Sperm retrieval rates and clinical outcomes for patients with different causes of azoospermia who undergo microdissection testicular sperm extraction-intracytoplasmic sperm injection.

Azoospermia, defined as the absence of spermatozoa in the ejaculate after assessment of centrifuged semen on at least two occasions, occurs in approximately 10% of male infertility cases. 1 Clinically, azoospermia is divided into obstructive azoospermia (OA) and nonobstructive azoospermia (NOA), which accounts for 60% of cases of defective spermatogenesis. 2 The etiology of NOA can be classified as acquired, congenital, or idiopathic. 3 Causes of congenital NOA mainly include genetic abnormalities, such as Klinefelter syndrome (KS), azoospermia Y chromosome factor ( AZFa , AZFb , and AZFc ) microdeletions, and cryptorchidism. Acquired causes of NOA mainly include trauma, testicular torsion, varicocele, inflammation, drugs, radiation, and testicular spermatogenetic dysfunction caused by other factors (cryptorchidism, pituitary tumor, and idiopathic hypogonadotropic sexual hypofunction). Idiopathic NOA is diagnosed when known etiologies are excluded. 4 Recently, intracytoplasmic sperm injection (ICSI) using microdissection sperm retrieval (micro-TESE) has been commonly applied to NOA patients, and is generally considered to be a standard method, with high sperm retrieval rate (SRR) and minimal tissue loss. 5 6 Most studies of micro-TESE-ICSI for NOA patients have mainly focused on comparing sperm retrieval and clinical pregnancy outcomes with those of OA patients. 4 7 8 However, there is limited data on clinical pregnancy outcomes, which are important in clinical practice for couples with OA, for distinct NOA with different etiologies. Therefore, the aim of our study was to compare the SRRs and clinical outcomes of patients who had different causes of NOA and who were undergoing micro-TESE-ICSI treatment.

HJ conceived and designed the study. KH supported substantial contributions to the study conception and design. LMZ analyzed and interpreted the data. HLZ drafted the article. DFL, JMM, WHT, HCL, LZ, and YL collected the data. All authors read and approved the final manuscript.

