Older Age Is Associated With Similar Improvements in Semen Parameters and Testosterone After Subinguinal Microsurgical Varicocelectomy Wayland Hsiao, James S. Rosoff, Joseph R. Pale, Eleni A. Greenwood and Marc Goldstein* From the Center for Male Reproductive Medicine, Department of Urology (WH, JSR, JRP, EAG, MG), and Institute for Reproductive Medicine (WH, MG), Weill Cornell Medical College and The Population Council Center for Biomedical Research (WH, MG), New York, New York Abbreviations and Acronyms TMC total motile sperm count Submitted for publication June 25, 2010. Study received institutional review board approval. Supported by a grant from The Frederick J. and Theresa Dow Wallace Fund of New York Community Trust (Hsiao). * Correspondence: 525 E. 68th St., Starr 900, New York, New York 10065. Purpose: It is generally accepted that men with clinically palpable varicocele are at high risk for a progressive decrease in fertility and testosterone levels with time. Varicocelectomy is thought to improve testicular function or at least halt the accelerated decrease in testicular function associated with varicocele. Substantial controversy exists as to whether varicocelectomy is effective in older men, possibly due to irreversible testicular damage or limited potential for recovery from varicocele induced damage. Materials and Methods: We retrospectively reviewed the records of men who underwent microsurgical subinguinal varicocelectomy, as done by a single surgeon. Demographics, patient questionnaires, operative notes, charts, testosterone and semen analysis were reviewed. Patients were divided into 3 groups based on age at surgery, including less than 30, 30 to 39 and 40 years or greater. Results: A total of 272 men met study inclusion criteria. In all 3 age groups we noted similar testosterone and baseline semen analysis parameters. There were significant increases in sperm concentration and total sperm count in all age groups. When analysis was restricted to men with baseline testosterone 400 ng/dl or less, there was a mean 110, 133 and 136 ng/dl increase in 21 men who were 40 years old or older, in 30 who were 30 to 39 years old and in 21 who were younger than 30 years, respectively. Conclusions: Microsurgical varicocelectomy resulted in significant increases in sperm concentration, total sperm count and testosterone in all age groups studied, including men in the fifth and sixth decades of life. Microsurgical varicocelectomy should be offered to older men for infertility and/or hypogonadism. Key Words: testis; varicocele; infertility, male; hypogonadism; aging VARICOCELES generally appear during or shortly after puberty and are found in 15% of men with up to 35% with primary infertility and 70% to 81% with secondary infertility presenting with varicocele. 1 It is generally accepted that men with varicocele are at risk for an accelerated, progressive decrease in fertility and testosterone with time if left unrepaired. 1 9 Varicocelectomy is thought to improve testicular function or at least halt the gradual decrease in testicular function associated with varicocele. Many studies show that varicocelectomy improves semen parameters, hormonal profiles and pregnancy rates. 1,5,10 12 Most of these studies were not controlled, randomized or stratified by patient age. 620 www.jurology.com 0022-5347/11/1852-0620/0 Vol. 185, 620-625, February 2011 THE JOURNAL OF UROLOGY Printed in U.S.A. 2011 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. DOI:10.1016/j.juro.2010.09.114
OLDER AGE AND SEMEN PARAMETERS AFTER VARICOCELECTOMY 621 Controversy exists as to whether varicocelectomy is as effective in older men due to irreversible testicular damage by the longstanding varicocele or limited potential for recovery from varicocele induced damage in older testes. The clinical implication is that if varicocelectomy is less effective in older men, perhaps it should not be offered with men electing androgen replacement and assisted reproduction instead. There are few studies of the effect of age at surgery on varicocelectomy outcomes. A small study by Ishikawa et al showed no significant difference in the response to varicocelectomy with respect to age. 13 Zini et al found similar improvements in semen analysis parameters in older men after varicocelectomy when a 40-year-old cutoff was used. 14 We tested the hypothesis that varicocelectomy is less effective in older men for treating infertility and/or male hypogonadism. We retrospectively reviewed the records of patients who underwent subinguinal microsurgical varicocelectomy at a tertiary referral center, and reviewed testosterone and semen analysis results. MATERIALS AND METHODS Patients and Evaluation After obtaining institutional