The Role of Sonography in Differentiating Full Versus Partial Distal Biceps Tendon Tears: Correlation With Surgical Findings

Transcription

1 Musculoskeletal Imaging Original Research Da Gama Lobo et al. Sonography in Biceps Tendon Tears Musculoskeletal Imaging Original Research Lucas Da Gama Lobo 1,2 David P. Fessell 1 Bruce S. Miller 1,3 Aine Kelly 1 Jee Young Lee 1,4 Catherine Brandon 1 Jon A. Jacobson 1 Da Gama Lobo L, Fessell DP, Miller BS, et al. Keywords: biceps, elbow, tendon tear, ultrasound DOI: /AJR Received May 31, 2011; accepted after revision May 18, Department of Radiology, University of Michigan Hospital, 1500 E Medical Center Dr, Ann Arbor, MI Address correspondence to D. P. Fessell (dfessell@med.umich.edu). 2 Present address: Department of Radiology, Mount Sinai Hospital, Toronto, ON, Canada. 3 Department of Orthopedic Surgery, University of Michigan Hospital, Ann Arbor, MI. 4 Present address: Department of Radiology, Dankook University Hospital, Dankook University School of Medicine, Chungchungnam-Do, Korea. AJR 2013; 200: X/13/ American Roentgen Ray Society The Role of Sonography in Differentiating Full Versus Partial Distal Biceps Tendon Tears: Correlation With Surgical Findings OBJECTIVE. The purpose of this study was to determine the accuracy of ultrasound for distinguishing complete rupture of the distal biceps tendon versus partial tear and versus a normal biceps tendon. Surgical findings were used as the reference standard in cases of tear. Clinical follow-up was used to assess the normal tendons. MATERIALS AND METHODS. The study population consisted of 45 consecutive elbow ultrasound cases with surgical confirmation and six cases of a clinically normal distal biceps tendon that underwent elbow ultrasound for suspicion of injury to a structure other than the biceps tendon. Cases underwent consensus review by two fellowship-trained musculoskeletal radiologists. Tendons were classified as normal biceps tendon, partial tear, or complete tear. The presence or absence of posterior acoustic shadowing at the distal biceps tendon was also assessed. The ultrasound findings were then compared with the surgical findings and clinical follow-up. RESULTS. Ultrasound showed 95% sensitivity, 71% specificity, and 91% accuracy for the diagnosis of complete versus partial distal biceps tendon tears. Posterior acoustic shadowing at the distal biceps had sensitivity of 97% and accuracy of 91% for indicating complete tear versus partial tear and sensitivity of 97%, specificity of 100%, and accuracy of 98% for indicating complete tear versus normal tendon. CONCLUSION. Ultrasound can play a role in the diagnosis of elbow injuries when a distal biceps brachii tendon tear is suspected. T ears of the distal biceps brachii tendon are less common than tears of the proximal long head of the biceps brachii tendon; however, complete tears are almost always treated surgically because of loss of strength and function with nonoperative management [1, 2]. The clinical diagnosis of a complete rupture of the distal biceps brachii tendon can be straightforward in cases in which there is retraction of the muscle but is more difficult in cases of nonretracted complete rupture with an intact lacertus fibrosus, which can prevent muscle retraction [3, 4]. Less severe forms of distal biceps brachii abnormality, such as partial tears, tendinosis, and inflammation of the bicipitoradial bursa, may be difficult to diagnose on clinical grounds. Accurate and timely diagnosis of complete ruptures of the distal biceps brachii tendon is important because optimal surgical results are achieved within the first 3 6 weeks after tear [5 8]. Both ultrasound and MRI have been shown to be useful for evaluating the distal biceps brachii tendon [9 12]. MRI has shown findings that strongly correlate with surgical findings for both partial and complete tears of the distal biceps brachii [10, 11]. However, MRI is more expensive than ultrasound, may be less accessible, and is contraindicated in some patients, including those with aneurysm clips. Ultrasound is less expensive, and the equipment is widely available. Furthermore, ultrasound enables dynamic examination as well as comparison with the contralateral extremity. In a small series of patients, ultrasound has been shown to be useful in the diagnosis of distal biceps brachii tendon tears, but, to our knowledge, there have been no published reports of a large patient population with surgical correlation [9, 12]. To determine the role of ultrasound, our study retrospectively compared ultrasound findings with surgical results for acute biceps tears and compared ultrasound findings with clinical follow-up for normal tendons. Materials and Methods This study was granted exemption by our institutional review board for both approval and patient 158 AJR:200, January 2013

