Reconstructive Urology National Patterns of Urethral Evaluation and Risk Factors for Urethral Injury in Patients With Penile Fracture Joseph J. Pariser, Shane M. Pearce, Sanjay G. Patel, and Gregory T. Bales OBJECTIVE MATERIALS AND METHODS
RESULTS
CONCLUSION
To examine the epidemiology and timing of penile fracture, patterns of urethral evaluation, and risk factors for concomitant urethral injury. The National Inpatient Sample (2003-2011) was used to identify patients with penile fractures. Clinical data included age, race, comorbidity, insurance, hospital factors, timing, hematuria, and urinary symptoms. Rates of formal urethral evaluation (cystoscopy or urethrogram) and urethral injury were calculated. Multivariate logistic regression was used to identify predictors of urethral evaluation and risk factors for urethral injury. A weighted population of 3883 patients with penile fracture was identified. Presentations during weekends (37%) and summers (30%) were overrepresented (both P 41 years: OR ¼ 2.25; 95% CI, 1.25-4.05; P ¼ .007), black race (OR ¼ 1.93; 95% CI, 1.12-3.34; P ¼ .018), and hematuria (OR ¼ 17.03; 95% CI, 3.20-90.54; P ¼ .001) were independent risk factors for urethral injury. Penile fractures, which occur disproportionately during summer and weekends, were associated with a 21% risk of urethral injury. Urethral evaluations were performed in a minority of patients. Even in patients with hematuria, 55% of patients underwent formal urethral evaluation. On multivariate analysis of patients with penile fracture, hematuria as well as older age and black race were independently associated with concomitant urethral injury. UROLOGY 86: 181e186, 2015. 2015 Elsevier Inc.
enile fractures occur when the tunica albuginea of the corpora cavernosum ruptures. Certain sexual positions are associated with increased risk,1 and immediate operative repair is considered standard of care. The incidence of penile fracture in the United States has been estimated at 500-600 cases per year.2 Perhaps due to mechanism of penile fracture, estimates of concomitant urethral injury in the setting of penile fracture have been reported as low as 1% in the Middle East and Asia with rates as high as 38% in the United States.1-8 Sequelae of urethral injuries in the setting of penile fracture can include urethral stricture or pseudodiverticula.9,10 Evaluation of the urethra can be performed with a variety of techniques including cystoscopy or urethrography. Magnetic resonance imaging, which is occasionally used to confirm penile fractures, can also identify urethral injuries.11 If severe or readily apparent, direct intraoperative
P
visualization can also identify urethral tears during exploration. With grade B evidence, the AUA Guidelines for Urotrauma recommend urethral evaluation in patients with blood at the meatus, gross hematuria, or inability to void,12 yet controversy persists regarding urethral evaluation in the majority of patients without these particular signs or symptoms. Whereas numerous case studies and single-institution series regarding urethral injury during penile fracture exist, few specifically address or are powered to identify risk factors for urethral injury. By using a national database, we sought to describe patterns of urethral evaluation and add to the management strategy by focusing on risk factors for concomitant urethral injury. We hypothesized that certain patient demographic factors or clinical signs may be associated with a higher risk of urethral injury during penile fracture.
Financial Disclosure: The authors declare that they have no relevant financial interests. From the Section of Urology, University of Chicago, Chicago, IL Address correspondence to: Joseph J. Pariser, M.D., Section of Urology, University of Chicago, 5841 S. Maryland Ave. MC 6038, Chicago, IL 60637. E-mail: pariserj@ gmail.com Submitted: January 27, 2015, accepted (with revisions): March 10, 2015
MATERIALS AND METHODS
ª 2015 Elsevier Inc. All Rights Reserved
The National Inpatient Sample (NIS) represents roughly 20% of all inpatient admissions in the United States and represents the largest all-payer inpatient care database. It is a portion of a http://dx.doi.org/10.1016/j.urology.2015.03.039 0090-4295/15
181
group of datasets managed by the Healthcare Cost and Utilization Project. The dataset is weighted to allow population-level estimates of the sampled observations. The International Classification of Diseases, Ninth Edition Clinical Modification (ICD-9) was used to identify diagnoses and procedures. The 2003-2011 NIS database was queried to identify patients with the diagnosis of penile fracture (ICD-9: 959.13). Institutional review board approval was not required as no identifiable patient information was used. The ICD-9 codes used for analysis are shown in Appendix A. Only patients who underwent surgical repair were included in the analysis. Urethral evaluation included performance of cystoscopy or urethrography. Clinical signs and symptoms included hematuria or lower urinary tract symptoms (LUTS; obstruction, retention, incontinence, frequency, oliguria, or other urinary symptoms). Urethral injury was defined as having a formal diagnosis of urethral injury or undergoing urethral repair (including any patient undergoing urethroplasty). Demographics examined included age, race, insurance status, and timing of admission. Age categories were created using three equally distributed tertiles. For race, “other” consisted of patients who were Asian, American Indian, and those coded “other” in the NIS dataset. Timing was defined in terms of weekend or weekday admission as well as season of presentation. Winter included December to February, spring was March to May, summer was June to August, and fall was September to November. Comorbidity was calculated utilizing the Elixhauser method, which has been well validated and is incorporated into the NIS.13 Hospital characteristics included bed size (small, medium, or large based on NIS criteria), teaching status, and region. Length of stay was categorized as 2 days or >2 days, with the latter representing the 90th percentile. All statistical analyses were performed using Stata 13.1 (Statacorp, College Station, TX). Survey-weighting was used for all analyses where applicable. Comparison of means according to urethral injury status was performed using an adjusted Wald test. To compare categorical variables, a chi-squared test with Rao-Scott correction was used.14 A one-sample t test of proportion with the weighted proportion was used to test whether weekend admission was significant