review board approval we retrospectively reviewed the records of patients who underwent subinguinal microsurgical varicocelectomy, as done by a single surgeon from January 1996 to January 2009. All patients were referred for infertility or male hypogonadism (low testosterone) and all were older than 16 years. In our practice indications for varicocelectomy include clinically palpable varicocele with semen parameter abnormalities, clinically palpable varicoceles associated with pain, clinically palpable varicocele associated with infertility without genetic abnormality, grade 3 varicocele associated with testicular atrophy and total testosterone less than 400 ng/dl on 2 morning laboratory draws. Only men with preoperative and postoperative semen analyses, and/or testosterone results were included in analysis. Patients who underwent concomitant varicocelectomy and vasectomy were excluded from analysis, as were men with a history of orchiectomy or solitary testis. We reviewed the charts of 1,469 patients, of whom 272 (18.5%) met study inclusion criteria and were included in the final analysis. Demographics, patient questionnaires, operative notes, clinical charts, laboratory reports and semen analyses were reviewed. A full history was obtained and physical examination was done by the primary surgeon in all cases. Physical examination included a full general and urological examination with particular emphasis on testicular volume, and varicocele presence and grade. All testicular volume measurements were made at physical examination by the attending physician using an orchidometer. Varicocele clinical grading was done according to the method of Dubin and Amelar. 11 Our surgical technique of subinguinal microscopic inguinal varicocelectomy with testicular delivery was described previously. 15 We reviewed all semen analysis data and hormonal profiles available in the clinical record. All testosterone levels were obtained before 10 a.m. To be included in our final analysis postoperative laboratory values had to be obtained at least 2 months after and within 3 years of surgery. In cases in which 2 or more analyses were available after surgery we averaged the results of the first 2 semen analyses and used this value in our final analysis. TMC per ejaculate was calculated by the formula, TMC per ejaculate ejaculate volume in ml concentration per ml in millions per ml motile fraction in % motile sperm. 16 All patients who were azoospermic and severely oligospermic (less than 1.0 10 7 spermatozoa) at initial semen analysis were excluded from semen analysis. If multiple postoperative hormone values were available, the test done closest to surgery was used. Statistical Analysis Patients were divided into 3 groups based on age at surgery, including less than 30, 30 to 39 and 40 years old or older. All patients were stratified by age at surgery and baseline characteristics were compared. Results are shown as the mean SE. We used 1-way ANOVA to compare continuous variables with the Tukey HSD post hoc test applied when ANOVA revealed statistical significance. The Pearson chi-square test was used to compare binary variables and the Kruskal-Wallis test was used to compare categorical variables across all age categories. To compare preoperative and postoperative hormonal and semen analysis results we applied the paired 2-sample t test. Analysis was restricted to patients with matched preoperative and postoperative semen analysis, and/or hormonal profile results. Patients on clomiphene citrate or aromatase inhibitors (anastrozole) were excluded from testosterone subanalysis, as were those with azoospermia at baseline. Continuous variables were evaluated for normal distribution with the Kolmogorov-Smirnov test. For semen analysis total sperm count and concentration we applied cube root transformation for paired t test analysis to obtain a normal distribution, 16 which was confirmed by the Kolmogorov-Smirnov test. All results are shown as the mean SE with p 0.05 considered statistically significant. Statistical analysis was done with JMP and SPSS 16.0. RESULTS Baseline Characteristics A total of 272 men met study inclusion criteria and were divided into 3 groups based on age at surgery, including 74, 187 and 85 who were younger than 30, 30 to 39 and 40 years old or older, respectively (mean 35.6 0.4). Average partner age was 32.3 0.3 years and pregnancy had been attempted for a mean of 23.7 1.2 months. As expected, stratification by age at surgery showed that older men tended to have older partners and the percent of patients with proven fertility with the current partner increased. In all 3 age groups baseline testosterone and baseline semen analysis parameters were similar (table 1).