2 Sonography in Biceps Tendon Tears consent. A retrospective database search of the radiology archive from 2002 to 2009 was completed to identify patients who had ultrasound evaluation of the distal biceps brachii tendon. Only those cases with surgical follow-up and no prior surgery on the distal biceps brachii were included in the final surgical subject group. An additional six cases of patients who underwent elbow ultrasound for nonbiceps abnormalities were also identified for inclusion as normal biceps cases. These cases were scanned for clinical concern for the following abnormalities: ulnar collateral ligament injury (two cases) and triceps, flexor tendon attachment, lateral epicondylitis, and concern for inflammatory arthritis (one case each). One surgical case was excluded because both elbows were scanned but all images were labeled as only one side. The inclusion criterion for the surgical cases was sonographic evaluation of the distal biceps brachii by one of nine fellowship-trained musculoskeletal radiologists (2 14 years of musculoskeletal ultrasound experience) as part of routine patient care using a commercially available scanner (Logiq 9, GE Healthcare, with a 6-15 or 7-11 MHz linear array transducer or iu22, Philips Healthcare, with a 12-5 or 17-5 MHz linear array transducer). In general, the biceps brachii muscle and tendon were evaluated from the musculotendinous junction to the insertion at the radial tuberosity. Longitudinal and transverse images were obtained with the elbow in extension or mild flexion. Ultrasound imaging during muscle contraction as well as comparison with the other side was performed as deemed necessary by the radiologist. Similarly, the use of dynamic scanning of the distal biceps from a medial or lateral approach during elbow flexion was used at the discretion of the radiologist [13, 14]. A Fig year-old man with complete biceps brachii tear correctly identified by ultrasound. A, Ultrasound image with extended FOV shows complete tear with 11 cm of retraction between thin arrow and thick arrow on radial tuberosity. B = biceps muscle, BR = brachialis muscle, L = lateral epicondyle of humerus. B, More focused longitudinal ultrasound image shows torn tendon end (white arrows) with posterior acoustic shadowing noted (black arrow). All ultrasound images were reviewed in consensus by two fellowship-trained musculoskeletal radiologists with 6 and 13 years of experience with musculoskeletal ultrasound. One reviewer had no special knowledge of the ultrasound cases and the other reviewer had not viewed any of the cases for approximately 6 months. Six elbow ultrasound examinations with a clinically normal distal biceps tendon were also included in the consensus review. All distal biceps tendon cases were viewed in randomized order and assessed as normal, partial tear, or complete tear. A complete distal biceps tendon tear was defined as a full-width, full-thickness tear. A partial tear was defined as a partial-width tear, partial thickness tear, or both. Posterior acoustic shadowing from the distal biceps was also noted to be present or absent in each case at the consensus review. Operative reports were retrospectively reviewed so that each surgical patient was categorized as having a partial or complete tear of the biceps brachii. Sensitivity, specificity, and accuracy were determined for distinguishing complete ruptures from partial tears. Time intervals from imaging to surgery with partial and complete tears were compared using the Mann Whitney U Test. Length of clinical follow-up ultrasound was also calculated for the clinically normal distal biceps tendon group. Results The subject group consisted of 45 consecutive patients with both ultrasound of the distal biceps brachii and subsequent surgical treatment and six patients with a normal biceps tendon who underwent clinical rather than surgical follow-up. Of the 45, there were 36 subjects with complete tears, seven with partial-thickness tears, and none with a normal finding at surgery. All patients in both the surgical and clinical follow-up groups were men, with an average age of 44 years (age range, years). For the majority of the surgical cases, the mechanism of injury was flexion of the elbow against a weight (eccentric contraction). Median time from ultrasound to surgery was 8 days for complete tears (average, 