Secondary Logo

Journal Logo

Arthrogryposis 2017 Supplement

Treatment of the Upper Extremity Contracture/Deformities

Oishi, Scott N. MD*; Agranovich, Olga MD; Pajardi, Giorgio E. MD; Novelli, Chiara MD; Baindurashvili, Alexey G. MD§; Trofimova, Svetlana I. MD; Abdel-Ghani, Hisham MD; Kochenova, Evgenia MD; Prosperpio, Giulietta MD; Jester, Andrea MD#; Yilmaz, Güney MD**; Şenaran, Hakan MD††,‡‡; Kose, Oksana MD§§; Butler, Lesley MPH*

Author Information
Journal of Pediatric Orthopaedics: July/August 2017 - Volume 37 - Issue - p S9-S15
doi: 10.1097/BPO.0000000000001002
  • Free


Arthrogryposis multiplex congenita (AMC) is estimated to occur in 1 in 10,000 live births.1 The underlying etiology still remains unknown. It affects both the upper and lower extremities in approximately 56% of patients, while the upper extremity is solely involved in approximately 17% of patients.2 In the upper extremity, a characteristic position includes shoulder internal rotation, elbow extension, wrist flexion/ulnar deviation, and thumb-in-palm deformity is usually present. The findings in the upper extremities should be symmetrical with sparing of the trunk. If not, a cervical spine work-up is warranted. The characteristic upper limb positioning makes performance of activities of daily living (ADL) much more difficult. These children are usually of average to above-average intelligence with a relatively normal life-span so any surgery that can change positioning can greatly enhance function as well as overall lifestyle. As the majority of these patients have lower extremity involvement as well, upper and lower extremity surgery should be coordinated to minimize anesthesia exposure and limit the number of surgeries. Initial surgeries are usually performed with goals of obtaining passive elbow flexion, improving wrist position, and getting the thumb out of the palm. At an older age, assessment for availability of functioning donor muscles for muscle transfer to obtain active elbow flexion is performed. There are several philosophies regarding treatment modalities as well as timing of surgical interventions. This article will review various philosophies including procedures to address the upper extremity, appropriate timing of said procedures, as well as postoperative rehabilitation.


Clearly the goals of treatment for the upper extremities is to enable the patient to be as independent as possible in performance of ADL’s as well as optimizing position for use in gainful employment. In most instances this involves procedures to obtain at least 90 degrees of passive elbow flexion, neutral wrist position, and thumb release. Every patient is different, so treatment must be tailored to match the clinical findings. It is important to note, however, that these patients have varying “catch-up” periods during the first year of life so therapy, including stretching and muscle strengthening, should be performed after initially evaluating these patients. As stated above, the majority of these patients also have lower extremity involvement so a coordinated approach can be used to limit the number of procedures and anesthetics. Fortunately, in most instances these can all be performed at the same setting. Some surgeons prefer to perform surgery at an early age (1 to 1½ year of age), while others prefer to wait until the child is older to perform these procedures. Surgery to obtain active elbow flexion in these patients can greatly enhance function, but requires good patient compliance with therapy for retraining of the muscle used to obtain the active elbow flexion. As a result, most of these surgeries are not performed until the child is near 5 years of age.


The typical internal rotation deformity present in many of these patients is very difficult to treat. Because of the paucity of muscle around the shoulder girdle in these patients, successful attempts at obtaining active shoulder external rotation or shoulder elevation in patients who lack this have not been reliably obtained. Instead, an external rotation osteotomy of the humerus can be performed. Although this does not result in obtaining active shoulder function, it can place the extremity in a better position for function. This can either be done proximally or distally, but it should be noted that a distal location is technically easier and the osteotomy can usually be stabilized with k-wires until healed. Positioning of the osteotomy is key, as over rotation can prevent mid-line functions and actually worsen function. Because of the inherent conditions present some recurrence of internal rotation deformity can occur as the child grows.

