- Email: [email protected]
Testosterone Levels Are Not Associated With Magnitude of Deformity in Men With Peyronie’s Disease
ORIGINAL RESEARCH & REVIEWS
PEYRONIE'S DISEASE
Testosterone Levels Are Not Associated With Magnitude of Deformity in Men With Peyronie’s Disease John P. Mulhall, MD, MSc, FECSM, FACS,1 Kazuhito Matsushita, MD,1,3 and Christian J. Nelson, PhD2
ABSTRACT
Introduction: There are some data suggesting that there is a relationship between the magnitude of Peyronie’s disease (PD)eassociated penile curvature and low serum testosterone. Aim: To evaluate the relationship between PD associated penile deformity and total testosterone (TT) and free testosterone (FT) levels. Methods: We identified patients diagnosed with PD who had early morning TT measurements. Because laboratories have different reference values, we analyzed testosterone values 2 ways, as raw values and as standardized values (TT and FT values converted into z-scores for each laboratory). Deformity assessment was conducted during an intracavernosal injectioneinduced rigid erection. The association between T levels and magnitude of penile curvature was analyzed in a number of ways. First, an independent measure t-test tested differences in curvature degree by TT groups (low T: <300 ng/dL; normal T: 300 ng/dL). Second, an association was sought between T levels based on quartiles of the normal range (<300, 300e450, 450e600, and >600 ng/dL). These analyses were conducted for both TT and FT levels and using both raw and standardized T values. Third, multivariable analysis using multiple linear regression was performed in an attempt to define predictors of the degree of penile curvature. Factors entered into the model included: TT level, FT level, patient age, race (white vs other), number of co-morbidities, body mass index, presence of erectile dysfunction, duration of PD, and prostatectomy (yes vs no). Main Outcome Measures: The relationship between T levels and the magnitude of penile curvature. Results: 184 subjects met all study criteria. Average age was 54 ± 12 years. Mean TT level was 425 ± 176 ng/dL, and mean curvature magnitude was 35 ± 19 . The percent of subjects in the first through fourth TT quartiles was 26%, 34%, 24%, and 16%. As a continuous variable, there was no association between TT levels (r ¼ 0.01, P ¼ .95) or FT levels (r ¼ 0.08, P ¼ .30) and the curvature magnitude. When examining TT groups, there was no difference in mean curvature between the low TT group and the normal TT group (35.4 ± 17 vs 34 ± 20 , P ¼ .70). When analyzing the TT and FT levels standardized into z-scores, there was no association between degree of curvature and TT z-scores (r ¼ 0.003, P ¼ .95) and FT z-scores (r ¼ 0.08, P ¼ .43). There was also no association between testosterone levels and degree of curvature in the multivariable model. Conclusions: There does not appear to be any association between testosterone levels and the magnitude of PDassociated penile deformity. Mulhall JP, Matsushita K, Nelson CJ. Testosterone Levels Are Not Associated With Magnitude of Deformity in Men With Peyronie’s Disease. J Sex Med 2019;16:1283e1289. Copyright 2019, International Society for Sexual Medicine. Published by Elsevier Inc. All rights reserved.
Key Words: Peyronie’s Disease; Testosterone; Deformity; Hypogonadism; Penis Received September 27, 2018. Accepted May 29, 2019. 1
Sexual & Reproductive Medicine Program, Urology Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA;
2
Departments of Psychiatry/Behavioral Sciences, Memorial Sloan-Kettering Cancer Center, New York, NY, USA;
3
Department of Urology, Juntendo University, Graduate School of Medicine, Tokyo, Japan
Copyright ª 2019, International Society for Sexual Medicine. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jsxm.2019.05.021
J Sex Med 2019;16:1283e1289
INTRODUCTION Peyronie’s disease (PD) is a benign fibrotic disorder of the tunica albuginea of the corpus cavernosum, which generally results in penile deformity and may result in penile pain and erectile dysfunction (ED), all of which may lead to difficulty engaging in satisfactory sexual relations. Although the pathophysiology remains unclear, a widely accepted hypothesis regarding the cause of PD is that it stems from repetitive 1283
1284
microtrauma to the erect or semi-erect penis with tunical delamination, fibrin deposition, and subsequent dysregulated healing of the tunica albuginea.1e5 The reported prevalence of PD has been discrepant, with a wide range in contemporary prevalence studies depending on the population studied and the definition of PD.6e11 The most recent estimates range from 3.2% in a questionnaire-based study of German men8 to 8.9% in a cohort of American men who were being screened for prostate cancer and examined by urologists,6 both significantly higher than previously reported. The mean age of the onset of PD in these studies was in the sixth decade of life, although it has also been documented in teenagers.12 Low serum testosterone occurs in 6e40% of men, with an age-related prevalence, with an approximate tripling of the prevalence between ages 40e80 years. Data from a number studies have identified a possible link between PD-associated deformity and low testosterone.13e17 Preliminary work on a link between low free testosterone and the presence of PD was undertaken by Moreno and Morgentaler.17 In a randomized controlled study, Cavallini et al16 reported that men with PD had lower free testosterone levels than healthy control subjects and testosterone supplementation in low-testosterone patients with PD improved the efficacy of intralesional verapamil injections. Because scepticism exists in the field regarding whether there is a true link between T and PD, we undertook the current study to define and evaluate the relationship between serum testosterone levels and PD.
METHODS Study Design This study is an Institutional Ethics Review Boardeapproved analysis of a prospectively constructed database. The sexual medicine database at our institution is updated in real time and includes patient demographics, medical comorbidities, medication exposure, as well as key PD factors, including duration and stability of the condition, nature of deformity, plaque features, penile trauma history, presence of pain and ED, and ability to participate in penetrative relations. The onset of PD was defined as the time of first recognition by the patient of either a penile plaque or penile deformity. Only patients deemed stable were included in this analysis. Stability was defined as 12 months duration and no deformity change for 3 months. Patients were asked to complete a comprehensive (although non-validated) PD evaluation questionnaire at the time of initial presentation. This information was combined with a clinical assessment, which included a rigorously conducted penile deformity assessment (see below).
Testosterone Measurement For the purpose of this study, we identified patients diagnosed with PD based on presence of a tunical plaque as defined by an experienced sexual medicine physician. Patients who had
Mulhall et al
morning testosterone levels measured before 12 noon, performed within 12 months of the first presentation for PD evaluation constituted the study population. Most of these men had total testosterone (TT) and free testosterone (FT) values.
Penile Deformity Assessment To be included in this analysis, patients were required to undergo an in-office deformity assessment. Penile deformity was assessed during maximum penile rigidity using a trimix (papaverine 30 mg/mL, phentolamine 1 mg/mL, prostaglandin E1 10 mg/mL) redosing schedule. If the patient failed to obtain an erection 80% rigid, equivalent to a grade 4 erection on the erection hardness score, they were excluded from this analysis. Patients with indentation(s) only without curvature were also excluded. To measure the degree of deformity, a goniometer was centered over the point of maximum curvature. The degree of curvature and its direction were recorded.
Statistics Correlation analysis was used to assess the univariate association among TT levels, FT levels, and degree of penile curvature. An independent measure t-test tested differences in curvature by TT level groups (low TT: <300 ng/dL; Normal TT: 300 ng/dL). ANOVA was used to test differences in curvature by TT levels quartiles: <300, 300e450, 450e600, and >600 ng/dL. Because the TT and FT values were reported from 11 different laboratories, all of which had different reference values, we analyzed testosterone values as raw values, as well as standardized values. 11 laboratories were used because many patients travel from afar, and some presented for evaluation in the afternoon, so we permitted patients to have their early morning laboratory testing done locally by their place of residence. We standardized the TT and FT values into z-scores for each laboratory. To ensure a stable mean and SD for each laboratory and for a laboratory’s values to be included, we required that 30 testosterone values be available to convert the raw values into z-scores. We also defined z-score groups as <1, 1 to 1, and >1 to represent scores falling <1 SD below the mean, between 1 SD below and above the mean, and >1 SD above the mean. Multivariable analysis used multiple linear regression predicting degree of penile curvature while simultaneously using the following variables as predictors: testosterone level, age, race (white vs other), number of comorbidities, body mass index, presence of erectile dysfunction, duration of PD, and prostatectomy (yes vs no).
