PEDIATRIC UROLOGY Predictivity of Clinical Findings and Doppler Ultrasound in Pediatric Acute Scrotum. Riccardo Lemini1, Riccardo Guanà1*, Nicola Tommasoni1, Alessandro Mussa2, Gianpaolo Di Rosa3, Jurgen Schleef1 Purpose: To evaluate the role of Doppler ultrasonography (DUS) in diagnosing pediatric testicular torsion (TT), and its diagnostic accuracy, and helping clinicians increase specificity and decrease negative exploration rates. Materials and Methods: We performed a retrospective study of all consecutive patients with acute testicular symptoms referring to our pediatric emergency department (ED) from January 2010 to December 2013. Results: We analyzed 1091 patients, with a mean age of 9 years. DUS was performed in 498 patients (40.8%); 107 patients (8.8%) underwent surgery and 41 patients (3.3%) had a TT. The following clinical findings were collected: presence of scrotal pain, erythema and swelling, spermatic cord pain and abnormal cremasteric reflex. The clinical findings significantly associated with TT were spermatic cord pain (OR = 37, 95% CI: 11.9-111.1, P < .001) and abnormal cremasteric reflex (OR = 47.6, 95% CI: 13.5-166.6, P < .001); the presence of swelling was not associated with TT (OR = 2.3, 95% CI: 0.7-8.4, P < .001). Scrotal pain was not significantly associated with TT (P = .9), while erythema made TT unlikely (OR = 0.22, 95% CI: 0.07-0.7, P = .0445). In all cases the DUS significantly increased the predictivity. Conclusion: TT was present in 3.3% of patients presenting with testicular symptoms. The predictivity based on clinical findings resulted high and the negative exploration rate for TT was 62%. DUS increased the predictivity in all patients. Keywords: acute scrotum; children; testicular torsion; ultrasonography; pediatric surgery INTRODUCTION Acute scrotum in the pediatric population is defined by the sudden onset of scrotal pain, erythema and swelling. Acute scrotum may become a surgical emer- gency, since it can lead to the loss of the affected tes- ticle, especially when a failure in diagnosing and an erroneous management occur. Despite the presence of recent guidelines(1), the management of acute scrotum is not uniform world-wide. In many institutions an im- mediate surgical exploration is performed in all males presenting with acute scrotum, in order to save the highest number of testicles; however other institutions use a more conservative approach, unless testicular tor- sion is evident(2-4). The main causes of acute scrotum in pediatric age are testicular torsion (TT), torsion of the appendage of the testis (TAT) and epididymo-or- chitis (EO). The annual incidence of TT is 1:4000 in males aged under 18 which accounts for 5-25% of acute scrotum in children. Furthermore, the bimodal distribution of TT, with peaks in the perinatal period and in adolescence, reflects the distinction between the extravaginal and the intravaginal torsion, which are respectively typical of newborns and older children. Anamnestic data and clinical findings are the corner- 1 Department of Pediatric General, Thoracic and Minimally Invasive Surgery, Regina Margherita Children's Hospital, Torino, Italy. 2 Department of Public Health and Pediatric Sciences, University of Torino, Torino, Italy. 3 Division of Pediatric Radiology, Regina Margherita Children’s Hospital, Torino, Italy. *Correspondence: Department of Pediatric General, Thoracic and Minimally Invasive Surgery, Regina Margherita Children's Hospital – Turin, Italy. Tel: +39 0113135276. Fax: + 39 0113135660.E-mail: riccardoguan@gmail.com. Received January 2016 & Accepted June 2016 stones to diagnose acute scrotum. Thus many studies have been performed in order to achieve clinical scores that could help surgeons in diagnosis and treatment(5-7). Typically, TT presents with a sudden onset of severe pain followed by scrotal and inguinal swelling. Often, a high riding testis, which can also lie transversely, is found. The absence of cremasteric reflex is one of the most con- sistent findings that supports the diagnosis of TT(2,8,9). An increased incidence of TT has been also associated with seasonal variations and lower temperatures(10,11). In addition to clinical examination, Doppler ultra- sonography (DUS) has been progressively used in TT diagnostic management. It can provide very good and useful information about anatomy and perfusion, with a reported sensitivity of 64-91% and specificity of 97-100%(1,12). DUS has become the standard imaging method