Putative Predictive Value of Etiological and Clinical Factors on Sperm Retrieval Outcomes in Non-Obstructive Azoospermia Using Micro-TESE: A Retrospective Study

Abstract

Non-obstructive azoospermia (NOA) is a condition of testicular failure caused by various factors. To retrieve sperm in NOA patients, microdissection testicular sperm extraction (micro-TESE) is a highly effective technique. The present study aimed to evaluate successful sperm retrieval (SSR) in NOA patients across different etiologies in the Iranian population to identify predictive factors. This retrospective analysis included 566 NOA patients undergoing micro-TESE from 2018 to 2023, with 58 histories of Cryptorchidism, 40 Klinefelter syndrome (KS), 6 Y chromosomal microdeletions (YCMDs), 6 histories of chemotherapy, 5 mumps orchitis, and 451 patients with idiopathic reasons. The overall SSR was 43.2%, with a significantly lower SSR in the KS group (p = .012). Patients with higher average ages tended to have higher SSR rates, especially in the idiopathic group. Hormone levels differed among the groups, with higher follicle-stimulating hormone and luteinizing hormone levels in the YCMDs group, higher testosterone levels in the idiopathic and Cryptorchidism groups, and higher prolactin levels in the KS group. There were no significant differences in other clinical characteristics between the SSR and sperm retrieval failure groups, except for a positive relationship between prolactin levels and SSR rates in the KS group (r = .45, p = .003). Our data underscores that underlying etiology and genetic background may reveal more valuable predictive value than age, hormone levels, and testicular volume. This finding suggests that no patient with NOA should be deprived of micro-TESE based on candidate predictors of SSR presented to date.

Introduction

Infertility is not just a medical issue but also a profound social and emotional challenge (A. Sharma & Shrivastava, 2022). The inability to conceive can cause stress, societal pressure, and anxiety (Simionescu et al., 2021). Male infertility, responsible for 30% to 50% of cases, often leads to additional psychological and relationship problems, including feelings of inadequacy and lowered self-esteem (Afrashteh & Mirkuhi, 2020; Eisenberg et al., 2023). Azoospermia, the most severe form of male infertility, affects 15% of infertile men, making it more challenging to achieve parenthood and often leading couples to pursue invasive and costly treatments (M. Sharma & Leslie, 2023; Tharakan et al., 2021). These factors together make infertility a complex issue that affects both individual well-being and social relationships (Simionescu et al., 2021).

Azoospermic patients are generally categorized into obstructive azoospermia (OA) and non-obstructive azoospermia (NOA) (Wu et al., 2021). In OA, spermatogenesis is preserved, and azoospermia occurs due to obstruction in any reproductive tract, such as the vas deferens, epididymis, or ejaculatory duct (Franco et al., 2022). NOA is characterized by the complete absence of sperm in the seminal fluid without any blocking factor and is caused by spermatogenic dysfunction in whole or in part, which includes three types of hypospermatogenesis, maturation arrest, and Sertoli cell-only syndrome (Cioppi et al., 2021; M. Sharma & Leslie, 2023). Therefore, accurate diagnosis of azoospermia is essential for clinical decisions and treatment options (Andrade et al., 2021). It helps to distinguish between OA and NOA, which can affect the choice of treatment and surgical interventions (Andrade et al., 2021).

NOA is a common and severe phenotype in infertile men with azoospermia, accounting for approximately 1% of the male population and 10% of infertile couples (M. Sharma & Leslie, 2023). Possible etiologies of NOA include genetic factors such as Klinefelter syndrome (KS) and Y chromosomal microdeletions (YCMDs), congenital anomalies such as cryptorchidism, acquired causes such as mumps orchitis and idiopathic factors (Kang et al., 2021).

Patients with NOA face limited options for biological parenthood, as assisted reproductive technologies typically have low success rates. Many must turn to donor insemination or adoption due to difficulties in sperm retrieval (Majzoub et al., 2024). Conventional testicular sperm extraction (cTESE) and microdissection testicular sperm extraction (micro-TESE), have been developed to obtain live sperm in azoospermia patients (Esteves et al., 2020). While both techniques are invasive, micro-TESE seems to be more effective than cTESE in sperm retrieval rates in NOA patients (57 vs. 37%) (Esteves et al., 2020). Micro-TESE allows for precise dissection of the testicular tissue, maximizing the chance of locating areas with active sperm production and reducing unnecessary tissue damage compared to cTESE (Gao et al., 2022). However, predicting the outcomes of micro-TESE in NOA patients remains challenging (Gao et al., 2022).

