- Email: [email protected]
Factors affecting extension lag after tendon reconstruction for finger extensor tendon rupture due to distal radioulnar lesion
Journal of Orthopaedic Science 21 (2016) 19e23
Contents lists available at ScienceDirect
Journal of Orthopaedic Science journal homepage: http://www.elsevier.com/locate/jos
Original article
Factors affecting extension lag after tendon reconstruction for finger extensor tendon rupture due to distal radioulnar lesion Toshiro Itsubo a, *, Shigeharu Uchiyama a, Hiroshi Yamazaki c, Masanori Hayashi a, Koichi Nakamura b, Kazuki Kuniyoshi d, Hiroyuki Kato a a
Department of Orthopaedic Surgery, Shinshu University School of Medicine, Japan Department of Orthopaedics, Azumi General Hospital, Japan Department of Orthopaedics, Aizawa Hospital, Japan d Department of Orthopaedic Surgery, Chiba University School of Medicine, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 November 2014 Received in revised form 4 October 2015 Accepted 10 October 2015 Available online 18 November 2015
Background: Although extensor tendon rupture associated with distal radioulnar joint disorder is often encountered, its treatment has not yet been established. We report the postoperative results for reconstruction of finger extensor tendon rupture due to distal radioulnar lesion and analyse the factors affecting postoperative extension lag. Methods: We examined 74 index, middle, ring, or little fingers with extensor tendon rupture of 34 hands. Primary diseases were rheumatoid arthritis in 24 hands and osteoarthritis in 10. Reconstruction methods included tendon graft in 45 fingers, extensor indicis proprius tendon transfer in 15, and end-to-side adjacent tendon suture in 14. At final postoperative follow-up ranging from 12 to 40 (mean: 18) months, we measured metacarpophalangeal (MCP) joint range of motion and extension lag and statistically evaluated the relationship between postoperative extension lag and several clinical factors. Results: We encountered no cases of re-rupture or worsening of finger flexion range after reconstruction. Mean postoperative active flexion of the MCP joint was 78.1 (range: 45e95) degrees. Mean postoperative extension lag was 10.3 (range: 0e50) degrees. We observed that postoperative extension lag was significantly larger in fingers associated with extensor tendon rupture in two or three additional fingers in the affected hand or in fingers of patients aged 80 years or over. The interval from rupture to reconstruction, reconstruction method, or arthritis type did not remarkably affect outcome. Conclusions: This study uncovered that surgical intervention for extensor tendon rupture should be performed before three fingers become affected. © 2015 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
1. Introduction In aging societies, the incidence of extensor tendon rupture associated with distal radioulnar joint (DRUJ) lesion in rheumatoid arthritis (RA) and osteoarthritis (OA) is increasing [1e8]. Tendon graft using the palmaris longus (PL) tendon, tendon transfer using the extensor indicis proprius (EIP) tendon, and end-to-side suture to an adjacent finger extensor tendon [9] have all been established as treatment methods for ruptured extensor tendon. However, few reports have analysed larger cohorts after reconstruction for such
lesions, nor have the long-term outcomes of tendon reconstruction or the effects of patient age, arthritis type, interval from rupture to reconstruction, number of affected fingers in the affected hand, reconstruction method, number of fingers reconstructed with one donor muscle, or postoperative rehabilitation on postoperative outcome been described in detail. We present the results of three methods of tendon reconstruction for fingers with extensor tendon rupture due to RA or OA and investigate the impact of clinical factors on treatment outcome. 2. Materials and methods
* Corresponding author. Department of Orthopaedic Surgery, Shinshu University School of Medicine, 3-1-1 Asahi, Matsumoto, Nagano 390-8621, Japan. Tel.: þ81 26 337 2659; fax: þ81 26 335 8844. E-mail address: [email protected] (T. Itsubo).
The present study included 74 index, middle, ring, or little fingers with extensor digitorum communis (EDC) tendon rupture of 34 hands of 34 patients who consecutively underwent tendon
http://dx.doi.org/10.1016/j.jos.2015.10.013 0949-2658/© 2015 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
20
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
reconstruction between 2003 and 2007. Mean patient age at reconstruction was 62.6 (range: 21e92) years. When divided by age, there were three patients aged 20e29 years with six affected fingers, one patient aged 30e39 years with three affected fingers, four patients aged 40e49 years with 11 affected fingers, two patients aged 50e59 years with four affected fingers, nine patients aged 60e69 years with 15 affected fingers, nine patients aged 70e79 years with 19 affected fingers, and six patients aged 80 years or more with 16 affected fingers. Eleven patients were male and 23 patients were female. Twenty-seven patients had affected fingers on their right hand and seven patients were affected on their left hand. Primary disease was RA in 24 subjects and OA in 10. Cases exhibiting deformity (i.e., Larsen grade S3) at the MCP joint or mutilans (i.e., Larsen grade 5) at the carpometacarpal or radiocarpal joint that influenced extension function were excluded from this study. Fingers with ulnar subluxation of the EDC tendons were excluded. The mean interval from awareness of the latest tendon rupture to tendon reconstruction was 5.9 months (range: 2 weeks to 24 months). The extensor tendon ruptures were as follows: isolated little finger, one finger of one hand; isolated ring finger, two fingers of two hands; both ring and little fingers, 46 fingers of 23 hands; both middle and ring fingers, two fingers of one hand; and three or four fingers, 23 fingers of seven hands. The 74 fingers were divided on the basis of reconstruction procedure, as follows: interposition tendon graft, 45 fingers; tendon transfer using the EIP tendon, 15 fingers; and adjacent tendon suture to the radial-sided intact EDC tendon, 14 fingers. Early controlled postoperative rehabilitation using dynamic splinting was performed for all fingers [2,10]. During postoperative periods ranging from 12 to 40 (mean: 18) months, we measured active range of motion (ROM) and postoperative extension lag of the MCP joint of the reconstructed finger at 0 degrees of wrist extension using a goniometer. Postoperative active ROM of the affected MCP joints was evaluated using a combination of mean flexion arc and mean extension lag in each joint using the modified Geldmacher's criterion [7,11]. We also calculated correlation coefficients for the degree of extension lag and the following clinical factors: 1) patient age at extensor tendon reconstruction, 2) arthritis type, 3) interval from latest extensor tendon rupture to reconstruction, 4) number of affected fingers in the affected hand, 5) tendon reconstruction procedure, i.e., tendon graft, tendon transfer, or adjacent tendon suture, and 6) wrist reconeKapandji procedure [12,13], Darstruction procedure, i.e., Sauve rach procedure [13,14], or resection of the bony spur at the DRUJ. We performed multiple regression analysis using multivariate analysis with extension lag as the responsive variable and the six above-mentioned factors as the explanatory variables to determine standard partial regression coefficients. Next, we assessed postoperative extension lag on the basis of patient age group at extensor tendon reconstruction. We also divided the cohort into groups of extensor tendon rupture of two or fewer fingers (51 fingers) or three or four fingers (23 fingers) in one hand and compared postoperative extension lag. Considering previous reports of ruptures in multiple fingers being unsatisfactorily reconstructed [7,15,16], we additionally categorised the 74 fingers into three groups according to the ratio of reconstructed extensor tendons to the number of motor sources: 17 fingers in which the extensor tendon was reconstructed using one motor source was defined as the 1:1 reconstruction group, 48 fingers in which another extensor tendon was reconstructed using the same motor source was classified as the 1:2 reconstruction group, and nine fingers in which three extensor tendons were reconstructed using the same motor source was categorised as the 1:3 reconstruction group. We compared postoperative extension lag among these groups.