in acute scrotum management, because of its wide availability, easy performance, non-invasiveness and low costs. However, it is limited by high operator dependency and non-negligible number of false-neg- ative results that may lead to loss of the testicle. The aim of the present study was to identify reliable clinical findings associated with TT and to evaluate the role of testicular DUS in diagnosing pediatric tes- ticular torsion, under the aspect of diagnostic accuracy Pediatric Urology 2779 Vol 13 No 04 July-August 2016 2780 and improvement, with an aim to achieve specificity increase and decrease in negative exploration rates. PATIENTS AND METHODS Study Population and inclusion criteria We performed a retrospective analysis of all consec- utive patients with acute testicular symptoms pre- senting to the emergency department (ED) of our hospital from January 2010 to December 2013. Pa- tients were selected from the hospital database us- ing ICD 9th edition codes: EO (6040, 60490, 60499), TAT (60823), TT (60820), testicular pain (6089). 1091 patients were included in the analysis, for a to- tal of 1219 ED admissions. Patients with a diagnosis of intrauterine testicular torsion, varicocele, hydrocele, and cryptorchidism were excluded from the study. Procedures and Evaluations Data were collected using patient charts and operating room records. Regarding medical history, data included the age of the patient, time of arrival at the ED, first evaluation (ED, another hospital, general practition- er), duration of symptoms, and history of possible trauma. All patients were physically examined by a pediatric surgeon, who evaluated erythema, swelling, spermatic cord pain and consistency, cremasteric re- flex, and fever (T > 38°C). A reduced or absent cre- masteric reflex was considered an abnormal finding. DUS was performed by a pediatric radiologist using Philips iU22 Ultrasound Machine (Philips Corpora- tion, Andover, MA, USA) adjusted and optimized for testicular structures; ultrasonography was performed with a 12.5 Hz linear transducer. The scanning meth- od consisted of the visualization of the scrotum and its contents with longitudinal and transverse axes. Both testicles were scanned in order to provide a compari- son of anatomy and blood flow. The entire testicle was evaluated from one extreme to another, then the scan was repeated after rotating the probe 90° to obtain a transverse image of the testicle. Once the gray-scale imaging was complete, the power Doppler examina- tion was performed. Power Doppler and pulsed Dop- pler parameters were optimized to adequately evaluate the blood flow in testicles and their adjacent structures. The color gain was calibrated carefully to avoid any artifactual appearance of flow. A diagnosis of TT was made in cases of uncertainty or absent central perfusion. All patients presenting with scrotal symptoms and a clinical or a DUS suspicion for TT underwent surgi- cal exploration. In TT cases, the testis was detorqued and placed in warm sponges for 15-20 minutes; if no sign of perfusion was observed, an orchiectomy was performed, then the surgeons proceeded with a contralateral orchiopexy using the three-stitch trian- gular technique with a reabsorbable suture. In case of a viable testis, it was fixed in with 3-4 stitches. Statistical Analysis Statistical analysis was performed using the SPSS soft- ware version 15.0 (SPSS, Inc, Chicago, IL). The differ- ence between groups was calculated using the Fisher's test and a binary logistic multivariate backward step- wise analysis with variable inclusion threshold P < .1, expressed as odds ratios (OR) and 95% confidence in- tervals (CI). Finally we calculated the R2 Nagelkerke index in order to evaluate the predictivity of the model. RESULTS The data from 1091 patients were analyzed, but since 128 patients were visited twice, 1219 ED visits were included in the final analysis. The mean age of the study population was 9.1 years (range 6 months – 16 years). The reported symptoms and clinical findings of the study population are summarized in Table 1. The mean time of presentation to the ED from the onset of symptoms was 21.49 hours (range 45 min- utes – 72 hours). For patients who underwent sur- gery, the mean duration was 19.89 hours (range Diagnostic management in pediatric acute scrotum-Lemini et al. Table 1. Reported symptoms and clinical findings Characteristic Value Percentage Mean age (year) 9.11 Side Right 592 48.56% Left 558 