Sonographic Appearance of the Epididymis in Pediatric Testicular Torsion

Transcription

1 Nussbaum Blask and Rushton Sonography of Testicular Torsion in Pediatric Patients Pediatric Imaging Original Research A C D E M N E U T R Y L I A M C A I G O F I N G Anna R. Nussbaum Blask 1 H. Gil Rushton 2 Nussbaum Blask AR, Rushton HG Keywords: color Doppler sonography, emergency radiology, epididymis, genitourinary tract imaging, pediatric radiology, testicular torsion DOI: /AJR Received April 14, 2005; accepted after revision November 10, Department of Radiology, Children s National Medical Center, 111 Michigan Ave., NW, Washington, DC Address correspondence to A. R. Nussbaum Blask (ablask@cnmc.org). 2 Department of Urology, Children s National Medical Center, Washington, DC AJR 2006; 187: X/06/ American Roentgen Ray Society Sonographic Appearance of the Epididymis in Pediatric Testicular Torsion OBJECTIVE. The purpose of this study was to determine the frequency of an enlarged epididymis in pediatric patients with testicular torsion and to determine whether an altered epididymis might be a helpful ancillary sonographic sign of testicular torsion. MATERIALS AND METHODS. The sonograms of 50 pediatric patients (age range, neonate 17 years) with testicular torsion were retrospectively reviewed for the size, appearance, and blood flow of the epididymis. Medical records were reviewed for surgical and pathologic findings and to determine whether testicular salvage had been possible. RESULTS. The epididymis was enlarged (maximum dimension, 5.5 cm) in 47 of the 50 patients with acute or late phase torsion and after manual or spontaneous detorsion. The average difference in volume between the ipsilateral epididymis and the contralateral epididymis was 30 cm 3, highly significant (p < ). The shape of the epididymis was altered in 92% of the cases (globular, bilobular, or multilobular). Seventy-three percent showed increased echogenicity and 27% appeared isoechoic. Of those with active torsion, 93% of the epididymides were avascular; 2%, hypovascular; and 5%, hypervascular. After detorsion, 100% of the epididymides (10/10) had blood flow. In patients with testicular loss due to infarction, pathology showed engorgement and enlargement of the epididymis with hemorrhagic infarction. In one patient with a hypervascular epididymis, surgery showed inflammation and erythema of the epididymis. Testicular loss occurred in each type of epididymal flow pattern. CONCLUSION. A markedly enlarged, echogenic, and avascular or hypovascular epididymis is an ancillary sonographic sign in pediatric patients with testicular torsion. A hypervascular enlarged epididymis infrequently occurs (5% of cases) and should not be mistaken for epididymitis. In addition, the return of epididymal blood flow is an ancillary sign of successful testicular detorsion. he acutely painful scrotum is a T common clinical occurrence in pediatric patients and is treated as a clinical emergency because of the possibility of testicular torsion. Testicular torsion, which requires immediate surgical intervention or manual detorsion, must be distinguished from the nonsurgical conditions of torsion of the appendix testis, epididymitis, and epididymoorchitis. If clinical suspicion of testicular torsion is high, surgery is performed without additional imaging studies. However, in most cases in the pediatric population, the absolute differentiation of testicular torsion from the nonsurgical conditions based on only physical signs and symptoms is difficult. In addition, up to 70% of patients with an acute scrotum have nonsurgical conditions [1, 2]. The age of the patient is not helpful in the differential diagnosis because testicular torsion can occur in patients of any age, including neonates [3, 4]. Radiologic imaging studies are performed frequently in the acutely symptomatic patient to avoid unnecessary surgical exploration. Over the past several years, testicular color Doppler sonography has become the initial imaging study of choice because of its reported high sensitivity ( %) and specificity (77 100%), coupled with its lack of radiation and rapidity of performance, replacing testicular scintigraphy in most cases [3 6]. In our institution, scintigraphy recently has served as an ancillary second study reserved for cases that were indeterminate for torsion on sonography or showed decreased but still present flow compared with the asymptomatic side. The role of power Doppler sonography in the examination of the testes has been investigated. Whether it provides an additional advantage over color Doppler sonography or might, in fact, increase the false-negative rate because of its increased sensitivity to AJR:187, December

