Outcomes following varicocele treatment: a comparative study of microsurgical and angioembolization approaches

Abstract

The aim of this study was to compare outcomes of microsurgical varicocelectomy (MV) versus percutaneous angioembolization (AE) in men with varicoceles causing infertility or scrotal pain. Endpoints included postoperative changes in semen parameters, complication rates, pain relief, need for additional interventions, and the influence of body mass index (BMI). We retrospectively reviewed 264 male patients treated for varicocele from July 2015 to July 2024 (MV: 214 patients, and AE: 50 patients). Only patients older than 18 years were included. Indications were infertility or chronic varicocele-related scrotal pain. Pre- and post-treatment semen analyses were compared in infertile men, and pain outcomes were assessed in those treated for pain. Complications and secondary interventions were recorded, and outcomes were stratified by BMI. MV significantly improved sperm concentration, motility, and total motile count within 3–6 months, while AE showed no significant change. Complications occurred only in the MV group (8.9%), including hydrocele, hematoma, and surgical site infection, while no complications were reported in the AE group. Persistent pain was observed in one-third of patients in both groups, and AE was associated with a higher rate of secondary intervention (14.0%) compared to MV (6.6%). BMI was not linked to complications or operative time but was associated with poorer baseline semen quality. Within 3–6 months post-treatment, varicocelectomy yielded greater improvement in semen quality, while AE offered a favorable safety profile but less reproductive benefit and a higher retreatment rate. Both approaches were safe across BMI categories, though longer-term studies are needed to clarify patient selection and delayed outcomes.

INTRODUCTION

Varicocele is a common vascular anomaly of the pampiniform plexus that is implicated in male infertility. It is found in over 15% of the general male population and up to 35% of men presenting with infertility.1 Not all varicoceles cause reproductive harm, but in many subfertile men, the varicocele contributes to impaired spermatogenesis. The rationale for treating a varicocele in adults is to improve fertility potential, as repair can restore or enhance semen quality and increase the chance of natural conception or successful assisted reproduction. Recent evidence confirms that varicocele treatment yields significant improvements in semen parameters and pregnancy rates when patients are appropriately selected.2,3

Several treatment approaches exist for varicocele. The two primary modalities are surgical varicocelectomy and percutaneous angioembolization.1 Different surgical techniques can be performed (retroperitoneal, inguinal, subinguinal, and high ligation), among which the subinguinal microsurgical approach is widely regarded as the gold standard due to its high success rates and low recurrence rates.3 Nonetheless, both techniques are effective in obliterating varicosities and improving hemodynamics, and each has dedicated proponents in the urologic and radiologic communities.

The relative merits of microsurgical varicocelectomy (MV) versus percutaneous angioembolization (AE) remain an important clinical question. Prior comparative studies have suggested that both approaches yield broadly similar outcomes in terms of semen quality improvement and fertility.4,5 Existing studies have been limited and have not deeply explored certain patient factors, for instance, the impact of body mass index (BMI) on varicocele presentation and treatment outcome.

Our objective was to compare outcomes of subinguinal MV versus percutaneous AE for varicocele treatment. The primary endpoints were difference in operative time, postoperative improvements in semen parameters, resolution of varicocele-related scrotal pain, need for second intervention, and postoperative complication rates. Secondary endpoints involved analyzing associations between obesity and procedure-related complication rates, operative time, and baseline semen parameters.

PATIENTS AND METHODS

Study design and patient selection

This study was a retrospective comparative analysis of men who underwent varicocele treatment at Cleveland Clinic Abu Dhabi, Abu Dhabi, United Arab Emirates between July 2015 and July 2024. After obtaining approval from the Institutional Review Board (IRB) of Cleveland Clinic Abu Dhabi (Approval No. T-052), we identified all patients who received either subinguinal MV or image-guided AE during the study period. Inclusion criteria were male patients aged >18 years with a palpable varicocele who met established clinical indications for treatment, including (1) documented infertility with abnormal semen parameters in couples of which the female partner has normal or treatable fertility, (2) abnormal semen parameters in patients not currently attempting to conceive, and (3) chronic testicular pain consistent with a varicocele. Exclusion criteria included patients who were lost to follow-up immediately after the procedure (i.e., no postoperative data available). The choice between MV and AE was based on patient preference following detailed counseling about both procedures. In cases of recurrence, a different approach from the initial treatment was typically recommended to maximize success rates. In keeping with the retrospective design, the requirement for informed consent was waived by the IRB. All data were de-identified, and institutional permission to publish the findings was obtained.