45.77% Bilateral 69 5.67% Trauma Yes 74 6.07% No 1145 93.93% Pain Yes 1144 93.92% No 75 6.08% Pain location Diffuse 475 41.52% Upper pole 31 2.71% Lower pole 638 55.77% Swelling Yes 528 43.31% No 691 56.69% Erythema Yes 430 35.27% No 789 64.73% Painful 50 4.10% Spermatic cord Thickened 4 0.33% Thickened and painful 20 1.64% Normal 1145 93.93% Cremasteric reflex Normal 1108 90.89% Reduced 62 5.08% Absent 49 4.03% Fever (T > 38°C) Yes 13 1.06% No 1206 98.94% Abbreviations: T = temperature 45 minutes – 48 hours). However, once there was a suspicion of a TT, the mean time between the physical examination and the operation was 1.07 hours (range 30 minutes - 2 hours and 15 minutes). Diagnostic results were: TT in 41 patients (all by surgical exploration), TAT in 323 (288 by clinical and DUS ex- amination, 35 by surgical exploration), EO in 506 (479 by clinical and DUS examination, 27 by surgical explo- ration), and other pathological diagnosis in 349 patients (including trauma, not-otherwise-specified testicle pain, idiopathic scrotal edema, inguinoscrotal hernia; 345 by clinical and DUS examination, 4 by surgical exploration). DUS was performed in 498 patients (40.85%), whose re- ports were compatible with the following diagnoses: EO 279 (56.02%), TT 27 (5.42%), TAT 41 (8.24%), normal 144 (28.92%), uncertain 5 (1%), hematoma 1 (0.20%), inguinoscrotal hernia 1 (0.20%). All 27 patients with a DUS report compatible with TT underwent surgical ex- ploration, however in 5 patients the diagnosis changed to a non-TT category (3 EO, 1 TAT, 1 hematoma). Of the 471 cases with DUS reports compatible with non- TT, 7 underwent surgical exploration due to a strong TT suspicion and the TT diagnosis was confirmed. Surgical exploration was performed in 107 cases (8.77%) and the diagnosis of TT was confirmed in 41 patients (3.36%). Orchiectomy rate was 14.02%. None of patients discharged from the ED returned with TT. The negative exploration rate was 62%. The overall sensitivity, specificity, positive predictive value (PPV), and diagnostic accuracy of both clinical and DUS findings have been presented in Table 3. Then we analyzed the association between the pres- ence of scrotal pain, erythema, swelling, spermatic cord pain and/or thickness, and abnormal cremaster- ic reflex. (Table 3) A prior selection was made using Fisher's test, then a multivariate analysis model was produced. The clinical findings associated with TT were spermatic cord pain and/or thickness (95% CI: 11.9-111.1, OR = 37, P < .001), and abnormal cre- masteric reflex (95% CI: 13.5-166.6, OR = 47.6, P < .001). The presence of erythema resulted associated with non-TT (95% CI: 0.07-0.7, OR = 0.22, P = .0445) while the presence of swelling was not associated with TT (95% CI: 0.7-8.4, OR = 2.3, P < .001). The pres- ence of pain was excluded from the analysis model be- cause of its low statistical specific relevance (P = .9). Furthermore, the presence of spermatic cord pain and/ or thickness (95% CI: 17.2-76.9, OR = 37, P < .001), abnormal cremasteric reflex (95% CI: 5.3-18.2, OR = 9.9, P < .001), and swelling (95% CI: 2.4-9.3, OR = 4.6, P < .001) were associated with surgical indication. In the second analysis we added the “DUS findings compatible with TT” to the model, in order to verify contingent differences in predictivity. The DUS vari- able was strongly associated with TT (95% CI: 11.9- 250, OR = 55.55, P < .001). Finally we calculated the R2 Nagelkerke index to evaluate the predictivity of the model. First only considering clinical findings, the R2 index was 0.649, then adding DUS variable, it increased to 0.784. This means that the DUS significantly increas- es the predictivity of acute scrotum diagnostic process. Discussion Children presenting at the ED with acute testicular symptoms require an immediate examination. The most frequent causes of acute scrotum are TT, EO, and TAT. In this study, the incidence of TT cases was lower (3.36%) than in previous reports(6,13-16). This can be explained by the fact that the current study includ- ed all patients with acute testicular symptoms, even those without abnormalities on physical examination. The optimal diagnostic management of acute scro- tum should identify patients who require an imme- diate surgical exploration, in order to recognize all TTs and save the highest number of twisted testicles. In our study we focused on spermatic cord pain and/ or thickness, and abnormal cremasteric reflex. The presence of pain and/or