Statistical analysis was performed using the SPSS statistical package version 21 (SPSS, Inc., Chicago, IL, USA). A P value of <0.05 was considered to be statistically significant. Paired t-tests were performed on changes in ROM before and after reconstruction, and the ManneWhitney U test or KruskaleWallis test was adopted for comparisons of postoperative extension lag among groups. If a difference was found to be significant, the ManneWhitney U test with Bonferroni correction was performed. This study was approved by the Ethical Review Board of our hospital.
3. Surgical techniques and postoperative rehabilitation All extensor tendon reconstruction procedures were done under general anaesthesia with a pneumatic tourniquet applied at the upper arm. An oblique incision was made at the dorsum and extended to the dorsoulnar forearm. The extensor retinaculum was raised as a radial-based flap at the level of the radiocarpal joint and as an ulnar-based flap at the level of the distal radioulnar joint. The distal radioulnar lesion was exposed and reconstructed eKapandji procedure [12,13], Darrach procedure by the Sauve [13,14], or resection of the bony spur at the DRUJ. In general, we resected the bony spur when deformity by arthritis was slight. If arthritic deformity was extensive, we typically performed the eKapandji procedure or the Darrach procedure for deforSauve mity at the DRUJ in patients uninvolved in heavy manual labour. We subsequently reconstructed the ruptured EDC tendon by one of three methods, namely, interposition tendon graft, tendon transfer, or adjacent tendon suture. We used the ipsilateral or contralateral PL tendon for interposition tendon grafting. First, we sutured the PL tendon at the dorsal hand with an interweaving suture to one to three refreshed distal stumps of ruptured EDC tendons. At the distal forearm, we selected one EDC tendon as a motor source by manually pulling on each proximal tendon stump. The stump that showed the longest excursion was chosen as the motor source tendon. The EIP tendon was used as the motor source for tendon transfer. It was cut proximally to the MCP joint, withdrawn at the wrist, passed over the dorsal extensor retinaculum, and then connected by an interweaving suture to one or two refreshed distal stumps of ruptured EDC tendons. For adjacent tendon suture, the refreshed distal stump of the EDC tendon was attached to an intact radial-sided EDC tendon by an end-to-side interweave suture at the dorsal hand. In little and ring fingers with extensor tendon rupture, we routinely reconstructed both the ring (EDC-4) and little (EDC-5) finger tendons with a single motor source either by tendon graft or EIP tendon transfer. We encountered two little fingers with both EDC-4 and EDM tendon rupture and an absent EDC-5 tendon. In these fingers, we first pulled proximally on the distal stump of the EDC-4 and checked the extension angle of the MCP joint. In one little finger showing an MCP joint extension angle similar to that of the ring finger after proximal pulling, we performed an EIP tendon transfer, sutured the EIP tendon to the EDC-4 alone, and left the EDM tendon untouched. In the other little finger that demonstrated a rather weak MCP joint extension angle after pulling on the EDC-4, we performed a tendon graft of both the EDC-4 and EDM to one motor source. The retinaculum was finally repaired to place the extensor tendon beneath the radial-based retinaculum flap and over the ulnar-based retinaculum flap. Using these three reconstruction techniques, the number of reconstructed EDC tendons using a single intact extensor tendon as a motor source was as follows: one EDC tendon reconstructed in 17 fingers, two EDC tendons reconstructed in 48 fingers, and three EDC tendons reconstructed in nine fingers (Table 1).
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
21
Table 1 Reconstruction methods for ruptured extensor tendons.
Tendon graft (45 fingers)
Tendon transfer (15 fingers) Tendon suture to adjacent finger (14 fingers)
Number of reconstructed fingers and tendon source
Number of reconstruction cases
One finger was reconstructed with one EDC Two fingers were reconstructed with one EDC Three fingers were reconstructed with one EDC One finger was reconstructed with one EIP Two fingers were reconstructed with one EIP One tendon was sutured to one adjacent intact EDC Two tendons were sutured to one adjacent intact EDC
12 30 3 5 10 8 6
EDC: extensor digitorum communis tendon; EIP: extensor indicis proprius tendon.
Table 2 Reconstruction procedure for each finger extensor tendon rupture. Number of fingers with extensor tendon rupture in each hand
Index Middle Ring Little
One (3 hands)
Two (24 hands)
Three (5 hands)
Four (2 hands)
e e TG: 1, TS: 1 TG: 1
e TS: 1 TG: 9, TS: 11, TT: 4 TG: 17, TT: 6
e TG: 3, TT: 1, TS: 1 TG: 3, TT: 2 TG: 3, TT: 2
TG: TG: TG: TG:
2 2 2 2
TG: tendon graft, TT: tendon transfer, TS: tendon suture to adjacent finger.