The elbow in the affected infant usually displays an elbow extension contracture. Although some patients may display active elbow flexion, most do not have this and only possess some degree of passive elbow flexion. This is a result of these patients having a functioning triceps muscle but lack of a functioning biceps muscle. One treatment philosophy includes initiating stretching and/or splinting at the initial evaluation in an attempt to obtain at least 90 degrees of elbow flexion. At about 1 year of age if this is not obtained, then an elbow release procedure may be indicated. The surgical procedure is shown in Figures 1A–E. It involves identification and transposition of the ulnar nerve followed by elbow capsulotomy and lengthening of the triceps tendon by either W-plasty or VY-plasty. As stated above, the goal of this procedure is to obtain at least 90 degrees of passive elbow flexion. Caution should be performed during manipulation of the elbow joint as too aggressive force may result in injury to the distal humeral physis. The elbow is then splinted in maximal flexion for 3 weeks followed by institution of range-of-motion exercises with flexion splinting. Longer periods of immobilization should be avoided to limit the risk of elbow flexion contracture.

A, Preoperative photograph. B, After capsulotomy and elbow release with the limbs of the triceps lengthening being shown; ribbon protects the ulnar nerve. C, Triceps lengthening. D, After completion of triceps lengthening. E, Postoperative view demonstrating passive elbow flexion obtained.

Obtaining active elbow flexion can enhance upper extremity function immensely for obvious reasons. It can be accomplished by different techniques including long-head triceps transfer, latissimus dorsi transfer, pectoralis transfer, or free gracilis transfer. Every patient needs to be evaluated individually as presence of recruitable muscles in these patients is highly variable. It is difficult to accurately assess the function of the muscles clinically in these young children, and thus a magnetic resonance imaging is usually recommended. It is also important that the patient have passive elbow flexion to at least 90 degrees with minimal flexion contracture. No matter what technique is used, a small increase in elbow flexion contracture will likely occur. In addition, patients with reasonable active shoulder elevation seem to have better outcomes from the transfer. There is no absolute optimal age for the procedure, however postoperative therapy is extensive and requires a cooperative patient. As a result usually this surgery is not performed until at least 3 years of age at the earliest, and most times closer to 5 years of age. If not done in the appropriate patient and setting, the procedure can result in a significant elbow flexion contracture which can worsen function.

In the patient that has robust long and lateral head of triceps muscles, a transfer of the long head of the triceps is an option for obtaining active elbow flexion (Figs. 2A, B). As in the elbow release procedure, identification and protection of the ulnar nerve is an imperative first step. Usually a good interval between the long and lateral heads can be developed and this can then be mobilized. Some surgeons use a fascia lata graft obtained from the thigh as an extension for the muscle, while others feel the surgery can performed without the extension graft. The long head is then transposed anteriorly and delivered into a separate incision in the antecubital fossa, and finally tunneled to the subcutaneous border of the ulna. Care must be taken when creating this tunnel to identify and avoid the median nerve and brachial artery. This is then fixed with either a bone anchor or the end of the fascia lata graft can be split with the ends brought around opposite sides of the ulna and attached to each other to set the tension. This is done with the elbow in 90 degrees of flexion and the shoulder in adduction (Figs. 3A–C). A long-arm cast is placed for 4 weeks. After removal, gravity-eliminated exercises are instituted with protective splinting. At 8 weeks, the splint is eliminated and against-gravity exercises are introduced. The patient must be monitored closely, as occasionally a significant elbow flexion contracture can occur and extension splinting must be instituted.

Proton density magnetic resonance imaging (MRI) of the right upper arm. A, Absence of brachialis and biceps muscles, but presence of robust triceps muscle. B, The 3 heads of the triceps are well defined.
A, Mobilization of the long head of the triceps with fascia lata graft attached. B, Passage of the long head of the triceps into the forearm. C, Attachment of the fascia lata graft to the ulna. Used with permission copyright @ 2017, Texas Scottish Rite Hospital for Children, Dallas, Texas, All Rights Reserved.

A study compared patients who underwent a long head of the triceps transfer with or without concomitant elbow release. Table 1 illustrates the comparison of the 34 patients included in the study. The study found that both groups regained excellent flexion; however, the group that also underwent elbow release had a comparatively increased risk of elbow flexion contracture. In addition, muscle strength was greater in the group that underwent long head triceps transfer alone. The authors separated their patients based on presumed spinal involvement, from C5 to T1. They concluded that in cases they classify as C6 and C6-C7 involvement, a latissimus dorsi muscle transfer was indicated. In cases of C5-C7 involvement latissimus dorsi or pectoralis major transfer was recommended, and in cases of C5-T1 involvement pectoralis major/ minor or long head of triceps transfer was recommended.