thickness of the spermatic cord showed to be highly predictive of TT, but we could not find reports in literature that consider this finding in their analysis. The abnormality of the cremasteric reflex was also strongly associated with an increased likelihood of TT. This result is supported by other studies(13,16,17,19-21), even if some of them(13,22) found an abnormal cremaster- ic reflex in all patients with TT. In the present study, 4 of the 41 patients with TT had a preserved cremaster- ic reflex. This difference can be explained by the large study population (1091 patients), in fact the previous series involved a smaller number of patients. Moreo- ver, a similar result can be found in recent reports(17,18). The presence of at least one of these findings should induce the suspicion of TT in children presenting at ED with acute testicular symptoms, in fact all patients with TT showed at least one of them. Furthermore, no TT was found in patients without either finding. Our negative exploration rate was of 62%, lower than in the known literature(17,18). The implementation of this approach would have further decreased it by 28%. The presence of erythema makes TT unlike- ly, as it is reported in several studies(17,18,23). The presence of swelling resulted slightly, but not significantly, associated with TT(19,23). The presence of pain was not statistically sig- nificant, because it is a common symptom of all causes of acute scrotum. However many stud- ies demonstrated the association of a pain du- Table 2: Sensitivity, specificity, positive predictive value and diagnostic accuracy of clinical data and Doppler ultrasonography in diagnosis of testicular torsion. Clinical DUS Sensitivity 94.11% 75.86% Specificity 95.37% 98.93% Positive predictive value 47.05% 81.48% Diagnostic Accuracy 95.32% 97.59% Abbreviations: DUS, Doppler ultrasonography Table 3: Analyzed clinical and DUS findings OR 95% CI Erythema 0.22 0.07-0.7 Swelling 2.3 0.7-8.4 Spermatic cord pain/thickness 37 11.9-111.1 Abnormal cremasteric reflex 47.6 13.5-166.6 Abbreviations: DUS = Doppler ultrasonography, OR = odd ratio, CI = confidence interval Diagnostic management in pediatric acute scrotum-Lemini et al. Pediatric Urology 2781 Vol 13 No 04 July-August 2016 2782 ration lower than 24 hours, with TT(2,17,18,20,22,24). In the last few years, DUS became an important part of the diagnostic process of acute scrotum, since it can evaluate the reduction/absence of central perfusion, it is a fast non-invasive procedure, and its costs are limited. However DUS is highly operator dependent and in many institutions is not available during night hours. In the current study, we recorded a 75.86% sen- sitivity and a 98.93 specificity, similarly to what is written in international literature(25,26). From the statis- tical analysis, we found a strong association between DUS findings and TT. Nevertheless, DUS cannot substitute the clinical examination, in fact 7 patients with TT had a preserved central perfusion. This cir- cumstance occurred in other reports(2,17) as well and confirms our results. A limitation of the current study was that DUS has been performed on 40.85% of pa- tients, even if groups were statistically comparable. These results allow us to claim that clinical and DUS findings must be enrolled together in order to increase the specificity, lowering the negative exploration rate and improving the model predictivity (R2 Nagelkerke index: 0.649 → 0.784). Thus we suggest to perform a DUS before surgical exploration, unless either the sus- picion of TT or the risk of a testicle loss are high. This approach is also consistent with other reports(17,27,28), even if there are studies that suggest an immediate sur- gical exploration in all patients with acute scrotum(4). Conclusion Acute scrotal symptoms are a common causes of pres- entation to a pediatric ED. TT incidence is low (3.36%). Clinical findings show high sensitivity, but low spec- ificity; thus there are no missed TT, but a high num- ber of negative exploration. The spermatic cord pain and/or thickness and an abnormal cremasteric reflex are two essential findings to search for when diagnos- ing TT. The DUS cannot replace the physical exami- nation, but it is an actual aid for all cases of uncertain clinical suspicion. The ensemble of clinical and DUS findings allow clinicians to improve diagnostic accura- cy and specificity, and lower negative exploration rate. CONFLICT OF INTEREST None declared. REFERENCES 1. Tekgul S, Riedmiller H, Gerharz E, et al. Guidelines on Paediatric Urology. ESPU/ EAU 2011. 