Risk Factors for Progressive Deterioration of Semen Quality in Patients With Varicocele
Article Outline
Objective
To assess the risk factors for progressive deterioration of semen quality (PDSQ) in adult patients with varicocele.
Methods
A total of 32 men with left varicocele and impaired semen quality (group 1) and 30 age-matched patients with left varicocele and normal semen quality (group 2) were recruited for the present study. All the subjects received conservative treatment, and the parameters for evaluation every 12 months included semen quality, peak retrograde flow (PRF) and spontaneous venous reflux by color Doppler ultrasonography, body mass index, serum concentration of follicle-stimulating hormone, luteinizing hormone, prolactin, testosterone, testicular volume and discrepancy, grade of varicocele, and scrotal temperature.
Results
The mean follow-up time was 63.2 months (range 60-66). The patients in group 1 had a significantly greater PRF, lower testicular volume, greater testicular volume discrepancy, lower testosterone level, higher scrotal temperature, and greater follicle-stimulating hormone level than those in group 2 at first. The semen quality deteriorated in 28 subjects (87.5%) in group 1, but in only 6 patients (20%) in group 2 during follow-up. Furthermore, the 6 subjects with PDSQ in group 2 had greater PRF and scrotal temperature than those without.
Conclusion
The rate of PDSQ was significantly greater in the varicocele patients with an initially abnormal semen quality than in those with initially normal semen quality (87.5% vs 20%). Furthermore, the varicocele patients with initially normal semen quality who had greater PRF and scrotal temperature might have a greater risk of PDSQ.
Varicocele is characterized by abnormal tortuosity and dilation of the veins of the pampiniform plexus within the spermatic cord; however, the definite etiology of varicocele is unknown. Varicocele is 1 of the causes of male infertility, and the prevalence of varicocele is known to be about 15%-20% in the general population and 30%-40% in infertile men.1 Zheng et al2 reported that experimental varicocele caused progressive impairment of the ipsilateral testis in an animal model. Witt and Lipshultz3 stated that varicocele in some men is a progressive and not a static lesion; however, Diamond et al4 argued against the concept that adolescent varicocele is a progressive disease process. Whether varicocele is a progressive or static disease is still under debate.
Scrotal temperature, testicular volume, pressure within the pampiniform plexus, and basal lamina thickness have been shown to correlate with the prognosis of varicocelectomy; however, they cannot be relied on as predictors of successful varicocelectomy because of the variation within patients.5 Levels of toxicants, reactive oxygen species, and hormones have been examined as predictors of varicocelectomy but only in small-scale studies.5 Zampieri et al6 reported that spontaneous venous reflux correlated with the onset of testicular hypotrophy and abnormal semen quality. Kozakowski et al7 demonstrated that patients with varicocele associated with a peak retrograde flow (PRF) >38 cm/s and 20% asymmetry of testicular volume should be considered for varicocelectomy. Handel et al8 reported that the prevalence of varicocele decreased with increasing body mass index (BMI). Irkilata et al9 suggested that the role of testosterone in the pathophysiology of varicocele has not been established and that testosterone might induce relaxation of the human internal spermatic vein.
Information concerning the risk factors for the progressive deterioration of semen quality (PDSQ) in adult patients with varicocele is limited. We therefore conducted the present prospective study to evaluate the possible risk factors by measuring the semen quality (including sperm density, motility, and morphology), PRF and spontaneous venous reflux by color Doppler ultrasonography (CDS), BMI, serum concentration of follicle-stimulating hormone (FSH), luteinizing hormone (LH), prolactin, testosterone, testicular volume and discrepancy, varicocele grade, and scrotal temperature.
Material and Methods
Patients
From January 2001 to August 2004, 32 young male patients with left varicocele and impaired semen quality (group 1) and 30 age-matched male patients with left varicocele and normal semen quality (group 2) were recruited for the present study. The patients were selected from the urology outpatient department, and those who wished to undergo surgery or who had severe oligospermia (<5 × 106/mL) or azoospermia were excluded. None of the patients had a history of smoking, maldescended testis, infection, previous surgery on the scrotum, or subclinical varicocele. The examinations were performed in a warm room with the patients in a standing position; the scrotum was inspected by palpation, and all the patients underwent CDS of the scrotum to confirm the diagnosis of clinical varicocele (palpable or visible). All the observations were blinded, all the subjects provided written informed consent, and the institutional review board of Taipei City Hospital approved the study.