Although factors such as hormone levels, testicular size, and the etiology of NOA may provide insights, they do not accurately determine the likelihood of success (Qi et al., 2021). These factors may vary concerning race, lifestyle, environmental pollution, and geographical context (Kimmins et al., 2024). Studies on this relationship within the Iranian population are lacking, particularly at the Yazd Research Institute of Reproductive Sciences, one of the largest infertility centers in Iran. Thus, this study aims to explore and compare the clinical characteristics and sperm retrieval rates in NOA patients with different etiologies who underwent micro-TESE treatment. In addition, clinical features will be compared between patients who achieved successful sperm retrieval (SSR) and those with sperm retrieval failure (SRF).

Methods

Patients

In this retrospective study, the electronic records of 10,170 men referred between 2018 and 2023 due to infertility to the Research Institute of Reproductive Sciences, Shahid Sadoqi University of Medical Sciences, Yazd, Iran, were collected. Among these, those who met the exclusion criteria were excluded from the study. Eventually, 566 cases, whose infertility was diagnosed as NOA, underwent micro-TESE, and met the inclusion criteria, were further investigated. Figure 1 shows a schematic diagram of our study design.

Figure 1.

Schematic diagram of the workflow for study design.

KS = Klinefelter syndrome; YCMDs = Y chromosome microdeletions; SSR = successful sperm retrieval; SSF = sperm retrieval failure.

Inclusion Criteria

Patients were included in the study if spermatozoa were not observed in at least two semen samples after centrifugation and screening using an inverted microscope, as well as whose clinical information was completed, which includes medical history, physical examination history, hormonal profile [follicle-stimulating hormone (FSH), luteinizing hormone (LH), testosterone (T), and prolactin (PRL)], genetic tests (karyotyping and Y Chromosome microdeletion analysis), measurement of testicular volume with scrotal ultrasound (even if the patient has only one testicle is included in the calculation). The presence of varicocele and any history of testicular cancer, congenital abnormalities (e.g., cryptorchidism), post-infectious (e.g., mumps orchitis), exposure to gonadotoxic (e.g., chemotherapy and radiotherapy) were considered.

Exclusion Criteria

OA patients with vasectomy, cystic fibrosis, ejaculatory duct obstruction, congenital sperm duct obstruction, anejaculation, bilateral hernia repair, complete AZFa or AZFb microdeletions, and those with incomplete records were excluded from the study.

Micro-TESE Procedure

Micro-TESE was performed under general anesthesia as described before (Gao et al., 2022). The procedure started in the larger testicle. After a transverse incision on the scrotum, the testis was delivered and the tunica albuginea was opened. Testicular parenchyma was observed directly at ×20–25 magnification to locate and examine the presence of spermatozoa. If no sperm was observed, the contralateral testis was thoroughly examined (Lan et al., 2022). Successful retrievals were defined as the detection of at least one sperm (Li et al., 2023).

Statistical Analysis

Statistical analysis was performed using GraphPad Prism version 10.2.3. Data were expressed as mean ± SD for numeric variables. Categorical variables were presented as proportions (%). The distribution of data was evaluated by the Shapiro-Wilk test. The group comparison of clinical data was done using Kruskal-Wallis analysis of variance (ANOVA) with post hoc contrasts by Dunn’s test. Fisher’s exact test was used for the associations between the etiology of NOA and the desired outcomes. The Mann-Whitney U test or independent sample t-test, depending on the normality of data distribution, was performed to compare SSR and SRF in each of the different clinical groups separately. In addition, relationships between rates of SSR and clinical characteristics were explored by Spearman correlation analysis. p < .05 was considered statistically significant.

Results

Clinical Characteristics of NOA Patients with Different Etiologies

Based on the clinical characteristics of the patients, 566 included patients undergone micro-TESE, had etiology histories of Cryptorchidism (n = 58, 10.2%), KS (n = 40, 7%), Y YCMDs with a history of chemotherapy (n = 6, 1%), patients with mumps orchitis (n = 5, 0.8%), and 451 (80%) patients with idiopathic reason (see Table 1).