against an expected 2 of 7 (28.6%) days representing weekend days. A chi-square goodness of fit test was used to test whether season of admission showed an unequal distribution against an expected distribution of onequarter of admissions in each season. Weighted multivariate logistic regression was used to identify independent factors associated with undergoing urethral evaluation and similarly, independent risk factors for urethral injury. Tests were considered significant if P 2 days (15% vs 7%, P ¼ .002). Overall, weekend (1438 patients or 37%) and summer (1025 patients or 30%) presentations were overrepresented compared to expected distributions (P .2 for both weekend and season). Overall, a small proportion of patients were diagnosed with hematuria (3%) or LUTS (2%). Hematuria was more common in patients with urethral injury (10% vs 0.5%, P .1). Compared to whites, Hispanics (OR ¼ 0.42; 95% CI, 0.22-0.82; P ¼ .011) and other race (OR ¼ 0.33; 95% CI, 0.11-0.98; P ¼ .046) were less likely to undergo urethral evaluation with a cystoscopy or urethrogram. The presence of hematuria increased the likelihood of undergoing urethral evaluation (OR ¼ 2.99; 95% CI, 1.03-8.73; P ¼ .045). As a predictor for urethral evaluation, urinary symptoms did not reach significance (OR ¼ 2.02; 95% CI, 0.61-6.73; P ¼ .252). UROLOGY 86 (1), 2015
Table 1. Characteristics of patients with penile fracture with comparisons by presence of urethral injury Parameter Number of patients Age in y (mean standard error) Age by tertile 41 y Race White Black Hispanic Other Unknown Elixhauser Comorbidity 0 1 2 Hospital size Small Medium Large Hospital teaching status Nonteaching Teaching Hospital region Northeast Midwest South West Primary payer Medicare Medicaid Private Other (eg, self-pay) Urethral evaluation Cystoscopy Retrograde urethrogram Hematuria Urinary symptoms Length of stay 2 d >2 d Weekend admission Season of admission Winter Spring Summer Fall
Total
No Urethral Injury
Urethral Injury
P Value
3883 (100%) 37.0 0.4
3070 (79%) 36.4 0.5
813 (21%) 39.4 0.8
— .1).
Table 3. Weighted multivariate logistic regression of risk factors of urethral injury* Factor
OR
95% CI
Age (y) Youngest tertile: 41 2.25 1.25-4.05 Race White Reference Reference Black 1.93 1.12-3.34 Hispanic 0.65 0.27-1.57 Other 0.70 0.21-2.31 Unknown 1.54 0.86-2.76 Hematuria 17.03 3.20-90.54 Urinary symptoms 2.24 0.67-7.46 Urethral evaluation 2.27 1.38-3.72
P Value — .027 .007 — .018 .333 .553 .145 .001 .188 .001
Abbreviations as in Table 2. * While controlling for Elixhauser comorbidity, hospital teaching status, hospital size, region, insurance, and timing (all P >.1).
of penile fracture to date. We studied the epidemiology of penile fracture and patterns of care for urethral evaluation and identified factors independently associated with urethral injury. Penile fracture is known to predominately occur in young, healthy men capable of relatively vigorous sexual activity, which is confirmed in our study with a mean age of 37 years and 79% of patients having no comorbidities. In our study, black race (22%) was overrepresented compared to the expected proportion given the 2010 US national census data (13%) in spite of unknown racial data for 736 (19%) patients.15 From a behavioral research standpoint, it is interesting to note that penile fractures disproportionately occur during weekends (37%) and summer months (30%). A study from Iran by Moslemi also demonstrated a trend toward increased incidence in summer months although the most common mechanism of injury being “manual habitual trauma” or taghaandan.16 Although our study lacks detailed information regarding mechanism of penile fracture, the mechanism is most likely intercourse, as this is the most common cause in Western series.1,7 Summer 184
months have also been shown to be overrepresented in epidemiologic studies of priapism.17,18 Perhaps the observed summer preponderance represents a ubiquitous increase in overall sexual behavior. We found a 21% rate of a urethral injury associated with penile fracture. Very small reports in the United States of 8 patients each have demonstrated somewhat higher rates of urethral injury of up to 38%.7,8 In Europe, rates of 14%-28% have been reported.1,19 Alternatively, series from Asia and Iran demonstrate a much lower rate of urethral injury (0%-2%).3,6,16 Our series confirms the higher rate of urethral injury in the United States, which is likely related to mechanism of injury. Previous series point to hematuria, urinary symptoms, or severity of penile fracture as risk factors for urethral injury and therefore recommend evaluation (by either urethrogram or cystoscopy) in these individuals.1,8,19,20 In our study, patients with hematuria (OR ¼ 2.99; 95% CI, 1.03-8.73; P ¼ .045) were more likely to undergo urethral evaluation. However, only 55% of patients with hematuria underwent urethral evaluation. This may have been a result of undercoding of hematuria or the relatively recent (August 2014) publication of guidelines.12 Regardless, 85% of patients with hematuria were ultimately diagnosed with urethral injury. This discrepancy may also represent intraoperative identification of urethral involvement on penile exploration, especially if the surgeon had an appropriately heightened suspicion of urethral injury in the setting of hematuria. Cystoscopy was performed over 3 times more often than urethrogram. This may represent urologist preference or availability of equipment given the emergent nature of penile fracture. Urethral evaluation was independently associated with race. Hispanic patients and other (eg, Asian, American Indian, etc) races were less likely to undergo evaluation than white patients. This may represent a disparity in care or could be related to hospital- and/or provider-level factors not accounted for in this study. Some minor, unrecognized urethral injuries associated with penile fracture may heal without repair. Muentener et al21 reported 29 penile fractures, in which 17 (59%) patients were treated conservatively. In total, urethral injury was identified in 9 (31%) patients. One-third of patients (2 of 6) with known urethral injury treated conservatively had suboptimal urinary outcomes in the form of urethral stricture or obstructive symptoms. In comparison, no patients (0 of 3) had urinary complaints after undergoing immediate repair. Early repair minimizes urine extravasation and related fibrosis, potentially avoiding negative outcomes. Additionally, patients in our study were more likely to experience a prolonged hospitalization if diagnosed with a urethral injury, further emphasizing the potential morbidity of urethral involvement. On multivariate analysis, 2 novel risk factors for urethral injury associated with penile fracture were identified: older age and black race. Potential explanations UROLOGY 86 (1), 2015
