Measuring Sperm DNA Fragmentation and Clinical Outcomes of Medically Assisted Reproduction: A Systematic Review and Meta-Analysis

Abstract

Sperm DNA fragmentation has been associated with reduced fertilization rates, embryo quality, pregnancy rates and increased miscarriage rates. Various methods exist to test sperm DNA fragmentation such as the sperm chromatin structure assay (SCSA), the sperm chromatin dispersion (SCD) test, the terminal deoxynucleotidyl transferase mediated deoxyuridine triphosphate nick end labelling (TUNEL) assay and the single cell gel electrophoresis (Comet) assay. We performed a systematic review and meta-analysis to assess the value of measuring sperm DNA fragmentation in predicting chance of ongoing pregnancy with IVF or ICSI. Out of 658 unique studies, 30 had extractable data and were thus included in the meta-analysis. Overall, the sperm DNA fragmentation tests had a reasonable to good sensitivity. A wide variety of other factors may also affect the IVF/ICSI outcome, reflected by limited to very low specificity. The constructed hierarchical summary receiver operating characteristic (HSROC) curve indicated a fair discriminatory capacity of the TUNEL assay (area under the curve (AUC) of 0.71; 95% CI 0.66 to 0.74) and Comet assay (AUC of 0.73; 95% CI 0.19 to 0.97). The SCSA and the SCD test had poor predictive capacity. Importantly, for the TUNEL assay, SCD test and Comet assay, meta-regression showed no differences in predictive value between IVF and ICSI. For the SCSA meta-regression indicated the predictive values for IVF and ICSI were different. The present review suggests that current sperm DNA fragmentation tests have limited capacity to predict the chance of pregnancy in the context of MAR. Furthermore, sperm DNA fragmentation tests have little or no difference in predictive value between IVF and ICSI. At this moment, there is insufficient evidence to recommend the routine use of sperm DNA fragmentation tests in couples undergoing MAR both for the prediction of pregnancy and for the choice of treatment. Given the significant limitations of the evidence and the methodological weakness and design of the included studies, we do urge for further research on the predictive value of sperm DNA fragmentation for the chance of pregnancy after MAR, also in comparison with other predictors of pregnancy after MAR.

Fig 1. Flowdiagram of search and selection strategy in a systematic review and meta-analysis of sperm DNA fragmentation tests and pregnancy rates after MAR.

Legend: not applicable.

Fig 2. Overall risk of bias in meta-analysis.

This figure illustrates the overall risk of bias in the meta-analysis. The horizontal axis represents the number of studies included. The color of the bars represent the risk of bias. Yellow: high risk, blue: low risk and grey: unclear risk.

Fig 3. HSROC curve.

Hierarchical summary receiver operating characteristic (HSROC) plot of sperm DNA fragmentation for prediction of (clinical) pregnancy. Each circle on the plot represents the pair of sensitivity and specificity from a study and the size of the circle is scaled according to the sample size of the study. The solid red block represents the summary sensitivity and specificity, and this summary point is surrounded by a 95% confidence region (yellow dashed line) and 95% prediction region (green dotted line). Sperm DNA fragmentation in the prediction of (clinical) pregnancy for all studies and all cut-off values of the DNA fragmentation index reported: (A) SCSA, (B) SCD test, (C) TUNEL assay and (D) alkaline Comet assay. AUC: Area under the curve; HSROC: Hierarchical summary receiver operating characteristics.

Fig 4. Forest plot.

Forest plot of sperm DNA fragmentation according to the DNA fragmentation index for predicting pregnancy. The plot shows study-specific estimates of sensitivity and specificity (with 95% confidence intervals). The studies are ordered according to the type of treatment: (A) SCSA, (B) SCD test, (C) TUNEL assay and (D) alkaline Comet assay. CI: confidence interval.