One hand surgeon (senior author) performed all tendon reconstruction procedures. The reconstruction procedure for each extensor tendon finger is shown in Table 2. In all hands, the tendon suture method involved triple weaving, with tension adjusted for greater extension at the MCP joint than that of the intact adjacent finger with the wrist in a neutral position. After surgery, a static palmar splint was applied from the tip of all four fingers to the proximal forearm to maintain the wrist at 45 degrees of extension, the finger MCP joint at 30 degrees of flexion, and the proximal and distal interphalangeal joints at 0 . Three or 4 days after surgery, active flexion and passive extension at the MCP joint combined with active flexion and extension at the proximal and distal interphalangeal joints were begun with a dynamic rubber-band reverse Kleinert splint [4] on the affected hand twice a day along with a static splint between sessions. Twenty-one days after surgery, active extension and flexion ROM exercises without the dynamic splint were performed during the day with static splinting at night. Forty-two days after extensor reconstruction, the patients were allowed to use the hand freely in normal activities of daily living. 4. Results Thirty-two cases (94.1%) achieved good results and 2 (5.9%) showed fair results according to the modified Geldmacher's criterion. The mean extension lag after extensor tendon reconstruction was 10.3 , which was significantly improved compared with the preoperative lag of 45.3 (Table 3). The mean flexion arc of the MCP joint after extensor tendon reconstruction was 78.1 and slightly reduced compared with the preoperative arc of 80.8 (Table 3). There were no significant differences between RA and OA fingers in terms of postoperative extension lag (Table 4). There was neither postoperative worsening of the MCP joint nor recurrence of tendon rupture. Active flexion of the MCP joint ranged from 60 to 90 in all
fingers, which indicated no cases of severe postoperative extensor tendon adhesion. A significant relationship was observed between patient age at tendon reconstruction and extension lag (r ¼ 0.4, P ¼ 0.001) (Fig. 1). In particular, the extension lag in patients aged 80 years or older was significantly greater than that in any other patient age group (P < 0.001) (Fig. 2). The extension lag of fingers of hands with three or more extensor tendon ruptures was significantly greater than that of fingers of hands with two or fewer extensor tendon ruptures (P < 0.01) (Fig. 3). In the two little fingers with ruptured EDC-4 and EDM tendons and an absent EDC-5 tendon, postoperative MCP joint extension lags were 5.0 and 10.0 degrees, respectively, and similar to the mean extension lag of all fingers (10.3 degrees). There was no correlation between interval from latest extensor tendon rupture to reconstruction and postoperative extension lag (P ¼ 0.15), type of tendon reconstruction procedure and postoperative extension lag (P ¼ 0.32), or primary disease and postoperative extension lag (P ¼ 0.41). In multiple regression analysis, the number of ruptured fingers significantly affected extension lag (r ¼ 0.490, P < 0.0001), followed next by patient age (r ¼ 0.450, P ¼ 0.0006). There was a significant, but weak, correlation between interval from latest rupture to reconstruction and extension lag (r ¼ 0.308, P ¼ 0.0004). There were no significant relationships between tendon reconstruction technique (r ¼ 0.183, P ¼ 0.072), DRUJ reconstruction technique (r ¼ 0.364, P ¼ 0.1826), or primary disease (r ¼ 0.068, P ¼ 0.5923) and extension lag (Table 5). There was also no significant difference in postoperative extension lag among the 1:1, 1:2, and 1:3 ratio reconstruction groups (P ¼ 0.06). 5. Discussion This study presents the postoperative results for reconstruction of finger extensor tendon rupture due to distal radioulnar lesion
Table 3 Changes in active range of motion.
Range (mean) of active flexion of the MCP joint Range (mean) of extension lag of the MCP joint
Before reconstruction (degrees)
After reconstruction (degrees)
P
70e90 (80.8) 30e80 (45.3)
60e90 (78.1) 0e50 (10.3)
0.62 0.004
There was no significant difference between the preoperative and postoperative degree of active flexion of the MCP joint. The angle of extension lag was significantly improved after surgery. MCP: metacarpophalangeal.
22
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
Table 4 Comparison of postoperative results between RA and OA fingers.
Range (mean) of active flexion of the MCP joint after reconstruction Range (mean) of extension lag of the MCP joint after reconstruction
RA (n ¼ 51)
OA (n ¼ 23)
P
60e90 (79.4) 0e45 (9.7)
60e90 (76.3) 0e50 (11.2)
0.62 0.38
There was no significant difference in postoperative results between RA and OA after tendon reconstruction (unpaired t-test). OA: osteoarthritis, RA: rheumatoid arthritis.
Fig. 1. A significant correlation is seen between patient age at extensor tendon reconstruction and postoperative extension lag (P ¼ 0.001; correlation coefficient ¼ 0.4).
and analyses the factors affecting postoperative extension lag. We observed that extension lag was significantly larger in fingers associated with extensor tendon rupture in two or three additional fingers in the affected hand or in fingers of patients aged 80 years or over. There have been numerous reports of extensor tendon rupture due to DRUJ lesions resulting from RA or OA [1e8]. At present, early mobilisation using a dynamic splint after extensor tendon reconstruction is advised and favourable outcomes have been described using this technique [2,10]. In the present study, extension arc was
Fig. 2. Comparison of postoperative extension lag by age group. The extension lag of patients aged 80 years or older is significantly larger than that of patients in younger groups in multiple-comparison post-hoc tests (ManneWhitney U test with Bonferroni correction, *P < 0.001). The number in each box represents the number of ruptured tendons.
Fig. 3. Comparison of postoperative extension lag between extensor tendon rupture of one or two fingers and of three or four fingers. Postoperative extension lag in cases of three or four affected fingers is significantly larger (*P < 0.01). The number in each box represents the number of ruptured tendons.
increased and flexion arc was maintained by various extensor tendon reconstruction procedures along with early mobilization. We noticed a mean postoperative extension lag of approximately 10 , which was similar to that in previous publications [3e5,10]. The reasons for extension lag after extensor tendon reconstruction are thought to be looseness at the suturing site, adhesion of the suturing site around the wound, elongation of the grafted tendon, lack of power in the motor source, and myostatic contracture of the motor source. However, since few reports have investigated the mechanisms of postoperative extension lag [15,16], we aimed to retrospectively elucidate the factors affecting outcome after reconstruction of extensor tendon rupture. In our cohort, the number of fingers affected by extensor tendon rupture was most strongly related to postoperative extension lag. This finding is consistent with previous reports [7,15]. We had suspected that the ratio of reconstructed extensor tendons to motor source was responsible for increased lag, but this was not the case in the present study. In hands with extensor tendon rupture from DRUJ lesions in RA or OA, the extensor digiti minimi (EDM) usually ruptures first, followed by successive ulnar-sided EDC tendons [5]. Therefore, it appears essential that surgical intervention be performed when only the little and/or ring fingers show signs of extensor tendon rupture. The interval from the latest episode of extensor tendon rupture to reconstruction, type of reconstruction, and arthritis type were not strongly related to postoperative extension lag. In contrast, we found a correlation between patient age and extension lag, whereby poorer outcomes were observed in patients aged 80 years or older. Rapid progression of joint contracture after surgery [17] is thought to account for a less favourable outcome in older patients after extensor tendon reconstruction. However, we encountered no
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
23
Table 5 Multivariate statistics from multiple regression analysis with objective variable set as degree of extension lag. Clinical factor
Standardised partial regression coefficient
P value
T value
Standard error
Number of affected fingers Patient age Interval from latest rupture to reconstruction Tendon reconstruction technique Wrist (DRUJ) reconstruction technique Primary disease (RA or OA)
0.490 0.450 0.308 0.183 0.364 0.068
<0.001 <0.001 <0.001 0.0724 0.1826 0.5923
4.648 3.623 3.767 2.021 3.668 0.538
3.004 0.088 0.137 2.976 4.691 3.689
RA: rheumatoid arthritis; OA: osteoarthritis. The number of affected fingers most strongly influenced extension lag, followed next by patient age. There was a mildly significant correlation between interval from rupture to reconstruction and extension lag.