Table 1.

Clinical Characteristics of Non-Obstructive Azoospermia Patients with Different Etiologies.

Characteristics	Whole cohort (n = 566)	Idiopathic (n = 451, 80%)	Cryptorchidism (n = 58, 10.2%)	KS (n = 40, 7%)	YCMDs (n = 6, 1%)	Chemotherapy (n = 6, 1%)	Mumps orchitis (n = 5, 0.8%)
Age (years)	34.06 ± 5.67	34.31 ± 5.76	33.32 ± 5.20	32.93 ± 0.76	29.86 ± 4.37	32.43 ± 5.38	36.80 ± 3.48
FSH (mIU/mL)	17.37 ± 15.72	16.73 ± 15.52	16.37 ± 11.63	25.43 ± 15.33*	17.29 ± 11.63	17.85 ± 15.42	17.94 ± 19.59
LH (mIU/mL)	10.36 ± 9.37	9.58 ± 8.79	11.57 ± 9.35	18.22 ± 12.99*	12.00 ± 6.58	10.53 ± 6.52	7.28 ± 7.10
T (ng/mL)	3.70 ± 3.36	3.80 ± 3.59	3.65 ± 1.87	2.49 ± 2.26*	3.43 ± 2.44	4.73 ± 2.49	3.86 ± 2.11
PRL (ng/mL)	12.85 ± 13.56	12.70 ± 10.64	11.05 ± 6.51	12.95 ± 10.01	8.71 ± 2.69	11.63 ± 4.29	11.37 ± 6.17
LTV (mL)	10.21 ± 4.92	10.87 ± 4.77	8.08 ± 4.15*	5.63 ± 4.17*	11.14 ± 4.22	9.71 ± 7.08	10.60 ± 4.33
RTV (mL)	10.24 ± 5.04	10.87 ± 4.74	8.15 ± 5.33*	5.19 ± 4.03*	11.43 ± 4.19	11.14 ± 8.09	11.40 ± 5.12

Note. The mean difference between groups was evaluated using Kruskal-Wallis analysis of variance (ANOVA) with post hoc contrasts by Dunn’s test. Values are presented as mean ± standard deviation. FSH = follicle-stimulating hormone; LH = luteinizing hormone; T = testosterone; PRL = prolactin; LTV = left testicular volume; RTV = right testicular volume; KS = Klinefelter syndrome; YCMDs = Y chromosome microdeletions.

^*Statistically significant.

The overall average age of the patients was 34.06 ± 5.67 (range of 20-58 years), and the levels of FSH (mIU/mL), LH (mIU/mL), T (ng/mL), and PRL (ng/mL) were 15.72 ± 17.37, 9.37 ± 10.36, 3.36 ± 3.70, and 13.56 ± 12.85, respectively. The volume of the left and right testis was measured as 4.92 ± 10.21 and 5.04 ± 10.21, respectively.

The serum levels of FSH (H = 14.89, p = .02), LH (H = 21.08, p = .001), T (H = 13.43, p = .03) and testicular volume on both right (H = 62.51, p < .0001) and left sides (H = 59.75, p < .0001) were significantly different among the study groups. Post hoc analysis revealed that the serum levels of FSH (p = .003) and LH (p = .001) were significantly higher in patients with KS compared to other groups. The least significant proportion belonged to T serum levels in the KS group (p = .008). Testicular volume on both right and left sides was lower in KS (p < .0001) and Cryptorchidism (p < .05) groups compared to other groups.

Comparison of Micro-TESE Results Among NOA Patients with Different Etiologies

The rate of SSR in different groups is shown in Table 2. Out of a total of 566 patients, 245 (43.2%) achieved SSR. The retrieval rate in different groups was as follows: Idiopathic (193/451, 43.3%), Cryptorchidism (34/58, 58.6), KS (9/40, 22.5%), YCMDs (3/6, 50%), Chemotherapy (3/6, 50%), and Mumps orchitis (3/5, 60%). Among these, a significantly lower SSR was seen in KS (p = .012). Also, 44% of infertile men with Cryptorchidism showed an SSR rate from the same side of their surgery.

Table 2.

Successful Sperm Retrieval of Non-Obstructive Azoospermia Patients with Different Etiologies.

