Real-Time Sonoelastography of Lateral Epicondylitis: Comparison of Findings Between Patients and Healthy Volunteers

Transcription

1 Musculoskeletal Imaging Original Research De Zordo et al. Sonoelastography of Lateral Epicondylitis Musculoskeletal Imaging Original Research Tobias De Zordo 1 Stephanie R. Lill 1 Christian Fink 2 Gudrun M. Feuchtner 1 Werner Jaschke 1 Rosa Bellmann-Weiler 3 Andrea S. Klauser 1 De Zordo T, Lill SR, Fink C, et al. Keywords: elastography, elbow, enthesis, hypervascularity, lateral epicondylitis, power Doppler ultrasound, overuse, real-time sonoelastography, sonography, tendinosis, tennis elbow, ultrasound DOI: /AJR Received October 28, 2008; accepted after revision January 2, Department of Radiology II, Medical University Innsbruck, Anichstrasse 35, 6020 Innsbruck, Austria. Address correspondence to A. S. Klauser (andrea.klauser@i-med.ac.at). 2 Department of Trauma Surgery and Sports Medicine, Medical University Innsbruck, Innsbruck, Austria. 3 Department of Internal Medicine, Medical University Innsbruck, Innsbruck, Austria. AJR 2009; 193: X/09/ American Roentgen Ray Society Real-Time Sonoelastography of Lateral Epicondylitis: Comparison of Findings Between Patients and Healthy Volunteers OBJECTIVE. The purpose of this study was to evaluate real-time sonoelastography in the assessment of the origins the common extensor tendon in healthy volunteers and in patients with symptoms of lateral epicondylitis. The findings were compared with those obtained at clinical examination, ultrasonography, and power Doppler sonography. SUBJECTS AND METHODS. Thirty-eight elbows of 32 consecutively registered patients with symptoms of lateral epicondylitis and 44 asymptomatic elbows of 28 healthy volunteers were assessed with ultrasound and real-time sonoelastography. A clinical examination was performed, and pain was classified with a visual analog scale. RESULTS. In healthy volunteers, real-time sonoelastographic images showed hard tendon structures in 96% of tendon thirds and mild alterations in 4%. Real-time sonoelastography of patients showed hard structures in 33% of tendon thirds but softening of different grades in 67%, a statistically significant difference in relation to the findings in healthy volunteers (p < 0.001). Lateral collateral ligament involvement and overlying fascial involvement were more commonly detected with real-time sonoelastography. The sensitivity of real-time sonoelastography was 100%, the specificity 89%, and the accuracy 94% with clinical examination as the reference standard. Good correlation with ultrasound findings was found (r 0.900). No correlation was observed between ultrasound or real-time sonoelastographic findings and power Doppler sonographic findings, but power Doppler sonographic findings had a strong correlation with the visual analog scale score. CONCLUSION. Real-time sonoelastography is valuable in the detection of the intratendinous and peritendinous alterations of lateral epicondylitis and facilitates differentiation between healthy and symptomatic extensor tendon origins with excellent sensitivity and excellent correlation with ultrasound findings. T he diagnosis of lateral epicondylitis, known as tennis elbow, is commonly based on clinical findings. However, imaging such as conventional sonography and MRI of the origin of the common extensor tendons often is used to confirm the diagnosis. Furthermore, imaging findings provide information about the severity of disease and differential diagnosis [1, 2]. MRI has been found more sensitive than ultrasound with a 5- to 10-MHz transducer in the detection of signs of lateral epicondylitis [3]. Ultrasound studies have shown sensitivities ranging between 72% and 88% and specificities ranging between 36% and 48.5% in comparison with clinical findings [1]. Tendon compressibility and compressibility of intratendinous vessels have been postulated as helpful diagnostic criteria in the sonographic assessment of lateral epicondylitis [4]. The aim of this study was to evaluate tendon compressibility with real-time sonoelastography, which has been found to add information in the diagnostic evaluation of cancer through the assessment of tissue elasticity [5 8]. Elastography was first described by Ophir et al. in [9] 1991, and in 1999, Pesavento et al. [10] developed a fast cross-sectional technique based on real-time elastographic imaging. The principle of real-time sonoelastography is that tissue compression produces strain (displacement) within tissue and that the strain is less in hard tissue than in soft tissue [11]. Preliminary results in the evaluation of Achilles tendinopathy showed that healthy tendons were characterized by harder tissue than were diseased tendons, which exhibited a softer tissue spectrum at real-time sonoelastography (De Zordo T et al., presented at the 2007 annual meeting of 180 AJR:193, July 2009

