, and FSH as predictors of success, and a nomogram (Figure 1) was produced with a concordance index of 0.68.
Conclusions: We generated a predictive nomogram based upon preoperative semen volume, pH, and FSH as variables associated with successful mTESE. While the mechanism by which volume and pH may predict SRR is unclear, this model may provide modest benefit when counseling patients regarding individualized success rates prior to undergoing mTESE.