2 Nussbaum Blask and Rushton decreased but still present arterial blood flow is not yet known [7 11]. In several reports, researchers have described false-negative color Doppler sonography findings in patients with testicular torsion despite its high reported sensitivities [12 15]. In addition, seeing flow in the normal prepubertal testis may be difficult or impossible, making the diagnosis of torsion on the contralateral symptomatic side more difficult [3, 4, 10, 11, 16]. In fact, some authors have suggested that radionuclide scintigraphy remain the imaging technique of choice in prepubertal children [10]. In our previous study of 46 pediatric patients with an acute scrotum, we observed that a markedly enlarged avascular epididymis was present in children with torsion of the testis; it was a source of confusion in several cases because of its masslike appearance [3]. In this study, we retrospectively reviewed a series of pediatric patients with the known diagnosis of testicular torsion, including 13 patients with torsion from our previous study, to determine the frequency of this finding on testicular sonography. Also, we wanted to determine whether an abnormal appearance of the epididymis might be a useful ancillary gray-scale and color Doppler sign of testicular torsion. In addition, we wanted to see whether there were any differences in the appearance of the epididymis during acute phase torsion, during late phase torsion, or after spontaneous or manual detorsion, and we wanted to correlate the appearance of the epididymis with pathologic and surgical findings. Our final aim was to determine whether the sonographic appearance of the epididymis could be of any assistance in predicting testicular salvage. Materials and Methods All patients with a hospital discharge diagnosis of testicular torsion between January 1992 and 2002 were identified by performing a computerized search of medical records. Patients who had undergone color Doppler sonography of the testis and epididymis as part of their diagnostic evaluation formed the study group, which was composed of 50 patients who ranged in age from neonate to 17 years. Institutional review board consent for the study was obtained. Each sonogram was retrospectively reviewed by one of the authors for all of the parameters being evaluated. The reviewer was not blinded to clinical information. The size of the epididymis on both the symptomatic and the contralateral asymptomatic sides was determined. Sonographic studies had been performed using a standardized method to obtain longitudinal images parallel to the long axis of the testis and transverse images perpendicular to the long axis of the testis. The length and depth of the entire visualized portion of the epididymis were measured with calipers on a single longitudinal image in the plane of maximal enlargement. The transverse measurement was obtained from a single transverse image in the plane of maximal enlargement. On the symptomatic side, if only the head was enlarged and the head was the only portion of the epididymis visualized, it was used for the measurements of epididymal size. A range of sizes was obtained for the symptomatic side and for the asymptomatic side. The epididymal size in the asymptomatic hemiscrotum served as the control sample in 41 of the 50 patients because there were no cases of bilateral synchronous testicular torsion. In the remaining nine patients, the epididymis in the asymptomatic hemiscrotum could not be measured retrospectively because images were insufficient or the epididymis was not seen or because of prior removal or atrophy of the contralateral testis and epididymis (two patients). In eight of these nine patients, the epididymis in the symptomatic hemiscrotum was matched to that of an age-matched control subject. In one of the nine patients, no quantitative analysis was performed because of the lack of hard-copy images. Evaluation was qualitative only and based on the written report. The shape of each epididymis in the symptomatic hemiscrotum was categorized as triangular, round, or ovoid, and an estimated volume was obtained using the formula for a pyramid, sphere, or prolate ellipsoid, respectively, in 49 of the 50 patients. The estimated volume of the normal epididymis was calculated using the formula for a pyramid because review of the images in our series indicated the shape of the normal epididymis was best approximated by the triangular shape; none was globular, bilobular, or multilobular. Other authors have previously described the shape of the normal epididymis as pyramidal [17, 18]. Leung et al. [18] in their sonographic analysis of 40 asymptomatic men suggested that the epididymis has a triangle, crescent, or teardrop shape. The volume of the epididymis on the symptomatic side was compared with that of the contralateral normal epididymis using a Student s t test of the difference between the means. Before statistical tests were applied, the data were checked for normality and homogeneity of variance. Normalizing variance-stabilizing data transformations were applied to satisfy the criteria for parametric analysis before testing. Thereafter, a Student s t test was applied to test whether the difference in epididymal volume between the symptomatic side and the asymptomatic side was a value other than zero. The portion of the epididymis that was enlarged was documented as the head, head and body, body and tail, head and tail, or entire epididymis. The gray-scale appearance of the epididymis was evaluated. The shape was further analyzed to determine if it maintained a triangular configuration or if it was distorted into a globular, bilobular, or multilobular shape. The echotexture of the epididymis on the side of testicular torsion was evaluated and characterized as isoechoic or hyperechoic and as homogeneous or heterogeneous in texture. The number of epididymides containing multiple prominent cystic spaces was tabulated. The vascularity of the ipsilateral epididymis and the symptomatic testis