2 Sonoelastography of Lateral Epicondylitis the Radiological Society of North America). Our specific goal was to evaluate realtime sonoelastography in the assessment of common extensor tendon origins in healthy volunteers and in patients with a clinically confirmed diagnosis of lateral epicondylitis. Correlation analysis of real-time sonoelastographic findings with the clinical examination, ultrasound, and power Doppler ultrasound findings was performed. Subjects and Methods The study was approved by the institutional review board, and informed oral and written consent was obtained from all patients and healthy volunteers. A prospective analysis with ultra sound and real-time sonoelastography was per formed on 38 elbows of 32 consecutively enrolled patients (six patients underwent imaging of both elbows) with clinical symptoms of lateral epi condylitis (10 men, 22 women; mean age, 52.6 years; range, years). In addition, 44 asymptomatic elbows (16 subjects, both elbows) of 28 healthy volunteers (11 men, 17 women; mean age, 43.6 years; range, years) were assessed. The healthy volun teers underwent a clinical exami nation to exclude lateral epicondylitis. Further exclusion criteria for participation of healthy volunteers were a history of tendon rupture or systemic in flammatory disorder, such as rheuma toid arthritis and hyper cholesterolemia. Clinical assessment included evaluation of local tenderness directly over the lateral epicondyle, evaluation of pain aggravation during resisted wrist extension and radial devi ation, and evaluation of decreased grip strength. Subjective pain was recorded with a visual analog scale (score, 0 100). Imaging The healthy volunteers and patients with lateral epicondylitis underwent a routine ultrasound scan, including power Doppler ultrasound (MyLab 90 scanner, Esaote) with a linear-array transducer (LA 435, Esaote) at a frequency of 6 18 MHz. A radiologist with 6 years experience in musculoskeletal ultrasound who was blinded to the clinical examination findings performed the examination. After the initial routine ultrasound scan, patients and healthy volunteers underwent an additional ultrasound exami nation with a real-time sono elastographic scanner (EUB 9000, EUP-L54M, Hitachi Medical) at a frequency range of 6 13 MHz. Realtime sonoelastography was performed by a radiologist with 3 years of experience in muscu loskeletal ultrasound who was blinded to the ultrasound, power Dop pler ultrasound, and clini cal examination findings. Real-time sono elasto graphic and ultrasound images were graded independently after 3 months by both radiologists in consensus who were unaware of the clinical results. The lateral aspect of the elbow was examined with thumbs up and the elbow in 90 flexion. Care was taken to allow a comfortable and relaxed position so that tension on the tendon was avoided. To improve transducer coupling, a generous amount of contact gel was placed on the skin over the tendon, and an additional acoustic standoff pad (Sonar Aid, Geistlich Pharma) was used. The transducer was positioned parallel to the longitudinal axis of the tendon to avoid anisotropy. Sonography, power Doppler ultrasound, and real-time sonoelastographic examinations were performed within 1 hour on the same day to avoid possible changes over time, exercise, and therapeutic intervention. Calculation of tissue elasticity distribution at real-time sonoelastography was performed in real time (up to 30 frames/s), and the examination results were represented in color over the conventional ultrasound image. The force applied to the tendon was adjusted appropriately according to the visual indicator seen on the