Sperm retrieval and the predictive parameter of non-obstructive azoospermia: a meta-analysis of literatures 1990 to 2008

Abstract

To investigate which sperm retrieval technique is suitable for the non-obstructive azoospermia (NOA) patient, and to identify the relevant predictive parameters. Literatures on NOA patients who had undergone sperm retrieval and pathological examination of testis were identified from Cochrane Library, CNKI and Medline (1990 to 2008) and analyzed. Twenty-five articles were enrolled. When testicular fine needle aspiration (TEFNA) was compared with testicular sperm extraction (TESE), the sperm retrieval rate of the former was 23.0%, significantly lower than that of the latter (52.2%, RR: 0.49, 95%CI: 0.41 - 0.60, P < 0.05); and when TESE was compared with micro-surgical testicular sperm extraction (mTESE), the sperm retrieval rate of the former was 35.7%, significantly lower than that of the latter (54.6%, RR: 0.70, 95%CI: 0.50 - 0.98, P < 0.05). Sperm retrieval rate was closely correlated with the testicular pathological category of the NOA patients. The sperm retrieval rates of the patients with hypospermatogenesis (HS), maturation arrest (MA), and Sertoli cell only syndrome (SCOS) were 76.7%, 46.2%, and 32.8% respectively (RR: 1.65, 2.40, 1.50; 95%CI: 1.21 - 2.91, 1.85 - 6.90, 1.02 - 2.26, P < 0.05). mTESE is the best sperm retrieval technique. A better to choice before deciding the treatment program of NOA patients is to identify the testicular pathological category in the NOA patients, and then to predict the outcome of TESE before assisted reproduction technology.