All the patients received annual checkups. The parameters for evaluation included semen quality (sperm density, motility, and morphology; total sperm count [TSC]; and total motile sperm count [TMSC]); PRF and spontaneous venous reflux according to CDS; BMI (<25 kg/m2, normal; ≥25 kg/m2; and <30 kg/m2, overweight; ≥30 kg/m2, obese); distance from renal hilum to scrotum; serum concentration of FSH, LH, prolactin, and testosterone; testicular volume and discrepancy; varicocele grade; and scrotal temperature. The change in semen quality was defined as [(initial data − data after 60 months)/initial data] × 100%. The varicocele grade was defined as follows: grade 1, varicocele palpable only during the Valsalva maneuver; grade 2, varicocele palpable in the absence of the Valsalva maneuver; and grade 3, varicocele detectable by visual scrutiny.10 Testicular volume (bilateral) was measured using ultrasonography and the following formula: volume = 0.53 × length × width × height. A discrepancy was calculated as follows: (right testicular volume − left testicular volume) × 100/right testicular volume.11 The normal ranges were as follows: testosterone, 241-827 ng/dL; prolactin, 2.0-14.7 ng/mL; FSH, 1-15 mIU/mL; and LH, 1-15 mIU/mL. The scrotal temperature was measured using thermistor temperature sensors attached to the skin of the left scrotum.12
Sperm Preparation
Two sets of semen samples were collected from the patients according to the World Health Organization guidelines.13 The patients were abstinent for 3-4 days, and leukospermia and viscous semen samples were excluded. All semen samples were analyzed for semen quality, including sperm density, motility, morphology, TSC, and TMSC within 1 hour after liquefaction at room temperature, and the mean values were used for statistical analysis. PDSQ was defined as a decrease in sperm density, TSC, or TMSC >45% or deterioration of motility or morphology >20% during follow-up.
Statistical Analysis
We used the Mann-Whitney U test and multivariate analysis to analyze the experimental data. The Statistical Package for Social Sciences, version 10.0, for Windows software (SPSS, Chicago, IL) was used for all statistical analyses. P < .05 was considered statistically significant when examining the differences between groups.
Results
A total of 80 patients with left varicocele (42 with impaired semen quality [group 1] and 38 with normal semen quality [group 2]) were included in the present study. However, 18 subjects (10 of 42 and 8 of 38 in groups 1 and 2, respectively) did not finish the evaluation. The mean age was 22.3 years (range 18-25) and 21.9 years (range 18-25) in groups 1 and 2, respectively. The mean follow-up period was 63.2 months (range 60-66). The data used for evaluation for all patients are listed in Table 1, Table 2. No patients had improved semen quality, increased left testicular volume, downgrading of left varicocele, or developed right varicocele during follow-up. Of the 32 patients in group 1, 6 had grade 1, 16 had grade 2, and 10 had grade 3. Of the 30 patients in group 2, 7 had grade 1, 15 had grade 2, and 8 had grade 3. Upgrading of varicocele was noted in 6 patients (18.8%) in group 1 (3 with grade 1 and 3 with grade 2) but only in 2 patients (6.7%) in group 2 (1 with grade 1 and 1 with grade 2). Spontaneous venous reflux was noted in all 18 patients with grade 3 and in 7 patients with grade 2 (5 in group 1 and 2 in group 2). The patients in group 1 had significantly poorer semen quality, greater PRF (32.4 ± 3.2 vs 27.9 ± 3.0 cm/s), lower testicular volume (24.1 ± 6.1 vs 29.5 ± 5.6 mL), greater testicular volume discrepancy (11.8% ± 4.2% vs 2.8% ± 2.1%), lower testosterone levels (337.5 ± 112.5 vs 410.2 ± 123.2 ng/dL), higher scrotal temperature (34.55° ± 0.92°C vs 33.92° ± 1.10°C), and higher FSH levels (16.9 ± 5.4 vs 12.1 ± 2.5 mIU/mL) than those in group 2. The differences were all significant on multivariate analysis. In addition, the amplitudes of change in sperm density, TSC, and TMSC were significantly greater in the patients in group 1 than in those in group 2. No significant differences were noted for the distance from renal hilum to scrotum, BMI, and serum levels of prolactin and LH between the 2 groups during follow-up (Table 2).