was evaluated in all 50 subjects with color flow Doppler imaging using low-flow settings (HDL 3000 or 5000, ATL). Each sonographic examination was performed by one of three experienced pediatric sonographers. The epididymis was characterized as vascular if color flow was seen, avascular if no color flow was detected, and hypovascular if only minimal flow was seen. If the epididymal flow was greater than the flow seen in the asymptomatic side, it was considered hypervascular. Pulsed Doppler tracings of the epididymis were not obtained except in two patients. The Doppler findings in the symptomatic testis were correlated with the findings in the ipsilateral epididymis. The number of cases with manual detorsion or spontaneous detorsion was tabulated, and color Doppler examinations were assessed both before and after detorsion, if available, or only after detorsion if sonography had not been performed before detorsion. The patients medical records were reviewed for surgical and pathologic findings and to determine whether testicular salvage had been possible. The diagnosis of testicular torsion was established by surgery in 48 of the 50 patients; by late testicular atrophy in one patient; and by nuclear scintigraphy performed 24 hours after the initial sonography in one patient, whose parents declined giving consent for surgical intervention. Scans were divided into those showing acute phase versus late phase torsion on the basis of the duration of each patient s symptoms, as described by Chen et al. [19]: Acute phase torsion was defined as < 24 hours of pain and late phase torsion as 24 hours of pain. Results Of the 50 cases, 34% were acute phase torsion and 66% were late phase torsion. Of the 17 patients with acute phase torsion, 10 (59%) underwent manual detorsion and spontaneous detorsion occurred in one (6%) AJR:187, December 2006

3 Sonography of Testicular Torsion in Pediatric Patients The average difference in volume between the epididymis ipsilateral to testicular torsion and the contralateral epididymis was 30 cm 3 with a 95% CI of cm 3. This difference is highly statistically significant (p < ). Enlargement was evident in 47 of the 50 patients with acute phase or late phase torsion and after manual and spontaneous detorsion. The size of the epididymis was normal in the remaining three patients, one with acute phase torsion and two with late phase torsion. Of the latter two cases, one was in a neonate. The epididymal size on the symptomatic side ranged from to 0.9 L 0.5 D 1 T cm 5.5 L 4.5 D 4.5 T cm. The measurements on the asymptomatic side ranged from to 0.3 L 0.3 D 0. 3 T cm 0.9 L 1.7 D 1.1 T cm, where L is length, D is depth, and T is transverse. In most cases, epididymal enlargement was visually obvious, even without obtaining A B C Fig. 1 Longitudinal images of epididymis in three patients with testicular torsion. A, Sonogram of 16-year-old boy shows diffuse enlargement of head, body, and tail of epididymis (E) with bilobular shape. Note diffusely increased echogenicity. B, Sonogram of 15-year-old boy shows heterogeneously increased echogenicity of epididymis (E) with globular shape. C, Sonogram of 15-year-old boy shows head of epididymis (E) is enlarged, has isoechoic texture, and is slightly triangular. T = testis. measurements. By contrast, epididymal enlargement was less pronounced in infants and in three boys with acute phase torsion. There were five patterns of enlargement of the epididymis: head only (39.1%); head and body (13%); body and tail (4.3%); head and tail (2.2%); and diffuse (41.3%) involving the head, body, and tail with increased tissue cephalad, lateral, and caudad, respectively, to the testis (Figs. 1A 1C). The shape of the epididymis was altered in acute phase or late phase torsion and after manual or spontaneous detorsion. The epididymis maintained a triangular configuration in the remaining 8% of cases. In the cases in which the normal triangular configuration was lost, the shape was globular (87%), bilobular (9%), or multilobular (4%) (Figs. 1A 1C). The echotexture of the epididymis was altered in patients with acute phase or late phase torsion and in those examined after manual or spontaneous detorsion. The most common pattern was markedly increased echogenicity (36/49 [73%]), either heterogeneous with increased foci (n = 29 cases) or homogeneous with a diffuse increase in echogenicity (n =7) (Figs. 1A 1C). Less commonly, the epididymis was isoechoic (13/49 [27%]). Multiple cystic spaces were seen in 13 cases. Six of these cases showed no flow in the epididymis or cystic spaces because they were evaluated after testicular detorsion only. Two of the cases with cystic spaces were in the patients with testicular torsion and hyperemic epididymides, and most of the areas of sonolucency were confirmed to be vascular channels on the available color Doppler images. Because of the retrospective nature of our study, we could not determine if the other spaces without color fill-in were vascular channels as well or if they were areas of necrosis, hemorrhage, or thrombus within vessels. In the five other cases examined after manual or spontaneous detorsion, most, but again not all, of the cystic and tubular spaces were confirmed to be vascular. In two of the latter patients examined both before and after detorsion, the cystic spaces were seen to be primarily vascular channels that filled in with color on Doppler imaging after detorsion (Figs. 2A 2C). Before undergoing surgery because of testicular torsion, 39 patients received no treatment, 10 patients underwent manual detorsion, and spontaneous detorsion occurred in one patient. Of the 10 patients with manual detorsion, seven were sonographically examined only after detorsion, one only before detorsion, and two were examined before and after detorsion. The patient with spontaneous detorsion was examined before and after detorsion. Epididymal blood flow was evaluated. Of the 43 patients with active torsion at the time of sonography, the epididymis was avascular in 40 patients, hypovascular in one patient, and hypervascular in two patients (Figs. 3A 3C). Of the 10 patients who were ex- AJR:187, December