include differences in mechanism of penile fracture and anatomic variation associated with age and race. Although it is clear that undergoing a urethral evaluation increased the likelihood of being diagnosed with an injury, this did not explain the age and racial associations with urethral injuries. Older patients were not more likely to undergo evaluation, and there was even a trend toward fewer evaluations in black patients. Additionally, our findings corroborate previous reports that hematuria is a risk factor for concomitant urethral injury, which underscores the need for evaluation in these patients. Urinary symptoms were not associated with urethral injury, but this may have represented undercoding given the very low rate of LUTS (2%). Our findings add to the relatively limited literature on patient selection for formal urethral evaluation in the setting of penile fracture. Whereas some surgeons may consider a 21% rate of urethral injury sufficient to support routine evaluation, others may argue that many urethral injuries may be found during exploration. Therefore, determining the rate of missed urethral injury by omitting formal evaluation remains a key question. Our findings allow for further risk stratification based on demographic data while controlling for clinical signs such as hematuria. At a minimum, our results suggest that surgeons should have a higher suspicion for urethral injury in certain patients based on age and race. Although this study represents the largest series of penile fractures and the first to identify age and race as risk factors for urethral injury, some limitations should be noted. Claims-based data are dependent on appropriate coding. As an example, urinary symptoms were only present in 87 (2%) patients, which may have limited its analysis as a risk factor. One reassuring finding was that the rate of urethral injury appeared to be consistent with that of previously published results. As only a minority (23%) of patients underwent formal urethral evaluation, the true rate of urethral injury remains unknown. A significant proportion of patients had missing racial data, which may have limited analysis. Other limitations of the database include a lack of information regarding the mechanism of penile fracture, specific details regarding operative findings, urethral evaluations performed other than cystoscopy or urethrography such as magnetic resonance imaging, type of repair performed, and long-term outcomes. Finally, some penile fractures may have been managed in an outpatient setting, which would have not been captured using the NIS dataset and would result in an underestimation of the true incidence of penile fracture.
CONCLUSION Penile fractures were associated with a significant risk of urethral injury and occurred disproportionately during summer and weekends. Urethral evaluations were performed in a minority of patients. Even in the setting of hematuria, only 55% of patients underwent formal urethral evaluation. On multivariate analysis of patients with UROLOGY 86 (1), 2015
penile fracture, hematuria as well as older age and black race were independently associated with concomitant urethral injury. References 1. Reis LO, Cartapatti M, Marmiroli R, et al. Mechanisms predisposing penile fracture and long-term outcomes on erectile and voiding functions. Adv Urol. 2014;2014:1-4. DOI:10.1155/2014/768158. 2. Aaronson DS, Shindel AW. U.S. national statistics on penile fracture. J Sex Med. 2010;7:3226. 3. Ishikawa T, Fujisawa M, Tamada H, et al. Fracture of the penis: nine cases with evaluation of reported cases in Japan. Int J Urol. 2003;10: 257-260. 4. Rivas JG, Dorrego JM, Hernandez MM, et al. Traumatic rupture of the corpus cavernosum: surgical management and clinical outcomes. A 30 years review. Cent European J Urol. 2014;67:88-92. 5. Amit A, Arun K, Bharat B, et al. Penile fracture and associated urethral injury: experience at a tertiary care hospital. Can Urol Assoc. 2013;7:E168-E170. 6. Zargooshi J. Penile fracture in Kermanshah, Iran: report of 172 cases. J Urol. 2000;164:364-366. 7. Fergany AF, Angermeier KW, Montague DK. Review of Cleveland Clinic experience with penile fracture. Urology. 1999;54:352-355. 8. Kamdar C, Mooppan UMM, Kim H, et al. Penile fracture: preoperative evaluation and surgical technique for optimal patient outcome. BJU Int. 2008;102:1640-1644; discussion 1644. 9. Raheem AA, El-Tatawy H, Eissa A, et al. Urinary and sexual functions after surgical treatment of penile fracture concomitant with complete urethral disruption. Arch Ital Urol Androl. 2014;86:15-19. 10. Di Pierro GB, Iannotta L, Innocenzi M, et al. Urethral pseudodiverticulum secondary to penile fracture and complete urethra dissection. Can Urol Assoc J. 2013;7:E347-E350. 11. Maubon AJ, Roux JO, Faix A, et al. Penile fracture: MRI demonstration of a urethral tear associated with a rupture of the corpus cavernosum. Eur Radiol. 1998;8:469-470. 12. Morey AF, Brandes S, Dugi DD, et al. Urotrauma: AUA guideline. J Urol. 2014;192:327-335. 13. Southern DA, Quan H, Ghali WA. Comparison of the Elixhauser and Charlson/Deyo methods of comorbidity measurement in administrative data. Med Care. 2004;42:355-360. 14. Rao J, Scott A. On chi-squared tests for multiway contingency tables with cell proportions estimated from survey data. Ann Stat. 1984;12:46-60. 15. Humes KR, Jones NA, Ramirez RR. Overview of Race and Hispanic Origin: 2010. 2011. 16. Moslemi MK. Evaluation of epidemiology, concomitant urethral disruption and seasonal variation of penile fracture: a report of 86 cases. Can Urol Assoc J. 2013;7:E572-E575. 17. Roghmann F, Becker A, Sammon JD, et al. Incidence of priapism in emergency departments in the United States. J Urol. 2013;190: 1275-1280. 18. Stein DM, Flum AS, Cashy J, et al. Nationwide emergency department visits for priapism in the United States. J Sex Med. 2013;10:2418-2422. 19. Hatzichristodoulou G, Dorstewitz A, Gschwend JE, et al. Surgical management of penile fracture and long-term outcome on erectile function and voiding. J Sex Med. 2013;10:1424-1430. 20. Jack GS, Garraway I, Reznichek R, et al. Current treatment options for penile fractures. Rev Urol. 2004;6:114-120. 21. Muentener M, Suter S, Hauri D, et al. Long-term experience with surgical and conservative treatment of penile fracture. J Urol. 2004; 172:576-579.