We collected medical data retrospectively from the Reproductive Medicine Center of Peking University Third Hospital in Beijing (China) from January 2014 to December 2017. Azoospermia was confirmed by testing at least two semen samples. Data regarding medical history; physical examination; assessments of sex hormones, including follicle-stimulating hormone (FSH), luteinizing hormone (LH), and testosterone (T) levels; and genetic abnormalities were recorded for each patient. All enrolled participants in our study fulfilled the following criteria: (1) complete clinical data; (2) absence of preexisting chronic diseases, including hypertension, diabetes mellitus, and heart, kidney, hematological, and autoimmune diseases; (3) absence of female infertility factors, including anovulation, hormonal infertility, tubal factors, and endometriosis; (4) ages of couples from 23 years to 35 years; and (5) body mass index (BMI) of all couples from 18 kg m −2 to 30 kg m −2 . In total, 769 cases of azoospermia who underwent 347 cycles of micro-TESE-ICSI were identified and studied. Patients were classified into Group A (KS, n = 284, 125 cycles), Group B (AZFc microdeletion, n = 91, 64 cycles), Group C (cryptorchidism, n = 52, 39 cycles), Group D (previous mumps and bilateral orchitis, n = 23, 23 cycles), and Group E (idiopathic azoospermia, n = 319, 96 cycles) according to different causes of NOA. The clinical characteristics, in particular, SRR, embryonic development, and pregnancy outcomes, of the patients were compared between all groups. The study protocol was approved by the Ethics Committee of Peking University Third Hospital, and all patients provided written informed consent before the study. Micro-TESE was performed under general anesthesia, and a mid-line scrotal incision was made on the median raphe of the scrotum. Testicular parenchyma was exposed and directly examined for dilated tubule areas under an operating microscope (×12 to ×24 magnification). The seminiferous tubules were teased with 21G needles and placed into a G-MOPS-plus buffer (Vitrolife, Goteborg, Sweden). The collected tubules were finely minced into a homogeneous suspension and examined for the presence of spermatozoa using an inverted microscope (TS100, Nikon, Tokyo, Japan). Samples were positive when at least one sperm was identified. If fresh oocytes could be obtained on the same or the next day of sperm retrieval, fresh sperm were used for the ICSI procedure. If not, sperm were cryopreserved with sperm-freezing solution (Vitrolife). The testicular suspension was diluted 1:1 with the sperm-freezing solution in a 2-ml straw. The straw was incubated at room temperature for 10 min, then placed horizontally on a styrofoam board in a liquid nitrogen bath for 30 min, and then the straw was stored in liquid nitrogen vapor. When thawing, the straws were put in a 37°C incubator for 15 min, and then the sperm were collected from the cryoprotectant by two washes in culture medium, each with centrifugation at 300g for 10 min (BY-160A, BY-CENTRIFUGE, Beijing, China). The resulting samples were incubated in culture medium (Vitrolife) for later use. Ovarian stimulation was a long protocol performed with a combination of gonadotrophin-releasing hormone (GnRH) analogs (Cetrotide, Merck Serono, Amsterdam, the Netherlands), FSH (Gonal-F alfa, Merck Serono), and human chorionic gonadotrophin (HCG; Choriogonadotropin alfa, Merck Serono). Oocyte retrieval involved vaginal ultrasound-guided puncture of follicles 36 h and 38 h after HCG administration. The oocytes were maintained in G-MOPS-plus (Vitrolife) and placed in a 37°C incubator with 5% CO 2 , 5% O 2 , and 90% N 2 . Cumulus cells were removed by pipetting and exposure to hyaluronidase 2 h after retrieval (Type VIII; Sigma Chemical Company, St. Louis, MO, USA). ICSI was performed as described in detail elsewhere. 9 Fertilization was assessed by the presence of two pronuclei (2pn) and two polar bodies 17–19 h after insemination. Embryo transfer was performed on day 3 after oocyte retrieval. Embryos were scored according to the Society for Assisted Reproductive Technology (SART) scoring system, 10 and only the best embryos were selected for transfer. The number of transferred embryos was usually limited to two to reduce the risk of a multiple pregnancy. The fertilization rate was the number of 2pn zygotes in all mature metaphase II (MII) stage oocytes. Clinical pregnancy was defined as a rising serum HCG level at least 12 days after embryo transfer. Clinical pregnancy was confirmed by the presence of a gestational sac during ultrasound examination in the 5 th week after transfer. The miscarriage rate was the number of miscarriage cycles divided by the number of clinical pregnancy cycles. Indicators evaluating embryonic development included the number of oocytes and MII oocytes that were retrieved, the fertilization rate, the number of transferred embryos, and the good-quality embryo rate. Clinical outcomes included the cumulative pregnancy, implantation, cumulative live birth, and cumulative miscarriage rates, as well as the cumulative number of birth defects. Statistical analyses were performed using Statistical Package for Social Sciences version 22.0 (IBM Corp., Armonk, NY, USA). Data were expressed as the mean ± standard deviation, and categorical data were expressed as percentages. P < 0.05 was considered statistically significant.