622 OLDER AGE AND SEMEN PARAMETERS AFTER VARICOCELECTOMY Table 1. Baseline clinical parameters in 272 study patients by age at surgery Less Than 30 30 39 40 3-Way Pairwise p Value No. pts 56 147 69 Mean SE age at surgery (range) 25.7 0.5 (16 29.9) 35.0 0.3 (30.1 39.4) 45.4 0.4 (40 65.4) 0.0001 (ANOVA)* Mean SE female age 24.7 0.6 32.2 0.3 37.7 0.5 0.0001 (ANOVA)* Mean SE pregnancy attempted (No. mos) 25.3 3.6 23.8 2.0 25.3 3.0 0.88 (ANOVA) No. previous pregnancy with current 8/56 (14.5) 33/147 (22.5) 26/69 (37.7) 0.03 (Pearson chi-square test) partner/total No. (%) Serum values: No. pts 52 144 66 Mean SE follicle-stimulating hormone (mu/ml) 8.5 1.3 8.0 0.8 7.6 1.2 0.87 (ANOVA) Mean SE luteinizing hormone (miu/ml) 6.8 0.8 5.3 0.5 4.3 0.8 0.06 (ANOVA) Mean SE testosterone (ng/dl) 418 21.3 422 12.8 401 18.9 0.65 (ANOVA) Sperm values: No. pts 45 140 63 Mean SE concentration (million/ml) 33.6 7.9 42.8 4.5 49.0 6.7 0.23 Mean SE total count 118.6 28.2 129.7 16.0 129.7 23.8 0.90 Mean SE % motility 41.1 3.1 40.5 1.7 45.6 2.6 0.25 (ANOVA) Mean SE % progression 2.3 0.1 2.3 0.1 2.4 0.1 0.37 (ANOVA) Mean SE % nl morphology 17.4 2.7 16.6 1.6 19.4 2.6 0.64 (ANOVA) Mean SE testicular vol (cc): Lt 18.1 0.7 16.9 0.4 16.4 0.7 0.22 (ANOVA) Rt 18.7 0.8 18.6 0.5 18.8 0.7 0.95 (ANOVA) * Tukey HSD 0.0001 for all pairs. Cube root transformation applied before comparison with ANOVA. In all age groups examined approximately 70% to 75% of patients presented with bilateral varicoceles and underwent bilateral varicocelectomy. In approximately 25% to 30% of patients an isolated left varicocele was repaired while an isolated right varicocele was repaired in fewer than 2% in all groups. In all age groups there was a percent of men with bilateral varicoceles and a similar varicocele grade distribution. Most left varicoceles were grades 2 and 3, and most right varicoceles were grades 1 and 2. Testicular volume was similar across all age groups (table 2). Varicocelectomy Effect Semen parameters. A total of 222 patients had matching preoperative and postoperative semen analysis results, and were included in this section of the investigation (table 3). There was similar followup in all 3 age groups (approximately 9 months, p 0.80) as well as a statistically significant increase in sperm concentration and total sperm count. The postoperative sperm concentration were similar at approximately 58 million per ml in each group. In 30 to 39- year-old men there was also a small but significant increase in the percent of motile sperm after varicocelectomy. In men younger than 30 and those 40 years or older there was no statistically significant increase in motility or progression. Surgery did not seem to affect the percent of normal forms in any age group (table 3). The percent of patients with a TMC of greater than 20 Table 2. Varicocele and operative characteristics by age No. Less Than 30 (%) No. 30 39 (%) No. 40 (%) p Value (Kruskal-Wallis test) Overall 56 147 69 Varicocelectomy: Bilat 38 (67.9) 104 (70.8) 52 (75.4) 0.64 Lt 17 (30.4) 41 (27.9) 16 (23.2) R 1 (1.8) 2 (1.4) 1 (1.5) Lt varicocele grade: 0 4 (7.1) 13 (8.8) 4 (5.8) 0.71 1 6 (10.7) 17 (11.6) 13 (18.8) 2 21 (37.5) 52 (35.4) 25 (36.2) 3 25 (44.6) 65 (44.2) 27 (39.1) Rt varicocele grade: 0 14 (25.0) 45 (30.6) 19 (27.5) 0.98 1 21 (37.5) 44 (29.9) 24 (34.8) 2 18 (32.1) 47 (32.0) 21 (30.4) 3 3 (5.4) 11 (7.5) 5 (7.3)
OLDER AGE AND SEMEN PARAMETERS AFTER VARICOCELECTOMY 623 Table 3. Change in semen analysis results by age Mean SE Less Than 30 Mean SE 30 39 Mean SE 40 No. pts 39 150 58 Followup (mos)* 10 1.3 9.4 0.7 10.8 1.1 Total sperm count (millions/ml): Preop 128.9 27.7 137.9 17.2 139.9 26.5 Postop 202.8 39.6 168.7 14.1 154.0 20.8 T T(38) 2.45 T(129) 3.86 T(57) 2.17 p Value (paired t test) 0.01 0.0002 0.03 Concentration: Preop 36.3 7.3 44.5 4.7 52.1 7.7 Postop 57.2 10.2 58.2 5.1 57.7 6.3 T T(38) 3.32 T(129) 4.77 T(56) 2.20 p Value (paired t test) 0.002 0.0001 0.03 % Motility: Preop 42.6 3.8 40.9 1.8 45.9 2.2 Postop 45.4 3.5 49.3 1.7 49.7 2.3 T T(36) 0.58 T(124) 4.35 T(56) 1.51 p Value (paired t test) 0.56 0.0001 0.13 Progression: Preop 2.3 0.1 2.3 0.1 2.4 0.1 Postop 2.5 0.1 2.4 0.1 2.3 0.1 T T(25) 1.28 T(91) 2.03 T(40) -0.28 p Value (paired t test) 0.21 0.05 0.78 % Normal morphology: Preop 18.4 3.5 17.0 1.5 20.5 2.8 Postop 19.1 2.1 17.8 1.3 19.8 2.6 T T(32) -0.20 T(98) -0.55 T(39) -0.68 p Value (paired t test) 0.84 0.57 0.49 * Paired sample t test p 0. million increased similarly in all 3 age groups, including 64.9% (24 