2. Yang C, Song B, Liu X, Wei G, Tan J, He D. Acute scrotum in children. An 18-year retrospective study. Pediatr Emer Care 2011; 27: 270-274. 3. Cass AS, Cass BP, Veeraraghavan K. Immediate surgical esploration of the unilateral acute scrotum in young male subjects. J Urol 1980; 124: 829-832. 4. Murphy FL, Fletcher L, Please P. Early scrotal exploration in all cases is the investigation and intervention of choice in the acute pediatric scrotum. Pediatr Surg Int 2006; 22: 413-416. 5. Gunther P, Rubben I. The acute scrotum in childhood and adolescence. Dtsch Artebl Int 2012; 109 (25): 449-58. DOI: 10.3238/ arztebl.2012.0449. 6. Mcandrew HF, Pemperton R, Kikiros CS, Gollow I. The incidence and investigation of acute scrotal problems in children. Pediatr Surg Int 2002;18: 435-437. 7. Gunter P, Schenk JP. Testicular torsion: diagnosis, differential diagnosis and treatment in children. Radiologe 2006; 46: 590-595 8. Sharp VJ, Kieran K, Arlen AM. Testicular Torsion: diagnosis, evaluation and management. American Family Physician 2013; 88 (12): 835-840 9. Caesar RE, Kaplan GW. The incidence of the cremasteric reflex in normal boys. J Urol 1994; 152: 779-780. 10. Srinivasan AK, Freyle J, Gitlin JS, Palmer LS. Climatic conditions and the risk of testicular torsion in adolescent males. J Urol 2007; 178: 2585-2588. 11. Korkes F, Cabral PR, Alves CD, Savioli ML, Pompeo AC. Testicular torsion and weather conditions: analysis of 21.289 cases in Brazil. Int Braz J Urol 2012; 38: 222-228. 12. Gearhart JG. Pediatric Urology. Sauders 2010; 42: 555-567. 13. Kadish HA, Bolte RG. A retrospective review of pediatric patients with epididymitis, testicular torsion, and torsion of testicular appendages. Pediatrics 1998; 102, 73-76. 14. Erikci VS, Hosgor M, Aksoy N, et al. Treatment of acute scrotum in children: 5-years experience. Turkish Journal of Trauma and Emergency Surgery 2013; 19: 333-336. 15. Liang T, Metcalfe P, Sevcik W, Noga M. Retrospective review of diagnosis and treatment in children presenting to the pediatric department with acute scrotum. AJR Am J Rentgenol 2013; 200: 444-449. 16. Van Glabeke E, Kairouni A, Larroquet M, Audry G, Gruner M. Acute scrotal pain in children: results of 543 surgical explorations. Pediatr Surg Int 1999; 15: 353-357. 17. Boettcher M, Krebs T, Bergholz R, Wenke K, Aronson D, Reinshagen K. Clinical and sonographic features predict testicular torsion in children: a prospective study. BJU Int 2013; 112: 1201-1206. 18. Boettcher M, Berghol R, Krebs TF, Wenke K, Aronson DC. Clinical Predictors of Testicular Torsion in Children. Pediatric Urology 2012; 79: 670-674. 19. Beni-Israel T, Goldman M, Bar Chaim S, Kozer E. Clinical predictors for testicular torsion as seen in the pediatric ED. Am J Emerg Med 2010; 28: 786-789. 20. Ciftci AO, Senocak ME, Tanyel FC, Buyukpamukcu N. Clinical predictors for differential diagnosis of acute scrotum. Eur J Pedriatr Surg 2004; 14: 333-338. Diagnostic management in pediatric acute scrotum-Lemini et al. 21. Kass EJ, Lundak B. The acute scrotum. Pedriatr Clin North Am 1997; 44: 1251-1266. 22. Rabinowitz R. The importance of the cremasteric reflex in acute scrotal swelling in children. J Urol 1984; 132: 89-90. 23. Karmazyn B, Steinberg R, Kornreich L, et al. Clinical and sonographic criteria of acute scrotal problems in children: a retrospective study of 172 boys. Pediatr Radiol 2005; 35: 302-310. 24. Jefferson RH, Perez LM, Joseph DB. Critical analysis of the clinical presentation of acute scrotum: a 9-year experience at a single institution. J Urol 1997; 158: 1198-1200. 25. Kalfa N, Veyrac C, Baud C, Couture A, Averous M, Galifer RB. Ultrasonography of the spermatic cord in children with testicular torsion: impact on the surgical strategy. J Urol 2004; 172: 1692-1695. 26. Baker LA, Singman D, Mathews RI, Benson J, Docimo SG. An analysis of clinical outcomes using color doppler testicular ultrasound for testicular torsion. Pediatrics 2000; 105: 604- 607. 27. Jequier S, Patriquin H, Filiatrault D, et al. Duplex doppler sonographic examinations of the tests in prepubertal boys. J Ultrasound Med 1993; 12: 317-322. 28. Lam WW, Yap TL, Jacobsen AS, Teo HJ. Colour doppler ultrasonography replacing surgical exploration for acute scrotum: myth or reality? Pediatr Radiol 2005; 35: 597-600. Diagnostic management in pediatric acute scrotum-Lemini et al. Pediatric Urology 2783