cases of decreased ROM of the joint after surgery, and radiographs uncovered no patients with advanced OA changes during followup. Thus, it is unlikely that extension lag occurred because of the progression of joint contracture or OA. Disuse muscle weakness occurs frequently in patients aged over 80 years [18], which might have been one of the reasons for a poor outcome in very elderly patients in our study group. There were no patients who complained of restriction of finger flexion at final follow-up; postoperative rehabilitation using dynamic splinting likely contributed to this favourable result. In contrast, mean postoperative extension lag remained at approximately 10 , indicating that extensor tendons may be reconstructed at a stronger tension. This study has several limitations. First, it was retrospective in nature and the period of postoperative evaluation varied. Second, patients with RA or OA were analysed using the same methods. Although different pathological variables exist between the two disorders, no statistical difference in clinical results was observed. In all cases of RA, disease activity was well controlled and bony or joint destruction did not progress after extensor tendon reconstruction. As we could not determine the precise duration of RA, we could not analyze for relationships between RA duration and postoperative extension lag. Third, there were no clear criteria for the surgeon to judge which tendon reconstruction procedure should be performed. Lastly, no patient-based evaluation was performed either before or after surgery. Thus, how the remaining extension lag affected activities of daily living and patient satisfaction remain unclear. The number of tendon ruptures and the age of the patient are not under the control of the surgeon. However, it appears that tendon graft, extensor indicis proprius tendon transfer, and end-toside adjacent tendon suture are possible during the operation based on what is available as a motor source. The findings revealed in this study may also enable more precise informed consent from patients and prompter intervention before the involvement of multiple fingers. Future studies are required to quantitatively determine the excursion of reconstructed tendons, reveal the relationship between tendon excursion and postoperative extension lag, and clarify the associations between postoperative extension lag and patient-evaluated outcome. Further improvements in reconstruction procedures are also needed for patients aged over 80 years and for hands with extensor tendon rupture affecting three or four fingers. Conflict of interest The authors declare that they have no conflict of interest.
Acknowledgement The authors would like to thank Mr. Trevor Ralph for his English editorial assistance. References [1] Carr AJ, Burge PD. Rupture of extensor tendons due to osteoarthritis of the distal radio-ulnar joint. J Hand Surg Br 1992;17:694e6. [2] Chow JA, Dovelle S, Thomes LJ, Ho PK, Saldana J. A comparison of results of extensor tendon repair followed by early controlled mobilisation versus static immobilization. J Hand Surg Br 1989;14:18e20. [3] Chu PJ, Lee HM, Hou YT, Hung ST, Chen JK, Shih JT. Extensor-tendons reconstruction using autogenous palmaris longus tendon grafting for rheumatoid arthritis patients. J Orthop Surg Res 2008;3:16. [4] Chung US, Kim JH, Seo WS, Lee KH. Tendon transfer or tendon graft for ruptured finger extensor tendons in rheumatoid hands. J Hand Surg Eur 2010;35:279e82. [5] Moore JR, Weiland AJ, Valdata L. Tendon ruptures in the rheumatoid hand: analysis of treatment and functional results in 60 patients. J Hand Surg Am 1987;12:9e14. [6] Ohshio I, Ogino T, Minami A, Kato H, Miyake A. Extensor tendon rupture due to osteoarthritis of the distal radio-ulnar joint. J Hand Surg Br 1991;16:450e3. [7] Sakuma Y, Ochi K, Iwamoto T, Saito A, Yano K, Naito Y, Yoshida S, Ikari K, Momohara S. Number of ruptured tendons and surgical delay as prognostic factors for the surgical repair of extensor tendon ruptures in the rheumatoid wrist. J Rheumatol 2014;41:265e9. [8] Yamazaki H, Uchiyama S, Hata Y, Murakami N, Kato H. Extensor tendon rupture associated with osteoarthritis of the distal radioulnar joint. J Hand Surg Eur 2008;33:469e74. [9] Wolfe Scott W, Hotchkiss Robert N, Pederson William C, Kozin Scott H. Green's operative hand surgery. In: Strauch Robert J, editor. Extensor tendon injury. 6th ed. Churchill Livingstone; 2011. p. 159e88. [10] Kuniyoshi K, Kato H, Minami A, Suenaga N, Iwasaki N. Early active mobilization with dynamic splintage after reconstruction for extensor tendon rupture in rheumatoid hands. J Jpn Soc Surg Hand 2003;20:593e7 [in Japanese]. [11] Geldmacher J, Plank M, Treuheit KD. Significance of the preoperative status in the evaluation of results of the reconstruction of extensor tendons. Handchir Mikrochir Plast Chir 1986;18:23e9 [in German]. [12] Kapandji IA. The Kapandji-Sauve operation: its techniques and indications in non rheumatoid disease. Ann Chir Main 1986;5:181e93. [13] Wolfe Scott W, Hotchkiss Robert N, Pederson William C, Kozin Scott H. Green's operative hand surgery. In: Stanley John, editor. Arthroplasty and arthodesis of the wrist. 6th ed. Churchill Livingstone; 2011. p. 429e741. [14] Darrach W. Forward dislocation at the inferior radio-ulnar joint, with fracture of the lower third of the shaft of the radius. Ann Surg 1912;56:801. [15] Ikegami H, Takayama S, Ishiguro T, Horiuchi Y, Ichikawa T. Extensor tendon rupture in the rheumatoid hands. J Jpn Soc Surg Hand 2002;19:336e41 [in Japanese]. [16] Matsuda S, Mukaiyama K, Kato H, Yamazaki H. Treatment for osteoarthritis of the distal radioulnar joint and combined rupture of finger extensor tendons. J Jpn Soc Surg Hand 2004;21:479e81 [in Japanese]. [17] Shephard RJ. Management of exercise in the elderly. Can J Appl Sports Sci 1984;9:109e20. [18] Rosenberg I. Summary comments: epidemiological and methodological problems in determining nutritional status of older persons. Am J Clin Nutr 1989;50:1231e3.