4 Nussbaum Blask and Rushton Fig year-old boy with acute phase torsion that resolved after spontaneous detorsion. A, Epididymis contains multiple cystic spaces and no flow on sonogram obtained during acute phase torsion. B, Testis is avascular on sonogram obtained during acute phase torsion. C, Sonogram obtained after spontaneous detorsion shows that epididymis has regained flow. Cystic spaces in epididymis are vascular channels. D, Sonogram obtained after spontaneous detorsion shows that testis also has regained flow. A C amined after detorsion, either manual or spontaneous, all epididymides were vascular and six of the 10 were hypervascular. Of the three cases examined before and after detorsion, the transition from no epididymal flow to epididymal flow was observed, similar to the observation of flow in the corresponding testis (Fig. 2). The testis was also avascular in 34 of the 40 patients, whereas it was hypovascular in the remaining six patients with an associated avascular epididymis. The testis also showed decreased flow in the one patient with a hypovascular epididymis and acute phase torsion. In the two patients with unexpected epididymal hyperemia, the testis was avascular in one and hypovascular in the second. In the former, scintigraphy also showed increased flow to the epididymis and a photon-deficient testis. Surgery and pathologic examination of 30 patients with testicular loss and avascular epididymis, as shown by the Doppler flow pattern, showed engorgement and enlargement of the epididymis with hemorrhagic infarction (Fig. 4). Two surgical cases after detorsion showed congestion of the distal spermatic cord. No other surgical or pathologic report included an evaluation of the size of the spermatic cord. Information about the degree of torsion of the spermatic cord was available in 11 cases. Seven patients with late phase torsion had from 540 to 720 or more of cord torsion with both an avascular epididymis and avascular testis at sonography; testicular salvage was not possible. In one patient with acute phase torsion with a twist of who had both avascular epididymis and avascular testis, tes- B D 1630 AJR:187, December 2006

5 Sonography of Testicular Torsion in Pediatric Patients ticular salvage was successful. In three patients, a 360 twist of the spermatic cord was seen. One of the three patients had late phase torsion and an avascular epididymis and avascular testis; testicular salvage was not possible. The second case was also in a patient with late phase torsion with an avascular epididymis, minimal flow in the testis, and no testicular salvage. The third was a patient with acute phase torsion and a hyperemic epididymis, avascular testis, and testicular salvage. Surgery revealed an inflamed erythematous epididymis in one of the two patients with a hyperemic epididymis. The testis, which appeared avascular at Doppler imaging, was ischemic with 360 of torsion but was salvaged; the patient had a 12-hour history of pain A C (Fig. 3C). In the second case of hyperemic epididymis, surgery revealed hemorrhagic infarction of the epididymis, and testis salvage was not possible; that patient had a 20-hour history of pain and minimal flow in the testis at Doppler imaging (Figs. 3A and 3B). The success rates of immediate testicular salvage were assessed. Of the 17 patients with acute phase torsion, two experienced testicular loss: One had a delay in treatment and the other had testicular infarction with a history of 20 hours of pain. Of the remaining 15 patients with acute torsion, the testis was salvaged. In 14 of 15, pain was present for 7 hours or less, and in the remaining patient, pain was present for 12 hours. Three of the 33 testes in late phase torsion were not removed because of Fig. 3 Acute phase torsion with hyperemic epididymis. A and B,15-year-old boy who underwent imaging after 20 hours of pain. Sonograms show hyperemic epididymis (E) (A) and decreased flow in testis (T) (B). Testis was not salvaged. C, 5-year-old boy who underwent imaging after 12 hours of pain. Sonogram shows hyperemic epididymis (E) with no flow in testis (T). Testis was salvaged. possible viability. These three testes subsequently atrophied, and testicular salvage was not possible in any of the remaining cases of late phase torsion. Testicular loss occurred in each type of epididymal flow pattern. Discussion Color Doppler sonography is the imaging technique that is most frequently used to differentiate between surgical (testicular torsion) versus nonsurgical (epididymitis, torsion of the appendix testis, or epididymoorchitis) conditions in pediatric patients with an acutely painful scrotum. In some cases, differentiation between surgical versus nonsurgical disease may be difficult because of low flow in the prepubertal testis and because arterial flow, albeit di- B AJR:187, December

