Prediction model for live birth in ICSI using testicular extracted sperm

Area of Science:

• Reproductive Medicine
• Infertility Treatment
• Andrology

Background:

• Prediction models are crucial in reproductive medicine for assessing treatment success.
• Currently, no specific prediction model exists for live birth following testicular sperm extraction with intracytoplasmic sperm injection (TESE-ICSI).

Purpose of the Study:

• To identify parameters with predictive value for live birth in couples undergoing TESE-ICSI.
• To develop and validate a prediction model for live birth after TESE-ICSI.

Main Methods:

• A retrospective cohort study involving 1559 TESE-ICSI cycles across two academic hospitals (2007-2015).
• Development of a prediction model using multivariable logistic regression in a dataset of 1006 cycles, with external validation on 553 cycles.

Main Results:

• Live birth was associated with younger female age, a first TESE-ICSI cycle, lower male LH, higher male testosterone, use of motile sperm for ICSI, and obstructive azoospermia.
• The prediction model demonstrated moderate discriminative capacity (c-statistic 0.67) and good calibration in the validation set.

Conclusions:

• The developed model can differentiate couples with low or high chances of live birth following TESE-ICSI.
• This tool can assist in patient counseling and clinical decision-making for TESE-ICSI procedures.