video screen that showed optimal strain at the region of interest. Each real-time sonoelastographic scan was repeated by compression and relaxation of the scan area several times (at least three compression decompression cycles) until the findings were confirmed to be reproducible. These representative images were sent to the local PACS and used for statistical evaluation. According to the system of Connell et al. [12], tendon abnormalities were divided into three sections: the anterior, middle, and posterior fibers. At least three scans of each tendon third were performed. The extensor tendon origins were evaluated for the presence of focal lesions, including areas of degeneration and partial rupture, which were defined as hypoechoic areas at ultrasound, and as red to yellow (soft) areas at real-time sonoelastography. Focal lesions were counted to evaluate the dimension of focal lesions according to the following grading system: 0, no focal lesion or blue to green (hard) tendon; 1, one focal lesion; 2, two focal lesions; 3, more than two focal lesions. Alterations in the collateral lateral ligament and abnormalities of the overlying fascia were assessed. Collateral ligament involvement was defined as thickening of the humeroradial ligament with or without evidence of partial- or full-thickness rupture on ultrasound images [12] or as softening of the ligamentous structure at realtime sonoelasto graphy. Involvement of the overlying fascia was diagnosed when ultrasound showed thickening of the peritendinous tissue layer but real-time sono elastography showed irregular elasticity changes in the form of distinct softening accompanied by softening of the underlying tendon. Power Doppler ultrasound was used to assess whether intratendinous hyperemia, defined as color-flow signal in the extensor tendon origin, was present. Standardized settings (transmit power < 500 mw/cm 2, low-pass wall filter, medium persistence) were used and remained fixed throughout the study. These settings were chosen to maximize the sensitivity to low-velocity and low-volume blood flow. The power Doppler ultrasound gain was optimized with an increase in gain until noise appeared; then the gain was reduced slightly, only enough to suppress the noise (usually ~ 60 70% gain). We applied the appropriate color velocity scale using the musculoskeletal program of our ultrasound unit. The window (color box) was restricted to the vascular area studied. Statistical Analysis Statistical analysis was performed with SPSS software (release 13.0, SPSS). Standard descriptive statistics were used to summarize characteristics of the patients and healthy volunteers, including median and SD for the continuous variables and counts and percentages for the categoric variables. Continuous variables were compared between patients and controls by use of the Mann-Whitney U test, and categorical variables by use of the Fisher exact test. Correlation of parameters was performed with Spearman s nonparametric correlation. Statistical significance was defined as two-sided p 0.05, and Bonferroni corrections were made for multiple comparisons when appropriate. After Bonferroni correction for three comparisons, a value of p < 0.01 was considered statistically significant. With clinical examination and pain as the reference standards, sensitivity, specificity, and accuracy were calculated. Cor relations of para meters were analyzed with Spearman s nonpara metric correlation for ultrasound, real-time sono elastographic, power Doppler ultrasound, and clinical findings. Healthy volunteers were sex-matched to patients (p = 0.643), but age matching was not possible (p < 0.001) because the healthy volunteer sample was from a younger population than the patient sample. To obtain relevant data, a new data sheet with a filter variable for age was created (only controls age 38 years and older were included). After that, the Mann-Whitney U test showed an asymptomatic significance of p = for age; thus both groups were regarded as balanced for age. Diagnostic accuracy, sensitivity, specificity, and positive and negative predictive values were calculated, and the results showed that the values were similar compared with the values previously calculated without matching for age, the difference being less than 2.9% (p > 0.01). AJR:193, July

