Does varicocelectomy improve semen analysis outcomes in adolescents without testicular asymmetry?

Summary

Purpose

The main indications for adolescent varicocelectomy are testicular hypotrophy or pain. However, we have previously shown that both serial total testicular volume and volume differential are weakly associated with semen quality. The ultimate patient goal is paternity, but semen analysis is critical to appropriate management of varicocele. We hypothesize that varicocelectomy improves total motile count (TMC) among patients who only have abnormal semen analysis (SA) parameters, but not among those with potential hormonal dysfunction such as Klinefelter syndrome or cryptorchidism.

Methods

We retrospectively reviewed our registry of adolescent males followed with a clinical left varicocele. For this study, subjects without sustained testicular asymmetry, who were Tanner V, and gave at least one preoperative SA were included. Subjects were excluded if they had embolization for their varicocele or no postoperative SA. Primary outcome was change in TMC before and after surgery, compared using the Wilcoxon signed rank test after stratifying by surgical indication. Secondary outcomes included rates of improved TMC and normalized TMC (> 20 million) after surgery, compared across covariates using the Fisher exact test.

Results

Seventeen patients met the eligibility criteria, 11 of whom underwent repair for only abnormal preoperative TMC. Overall, median age (interquartile range [IQR]) at first preoperative SA was 17.6 (15.9–17.9) years. The median preoperative TMC across all SA was 2.8 (0.7–7.4) million. The median age at surgery was 18.2 (16.8–18.9) years. Postoperatively, the median TMC across all SA increased to 18.2 (3.6–18.2) million (Wilcoxon signed rank test, p < 0.01; see figure). The improvement in TMC occurred primarily in the group who only had abnormal preoperative TMC (82% improved, 55% normalized); lack of improvement was seen in patients who had a history of Klinefelter or orchiopexy for cryptorchidism.

Conclusions

Adolescent varicocele patients should undergo SA after development of Tanner V. Varicocelectomy has a high success rate for improving TMC in adolescent or young adult males who only have abnormal TMC and no history of cryptorchidism.

Introduction

Approximately 15% of adolescent males ages 15–19 years are expected to develop varicocele [1]. Of these, 85% with uncorrected varicoceles have been able to achieve paternity [2]. Traditionally, indications for surgical repair have been testicular hypotrophy or pain, but in otherwise asymptomatic adolescent males without testicular asymmetry, surgical repair and its timing remain controversial. These patients may be potentially at most risk for progressive damage to their testicular function, given their lack of signs and symptoms, and therefore require close monitoring.

Semen analysis (SA) has been advocated as a critical biomarker for assessing future paternity potential [3–5]. Calculation of total motile count (TMC) has been most commonly used with a cutoff of 20 million separating normal versus abnormal sperm quality [5, 6]. The effects of varicocelectomy on TMC and pregnancy outcomes have been well studied in adult males presenting with infertility [7–9], but more evidence on the treatment effect among adolescent males is needed.

Recently, a systematic review and meta-analysis found 10 published studies comprising 379 adolescent males treated for varicocele and 270 control subjects who could provide data on SA outcomes after varicocelectomy [10]. Overall, surgical repair improved various semen parameters, but indications for surgery were not addressed. As such, whether varicocelectomy improves SA outcomes specifically among asymptomatic adolescent males without testicular asymmetry remains unclear. We sought to address this question by comparing pre- and postoperative TMC among all Tanner V adolescent males referred to our institution, stratifying by indication for repair. We hypothesized that varicocelectomy would improve TMC among patients who only have abnormal TMC, but not among those with potential hormonal dysfunction such as Klinefelter syndrome or history of orchiopexy for cryptorchidism.

Materials and methods

Study design

After local institutional review board approval, we queried our registry of adolescent males referred to our institution for varicocele. This registry includes patients prospectively enrolled from January 2003 to September 2015 and older patients whose records were retrospectively entered.