RESULTS Patient Population Based on the aforementioned parameters, a total of 184 subjects met all study criteria. Mean patient age was 54 ± 12 (range 17e77) years. The mean duration of PD symptoms at presentation was 18 ± 33 (range 12e180) months. 76% of patients were still capable of having sexual intercourse. The comorbidity profile included J Sex Med 2019;16:1283e1289
1285
Testosterone & Peyronie’s Disease
Table 1. Comorbidity profiles based on testosterone quartiles
Age (y) Duration of PD (mo) Erectile dysfunction Hyperlipidemia Hypertension Radical prostatectomy Curvature (degree) Uni-planar curvature Direction Dorsal Ventral Lateral Bi-planar curvature
<300 (N ¼ 33)
300e450 (N ¼ 43)
451e600 (N ¼ 38)
>600 (N ¼ 24)
P value
53 ± 10.0 10.1 ± 13.7 5 (17%) 14 (44%) 8 (25%) 6 (18%) 35.6 ± 17 20 (61%)
58 ± 9.9 11.3 ± 18.6 6 (14%) 21 (50%) 16 (38%) 22 (51%) 37.9 ± 18 31 (72%)
51 ± 13.4 24.3 ± 48.1 9 (26%) 15 (42%) 10 (28%) 14 (37%) 36.1 ± 21 28 (74%)
48 ± 15.8 22.7 ± 35.7 8 (33%) 6 (25%) 1 (4%) 9 (38%) 32.9 ± 20 12 (50%)
.05 .13 .08 .15 .11 .22 .74 .18
10 3 7 12
12 6 13 12
12 5 11 10
4 3 5 12
.8 .9 .97 .2
(50%) (15%) (35%) (36%)
(39%) (19%) (42%) (28%)
hyperlipidemia (42%), cigarette smoking (25%), hypertension (25%), and diabetes (10%). Mean curvature magnitude was 34 ± 19 . Breakdown of curvature by degree was as follows: 42% were <30 , 35% were 30e45 , 13% were 46e60 , and 10% were >60 . 31% of the patients had multiplanar curvature. We chose the magnitude of curvature as the degree of curvature present in the primary direction of curvature.
Outcomes Mean TT was 425 ± 176 ng/dL. The percent of subjects in the afore-defined TT level quartiles was 26%, 34%, 24%, and 16%, respectively. Patient characteristics based on the quartiles are also presented in Table 1. Breaking down the study population by decade of age (<40 years, 40e49 years, 50e59 years, and 60 years), the percentage of subjects with low TT (<300 ng/dL) in each subgroup was 19%, 38%, 26%, and 21% (P ¼ .19), respectively (Table 2).
Total Testosterone When analyzing the TT levels as raw scores, there was no association between TT levels and degree of curvature (r ¼ 0.01, P ¼ .95). In multivariable analysis, TT was not a significant predictor of degree of curvature (b ¼ 0.01, P ¼ .95). When examining TT groups, there was no difference in mean curvature between the low TT group and the normal TT group, 35 ± 17 vs 34 ± 20 (P ¼ .70), respectively. Table 2. Proportion of age ranges of patients with PD and prevalence of low testosterone levels in patients with PD Age (years)
Peyronie’s disease
Low T (<300)
<40 40e49 50e59 60
23 22 46 47
4 10 11 8
(17%) (16%) (33%) (34%)
J Sex Med 2019;16:1283e1289
(17%) (46%) (24%) (17%)
(43%) (18%) (39%) (26%)
(33%) (25%) (42%) (50%)
Furthermore, there was no significant difference in mean curvature degree by TT quartiles: 35 ± 17 , 33 ± 19 , 36 ± 21 , 33 ± 20 (P ¼ .84). There was also no significant difference in mean TT levels when comparing men with uniplanar curvature (420 ± 175 ng/dL) vs men with multiplanar curvature (435 ± 178 ng/dL, P ¼ .62). When analyzing the TT levels standardized into z-scores, 2 laboratories had 30 values reported and met criteria to convert the raw scores into z-scores (n ¼ 145). There was no association between TT z-scores and degree of curvature (r ¼ 0.003, P ¼ .95). In multivariable analysis, the TT z-scores were not a significant predictor of degree of curvature (b ¼ 0.03, P ¼ .71). When examining TT z-score groups, there was no difference in mean curvature among the 3 groups: 33 ± 13 , <1 SD; 34 ± 20 , 1 SD to 1 SD; and 35 ± 19 , >1 SD (P ¼ .92).
Free Testosterone When analyzing the FT levels as raw scores (n ¼ 154), there was no association between FT levels and degree of curvature (r ¼ 0.08, P ¼ .3). In multivariable analysis, the FT levels were not a significant predictor of degree of curvature (b ¼ 0.01, P ¼ .95). When analyzing the FT levels standardized into z-scores, 2 laboratories had 30 values reported and met the criteria to convert them into z-scores (n ¼ 114). There was no association between FT z-scores and degree of curvature (r ¼ 0.08, P ¼ .43). In multivariable analysis, the FT z-scores were not a significant predictor of degree of curvature (b ¼ 0.03, P ¼ .77). When examining TT z-scores groups, there was no difference in mean curvature among the 3 groups: 24 ± 8 , <1 SD; 35 ± 19 , 1 SD to 1 SD; and 31 ± 23 >1 SD (P ¼ .30).