APPENDIX SUPPLEMENTARY DATA
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.urology. 2015.03.039. 185
EDITORIAL COMMENT The authors queried the National Inpatient Sample to determine the patterns of urethral injury evaluation in penile fracture patients.1 They should be commended as research in this field is mainly restricted to single-institution series. Three percent of patients presented with gross hematuria with a urethral injury rate of 85%. The most alarming finding in this study is that only 55% of those patients underwent an evaluation of injury to the urethra. This is despite the fact that gross hematuria is a known indicator of injury to the urethra.2 Of all patients identified with urethral injuries, 66% had no formal evaluation, suggesting that the majority were discovered intraoperatively. In August 2014, the American Urological Association released guidelines on urotrauma which include penile fracture.3 Statement number 29 indicates, “Clinicians must perform evaluation for concomitant urethral injury in patients with penile fracture or penetrating trauma who present with blood at the urethral meatus, gross hematuria, or inability to void (standard; evidence strength B).” Patients were sampled in this series from 2003-2011. The lack of workup for urethral injuries may reflect older practice patterns which could have changed before the publication of the guidelines. This seems doubtful given the relatively short time span and the historically wellestablished relationship of hematuria to urethral injury. It raises the question, why would high-risk patients not have a diagnostic workup? First, there may be coding errors in the dataset which do not capture the relevant urethral studies. Alternatively, there may be a high index of suspicion based on clinical findings and determination that a workup is not needed. Finally, the injury may be discovered incidentally at the time of surgery, which has the potential to impact both patient counseling and surgical planning. Bullock et al4 showed that nearly 58% of board-certified urologists in the United States do not perform urethroplasties. Urethral repair at the time of fracture is not as technically challenging compared to a formal urethroplasty, but one has to wonder whether the early detection of urethral injury would change the management of patients. One such change is the transfer of patients to centers with more expertise in urethral surgery. Adherence to evidenced-based, national, guidelines should be the gold standard in the majority of patients. Although exceptions do occur, a large percentage of patients are not meeting the recommended guidelines, leaving room for improvement in the delivery of care. Joshua A. Broghammer, M.D., F.A.C.S., Department of Urology, University of Kansas Medical Center, Kansas City, KS
References 1. Pariser JJ, Pearce SM, Patel SG, Bales GT. National patterns of urethral evaluation and risk factors for urethral injury in patients with penile fracture. Urology. 2015;86:181-186. 2. Derouiche A, Belhaj K, Hentati H, et al. Management of penile fractures complicated by urethral rupture. Int J Impot Res. 2008;20: 111-114. 3. Morey AF, Brandes S, Dugi DD 3rd, et al; American Urological Association. Urotrauma: AUA guideline. J Urol. 2014;192:327-335.
186
4. Bullock TL, Brandes SB. Adult anterior urethral strictures: a national practice patterns survey of board certified urologists in the United States. J Urol. 2007;177:685-690.
http://dx.doi.org/10.1016/j.urology.2015.03.040 UROLOGY 86: 186, 2015. 2015 Elsevier Inc.
REPLY We appreciate the editorial comment regarding the underutilization of formal urethral evaluation in certain high-risk clinical settings such as hematuria.1 As mentioned, only 55% of patients underwent urethroscopy or urethrogram with 85% of these patients ultimately being diagnosed with a urethral injury. We agree that possible explanations include undercoding of evaluations or identification during exploration. It is our impression that some practitioners do not consider preoperative diagnostic workup necessary given that many of the urethral tears can be identified in close proximity to the corporal injury. However, we highlight that this would be in disagreement with American Urological Association guidelines on urotrauma.2 In our experience, one challenging aspect of the management of penile fracture and possible concomitant urethral involvement remains the localization of injuries. Especially in severe cases, a sizable hematoma may obfuscate the surgical field and predispose patients to missed urethral injuries. As an additional benefit of formal urethral evaluation before exploration, identification of the site of a urethral injury can direct the surgeon to the site of the corporal fracture. Our study demonstrated an overall urethral injury rate of 21% for patients with penile fracture. Some may consider this finding sufficient to warrant routine evaluation for all patients. As additional factors to aid in risk stratification, we found that older age and black race were associated with an increased risk of urethral injury. For patients with penile fracture, we tend to favor the liberal use of flexible cystoscopy immediately before exploration to minimize the risk of a missed diagnosis of a concomitant urethral injury, which could lead to negative outcomes if not repaired.3 Joseph J. Pariser, M.D., Shane M. Pearce, M.D., Sanjay G. Patel, M.D., and Gregory T. Bales, M.D., Section of Urology, University of Chicago, Chicago, IL
References 1. Broghammer JA. National patterns of urethral evaluation and risk factors for urethral injury in patients with penile fracture [Editorial Comment]. Urology. 2015;86:181-186. 2. Morey AF, Brandes S, Dugi DD 3rd, et al; American Urological Association. Urotrauma: AUA guideline. J Urol. 2014;192:327-335. 3. Muentener M, Suter S, Hauri D, et al. Long-term experience with surgical and conservative treatment of penile fracture. J Urol. 2004; 172:576-579.
http://dx.doi.org/10.1016/j.urology.2015.03.041 UROLOGY 86: 186, 2015. 2015 Elsevier Inc.