Intervention details

All varicocelectomies were performed by three urologists (KS, AB, and OR) using a standard subinguinal microsurgical approach. The procedure was repeated on the contralateral side in bilateral cases during the same anesthetic session. Spermatic cord denervation was not performed in patients undergoing MV, even in those presenting primarily with scrotal pain. Gubernacular veins and external spermatic vein were ligated when the testis was delivered, typically in redo procedures for recurrence (14 patients, 6.5%). The testis was not routinely delivered in primary cases. In all patients, lymphatic channels were meticulously preserved to minimize hydrocele formation.

Two interventional radiologists performed AE procedures in an angiography suite. Under local anesthesia and intravenous sedation, a catheter was introduced (typically via a jugular or femoral vein access) and guided to the internal spermatic vein under fluoroscopic guidance. Varicocele embolization was achieved using coil embolization of the gonadal vein (with sclerosant agents) until venous flow to the pampiniform plexus was occluded. Gonadal shielding was routinely used. All patients undergoing MV and AE were observed briefly after the procedure and typically discharged the same day.

Baseline and follow-up assessments

Patient demographics and clinical presentation details were recorded from the medical records. We noted whether the primary indication for treatment was infertility (defined as failure to conceive after ≥12 months of trying, including both primary and secondary infertility) or scrotal pain (varicocele-associated discomfort). The laterality of the varicocele (unilateral vs bilateral) and varicocele grade (when documented) were also recorded. Preoperative semen analysis results were collected for patients who presented with infertility or as part of their initial workup. Diagnosis of varicocele was primarily clinical, based on scrotal examination performed in both standing and supine positions with the Valsalva maneuver. Doppler ultrasound was selectively used to confirm equivocal findings, applying the standard criterion of reflux lasting >2 s. Semen parameters of interest included sperm concentration (10⁶ ml⁻¹), total motile sperm count (TMSC; 10⁶), percent of progressive motility (%), sperm morphology (percentage of normal forms ; %), ejaculate volume (ml), and pH. Postoperative semen analyses were obtained at the discretion of the treating physician, typically around 3–6 months after the procedure for infertility patients. In patients whose primary presentation was pain, semen analyses were not mandatory unless they were also assessing fertility.

Outcome measures

The primary outcomes for comparison between the MV and AE groups were difference in operative time, postoperative improvements in semen parameters, resolution of varicocele-related scrotal pain, need for second intervention, and postoperative complication rates. Complications were defined as any undesirable postoperative events attributable to the procedure. An overall complication rate (any complication vs none) was calculated for each group. Testicular atrophy was analyzed separately as a distinct adverse outcome. Improvement in semen parameters was evaluated in the subset of patients with available pre- and post-varicocele repair semen analyses. Pain outcome was evaluated in the subset of patients who reported scrotal pain pre-treatment, by noting the proportion with persistent pain at the last follow-up. The associations between patient BMI and procedure-related complication rates, operative times, and baseline semen parameters were analyzed as secondary outcomes.

Statistical analyses

All data were compiled and analyzed using SPSS Statistics (version 29; IBM Corp., Armonk, NY, USA). Continuous variables were checked for normality using the Shapiro–Wilk test. Because several key continuous variables (such as sperm count) showed non-normal distributions, non-parametric statistical tests were employed. The Mann–Whitney U test was used to compare continuous outcomes between the two treatment groups (MV vs AE), such as preoperative semen parameters, and also to compare semen parameters between BMI categories (normal-weight vs overweight and obese). Wilcoxon signed-rank tests were used for paired comparisons of pre- vs post-treatment semen values within the same patient group. For descriptive purposes, continuous data are presented as mean ± standard deviation (s.d.), and categorical data are presented as frequencies and percentages. Categorical variables were compared using the chi-square test (or Fisher’s exact test when appropriate). To measure the magnitude of differences, effect sizes were calculated where relevant: for non-parametric comparisons, we report the rank biserial correlation coefficient r, and for certain comparisons of means, we report Cohen’s d. A two-tailed P < 0.05 was considered statistically significant for all analyses.