DISCUSSION The results of this analysis of prospectively collected data demonstrate that there is no relationship between low
1286
testosterone levels and the magnitude of PD-associated penile deformity. Because of an extensive amount of basic science and clinical research, our understanding of PD has evolved significantly in recent years.6, 18 A recent study showed many physicians who encounter men with PD have incorrect assumptions about the disease even in light of the growing awareness of PD.19 The survey found that a large number of primary-care physicians and urologists have a limited understanding of the prevalence, pathophysiology and appropriate management of PD. Studies suggest a prevalence of PD that may be as high as 9%, significantly higher than initially believed. Schwarzer et al,8 in a community based study from Germany, found an overall incidence of 3.2%, with a steep age-related increase, with an incidence of 6.5% in those older than 70 years of age. In a study from Brazil, Rhoden et al9 reported an incidence of 3.7% in the population of men aged >50 who presented for prostate cancer screening. A study from Italy based on a questionnaire administered by andrologists at each of the 10 centers found a 7.1% prevalence of PD and a significant association with cigarette smoking (odds ratio ¼ 7.2), but no association with cardiovascular disease, diabetes, hypertension, or alcohol consumption.11 Among men presenting with ED of various severities, 7.9% of the patients had PD.20 The greatest PD incidence has been cited by Mulhall et al,6 with an 8.9% incidence in a population of men presenting for prostate cancer screening. These authors also found a significant association with older age, hypertension, and diabetes. Some studies have revealed a significant association with ED in comparison to the non-PD population.20e22 Indeed, epidemiological studies showed that PD is also significantly associated with diabetes,6,11,12,20,21,23,24 a condition known to be associated with low testosterone levels. Low testosterone is a prevalent condition.25e28 Estimates of the prevalence of low testosterone depend on the diagnostic criteria used, and thus vary between studies. The Hypogonadism in Males study reported that the estimated overall prevalence of HG was 38.7% in men aged 45 years or older.29 Older men generally have lower testosterone levels. In the Hypogonadism in Males study, the prevalence of low testosterone in the 45e54 years age group was 34%, whereas it was 50% in men >85 years.29 Likewise, in the Baltimore Longitudinal Study, using a serum T level <325 ng/dL, the percentage of men with reduced testosterone levels increased from 12% in men in their 50s to 19%, 28%, and 49% in men in their 60s, 70s, and 80s, respectively.30 Studies have consistently found that low T increases with age31; however, in the cohort of our study, the highest prevalence occurred in men in their 40s with a rate of 46%. Although it has been suggested that a strong association between PD deformity and low serum T concentrations exists, there is very little evidence in support of such a concept. In the study by El-Sakka et al,20 11% of PD patients had low T levels, but the definition of low T in this study was not reported, and no control group existed. Although there is no universally accepted cutoff for normal TT levels, 2 major medical societies
Mulhall et al
(American Urological Association, The Endocrine Society) have recommended 300 ng/dL as a reasonable level to consider as the lower limit of normal TT.13 Much of the weight in favor of the link between PD and low T comes from the Moreno study.17 In that study, low T was defined as TT <300 ng/dL32 or free testosterone (FT) <1.5 ng/dL.33 The laboratory measurements were taken any time between 8:30 AM and 5:30 PM, with no clear indication what the distribution of the timing of the blood draws was in the study population. Serum testosterone levels exhibit a diurnal variation, with peak concentrations occurring between 6e8 AM and trough concentrations at 6e8 PM.34 Because testosterone is subject to circadian rhythms, it is recommended to draw the blood samples during the early morning, when testosterone concentrations are at their maximum,35 and, thus, this is a major limitation of the study by Moreno et al.17 Using a cutoff <300 ng/dL for TT, the prevalence of low T among men with PD in the Moreno study was 29%, which is approximately similar to that in our study (24%), using the same definition. In the analysis by Moreno et al,17 low T was also diagnosed when FT levels were <1.5 ng/dL. Using this cutoff, a staggering 74% of the patients were diagnosed with low T levels, raising significant concerns about the validity of this cutoff. Moreno et al17 compared men with low T to those with normal T levels and reported that men with low T had significantly greater degree of penile curvature than men with normal T (54 ± 25 vs 37 ± 12 , respectively, P ¼ .006), but, when analyzed using only TT, the correlation with severity of curvature was not significant (r ¼ 0.199, P ¼ .138). Most importantly, in the analysis by Moreno et al,17 the assessment of penile deformity was determined in varied fashion, including, using photographs, in-office intracavernosal injection assisted erections and by patient self-report. The proportion in each assessment group was not reported. Although the methods to define the deformity and the curvature vary between the studies, the least reliable approaches to assess penile deformity are likely to be patient self-report and at-home photography.36 For deformity assessment, it is ideal if the urologist conducts the measurement at the time of a rigid erection, preferably using intracavernosal injections in a re-dosing fashion to induce a good quality erection. Ohebshalom et al37 suggested that photography significantly underestimates the degree of curvature, with escalating problems with increasing degree of curvature and increasing degree of ED. In the study by Moreno et al,17 it was not stated whether the photographs were taken by the patient himself or the physician; however, the challenge in using at-home photography is the potentially limited quality of the photograph, as well as defining how rigid the erection was at the time of the photograph. Another analysis by Cavallini et al16 compared 106 men with PD and 99 healthy control subjects and showed that the mean free and bioavailable levels were lower in the PD group. There was no control for time of day in testing T levels, and the assays used could be called into question. Furthermore, almost half the PD patients J Sex Med 2019;16:1283e1289
1287
Testosterone & Peyronie’s Disease
had normal T levels. Second, the authors compared PD parameters among the 54 PD patients with low T and the 52 PD patients with normal T and found the only difference between the 2 groups being a greater plaque volume in those PD patients with low T levels. Plaque area/volume are notoriously challenging to measure, even on ultrasound scanning, with high inter observer and intraobserver variability. Finally, the authors compared the efficacy of intralesional verapamil between 23 PD men with T who received T therapy and 23 PD men with low T who did not receive T therapy. They suggested that plaque area and penile curvature improved better in those men receiving T therapy. Obviously, such small patients cannot answer such an efficacy question. One must also consider what the potential mechanisms would be linking low T and PD. Recent work has shown that comorbidities such as ED, hypertension, and diabetes are commonly linked to both low T and PD.20,23,29 Among those, ED is the most common comorbidity for both, and its most common cause is penile vascular insufficiency. This is usually part of a generalized endothelial dysfunction and is related to several conditions, including diabetes mellitus, hypertension, hyperlipidemia, and obesity. These conditions underlie the pathophysiology of metabolic syndrome. Low T has been found to be an integral component of the pathology underlying endothelial dysfunction and metabolic syndrome.38 One might posit that vascular risk factors might influence the development of PD by increasing susceptibility to penile trauma due to attempts at sexual intercourse with sub-maximal rigidity. Of note, the study by Moreno et al17 failed to use a multivariable analysis to define whether the actual T levels were a predictor of the magnitude of penile deformity or whether the underlying comorbidities (which by themselves may be associated with low T), were the link. In our study, we looked at the impact of the differences in curvature using a variety of measures of TT an FT, including raw scores, z-scores, normal vs abnormal TT (300 vs <300), TT quartiles, and, using a multivariable model for TT, FT using raw and z-scores. In none of the analyses were T levels associated with degree of penile deformity in men with PD. Our analysis is not without limitations; specifically, the study population is relatively small, and there is an absence of a control (non-PD) group. Despite including only 184 patients, the statistical analysis has more than sufficient power to detect small associations, and it is crystal clear in demonstrating no association between T level and degree of deformity. It is extremely unlikely that more patients would change the statistical outcomes. There is no consensus regarding within what period of time from the diagnosis of PD T levels should be measured; thus, a time period of 12 months was chosen, and this was done for 2 main reasons: (i) any duration longer than this may lead to the T levels not being representative, because T levels may have changed over this period of time (ii) windows shorter than this may have led to smaller patient numbers, because many of our patients have seen other urologists before seeing us and are >6 months of having PD at presentation. Another limitation is that J Sex Med 2019;16:1283e1289
our estimate of actuarial association between PD and HG assumes the PD study population in our analysis is representative of all PD patients. Our study cohort included 17% of patients <40 years of age, and 41% of patients had developed PD after radical pelvic surgery. In the absence of a population-based analysis, this issue cannot be resolved with any certainty. Furthermore, we excluded all patients with deformity other than curvature. Such patients were excluded primarily because deformity measurement in such cases is notoriously challenging. For example, there exists no standard means of assessing degree of indentation or hourglass deformity. On the other hand, the strengths of the study include a robust definition of PD (all patients were examined by a highly experienced PD expert), rigorous assessment of penile deformity using intracavernosal injections, highly sophisticated statistical analysis and most importantly a structured measurement of serum T levels as it pertains to circadian rhythm (all laboratory assessments done before noon).