Our general institutional varicocele clinical pathway has been previously described [11]. Briefly, an adolescent with a clinical varicocele was offered surgical repair at initial presentation if pain or obvious testicular asymmetry were present. Testicular volumes were formerly measured using scrotal ultrasonography, but a poor association was found between serial ultrasound-based testicular volumes and SA outcomes [11]. Our pathway was subsequently changed to orchidometer-based assessment for cost savings. Asymptomatic adolescent males with adequate total testicular volume and without significant testicular size discrepancy were followed annually with orchidometry until the patient reached the Tanner V stage of sexual development and age 15 years. At that point, an initial SA was recommended. Each SA was obtained after at least 2 day's abstinence and was repeated at least 3 months later if the initial one had abnormal TMC. Surgical repair was recommended after two consecutive poor SA results. Postoperative SA was obtained at least 3 months after repair. All SAs were analyzed according to standard WHO criteria [12].

Inclusion and exclusion criteria

The eligibility criteria included adolescent Tanner V males with varicocele who had insignificant testicular size discrepancy (defined as symmetric on palpation and orchidometry), who provided at least one preoperative SA, and who underwent repair. Since we sought to explore how varicocele repair outcomes varied by surgical indication, we included patients who underwent repair for various indications. Exclusion criteria included patients who did not complete a postoperative SA and those who underwent embolization with interventional radiology for their varicocele, as these patients may have different risk factors than others referred for surgical repair.

Outcome measures

The primary outcome was TMC, calculated as the product of total semen volume (mL) × percent normal motility × sperm density (million sperm/mL). Preoperative TMC was averaged across all SA provided per patient to obtain a mean preoperative TMC per patient. The mean postoperative TMC was similarly calculated per patient.

Secondary outcomes were the rate of improved TMC and rate of normalized TMC following surgical repair. If the mean postoperative TMC was greater than the mean preoperative TMC for a patient, then the TMC improved. For the normalization of TMC outcome, TMC was dichotomized into two groups: normal if at least 20 million, or abnormal if less than 20 million. This cutoff value has been generally accepted as an indicator of sperm quality [5,6].

Exposure measure and covariates

The primary exposure was indication for varicocele repair. These included abnormal preoperative TMC only, abnormal TMC with prior history of orchiopexy for cryptorchidism, pain/patient discomfort, and Klinefelter syndrome.

Covariates included age at initial referral, ages at each preoperative and postoperative SA, age at surgical repair, grade of varicocele, and type of repair. Repairs in the last decade have been performed via the microsurgical subinguinal approach; older repairs include the traditional open Palomo and Ivannisovich approaches. Follow-up was determined by the duration of time from the age at initial referral to the age at the last recorded postoperative SA.

Statistical analysis

Descriptive statistics with medians and interquartile ranges (IQR) were generated for all covariates. The mean preoperative and postoperative TMC values were calculated for each patient as described above. These primary outcomes were then stratified by indication for repair and because of non-parametric distribution were compared using the Wilcoxon signed rank test. Associations between covariates and secondary outcomes were tested with the Fisher exact test and the Wilcoxon rank sum test for categorical and continuous variables, respectively. No values were missing. All statistical tests were performed using Stata version 14.1 (StataCorp,College Station, TX, USA) with a two-tailed alpha of 0.05.

Results

Initial query of our varicocele registry generated 28 patients who met our inclusion criteria. After exclusion of three (11%) patients for embolization of their varicocele and eight (29%) patients for no postoperative SA, our final study cohort consisted of 17 patients. The median follow-up for this cohort from initial referral was 3.7 years (IQR 2.8–5.8).

Cohort characteristics are described in Table 1. The median age at surgical repair was 18.2 years (IQR 16.8–18.9). Most patients (11 of 17, 65%) underwent surgery for abnormal preoperative SA only, although two had abnormal SA with prior orchiopexy for cryptorchidism (1 with right orchiopexy at age 9, 1 with bilateral orchiopexies at age 3), two had pain that developed after the preoperative SA, one has Klinefelter syndrome, and one underwent repair for family preference despite a normal preoperative TMC.

Table 1.

Cohort characteristics.