Contents lists available at ScienceDirect
Journal of Orthopaedic Science journal homepage: http://www.elsevier.com/locate/jos
Original article
Factors affecting extension lag after tendon reconstruction for finger extensor tendon rupture due to distal radioulnar lesion Toshiro Itsubo a, *, Shigeharu Uchiyama a, Hiroshi Yamazaki c, Masanori Hayashi a, Koichi Nakamura b, Kazuki Kuniyoshi d, Hiroyuki Kato a a
Department of Orthopaedic Surgery, Shinshu University School of Medicine, Japan Department of Orthopaedics, Azumi General Hospital, Japan Department of Orthopaedics, Aizawa Hospital, Japan d Department of Orthopaedic Surgery, Chiba University School of Medicine, Japan b c
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 November 2014 Received in revised form 4 October 2015 Accepted 10 October 2015 Available online 18 November 2015
Background: Although extensor tendon rupture associated with distal radioulnar joint disorder is often encountered, its treatment has not yet been established. We report the postoperative results for reconstruction of finger extensor tendon rupture due to distal radioulnar lesion and analyse the factors affecting postoperative extension lag. Methods: We examined 74 index, middle, ring, or little fingers with extensor tendon rupture of 34 hands. Primary diseases were rheumatoid arthritis in 24 hands and osteoarthritis in 10. Reconstruction methods included tendon graft in 45 fingers, extensor indicis proprius tendon transfer in 15, and end-to-side adjacent tendon suture in 14. At final postoperative follow-up ranging from 12 to 40 (mean: 18) months, we measured metacarpophalangeal (MCP) joint range of motion and extension lag and statistically evaluated the relationship between postoperative extension lag and several clinical factors. Results: We encountered no cases of re-rupture or worsening of finger flexion range after reconstruction. Mean postoperative active flexion of the MCP joint was 78.1 (range: 45e95) degrees. Mean postoperative extension lag was 10.3 (range: 0e50) degrees. We observed that postoperative extension lag was significantly larger in fingers associated with extensor tendon rupture in two or three additional fingers in the affected hand or in fingers of patients aged 80 years or over. The interval from rupture to reconstruction, reconstruction method, or arthritis type did not remarkably affect outcome. Conclusions: This study uncovered that surgical intervention for extensor tendon rupture should be performed before three fingers become affected. © 2015 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
1. Introduction In aging societies, the incidence of extensor tendon rupture associated with distal radioulnar joint (DRUJ) lesion in rheumatoid arthritis (RA) and osteoarthritis (OA) is increasing [1e8]. Tendon graft using the palmaris longus (PL) tendon, tendon transfer using the extensor indicis proprius (EIP) tendon, and end-to-side suture to an adjacent finger extensor tendon [9] have all been established as treatment methods for ruptured extensor tendon. However, few reports have analysed larger cohorts after reconstruction for such
lesions, nor have the long-term outcomes of tendon reconstruction or the effects of patient age, arthritis type, interval from rupture to reconstruction, number of affected fingers in the affected hand, reconstruction method, number of fingers reconstructed with one donor muscle, or postoperative rehabilitation on postoperative outcome been described in detail. We present the results of three methods of tendon reconstruction for fingers with extensor tendon rupture due to RA or OA and investigate the impact of clinical factors on treatment outcome. 2. Materials and methods
* Corresponding author. Department of Orthopaedic Surgery, Shinshu University School of Medicine, 3-1-1 Asahi, Matsumoto, Nagano 390-8621, Japan. Tel.: þ81 26 337 2659; fax: þ81 26 335 8844. E-mail address: [email protected] (T. Itsubo).
The present study included 74 index, middle, ring, or little fingers with extensor digitorum communis (EDC) tendon rupture of 34 hands of 34 patients who consecutively underwent tendon
http://dx.doi.org/10.1016/j.jos.2015.10.013 0949-2658/© 2015 The Japanese Orthopaedic Association. Published by Elsevier B.V. All rights reserved.
20
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
reconstruction between 2003 and 2007. Mean patient age at reconstruction was 62.6 (range: 21e92) years. When divided by age, there were three patients aged 20e29 years with six affected fingers, one patient aged 30e39 years with three affected fingers, four patients aged 40e49 years with 11 affected fingers, two patients aged 50e59 years with four affected fingers, nine patients aged 60e69 years with 15 affected fingers, nine patients aged 70e79 years with 19 affected fingers, and six patients aged 80 years or more with 16 affected fingers. Eleven patients were male and 23 patients were female. Twenty-seven patients had affected fingers on their right hand and seven patients were affected on their left hand. Primary disease was RA in 24 subjects and OA in 10. Cases exhibiting deformity (i.e., Larsen grade S3) at the MCP joint or mutilans (i.e., Larsen grade 5) at the carpometacarpal or radiocarpal joint that influenced extension function were excluded from this study. Fingers with ulnar subluxation of the EDC tendons were excluded. The mean interval from awareness of the latest tendon rupture to tendon reconstruction was 5.9 months (range: 2 weeks to 24 months). The extensor tendon ruptures were as follows: isolated little finger, one finger of one hand; isolated ring finger, two fingers of two hands; both ring and little fingers, 46 fingers of 23 hands; both middle and ring fingers, two fingers of one hand; and three or four fingers, 23 fingers of seven hands. The 74 fingers were divided on the basis of reconstruction procedure, as follows: interposition tendon graft, 45 fingers; tendon transfer using the EIP tendon, 15 fingers; and adjacent tendon suture to the radial-sided intact EDC tendon, 14 fingers. Early controlled postoperative rehabilitation using dynamic splinting was performed for all fingers [2,10]. During postoperative periods ranging from 12 to 40 (mean: 18) months, we measured active range of motion (ROM) and postoperative extension lag of the MCP joint of the reconstructed finger at 0 degrees of wrist extension using a goniometer. Postoperative active ROM of the affected MCP joints was evaluated using a combination of mean flexion arc and mean extension lag in each joint using the modified Geldmacher's criterion [7,11]. We also calculated correlation coefficients for the degree of extension lag and the following clinical factors: 1) patient age at extensor tendon reconstruction, 2) arthritis type, 3) interval from latest extensor tendon rupture to reconstruction, 4) number of affected fingers in the affected hand, 5) tendon reconstruction procedure, i.e., tendon graft, tendon transfer, or adjacent tendon suture, and 6) wrist reconeKapandji procedure [12,13], Darstruction procedure, i.e., Sauve rach procedure [13,14], or resection of the bony spur at the DRUJ. We performed multiple regression analysis using multivariate analysis with extension lag as the responsive variable and the six above-mentioned factors as the explanatory variables to determine standard partial regression coefficients. Next, we assessed postoperative extension lag on the basis of patient age group at extensor tendon reconstruction. We also divided the cohort into groups of extensor tendon rupture of two or fewer fingers (51 fingers) or three or four fingers (23 fingers) in one hand and compared postoperative extension lag. Considering previous reports of ruptures in multiple fingers being unsatisfactorily reconstructed [7,15,16], we additionally categorised the 74 fingers into three groups according to the ratio of reconstructed extensor tendons to the number of motor sources: 17 fingers in which the extensor tendon was reconstructed using one motor source was defined as the 1:1 reconstruction group, 48 fingers in which another extensor tendon was reconstructed using the same motor source was classified as the 1:2 reconstruction group, and nine fingers in which three extensor tendons were reconstructed using the same motor source was categorised as the 1:3 reconstruction group. We compared postoperative extension lag among these groups.