Variables (n/%)	Whole cohort (n = 566)	Idiopathic (n = 451)	Cryptorchidism (n = 58)	KS (n = 40)	YCMDs (n = 6)	Chemotherapy (n = 6)	Mumps orchitis (n = 5)
SSR	245 (43.3%)	195 (43.2%)	34 (58.6%)	9 (22.5%)*	3 (50%)	3 (50%)	3 (60%)

Note. Fisher’s exact test was used for the associations between the etiology of non-obstructive azoospermia patients and the desired outcomes. Data were expressed as proportions (%). SSR = successful sperm retrieval; KS = Klinefelter syndrome; YCMDs = Y chromosome microdeletions.

^*Statistically significant.

The comparison of successful and unsuccessful sperm retrieval concerning clinical factors and different etiologies is given in Table 3. For SSR patients, the average age overall (U = 32,616, p = .0006) and in the idiopathic group (U = 19,504, p < .0001) was higher. There were statistically higher in FSH (t = 4.105, df = 5, p = .009) and LH (t = 2.714, df = 5, p = .04) in the YCMDs group. A significant higher was seen in T levels overall (U = 30,288, p < .0001), in the idiopathic (U = 20,391, p = .0009) and Cryptorchidism (U = 294, p = .04) groups. Regarding the KS group, there were significantly higher PRL levels (U = 54.50, p = .003). The volume of the right testicular was also higher overall (U = 35,505, p = .04). In the rest of the groups and clinical characteristics, no significant difference was observed between the SSR and SSF groups.

Table 3.

Predictors of Successful Sperm Retrieval for Non-Obstructive Azoospermia Patients with Different Etiologies.

SSR predictors	Whole cohort SSR (245)/SSF (321)	Idiopathic SSR (195)/SSF (256)	Cryptorchidism SSR (34)/SSF (24)	KS SSR (9)/SSF (31)	YCMDs SSR (3)/SSF (3)	Chemotherapy SSR (3)/SSF (3)	Mumps orchitis SSR (3)/SSF (2)
Age (years)	35.04 ± 5.89* 33.34 ± 5.39	35.51 ± 5.97* 33.40 ± 5.44	33.24 ± 5.07 33.44 ± 5.65	31.78 ± 3.76 33.25 ± 5.21	28.33 ± ± 3.05 31.00 ± 5.29	31.00 ± 6.08 33.50 ± 5.44	40.67 ± 7.02 31.00 ± 5.65
FSH (mIU/mL)	17.01 ± 16.20 17.65 ± 15.37	16.19 ± 15.45 17.14 ± 15.59	17.96 ± 11.92 14.21 ± 10.39	24.33 ± 16.48 25.73 ± 15.25	28.20 ± 7.58* 9.10 ± 4.55	6.82 ± 5.38 26.13 ± 15.59	25.27 ± 23.77 6.95 ± 1.48
LH (mIU/mL)	9.80 ± 8.61 10.78 ± 9.90	9.09 ± 8.01 9.96 ± 9.34	12.15 ± 10.45 10.79 ± 7.76	16.41 ± 11.18 18.73 ± 13.57	17.43 ± 6.15* 7.92 ± 3.13	9.06 ± 9.46 11.63 ± 4.65	9.26 ± 9.32 4.30 ± 1.27
T (ng/mL)	3.93 ± 1.92* 3.51 ± 4.12	3.93 ± 1.93* 3.71 ± 4.44	3.98 ± 1.57* 3.20 ± 2.17	3.40 ± 2.20 2.23 ± 2.25	5.83 ± 0.28 1.62 ± 1.32	3.60 ± 0.30 5.57 ± 3.18	4.76 ± 1.36 2.50 ± 2.82
PRL (ng/mL)	13.16 ± 15.59 12.62 ± 11.80	12.05 ± 6.66 13.20 ± 12.87	11.75 ± 8.57 10.64 ± 5.14	23.36 ± 14.34* 10.03 ± 5.99	9.66 ± 3.78 8.00 ± 1.82	9.76 ± 6.01 13.03 ± 2.59	11.22 ± 2.38 11.60 ± 11.88
LTV (mL)	10.48 ± 4.90 10.01 ± 4.93	11.08 ± 4.76 10.70 ± 4.78	7.97 ± 4.10 8.24 ± 4.30	5.66 ± 4.18 5.62 ± 4.24	9.00 ± 5.56 12.75 ± 2.63	13.33 ± 9.07 7.00 ± 4.76	9.67 ± 5.85 12.00 ± 0.00
RTV (mL)	10.73 ± 4.96* 9.86 ± 5.07	11.26 ± 4.62 10.58 ± 4.81	9.12 ± 5.65 6.84 ± 4.66	4.55 ± 2.29 5.37 ± 4.41	9.00 ± 5.56 13.25 ± 2.06	13.33 ± 9.07 9.50 ± 8.22	9.00 ± 5.56 15.00 ± 0.00