6 Nussbaum Blask and Rushton Fig. 4 Pathologic specimen of infarcted testis (T) and epididymis (E) from 17-year-old boy with late phase torsion shows diffusely swollen epididymis. C = spermatic cord. minished in comparison with the asymptomatic side, is still detected on sonograms. In cases of partial torsion ( ), subtle asymmetry in color blood flow may be overlooked [20]. Bentley et al. [21] suggested that a testicular color Doppler pattern symmetric with the asymptomatic testis initially may be seen in patients with a thick spermatic cord and a long helix that is twisted. However, patients with intermittent testicular torsion or spontaneous detorsion may undergo the study when they are pain free, so imaging may show a normal testis or increased flow in the testis [2]. Our sonographic study suggests that an analysis of the appearance and vascularity of the epididymis will add ancillary information that may assist in the diagnosis of testicular torsion in pediatric patients and that an awareness of the associated finding of epididymal enlargement may avoid confusion with a mass or erroneous diagnosis of epididymitis, even though the diagnosis of testicular torsion is still primarily based on the flow pattern in the testis. Enlargement of the spermatic cord may also be a finding associated with testicular torsion [20 24]. Our study population consisted of pediatric patients with a proven diagnosis of either acute or late phase testicular torsion over a 10-year period. This review showed that enlargement of the epididymis in conjunction with altered echotexture, most frequently increased echogenicity, was a consistent finding in almost all of the cases. This appearance was also present in the patients who were examined after spontaneous or manual detorsion of the testis. Although the retrospective nature of the study necessitated epididymal measurements from 2D images, rather than the more ideal 3D volumetric measurements, the same methodology was applied to evaluate both the symptomatic and the asymptomatic hemiscrotum and the differences between the sides were highly significant. Also, although the estimated measurements were necessary for statistical analysis, epididymal enlargement in conjunction with altered shape and echotexture in the symptomatic hemiscrotum was visually obvious without comparison of epididymal measurements and was marked in most of the cases. In the absence of or before manipulation of the testis or spontaneous detorsion, the epididymis was avascular or hypovascular on color Doppler sonography in 95% of the cases of acute or late phase torsion. Surprisingly, the epididymis was hyperemic in 5% of the cases of acute or late phase torsion (12% of the cases of acute torsion). Our sonographic evaluations of patients examined both before and after detorsion indicated that return of blood flow to the epididymis can be observed in real time, similar to flow to the testis, and can be used as addi- tional evidence of a successful manual detorsion. Epididymal hyperemia was the predominant finding after testicular detorsion. Spontaneous detorsion should be a strong consideration in a patient whose pain has abated suddenly or who reports episodes of intermittent pain because epididymitis can have the same sonographic appearance. Marked enlargement of the epididymis when the studies were initially performed was a confusing finding in several of our cases, and the possibility of an extratesticular mass or hematoma was considered. Enlargement of the epididymis in patients with testicular torsion has not been previously emphasized. However, our observations confirmed findings described previously by a few investigators. In one study, epididymal enlargement with hemorrhage was identified on gray-scale sonography in seven adult patients with testicular torsion [25]. The sonographic findings in seven pediatric patients with testicular torsion suggested a misleading diagnosis of tumor or epididymitis that resulted in delay of surgery and in testicular loss [12 15]. Four of these cases evaluated on color Doppler imaging showed testicular blood flow that was thought to be normal on the symptomatic side. In a recent case of testicular torsion that we reviewed, not part of this series, sonography showed avascular epididymal enlargement that was thought to be a tumor because flow was present within the testis in the symptomatic hemiscrotum. Diagnosis was delayed; a repeat study 9 hours after the first at our institution showed no testicular flow, and testicular loss occurred. Retrospective review of the cases in our study group suggests that the diagnosis would have been altered or made earlier in two cases of testicular torsion, both of which resulted in testicular loss. The two cases of hyperemic epididymides had potential for delay in diagnosis because epididymitis with testicular ischemia was the leading sonographic diagnosis, but testicular torsion was thought to be the most likely diagnosis after testicular scintigraphy. Although two additional patients were prospectively thought to have traumatic testicular contusion with epididymal hemorrhage, the outcome would not have been different because both underwent surgical exploration and had late phase testicular torsion. As these cases point out, an enlarged avascular epididymis has the potential to change or expedite the diagnosis in cases of acute torsion in which testicular flow is still present but is decreased or is still visually normal, because these cases may be interpreted as a 1632 AJR:187, December 2006

