Remembered International Index of Erectile Function Domain Scores Are Not Accurate in Assessing Preoperative Potency in Candidates for Bilateral Nerve-Sparing Radical Retropubic Prostatectomy

INTRODUCTION: The reliability of reported postoperative data in patients undergoing nerve-sparing radical retropubic prostatectomy is often limited because the degree of sexual function (SF) has not been assessed objectively both before and after treatment. Most reports include only a retrospective chart review, and there is a question of whether such data are accurate. AIM: To test the agreement between a remembered International Index of Erectile Function (IIEF) score, which targeted SF regarding a period preceding the surgery by 6 months and a real-time IIEF, 4 weeks prior to surgery, in candidates for bilateral nerve-sparing radical retropubic prostatectomy (BNSRRP). METHODS: From May 2005 to May 2006, 333 consecutive patients were candidates for BNSRRP at our institution. Upon admission on the day prior to surgery, each patient was asked to complete a set of validated questionnaires including both a remembered and a real-time IIEF. Two-tailed Student's t-test, chi2 test, Pearson correlation coefficient, multivariate regression analyses, and interrater agreement (kappa) were used to test the agreement between the two assessments. MAIN OUTCOME MEASURES: Assessing the preoperative SF characteristics of candidates for a BNSRRP, and testing the reliability of a remembered IIEF with the interrater agreement (kappa) test. RESULTS: Mean scores for the remembered IIEF were overall better than the real-time IIEF scores, as supported by direct comparison of the mean IIEF domain scores. Univariate correlation analysis and multivariate regression analysis indicated a significant correlation in the quality of the SF during the two periods. However, the remembered IIEF scores did not show a good statistical agreement with those of the real-time assessment, as demonstrated by the interrater agreement analysis. CONCLUSIONS: Because of the lack of significant agreement between remembered and real-time IIEF scores, the present findings indicate that remembered IIEF should not be used to assess SF in a real-life clinical setting in candidates for BNSRRP.