Note. The Mann-Whitney test was performed to compare SSR and SRF in different clinical groups. The relationships between clinical characteristics and rates of SSR were explored by Spearman’s correlation analysis. Values are presented as mean ± standard deviation. SSR = successful sperm retrieval; SSF = sperm retrieval failure; FSH = follicle-stimulating hormone; LH = luteinizing hormone; T = testosterone; PRL = prolactin; LTV = left testicular volume; RTV = right testicular volume; KS = Klinefelter syndrome; YCMDs = Y chromosome microdeletions.

^*Statistically significant.

Negligible positive relationships were found between age and rates of SSR overall (r = .14, p = .0006) and in the idiopathic group (r = .18, p < .0001). Similarly, weak positive relationships were observed between T levels and rates of SSR overall (r = .19, p < .0001), idiopathic (r = .15, p = .0008), and Cryptorchidism (r = .26, p = .4) groups. Regarding the KS group, positive relationships were found between PRL levels and rates of SSR (r = .45, p = .003), and negligible positive relationships were observed between RTV levels and rates of SSR overall (r = .08, p = .046).

Comparison Outcomes of Micro-TESE in Idiopathic NOA Patients

Due to the high number of idiopathic patients, these were briefly examined in terms of underlying factors that may affect azoospermia. Of the 451 idiopathic patients, 3% were exposed to heat/toxic environmental/occupational risks, 30% had various diseases such as Neurologic, Respiratory, Gl Tract, diabetes, renal, Endocrinopathy, liver, Heart, urological, Hematologic; 4% with history of Inguinal Hernia, 1% with Orchiectomy, 1% had trauma, 22% drug abuse, 22% history of Varicocele, and 17% Varicocelectomy.

Discussion

Many studies have been conducted so far, to provide a predictive factor on SSR in NOA patients (Caroppo & Colpi, 2021a). While some studies reported a potential predictive role, some found no association between clinical factors and etiologies of NOA with outcomes of micro-TESE (Chen et al., 2019; Deng et al., 2023; Gao et al., 2022; Zhang et al., 2021). It is difficult to propose a single factor as a predictor of micro-TESE outcome. The difference in race and population may be one of the effective factors in this conflict (Wellons et al., 2012). Since a similar study has not been conducted in the Iranian population, the present study aimed to investigate the potential role of clinical factors and different etiologies of NOA on SRR in an infertility center of the Iranian population.

In a review of 4,895 patients, SRR with micro-TESE of men with NOA was 46.6% and was mainly related to the characteristics of the treated population (Achermann et al., 2021). Similarly, our data reveal that the overall SRR with micro-TESE in NOA patients was 43.5%, though this rate varied by etiology, with the lowest significant rate in the KS group, and higher nonsignificantly rates observed in cryptorchidism and mumps orchitis patients.

KS was identified as the most prevalent chromosomal abnormality, accounting for 2,239 cases (75.5%). This condition is characterized by infertility, increased levels of FSH and LH, normal or reduced T level, and reduced testicular volume (Majzoub et al., 2022; Zhang et al., 2021). Our data showed the highest FSH and LH levels and the lowest T level and testicular volume in KS patients compared to other groups, which is following with the characteristics of KS patients. The elevated levels of FSH and LH serve as indicators of primary testicular failure (Ladjouze & Donaldson, 2019). The hypothalamic-pituitary response to low T levels triggers these increases (Xia et al., 2017). Reduced T levels and testicular volume are consistent with impaired Leydig cell function and spermatogenic failure (Xia et al., 2017), explaining the low SRR observed in KS cases. The higher level of PRL hormone in SSR-KS patients was contrary to expectations. Maintaining normal PRL levels is essential for male fertility, and abnormalities significantly impact sperm production and overall reproductive health (Spaggiari et al., 2023). In the current study, elevated PRL may reflect a compensatory response to the unique genetic and endocrine profile of KS patients (Mahdi et al., 2018), while psychological or social factors could further influence PRL levels (Fiala et al., 2021). Although PRL’s role in spermatogenesis remains unclear (Raut et al., 2019), its moderate correlation with SSR in KS may indicate its possible relevance in compensating for hormonal imbalances.