UROLOGY 86 (1), 2015
RESULTS
CONCLUSION
To examine the epidemiology and timing of penile fracture, patterns of urethral evaluation, and risk factors for concomitant urethral injury. The National Inpatient Sample (2003-2011) was used to identify patients with penile fractures. Clinical data included age, race, comorbidity, insurance, hospital factors, timing, hematuria, and urinary symptoms. Rates of formal urethral evaluation (cystoscopy or urethrogram) and urethral injury were calculated. Multivariate logistic regression was used to identify predictors of urethral evaluation and risk factors for urethral injury. A weighted population of 3883 patients with penile fracture was identified. Presentations during weekends (37%) and summers (30%) were overrepresented (both P 41 years: OR ¼ 2.25; 95% CI, 1.25-4.05; P ¼ .007), black race (OR ¼ 1.93; 95% CI, 1.12-3.34; P ¼ .018), and hematuria (OR ¼ 17.03; 95% CI, 3.20-90.54; P ¼ .001) were independent risk factors for urethral injury. Penile fractures, which occur disproportionately during summer and weekends, were associated with a 21% risk of urethral injury. Urethral evaluations were performed in a minority of patients. Even in patients with hematuria, 55% of patients underwent formal urethral evaluation. On multivariate analysis of patients with penile fracture, hematuria as well as older age and black race were independently associated with concomitant urethral injury. UROLOGY 86: 181e186, 2015. 2015 Elsevier Inc.
enile fractures occur when the tunica albuginea of the corpora cavernosum ruptures. Certain sexual positions are associated with increased risk,1 and immediate operative repair is considered standard of care. The incidence of penile fracture in the United States has been estimated at 500-600 cases per year.2 Perhaps due to mechanism of penile fracture, estimates of concomitant urethral injury in the setting of penile fracture have been reported as low as 1% in the Middle East and Asia with rates as high as 38% in the United States.1-8 Sequelae of urethral injuries in the setting of penile fracture can include urethral stricture or pseudodiverticula.9,10 Evaluation of the urethra can be performed with a variety of techniques including cystoscopy or urethrography. Magnetic resonance imaging, which is occasionally used to confirm penile fractures, can also identify urethral injuries.11 If severe or readily apparent, direct intraoperative
P
visualization can also identify urethral tears during exploration. With grade B evidence, the AUA Guidelines for Urotrauma recommend urethral evaluation in patients with blood at the meatus, gross hematuria, or inability to void,12 yet controversy persists regarding urethral evaluation in the majority of patients without these particular signs or symptoms. Whereas numerous case studies and single-institution series regarding urethral injury during penile fracture exist, few specifically address or are powered to identify risk factors for urethral injury. By using a national database, we sought to describe patterns of urethral evaluation and add to the management strategy by focusing on risk factors for concomitant urethral injury. We hypothesized that certain patient demographic factors or clinical signs may be associated with a higher risk of urethral injury during penile fracture.
Financial Disclosure: The authors declare that they have no relevant financial interests. From the Section of Urology, University of Chicago, Chicago, IL Address correspondence to: Joseph J. Pariser, M.D., Section of Urology, University of Chicago, 5841 S. Maryland Ave. MC 6038, Chicago, IL 60637. E-mail: pariserj@ gmail.com Submitted: January 27, 2015, accepted (with revisions): March 10, 2015
MATERIALS AND METHODS
ª 2015 Elsevier Inc. All Rights Reserved
The National Inpatient Sample (NIS) represents roughly 20% of all inpatient admissions in the United States and represents the largest all-payer inpatient care database. It is a portion of a http://dx.doi.org/10.1016/j.urology.2015.03.039 0090-4295/15
181
group of datasets managed by the Healthcare Cost and Utilization Project. The dataset is weighted to allow population-level estimates of the sampled observations. The International Classification of Diseases, Ninth Edition Clinical Modification (ICD-9) was used to identify diagnoses and procedures. The 2003-2011 NIS database was queried to identify patients with the diagnosis of penile fracture (ICD-9: 959.13). Institutional review board approval was not required as no identifiable patient information was used. The ICD-9 codes used for analysis are shown in Appendix A. Only patients who underwent surgical repair were included in the analysis. Urethral evaluation included performance of cystoscopy or urethrography. Clinical signs and symptoms included hematuria or lower urinary tract symptoms (LUTS; obstruction, retention, incontinence, frequency, oliguria, or other urinary symptoms). Urethral injury was defined as having a formal diagnosis of urethral injury or undergoing urethral repair (including any patient undergoing urethroplasty). Demographics examined included age, race, insurance status, and timing of admission. Age categories were created using three equally distributed tertiles. For race, “other” consisted of patients who were Asian, American Indian, and those coded “other” in the NIS dataset. Timing was defined in terms of weekend or weekday admission as well as season of presentation. Winter included December to February, spring was March to May, summer was June to August, and fall was September to November. Comorbidity was calculated utilizing the Elixhauser method, which has been well validated and is incorporated into the NIS.13 Hospital characteristics included bed size (small, medium, or large based on NIS criteria), teaching status, and region. Length of stay was categorized as 2 days or >2 days, with the latter representing the 90th percentile. All statistical analyses were performed using Stata 13.1 (Statacorp, College Station, TX). Survey-weighting was used for all analyses where applicable. Comparison of means according to urethral injury status was performed using an adjusted Wald test. To compare categorical variables, a chi-squared test with Rao-Scott correction was used.14 A one-sample t test of proportion with the weighted proportion was used to test whether weekend admission was significant against an expected 2 of 7 (28.6%) days representing weekend days. A chi-square goodness of fit test was used to test whether season of admission