3 De Zordo et al. TABLE 1: Overall Findings in Elbows of Patients: Incidence of Focal Lesions and Involvement of Single Tendon Thirds No. of focal lesions Finding Ultrasound (n = 38) Real-Time Sonoelastography (n = 38) 0 2 (5) 0 (0) 1 10 (26) 10 (26) 2 15 (40) 18 (48) 3 11 (29) 10 (26) Location Anterior third 19 (50) 20 (53) Middle third 34 (89) 36 (95) Posterior third 20 (53) 20 (53) Note Values in parentheses are percentages. A C E Results Among the patients, the median duration of symptoms was 9 ± (SD) months (range, months). Median pain strength was a visual analog scale score of 77.5 ± (range, ). In healthy volunteers, no symptomatic common extensor tendon origin was detected. The overall ultrasound and real-time sonoelastographic findings are shown in Table 1. In healthy volunteers, real-time sonoelastography showed hard tendon structures, corresponding to blue coloring, in 96.3% (127/132) of common extensor tendon thirds (Fig. 1). Grade 1 lesions were found in the other 3.7% (5/132) (Table 2). In patients, real-time sonoelastography showed elastic alterations of the pathologic tendon in 66.7% (76/114) of tendon sections and normal elasticity in 33.3% (38/114). Grade 1 changes were detected in 34.2% (39/114), two focal lesions in 21.1% (24/114), and three or more focal lesions in 11.4% (13/114) (Fig. 2). Statistical analysis of real-time sonoelastographic findings showed a significant difference between the findings in healthy volunteers and those in patients (p < 0.001). In healthy volunteers, ultrasound showed 96.3% (127/132) of tendon thirds were normal. One focal lesion was detected in the other 3.7% (5/132). In patients, ultrasound showed 36.0% (41/114) of tendon thirds were normal, 33.3% (38/114) had one focal lesion, 18.4% (21/114) had two focal lesions, and 12.3% (14/114) had more than two focal lesions. In healthy volunteers, no involvement of the collateral ligament or of the overlying fascia was found. Symptomatic collateral ligament involvement was found in 10 elbows (26.3%) at real-time sonoelastography and in eight (21.1%) at ultrasound (Fig. 3). Overlying fascial involvement was found at real-time sonoelastography in 11 (28.9%) and at ultrasound in five (13.2%) elbows with symptomatic abnormalities. Real-time sonoelastography had a sensitivity of 100%, specificity of 89%, and accuracy of 94% compared with clinical examination. The positive predictive value was 88%, and the negative predictive value was 100%. Ultrasound had a sensitivity of 95%, specificity of 89%, accuracy of 91%, positive predictive value of 88%, and negative predictive value of 95%. The correlation between real-time sonoelastography and ultrasound was high, with correlation coefficients of (p < 0.001) for the anterior section, (p < 0.001) for the middle section, and (p < 0.001) for the posterior section of the tendon. No correlation between realtime sonoelastography and power Doppler ultrasound or between ultrasound and power Doppler ultrasound was observed. However, power Doppler ultrasound had a significant correlation with the visual analog scale score (r = 0.844; p < 0.001). Discussion The cause of epicondylitis is considered to be repetitive microtrauma sustained during supination of the forearm and dorsiflexion of the wrist that results in tendon degeneration with rupture of individual collagen fibers that stimulates a reparative response [12, 13]. In chronic disease, a cycle of tendon degeneration and repair results in weakening of the common extensor origin and leads to risk of rupture [13]. With conventional ultrasound, it can be difficult or even impossible to differentiate tissue affected by degenerative disease from healthy tendon because damaged tissue often has the same echogenicity as the surrounding healthy tissue [6]. However, it is well known that inflammation and tumors cause changes in tissue elasticity [6]. Increased compressibility has been described [4] as a new ultrasound Fig year-old healthy volunteer with normal extensor tendons. A F, Longitudinal ultrasound (A, C, and E) and real-time sonoelastographic (B, D, and F) scans show normal anterior third (A and B), middle third (C and D), and posterior third (E and F) of common extensor tendon. Elasticity spectrum was between red, representing soft tissue, and blue, representing hard tissue. Ultrasound findings of normal tendons were those of hard tissue (stars) on real-time sonoelastographic images. Surrounding tissue can be appreciated as having soft structure but clearly discernible from tendon. Bony artifact shown as red areas inside bone are present on all real-time sonoelastographic images. r = radial head, lat. epi. = lateral epicondyle. B D F 182 AJR:193, July 2009