7 Sonography of Testicular Torsion in Pediatric Patients TABLE 1: Sonographic Patterns and Differential Diagnosis in Pediatric Testicular Torsion Sonographic Pattern Diagnosis Epididymis Testis Testicular torsion Enlarged No change echotexture (A) Hyperechoic Altered echotexture (L) Altered shape Avascular (A or L) Congested or twisted spermatic Hypovascular (A or L) cord Avascular (A or L) Hypervascular halo (L) Hypovascular (A or L) Resistive index elevated Hypervascular (A or L), uncommon Spontaneous detorsion (pain resolved or intermittent) Partial torsion or thick spermatic cord Epididymitis or torsion of appendix testis Enlarged Hypervascular or normal color flow Enlarged Hypovascular or avascular Enlarged Hypervascular Hypervascular or normal color flow Avascular, hypovascular, or normal color flow Hypervascular or normal color flow Extratesticular true solid mass Separate normal epididymis Normal color flow Note A = acute phase torsion, L = late phase torsion. TABLE 2: Suggested Management Protocol Sonographic Diagnosis Management Acute phase testicular torsion Emergent surgical exploration or may attempt manual detorsion Late phase testicular torsion Nonemergent surgical exploration Spontaneous detorsion Nonemergent surgical exploration Partial testicular torsion Emergent surgical exploration Indeterminate testicular sonogram If < 12 h of symptoms, surgical exploration If > 12 h of symptoms, scintigraphy or surgical exploration Epididymitis or torsion of appendix testis Nonsurgical management Extratesticular true solid mass Nonemergent surgical exploration false-negative. These cases are likely ones in which there are fewer degrees of spermatic cord torsion (partial torsion) and the patient presents earlier in the time course or has a thicker spermatic cord [21]. This sign theoretically has the potential to be of value when no testicular flow can be seen in the asymptomatic testis, although this did not occur in our series. Sanelli et al. [20] reported a case of partial torsion (360 ) of the spermatic cord with visually normal color flow in the testis. They noted a masslike lobular appearance of the spermatic cord with a coiled configuration of the vessels and an elevated resistive index in the affected testis. Other authors have suggested that analysis of Doppler waveform patterns may also be helpful in cases of incomplete or partial torsion [26]. A study by Baud et al. [22] of patients with testicular torsion showed features that paralleled our results. They also described an extratesticular mass present in cases of testicular torsion, but not detorsion. They indicated the mass was due to the engorged, twisted spermatic cord. Kalfa et al. [23] suggested examining the spermatic cord along its complete length to detect a twist. They found that in all cases of torsion, even those with ambiguous testicular flow findings, sonography depicted the twist as a snail shell shaped mass. Arce et al. [24] also described additional sonographic findings related to specific color Doppler assessment of the spermatic cord. In our series, the surgical, pathologic, and sonographic findings suggested that the extratesticular mass was primarily due to an enlarged epididymis. The mass not only was supratesticular in location, but also was frequently paratesticular and infratesticular. Also, a separate epididymal structure was not seen, and epididymal enlargement was present at pathologic examination. However, a portion or portions of the tissue cephalad to the testis could possibly have been an engorged spermatic cord merged with the epididymal head, even though this possibility could not be confirmed by retrospective review of the pathologic or surgical reports or by reanalysis of the sonograms. All of the sonograms were of the scrotum only; the inguinal regions were not examined. Other causes of supratesticular masses need to be considered. Torsion of the testicular appendages may be associated with a supratesticular mass. However, the complete constellation of findings are dissimilar to those associated with testicular torsion; instead, a distinct 4- to 16-mm extratesticular extraepididymal avascular mass is identified in most cases at the upper pole of the testis (the epididymis testis or the appendix testis) in conjunction with an enlarged epididymis and testis with increased flow [27]. Isolated torsion of the epididymis has also been reported, although it seems to be a rare occurrence. A 9-year-old boy with acute scrotal pain, associated with a long, tortuous epididymis and long mesorchium, had this diagnosis established at surgery [28]. Clinically, tender swelling at the inferior pole of the testis was present and the testis was not tender to palpation. A sonogram was not obtained. Two other patients with an acute scrotum and with dissociation of the testis and epididymis have been reported, one in a 6 month old and one in an 18 year old [29]. The epididymis was completely infarcted in the infant, requiring epididymectomy. In most of our patients, the absence of color Doppler flow in the enlarged epididymis was the key feature differentiating testicular torsion from inflammatory epididymitis. However, the epididymis infrequently was hyperemic even when testicular torsion was present. Spectral analysis was not performed in either of our two patients with hyperemic epididymis, so whether the increase in flow was arterial or AJR:187, December