Results After Long Head of Triceps to Biceps Transfer in Patients With Arthrogryposis


Many times the wrist and thumb can be treated at the same operative procedure as the elbow release. The procedure for the wrist depends on the clinical examination. Most frequently a flexion contracture is present which requires a carpal wedge osteotomy to correct, however at times an extensor carpi ulnaris (ECU) to extensor carpi radialis brevis (ECRB) tendon transfer may suffice. Although the radial wrist extensors are usually atretic with little (if any) extensor function, the ECU tendon is frequently spared but acts as a pure ulnar deviator of the wrist instead of extensor. The carpal wedge osteotomy procedure is based on the fact that these patients have radiocarpal motion, but usually develop carpal coalitions between the proximal and distal rows long-term. In any case, release of the volar fascia is usually required to remove a significant deforming force.

The technique for a carpal wedge osteotomy has previously been reported with reproducible results that largely maintain position over time.3,4 First, an incision is made on the volar side of the distal forearm and the volar fascia is released. At times fractional lengthening of wrist flexors and/or finger flexors is required as these structures may also be tight and contributing to the deformity. Next, the dorsal wrist is exposed and capsule opened. A bi-planar osteotomy is then performed and stabilized with pin fixation and nonabsorbable suture or wire. An ECU to ECRB tendon transfer is also performed at the same setting (Fig. 4).3 Care must be taken when performing the proximal osteotomy to leave enough proximal row for radiocarpal motion. Pin fixation is important during the healing of the osteotomy so as to avoid breakage of the wire or suture. The wrist is usually immobilized for 6 weeks postoperatively with protective splinting after pin removal. When performed with elbow release the long-arm cast is initially changed at 3 weeks with fabrication of the elbow splint and application of a short-arm cast for another 3 weeks.

Technique for carpal wedge osteotomy, showing the anteroposterior A, and lateral B, view of the location of the osteotomy and the anteroposterior C, and lateral D, view after wedge osteotomy. MC indicates metacarpal.3 Used with permission from Foy et al.3

One proposed algorithm for wrist treatment, based on a review of 172 patients (334 upper extremities), stratifies the wrists into 3 treatment groups: isolated flexion contracture of the wrist (64%), flexion contracture associated with ulnar deviation (24%), and isolated ulnar deviation of the wrist (12%). Table 2 separates patients into 4 groups based on level of perceived spinal cord lesion. Increasing arthrogrypotic involvement at the wrist led to decreased amplitude of passive and active movements along with decreased muscle power and wrist function. The frequency of upper extremity and hand contractures progressively increased with increasing level of perceived spinal cord involvement. If passive wrist extension to neutral or past neutral is present, then tendon transfers are performed (flexor carpi radialis to ECRB) with shortening of the ECRB and extensor carpi radialis longus. If the flexor carpi radialis (FCR) and flexor carpi ulnaris are transected and wrist extension to neutral is not obtained, then a carpal wedge osteotomy is performed with tendon transfers. In severe wrist flexion contractures, a carpal wedge osteotomy with tendon transfers as well as shortening osteotomy of the forearm is performed. If isolated ulnar deviation is present then tendon transfers without carpal wedge osteotomy is performed. After tendon transfer surgery alone casting for 4 weeks was performed. In the patients that received carpal wedge osteotomy casting was performed for 5 to 6 weeks, with longer immobilization (6 to 8 wk) for the patients that received carpal wedge osteotomy and forearm shortening osteotomies. No matter the type of surgery performed, removable splints were instituted after removal of the casts. Results were assessed by analyzing the functional position of the wrist in frontal and sagittal planes, amplitude of active wrist extension, cosmetic appearance, and functional ability to grasp. Comprised results consisted of 53% good result, 42% satisfactory result, and 5% unsatisfactory result. Results were then categorized by perceived spinal cord lesion: C6-C7 had 100% good outcome, C5-C8 had 84% good and 16% satisfactory outcome. In the C5-T1 group 11% had good, 79% satisfactory, and 10% unsatisfactory outcome.

Variants of Upper Limb Deformities Due to Arthrogryposis Multiplex Congenita

The thumb-in-palm deformity can be very difficult to treat as it can involve a skin deficit, muscle/ fascia contracture, possible FPL contracture, and possible extensor weakness. Failure to address each of these components will lead to a suboptimal result. The goal is to allow the thumb to at least clear the second metacarpal. One approach for surgical treatment of the muscle/ fascia contracture is a thenar slide. An incision is made next to the thenar crease, and the origin of the thenar muscles is released after first identifying and protecting the motor branch of the median nerve. This allows the first metacarpal to abduct out of the palm. As the soft tissue contracture is 2-dimensional in nature and is not usually adequately treated with z-plasty alone, an index transposition flap can be included to effectively improve active thumb function.5 In all cases, temporary pin fixation across the MP joint is imperative for wound healing and positioning of the thumb. If the FPL is tight and prevents thumb abduction, a fractional lengthening can be performed but must be done carefully. The majority of these patients have very small tendons and muscles so care to not weaken their function must be kept in mind.