CONCLUSIONS This study demonstrates that there is no relationship between T levels and degree of deformity in PD patients. However, neither the Moreno study nor ours can answer whether HG is associated with PD because they are both non-controlled. Further prospective studies are needed to ascertain the relationship. Corresponding Author: John P. Mulhall, MD, MSc, FECSM, FACS, Sexual and Reproductive Medicine Program, Urology Service, 353 E 68th Street, New York, NY 10065, USA. Tel: (646) 422-4359; Fax: (212) 988-0768; E-mail: mulhalj1@ mskcc.org Conflicts of Interest: The authors report no conflicts of interest. Funding: Supported by the Sidney Kimmel Center for Prostate and Urologic Cancers and the National Institutes of Health/ National Cancer Institute to Memorial Sloan Kettering Cancer Center through the Cancer Center Support Grant, award number P30 CA008748
STATEMENT OF AUTHORSHIP Category 1 (a) Conception and Design John P. Mulhall (b) Analysis of Data Christian J. Nelson (c) Interpretation of Data Kazuhito Matsushita, Christian J. Nelson Category 2 (a) Drafting the Article John P. Mulhall; Kazuhito Matsushita (b) Revising it critically for Important Intellectual Content John P. Mulhall
1288 Category 3
Mulhall et al
(a) Final Approval of the Version to be published John P. Mulhall; Kazuhito Matsushita; Christian J. Nelson
17. Moreno SA, Morgentaler A. Testosterone deficiency and Peyronie’s disease: Pilot data suggesting a significant relationship. J Sex Med 2009;6:1729-1735.
REFERENCES
18. Mulhall JP, Anderson MS, Lubrano T, et al. Peyronie’s disease cell culture models: phenotypic, genotypic and functional analyses. Int J Impot Res 2002;14:397-405.
1. Chung E, Ralph D, Kadioglu A, et al. Evidence-based management of Peyronie’s disease. J Sex Med 2016;13:905-923. 2. Mulhall JP, Alex B, Choi JM. Predicting delay in presentation in men with Peyronie’s disease. J Sex Med 2010;7:2226-2230. 3. Mulhall JP. Expanding the paradigm for plaque development in Peyronie’s disease. Int J Impot Res 2003;15(Suppl. 5):S93-S102. 4. Devine CJ Jr, Somers KD, Jordan SG, et al. Proposal: Trauma as the cause of the Peyronie’s lesion. J Urol 1997; 157:285-290. 5. Brock G, Hsu GL, Nunes L, et al. The anatomy of the tunica albuginea in the normal penis and Peyronie’s disease. J Urol 1997;157:276-281. 6. Mulhall JP, Creech SD, Boorjian SA, et al. Subjective and objective analysis of the prevalence of Peyronie’s disease in a population of men presenting for prostate cancer screening. J Urol 2004;171:2350-2353.
19. LaRochelle JC, Levine LA. A Survey of primary-care physicians and urologists regarding Peyronie’s disease. J Sex Med 2007; 4:1167-1173. 20. El-Sakka AI. Prevalence of Peyronie’s disease among patients with erectile dysfunction. Eur Urol 2006;49:564-569. 21. El-Sakka AI, Tayeb KA. Vascular impairment of erection in patients with diabetes and Peyronie’s disease: Is that accumulative? J Sex Med 2009;6:1736-1742. 22. El-Sakka AI, Tayeb KA. Peyronie’s disease in diabetic patients being screened for erectile dysfunction. J Urol 2005; 174:1026-1030. 23. Usta MF, Bivalacqua TJ, Jabren GW, et al. Relationship between the severity of penile curvature and the presence of comorbidities in men with Peyronie’s disease. J Urol 2004; 171:775-779.
7. Sommer F, Schwarzer U, Wassmer G, et al. Epidemiology of Peyronie’s disease. Int J Impot Res 2002;14:379-383.
24. Kadioglu A, Tefekli A, Erol B, et al. A retrospective review of 307 men with Peyronie’s disease. J Urol 2002; 168:1075-1079.
8. Schwarzer U, Sommer F, Klotz T, et al. The prevalence of Peyronie’s disease: Results of a large survey. BJU Int 2001; 88:727-730.
25. Tal R, Heck M, Teloken P, et al. Peyronie’s disease following radical prostatectomy: incidence and predictors. J Sex Med 2010;7:1254-1261.
9. Rhoden EL, Teloken C, Ting HY, et al. Prevalence of Peyronie’s disease in men over 50-y-old from Southern Brazil. Int J Impot Res 2001;13:291-293.
26. U.S. Census Bureau. Population; Available from: http://www. census.gov/compendia/statab/cats/population.html. Accessed June 8, 2011.
10. Lindsay MB, Schain DM, Grambsch P, et al. The incidence of Peyronie’s disease in Rochester, Minnesota, 1950 through 1984. J Urol 1991;146:1007-1009.
27. Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy in men with androgen deficiency syndromes: An Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2010;95:2536-2559.
11. La Pera G, Pescatori ES, Calabrese M, et al. Peyronie’s disease: prevalence and association with cigarette smoking. A multicenter population-based study in men aged 50-69 years. Eur Urol 2001;40:525-530. 12. Deveci S, Hopps CV, O’Brien K, et al. Defining the clinical characteristics of Peyronie’s disease in young men. J Sex Med 2007;4:485-490. 13. Dandona P, Rosenberg MT. A practical guide to male hypogonadism in the primary care setting. Int J Clin Pract 2010; 64:682-696. 14. Araujo AB, Esche GR, Kupelian V, et al. Prevalence of symptomatic androgen deficiency in men. J Clin Endocrinol Metab 2007;92:4241-4247. 15. Iacono F, Prezioso D, Ruffo A, et al. Testosterone deficiency causes penile fibrosis and organic erectile dysfunction in aging men. Evaluating association among Age, TDS and ED. BMC Surg 2012;12(Suppl. 1):S24. 16. Cavallini G, Biagiotti G, Lo Giudice C. Association between peyronie disease and low serum testosterone levels: Detection and therapeutic considerations. J Androl 2012;33:381-388.
28. Wang C, Nieschlag E, Swerdloff R, et al. Investigation, treatment and monitoring of late-onset hypogonadism in males: ISA, ISSAM, EAU, EAA and ASA recommendations. Eur J Endocrinol 2008;159:507-514. 29. Mulligan T, Frick MF, Zuraw QC, et al. Prevalence of hypogonadism in males aged at least 45 years: the HIM study. Int J Clin Pract 2006;60:762-769. 30. Harman SM, Metter EJ, Tobin JD, et al. Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal Study of Aging. J Clin Endocrinol Metab 2001;86:724-731. 31. Wang C, Nieschlag E, Swerdloff R, et al. Investigation, treatment, and monitoring of late-onset hypogonadism in males: ISA, ISSAM, EAU, EAA, and ASA recommendations. J Androl 2009;30:1-9. 32. Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy in adult men with androgen deficiency syndromes: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2006;91:1995-2010. J Sex Med 2019;16:1283e1289
Testosterone & Peyronie’s Disease
1289
33. Morgentaler A. Commentary: Guideline for male testosterone therapy: A clinician’s perspective. J Clin Endocrinol Metab 2007;92:416-417.
36. Muller A, Mulhall JP. Peyronie’s disease intervention trials: Methodological challenges and issues. J Sex Med 2009; 6:848-861.
34. Diver MJ, Imtiaz KE, Ahmad AM, Vora JP, Fraser WD. Diurnal rhythms of serum total, free and bioavailable testosterone and of SHBG in middle-aged men compared with those in young men. Clin Endocrinol (Oxf) 2003;58:710-717.
37. Ohebshalom M, Mulhall J, Guhring P, et al. Measurement of penile curvature in Peyronie’s disease patients: comparison of three methods. J Sex Med 2007;4:199-203.
35. Rhoden EL, Morgentaler A. Risks of testosterone-replacement therapy and recommendations for monitoring. N Engl J Med 2004;350:482-492.
38. Guay AT, Traish A. Testosterone deficiency and risk factors in the metabolic syndrome: implications for erectile dysfunction. Urol Clin North Am 2011;38:175-183.