8 Nussbaum Blask and Rushton venous flow is not known. Recently, Arce et al. [24] described increased flow in the spermatic cord distal to the site of rotation and attributed it to venous engorgement. That theory could also explain the hyperemic epididymis cases in our series. Those cases were a diagnostic dilemma, and the possibility of epididymitis with secondary ischemia of the testis from venous outflow compromise was entertained [30]. The fact that the confirmed diagnosis was testicular torsion in both of these patients suggests that surgical exploration is warranted when an avascular or hypovascular testis is seen in conjunction with what appears to be a hyperemic epididymis, distal spermatic cord, or both. If a patient has undergone known manual detorsion of the testis, the presence of flow within an enlarged epididymis and in the testis suggests manipulation was successful, as shown by the resumption of blood flow. In the patient whose acute symptoms have resolved, spontaneous detorsion should be considered when an enlarged hyperemic epididymis and testicular flow are present. An analysis of the vascular supply to the scrotum and its contents explains these sonographic findings. The spermatic cord contains the testicular, deferential (artery of the ductus deferens), and cremasteric arteries. The testicular artery supplies the testis and epididymis, and the artery of the ductus deferens and cremasteric arteries supply the epididymis, vas deferens, and peritesticular tissues [19, 30]. The testicular artery is a branch of the abdominal aorta, and the artery of the ductus deferens is a branch of the superior vesical artery [31]. The epididymis therefore has a dual blood supply via the testicular artery and via its anastomosis with the artery of the ductus deferens [31]. The venous drainage of the scrotal contents also passes through the spermatic cord. Because the vascular supply of the epididymis passes through the spermatic cord, the twisting of the spermatic cord, similar to twisting of the testis, will compromise its blood supply and venous drainage. This explains the presence of epididymal enlargement and engorgement seen on sonographic examination and the ischemic changes identified at pathologic examination. Our patients with hypervascular epididymides and lower cord complex associated with testicular torsion are more difficult to explain unless the artery to the ductus deferens, which travels centrally in the cord, cannot be compressed to the same degree as the testicular artery. Theoretically, vessels that do not pass through the spermatic cord, such as the pudendal vessels, may provide blood flow, although no anastomoses have been described. The scrotal wall derives its blood supply from branches of the pudendal artery, which do not pass through the spermatic cord and therefore maintain its supply during testicular torsion [19, 32]. In late phase torsion, peritesticular inflammatory changes occur and account for the halo of peripheral hypervascularity seen at both sonography and scintigraphy [32]. Alternatively, venous engorgement, due to obstruction of the distal spermatic cord as proposed by Arce et al. [24], is a possibility for the hypervascular epididymis. Testicular viability depends on both the degree and the duration of torsion. After 10 hours of symptoms, most testes cannot be saved [33]. Some testes that are saved at surgery atrophy over time [33]. This occurred in three of our patients. In our series, an enlarged avascular epididymis was present in cases of both acute and late phase torsion; therefore, it was not a helpful sign for predicting which testes could be salvaged. Because we had only two cases of hyperemic enlarged epididymides with testicular torsion, one with and one without testicular salvage, whether this sign could be of any predictive value for testicular salvage is not known. In all of our patients with manual or spontaneous detorsion and reestablishment of blood supply to the testis and epididymis, immediate testicular salvage was achieved, even in those cases in which testicular flow after detorsion was diminished in comparison with the asymptomatic side. The enlarged hyperechoic epididymis is a consistent finding in pediatric testicular torsion that may assist in the diagnosis of torsion, especially in the prepubertal testis and in difficult cases in which arterial supply to the testis is still maintained but is decreased. Sonographic findings and the associated management suggestions are summarized in Tables 1 and 2. The primary diagnosis of testicular torsion is, nevertheless, still based on testicular flow pattern. The epididymis can resemble an extratesticular mass or hematoma. The avascular or hypovascular nature of the epididymis in most cases of testicular torsion (95%) allows these cases to be distinguished from epididymitis, which can have a similar gray-scale