LAUSR

BACKGROUND Recurrent pregnancy loss (RPL) and unexplained infertility are the current indications to test sperm DNA fragmentation (SDF) according to the EAU Guidelines on sexual and reproductive health. OBJECTIVE To identify a novel and better performing model to diagnose primary infertile men presenting with altered SDF and to outlie its predictive ability in respect to current EAU Guidelines' recommendations. MATERIALS AND METHODS Data from the latest 515 consecutive primary infertile men as for World Health Organization (WHO) criteria were analysed. Semen analysis, SDF (according to SCSA) and serum hormones were considered in every patient. Altered SDF was defined with levels >30%. Descriptive statistics was applied to compare patients with normal vs. SDF>30%. The new predicting model was identified through logistic regression analysis exploring potential predictors of SDF>30% at first clinical presentation. Diagnostic accuracy between the two predictive models (EAU Guidelines Vs. New) was assessed and decision curve analyses (DCA) tested their clinical benefit. RESULTS Of 515, 268 (51.9%) patients had SDF>30% at clinical presentation. Patients with SDF>30% were older (median (IQR) 39 (35 - 43) vs. 37 (34 - 41) years), had lower mean testicular volume (TV) (Prader 15 (12 - 20) vs. 17.5 (13.5 - 20) and lower total motile sperm count (TMSC) (1.80 (0.7 - 13.2) vs. 11.82 (4.2 - 44.5) x 106 ), all p<0.001. No other clinical differences were depicted. The two groups showed similar rates of history of RPL and unexplained infertility. At multivariable logistic regression analysis, age >38 yrs (OR: 2.43) and baseline TMSC < 20×106 (OR: 3.72) were associated with SDF>30%, after adjusting for Prader <15, history of miscarriages and unexplained infertility, all p<0.0001. The newly identified model (unexplained infertility + history of poli-abortions + Prader <15 + age ≥ 38 years + TMSC < 20×106 ) showed higher accuracy to identify SDF >30% at baseline in respect to EAU guidelines (AUC: 72.1 vs. AUC: 52.7), with superior clinical net benefit use. CONCLUSIONS The application of the EAU sexual and reproductive health guidelines does not ensure proper identification of primary infertile men with pathological SDF. We propose a novel and better performing predictive model to identify those infertile men with altered SDF at first clinical assessment. DISCUSSION As altered SDF has been widely linked with the inability to conceive, this second-level test could be further implemented over the diagnostic work-up of a broader sub-set of patients presenting for MFI. We propose a better performing model to identify this specific category of patients. This article is protected by copyright. All rights reserved.