Ultrasonographic Assessment of the Flexor Pollicis Longus Tendon After Plate Fixation Chieko Kadoma, MD; Masatoshi Takahara, MD, PhD; Masahiro Maruyama, MD; Hiroshi Satake, MD, PhD; Michiaki Takagi, MD, PhD abstract Rupture of the flexor pollicis longus tendon is a major complication after volar locking plate fixation of distal radius fracture. This study used ultrasonography to assess the flexor pollicis longus tendon and intermediate tissue. The study assessed 27 patients (28 wrists) who underwent removal of the volar locking plate. Before plate removal, radiography and ultrasonography were performed to assess the relation and the volar locking plate. Intraoperatively, the authors evaluated the intermediate tissues and the distal volar margin of the plate. Preoperative and intraoperative findings were compared. Intraoperative findings were strongly related to the distance between the flexor pollicis longus tendon and the volar locking plate on ultrasonography. The sensitivity of ultrasonography in detecting thin, membrane-like intermediate tissue through which the plate was visible was 95%, and the specificity was 89% if the distance and the plate was less than 0.7 mm. Compression of the flexor pollicis longus tendon was seen in 11 cases (39.3%), and this finding suggested the presence of thin, membrane-like intermediate tissue. The study results showed that ultrasonography could be used to identify the type of intermediate tissue between the flexor pollicis longus tendon and the volar locking plate. [Orthopedics. 2017; 40(1):e104-e108.] During the past decade, volar plating of distal radius fractures has become an increasingly popular surgical technique. 1-3 However, rupture of the flexor pollicis longus tendon has been reported after plate fixation. The reported prevalence of flexor pollicis longus tendon rupture ranged from 0.17% to 12%. 4-6 Adham et al 7 suggested that some ruptures of the flexor pollicis longus tendon could be attributed to the design of the plate or screw, long-term use of steroids, collapse of the fracture, suboptimal placement of plates, or incorrect use of plates. On the other hand, these authors suggested that proper placement of the volar plate, with the distal edge in the concavity of the distal ends of the radius, is necessary to prevent complications. 7 This placement eliminates the possibility of contact between the plate and the flexor pollicis longus tendon. 7 Repair of the pronator quadratus muscle provides soft tissue coverage between the plate and the flexor tendon. 7 The average distance between the distal edge of the normal radius and the flexor pollicis longus tendon was reported as 1.94 mm (range, 1.2-2.6 mm), and contact The authors are from the Center for Hand, Elbow and Sports Medicine (CK, MTakahara), Izumi Orthopaedic Hospital, and the Department of Orthopaedic Surgery (MM, HS, MTakagi), Yamagata University Faculty of Medicine, Sendai, Japan. The authors have no relevant financial relationships to disclose. Correspondence should be addressed to: Masatoshi Takahara, MD, PhD, Center for Hand, Elbow and Sports Medicine, Izumi Orthopaedic Hospital, Maruyama 6-1, Kamiyagari, Izumi-ku, Sendai, 981-3121, Japan (tehiji@izumiseikei. com). Received: March 7, 2016; Accepted: August 23, 2016. doi: 10.3928/01477447-20161017-01 e104
between the plate and the flexor pollicis longus tendon is difficult to avoid if the plate is placed more distally. 8,9 Although implant removal has been performed to prevent rupture of the flexor pollicis longus tendon, it is not always necessary. If the risk of tendon rupture could be predicted, it would be possible to select cases for plate removal. The current authors hypothesized that they could estimate the degree of risk of flexor pollicis longus tendon rupture if ultrasonography correctly showed the relation and the plate. The authors assessed the distance between the plate and the flexor pollicis longus tendon with ultrasonography and compared the distance with intraoperative findings. The goal of this study was to assess the flexor pollicis longus tendon and intermediate tissue with ultrasonography. Materials and Methods Patients and families were informed that data from the case would be submitted for publication and gave their consent. The study was approved by the ethics committee of the study hospital. During the past 2 years, 79 patients (81 wrists) underwent volar locking plate (LC-DRP VA Plate; Synthes GmbH, Solothurn, Switzerland) fixation for distal radius fracture. All patients had follow-up with radiography and ultrasonography. Among 79 patients (81 wrists), 27 patients (28 wrists) underwent removal of the plate because of symptoms or displacement of the plate. In other cases, the plate was removed to avoid future complications, including rupture of the flexor pollicis longus tendon. Mean age of the study participants was 52 years (range, 16-86 years). The right wrist was affected in 15 cases, and the left wrist was affected in 13 cases. Implant removal was performed an average of 210 days (range, 70-389 days) after fixation. The prominence of the plate from the radius was measured on a lateral radiograph. A critical line (plate) was drawn A tangential to the most volar extent of the volar rim of the plate, parallel to the volar cortical bone of the radial shaft. 