showed an unequal distribution against an expected distribution of onequarter of admissions in each season. Weighted multivariate logistic regression was used to identify independent factors associated with undergoing urethral evaluation and similarly, independent risk factors for urethral injury. Tests were considered significant if P 2 days (15% vs 7%, P ¼ .002). Overall, weekend (1438 patients or 37%) and summer (1025 patients or 30%) presentations were overrepresented compared to expected distributions (P .2 for both weekend and season). Overall, a small proportion of patients were diagnosed with hematuria (3%) or LUTS (2%). Hematuria was more common in patients with urethral injury (10% vs 0.5%, P .1). Compared to whites, Hispanics (OR ¼ 0.42; 95% CI, 0.22-0.82; P ¼ .011) and other race (OR ¼ 0.33; 95% CI, 0.11-0.98; P ¼ .046) were less likely to undergo urethral evaluation with a cystoscopy or urethrogram. The presence of hematuria increased the likelihood of undergoing urethral evaluation (OR ¼ 2.99; 95% CI, 1.03-8.73; P ¼ .045). As a predictor for urethral evaluation, urinary symptoms did not reach significance (OR ¼ 2.02; 95% CI, 0.61-6.73; P ¼ .252). UROLOGY 86 (1), 2015
Table 1. Characteristics of patients with penile fracture with comparisons by presence of urethral injury Parameter Number of patients Age in y (mean standard error) Age by tertile 41 y Race White Black Hispanic Other Unknown Elixhauser Comorbidity 0 1 2 Hospital size Small Medium Large Hospital teaching status Nonteaching Teaching Hospital region Northeast Midwest South West Primary payer Medicare Medicaid Private Other (eg, self-pay) Urethral evaluation Cystoscopy Retrograde urethrogram Hematuria Urinary symptoms Length of stay 2 d >2 d Weekend admission Season of admission Winter Spring Summer Fall
Total
No Urethral Injury
Urethral Injury
P Value
3883 (100%) 37.0 0.4
3070 (79%) 36.4 0.5
813 (21%) 39.4 0.8
— .1).
Table 3. Weighted multivariate logistic regression of risk factors of urethral injury* Factor
OR
95% CI
Age (y) Youngest tertile: 41 2.25 1.25-4.05 Race White Reference Reference Black 1.93 1.12-3.34 Hispanic 0.65 0.27-1.57 Other 0.70 0.21-2.31 Unknown 1.54 0.86-2.76 Hematuria 17.03 3.20-90.54 Urinary symptoms 2.24 0.67-7.46 Urethral evaluation 2.27 1.38-3.72
P Value — .027 .007 — .018 .333 .553 .145 .001 .188 .001
Abbreviations as in Table 2. * While controlling for Elixhauser comorbidity, hospital teaching status, hospital size, region, insurance, and timing (all P >.1).
of penile fracture to date. We studied the epidemiology of penile fracture and patterns of care for urethral evaluation and identified factors independently associated with urethral injury. Penile fracture is known to predominately occur in young, healthy men capable of relatively vigorous sexual activity, which is confirmed in our study with a mean age of 37 years and 79% of patients having no comorbidities. In our study, black race (22%) was overrepresented compared to the expected proportion given the 2010 US national census data (13%) in spite of unknown racial data for 736 (19%) patients.15 From a behavioral research standpoint, it is interesting to note that penile fractures disproportionately occur during weekends (37%) and summer months (30%). A study from Iran by Moslemi also demonstrated a trend toward increased incidence in summer months although the most common mechanism of injury being “manual habitual trauma” or taghaandan.16 Although our study lacks detailed information regarding mechanism of penile fracture, the mechanism is most likely intercourse, as this is the most common cause in Western series.1,7 Summer 184
months have also been shown to be overrepresented in epidemiologic studies of priapism.17,18 Perhaps the observed summer preponderance represents a ubiquitous increase in overall sexual behavior. We found a 21% rate of a urethral injury associated with penile fracture. Very small reports in the United States of 8 patients each have demonstrated somewhat higher rates of urethral injury of up to 38%.7,8 In Europe, rates of 14%-28% have been reported.1,19 Alternatively, series from Asia and Iran demonstrate a much lower rate of urethral injury (0%-2%).3,6,16 Our series confirms the higher rate of urethral injury in the United States, which is likely related to mechanism of injury. Previous series point to hematuria, urinary symptoms, or severity of penile fracture as risk factors for urethral injury and therefore recommend evaluation (by either urethrogram or cystoscopy) in these individuals.1,8,19,20 In our study, patients with hematuria (OR ¼ 2.99; 95% CI, 1.03-8.73; P ¼ .045) were more likely to undergo urethral evaluation. However, only 55% of patients with hematuria underwent urethral evaluation. This may have been a result of undercoding of hematuria or the relatively recent (August 2014) publication of guidelines.12 Regardless, 85% of patients with hematuria were ultimately diagnosed with urethral injury. This discrepancy may also represent intraoperative identification of urethral involvement on penile exploration, especially if the surgeon had an appropriately heightened suspicion of urethral injury in the setting of hematuria. Cystoscopy was performed over 3 times more often than urethrogram. This may represent urologist preference or availability of equipment given the emergent nature of penile fracture. Urethral evaluation was independently associated with race. Hispanic patients and other (eg, Asian, American Indian, etc) races were less likely to undergo evaluation than white patients. This may represent a disparity in care or could be related to hospital- and/or provider-level factors not accounted for in this study. Some minor, unrecognized urethral injuries associated with penile fracture may heal without repair. Muentener et al21 reported 29 penile fractures, in which 17 (59%) patients were treated conservatively. In total, urethral injury was identified in 9 (31%) patients. One-third of patients (2 of 6) with known urethral injury treated conservatively had suboptimal urinary outcomes in the form of urethral stricture or obstructive symptoms. In comparison, no patients (0 of 3) had urinary complaints after undergoing immediate repair. Early repair minimizes urine extravasation and related fibrosis, potentially avoiding negative outcomes. Additionally, patients in our study were more likely to experience a prolonged hospitalization if diagnosed with a urethral injury, further emphasizing the potential morbidity of urethral involvement. On multivariate analysis, 2 novel risk factors for urethral injury associated with penile fracture were identified: older age and black race. Potential explanations UROLOGY 86 (1), 2015
