Nomograms for predicting changes in semen parameters in infertile men after varicocele repair

Objective: To formulate nomograms based on pre-repair characteristics to predict improvements in semen parameters after varicocele repair. Design: Model using multivariable linear regression based on prospectively collected database, with performance was quantified by concordance correlation coefficient and Pearson correlation coefficient after internal validation with bootstrapping. Setting: A male infertility specialty clinic. Patient(s): Men presenting for fertility evaluation from 2003-2012 having varicocele repair. Intervention(s): None. Main Outcome Measure(s): Semen parameters before and after varicocele repair. Result(s): Men undergoing varicocele repair (surgical or embolization) were identified via a prospectively collected database. The relationship of pre-repair semen and clinical characteristics to improvements in semen parameters was modeled using multivariable linear regression, then the model performance was quantified by concordance correlation coefficient and Pearson correlation coefficient after internal validation with bootstrapping. A total of 376 men who had undergone varicocele repair had data available for analysis. After varicocelectomy, the total motile count (TMC) varied depending on the initial left varicocele grade, ejaculate volume, sperm concentration, and motility. The final sperm concentration depended on the initial left varicocele grade, sperm concentration, and motility. The postvaricocelectomy sperm motility varied depending on the patient's age, left varicocele grade, sperm motility, morphology, and TMC. The final percentage of normal forms depended on the prevaricocelectomy sperm morphology, age, right varicocele grade, normal morphology, and TMC. Nomograms using prevaricocelectomy semen parameters and clinical features were developed to predict postvaricocelectomy TMC, sperm concentration, motility, and morphology. The concordance correlation coefficients were 0.45, 0.47, 0.65, and 0.36, respectively. Conclusion(s): Clinical factors provide substantial ability to predict postvaricocele repair semen parameters. These nomograms may be used by clinicians to predict postvaricocele repair semen parameters. (C) 2014 by American Society for Reproductive Medicine.