12 days; range, 3 90 days) and 38 days for partial tears (average, 57 days; range, days), which was significantly different (Mann Whitney U Test, p < 0.001). The average clinical follow-up time after ultrasound for those with a normal biceps tendon at ultrasound was 182 days. Of the 38 surgically confirmed complete tears, ultrasound correctly reported a complete tear preoperatively in 95% (Figs. 1 and 2). In two cases, a complete tear at surgery was interpreted as a partial tear with ultrasound (Fig. 3). Of the seven partial tears confirmed at surgery, ultrasound correctly identified five (Fig. 4). In two cases, ultrasound described a complete tear, but a partial tear was noted at surgery. These results are summarized in Table 1. In no cases did ultrasound describe a tear where no tear (partial or complete) was found at surgery. In no cases did the consensus review identify a normal tendon as torn (partial tear or complete tear). Table 2 tabulates the presence of shadowing in complete tears, partial tears, and normal tendons. Shadowing is illustrated in Figures 1 and 5. The sensitivity for shadowing to indicate complete tear versus partial tear was B AJR:200, January

3 Da Gama Lobo et al. Fig year-old man with complete biceps brachii tendon tear correctly diagnosed by ultrasound. Ultrasound image of long axis to biceps brachii tendon shows retracted tendon (arrows) and intervening hypoechoic fluid and hematoma. RT = radial tuberosity, RH = radial head. 97% with accuracy of 91%. The specificity (four of seven cases) was low and not of great value given such small numbers. The sensitivity for shadowing to indicate complete tear versus normal tendon was 97% (37/38) with specificity of 100% (6/6) and accuracy of 98% (43/44). The sensitivity for shadowing to indicate partial tear versus normal was low and not of great value given the small numbers (3/7). The accuracy of shadowing to correctly diagnose partial tears versus normal tendon was also of limited value given the small numbers (9/10 cases). In all partial tears, the remaining intact fibers were markedly thickened and hypoechoic with fluid, debris, or hematoma filling the gap at the site of the torn fibers. Thirty-four of the 38 cases of complete rupture were surgically treated with anatomic reinsertion of the distal biceps brachii tendon to the radial tuberosity using a single anterior incision (20 cases) or the Boyd- Anderson two-incision technique (14 cases). The two-incision technique uses a small incision at the radial tuberosity and a small incision more proximally. The torn tendon is tunneled through the soft tissues from the proximal to the distal incision. Four cases of complete tear were treated with a biceps tenodesis to the brachialis tendon muscle. Six of the seven patients with a partial tear were treated with anatomic reinsertion (five patients with the single incision and one with the two-incision technique). The other partial tear was treated with biceps tenodesis to the brachialis. Five orthopedic surgeons performed the 45 surgeries, with two surgeons performing 37 of the 45 studies. Fig year-old man with complete biceps brachii tendon tear incorrectly interpreted as partial-thickness tear by ultrasound. Ultrasound image of long axis to biceps brachii tendon shows heterogeneous scar tissue (thin arrows) between torn tendon end (thick arrow) and radial tuberosity (RT). Fig year-old man with partial-thickness biceps brachii tendon tear correctly identified by ultrasound. A and B, Ultrasound images of long axis (A) and short axis (B) to biceps brachii tendon show hypoechoic and thickened short head of biceps tendon (thin arrows) with sparing of more lateral long head of biceps tendon (thick arrows, B) Degree of thickening and hypoechogenicity and irregularity of deep margin of tendon in longitudinal plane favored partial tear over tendinosis. RH = radial head, RT= radial tuberosity, L = lateral epicondyle. A Discussion Distal biceps tendon tears tend to occur in middle-aged men and are usually of traumatic origin. The mechanism of injury often involves large force acting against resistance from a flexed elbow, such as in weightlifting or gymnastics [6 8, 11]. Accurate diagnosis is important because prompt surgical repair of complete tears is optimal. Our results show that ultrasound can differentiate complete versus partial tear with 95% sensitivity, 71% specificity, and 91% accuracy. The presence of posterior acoustic shadowing has been shown to have a high correlation with complete rupture of the Achilles tendon [15]. This was also seen in the current study, with shadowing showing high sensitivity (97%), specificity (100%), and accuracy (98%) for complete tear of the distal biceps tendon tears versus a normal tendon (Figs.1 and 5). The high specificity suggests that lack of shadowing can help exclude a complete tear. Further study with a larger number of healthy tendons B 160 AJR:200, January 2013