4 Sonoelastography of Lateral Epicondylitis TABLE 2: Detection of Focal Lesions With Ultrasound and Real-Time Sonoelastography in Elbows of Healthy Volunteers and Patients Ultrasound Grade Healthy Volunteers (n = 44) Patients (n = 38) p Anterior section < (100) 19 (50.00) 1 0 (0) 10 (26.32) 2 0 (0) 5 (13.16) 3 0 (0) 4 (10.52) Middle section < (88.64) 4 (10.52) 1 5 (11.36) 18 (47.37) 2 0 (0) 10 (26.32) 3 0 (0) 6 (15.79) Posterior section < (100) 18 (47.37) 1 0 (0) 10 (26.32) 2 0 (0) 6 (15.79) 3 0 (0) 4 (10.52) Real-time sonoelastography Anterior section < (100) 18 (47.37) 1 0 (0) 10 (26.32) 2 0 (0) 7 (18.42) 3 0 (0) 3 (7.89) Middle section < (90.91) 2 (5.26) 1 4 (9.09) 17 (44.74) 2 0 (0) 13 (34.21) 3 0 (0) 6 (15.79) Posterior section < (97.73) 18 (47.37) 1 1 (2.27) 12 (31.58) 2 0 (0) 4 (10.52) 3 0 (0) 4 (10.52) Note Values in parentheses are percentages. sign of lateral epicondylitis. Thus, estimation of tissue softening may be a useful tool for characterization of an intratendinous focal lesion or peritendinous involvement in patients with lateral epicondylitis. In our study, healthy common extensor tendon origins were found to have harder tissue characteristics (grade 0) at real-time sonoelastography; only 3% of tendon thirds had mild alterations (grade 1). In tendon origins with symptomatic abnormalities, distinct focal lesions were detected with real-time sonoelastography and scored grade 1 in 34% of cases, grade 2 in 21%, and grade 3 in 11%. Accordingly, it can be presumed that lateral epicondylitis is associated with considerable softening of intratendinous tissue. The histologic changes of lateral epicondylitis are described as collagen fibrillar degeneration, angiofibroblastic proliferation, tissue necrosis with myxoid and hyaline degeneration, and fibrosis [14]. These histopathologic alterations of lateral epicondylitis have been found to increase the compressibility of tissue at ultrasound [4] and may cause softening of tissue at real-time sonoelastography. However, no histopathologic analysis has been performed, to our knowledge, and further studies are needed. The most common ultrasound finding of lateral epicondylitis is a focal area of low echogenicity corresponding to areas of collagen degeneration and intrasubstance fiber rupture that can fill with reparative granulation tissue [12], typically at the origin of the extensor carpi radialis brevis tendon and less commonly at the anterior aspect of the extensor digitorum communis tendon [15]. In our study, slightly more focal lesions were found with real-time sonoelastography (81 lesions) than with ultrasound (78 lesions). Unlike Connell et al. [12], we found most changes in the middle part of the tendon (95% at realtime sonoelastography, 90% at ultrasound) and not in the anterior part (53% at real-time sonoelastography, 50% at ultrasound). However, both parts represent fibers of the extensor carpi radialis brevis tendon. When focal areas enlarge and extend to the surface, partial rupture or even complete tears ensue [12]. Partial tears are thought to be less likely to respond to conservative treatment [16], but discrimination between focal areas of tendinopathy and partial tears can be difficult with ultrasound [1] and was not possible with real-time sonoelastography in our study. However, in all patients with partial ruptures detected with ultrasound, grade 3 alterations (more than two focal lesions) were identified with real-time sonoelastography. Therefore, high-grade alterations found at real-time sonoelastography should be addressed in follow-up studies. Lateral collateral ligament injury is a common cause of therapeutic failure and should be routinely assessed with ultrasound [1]. In our study, more cases of ligament involvement were detected with real-time sonoelastography (26%) than with ultrasound (21%). These patients may need surgical therapy if conservative treatment fails. Follow-up studies are needed to confirm this hypothesis. Injections of corticosteroids are a known treatment of patients with lateral epicondylitis [17]. Although the exact healing mechanism is not known, corticosteroids are assumed to prevent spread of neovessels in the tendon substance and to loosen adhesions between peritendinous and tendinous tissue. Therefore, we investigated involvement of the overlying fascial structure. Real-time sonoelastography showed clear differentiation AJR:193, July