The characteristics of all patients are described in Table 1 . The mean male age was 28.6 ± 4.0 years, and the mean T level was 6.08 ± 9.97 ng ml −1 in Group A; both of these parameters were the lowest in all groups. Significance was reached compared with Groups B, C, and E ( P < 0.05). Female age and testicular volume in Group A were 26.7 ± 3.5 years and 1.92 ± 0.8 ml, respectively, which were the lowest in all groups. The FSH and LH levels were higher in Group A than for any other groups ( P < 0.05). There were no significant differences in other patient characteristics among the different groups. Characteristics of all patients Data are expressed as mean±s.d.. a Values in Group A was the lowest among all the groups, and significance was reached compared with Groups B, C, or E ( P <0.05); b values in Group A was the lowest of all the groups, and significance was reached compared with Groups B, C, D, or E ( P <0.05); c values in Group A was the highest of all the groups ( P <0.05). s.d.: standard deviation; FSH: follicle-stimulating hormone; LH: luteinizing hormone; T: testosterone; BMI: body mass index Table 2 presents the SRRs of all groups. Patients in Group D had the highest and most successful SRR, which was 100%. The average SRR for all patients in our study was 46.0%. The sperm retrieval rate using micro-testicular sperm extraction TESE: testicular sperm extraction; SRR: sperm retrieval rate Embryonic development for all groups is summarized in Table 3 . The fertilization rate was 27.9% ± 19.1% in Group B, which was the lowest among the groups, and 65.1% ± 19.3% in Group D, which was the highest among the groups (both P < 0.05). No difference was found for other aspects of embryonic development, including the average number of oocytes retrieved, average number of MII oocytes, oocyte maturation rate, and average number of transferred embryos for all the groups. Embryonic development for all groups Data are expressed as mean±s.d.. a Values in Group B was the lowest of all the groups ( P <0.05); b values in Group D was the highest of all the groups ( P <0.05). MII oocytes: metaphase II oocytes; s.d.: standard deviation The pregnancy outcomes for all patients are shown in Table 4 . The rates of clinical pregnancy, implantation, and live birth in Group D were 78.3%, 65.0%, and 74.0%, respectively, which were higher than those in any other groups (all P < 0.05). Patients in Group B had the lowest clinical pregnancy, implantation, and live birth rates for all the groups (all P 0.05). Pregnant outcomes for all groups a Values in Group B was the lowest of all the groups ( P <0.05); b values in Group D was the highest of all the groups ( P <0.05)

All authors declare no competing interests.

Our study found that patients with orchitis have the highest SRR and best clinical outcomes. Although AZFc microdeletion patients had a higher SRR, their clinical outcomes were worse. A higher SRR does not indicate good outcomes nor does a lower SRR indicate poor outcomes. These findings will help formulate our suggestions to patients and provide relevant guidance for clinical work. As related data are very limited, more evidence is needed through studies with larger samples and different communities to investigate this aspect in future.