Characteristic	n	Median (IQR)
Preoperative
Age at urology referral, years	17	14.4 (13.2–15.7)
Varicocele grade (%)
I	1 (6%)
II	5 (29%)
III	11 (65%)
Age at 1st preop SA, years	17	17.6 (15.9–17.9)
1st preop TMC, million	17	2.1 (0.5–10.6)
Age at 2nd preop SA, years	14	18.0 (17.1–18.5)
2nd preop TMC, million	14	3.2 (0.8–9.3)
Age at 3rd preop SA, years	3	17.6 (17.6–18.5)
3rd preop TMC, million	3	4.2 (0.9–12.9)
Perioperative
Age at surgery, years	17	18.2 (16.8–18.9)
Indication for repair (%)
Abnormal SA only	11 (65%)
Abnormal SA + UDT	2 (12%)
Pain	2 (12%)
Klinefelter syndrome	1 (6%)
Family preference	1 (6%)
Type of repair (%)
Microscopic subinguinal	15 (88%)
Ivanissevich	1 (6%)
Palomo	1 (6%)
Postoperative
Age at 1st postop SA, y	17	18.7 (17.9–19.4)
1st postop TMC, million	17	12.8 (3.6–34.5)
Age at 2nd postop SA, years	5	18.9 (17.9–19.0)
2nd postop TMC, million	5	23.6 (6.1–27.7)
Age at 3rd postop SA, years	1	19.6
3rd postop TMC, million	1	12.2

IQR = interquartile range; SA = semen analysis; TMC = total motile count; UDT = history of undescended testicle status after orchiopexy.

Table 2 demonstrates the changes in TMC before and after surgery with stratification by surgical indication. Overall, the preoperative TMC significantly increased from a median of 2.8 million (IQR 0.7–7.4) to 18.2 million (IQR 3.6–18.2) after surgical repair (Wilcoxon signed-rank test, p < 0.01). After stratification by surgical indication, significant improvement in TMC occurred primarily in those patients who had abnormal SA only (p = 0.01), with the other indication categories limited by small numbers.

Table 2.

Changes in total motile count by indication for surgical repair.

Indication	N	Mean preop TMC	Mean postop TMC	p *	N (%) with improved TMC	N (%) with Postop TMC≥ 20
Overall	17	2.8 (0.7–7.4)	18.2 (3.6–18.2)	< 0.01	12 (71%)	8 (47%)
Abnormal SA only	11	2.4 (0.7–6.1)	23.4 (3.6–34.5)	0.01	9 (82%)	6 (55%)
Abnormal SA + UDT	2	2.1 (0.6–3.6)	5.2 (3.1–7.3)	0.65	1 (50%)	0 (0%)
Pain	2	50.9 (46.3–55.5)	84.2 (59.7–108.7)	0.18	2 (100%)	2 (100%)
Klinefelter	1	0	0	–	0 (0%)	0 (%)
Family preference	1	30.4	17.2	–	0 (0%)	0 (%)

Note. All values are median (interquartile range) unless otherwise noted. SA = semen analysis; TMC = total motile count, in millions; UDT = history of undescended testicle status after orchiopexy.

^*Wilcoxon signed rank test between preop TMC and postop TMC.

Figure 1 shows a scatterplot of mean preoperative versus mean postoperative TMC per patient, with color stratification by surgical indication. Red reference lines mark the 20 million cutoff for TMC. The blue diagonal reference line marks no change in TMC after surgery. Thus, markers above the blue diagonal line demonstrate any increase in TMC after surgery. Similarly, markers located in the left upper quadrant of the red reference lines correspond to normalization of TMC from < 20 million before surgery to ≥ 20 million after surgery.

Figure 1.

Scatterplot of preoperative and postoperative mean total motile counts, stratified by surgical indication.

Secondary outcomes of TMC improvement and TMC normalization following surgery were not significantly different across varicocele grade, repair type, and indication for repair (Fisher exact test, all p > 0.05, not shown). Age at surgery was not significantly different among patients whose TMC improved or did not improve, and among patients whose TMC normalized and did not normalize (Wilcoxon rank-sum test, all p > 0.05, not shown).

Discussion

Among adolescent Tanner V males with an asymptomatic left varicocele, no history of cryptorchidism, and no significant testicular asymmetry, varicocelectomy significantly improved TMC in 82% of patients, with normalization of TMC in 55% of patients. Varicocelectomy did not appear as effective in males with a history of cryptorchidism requiring orchiopexy. The numbers of our study are small, but the results demonstrate the potential benefit surgical intervention may have on specific subpopulations of at-risk adolescent or young males.