4 Sonography in Biceps Tendon Tears TABLE 1: Ultrasound Findings Compared With Surgical Findings in Partial and Complete Tears Ultrasound Findings TABLE 2: Shadowing at Ultrasound Ultrasound Findings Complete Surgical or Clinical Follow-Up Findings Complete Partial Normal Shadowing No shadowing Total Surgical Findings Partial Complete Partial Total Total Total Fig year-old man with complete distal biceps rupture correctly identified by ultrasound. Longitudinal ultrasound image shows torn and markedly retracted and redundant tendon (thick arrow) with marked posterior shadowing from torn tendon end (thin arrows). can help confirm these findings. Shadowing has poor sensitivity for partial versus normal tears (43%) because four of seven partial tears in our study did not show posterior acoustic shadowing from the torn tendon. Because only a portion of the tendon is torn with a partialthickness tear, shadowing is more variable. Our findings suggest a lack of shadowing on ultrasound does not reliably exclude a partial tear. More study is needed with a greater number of partial tears. Seiler et al. [16] studied the potential sites of biceps brachii tears on the basis of anatomic and CT studies to identify causes of tendon tears. They hypothesized that a combination of mechanical impingement and arterial supply was the primary cause of tears. There are three vascular zones within the distal biceps brachii with proximal zone 1 supplied by the brachial artery and distal zone 3 supplied by the posterior recurrent artery. Zone 2, approximately 2 cm proximal to insertion, represents a relatively hypovascular zone. During pronation, the proximal radioulnar space is reduced by half. Thus mechanical impingement during rotation and a relatively hypovascular region (zone 2) may contribute to tendon tears. In several small studies, MRI has shown a strong correlation with surgical findings for both partial and complete tears of the distal biceps brachii tendon [10, 11]. Using MRI, Fitzgerald et al. [11] studied 21 patients with clinically suspected injury of the distal biceps tendon, and they identified 12 complete distal biceps tears, four partial tears, one brachialis tear, and one ganglion. Surgical correlation was performed in 15 patients, with 100% agreement between MRI and surgical findings. MRI findings led to changes in clinical treatment plans in 38% of patients. Falchook et al. [10] studied distal biceps injuries with MRI. Complete tendon rupture was diagnosed in 10 patients and partial tears in six. Nine patients had surgical correlation, and the MRI findings were confirmed at surgery. Ultrasound has also shown a similar high correlation with surgical findings in several small series of distal biceps brachii tears. Belli et al. [9] examined 25 patients with clinically indicated distal biceps tendon ruptures. Ultrasound noted 17 of 18 full-thickness tears, with 14 of the 18 having surgical confirmation [9]. Using ultrasound, Miller and Adler [12] examined seven patients for distal biceps abnormalities. Ultrasound correctly diagnosed four of the five surgically confirmed complete tears. Two partial tears were correctly diagnosed using ultrasound, one of which was surgically confirmed [12]. Our larger series of 45 patients shows sensitivity of 95% for the sonographic detection of complete tears and accuracy of 91% for diagnosing tears as complete versus partial. There were two cases in which a complete tear was interpreted as a partial-thickness tear at ultrasound. Scar tissue was described at the site of the tendon tear in the operative report of one of these cases and may have been mistaken at ultrasound for intact tendon fibers. Because only seven surgically confirmed partial tears were found at surgery in our study, we cannot draw strong conclusions regarding the accuracy of ultrasound for identifying partial tears. The time interval between ultrasound and surgery for partial tears had a median of 38 days and was significantly greater than the median time of 8 days for complete ruptures. This also limits evaluation because interval healing or further tearing could take place if there is an extended time period between ultrasound and surgery. The recommended treatment for complete and high-grade partial tears