Table 1. Mean semen quality (sperm motility, morphology, density, total sperm count, and total motile sperm count)
| Semen Analysis | Initial | Follow-up (mo) | ||||
|---|---|---|---|---|---|---|
| 12 | 24 | 36 | 48 | 60 | ||
| Motility (%) | ||||||
 G1 | 44.22±5.44 | 42.52±5.12 | 41.02±4.99 | 40.42±4.81 | 39.02±4.67 | 38.72±4.59 |
| G2 | 63.04±9.75 | 63.02±9.73 | 61.64±9.34 | 59.88±9.02 | 59.02±8.79 | 55.76±7.98 |
| P value | < .01 | < .01 | < .01 | < .01 | < .01 | < .01 |
| Morphology (%) | ||||||
| G1 | 42.73±6.12 | 40.61±6.01 | 39.01±5.32 | 38.21±5.01 | 37.57±4.51 | 37.02±4.01 |
| G2 | 61.23±7.92 | 61.21±7.90 | 60.02±7.45 | 58.56±7.12 | 57.91±6.99 | 54.89±6.51 |
| P value | < .01 | < .01 | < .01 | < .01 | < .01 | < .01 |
| Density (×106/mL) | ||||||
| G1 | 13.78±5.34 | 10.99±5.02 | 9.28±4.99 | 8.97±4.21 | 7.98±4.01 | 7.47±3.79 |
| G2 | 39.87±13.04 | 39.84±13.01 | 38.27±12.56 | 36.39±12.01 | 35.98±11.59 | 31.94±10.78 |
| P value | < .01 | < .01 | < .01 | < .01 | < .01 | < .01 |
| TSC (× 106) | ||||||
| G1 | 28.98±10.52 | 23.32±10.01) | 19.51±8.99 | 18.90±8.21) | 16.82±7.99) | 15.65±7.85 |
| G2 | 87.95±22.51 |
87.68±22.31 | 84.22±20.51 | 80.45±20.11) | 79.21±19.85) | 70.28±18.99 |
| P value | < .01 | < .01 | < .01 | < .01 | < .01 | < .01 |
| TMSC (× 106) | ||||||
| G1 | 12.81±5.21 | 9.92±4.95 | 8.00±4.85 | 7.64±3.91 | 6.56±3.65 | 6.06±3.42 |
| G2 | 55.44±15.21 | 55.26±15.01 | 51.91±14.62 | 48.14±14.42 | 46.75±14.01 | 39.19±13.12 |
| P value | < .01 | < .01 | < .01 | < .01 | < .01 | < .01 |
Table 2. Mean values of parameters studied in groups 1 and 2
| Parameter | Group 1(n = 32) | Group 2(n = 30) | P Value |
|---|---|---|---|
| PRF⁎ | 32.4±3.2 | 27.9±3.0 | .03 |
| BMI | 21.9±2.2 | 23.4±3.1 | .08 |
| FSH⁎ | 16.9±5.4 | 12.1±2.5 | .02 |
| LH | 21.7±15.4 | 23.3±12.8 | .08 |
| Testosterone⁎ | 337.5±112.5 | 410.2±123.2 | .02 |
| Prolactin | 15.1±7.5 | 15.9±8.2 | .09 |
| Testicular volume⁎ | 24.1±6.1 | 29.5±5.6 | .02 |
| Right | 12.8±3.4 | 15.1±2.8 | .06 |
| Left⁎ | 11.4±2.9 | 14.6±2.3 | .01 |
| TVD⁎ | 11.8±4.2 | 2.8±2.1 | .01 |
| ST⁎ | 34.55±0.92 | 33.92±1.10 | .04 |
| Motility change (%) | 12.4±1.4 | 11.5±1.2 | .08 |
| Morphology change (%) | 13.4±1.5 | 10.4±1.1 | .07 |
| Density⁎ change (%) | 45.8±5.2 | 19.9±2.5 | .01 |
| TSC⁎ change (%) | 46.0±5.5 | 20.1±2.7 | .01 |
| TMSC⁎ change (%) | 52.7±5.8 | 29.3±2.8 | .01 |
⁎ P < .05 considered statistically significant. |
The data used for evaluation of the patients who did and did not develop PDSQ are listed in Table 3, Table 4. The semen quality deteriorated progressively in 28 subjects (87.5%) in group 1 within 5 years (10 in the first year, 18 in the second year, 23 in the third year, and 27 in the fourth year; 5 with grade 1, 14 with grade 2, and 9 with grade 3). A total of 28 patients with PDSQ in group 1 had greater PRF, higher scrotal temperature, lower testicular volume, greater testicular volume discrepancy, lower BMI, lower testosterone, and greater FSH than those without PDSQ in group 1. On multivariate analysis, significant differences were found only in the PRF and scrotal temperature (Table 3). Six patients (20%) in group 2 developed PDSQ during follow-up (1 in the second year, 3 in the third year, and 4 in the fourth year; 1 with grade 1, 2 with grade 2, and 3 with grade 3). Six patients with PDSQ in group 2 had greater PRF, higher scrotal temperature, lower testicular volume, greater testicular volume discrepancy, lower BMI, and lower testosterone levels than those without PDSQ in group 2. On multivariate analysis, significant differences were found in PRF and scrotal temperature (Table 4). In addition, patients with PDSQ had significantly greater amplitudes of change in the sperm density, motility, and morphology and TSC and TMSC than those without PDSQ (Table 3, Table 4).