Statistical analysis was performed using the SPSS statistical package version 21 (SPSS, Inc., Chicago, IL, USA). A P value of <0.05 was considered to be statistically significant. Paired t-tests were performed on changes in ROM before and after reconstruction, and the ManneWhitney U test or KruskaleWallis test was adopted for comparisons of postoperative extension lag among groups. If a difference was found to be significant, the ManneWhitney U test with Bonferroni correction was performed. This study was approved by the Ethical Review Board of our hospital.
3. Surgical techniques and postoperative rehabilitation All extensor tendon reconstruction procedures were done under general anaesthesia with a pneumatic tourniquet applied at the upper arm. An oblique incision was made at the dorsum and extended to the dorsoulnar forearm. The extensor retinaculum was raised as a radial-based flap at the level of the radiocarpal joint and as an ulnar-based flap at the level of the distal radioulnar joint. The distal radioulnar lesion was exposed and reconstructed eKapandji procedure [12,13], Darrach procedure by the Sauve [13,14], or resection of the bony spur at the DRUJ. In general, we resected the bony spur when deformity by arthritis was slight. If arthritic deformity was extensive, we typically performed the eKapandji procedure or the Darrach procedure for deforSauve mity at the DRUJ in patients uninvolved in heavy manual labour. We subsequently reconstructed the ruptured EDC tendon by one of three methods, namely, interposition tendon graft, tendon transfer, or adjacent tendon suture. We used the ipsilateral or contralateral PL tendon for interposition tendon grafting. First, we sutured the PL tendon at the dorsal hand with an interweaving suture to one to three refreshed distal stumps of ruptured EDC tendons. At the distal forearm, we selected one EDC tendon as a motor source by manually pulling on each proximal tendon stump. The stump that showed the longest excursion was chosen as the motor source tendon. The EIP tendon was used as the motor source for tendon transfer. It was cut proximally to the MCP joint, withdrawn at the wrist, passed over the dorsal extensor retinaculum, and then connected by an interweaving suture to one or two refreshed distal stumps of ruptured EDC tendons. For adjacent tendon suture, the refreshed distal stump of the EDC tendon was attached to an intact radial-sided EDC tendon by an end-to-side interweave suture at the dorsal hand. In little and ring fingers with extensor tendon rupture, we routinely reconstructed both the ring (EDC-4) and little (EDC-5) finger tendons with a single motor source either by tendon graft or EIP tendon transfer. We encountered two little fingers with both EDC-4 and EDM tendon rupture and an absent EDC-5 tendon. In these fingers, we first pulled proximally on the distal stump of the EDC-4 and checked the extension angle of the MCP joint. In one little finger showing an MCP joint extension angle similar to that of the ring finger after proximal pulling, we performed an EIP tendon transfer, sutured the EIP tendon to the EDC-4 alone, and left the EDM tendon untouched. In the other little finger that demonstrated a rather weak MCP joint extension angle after pulling on the EDC-4, we performed a tendon graft of both the EDC-4 and EDM to one motor source. The retinaculum was finally repaired to place the extensor tendon beneath the radial-based retinaculum flap and over the ulnar-based retinaculum flap. Using these three reconstruction techniques, the number of reconstructed EDC tendons using a single intact extensor tendon as a motor source was as follows: one EDC tendon reconstructed in 17 fingers, two EDC tendons reconstructed in 48 fingers, and three EDC tendons reconstructed in nine fingers (Table 1).
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
21
Table 1 Reconstruction methods for ruptured extensor tendons.
Tendon graft (45 fingers)
Tendon transfer (15 fingers) Tendon suture to adjacent finger (14 fingers)
Number of reconstructed fingers and tendon source
Number of reconstruction cases
One finger was reconstructed with one EDC Two fingers were reconstructed with one EDC Three fingers were reconstructed with one EDC One finger was reconstructed with one EIP Two fingers were reconstructed with one EIP One tendon was sutured to one adjacent intact EDC Two tendons were sutured to one adjacent intact EDC
12 30 3 5 10 8 6
EDC: extensor digitorum communis tendon; EIP: extensor indicis proprius tendon.
Table 2 Reconstruction procedure for each finger extensor tendon rupture. Number of fingers with extensor tendon rupture in each hand
Index Middle Ring Little
One (3 hands)
Two (24 hands)
Three (5 hands)
Four (2 hands)
e e TG: 1, TS: 1 TG: 1
e TS: 1 TG: 9, TS: 11, TT: 4 TG: 17, TT: 6
e TG: 3, TT: 1, TS: 1 TG: 3, TT: 2 TG: 3, TT: 2
TG: TG: TG: TG:
2 2 2 2
TG: tendon graft, TT: tendon transfer, TS: tendon suture to adjacent finger.
