OBJECTIVE To examine the economic impact of initial treatments for varicocele-associated nonobstructive azoospermia, specifically varicocelectomy versus microsurgical testicular sperm extraction (TESE) with IVF/intracytoplasmic sperm injection (ICSI). METHODS Decision analytic model based on 1) outcomes data from Society for Assisted Reproductive Technology (SART) database and peer-reviewed literature and 2) costing data from Medicare Resource-Based Relative Value Scale and sampling of high volume US IVF centers. METHODS Academic medical center. METHODS Simulation with a decision analytic model. METHODS Variation of successful spontaneous live delivery after varicocelectomy versus rate of successful live delivery after IVF/ICSI. METHODS Cost-effectiveness. RESULTS Microsurgical TESE was more cost effective than varicocelectomy. In 1999, initial treatment with microsurgical TESE was more cost effective ($65,515) than varicocelectomy ($76,878). Relative cost-effectiveness was unchanged in 2005: $69,731 versus $79,576. The cost-effectiveness of both treatments improved in relation to projections by inflation. Sensitivity analyses suggest that the relative cost-effectiveness of TESE versus varicocelectomy can only be changed with either substantial improvement in spontaneous live delivery rates after varicocelectomy or with deterioration in IVF success rates. CONCLUSIONS Microsurgical TESE appears to be more cost effective than varicocelectomy for treatment of varicocele-associated nonobstructive azoospermia when indirect costs are considered. The cost-effectiveness of both treatments has improved with time. These results may be tailored with institution-specific data to allow more individualized results.