10,11 In the same way, a critical line (radius) was drawn tangential to the most volar extent of the volar rim of the radius, and the distance between these lines was measured. The authors defined this distance as plate prominence (radiograph) and expressed it as positive if the plate protruded from the radius (Figure 1A). Patients underwent radiographic evaluation an average of 49 days (range, 22-183 days) after volar plate fixation. Ultrasonography (HI VISION Avius; Hitachi, Tokyo, Japan) and a linear probe with a frequency of 14 to 16 MHz were used. All examinations were performed by an author (M. Takahara) who was familiar with ultrasonography. Longitudinal images of the flexor pollicis longus tendon and the distal volar margin of the plate were obtained with the wrist in neutral position and maximum extension (Figures 1B-C). The same author also examined whether the dorsal surface of the flexor pollicis longus tendon was compressed volarly at the distal volar margin of the plate. The minimal distance between the flexor pollicis longus tendon and the plate was measured. This distance was defined as tendon-plate B C Figure 1: Preoperative findings. Lateral radiograph of the wrist. Quantification (< >) of the plate prominence from the volar prominent edge of the distal radius (millimeters) (A). Ultrasonographic longitudinal evaluation of the flexor pollicis longus tendon with the wrist in neutral position (B). Ultrasonographic longitudinal evaluation of the flexor pollicis longus tendon with the wrist in maximum extension (C). Longitudinal ultrasonography. The flexor pollicis longus tendon-plate distance is measured as the minimum distance and the plate. The flexor pollicis longus tendon-plate distance was 0.9 mm. Symbols: asterisks, flexor pollicis longus tendon; black arrows, plate surface; arrowheads, distal edge of the radius; yellow arrow with lines, tendon-plate distance (D). distance (ultrasound) (Figure 1D). Ultrasonography was performed an average of 184 days (range, 62-404 days) after volar plate fixation. During plate removal, patients were evaluated to identify injury of the flexor pollicis longus tendon. The same author who performed the ultrasound examinations (M. Takahara) grossly assessed the intermediate tissues between the flexor pollicis longus tendon and the distal volar margin of the plate. The tissue was divided into 2 groups: (1) thick tissue that included muscle or fibrous tissue that covered the plate and (2) thin, membrane-like tissue through which the plate was visible (Figure 2). Before removal of the plate, no patient had pain, crepitus, or functional loss during thumb motion. Changes in symptoms after removal of the implant and their relation to intraoperative findings were examined. The correlation between plate prominence (radiograph) and tendon-plate distance (ultrasound) was statistically examined with Spearman s rank correlation coefficient. The relation between intraoperative findings and plate prominence (radiograph) and the relation between intraoperative findings and tendon-plate distance (ultrasound) were analyzed with the Mann-Whitney U test D JANUARY/FEBRUARY 2017 Volume 40 Number 1 e105
Figure 2: Intraoperative findings. The intermediate tissue and the plate (arrow) was thin, membrane-like tissue through which the plate was visible (A). The plate was not visible (arrow) through dense, fibrous tissue (B). In a 78-year-old woman, the flexor pollicis longus tendon was compressed. The tendon-plate distance was 0 mm. Symbols: asterisks, flexor pollicis longus tendon; arrow, plate surface (C). In the same patient as Figure 2C, thin, membrane-like tissue (arrow) is seen (D). Figure 3: Relation between plate prominence and tendon-plate distance. Plate prominence was significantly related to tendon-plate distance (r=-0.535, P=.0055). Abbreviation: US, ultrasound. and Fisher s exact test. P<.05 was considered statistically significant. Results Mean plate prominence (radiograph) was 1.6 mm (range, -0.8 to 4.3 mm). Ultrasonography showed localized deformity of the flexor pollicis longus tendon in 2 cases (7.1%) and compression of the flexor pollicis longus tendon at the distal volar margin of the plate in 11 cases (39.3%). Mean tendon-plate distance (ultrasound) was 0.7 mm (range, 0-2.6 mm) Figure 4: Relation between plate prominence and intraoperative findings. No significant correlation was found (P=.1149). in neutral position and 0.5 mm (range, 0-2.6 mm) in maximum wrist extension. The latter distance tended to be shorter (P=.38), although the difference did not reach statistical significance. The shorter distance, representing the minimum tendon-plate distance, was a mean of 0.5 mm (range, 0-2.6 mm), and this value was used for statistical analysis. Age did not statistically correlate with the distance between the flexor pollicis longus tendon and the volar margin of the plate (r=-0.117, P=.5767). Intraoperative findings showed that the intermediate tissue was the pronator quadratus muscle in 5 cases (18%); dense, fibrous tissue in 4 cases (14%); and thin, membrane-like tissue through which the plate was visible in 19 cases (68%). In 1 case, the flexor pollicis longus tendon had a partial rupture of 8 mm, representing 10% of its width on the dorsal side of the tendon. This patient had no preoperative symptoms involving the flexor pollicis longus tendon. Plate prominence (radiograph) was 1.8 mm, tendon-plate distance (ultrasound) was 0.4 mm, and thin, membrane-like intermediate tissue was found. Changes in symptoms after implant removal were reviewed in the medical charts. The relation between changes in symptoms and intraoperative findings was examined. One patient had relief of pain during active flexion of the index finger, and another had relief of wrist dullness. The intermediate tissue was dense, fibrous tissue in the former case and pronator quadratus muscle in the latter case, and the flexor pollicis longus tendons were intact in both patients. A negative relation was found between plate prominence (radiograph) and tendon-plate distance (ultrasound) (r=-0.535, P=.0055). The more the plate protruded from the radius on radiography, the closer to the plate the flexor pollicis longus tendon appeared on ultrasonography (Figure 3). No significant relation was found between plate prominence (radiograph) and the intermediate tissue between the flexor pollicis longus tendon and the plate (P=.1149). However, the 2 patients who showed no plate prominence (radiograph) had dense, fibrous tissue over the plate (Figure 4). A strong relation was found between tendon-plate distance (ultrasound) and the intermediate tissue (intraoperative) (P=.0001). When the tendon-plate distance (ultrasound) was shorter, the intermediate tissue (intraoperative) was thinner (Figure 5). Thin, membrane-like intermediate tissue was found in 10 of the e106
11 cases (91%) with compression of the flexor pollicis longus tendon shown on ultrasonography (P=.0491). Discussion Soong et al 4 focused on plate implant prominence on postoperative lateral radiographs. They found no ruptures in a group of patients treated with the DVR plate (DePuy Orthopaedics, Inc, Warsaw, Indiana), perhaps as a result of the lower profile of the plate. These authors concluded that prominence of the plate at the watershed line of the distal part of the radius may increase the risk of tendon injury. 11 Brown and Lifchez 12 reported a 75-yearold woman who had rupture of the flexor pollicis longus tendon, even though the pronator quadratus was returned to its native position after volar locking plate fixation. According to Douthit, 13 the overall bulk of the pronator quadratus is variable, and the muscle can be relatively thin in middleaged women. Brown and Lifchez 12 speculated that pronator quadratus plate coverage may not protect tendons adequately. Some reports have described the use of ultrasonography to assess the flexor pollicis longus tendon after volar locking plate fixation in cases of distal radius fracture. In 1 study, ultrasonography showed that the distal part of the plate and the flexor pollicis longus tendon were apposed in 19 of 20 cases. 8 In another study, ultrasonography showed that the tendon and the plate were apposed in 62.5% of cases treated with fixation of a distally placed plate, whereas the tendon was never apposed with a proximally placed plate. 12 The current study assessed the distance and the plate with ultrasonography preoperatively and assessed the tendon-plate intermediate tissue intraoperatively. The tendon-plate distance (ultrasound) had a strong relation to the intermediate tissue and a negative relation to plate prominence (radiograph). These findings suggested that ultrasonographic evaluation Figure 5: Relation between tendon-plate distance and intraoperative findings. A significant relation was seen (P=.0001). Abbreviation: US, ultrasound. of the flexor pollicis longus tendon and the plate was more useful than radiography in identifying the intermediate tissue and the volar locking plate. No previous data were found on the relation between preoperative assessment and intraoperative findings on rupture of the flexor pollicis longus tendon. Thin, membrane-like tissue between the flexor pollicis longus tendon and the plate was histologically composed of synovial tissue. If the flexor pollicis longus tendon-plate distance (ultrasound) was less than 0.7 mm, ultrasonography showed sensitivity of 95%, specificity of 89%, accuracy of 93%, positive predictive value of 95%, and negative predictive value of 89% in detecting the intermediate tissue of thin synovial membrane through which the plate was visible (P<.0001). Although both sensitivity and specificity were high, 1 false-negative finding occurred (Table). To reach 100% sensitivity without any false-negative findings, the flexor pollicis longus tendon-plate distance (ultrasound) had to be lower than 0.9 mm, and its specificity was 44%, accuracy was 82%, positive predictive value was 79%, and negative predictive value was 100% (P=.0062). Of 10 patients who had focal compression of the flexor pollicis longus tendon on ultrasonography, 9 (90%) had thin, membrane-like