include differences in mechanism of penile fracture and anatomic variation associated with age and race. Although it is clear that undergoing a urethral evaluation increased the likelihood of being diagnosed with an injury, this did not explain the age and racial associations with urethral injuries. Older patients were not more likely to undergo evaluation, and there was even a trend toward fewer evaluations in black patients. Additionally, our findings corroborate previous reports that hematuria is a risk factor for concomitant urethral injury, which underscores the need for evaluation in these patients. Urinary symptoms were not associated with urethral injury, but this may have represented undercoding given the very low rate of LUTS (2%). Our findings add to the relatively limited literature on patient selection for formal urethral evaluation in the setting of penile fracture. Whereas some surgeons may consider a 21% rate of urethral injury sufficient to support routine evaluation, others may argue that many urethral injuries may be found during exploration. Therefore, determining the rate of missed urethral injury by omitting formal evaluation remains a key question. Our findings allow for further risk stratification based on demographic data while controlling for clinical signs such as hematuria. At a minimum, our results suggest that surgeons should have a higher suspicion for urethral injury in certain patients based on age and race. Although this study represents the largest series of penile fractures and the first to identify age and race as risk factors for urethral injury, some limitations should be noted. Claims-based data are dependent on appropriate coding. As an example, urinary symptoms were only present in 87 (2%) patients, which may have limited its analysis as a risk factor. One reassuring finding was that the rate of urethral injury appeared to be consistent with that of previously published results. As only a minority (23%) of patients underwent formal urethral evaluation, the true rate of urethral injury remains unknown. A significant proportion of patients had missing racial data, which may have limited analysis. Other limitations of the database include a lack of information regarding the mechanism of penile fracture, specific details regarding operative findings, urethral evaluations performed other than cystoscopy or urethrography such as magnetic resonance imaging, type of repair performed, and long-term outcomes. Finally, some penile fractures may have been managed in an outpatient setting, which would have not been captured using the NIS dataset and would result in an underestimation of the true incidence of penile fracture.
CONCLUSION Penile fractures were associated with a significant risk of urethral injury and occurred disproportionately during summer and weekends. Urethral evaluations were performed in a minority of patients. Even in the setting of hematuria, only 55% of patients underwent formal urethral evaluation. On multivariate analysis of patients with UROLOGY 86 (1), 2015
penile fracture, hematuria as well as older age and black race were independently associated with concomitant urethral injury. References 1. Reis LO, Cartapatti M, Marmiroli R, et al. Mechanisms predisposing penile fracture and long-term outcomes on erectile and voiding functions. Adv Urol. 2014;2014:1-4. DOI:10.1155/2014/768158. 2. Aaronson DS, Shindel AW. U.S. national statistics on penile fracture. J Sex Med. 2010;7:3226. 3. Ishikawa T, Fujisawa M, Tamada H, et al. Fracture of the penis: nine cases with evaluation of reported cases in Japan. Int J Urol. 2003;10: 257-260. 4. Rivas JG, Dorrego JM, Hernandez MM, et al. Traumatic rupture of the corpus cavernosum: surgical management and clinical outcomes. A 30 years review. Cent European J Urol. 2014;67:88-92. 5. Amit A, Arun K, Bharat B, et al. Penile fracture and associated urethral injury: experience at a tertiary care hospital. Can Urol Assoc. 2013;7:E168-E170. 6. Zargooshi J. Penile fracture in Kermanshah, Iran: report of 172 cases. J Urol. 2000;164:364-366. 7. Fergany AF, Angermeier KW, Montague DK. Review of Cleveland Clinic experience with penile fracture. Urology. 1999;54:352-355. 8. Kamdar C, Mooppan UMM, Kim H, et al. Penile fracture: preoperative evaluation and surgical technique for optimal patient outcome. BJU Int. 2008;102:1640-1644; discussion 1644. 9. Raheem AA, El-Tatawy H, Eissa A, et al. Urinary and sexual functions after surgical treatment of penile fracture concomitant with complete urethral disruption. Arch Ital Urol Androl. 2014;86:15-19. 10. Di Pierro GB, Iannotta L, Innocenzi M, et al. Urethral pseudodiverticulum secondary to penile fracture and complete urethra dissection. Can Urol Assoc J. 2013;7:E347-E350. 11. Maubon AJ, Roux JO, Faix A, et al. Penile fracture: MRI demonstration of a urethral tear associated with a rupture of the corpus cavernosum. Eur Radiol. 1998;8:469-470. 12. Morey AF, Brandes S, Dugi DD, et al. Urotrauma: AUA guideline. J Urol. 2014;192:327-335. 13. Southern DA, Quan H, Ghali WA. Comparison of the Elixhauser and Charlson/Deyo methods of comorbidity measurement in administrative data. Med Care. 2004;42:355-360. 14. Rao J, Scott A. On chi-squared tests for multiway contingency tables with cell proportions estimated from survey data. Ann Stat. 1984;12:46-60. 15. Humes KR, Jones NA, Ramirez RR. Overview of Race and Hispanic Origin: 2010. 2011. 16. Moslemi MK. Evaluation of epidemiology, concomitant urethral disruption and seasonal variation of penile fracture: a report of 86 cases. Can Urol Assoc J. 2013;7:E572-E575. 17. Roghmann F, Becker A, Sammon JD, et al. Incidence of priapism in emergency departments in the United States. J Urol. 2013;190: 1275-1280. 18. Stein DM, Flum AS, Cashy J, et al. Nationwide emergency department visits for priapism in the United States. J Sex Med. 2013;10:2418-2422. 19. Hatzichristodoulou G, Dorstewitz A, Gschwend JE, et al. Surgical management of penile fracture and long-term outcome on erectile function and voiding. J Sex Med. 2013;10:1424-1430. 20. Jack GS, Garraway I, Reznichek R, et al. Current treatment options for penile fractures. Rev Urol. 2004;6:114-120. 21. Muentener M, Suter S, Hauri D, et al. Long-term experience with surgical and conservative treatment of penile fracture. J Urol. 2004; 172:576-579.