In recent years, as micro-TESE technology has increasingly improved, more NOA patients have been able to conceive their offspring through this technology. However, no systematic study has reported the SRRs or related clinical outcomes for patients with different causes of NOA, key parameters that also concern NOA patients in clinical practice. Therefore, our center analyzed relevant recent data to study the SRRs and clinical outcomes for patients with different causes of NOA and who underwent micro-TESE-ICSI, in order to provide relevant guidance for clinical work. Klinefelter syndrome is the most frequent chromosomal abnormality associated with NOA, with an estimated prevalence ranging from 1:500 to 1:700 in newborn males, and is characterized by infertility, elevated FSH and LH levels, normal or reduced T level, small testicular volume, and testicular hypofunction. 11 As these characteristics are quite apparent, patients with KS can be diagnosed and treated early. 12 Our data showed that the FSH and LH levels were the highest and that female age and testicular volume were the lowest in KS patients compared with all other groups (both P < 0.05). Both male age (28.6 ± 4.0 years) and T levels in KS were the lowest compared with all the groups, and significance was reached compared with Groups B, C, or E ( P < 0.05). Our data were in accordance with the characteristics of KS patients. We also investigated the SRRs of patients with different causes of NOA. We found that the average SRR was 46.0%, which is in agreement with previous studies showing SRRs from 40%–60%. 13 14 In our study, the highest SRR (100%) was for patients with orchitis. So far, we have not found a report describing the SRR for patients with orchitis who undergo micro-TESE treatment. We speculate that the reason for the high SRR may be that NOA in most of these patients was caused by mumps, and therefore the infertility caused by this acquired infection was less likely to affect the fertility of their offspring. 15 Moreover, the SRR for patients with cryptorchidism in our study was also very high (75.0%). Previous studies have shown that men with cryptorchidism had a good chance of sperm retrieval when undergoing micro-TESE. 16 17 Bernie et al . 14 found that men with cryptorchidism tended to have a slightly higher SRR (74%) than that of other NOA patients, which is consistent with our results. Interestingly, the SRR was 73.3% for patients with the AZFc microdeletion in our study. Y chromosome microdeletions are the most important genetic etiology of male infertility, and AZFc is the most frequently microdeleted region in these NOA patients, accounting for more than 60% of all AZF microdeletions. 18 Many studies recommend TESE for patients with AZFc microdeletion, as sperm is typically present in the testes of these men. 18 For instance, Oates et al . 18 discovered sperm in 67% of testes in these patients undergoing TESE. However, the SRR of AZFc microdeletion patients who underwent micro-TESE treatment was limited. Recently, Sabbaghian et al . 19 showed that patients with the AZFc microdeletion had a more successful SRR of 43.7%, which was lower than that in our study. Consistent with our results, An et al . 8 found that AZFc microdeletion patients had an SRR of 70%. The SRR for KS patients was slightly lower than that of other patients (44.7%), which is in line with a previous study (40%). 20 Testicular hypofunction and the high FSH levels of KS patients may be a possible explanation for the low SRR, as reports have shown that men with azoospermia with high FSH levels have lower SRRs, and FSH was a suggested predictive measure for sperm retrieval. 20 The SRR for KS patients in our study was consistent with results from a previous meta-analysis, which showed that the SRR for KS patients was close to 50%. 21 In our study, patients with idiopathic azoospermia had the lowest SRR (30.7%), which was consistent with that of a previous report. 22 As related data were limited, the possible reasons for this need to be explored in further research. In terms of clinical outcomes, our results demonstrated that patients with orchitis had the highest rates of fertilization, clinical pregnancy, implantation, and live birth, and that the patients with AZFc microdeletion had the lowest rates of fertilization and clinical pregnancy among all patients ( P < 0.05). As mentioned above, the better clinical outcomes of patients with orchitis may be attributed to their infertility being caused by an acquired infection. 15 We found only one report concerning AZFc microdeletions. In 2018, Sabbaghian et al . 19 reported a fertilization rate of approximately 36% for patients with AZFc microdeletions who underwent micro-TESE-ICSI treatment, but no pregnancy was observed in the authors' center. The low fertilization and pregnancy rates for these patients were in agreement with our results. Notably, Liu et al . 23 reported that patients with oligozoospermia and an AZFc microdeletion and who were undergoing conventional TESE-ICSI treatment also had lower fertilization rates, which may indicate that the clinical reproductive outcomes for AZFc microdeletion patients remain adverse despite the use of sperm retrieval methods. In addition, although KS patients had a relatively low SRR in our study, their clinical outcomes were relatively good. A recent meta-analysis 21 showed that the current micro-TESE method for patients with KS resulted in clinical pregnancy and live birth rates close to 50%, which is consistent with our data. Our results also showed that patients with cryptorchidism and idiopathic azoospermia had relatively good clinical outcomes. One study using TESE-ICSI for patients with cryptorchidism reported fertilization, pregnancy, and live birth rates of 43.3%, 30%, and 20%, respectively. 24 To date, no report has addressed the outcomes from micro-TESE-ICSI for patients with cryptorchidism or idiopathic azoospermia. Presently, new problems are arising as preimplantation genetic diagnosis (PGD) provides patients with the possibility to select female embryos for transfer, especially for KS and AZFc microdeletion patients who have a genetic etiology of male infertility. In our clinical practice, we are concerned with helping and adequately counseling patients with infertility as they make their own choices to become parents, especially for couples who decide to pursue PGD. Because of the high economic burden on patients and the debatable risks to offspring from PGD, our center does not recommend PGD for KS patients. None of the patients who enrolled in our study underwent the PGD procedure. To the best of our knowledge, the current study is the first to systematically explore the SRRs and clinical outcomes for patients with NOA due to different etiologies who undergo micro-TESE-ICSI treatment. Our study provided the largest sample size to date, although the retrospective and single-center nature of our data present key limitations for further consideration. We also lacked a long-term follow-up plan after delivery.