Regarding SSR patients, in a retrospective study, it was found that age did not affect the micro-TESE sperm recovery rate, although other studies showed that retrieval at a younger age can have better results (Gao et al., 2022). Conflicting results were shown regarding the relationship between testicular volume and SSR. Although some studies correlate larger testicular size with better outcomes, some suggest that sperm may be retrieved even in patients with testicular volumes of less than 2 mL (Caroppo & Colpi, 2021a; Mahdy et al., 2024). Our overall analysis of age and testicular volume showed a negligible positive relationship between age and rates of SSR as well as testicular volume and rates of SSR, highlighting that smaller testes can still harbor focal spermatogenesis. However, a slight increase and decrease in testicular volume were observed in the chemotherapy and YCMD/mumps orchitis groups, respectively. Due to the small number of statistics in these three groups, these results cannot be cited and interpreted.

It was found from 421 NOA patients in a retrospective study that the serum T level and FSH level did not seem to be significantly associated with micro-TESE outcomes (Althakafi et al., 2017). However, in another study of individuals with cryptorchidism, serum T levels in the successful group were significantly higher than the failed group (Saber-Khalaf et al., 2022). FSH and LH have moderate value in the independent prediction of SRR in men with NOA based on a systematic review study in 2021 (Zarezadeh et al., 2021). Similarly, our data indicates the T levels in SSR patients, particularly in cryptorchidism and idiopathic cases, were higher. Also, a nonsignificant decrease and increase in FSH and LH were observed in patients with SSR-chemotherapy and YCMDs/mumps orchitis groups, respectively. Due to the low number of statistics in these three groups, the obtained results cannot be relied upon and interpreted. Based on findings in other studies (Caroppo & Colpi, 2021a; Mahdy et al., 2024) along with ours, it seems that age, hormone levels, and testicular volume could not be reliably considered to predict SSR.

Due to the high number of idiopathic patients, these were briefly examined in terms of underlying factors that may affect azoospermia. In the idiopathic group, several possible factors emerge, including environmental and lifestyle risks, systemic health conditions, substance use, histories of surgical interventions, and specific genetic etiology. These findings align with studies linking such factors to impaired spermatogenesis and reduced SSR in idiopathic NOA (Caroppo & Colpib, 2021b). It should be noted that the mentioned factors are not considered as distinct etiologies of azoospermia, but rather as factors that may affect azoospermia along with other factors or indicate a specific condition that should be considered. However, a study of idiopathic NOA showed that a higher proportion of men in the idiopathic group had varicocele, compared to other groups. This could raise the possibility of varicocele as a definite cause of NOA (Alghobary & Mostafa, 2022; Biggs et al., 2023; Das et al., 2020; Jensen & Ko, 2021; Peña et al., 2020). Our results also confirm this finding. In the present study, a large number of individuals with varicocele were classified as idiopathic. Examining these exposures and medical histories in greater detail may uncover specific risk factors and improve the precision of diagnoses within this group.

This study has several limitations. First, the retrospective design may impact the quality of demographic and clinical data. Second, the relatively small sample size and the low number of samples in some groups could reduce the statistical power needed to detect intergroup differences. Additionally, limited clinical data restricts the ability to identify predictors for micro-TESE outcomes in NOA patients with varying etiologies. Future multicenter studies with larger cohorts and detailed assessments of lifestyle and genetic factors across diverse populations could improve the accuracy of predictive models for sperm recovery. Meta-analyses may provide insights by consolidating findings across different communities.

Conclusion

Our data underscores that etiology significantly influences SSR in NOA patients, like KS, with the lowest rates of micro-TESE outcome. It seems that the underlying etiology and genetic background may reveal more valuable predictive value rather than age, hormone levels, and testicular volume. This finding suggests that no patient with NOA should be deprived of micro-TESE based on candidate predictors of SSR presented to date.