One hand surgeon (senior author) performed all tendon reconstruction procedures. The reconstruction procedure for each extensor tendon finger is shown in Table 2. In all hands, the tendon suture method involved triple weaving, with tension adjusted for greater extension at the MCP joint than that of the intact adjacent finger with the wrist in a neutral position. After surgery, a static palmar splint was applied from the tip of all four fingers to the proximal forearm to maintain the wrist at 45 degrees of extension, the finger MCP joint at 30 degrees of flexion, and the proximal and distal interphalangeal joints at 0 . Three or 4 days after surgery, active flexion and passive extension at the MCP joint combined with active flexion and extension at the proximal and distal interphalangeal joints were begun with a dynamic rubber-band reverse Kleinert splint [4] on the affected hand twice a day along with a static splint between sessions. Twenty-one days after surgery, active extension and flexion ROM exercises without the dynamic splint were performed during the day with static splinting at night. Forty-two days after extensor reconstruction, the patients were allowed to use the hand freely in normal activities of daily living. 4. Results Thirty-two cases (94.1%) achieved good results and 2 (5.9%) showed fair results according to the modified Geldmacher's criterion. The mean extension lag after extensor tendon reconstruction was 10.3 , which was significantly improved compared with the preoperative lag of 45.3 (Table 3). The mean flexion arc of the MCP joint after extensor tendon reconstruction was 78.1 and slightly reduced compared with the preoperative arc of 80.8 (Table 3). There were no significant differences between RA and OA fingers in terms of postoperative extension lag (Table 4). There was neither postoperative worsening of the MCP joint nor recurrence of tendon rupture. Active flexion of the MCP joint ranged from 60 to 90 in all
fingers, which indicated no cases of severe postoperative extensor tendon adhesion. A significant relationship was observed between patient age at tendon reconstruction and extension lag (r ¼ 0.4, P ¼ 0.001) (Fig. 1). In particular, the extension lag in patients aged 80 years or older was significantly greater than that in any other patient age group (P < 0.001) (Fig. 2). The extension lag of fingers of hands with three or more extensor tendon ruptures was significantly greater than that of fingers of hands with two or fewer extensor tendon ruptures (P < 0.01) (Fig. 3). In the two little fingers with ruptured EDC-4 and EDM tendons and an absent EDC-5 tendon, postoperative MCP joint extension lags were 5.0 and 10.0 degrees, respectively, and similar to the mean extension lag of all fingers (10.3 degrees). There was no correlation between interval from latest extensor tendon rupture to reconstruction and postoperative extension lag (P ¼ 0.15), type of tendon reconstruction procedure and postoperative extension lag (P ¼ 0.32), or primary disease and postoperative extension lag (P ¼ 0.41). In multiple regression analysis, the number of ruptured fingers significantly affected extension lag (r ¼ 0.490, P < 0.0001), followed next by patient age (r ¼ 0.450, P ¼ 0.0006). There was a significant, but weak, correlation between interval from latest rupture to reconstruction and extension lag (r ¼ 0.308, P ¼ 0.0004). There were no significant relationships between tendon reconstruction technique (r ¼ 0.183, P ¼ 0.072), DRUJ reconstruction technique (r ¼ 0.364, P ¼ 0.1826), or primary disease (r ¼ 0.068, P ¼ 0.5923) and extension lag (Table 5). There was also no significant difference in postoperative extension lag among the 1:1, 1:2, and 1:3 ratio reconstruction groups (P ¼ 0.06). 5. Discussion This study presents the postoperative results for reconstruction of finger extensor tendon rupture due to distal radioulnar lesion
Table 3 Changes in active range of motion.
Range (mean) of active flexion of the MCP joint Range (mean) of extension lag of the MCP joint
Before reconstruction (degrees)
After reconstruction (degrees)
P
70e90 (80.8) 30e80 (45.3)
60e90 (78.1) 0e50 (10.3)
0.62 0.004
There was no significant difference between the preoperative and postoperative degree of active flexion of the MCP joint. The angle of extension lag was significantly improved after surgery. MCP: metacarpophalangeal.
22
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
Table 4 Comparison of postoperative results between RA and OA fingers.
Range (mean) of active flexion of the MCP joint after reconstruction Range (mean) of extension lag of the MCP joint after reconstruction
RA (n ¼ 51)
OA (n ¼ 23)
P
60e90 (79.4) 0e45 (9.7)
60e90 (76.3) 0e50 (11.2)
0.62 0.38
There was no significant difference in postoperative results between RA and OA after tendon reconstruction (unpaired t-test). OA: osteoarthritis, RA: rheumatoid arthritis.
Fig. 1. A significant correlation is seen between patient age at extensor tendon reconstruction and postoperative extension lag (P ¼ 0.001; correlation coefficient ¼ 0.4).
and analyses the factors affecting postoperative extension lag. We observed that extension lag was significantly larger in fingers associated with extensor tendon rupture in two or three additional fingers in the affected hand or in fingers of patients aged 80 years or over. There have been numerous reports of extensor tendon rupture due to DRUJ lesions resulting from RA or OA [1e8]. At present, early mobilisation using a dynamic splint after extensor tendon reconstruction is advised and favourable outcomes have been described using this technique [2,10]. In the present study, extension arc was
Fig. 2. Comparison of postoperative extension lag by age group. The extension lag of patients aged 80 years or older is significantly larger than that of patients in younger groups in multiple-comparison post-hoc tests (ManneWhitney U test with Bonferroni correction, *P < 0.001). The number in each box represents the number of ruptured tendons.
Fig. 3. Comparison of postoperative extension lag between extensor tendon rupture of one or two fingers and of three or four fingers. Postoperative extension lag in cases of three or four affected fingers is significantly larger (*P < 0.01). The number in each box represents the number of ruptured tendons.
increased and flexion arc was maintained by various extensor tendon reconstruction procedures along with early mobilization. We noticed a mean postoperative extension lag of approximately 10 , which was similar to that in previous publications [3e5,10]. The reasons for extension lag after extensor tendon reconstruction are thought to be looseness at the suturing site, adhesion of the suturing site around the wound, elongation of the grafted tendon, lack of power in the motor source, and myostatic contracture of the motor source. However, since few reports have investigated the mechanisms of postoperative extension lag [15,16], we aimed to retrospectively elucidate the factors affecting outcome after reconstruction of extensor tendon rupture. In our cohort, the number of fingers affected by extensor tendon rupture was most strongly related to postoperative extension lag. This finding is consistent with previous reports [7,15]. We had suspected that the ratio of reconstructed extensor tendons to motor source was responsible for increased lag, but this was not the case in the present study. In hands with extensor tendon rupture from DRUJ lesions in RA or OA, the extensor digiti minimi (EDM) usually ruptures first, followed by successive ulnar-sided EDC tendons [5]. Therefore, it appears essential that surgical intervention be performed when only the little and/or ring fingers show signs of extensor tendon rupture. The interval from the latest episode of extensor tendon rupture to reconstruction, type of reconstruction, and arthritis type were not strongly related to postoperative extension lag. In contrast, we found a correlation between patient age and extension lag, whereby poorer outcomes were observed in patients aged 80 years or older. Rapid progression of joint contracture after surgery [17] is thought to account for a less favourable outcome in older patients after extensor tendon reconstruction. However, we encountered no
T. Itsubo et al. / Journal of Orthopaedic Science 21 (2016) 19e23
23
Table 5 Multivariate statistics from multiple regression analysis with objective variable set as degree of extension lag. Clinical factor
Standardised partial regression coefficient
P value
T value
Standard error
Number of affected fingers Patient age Interval from latest rupture to reconstruction Tendon reconstruction technique Wrist (DRUJ) reconstruction technique Primary disease (RA or OA)
0.490 0.450 0.308 0.183 0.364 0.068
<0.001 <0.001 <0.001 0.0724 0.1826 0.5923
4.648 3.623 3.767 2.021 3.668 0.538
3.004 0.088 0.137 2.976 4.691 3.689
RA: rheumatoid arthritis; OA: osteoarthritis. The number of affected fingers most strongly influenced extension lag, followed next by patient age. There was a mildly significant correlation between interval from rupture to reconstruction and extension lag.