of the distal biceps is surgical repair because it is known that patients treated with conservative management lose approximately 30% of flexion strength and 40% of supination strength [2, 8, 17]. This corresponds with our data because the median time from ultrasound to surgery for complete tears was 8 days versus 38 days for partial tears. Nonoperative management of complete tears is usually recommended only to elderly patients with low-demand activities who are also not concerned by cosmetic considerations (e.g., a mass in the upper arm due to tendon retraction) [17]. Limitations of this study include its retrospective nature. Surgery was used as the standard of reference for the majority of cases and introduces bias; however, the direct visualization of the tendon at surgery is a stronger reference standard than clinical follow-up. The small number of partial tears and nor- AJR:200, January

5 Da Gama Lobo et al. mal tendons also limits statistical evaluation. Further study is needed with larger numbers of partial tears and normal tendons. An additional limitation is the multiple radiologists and surgeons involved in this retrospective study. This could introduce some level of variability in the sonographic and surgical interpretations. Such variability, however, provides a more accurate picture of the range of interpretations and accuracy that would be seen in an average clinical practice. In conclusion, in a series of 45 surgically confirmed cases, ultrasound has shown 95% sensitivity, 71% specificity, and 91% accuracy in the diagnosis of complete tears of the distal biceps brachii tendon versus partial tears. The presence of posterior acoustic shadowing can help indicate a complete tear (sensitivity, 97%) versus both partial tears and normal tendons. However, lack of shadowing does not reliably exclude a partial tear. Ultrasound can play a role in the diagnosis of elbow injuries in which distal biceps brachii tendon tear is suspected. References 1. Agins HJ, Chess JL, Hoekstra DV, Teitge RA. Rupture of the distal insertion of the biceps brachii tendon. Clin Orthop Relat Res 1988; Baker BE, Bierwagen D. Rupture of the distal tendon of the biceps brachii: operative versus nonoperative treatment. J Bone Joint Surg Am 1985; 67: Le Huec JC, Moinard M, Liquois F, Zipoli B, Chauveaux D, Le Rebeller A. Distal rupture of the tendon of biceps brachii: evaluation by MRI and the results of repair. J Bone Joint Surg Br 1996; 78: Teh J. Imaging of the elbow. Imaging 2007; 19: Bourne MH, Morrey BF. Partial rupture of the distal biceps tendon. Clin Orthop Relat Res 1991; D Alessandro DF, Shields CL Jr, Tibone JE, Chandler RW. Repair of distal biceps tendon ruptures in athletes. Am J Sports Med 1993; 21: Klonz A, Reilmann H. Biceps tendon: diagnosis, therapy and results after proximal and distal rupture [in German]. Orthopade 2000; 29: Rantanen J, Orava S. Rupture of the distal biceps tendon: a report of 19 patients treated with anatomic reinsertion, and a meta-analysis of 147 cases found in the literature. Am J Sports Med 1999; 27: Belli P, Costantini M, Mirk P, Leone A, Pastore G, Marano P. Sonographic diagnosis of distal biceps tendon rupture: a prospective study of 25 cases. J Ultrasound Med 2001; 20: Falchook FS, Zlatkin MB, Erbacher GE, Moulton JS, Bisset GS, Murphy BJ. Rupture of the distal biceps tendon: evaluation with MR imaging. Radiology 1994; 190: Fitzgerald SW, Curry DR, Erickson SJ, Quinn SF, Friedman H. Distal biceps tendon injury: MR imaging diagnosis. Radiology 1994; 191: Miller TT, Adler RS. Sonography of tears of the distal biceps tendon. AJR 2000; 175: Kalume Brigido M, De Maeseneer M, Jacobson JA, Jamadar DA, Morag Y, Marcelis S. Improved visualization of the radial insertion of the biceps tendon at ultrasound with a lateral approach. Eur Radiol 2009; 19: Smith J, Finnoff JT, O Driscoll SW, Lai JK. Sonographic evaluation of the distal biceps tendon using a medial approach: the pronator window. J Ultrasound Med 2010; 29: Hartgerink P, Fessell DP, Jacobson JA, van Holsbeeck MT. Sonographic accuracy and characterization of full thickness tears of the Achilles versus tendinosis/partial tears in 26 cases with surgical correlation. Radiology 2001; 220: Seiler JG 3rd, Parker LM, Chamberland PD, Sherbourne GM, Carpenter WA. The distal biceps tendon: two potential mechanisms involved in its rupture arterial supply and mechanical impingement. J Shoulder Elbow Surg 1995; 4: Morrey BF, Askew LJ, An KN, Dobyns JH. Rupture of the distal tendon of the biceps brachii: a biomechanical study. J Bone Joint Surg Am 1985; 67: AJR:200, January 2013