J Sex Med 2019;16:1283e1289
PEYRONIE'S DISEASE
Testosterone Levels Are Not Associated With Magnitude of Deformity in Men With Peyronie’s Disease John P. Mulhall, MD, MSc, FECSM, FACS,1 Kazuhito Matsushita, MD,1,3 and Christian J. Nelson, PhD2
ABSTRACT
Introduction: There are some data suggesting that there is a relationship between the magnitude of Peyronie’s disease (PD)eassociated penile curvature and low serum testosterone. Aim: To evaluate the relationship between PD associated penile deformity and total testosterone (TT) and free testosterone (FT) levels. Methods: We identified patients diagnosed with PD who had early morning TT measurements. Because laboratories have different reference values, we analyzed testosterone values 2 ways, as raw values and as standardized values (TT and FT values converted into z-scores for each laboratory). Deformity assessment was conducted during an intracavernosal injectioneinduced rigid erection. The association between T levels and magnitude of penile curvature was analyzed in a number of ways. First, an independent measure t-test tested differences in curvature degree by TT groups (low T: <300 ng/dL; normal T: 300 ng/dL). Second, an association was sought between T levels based on quartiles of the normal range (<300, 300e450, 450e600, and >600 ng/dL). These analyses were conducted for both TT and FT levels and using both raw and standardized T values. Third, multivariable analysis using multiple linear regression was performed in an attempt to define predictors of the degree of penile curvature. Factors entered into the model included: TT level, FT level, patient age, race (white vs other), number of co-morbidities, body mass index, presence of erectile dysfunction, duration of PD, and prostatectomy (yes vs no). Main Outcome Measures: The relationship between T levels and the magnitude of penile curvature. Results: 184 subjects met all study criteria. Average age was 54 ± 12 years. Mean TT level was 425 ± 176 ng/dL, and mean curvature magnitude was 35 ± 19 . The percent of subjects in the first through fourth TT quartiles was 26%, 34%, 24%, and 16%. As a continuous variable, there was no association between TT levels (r ¼ 0.01, P ¼ .95) or FT levels (r ¼ 0.08, P ¼ .30) and the curvature magnitude. When examining TT groups, there was no difference in mean curvature between the low TT group and the normal TT group (35.4 ± 17 vs 34 ± 20 , P ¼ .70). When analyzing the TT and FT levels standardized into z-scores, there was no association between degree of curvature and TT z-scores (r ¼ 0.003, P ¼ .95) and FT z-scores (r ¼ 0.08, P ¼ .43). There was also no association between testosterone levels and degree of curvature in the multivariable model. Conclusions: There does not appear to be any association between testosterone levels and the magnitude of PDassociated penile deformity. Mulhall JP, Matsushita K, Nelson CJ. Testosterone Levels Are Not Associated With Magnitude of Deformity in Men With Peyronie’s Disease. J Sex Med 2019;16:1283e1289. Copyright 2019, International Society for Sexual Medicine. Published by Elsevier Inc. All rights reserved.
Key Words: Peyronie’s Disease; Testosterone; Deformity; Hypogonadism; Penis Received September 27, 2018. Accepted May 29, 2019. 1
Sexual & Reproductive Medicine Program, Urology Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA;
2
Departments of Psychiatry/Behavioral Sciences, Memorial Sloan-Kettering Cancer Center, New York, NY, USA;
3
Department of Urology, Juntendo University, Graduate School of Medicine, Tokyo, Japan
Copyright ª 2019, International Society for Sexual Medicine. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jsxm.2019.05.021
J Sex Med 2019;16:1283e1289
INTRODUCTION Peyronie’s disease (PD) is a benign fibrotic disorder of the tunica albuginea of the corpus cavernosum, which generally results in penile deformity and may result in penile pain and erectile dysfunction (ED), all of which may lead to difficulty engaging in satisfactory sexual relations. Although the pathophysiology remains unclear, a widely accepted hypothesis regarding the cause of PD is that it stems from repetitive 1283
1284
microtrauma to the erect or semi-erect penis with tunical delamination, fibrin deposition, and subsequent dysregulated healing of the tunica albuginea.1e5 The reported prevalence of PD has been discrepant, with a wide range in contemporary prevalence studies depending on the population studied and the definition of PD.6e11 The most recent estimates range from 3.2% in a questionnaire-based study of German men8 to 8.9% in a cohort of American men who were being screened for prostate cancer and examined by urologists,6 both significantly higher than previously reported. The mean age of the onset of PD in these studies was in the sixth decade of life, although it has also been documented in teenagers.12 Low serum testosterone occurs in 6e40% of men, with an age-related prevalence, with an approximate tripling of the prevalence between ages 40e80 years. Data from a number studies have identified a possible link between PD-associated deformity and low testosterone.13e17 Preliminary work on a link between low free testosterone and the presence of PD was undertaken by Moreno and Morgentaler.17 In a randomized controlled study, Cavallini et al16 reported that men with PD had lower free testosterone levels than healthy control subjects and testosterone supplementation in low-testosterone patients with PD improved the efficacy of intralesional verapamil injections. Because scepticism exists in the field regarding whether there is a true link between T and PD, we undertook the current study to define and evaluate the relationship between serum testosterone levels and PD.
METHODS Study Design This study is an Institutional Ethics Review Boardeapproved analysis of a prospectively constructed database. The sexual medicine database at our institution is updated in real time and includes patient demographics, medical comorbidities, medication exposure, as well as key PD factors, including duration and stability of the condition, nature of deformity, plaque features, penile trauma history, presence of pain and ED, and ability to participate in penetrative relations. The onset of PD was defined as the time of first recognition by the patient of either a penile plaque or penile deformity. Only patients deemed stable were included in this analysis. Stability was defined as 12 months duration and no deformity change for 3 months. Patients were asked to complete a comprehensive (although non-validated) PD evaluation questionnaire at the time of initial presentation. This information was combined with a clinical assessment, which included a rigorously conducted penile deformity assessment (see below).
Testosterone Measurement For the purpose of this study, we identified patients diagnosed with PD based on presence of a tunical plaque as defined by an experienced sexual medicine physician. Patients who had
Mulhall et al
morning testosterone levels measured before 12 noon, performed within 12 months of the first presentation for PD evaluation constituted the study population. Most of these men had total testosterone (TT) and free testosterone (FT) values.
Penile Deformity Assessment To be included in this analysis, patients were required to undergo an in-office deformity assessment. Penile deformity was assessed during maximum penile rigidity using a trimix (papaverine 30 mg/mL, phentolamine 1 mg/mL, prostaglandin E1 10 mg/mL) redosing schedule. If the patient failed to obtain an erection 80% rigid, equivalent to a grade 4 erection on the erection hardness score, they were excluded from this analysis. Patients with indentation(s) only without curvature were also excluded. To measure the degree of deformity, a goniometer was centered over the point of maximum curvature. The degree of curvature and its direction were recorded.
Statistics Correlation analysis was used to assess the univariate association among TT levels, FT levels, and degree of penile curvature. An independent measure t-test tested differences in curvature by TT level groups (low TT: <300 ng/dL; Normal TT: 300 ng/dL). ANOVA was used to test differences in curvature by TT levels quartiles: <300, 300e450, 450e600, and >600 ng/dL. Because the TT and FT values were reported from 11 different laboratories, all of which had different reference values, we analyzed testosterone values as raw values, as well as standardized values. 11 laboratories were used because many patients travel from afar, and some presented for evaluation in the afternoon, so we permitted patients to have their early morning laboratory testing done locally by their place of residence. We standardized the TT and FT values into z-scores for each laboratory. To ensure a stable mean and SD for each laboratory and for a laboratory’s values to be included, we required that 30 testosterone values be available to convert the raw values into z-scores. We also defined z-score groups as <1, 1 to 1, and >1 to represent scores falling <1 SD below the mean, between 1 SD below and above the mean, and >1 SD above the mean. Multivariable analysis used multiple linear regression predicting degree of penile curvature while simultaneously using the following variables as predictors: testosterone level, age, race (white vs other), number of comorbidities, body mass index, presence of erectile dysfunction, duration of PD, and prostatectomy (yes vs no).
