Wrist arthrodesis and soft tissue rebalancing in the spastic hand

Introduction

Upper limb spasticity is a common condition often associated with upper motor neuron pathologies, with the wrist being the most affected joint. This typically leads to a flexed position of the wrist, significantly limiting hand function and causing disability (Seruya et al., 2016).

Various surgical intervention options are available for upper limb spasticity, including soft tissue releases, tendon transfers (Pontén et al., 2019; Schibli and Fridén, 2022), selective peripheral neurectomy (Kwak et al., 2011) and wrist arthrodesis (Van Heest and Strothman, 2009). However, decision-making in the treatment of wrist spasticity can be challenging due to the heterogeneous nature of the patient population and factors such as skeletal maturity, degree of contracture, active hand control, and differentiation between spasticity and fibrosis within the muscles.

A key concern in performing a wrist arthrodesis in patients with long-standing flexion contractures due to spasticity is the effect of repositioning and immobilizing the wrist on the function of the extrinsic tendons to the fingers and the thumb. Although in the spastic hand the fingers are typically straight when the wrist is in a flexed posture, correction of the wrist position to neutral may unmask tightness in the extrinsic tendons to the digits. Furthermore, some patients rely on their limited wrist motion for grasp and release (Seruya et al., 2016). Arthrodesis of the spastic wrist in a neutral position without addressing the extrinsic tendons can inadvertently lead to a clasp hand with poor release and reduced function. For this reason, we advocate the importance of careful patient selection, and in those patients suitable for arthrodesis, the simultaneous rebalancing of the extrinsic tendons to optimize the arc of movement and function in the digits.

In this study, we assessed the outcome of patients who underwent simultaneous procedures of wrist arthrodesis followed by digit rebalancing. We addressed the criteria for appropriate patient selection for wrist arthrodesis, patient-reported goals for surgery, incidence and types of digital rebalancing procedures simultaneously performed, and outcomes of these procedures in the spastic wrist.

Methods

This retrospective study involved a review of 43 wrists in 40 consecutive patients diagnosed with spasticity who underwent wrist arthrodesis surgery. The surgeries were performed by an upper limb spasticity surgeon (MN) with a level 4 expertise (Tang and Giddins, 2016) between July 2013 and September 2021 at two hospital sites. Ethical approval to review the outcomes was obtained from the audit and research committees at each hospital. Demographic data collection by the surgical and hand therapy teams included patient age, underlying diagnosis (cerebral palsy, stroke and other upper motor neurone pathologies) and the Modified Ashworth Score (MAS) for assessment of generalized spasticity.

Outcome measurements

The preoperative range of wrist motion was assessed by measuring passive extension with the hand clasped, to exclude the influence of the extrinsic digit tendons. Due to normal variations in the range of motion of the wrist joint (Costa et al., 2020), the degree of passive movement was expressed as being above or below the wrist in neutral (Figure 1). Further assessment of associated contractures of the extrinsic muscles of the fingers and thumb tendons was performed after surgical correction of the wrist contracture.

OPEN IN VIEWER

Figure 1.

Preoperative photo showing passive wrist extension to 20° beyond neutral, but with evidence of extrinsic flexor tendon contracture.

Patient-specific goals for surgery were assessed using the Canadian Orthotic Performance Measure (COPM), categorized into function, resting position, grip strength, hand hygiene, appearance and pain. These were assessed preoperatively for baseline comparison and reassessed at 3 and 12 months postoperatively to calculate the Goal Attainment Score (GAS; see below for details). The final outcomes were classified into four categories based on complications and attainment of surgical indications:

Excellent: no complication, all indications met

Good: resolved complication and/or one indication not met

Fair: recurrent deformity needing surgery, two or more indications not met

Poor: worse than preoperatively.

Surgical technique

Surgery was usually performed as part of Single Event Multi-Level Surgery (SEMLS), with the patient positioned in a supine position with a tourniquet applied and loupe magnification used. Shoulder and elbow releases were performed first to improve patient positioning.

Wrist arthrodesis

Wrist arthrodesis was performed using a standard dorsal approach to the distal radius and carpus (Figure 2). The proximal row of the carpus was excised and morselized for use as bone graft. The amount of bone excised during the preparation of the distal radius depended on the degree of fixed flexion contracture. In more severe cases flexion contracture >90° below neutral, sufficient bone was excised to enable the wrist to be brought into a neutral position with minimal tension on the volar soft tissues. If after preparation for the wrist arthrodesis tension was still noted in the wrist flexors, these were percutaneously released to minimize deforming forces on the plate to prevent cut out. The wrist was fixed in a neutral or slightly extended (10°–20°) position using a Synthes straight wrist fusion plate (Synthes, Leeds, UK). We chose this position as it restored more normal aesthetics, while providing a stable base to rebalance the digits to enable grasp release and flexion function (Bozon et al., 2022).

OPEN IN VIEWER

Figure 2.

Dorsal wrist arthrodesis with a locking plate. (a) Note detensioning of wrist extensor tendons and the flexed posture of the fingers at the MCPJs, indicating detensioning of the EDC tendons and (b) Side view showing wrist position in neutral. ECRB: extensor carpi radialis brevis; ECRL: extensor carpi radialis longus; EDC: extensor digitorum communis; MCPJ: metacarpophalangeal joint.