appearance. An enlarged hyperemic epididymis that is not due to epididymitis may be seen infrequently in cases of torsion (5%). The epididymis, in addition to the testis, should be sonographically examined for return of color Doppler vascularity after spontaneous or manual detorsion; the presence of epididymal flow in conjunction with flow in the testis is additional evidence of successful detorsion. After detorsion, either manual or spontaneous, the epididymis is frequently hypervascular. The appearance of the epididymis is abnormal in both acute and late phase torsion, and its appearance does help not predict testicular salvage. Acknowledgment We thank Dr. Robert J. McCarter for performing the statistical analysis for this project. References 1. Lewis AG, Bukowski TP, Jarvis PD, Wacksman J, Sheldon CA. Evaluation of the acute scrotum in the emergency department. J Pediatr Surg 1995; 30: Kass EJ, Stone KT, Cacciarelli AA, Mitchell B. Do all children with an acute scrotum require exploration? J Urol 1993; 150: Nussbaum Blask AR, Bulas D, Shalaby-Rana E, Rushton G, Shao C, Majd M. Color Doppler sonography and scintigraphy of the testis: a prospective, comparative analysis in children with acute scrotal pain. Pediatr Emerg Care 2002; 18: Paltiel HJ, Connolly LP, Atala A, Paltiel AD, Zurakowski D, Treves ST. Acute scrotal symptoms in boys with an indeterminate clinical presentation: comparison of color Doppler sonography and scintigraphy. Radiology 1998; 207: Baker LA, Sigman D, Mathews RI, Benson J, Docimo SG. An analysis of clinical outcomes using color Doppler testicular ultrasound for testicular torsion. Pediatrics 2000; 105: Patriquin HB, Yazbeck S, Trinh B, et al. Testicular torsion in infants and children: diagnosis with Doppler sonography. Radiology 1993; 188: Zoller G, Kugler A, Ringert RH. False positive testicular perfusion in testicular torsion in power Doppler ultrasound [in German]. Urologe A 2000; 39: Coley BD, Frush DP, Babcock DS, et al. Acute testicular torsion: comparison of unenhanced and contrast-enhanced power Doppler US, color Doppler US, and radionuclide imaging. Radiology 1996; 199: Bader TR, Kammerhuber F, Herneth AM. Testicular blood flow in boys as assessed at color Doppler and power Doppler sonography. Radiology 1997; 202: Albrecht T, Lotzof K, Hussain HK, Shedden D, Cosgrove DO, de Bruyn R. Power Doppler US of the normal prepubertal testis: does it live up to its promises? Radiology 1997; 203: Luker GD, Siegel MJ. Scrotal US in pediatric patients: comparison of power and standard color Doppler US. Radiology 1996; 198: AJR:187, December 2006

9 Sonography of Testicular Torsion in Pediatric Patients 12. Steinhardt GF, Boyarsky S, Mackey R. Testicular torsion: pitfalls of color Doppler sonography. J Urol 1993; 150: Ingram S, Hollman AS, Azmy A. Testicular torsion: missed diagnosis on colour Doppler sonography. Pediatr Radiol 1993; 23: Pryor L, Watson LR, Day DL, et al. Scrotal ultrasound for evaluation of subacute testicular torsion: sonographic findings and adverse clinical implications. J Urol 1994; 151: Allen TD, Elder JS. Shortcomings of color Doppler sonography in the diagnosis of testicular torsion. J Urol 1995; 154: Ingram S, Hollman AS. Colour Doppler sonography of the normal paediatric testis. Clin Radiol 1994; 49: Dogra VS, Gottlieb RH, Oka M, Rubens DJ. Sonography of the scrotum. Radiology 2003; 227: Leung ML, Gooding GA, Williams RD. High-resolution sonography of scrotal contents in asymptomatic subjects. AJR 1984; 143: Chen DC, Holder LE, Melloul M. Radionuclide scrotal imaging: further experience with 210 patients. J Nucl Med 1983; 24: Sanelli PC, Burke BJ, Lee L. Color and spectral Doppler sonography of partial torsion of the spermatic cord. AJR 1999; 172: Bentley DF, Ricchiuti DJ, Nasrallah PF, McMahon DR. Spermatic cord torsion with preserved testis perfusion: initial anatomical observations. J Urol 2004; 172(6 Pt 1): Baud C, Veyrac C, Couture A, Ferran JL. Spiral twist of the spermatic cord: a reliable sign of testicular torsion. Pediatr Radiol 1998; 28: 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(4 Pt 2): Arce JD, Cortes M, Vargas JC. Sonographic diagnosis of acute spermatic cord torsion. Pediatr Radiol 2002; 32: Vick CW, Bird K, Rosenfield AT, et al. Extratesticular hemorrhage associated with torsion of the spermatic cord: sonographic demonstration. Radiology 1986; 158: Dogra VS, Rubens DJ, Gottlieb RH, et al. Torsion and beyond: new twists in spectral Doppler evaluation of the scrotum. J Ultrasound Med 2004; 23: Baldisserotto M, de Souza JC, Pertence AP, Dora MD. Color Doppler sonography of normal and torsed testicular appendages in children. AJR 2005; 184: Ravichandran S, Blades RA, Watson ME. Torsion of the epididymis: a rare cause of acute scrotum. Int J Urol 2003; 10: Elert A, Hegele A, Olbert P, Heidenreich A, Hofmann R. Isolated epididymal torsion in dissociation of testis-epididymis [in German]. Urologe A 2002; 41: Horstman WG, Middleton WD, Melson GL, Siegel BA. Color Doppler US of the scrotum. Radio- Graphics 1991; 11: Bartleby.com Web site. Gray s anatomy of the human body, 20th ed. Available at: Accessed August 5, Mendel JB, Taylor GA, Treves S, Cheng TH, Retik A, Bauer S. Testicular torsion in children: scintigraphic assessment. Pediatr Radiol 1985; 15: Middleton WD. Scrotal sonography in Ultrasound Q 1991; 9:61 87 AJR:187, December