Table 3. Mean values of parameters studied in group 1 patients with and without progressive deterioration of semen quality
| Parameter | Progressive (n = 28) | Nonprogressive (n = 4) | P Value |
|---|---|---|---|
| PRF⁎ | 32.9±3.6 | 27.4±2.9 | .02 |
| BMI | 21.3±2.1 | 23.2±2.9 | .07 |
| FSH | 17.1±5.6 | 16.0±3.7 | .08 |
| LH | 21.6±15.1 | 22.4±15.8 | .09 |
| Testosterone | 330.5±101.4 | 386.4±145.2 | .06 |
| Prolactin | 14.9±7.1 | 16.5±8.6 | .06 |
| Testicular volume | 23.6±5.9 | 27.6±6.7 | .06 |
| Right | 12.5±3.0 | 14.4±3.3 | .07 |
| Left | 11.1±2.6 | 13.2±3.1 | .06 |
| TVD | 12.1±4.1 | 9.6±4.3 | .06 |
| ST⁎ | 35.02±0.21 | 34.05±0.18 | .04 |
| Motility (%) | |||
| Initial⁎ | 43.31±5.21 | 50.60±5.74 | .04 |
| 60 mo⁎ | 37.10±3.52 | 50.06±5.71 | .02 |
| Change⁎ | 14.4±1.5 | 1.1±0.2 | .01 |
| Morphology (%) | |||
| Initial⁎ | 42.32±4.99 | 45.62±5.82 | .08 |
| 60 mo⁎ | 35.80±3.21 | 45.56±5.82 | .03 |
| Change⁎ | 15.4±1.3 | 5.0±0.5 | .04 |
| Density (× 106/mL) | |||
| Initial⁎ | 13.70±5.30 | 14.34±5.41 | .04 |
| 60 mo⁎ | 6.92±1.41 | 11.32±3.82 | .01 |
| Change⁎ | 49.5±4.1 | 21.1±2.9 | .01 |
| TSC (×106) | |||
| Initial⁎ | 28.80±9.51 | 30.21±10.53 | .04 |
| 60 mo⁎ | 14.82±5.38 | 23.81±8.52 | .01 |
| Change⁎ | 48.5±3.9 | 21.2±2.7 | .01 |
| TMSC (×106) | |||
| Initial⁎ | 12.47±5.17 | 15.29±5.45 | .04 |
| 60 mo⁎ | 5.50±2.62 | 11.92±4.98 | .01 |
| Change⁎ | 55.9±4.9 | 22.0±2.3 | .01 |
⁎ P < .05 considered statistically significant. |
Table 4. Mean values of parameters studied in group 2 patients with and without progressive deterioration of semen quality
| Parameter | Progressive (n = 6) | Nonprogressive (n = 24) | P Value |
|---|---|---|---|
| PRF⁎ | 31.8±3.5 | 26.9±2.8 | .03 |
| BMI | 22.8±2.8 | 23.5±3.1 | .09 |
| FSH | 12.8±2.6 | 11.9±2.2 | .07 |
| LH | 22.9±12.2 | 23.4±12.9 | .09 |
| Testosterone | 385.2±101.2 | 416.3±129.5 | .08 |
| Prolactin | 15.4±7.6 | 16.0±8.4 | .12 |
| Testicular volume | 26.1±5.1 | 30.3±5.9 | .07 |
| Right | 13.3±2.7 | 15.4±2.7 | .06 |
| Left | 12.9±2.3 | 14.9±2.6 | .06 |
| TVD | 3.1±2.2 | 2.7±1.8 | .06 |
| ST⁎ | 34.41±0.21 | 33.25±0.19 | .04 |
| Motility (%) | |||
| Initial⁎ | 62.98±9.68 | 63.06±9.76 | .09 |
| 60 mo⁎ | 49.47±7.01 | 57.32±9.32 | .04 |
| Change⁎ | 21.5±2.1 | 9.1±1.3 | .02 |
| Morphology (%) | |||
| Initial⁎ | 61.10±7.78 | 61.26±7.98 | .11 |
| 60 mo⁎ | 51.25±5.82 | 55.81±7.05 | .04 |
| Change⁎ | 16.1±1.9 | 8.9±0.8 | .03 |
| Density (× 106/mL) | |||
| Initial⁎ | 39.52±13.01 | 39.96±13.15 | .13 |
| 60 mo⁎ | 21.49±7.27 | 34.55±12.91 | .01 |
| Change⁎ | 45.6±2.9 | 13.5±1.5 | .01 |
| TSC (×106) | |||
| Initial⁎ | 83.01±26.21 | 84.02±26.82 | .12 |
| 60 mo⁎ | 45.22±13.72 | 72.63±19.12 | .01 |
| Change⁎ | 46.1±3.1 | 13.6±0.9 | .01 |
| TMSC (×106) | |||
| Initial⁎ | 52.28±14.28 | 53.01±14.31 | .11 |
| 60 mo⁎ | 22.63±8.41 | 41.61±13.31 | .01 |
| Change⁎ | 56.7±5.1 | 21.5±1.4 | .01 |
⁎ P < .05 considered statistically significant. |
Comment