RESULTS

Clinical characteristics

After excluding 17 patients who were lost to follow-up (4 from the MV group and 13 from the AE group), a total of 214 men underwent MV and 50 underwent AE and were included in the final analysis. The age at surgery (mean ± s.d.) was 28.7 ± 9.3 years in the MV group and 31.5 ± 11.1 years in the AE group. BMI values (mean ± s.d.) were 27.0 ± 6.1 kg m⁻² and 26.7 ± 6.4 kg m⁻², respectively (Table 1). A total of 99 (46.3%) patients in the MV group and 20 (40.0%) patients in the AE group were treated for infertility (including primary and secondary types), while the remaining 115 (53.7%) patients in the MV group and 30 (60.0%) patients in the AE group primarily presented with scrotal pain.

Table 1.

Baseline demographic and clinical characteristics of patients undergoing microscopic varicocelectomy versus percutaneous angioembolization for varicocele treatment

Variable	Microscopic varicocelectomy (n=217)	Percutaneous angioembolization (n=63)
Age (year), mean±s.d.	28.7±9.3	31.5±11.1
BMI (kg m−2), mean±s.d.	27.0±6.1	26.7±6.4
Smoker, n (%)	85 (39.2)	21 (33.3)
Alcoholic, n (%)	4 (1.8)	2 (3.2)
Diabetes (n)	2	2
Hypertension (n)	3	0
CAD (n)	0	1
Stroke (n)	0	1
BMI class, n(%)
Underweight	13 (6.0)	4 (6.4)
Healthy weight	59 (27.2)	14 (22.2)
Overweight	85 (39.2)	27 (42.8)
Obese I	40 (18.4)	13 (20.6)
Obese II	14 (6.4)	3 (4.8)
Obese III	6 (2.8)	2 (3.2)
Varicocele grade, n (%)
Subclinical	12 (5.5)	2 (3.2)
Grade 1	32 (14.8)	20 (31.8)
Grade 2	79 (36.4)	18 (28.6)
Grade 3	94 (43.3)	23 (36.5)
Laterality, n (%)
Left	155 (71.4)	45 (71.4)
Right	6 (2.8)	0 (0)
Bilateral	56 (25.8)	18 (28.6)

This table includes all the patients before excluding 17 patients who were lost to follow-up (4 from the microscopic varicocelectomy group and 13 from the percutaneous angioembolization group). s.d.: standard deviation; BMI: body mass index; CAD: coronary artery disease

Operative time and procedural parameters

Operative time was longer in the MV group overall (mean ± s.d.: 83.4 ± 29.3 min) compared to the AE group (mean ± s.d.: 72.0 ± 24.8 min). When stratified by laterality, mean operative time (mean ± s.d.) for unilateral MV was 77.7 ± 24.8 min versus 69.9 ± 22.6 min for unilateral AE. For bilateral cases, MV required significantly more time than AE (mean ± s.d.: 101.8 ± 33.8 min vs 78.2 ± 30.3 min). For AE, the average fluoroscopy time was 22.2 min, with a mean radiation dose of 65.5 mGy. The relatively prolonged procedural time (mean: 70.4 min) primarily reflected anesthesia preparation, vascular access, and coil selection, rather than technical difficulty during embolization.

Complications and testicular atrophy

The overall complication rate was 8.9% (19/214) in the MV group. The most common complications were hydrocele (9 patients, 4.2%) and hematoma (9 patients, 4.2%), followed by a single case of surgical site infection (0.5%). No complications (0/50) were reported in the AE group. Testicular atrophy was observed in 2 (0.9%) patients in the MV group and in none of the AE patients. Given the clinical significance of this complication, the two cases of atrophy are described in detail. The first patient presented on postoperative day 4 with acute inguinoscrotal hematoma and absent intratesticular flow on Doppler ultrasound; intraoperative findings showed spontaneous testicular rupture with congestion, and orchiectomy was performed. The second patient developed an inguinoscrotal hematoma immediately postoperatively due to active arterial bleeding, which was repaired; however, he subsequently developed surgical site infection, and repeat exploration revealed non-viable testicular tissue necessitating orchiectomy. In both cases, the deferential artery was preserved intraoperatively, but thrombosis and secondary vascular compromise could not be reversed.