RESULTS Patient Population Based on the aforementioned parameters, a total of 184 subjects met all study criteria. Mean patient age was 54 ± 12 (range 17e77) years. The mean duration of PD symptoms at presentation was 18 ± 33 (range 12e180) months. 76% of patients were still capable of having sexual intercourse. The comorbidity profile included J Sex Med 2019;16:1283e1289
1285
Testosterone & Peyronie’s Disease
Table 1. Comorbidity profiles based on testosterone quartiles
Age (y) Duration of PD (mo) Erectile dysfunction Hyperlipidemia Hypertension Radical prostatectomy Curvature (degree) Uni-planar curvature Direction Dorsal Ventral Lateral Bi-planar curvature
<300 (N ¼ 33)
300e450 (N ¼ 43)
451e600 (N ¼ 38)
>600 (N ¼ 24)
P value
53 ± 10.0 10.1 ± 13.7 5 (17%) 14 (44%) 8 (25%) 6 (18%) 35.6 ± 17 20 (61%)
58 ± 9.9 11.3 ± 18.6 6 (14%) 21 (50%) 16 (38%) 22 (51%) 37.9 ± 18 31 (72%)
51 ± 13.4 24.3 ± 48.1 9 (26%) 15 (42%) 10 (28%) 14 (37%) 36.1 ± 21 28 (74%)
48 ± 15.8 22.7 ± 35.7 8 (33%) 6 (25%) 1 (4%) 9 (38%) 32.9 ± 20 12 (50%)
.05 .13 .08 .15 .11 .22 .74 .18
10 3 7 12
12 6 13 12
12 5 11 10
4 3 5 12
.8 .9 .97 .2
(50%) (15%) (35%) (36%)
(39%) (19%) (42%) (28%)
hyperlipidemia (42%), cigarette smoking (25%), hypertension (25%), and diabetes (10%). Mean curvature magnitude was 34 ± 19 . Breakdown of curvature by degree was as follows: 42% were <30 , 35% were 30e45 , 13% were 46e60 , and 10% were >60 . 31% of the patients had multiplanar curvature. We chose the magnitude of curvature as the degree of curvature present in the primary direction of curvature.
Outcomes Mean TT was 425 ± 176 ng/dL. The percent of subjects in the afore-defined TT level quartiles was 26%, 34%, 24%, and 16%, respectively. Patient characteristics based on the quartiles are also presented in Table 1. Breaking down the study population by decade of age (<40 years, 40e49 years, 50e59 years, and 60 years), the percentage of subjects with low TT (<300 ng/dL) in each subgroup was 19%, 38%, 26%, and 21% (P ¼ .19), respectively (Table 2).
Total Testosterone When analyzing the TT levels as raw scores, there was no association between TT levels and degree of curvature (r ¼ 0.01, P ¼ .95). In multivariable analysis, TT was not a significant predictor of degree of curvature (b ¼ 0.01, P ¼ .95). When examining TT groups, there was no difference in mean curvature between the low TT group and the normal TT group, 35 ± 17 vs 34 ± 20 (P ¼ .70), respectively. Table 2. Proportion of age ranges of patients with PD and prevalence of low testosterone levels in patients with PD Age (years)
Peyronie’s disease
Low T (<300)
<40 40e49 50e59 60
23 22 46 47
4 10 11 8
(17%) (16%) (33%) (34%)
J Sex Med 2019;16:1283e1289
(17%) (46%) (24%) (17%)
(43%) (18%) (39%) (26%)
(33%) (25%) (42%) (50%)
Furthermore, there was no significant difference in mean curvature degree by TT quartiles: 35 ± 17 , 33 ± 19 , 36 ± 21 , 33 ± 20 (P ¼ .84). There was also no significant difference in mean TT levels when comparing men with uniplanar curvature (420 ± 175 ng/dL) vs men with multiplanar curvature (435 ± 178 ng/dL, P ¼ .62). When analyzing the TT levels standardized into z-scores, 2 laboratories had 30 values reported and met criteria to convert the raw scores into z-scores (n ¼ 145). There was no association between TT z-scores and degree of curvature (r ¼ 0.003, P ¼ .95). In multivariable analysis, the TT z-scores were not a significant predictor of degree of curvature (b ¼ 0.03, P ¼ .71). When examining TT z-score groups, there was no difference in mean curvature among the 3 groups: 33 ± 13 , <1 SD; 34 ± 20 , 1 SD to 1 SD; and 35 ± 19 , >1 SD (P ¼ .92).
Free Testosterone When analyzing the FT levels as raw scores (n ¼ 154), there was no association between FT levels and degree of curvature (r ¼ 0.08, P ¼ .3). In multivariable analysis, the FT levels were not a significant predictor of degree of curvature (b ¼ 0.01, P ¼ .95). When analyzing the FT levels standardized into z-scores, 2 laboratories had 30 values reported and met the criteria to convert them into z-scores (n ¼ 114). There was no association between FT z-scores and degree of curvature (r ¼ 0.08, P ¼ .43). In multivariable analysis, the FT z-scores were not a significant predictor of degree of curvature (b ¼ 0.03, P ¼ .77). When examining TT z-scores groups, there was no difference in mean curvature among the 3 groups: 24 ± 8 , <1 SD; 35 ± 19 , 1 SD to 1 SD; and 31 ± 23 >1 SD (P ¼ .30).