of 37 men) to 70.3% (26 of 37) in those younger than 30 years (p 0.0001), 61.6% (77 of 125) to 70.4% (88 of 125) in the middle age group (p 0.0001) and 63.2% (36 of 57) to 71.9% (41 of 57) in group older than 40 years (p 0.01). Pregnancy was achieved by 33.3% of the men (13 of 39) younger than 30 years, by 39.2% (51 of 130) of those 30 to 39 years old and by 24.1% (14 of 58) of those 40 years old or older. Testosterone. A total of 106 men had matching preoperative and postoperative serum testosterone levels available and were included in this subanalysis (see figure). Patients on clomiphene citrate or anastrozole were excluded from analysis. There was a statistically significant increase in testosterone in the older, middle and younger age groups (mean 93 25, 59 25 and 73 32 ng/dl, respectively). Mean followup was 10 1.1 months in all patients in this subgroup with similar followup in all 3 groups (p 0.18). When patients were stratified by preoperative testosterone 400 ng/dl without stratification by age, those with testosterone less than 400 ng/dl at baseline had significant increases in testosterone (mean 309 7.4 to 431 16.2 ng/dl, p 0.001) while those with baseline testosterone 400 ng/dl or greater did not (498 17 to 463 30.5, p 0.29). Thus, analysis was further restricted to the 72 patients with baseline testosterone 400 ng/dl or greater, in whom the mean increase was 110, 133 and 136 ng/dl in 21 who were 40 years or older, in 30 who were 30 to 39 years old and in 21 who were younger than 30 years, respectively. Mean followup was 6.8 0.8 months with similar followup in each group (p 0.09). Testosterone and semen parameters. To evaluate whether there was a relationship between improved testosterone and improved semen parameters we examined 66 of our 272 patients (24.3%) who had preoperative and postoperative testosterone, and semen analysis results available, and were not on anastrozole or clomiphene citrate. Of this subgroup 41 men (62.1%) had concomitant improvements in TMC and testosterone, 7 (10.6%) had no improvement in TMC or testosterone, 8 (12.1%) had improvement in TMC only and 10 (15.2%) had improvement in testosterone only. When analysis of this subgroup was further restricted to 49 men with baseline testosterone less than 400 ng/dl, 34 (69.4%) had concomitant improvements in TMC and testosterone. DISCUSSION Varicocelectomy may be indicated for infertility and androgen deficiency. To determine whether varicocelectomy is less effective in older men we reviewed the records of 272 patients who underwent subinguinal
624 OLDER AGE AND SEMEN PARAMETERS AFTER VARICOCELECTOMY Testosterone changes after varicocelectomy in all men (A) and in men with baseline testosterone 400 ng/dl or less (B). Light gray bars indicate mean testosterone. Dark gray bars indicate mean postoperative testosterone. Whiskers indicate SEM. microscopic varicocelectomy at a tertiary referral center, as done by a single surgeon, and had matching semen analysis and/or testosterone results available. To evaluate the effect of age at surgery we divided our patients into 3 groups based on age at surgery, including less than 30, 30 to 39 and 40 years or greater. As expected, we found older partner age and a higher prevalence of secondary infertility in the oldest age group (40 years or older). Across all age groups studied varicocelectomy was associated with significant increases in sperm concentration, total sperm count and serum testosterone. We studied a surgical treatment for 2 indications and determined whether improvements in semen parameters and testosterone correlated. To answer this we evaluated a subset of 66 patients with matching testosterone and semen analysis results who were not on anatrozole or clomiphene citrate. In 41 patients (62.1%) there were concomitant improvements in semen analysis results and testosterone. An additional 15 men (22.7%) achieved improved semen analysis parameters or testosterone. Thus, in more than half of our subgroup varicocelectomy improved testosterone levels and TMC. The efficacy of varicocelectomy has been questioned in older men. The argument against varicocelectomy in older men is that if they do not respond as well to varicocelectomy, perhaps they should not be offered this surgical option. In fact, in an older man with an older female partner seeking treatment for infertility and androgen deficiency one can argue that assisted reproduction should be done and testosterone replacement should be started. However, if