Whether varicocele is a progressive or static disease remains controversial. Furthermore, the natural history of varicocele remains unclear. Keel14 reported that the coefficients of variation were about 45% for sperm counts and 20% for velocity. Therefore, in the present study, we defined PDSQ as a decrease in the sperm density, TSC, or TMSC >45% or deterioration of motility or morphology >20% to avoid the effects of laboratory variation. Spinelli et al15 suggested that varicocelectomy increases the testicular volume and improves testicular catch-up growth in adolescents and emphasized the importance of testicular volume in determining the time of surgical treatment. Kozakowski et al7 stated that PRF is a valuable tool in predicting persistent, progressive varicocele in adolescents. In the present study, the rate of PDSQ was significantly greater in varicocele patients with initially impaired semen quality than in those with normal semen quality in the first 5 years. In addition, no patient had improved semen quality, testicular catch-up growth, or downgrading of varicocele during follow-up within 5 years. Therefore, clinicians should pay particular attention to patients with varicocele and impaired semen quality, especially to those with a greater PRF, and early surgery should be considered. Furthermore, there is still a possibility that those with an initially normal semen quality will develop PDSQ, especially for those with a greater PRF. It is worth noting that the patients in the present study were adult men, in contrast to previous studies, which focused on adolescents.
Tsao et al16 reported that the prevalence and severity of varicocele correlated inversely with obesity and supported the concept that obesity might result in a decreased nutcracker effect. However, Paasch et al17 found that total sperm count correlated negatively with the BMI. In the present study, the patients with a lower BMI had poorer semen quality and a greater risk of developing PDSQ, but no significant differences were noted. Elevation of the scrotal temperature impairs spermatogenesis through induction of genes associated with oxidative stress and hypoxia.18 In the present study, the patients with a higher scrotal temperature had significantly poorer semen quality and a greater risk of developing PDSQ. Levinger et al19 suggested that the prevalence of varicocele increases over time and that the incidence of risk is about 10% for each decade of life. Cervellione et al20 suggested that one third of patients with subclinical varicocele would develop clinical varicocele within 4 years. Zampieri et al21 suggested that subclinical varicocele should be considered as the starting point of varicocele. We used CDS to confirm the diagnosis of varicocele and excluded the patients with subclinical varicocele in the present study. The limitations of our study included the relatively low number of patients. In addition, some selection bias existed because the patients who wished to undergo surgery or refused close follow-up were excluded.