5 De Zordo et al. between tendon and fascial structures in healthy volunteers, but involvement of the common tendon and fascia was found in patients (11 of 38 elbows with symptomatic abnormalities). Whether these findings affect therapy is being addressed in additional studies. Furthermore, other treatment options, such as percutaneous needle tenotomy and platelet-enriched plasma injection, have shown great promise not only in eliminating pain but also in stimulating regeneration of A C E Fig. 2 Findings in patients. All alterations (arrow) on ultrasound images (A, C, and E) appear as areas of softening on real-time sonoelastographic images (B, D, and F). All images are longitudinal scans. lat. epi. = lateral epicondyle, r = radial head. A and B, 41-year-old woman with intratendinous tendinopathy. C and D, 27-year-old man with partial tear. E and F, 47-year-old man with large area of tendinopathy and involvement of superficial structures. A C Fig. 3 Comparison of normal and abnormal findings. r = radial head, lat. epi. = lateral epicondyle. A and B, 26-year-old healthy volunteer. Longitudinal ultrasound (A) and real-time sonoelastographic (B) images show radial collateral ligament (arrow). B shows hard ligamentous tissue and small area of softening at insertion of common extensor tendon. C and D, 26-year-old woman with abnormal ligament. Longitudinal ultrasound scan (C) shows no alteration in radial collateral ligament (arrow), but real-time sonoelastographic image (D) shows ligament (arrow) is soft. B D F B D tendon and adjacent soft tissue [18, 19]. Because of the great utility of ultrasound in direct guidance of injections at specific sites in tendinopathy, further studies should prove whether real-time sonoelastography can be used for follow-up examinations. Levin et al. [1] found high sensitivity (72 88%) but low specificity ( %) of ultrasound in the diagnosis of symptomatic lateral epicondylitis. Our real-time sonoelastographic findings had a sensitivity of 100%, a specificity equal to that of ultrasound (89%), and greater accuracy than ultrasound (real-time sonoelastography, 94%; ultrasound, 91%). Furthermore, good correlation with ultrasound findings was found (r = ). Our findings were distinctly better than those of Levin et al., but the specificity was lower. A potential cause of the low specificity is that a spectrum of pathologic imaging findings may be present before the onset of symptoms, leading to false-positive findings if clinical examination is the reference standard. Connell et al. [12] argued that power Doppler ultrasound failed to show vascularization in epicondylitis, and Zanetti et al. [20] found a limited role of power Doppler ultrasound of tendons in outcome prediction. However, it has been suggested [21, 22] that chronic degenerative tendinopathy is associated with hyperemia of uncertain origin. Interestingly, we did not find any correlation between ultrasound and power Doppler ultrasound or between real-time sonoelastography and power Doppler ultrasound, but we did find that among patients power Doppler ultrasound findings had a strong correlation with visual analog scale pain score (r = 0.844). Several limitations have to be mentioned. First, ultrasound examinations in general depend on the individual operator. Although care was taken to obtain reproducible images, we did not calculate interobserver or intraobserver variability. We avoided high pressure and overly low pressure to obtain a relevant conclusion because even on images for which pressure was decreased below a certain level, the pattern of elasticity started to change markedly. Furthermore, real-time sonoelastographic construction of