cases of decreased ROM of the joint after surgery, and radiographs uncovered no patients with advanced OA changes during followup. Thus, it is unlikely that extension lag occurred because of the progression of joint contracture or OA. Disuse muscle weakness occurs frequently in patients aged over 80 years [18], which might have been one of the reasons for a poor outcome in very elderly patients in our study group. There were no patients who complained of restriction of finger flexion at final follow-up; postoperative rehabilitation using dynamic splinting likely contributed to this favourable result. In contrast, mean postoperative extension lag remained at approximately 10 , indicating that extensor tendons may be reconstructed at a stronger tension. This study has several limitations. First, it was retrospective in nature and the period of postoperative evaluation varied. Second, patients with RA or OA were analysed using the same methods. Although different pathological variables exist between the two disorders, no statistical difference in clinical results was observed. In all cases of RA, disease activity was well controlled and bony or joint destruction did not progress after extensor tendon reconstruction. As we could not determine the precise duration of RA, we could not analyze for relationships between RA duration and postoperative extension lag. Third, there were no clear criteria for the surgeon to judge which tendon reconstruction procedure should be performed. Lastly, no patient-based evaluation was performed either before or after surgery. Thus, how the remaining extension lag affected activities of daily living and patient satisfaction remain unclear. The number of tendon ruptures and the age of the patient are not under the control of the surgeon. However, it appears that tendon graft, extensor indicis proprius tendon transfer, and end-toside adjacent tendon suture are possible during the operation based on what is available as a motor source. The findings revealed in this study may also enable more precise informed consent from patients and prompter intervention before the involvement of multiple fingers. Future studies are required to quantitatively determine the excursion of reconstructed tendons, reveal the relationship between tendon excursion and postoperative extension lag, and clarify the associations between postoperative extension lag and patient-evaluated outcome. Further improvements in reconstruction procedures are also needed for patients aged over 80 years and for hands with extensor tendon rupture affecting three or four fingers. Conflict of interest The authors declare that they have no conflict of interest.
Acknowledgement The authors would like to thank Mr. Trevor Ralph for his English editorial assistance. References [1] Carr AJ, Burge PD. Rupture of extensor tendons due to osteoarthritis of the distal radio-ulnar joint. J Hand Surg Br 1992;17:694e6. [2] Chow JA, Dovelle S, Thomes LJ, Ho PK, Saldana J. A comparison of results of extensor tendon repair followed by early controlled mobilisation versus static immobilization. J Hand Surg Br 1989;14:18e20. [3] Chu PJ, Lee HM, Hou YT, Hung ST, Chen JK, Shih JT. Extensor-tendons reconstruction using autogenous palmaris longus tendon grafting for rheumatoid arthritis patients. J Orthop Surg Res 2008;3:16. [4] Chung US, Kim JH, Seo WS, Lee KH. Tendon transfer or tendon graft for ruptured finger extensor tendons in rheumatoid hands. J Hand Surg Eur 2010;35:279e82. [5] Moore JR, Weiland AJ, Valdata L. Tendon ruptures in the rheumatoid hand: analysis of treatment and functional results in 60 patients. J Hand Surg Am 1987;12:9e14. [6] Ohshio I, Ogino T, Minami A, Kato H, Miyake A. Extensor tendon rupture due to osteoarthritis of the distal radio-ulnar joint. J Hand Surg Br 1991;16:450e3. [7] Sakuma Y, Ochi K, Iwamoto T, Saito A, Yano K, Naito Y, Yoshida S, Ikari K, Momohara S. Number of ruptured tendons and surgical delay as prognostic factors for the surgical repair of extensor tendon ruptures in the rheumatoid wrist. J Rheumatol 2014;41:265e9. [8] Yamazaki H, Uchiyama S, Hata Y, Murakami N, Kato H. Extensor tendon rupture associated with osteoarthritis of the distal radioulnar joint. J Hand Surg Eur 2008;33:469e74. [9] Wolfe Scott W, Hotchkiss Robert N, Pederson William C, Kozin Scott H. Green's operative hand surgery. In: Strauch Robert J, editor. Extensor tendon injury. 6th ed. Churchill Livingstone; 2011. p. 159e88. [10] Kuniyoshi K, Kato H, Minami A, Suenaga N, Iwasaki N. Early active mobilization with dynamic splintage after reconstruction for extensor tendon rupture in rheumatoid hands. J Jpn Soc Surg Hand 2003;20:593e7 [in Japanese]. [11] Geldmacher J, Plank M, Treuheit KD. Significance of the preoperative status in the evaluation of results of the reconstruction of extensor tendons. Handchir Mikrochir Plast Chir 1986;18:23e9 [in German]. [12] Kapandji IA. The Kapandji-Sauve operation: its techniques and indications in non rheumatoid disease. Ann Chir Main 1986;5:181e93. [13] Wolfe Scott W, Hotchkiss Robert N, Pederson William C, Kozin Scott H. Green's operative hand surgery. In: Stanley John, editor. Arthroplasty and arthodesis of the wrist. 6th ed. Churchill Livingstone; 2011. p. 429e741. [14] Darrach W. Forward dislocation at the inferior radio-ulnar joint, with fracture of the lower third of the shaft of the radius. Ann Surg 1912;56:801. [15] Ikegami H, Takayama S, Ishiguro T, Horiuchi Y, Ichikawa T. Extensor tendon rupture in the rheumatoid hands. J Jpn Soc Surg Hand 2002;19:336e41 [in Japanese]. [16] Matsuda S, Mukaiyama K, Kato H, Yamazaki H. Treatment for osteoarthritis of the distal radioulnar joint and combined rupture of finger extensor tendons. J Jpn Soc Surg Hand 2004;21:479e81 [in Japanese]. [17] Shephard RJ. Management of exercise in the elderly. Can J Appl Sports Sci 1984;9:109e20. [18] Rosenberg I. Summary comments: epidemiological and methodological problems in determining nutritional status of older persons. Am J Clin Nutr 1989;50:1231e3.