Kondo et al22 reported that low serum FSH and high testosterone levels are good prognostic factors for varicocelectomy, and Ishikawa and Fujisawa23 stated that serum-free testosterone will increase after varicocele repair. Tanrikut and Goldstein24 suggested that the effect of varicocele on testosterone is not well understood. In the present study, patients with normal semen quality had a significantly greater testosterone level and lower FSH level than those with impaired semen quality. We propose that Leydig cell function is compromised in patients with varicocele and impaired semen quality, leading to a lower testosterone level. However, the role that testosterone plays in developing PDSQ requires additional evaluation. Furthermore, no significant differences were found in the FSH, LH, and prolactin levels between the patients who did and did not have PDSQ. Nevertheless, in the present study, the patients with normal semen quality had slightly greater FSH and LH compared with the data reported by Kondo et al,22 and the etiology needs additional evaluation.
Conclusions
The rate of PDSQ was significantly greater in patients with varicocele and initially abnormal semen quality than in those with initially normal semen quality (87.5% vs 20%) within 5 years. Furthermore, patients with varicocele and initially normal semen quality who had greater PRF and scrotal temperature had a greater risk of developing PDSQ.
References
- . Effects of varicocele on male infertility . Hum Reprod Update . 2001;7:59–64
- Effects on the ipsilateral testis during progression of experimental varicocele in rat . Med Sci Monit . 2008;14:122–126
- . Varicocele: a progressive or static lesion? . Urology . 1993;42:541–543
- Is adolescent varicocele a progressive disease process? . J Urol . 2004;172:1746–1748
- . Molecular and other predictors for infertility in patients with varicoceles . Front Biosci . 2009;14:3641–3672
- . Varicocele in adolescents: a 6-year longitudinal and follow-up observational study . J Urol . 2008;180:1653–1656
- Peak retrograde flow: a novel predictor of persistent, progressive and new onset asymmetry in adolescent varicocele . J Urol . 2009;181:2717–2723
- . The relationship between varicoceles and obesity . J Urol . 2006;176:2138–2140
- The vasodilator effect of testosterone on the human internal spermatic vein and its relation to varicocele grade . J Urol . 2008;180:772–776
- . Treatment of palpable varicocele in infertile men: a meta-analysis to define the best technique . J Androl . 2009;30:33–40
- . Paternity after varicocelectomy: preoperative sonographic parameters of success . J Ultrasound Med . 2007;26:593–599
- Influence of moderate cycling on scrotal temperature . Int J Androl . 2008;31:403–407
- . In: Laboratory Manual for the Examination of Human Semen and Semen-Cervical Mucus Interaction . 2nd ed.. New York: Cambridge University Publishing Group; 1992;p. 34
- . Within- and between-subject variation in semen parameters in infertile men and normal semen donors . Fertil Steril . 2006;85:128–134
- The role of testicular volume in adolescents with varicocele: the better way and time of surgical treatment . J Urol . 2010;184:1722–1726
- The relationship between varicoceles and obesity in a young adult population . Int J Androl . 2009;32:385–390
- Obesity and age affect male fertility potential . Fertil Steril . 2010;94:2898–2901
- . A single, mild, transient scrotal heat stress causes hypoxia and oxidative stress in mouse testes, which induces germ cell death . Biol Reprod . 2009;80:913–919
- Is varicocele prevalence increasing with age? . Andrologia . 2007;39:77–80
- . Subclinical varicocele in the pediatric age group . J Urol . 2008;179:717–719
- Effects of bioflavonoid in the management of subclinical varicocele . Pediatr Surg Int . 2010;26:505–508
- Predictors of improved seminal characteristics by varicocele repair . Andrologia . 2009;41:20–23
- . Varicocele ligation on free testosterone levels in infertile men with varicocele . Arch J Androl . 2004;50:443–448
- . Varicocele repair for treatment of androgen deficiency . Curr Opin Urol . 2010;20:500–502
Funding Support: This study was supported by a grant from Taipei City Hospital (grant 99002-62-096), Taipei, Taiwan.
PII: S0090-4295(11)02341-7
doi:10.1016/j.urology.2011.08.063
© 2012 Elsevier Inc. All rights reserved.