images led to artifacts; the red of soft tissue was found in the elbow and surrounding bony structures. With practice, however, it became easy to differentiate artifacts from reliable images. Another limitation was the problem that with a certain extent of rupture, several focal lesions on real-time sonoelastography merged and looked like one single large focal lesion. We conclude that real-time sonoelastography can be used to differentiate healthy and symptomatic diseased extensor tendon origins at the lateral epicondyle with excellent correlation with ultrasound and clinical examination findings. Real-time sonoelastography also may be a powerful diagnostic adjunct to ultrasound and power Doppler ultrasound in a detailed, accurate, and sensitive combined diagnostic approach to suspected lateral epicondylitis. 184 AJR:193, July 2009

6 Sonoelastography of Lateral Epicondylitis FOR YOUR INFORMATION References 1. Levin D, Nazarian LN, Miller TT, et al. Lateral epicondylitis of the elbow: US findings. Radiology 2005; 237: Bredella MA, Tirman PF, Fritz RC, et al. MR imaging findings of lateral ulnar collateral ligament abnormalities in patients with lateral epicondylitis. AJR 1999; 173: Miller TT, Shapiro MA, Shults E, et al. Comparison of sonography and MRI findings for diagnosing epicondylitis. J Clin Ultrasound 2002; 30: Khoury V, Cardinal E. Tenomalacia : a new sonographic sign of tendinopathy? Eur Radiol 2008; 19: Varghese T, Ophir J, Konofagou E, et al. Tradeoffs in elastographic imaging. Ultrason Imaging 2001; 23: Frey H. Realtime elastography: a new ultrasound procedure for the reconstruction of tissue elasticity [in German]. Radiologe 2003; 43: Pallwein L, Mitterberger M, Pinggera G, et al. Sonoelastography of the prostate: comparison with systematic biopsy findings in 492 patients. Eur J Radiol 2008; 65: Lyshchik A, Higashi T, Asato R, et al. Thyroid gland tumor diagnosis at US elastography. Radiology 2005; 237: Ophir J, Cespedes I, Ponnekanti H, et al. Elastography: a quantitative method for imaging the elasticity of biological tissues. Ultrason Imaging 1991; 13: Pesavento A, Perrey C, Krueger M, Ermert H. A time-efficient and accurate strain estimation concept for ultrasonic elastography using iterative phase zero estimation. IEEE Trans Ultrason Ferroelectr Freq Contr 1999; 46: Itoh A, Ueno E, Tohno E, et al. Breast disease: clinical application of US elastography for diagnosis. Radiology 2006; 239: Connell D, Burke F, Coombes P, et al. Sonographic examination of lateral epicondylitis. AJR 2001; 176: Nirschl RP, Perrone FA. Tennis elbow: the surgical treatment of lateral epicondylitis. J Bone Joint Surg Am 1979; 61: Chard MD, Cawston TE, Riley GP, et al. Rotator cuff degeneration and lateral epicondylitis: a comparative histological study. Ann Rheum Dis 1994; 53: Bianchi S, Martinoli C. Elbow. In: Bianchi S, Martinioli C, eds. Ultrasound of the musculoskeletal system. Berlin, Germany: Springer-Verlag, 2007: Regan W, Wold LE, Coonrad R, et al. Microscopic histopathology of chronic refractory lateral epicondylitis. Am J Sports Med 1992; 20: Tonks JH, Pai SK, Murali SR. Steroid injection therapy is the best conservative treatment for lateral epicondylitis: a prospective randomised controlled trial. Int J Clin Pract 2007; 61: Connell DA, Ali KE, Ahmad M, et al. Ultrasoundguided autologous blood injection for tennis elbow. Skeletal Radiol 2006; 35: McShane JM, Nazarian LN, Harwood MI. Sonographically guided percutaneous needle tenotomy for treatment of common extensor tendinosis in the elbow. J Ultrasound Med 2006; 25: Zanetti M, Metzdorf A, Kundert HP, et al. Achilles tendons: clinical relevance of neovascularization diagnosed with power Doppler US. Radiology 2003; 227: Astrom M. Laser Doppler flowmetry in the assessment of tendon blood flow. Scand J Med Sci Sports 2000; 10: du Toit C, Stieler M, Saunders R, et al. Diagnostic accuracy of power-doppler ultrasound in patients with chronic tennis elbow. Br J Sports Med 2008; 42: The AJR has made getting the articles you really want really easy with an online tool, Really Simple Syndication, available at It s simple. Click the yellow RSS button located in the menu on the left of the page. You ll be on your way to syndicating your AJR content in no time. AJR:193, July