Pain and second procedures

Among those with preoperative scrotal pain, postoperative pain persisted in 32.6% (43/132) of MV patients and 39.4% (13/33) in AE patients. Among patients who underwent MV, 14 (6.5%) patients later required a second intervention. Of these, 7 (3.3%) underwent AE and 7 (3.3%) underwent repeat MV. In the AE group, 7 (14.0%) patients subsequently underwent MV due to unresolved or persistent symptoms following initial angioembolization.

BMI-based outcomes

BMI was not significantly associated with overall complication risk. Complications occurred in 10.3% (10/97) of normal-BMI patients and 5.4% (9/167) of overweight and obese patients (χ² [1] = 2.224, P = 0.1; Fisher’s exact P = 0.1). However, higher BMI was strongly associated with infertility: 53.0% of overweight/obese patients were infertile at baseline compared to 26.8% of normal-BMI men (χ²[1] = 17.1, P < 0.001). Operative duration (mean ± s.d.) did not differ significantly by BMI group (normal-BMI: 80.5 ± 27.7 min vs overweight and obese 81.7 ± 29.5 min; P = 0.8; Cohen’s d = –0.04).

Semen quality outcomes

Overweight/obese men had significantly poorer preoperative semen parameters. They demonstrated lower total TMSC (Z = –2.429, P = 0.02, r = 0.178), sperm concentration (Z = –2.709, P = 0.007, r = 0.198), progressive motility (Z = –2.900, P = 0.004, r = 0.209), and semen volume (Z = –2.176, P = 0.03, r = 0.155). Semen pH and morphology were unaffected. Following microscopic varicocelectomy, patients exhibited statistically significant improvements in sperm concentration (Z = –4.17, P < 0.001), progressive motility (Z = –3.65, P < 0.001), and TMSC (Z = –2.52, P = 0.01), as shown in Figure 1. In the AE group, no significant postoperative improvements were observed: sperm concentration (P = 0.445), progressive motility (P = 0.262), TMSC (P = 0.600), semen volume (P = 0.959), pH (P = 0.276), and morphology (P = 0.514), as shown in Figure 2.

Figure 1.

Changes in semen parameters pre- and post-microscopic varicocelectomy. TMSC: total motile sperm count.

Figure 2.

Changes in semen parameters pre- and post-angioembolization. TMSC: total motile sperm count.

DISCUSSION

This comparative analysis of subinguinal MV and image-guided AE provides important insight into the relative strengths and limitations of each modality. Microsurgical ligation under magnification allows meticulous identification and ligation of all offending veins while sparing the testicular artery and lymphatics, leading to minimal risk of hydrocele or testicular damage. In contrast, radiographic occlusion of the internal spermatic veins (AE) offers a minimally invasive alternative that avoids a surgical incision.4,5 Endovascular embolization is particularly popular in some centers because it can be done under local anesthesia with quick recovery, but historically, it has been associated with slightly higher failure or recurrence rates than microsurgical surgery.4,5 For example, AE techniques have reported varicocele recurrence in approximately 3%–11% of cases, compared to recurrence rates of around 2% with MV.1,4

Our results showed that only MV led to statistically significant improvements in semen parameters postoperatively, while AE failed to demonstrate such gains despite its role in symptom resolution and minimally invasive appeal. These findings are based on semen analyses obtained approximately 3–6 months after the intervention. It is important to acknowledge that semen quality may continue to improve beyond this time frame, particularly after surgical repair, with a previous study suggesting that optimal sperm recovery can take up to 6–12 months following varicocelectomy.3

Kumar and Gupta1 reported statistically significant improvements in sperm count, normal motility, and normal morphology in 71%, 60%, and 49% of patients, respectively, following MV, with a spontaneous pregnancy rate of 34%. Goldstein2 emphasized that subinguinal microsurgical techniques allow artery- and lymphatic-sparing dissection, minimizing hydrocele formation and optimizing fertility outcomes.

In contrast, AE was not associated with a statistically significant improvement in semen quality in our cohort. While the procedure benefits from shorter operative time and minimal invasiveness, these advantages appear to come at the cost of reduced reproductive efficacy. Bou Nasr et al.,3 however, presented a more optimistic picture of AE, showing comparable improvements in semen parameters and pregnancy rates to MV, though their study evaluated outcomes over a longer follow-up period (up to 12 months). This raises the possibility that AE may also yield delayed benefits that were not captured within the 3–6-month assessment window in our study.