varicocelectomy is as effective in older men, we should continue to offer varicocelectomy to older men who have androgen deficiency or a younger partner, or elect not to undergo assisted reproduction. Mean age in our oldest group was only 45.4 years. Most patients in this group were in the fourth or fifth decade of life (data not shown). Thus, our study shows the efficacy of varicocelectomy in patients in these men. While most men presenting with clinical hypogonadism are generally in the seventh decade of life or older, to our knowledge it remains to be shown that varicocelectomy is equally effective in this age group. Also, while we noted that varicocelectomy can improve serum testosterone in men with baseline testosterone less than 400 ng/dl, we did not find such improvement in men with baseline testosterone 400 ng/dl or greater. Thus, varicocelectomy is unlikely to improve testosterone production in men with normal testosterone, which in our study was 400 ng/dl or greater. Our study is not without limitations. In our population the indication for varicocelectomy was heterogeneous, ie infertility vs hypogonadism, which may have limited our ability to detect indication specific differences in surgical outcome. Also, while we report pregnancy outcomes, we did not compare them across age groups. We believe that with the different partner ages across groups and differing female reproductive potential any comparison of fertility outcomes across groups would have been somewhat artificial. Few groups have addressed this topic. Of those studies our results agree with those of Ishikawa and Fujisawa, 13 and Zini et al, 14 who noted that age at varicocelectomy did not affect semen analysis results or pregnancy rates. Our study confirms the uniformly positive effect of varicocelectomy on sperm concentration and total sperm count. Furthermore, to our knowledge we report for the first
OLDER AGE AND SEMEN PARAMETERS AFTER VARICOCELECTOMY 625 time that testosterone increased significantly after varicocelectomy across all age groups examined. CONCLUSIONS Subinguinal microsurgical varicocelectomy is associated with significant increases in sperm concentration, total sperm count and testosterone across all age groups studied. The increase in testosterone was greatest in men with baseline serum testosterone 400 ng/dl or less with a mean testosterone increase of more than 100 ng/dl. Microsurgical varicocelectomy should be offered for infertility and/or androgen deficiency. REFERENCES 1. Gorelick JI and Goldstein M: Loss of fertility in men with varicocele. Fertil Steril 1993; 59: 613. 2. Russell JK: Varicocele, age, and fertility. Lancet 1957; 273: 222. 3. Lipshultz LI and Corriere JN Jr: Progressive testicular atrophy in the varicocele patient. J Urol 1977; 117: 175. 4. Nagler HM, Li XZ, Lizza EF et al: Varicocele: temporal considerations. J Urol 1985; 134: 411. 5. Kass EJ and Belman AB: Reversal of testicular growth failure by varicocele ligation. J Urol 1987; 137: 475. 6. Hadziselimovic F, Herzog B, Liebundgut B et al: Testicular and vascular changes in children and adults with varicocele. J Urol 1989; 142: 583. 7. Harrison RM, Lewis RW and Roberts JA: Pathophysiology of varicocele in nonhuman primates: long-term seminal and testicular changes. Fertil Steril 1986; 46: 500. 8. Chehval MJ and Purcell MH: Deterioration of semen parameters over time in men with untreated varicocele: evidence of progressive testicular damage. Fertil Steril 1992; 57: 174. 9. Witt MA and Lipshultz LI: Varicocele: a progressive or static lesion? Urology 1993; 42: 541. 10. Schlegel PN and Goldstein M: Anatomical approach to varicocelectomy. Semin Urol 1992; 10: 242. 11. Dubin L and Amelar RD: Varicocelectomy: 986 cases in a twelve-year study. Urology 1977; 10: 446. 12. Su LM, Goldstein M and Schlegel PN: The effect of varicocelectomy on serum testosterone levels in infertile men with varicoceles. J Urol 1995; 154: 1752. 13. Ishikawa T and Fujisawa M: Effect of age and grade on surgery for patients with varicocele. Urology 2005; 65: 768. 14. Zini A, Boman J, Jarvi K et al: Varicocelectomy for infertile couples with advanced paternal age. Urology 2008; 72: 109. 15. Goldstein M, Gilbert BR, Dicker AP et al: Microsurgical inguinal varicocelectomy with delivery of the testis: an artery and lymphatic sparing technique. J Urol 1992; 148: 1808. 16. Handelsman DJ: Optimal power transformations for analysis of sperm concentration and other semen variables. J Androl 2002; 23: 629.