intermediate tissue Table Relation Between Tendon-Plate Distance and Intraoperative Intermediate Tissue a Tendon- Plate Distance, mm Dense, Fibrous Tissue or Muscle No. Thin, Membranelike Tissue <0.7 1 18 0.7 8 1 a Cutoff tendon-plate distance on ultrasound was 0.7 mm. Sensitivity was 95%, specificity was 89%, accuracy was 93%, positive predictive value was 95%, and negative predictive value was 89% (P<.0001). Figure 6: Ultrasound (US) assessment of intermediate tissue between the flexor pollicis longus (FPL) tendon and the plate. If the distance is less than 0.7 mm or if the FPL tendon is compressed by the plate, the intermediate tissue is assessed as thin, membrane-like tissue. through which the plate was visible. The tendon-plate distance (ultrasound) can be used to identify whether the patient has thin, membrane-like intermediate tissue or dense, fibrous scar. The current authors are often asked about future problems with volar locking plates when patients have achieved bone union and almost normal wrist function. When patients are told that there is a potential complication of flexor pollicis longus tendon rupture associated with JANUARY/FEBRUARY 2017 Volume 40 Number 1 e107
the plate if the flexor pollicis longus tendon is attenuated, most patients ask how they can avoid this risk. In this case, the flexor pollicis longus tendon and the intermediate tissue are assessed with lateral radiography and ultrasonography. If the tendon-plate distance is less than 0.7 mm or if the flexor pollicis longus tendon is compressed by the plate, the intermediate tissue is assessed as a synovial membrane or none (Figure 6). The tendon may have a risk of attenuation or even rupture if the floor is a plate or only a thin membrane. Plate removal is recommended for patients who are found to have only thin, membrane-like tissue. Conclusion This study assessed the flexor pollicis longus tendon and the intermediate tissue. However, the practical incidence of tendon attenuation in these cases is unknown. Follow-up is necessary with patients who have not undergone plate removal even though ultrasonography showed thinner intermediate tissue. Intermediate tissue can change over time. Serial ultrasonography should be performed to re-evaluate the intermediate tissue. Repeated ultrasonographic evaluations were not performed in this study; therefore, no data are available on intra- or interexaminer differences. This preliminary report compared ultrasonographic findings with the intermediate tissue between the flexor pollicis longus tendon and the plate. The results showed that ultrasonography could be used to identify the intermediate tissue, suggesting that ultrasonography has the potential to predict the risk of attenuation of the flexor pollicis longus tendon. Further studies with larger samples are necessary to investigate intraor interexaminer differences, changes over time, and diagnostic values. References 1. Orbay JL. The treatment of unstable distal radius fractures with volar fixation. Hand Surg. 2000; 5(2):103-112. 2. Keating JF, Court-Brown CM, McQueen MM. Internal fixation of volar-displaced distal radial fractures. J Bone Joint Surg Br. 1994; 76(3):401-405. 3. Diaz-Garcia RJ, Oda T, Shauver M, Chung KC. A systemic review of outcomes and complications of treating unstable distal radius fractures in the elderly. J Hand Surg Am. 2011; 36(5):824-835. 4. Soong M, van Leerdam R, Guitton TG, Got C, Katarincic J, Ring D. Fracture of the distal radius: risk factors for complications after locked volar plate fixation. J Hand Surg Am. 2011; 36(1):3-9. 5. Minegishi H, Dohi O, An S, Sato H. Treatment of unstable distal radius fractures with the volar locking plate. Ups J Med Sci. 2011; 116(4):280-284. 6. Drobetz H, Kutscha-Lissberg E. Osteosynthesis of distal radial fractures with a volar locking screw plate system. Int Orthop. 2003; 27(1):1-6. 7. Adham MN, Porembski M, Adham C. Flexor tendon problems after volar plate fixation of distal radius fractures. Hand (N Y). 2009; 4(4):406-409. 8. Dohi D. Evaluation of flexor pollicis longus tendon rupture after volar locking plate fixation of distal radius fractures by ultrasonography [in Japanese]. J Jpn Soc Surg Hand. 2010; 27(2):70-73. 9. Kameyama M, Takeda K, Komiyama T, et al. Ultrasonographic study of flexor pollicis longus tendon after volar locking plate fixation in the treatment of fractures of the distal radius [in Japanese]. J Jpn Soc Fracture Repair. 2012; 34(1):25-27. 10. White BD, Nydick JA, Karsky D, Williams BD, Hess AV, Stone JD. Incidence and clinical outcomes of tendon rupture following distal radius fracture. J Hand Surg Am. 2012; 37(10):2035-2040. 11. Soong M, Earp BE, Bishop G, Leung A, Blazar P. Volar locking plate implant prominence and flexor tendon rupture. J Bone Joint Surg Am. 2011; 93(4):328-335. 12. Brown EN, Lifchez SD. Flexor pollicis longus tendon rupture after volar plating of a distal radius fracture: pronator quadratus plate coverage may not adequately protect tendons. eplasty. 2011; 11:e43. 13. Douthit JD. Volar plating of dorsally comminuted fractures of the distal radius: a 6-year study. Am J Orthop (Belle Mead NJ). 2005; 34(3):140-147. e108