The absence of complications in the AE group observed in our cohort is consistent with prior literature suggesting a superior safety profile of embolization. Cassidy et al.5 reported no risk of hydrocele formation with AE, contrasting with the 8.24% hydrocele rate seen with surgical approaches. In our data, hydrocele formation occurred in 4.2% of MV patients, lower but consistent with the rates reported in surgical literature. However, AE carries unique risks such as technical failure and coil migration. Although not observed in our series, rare but severe complications of AE have been described, including bowel ischemia and thromboembolic events leading to lower limb ischemia.6,7

Our findings also reveal an important association between BMI and fertility outcomes. Overweight and obese men had significantly poorer baseline semen quality, specifically lower total motile sperm count, sperm concentration, progressive motility, and semen volume. These findings are consistent with evidence from Palmer et al.,8 who reported that obesity is associated with hormonal imbalances, increased oxidative stress, and impaired spermatogenesis. Similar demographic distributions have also been reported in other varicocele cohorts.9 The relatively high proportion of overweight and obese patients in our series reflects the broader population trend in the United Arab Emirates, where around 40% of men are overweight, and 26% are obese. In other words, nearly two-thirds of adult males in the country are living with excess weight.10

Our study did not identify a statistically significant increase in complication rates or operative time among overweight and obese patients, indicating that both MV and AE are technically feasible and safe across BMI categories. However, the consistently poorer baseline semen parameters observed in overweight and obese men highlight the need to interpret postoperative semen outcomes in this subgroup with consideration of their underlying reproductive impairment.

Persistent pain was observed in approximately one-third of patients in both groups postoperatively, with a higher proportion in the AE group. This underscores the variable efficacy of varicocele treatment in resolving scrotal discomfort, particularly with embolization. As noted by Goldstein,2 successful pain resolution depends on appropriate patient selection and the exclusion of alternative sources of orchialgia.

Moreover, reintervention was more frequently required in the AE group, with 14.0% eventually undergoing MV after failed embolization. This emphasizes the need for thorough counseling about the potential need for a second procedure, particularly in patients with bilateral varicoceles or complex venous anatomy. Recurrence in our cohort was assessed clinically during follow-up visits using standard physical examination techniques, including supine and standing assessment with Valsalva maneuver. Doppler ultrasound was not routinely used, reflecting common real-world practice, although this approach may underestimate subclinical recurrence. In particularly challenging cases where venous dissection is not feasible or recurrence risk is high, alternative intraoperative strategies such as internal spermatic–superficial epigastric vein bypass have been described.11 While not part of our practice, these techniques may be useful adjuncts in selected scenarios.

This study has several limitations. Its retrospective, single-center design limits generalizability, and the non-randomized treatment allocation introduces potential selection bias. Follow-up was limited to 3–6 months, which may not capture long-term outcomes such as recurrence or delayed improvements in semen quality. The smaller sample size in the AE group may also reduce comparative statistical power. Further studies are required to assess long-term outcomes, including fertility and hormonal effects, which were not captured in the present analysis.

CONCLUSION

Our study indicates that subinguinal MV may be more effective than AE in improving semen parameters within a 3–6-month postoperative period. While AE offers faster recovery and fewer complications, it did not yield statistically significant semen improvements in the short term. Given that prior studies report delayed reproductive gains, longer follow-up may reveal additional benefits for both procedures. MV remains the preferred option for men seeking definitive fertility improvement, particularly in bilateral cases, while AE may be considered in selected patients prioritizing minimally invasive care. Obesity was linked to poorer baseline semen quality but not increased complication risk or longer operative time. Further studies with extended follow-up are needed to better characterize long-term outcomes.

AUTHOR CONTRIBUTIONS

AAH conceived and designed the study, oversaw data collection and statistical analysis, and drafted the initial manuscript. O Almidani assisted in study design, participated in data collection, and contributed to manuscript drafting and revisions. FY, RR, AA, and O Aljasmi assisted in data collection and contributed to drafting and revising the manuscript. AM, KS, and AB contributed to manuscript drafting and revision. OR conceived the study, supervised the project, contributed to data interpretation, and participated in drafting and revising the manuscript. All authors read and approved the final manuscript.

COMPETING INTERESTS

The author Omer Raheem is a member of the Editorial Board of Asian Journal of Andrology, but he has not been involved in the peer-review or decision-making process for this manuscript. Other authors declare no competing interests.