DISCUSSION The results of this analysis of prospectively collected data demonstrate that there is no relationship between low
1286
testosterone levels and the magnitude of PD-associated penile deformity. Because of an extensive amount of basic science and clinical research, our understanding of PD has evolved significantly in recent years.6, 18 A recent study showed many physicians who encounter men with PD have incorrect assumptions about the disease even in light of the growing awareness of PD.19 The survey found that a large number of primary-care physicians and urologists have a limited understanding of the prevalence, pathophysiology and appropriate management of PD. Studies suggest a prevalence of PD that may be as high as 9%, significantly higher than initially believed. Schwarzer et al,8 in a community based study from Germany, found an overall incidence of 3.2%, with a steep age-related increase, with an incidence of 6.5% in those older than 70 years of age. In a study from Brazil, Rhoden et al9 reported an incidence of 3.7% in the population of men aged >50 who presented for prostate cancer screening. A study from Italy based on a questionnaire administered by andrologists at each of the 10 centers found a 7.1% prevalence of PD and a significant association with cigarette smoking (odds ratio ¼ 7.2), but no association with cardiovascular disease, diabetes, hypertension, or alcohol consumption.11 Among men presenting with ED of various severities, 7.9% of the patients had PD.20 The greatest PD incidence has been cited by Mulhall et al,6 with an 8.9% incidence in a population of men presenting for prostate cancer screening. These authors also found a significant association with older age, hypertension, and diabetes. Some studies have revealed a significant association with ED in comparison to the non-PD population.20e22 Indeed, epidemiological studies showed that PD is also significantly associated with diabetes,6,11,12,20,21,23,24 a condition known to be associated with low testosterone levels. Low testosterone is a prevalent condition.25e28 Estimates of the prevalence of low testosterone depend on the diagnostic criteria used, and thus vary between studies. The Hypogonadism in Males study reported that the estimated overall prevalence of HG was 38.7% in men aged 45 years or older.29 Older men generally have lower testosterone levels. In the Hypogonadism in Males study, the prevalence of low testosterone in the 45e54 years age group was 34%, whereas it was 50% in men >85 years.29 Likewise, in the Baltimore Longitudinal Study, using a serum T level <325 ng/dL, the percentage of men with reduced testosterone levels increased from 12% in men in their 50s to 19%, 28%, and 49% in men in their 60s, 70s, and 80s, respectively.30 Studies have consistently found that low T increases with age31; however, in the cohort of our study, the highest prevalence occurred in men in their 40s with a rate of 46%. Although it has been suggested that a strong association between PD deformity and low serum T concentrations exists, there is very little evidence in support of such a concept. In the study by El-Sakka et al,20 11% of PD patients had low T levels, but the definition of low T in this study was not reported, and no control group existed. Although there is no universally accepted cutoff for normal TT levels, 2 major medical societies
Mulhall et al
(American Urological Association, The Endocrine Society) have recommended 300 ng/dL as a reasonable level to consider as the lower limit of normal TT.13 Much of the weight in favor of the link between PD and low T comes from the Moreno study.17 In that study, low T was defined as TT <300 ng/dL32 or free testosterone (FT) <1.5 ng/dL.33 The laboratory measurements were taken any time between 8:30 AM and 5:30 PM, with no clear indication what the distribution of the timing of the blood draws was in the study population. Serum testosterone levels exhibit a diurnal variation, with peak concentrations occurring between 6e8 AM and trough concentrations at 6e8 PM.34 Because testosterone is subject to circadian rhythms, it is recommended to draw the blood samples during the early morning, when testosterone concentrations are at their maximum,35 and, thus, this is a major limitation of the study by Moreno et al.17 Using a cutoff <300 ng/dL for TT, the prevalence of low T among men with PD in the Moreno study was 29%, which is approximately similar to that in our study (24%), using the same definition. In the analysis by Moreno et al,17 low T was also diagnosed when FT levels were <1.5 ng/dL. Using this cutoff, a staggering 74% of the patients were diagnosed with low T levels, raising significant concerns about the validity of this cutoff. Moreno et al17 compared men with low T to those with normal T levels and reported that men with low T had significantly greater degree of penile curvature than men with normal T (54 ± 25 vs 37 ± 12 , respectively, P ¼ .006), but, when analyzed using only TT, the correlation with severity of curvature was not significant (r ¼ 0.199, P ¼ .138). Most importantly, in the analysis by Moreno et al,17 the assessment of penile deformity was determined in varied fashion, including, using photographs, in-office intracavernosal injection assisted erections and by patient self-report. The proportion in each assessment group was not reported. Although the methods to define the deformity and the curvature vary between the studies, the least reliable approaches to assess penile deformity are likely to be patient self-report and at-home photography.36 For deformity assessment, it is ideal if the urologist conducts the measurement at the time of a rigid erection, preferably using intracavernosal injections in a re-dosing fashion to induce a good quality erection. Ohebshalom et al37 suggested that photography significantly underestimates the degree of curvature, with escalating problems with increasing degree of curvature and increasing degree of ED. In the study by Moreno et al,17 it was not stated whether the photographs were taken by the patient himself or the physician; however, the challenge in using at-home photography is the potentially limited quality of the photograph, as well as defining how rigid the erection was at the time of the photograph. Another analysis by Cavallini et al16 compared 106 men with PD and 99 healthy control subjects and showed that the mean free and bioavailable levels were lower in the PD group. There was no control for time of day in testing T levels, and the assays used could be called into question. Furthermore, almost half the PD patients J Sex Med 2019;16:1283e1289
1287
Testosterone & Peyronie’s Disease
had normal T levels. Second, the authors compared PD parameters among the 54 PD patients with low T and the 52 PD patients with normal T and found the only difference between the 2 groups being a greater plaque volume in those PD patients with low T levels. Plaque area/volume are notoriously challenging to measure, even on ultrasound scanning, with high inter observer and intraobserver variability. Finally, the authors compared the efficacy of intralesional verapamil between 23 PD men with T who received T therapy and 23 PD men with low T who did not receive T therapy. They suggested that plaque area and penile curvature improved better in those men receiving T therapy. Obviously, such small patients cannot answer such an efficacy question. One must also consider what the potential mechanisms would be linking low T and PD. Recent work has shown that comorbidities such as ED, hypertension, and diabetes are commonly linked to both low T and PD.20,23,29 Among those, ED is the most common comorbidity for both, and its most common cause is penile vascular insufficiency. This is usually part of a generalized endothelial dysfunction and is related to several conditions, including diabetes mellitus, hypertension, hyperlipidemia, and obesity. These conditions underlie the pathophysiology of metabolic syndrome. Low T has been found to be an integral component of the pathology underlying endothelial dysfunction and metabolic syndrome.38 One might posit that vascular risk factors might influence the development of PD by increasing susceptibility to penile trauma due to attempts at sexual intercourse with sub-maximal rigidity. Of note, the study by Moreno et al17 failed to use a multivariable analysis to define whether the actual T levels were a predictor of the magnitude of penile deformity or whether the underlying comorbidities (which by themselves may be associated with low T), were the link. In our study, we looked at the impact of the differences in curvature using a variety of measures of TT an FT, including raw scores, z-scores, normal vs abnormal TT (300 vs <300), TT quartiles, and, using a multivariable model for TT, FT using raw and z-scores. In none of the analyses were T levels associated with degree of penile deformity in men with PD. Our analysis is not without limitations; specifically, the study population is relatively small, and there is an absence of a control (non-PD) group. Despite including only 184 patients, the statistical analysis has more than sufficient power to detect small associations, and it is crystal clear in demonstrating no association between T level and degree of deformity. It is extremely unlikely that more patients would change the statistical outcomes. There is no consensus regarding within what period of time from the diagnosis of PD T levels should be measured; thus, a time period of 12 months was chosen, and this was done for 2 main reasons: (i) any duration longer than this may lead to the T levels not being representative, because T levels may have changed over this period of time (ii) windows shorter than this may have led to smaller patient numbers, because many of our patients have seen other urologists before seeing us and are >6 months of having PD at presentation. Another limitation is that J Sex Med 2019;16:1283e1289
our estimate of actuarial association between PD and HG assumes the PD study population in our analysis is representative of all PD patients. Our study cohort included 17% of patients <40 years of age, and 41% of patients had developed PD after radical pelvic surgery. In the absence of a population-based analysis, this issue cannot be resolved with any certainty. Furthermore, we excluded all patients with deformity other than curvature. Such patients were excluded primarily because deformity measurement in such cases is notoriously challenging. For example, there exists no standard means of assessing degree of indentation or hourglass deformity. On the other hand, the strengths of the study include a robust definition of PD (all patients were examined by a highly experienced PD expert), rigorous assessment of penile deformity using intracavernosal injections, highly sophisticated statistical analysis and most importantly a structured measurement of serum T levels as it pertains to circadian rhythm (all laboratory assessments done before noon).