APPENDIX SUPPLEMENTARY DATA
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.urology. 2015.03.039. 185
EDITORIAL COMMENT The authors queried the National Inpatient Sample to determine the patterns of urethral injury evaluation in penile fracture patients.1 They should be commended as research in this field is mainly restricted to single-institution series. Three percent of patients presented with gross hematuria with a urethral injury rate of 85%. The most alarming finding in this study is that only 55% of those patients underwent an evaluation of injury to the urethra. This is despite the fact that gross hematuria is a known indicator of injury to the urethra.2 Of all patients identified with urethral injuries, 66% had no formal evaluation, suggesting that the majority were discovered intraoperatively. In August 2014, the American Urological Association released guidelines on urotrauma which include penile fracture.3 Statement number 29 indicates, “Clinicians must perform evaluation for concomitant urethral injury in patients with penile fracture or penetrating trauma who present with blood at the urethral meatus, gross hematuria, or inability to void (standard; evidence strength B).” Patients were sampled in this series from 2003-2011. The lack of workup for urethral injuries may reflect older practice patterns which could have changed before the publication of the guidelines. This seems doubtful given the relatively short time span and the historically wellestablished relationship of hematuria to urethral injury. It raises the question, why would high-risk patients not have a diagnostic workup? First, there may be coding errors in the dataset which do not capture the relevant urethral studies. Alternatively, there may be a high index of suspicion based on clinical findings and determination that a workup is not needed. Finally, the injury may be discovered incidentally at the time of surgery, which has the potential to impact both patient counseling and surgical planning. Bullock et al4 showed that nearly 58% of board-certified urologists in the United States do not perform urethroplasties. Urethral repair at the time of fracture is not as technically challenging compared to a formal urethroplasty, but one has to wonder whether the early detection of urethral injury would change the management of patients. One such change is the transfer of patients to centers with more expertise in urethral surgery. Adherence to evidenced-based, national, guidelines should be the gold standard in the majority of patients. Although exceptions do occur, a large percentage of patients are not meeting the recommended guidelines, leaving room for improvement in the delivery of care. Joshua A. Broghammer, M.D., F.A.C.S., Department of Urology, University of Kansas Medical Center, Kansas City, KS
References 1. Pariser JJ, Pearce SM, Patel SG, Bales GT. National patterns of urethral evaluation and risk factors for urethral injury in patients with penile fracture. Urology. 2015;86:181-186. 2. Derouiche A, Belhaj K, Hentati H, et al. Management of penile fractures complicated by urethral rupture. Int J Impot Res. 2008;20: 111-114. 3. Morey AF, Brandes S, Dugi DD 3rd, et al; American Urological Association. Urotrauma: AUA guideline. J Urol. 2014;192:327-335.
186
4. Bullock TL, Brandes SB. Adult anterior urethral strictures: a national practice patterns survey of board certified urologists in the United States. J Urol. 2007;177:685-690.
http://dx.doi.org/10.1016/j.urology.2015.03.040 UROLOGY 86: 186, 2015. 2015 Elsevier Inc.
REPLY We appreciate the editorial comment regarding the underutilization of formal urethral evaluation in certain high-risk clinical settings such as hematuria.1 As mentioned, only 55% of patients underwent urethroscopy or urethrogram with 85% of these patients ultimately being diagnosed with a urethral injury. We agree that possible explanations include undercoding of evaluations or identification during exploration. It is our impression that some practitioners do not consider preoperative diagnostic workup necessary given that many of the urethral tears can be identified in close proximity to the corporal injury. However, we highlight that this would be in disagreement with American Urological Association guidelines on urotrauma.2 In our experience, one challenging aspect of the management of penile fracture and possible concomitant urethral involvement remains the localization of injuries. Especially in severe cases, a sizable hematoma may obfuscate the surgical field and predispose patients to missed urethral injuries. As an additional benefit of formal urethral evaluation before exploration, identification of the site of a urethral injury can direct the surgeon to the site of the corporal fracture. Our study demonstrated an overall urethral injury rate of 21% for patients with penile fracture. Some may consider this finding sufficient to warrant routine evaluation for all patients. As additional factors to aid in risk stratification, we found that older age and black race were associated with an increased risk of urethral injury. For patients with penile fracture, we tend to favor the liberal use of flexible cystoscopy immediately before exploration to minimize the risk of a missed diagnosis of a concomitant urethral injury, which could lead to negative outcomes if not repaired.3 Joseph J. Pariser, M.D., Shane M. Pearce, M.D., Sanjay G. Patel, M.D., and Gregory T. Bales, M.D., Section of Urology, University of Chicago, Chicago, IL
References 1. Broghammer JA. National patterns of urethral evaluation and risk factors for urethral injury in patients with penile fracture [Editorial Comment]. Urology. 2015;86:181-186. 2. Morey AF, Brandes S, Dugi DD 3rd, et al; American Urological Association. Urotrauma: AUA guideline. J Urol. 2014;192:327-335. 3. Muentener M, Suter S, Hauri D, et al. Long-term experience with surgical and conservative treatment of penile fracture. J Urol. 2004; 172:576-579.
http://dx.doi.org/10.1016/j.urology.2015.03.041 UROLOGY 86: 186, 2015. 2015 Elsevier Inc.
UROLOGY 86 (1), 2015