Lastly, careful examination may reveal the status of active thumb extension. In some cases extensor pollicis longus (EPL) function is present, therefore thenar release and thumb soft-tissue release will allow adequate thumb extension to be obtained. However, in some cases the EPL tendon must either be plicated or a tendon transfer be contemplated to increase extensor power. Failure to address this will lead to recurrence of the thumb deformity. Sadly, in many cases, EPL function is not adequate and no adequate tendon is available for transfer. Splinting and therapy after pin removal are important to maintain and optimize the results from surgery.

An option to treat an unstable thumb by chondrodesis of the MP joint was highlighted in a study of 69 complex clasped thumbs in 39 patients at a single center who were operated on at an average age of 23 months (15 to 36 mo). Surgical technique included skin augmentation of the deficient first web using the the Ghani flap,6 release of the tight structures that were present in the first web-space, and chondrodesis of the MP joint to achieve stability. At an average follow-up of 3 years (2 to 7 y), all parents were satisfied with the outcome. No recurrence of web narrowing occurred. Significant improvement of thumb position and function was noted. Nonunion of the chondrodesis occurred in 7 thumbs with 2 exhibiting fair stability and 5 with poor stability. Of note, release of the first web space involved release of the intermetacarpal fascia in all 69 thumbs, adductor pollicis in 41 thumbs, first dorsal interosseous in 35 thumbs, and CMC joint in 30 thumbs.

Another study assessed the outcomes of an individually tailored approach of soft tissue and osteotomies for 41 patients treated between 2000 and 2014. The average age at surgery was 7.5 years. Specific aims of the surgery included correction of anatomically distinct factors contributing to the deformity, web space deepening and expanding, release of first web muscles, as well as osteotomies if indicated. Bony procedures were common in the young patient with severe deformity in addition to the soft-tissue releases and reconstructions. All patients reported functional and esthetic improvement after surgery and appropriate therapy. This study led to the conclusion that an aggressive and early approach combining splinting and surgery could improve patient function and independence (Figs. 5A–C).

A, Preoperative view of thumb-in-palm deformity. B, Postoperative 10-year follow-up. C, Postoperative 10-year follow-up.


Patients with arthrogryposis present challenging reconstructive problems for the surgeon. These patients are usually of near-normal or above-normal intelligence so any surgery to place their extremities in better position can be extremely useful and important for independence and performance of ADL’s. Timing of surgery varies among surgeons, but is usually accomplished before starting school. Types of surgery are also varied although the goals are the same: passive elbow flexion to enable the hand to reach the mouth, wrist positioning in neutral or slight extension, thumb out of the palm with adequate first web-space, and active elbow flexion (in the suitable patient). As many of these patients have lower extremity involvement, coordination of procedures is important to limit anesthetic exposure.


1. Bamshad M, Van Heest AE, Pleasure D. Arthrogryposis: a review and update. J Bone Joint Surg Am. 2009;91(suppl 4):40–46.
2. Hall JG, Aldinger KA, Tanaka KI. Amyoplasia revisited. Am J Med Genet A. 2014;164A:700–730.
3. Foy CA, Mills J, Wheeler L, et al. Long-term outcome following carpal wedge osteotomy in the arthrogrypotic patient. J Bone Joint Surg Am. 2013;95:e150.
4. Ezaki M, Carter PR. Carpal wedge osteotomy for the arthrogrypotic wrist. Tech Hand Up Extrem Surg. 2004;8:224–228.
5. Ezaki M, Oishi SN. Index rotation flap for palmar thumb release in arthrogryposis. Tech Hand Up Extrem Surg. 2010;14:38–40.
6. Ghani HA. Modified dorsal rotation advancement flap for release of the thumb web space. J Hand Surg Br. 2006;31:226–229.

arthrogryposis; arthrogryposis multiplex congenital; amyoplasia; congenital contracture syndromes; distal arthrogryposis

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.