CONCLUSIONS This study demonstrates that there is no relationship between T levels and degree of deformity in PD patients. However, neither the Moreno study nor ours can answer whether HG is associated with PD because they are both non-controlled. Further prospective studies are needed to ascertain the relationship. Corresponding Author: John P. Mulhall, MD, MSc, FECSM, FACS, Sexual and Reproductive Medicine Program, Urology Service, 353 E 68th Street, New York, NY 10065, USA. Tel: (646) 422-4359; Fax: (212) 988-0768; E-mail: mulhalj1@ mskcc.org Conflicts of Interest: The authors report no conflicts of interest. Funding: Supported by the Sidney Kimmel Center for Prostate and Urologic Cancers and the National Institutes of Health/ National Cancer Institute to Memorial Sloan Kettering Cancer Center through the Cancer Center Support Grant, award number P30 CA008748
STATEMENT OF AUTHORSHIP Category 1 (a) Conception and Design John P. Mulhall (b) Analysis of Data Christian J. Nelson (c) Interpretation of Data Kazuhito Matsushita, Christian J. Nelson Category 2 (a) Drafting the Article John P. Mulhall; Kazuhito Matsushita (b) Revising it critically for Important Intellectual Content John P. Mulhall
1288 Category 3
Mulhall et al
(a) Final Approval of the Version to be published John P. Mulhall; Kazuhito Matsushita; Christian J. Nelson
17. Moreno SA, Morgentaler A. Testosterone deficiency and Peyronie’s disease: Pilot data suggesting a significant relationship. J Sex Med 2009;6:1729-1735.
REFERENCES
18. Mulhall JP, Anderson MS, Lubrano T, et al. Peyronie’s disease cell culture models: phenotypic, genotypic and functional analyses. Int J Impot Res 2002;14:397-405.
1. Chung E, Ralph D, Kadioglu A, et al. Evidence-based management of Peyronie’s disease. J Sex Med 2016;13:905-923. 2. Mulhall JP, Alex B, Choi JM. Predicting delay in presentation in men with Peyronie’s disease. J Sex Med 2010;7:2226-2230. 3. Mulhall JP. Expanding the paradigm for plaque development in Peyronie’s disease. Int J Impot Res 2003;15(Suppl. 5):S93-S102. 4. Devine CJ Jr, Somers KD, Jordan SG, et al. Proposal: Trauma as the cause of the Peyronie’s lesion. J Urol 1997; 157:285-290. 5. Brock G, Hsu GL, Nunes L, et al. The anatomy of the tunica albuginea in the normal penis and Peyronie’s disease. J Urol 1997;157:276-281. 6. Mulhall JP, Creech SD, Boorjian SA, et al. Subjective and objective analysis of the prevalence of Peyronie’s disease in a population of men presenting for prostate cancer screening. J Urol 2004;171:2350-2353.
19. LaRochelle JC, Levine LA. A Survey of primary-care physicians and urologists regarding Peyronie’s disease. J Sex Med 2007; 4:1167-1173. 20. El-Sakka AI. Prevalence of Peyronie’s disease among patients with erectile dysfunction. Eur Urol 2006;49:564-569. 21. El-Sakka AI, Tayeb KA. Vascular impairment of erection in patients with diabetes and Peyronie’s disease: Is that accumulative? J Sex Med 2009;6:1736-1742. 22. El-Sakka AI, Tayeb KA. Peyronie’s disease in diabetic patients being screened for erectile dysfunction. J Urol 2005; 174:1026-1030. 23. Usta MF, Bivalacqua TJ, Jabren GW, et al. Relationship between the severity of penile curvature and the presence of comorbidities in men with Peyronie’s disease. J Urol 2004; 171:775-779.
7. Sommer F, Schwarzer U, Wassmer G, et al. Epidemiology of Peyronie’s disease. Int J Impot Res 2002;14:379-383.
24. Kadioglu A, Tefekli A, Erol B, et al. A retrospective review of 307 men with Peyronie’s disease. J Urol 2002; 168:1075-1079.
8. Schwarzer U, Sommer F, Klotz T, et al. The prevalence of Peyronie’s disease: Results of a large survey. BJU Int 2001; 88:727-730.
25. Tal R, Heck M, Teloken P, et al. Peyronie’s disease following radical prostatectomy: incidence and predictors. J Sex Med 2010;7:1254-1261.
9. Rhoden EL, Teloken C, Ting HY, et al. Prevalence of Peyronie’s disease in men over 50-y-old from Southern Brazil. Int J Impot Res 2001;13:291-293.
26. U.S. Census Bureau. Population; Available from: http://www. census.gov/compendia/statab/cats/population.html. Accessed June 8, 2011.
10. Lindsay MB, Schain DM, Grambsch P, et al. The incidence of Peyronie’s disease in Rochester, Minnesota, 1950 through 1984. J Urol 1991;146:1007-1009.
27. Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy in men with androgen deficiency syndromes: An Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2010;95:2536-2559.
11. La Pera G, Pescatori ES, Calabrese M, et al. Peyronie’s disease: prevalence and association with cigarette smoking. A multicenter population-based study in men aged 50-69 years. Eur Urol 2001;40:525-530. 12. Deveci S, Hopps CV, O’Brien K, et al. Defining the clinical characteristics of Peyronie’s disease in young men. J Sex Med 2007;4:485-490. 13. Dandona P, Rosenberg MT. A practical guide to male hypogonadism in the primary care setting. Int J Clin Pract 2010; 64:682-696. 14. Araujo AB, Esche GR, Kupelian V, et al. Prevalence of symptomatic androgen deficiency in men. J Clin Endocrinol Metab 2007;92:4241-4247. 15. Iacono F, Prezioso D, Ruffo A, et al. Testosterone deficiency causes penile fibrosis and organic erectile dysfunction in aging men. Evaluating association among Age, TDS and ED. BMC Surg 2012;12(Suppl. 1):S24. 16. Cavallini G, Biagiotti G, Lo Giudice C. Association between peyronie disease and low serum testosterone levels: Detection and therapeutic considerations. J Androl 2012;33:381-388.
28. Wang C, Nieschlag E, Swerdloff R, et al. Investigation, treatment and monitoring of late-onset hypogonadism in males: ISA, ISSAM, EAU, EAA and ASA recommendations. Eur J Endocrinol 2008;159:507-514. 29. Mulligan T, Frick MF, Zuraw QC, et al. Prevalence of hypogonadism in males aged at least 45 years: the HIM study. Int J Clin Pract 2006;60:762-769. 30. Harman SM, Metter EJ, Tobin JD, et al. Longitudinal effects of aging on serum total and free testosterone levels in healthy men. Baltimore Longitudinal Study of Aging. J Clin Endocrinol Metab 2001;86:724-731. 31. Wang C, Nieschlag E, Swerdloff R, et al. Investigation, treatment, and monitoring of late-onset hypogonadism in males: ISA, ISSAM, EAU, EAA, and ASA recommendations. J Androl 2009;30:1-9. 32. Bhasin S, Cunningham GR, Hayes FJ, et al. Testosterone therapy in adult men with androgen deficiency syndromes: an endocrine society clinical practice guideline. J Clin Endocrinol Metab 2006;91:1995-2010. J Sex Med 2019;16:1283e1289
Testosterone & Peyronie’s Disease
1289
33. Morgentaler A. Commentary: Guideline for male testosterone therapy: A clinician’s perspective. J Clin Endocrinol Metab 2007;92:416-417.
36. Muller A, Mulhall JP. Peyronie’s disease intervention trials: Methodological challenges and issues. J Sex Med 2009; 6:848-861.
34. Diver MJ, Imtiaz KE, Ahmad AM, Vora JP, Fraser WD. Diurnal rhythms of serum total, free and bioavailable testosterone and of SHBG in middle-aged men compared with those in young men. Clin Endocrinol (Oxf) 2003;58:710-717.
37. Ohebshalom M, Mulhall J, Guhring P, et al. Measurement of penile curvature in Peyronie’s disease patients: comparison of three methods. J Sex Med 2007;4:199-203.
35. Rhoden EL, Morgentaler A. Risks of testosterone-replacement therapy and recommendations for monitoring. N Engl J Med 2004;350:482-492.
38. Guay AT, Traish A. Testosterone deficiency and risk factors in the metabolic syndrome: implications for erectile dysfunction. Urol Clin North Am 2011;38:175-183.
J Sex Med 2019;16:1283e1289