The effectiveness of varicocelectomy in improving semen parameters and, ultimately, pregnancy rates has been controversial and often depends on the population being studied. Even among adult men who present with infertility, who have been the best-studied group of patients, the benefit of varicocele repair has not been clear. Some systematic reviews, meta-analyses, and guidelines cast doubt on the treatment effect of varicocele repair [13,14]. Others, however, have shown a significant benefit on both semen parameters [7,15] and pregnancy rates [9]. A joint best practice policy statement from the American Urological Association (AUA) and the American Society of Reproductive Medicine (ASRM) on varicoceles and infertility recommends that repair be performed in infertile males because of the potentially greater benefits that outweigh the low perioperative risks [16].

Given the controversy of varicocelectomy among infertile adult males, the debate on varicocelectomy among adolescent males is even greater and mostly centers on indications for repair. Pregnancy rates are almost impossible to assess as an outcome, and, as such, surrogate markers for pregnancy become vitally important. Among these, TMC on a SA has been considered one of the best surrogates for future paternity [6]. The ethical dilemma, and potentially key barrier, in using TMC as an outcome is that this requires obtaining SA from adolescents. This dilemma has both severely limited the utilization rate of SA and impaired discussion of obtaining SA among practitioners, parents, and patients [17]. In a survey study with 168 respondents out of 315 pediatric urologists, only 13% routinely used SA for management of adolescent varicoceles, with 53% of respondents never asking for an SA [17]. Indeed, the AUA/ASRM best-practice policy for adolescent males with varicoceles currently considers SA optional, with the recommendation to perform annual surveillance with “objective measurements of testis size and/or semen analyses” [16]. However, we have previously shown that serial measurements of testis size do not correlate well with SA outcomes [11]. Hence, at our institution, we address the ethical issues of obtaining a semen sample from a pubescent male by requiring the patient to be Tanner V and at least 15 years old before SA is recommended.

Within this context of potential underutilization of SA among adolescent males with varicocele, surgical repair has shown success. A recent systematic review and meta-analysis of studies focused on adolescent and young adults found a significant improvement in semen parameters following surgical repair [10]. However, the studies included in the meta-analysis were highly heterogeneous and included patients who had other indications for repair besides abnormal semen parameters, such as pain or testicular hypotrophy. Importantly, the meta-analysis did not stratify results by surgical indication. Our study thus attempted to fill in this knowledge gap with specific focus on males with abnormal TMC as the only indication for surgery. In this particular subpopulation, our results showed profound improvement in SA outcomes after surgical repair. Over 80% of patients showed an increase in TMC following varicocelectomy, with over 50% demonstrating normalization of TMC to at least 20 million. Additionally, our results suggest that patients with potential disruption of the normal endocrine pathway, such as with cryptorchidism, may not benefit as much from varicocele repair, although our sample size is too limited to make definitive conclusions. These results may help with risk stratification and preoperative counseling.

The results of our study must be interpreted within its limitations. First, our sample size may be too small to preclude selection bias and too limited to assess predictors of TMC improvement or normalization on a multivariable analysis. However, our inclusion criteria specified only patients who did not have testicular asymmetry and who had at least one preoperative TMC, which restricts the population undergoing surgery. Frequently, including at our institution historically, children undergoing repair for significant testicular asymmetry do not have preoperative SA. Regardless, our results must be considered exploratory and must be repeated in a larger cohort to be reliable. Additionally, given the aforementioned barriers to obtaining preoperative and postoperative SA among adolescents, we believe our results provide at the very least preliminary data for future, multi-institutional studies. Second, most of the patients underwent subinguinal microscopic varicocelectomy, which prevents comparison of success rates by surgical technique. A systematic review of varicocelectomy techniques, however, suggested that the subinguinal microscopic technique had the highest pregnancy rates and the lowest complication rates [18]. As such, our institution has more recently performed repairs this way only. Lastly, our results cannot be extrapolated to pregnancy rates. Even though TMC has been considered an excellent surrogate for fertility potential [6], our cohort must be followed for at least another 10 years to address issues with paternity-proven fertility.

In summary, we have shown that among adolescent Tanner V males with asymptomatic left varicocele, testicular symmetry, and at least one abnormal SA, varicocelectomy offers an excellent chance at improving TMC. Future multi-institutional longitudinal studies could increase the sample size to study the effectiveness of varicocele on pregnancy outcomes.

Figure.

Change in total motile count after varicocelectomy by patient.

Abbreviations

ASRM

American Society of Reproductive Medicine

AUA

American Urological Association

IQR

interquartile range

SA

semen analysis

TMC

total motile count