Digital rebalancing

After arthrodesis, digital rebalancing was addressed. The position of the hand and fingers were reassessed, considering factors such as residual tightness in the intrinsics and flexor digitorum superficialis (FDS) tendons, and laxity of the extensor digitorum communis (EDC) tendons.

Typical rebalancing procedures for the fingers included:

FDS fractional lengthening, involving multiple non-linear partial tenotomies of the musculotendinous junction (Arnaout and Leclercq, 2022), followed by stretching of the digit to allow the fibres to elongate without completely rupturing (Figure 3).

OPEN IN VIEWER

Figure 3.

Fractional lengthening of the flexor digitorum superficialis tendons at the musculotendinous junction.

Intrinsic release, with our preference being the release of the interossei from their metacarpal origin since the metacarpals are already exposed during wrist arthrodesis. However, alternative methods of intrinsic release include releasing them from their distal insertion into the extensor mechanism.

Restoring metacarpal extension using a tendon transfer technique involving the extensor carpi radialis longus (ECRL) tendon to the EDC (Figure 4). This transfer was carefully adjusted to maintain the metacarpophalangeal joint (MCPJ) at approximately 30° of flexion instead of full extension. Given the patients’ limited hand control and weak extension, we view this transfer as functioning as a tenodesis. Although we considered other potential donor tendons for transfer, we found that the ECRL tendon was conveniently located near the EDC, redundant after the wrist arthrodesis, and capable of providing appropriate tension to keep the digits extended away from the palm.

OPEN IN VIEWER

Figure 4.

Tenodesis of the divsions of the extensor digitorum communis tendons, with the fingers held extended.

Typical rebalancing procedures for the thumb included:

Addressing metacarpal adduction caused by intrinsic spasticity. The adductor pollicis muscle was released from the middle finger metacarpal using a palmar approach.

Addressing any flexor pollicis longus (FPL) contracture, which presented as interphalangeal joint (IPJ) contracture. This was corrected by performing a z-lengthening of the FPL tendon.

Extensor rebalancing was addressed by shortening the extensor pollicis longus tendon and rerouted it around the abductor pollicis longus tendon (Figure 5).

OPEN IN VIEWER

Figure 5.

(a) Transfer of extensor carpi radialis longus to extensor digitorum communis to maintain extension at the metacarpophalangeal joint and (b) Extensor policis longus shortened and rerouted around the abductor pollicis longus tendon to maintain a more extended thumb position at the level of index finger.

In cases of severe MCPJ contracture, joint arthrodesis was performed.

The ultimate aim of surgery was to stabilize the wrist in a neutral position (Figure 6), with the fingers in a neutral cascade and the thumb opposable to the index finger.

OPEN IN VIEWER

Figure 6.

Postoperative radiographs of wrist arthrodesis. (a) Anteroposterior and (b) lateral.

Postoperative follow-up involved placing patients in a volar plaster that extended to include the MCPJ for 6 weeks. Rehabilitation and reviews were conducted at 2, 6, 12 and 52 weeks postoperatively. The minimum follow-up was 1 year.

Results

Patient demographics

A total of 43 wrists were arthrodesed in 40 patients with a median follow-up of 12 months. The majority of cases had a diagnosis of cerebral palsy, with a mean age of 17.2 years (standard deviation [SD] 6.2); five cases were due to stroke or acquired brain injury and had a mean age of 54.1 years (SD 27.8) (Table 1).

OPEN IN VIEWER

Table 1.

Patient demographics, preoperative information and surgical indications.

Demographics and preoperative findings	Different parameters
No. of patients (wrists)	40 (43)
Age a (years)	21.5 (16)
Follow-up b (months)	12.8 (0.6–77.5)
Diagnosis
CP	38 (88)
Stroke	2 (5)
Other	3 (7)
Modified Ashworth Score b	3.0 (1–4)
Passive extension a (°)	–46 (46)
Active control
Good	2 (5)
Limited	5 (12)
None	36 (84)
Surgical indications
Activities of daily living	22 (51)
Position	43 (100)
Grip	29 (67)
Pain	27 (63)
Hygiene	39 (91)
Appearance	31 (72)

Data are presented as n (%), unless otherwise indicated.

a

Data are presented as mean (SD).

b

Data are presented as median (range).

CP: cerebral palsy; SD: standard deviation

Patients typically exhibited high levels of both generalised and wrist spasticity, with a mean MAS score of 3.0 (SD 1.0). In total, 40 patients underwent SEMLS, at a mean of 3.4 levels (SD 1.1). Preoperatively, the mean passive extension was 44° below neutral. Five patients had a greater range of motion preoperatively (mean 38° passive extension above neutral): two of these had a hypotonic form of cerebral palsy and three had recurrent deformity after previous soft tissue rebalancing. Of the 40 patients, only two had good active control in their hand.

Wrist repositioning to neutral or near neutral was a desired surgical goal by all patients. Among other goals, 39 individuals desired improved hand hygiene, washing or dressing, while 31 patients sought surgery for aesthetic reasons. Notably, patients who cited appearance as a motivating factor for surgery were considerably younger, with a median age of 14.7 years compared to 21.0 years (p < 0.001, Mann–Whitney-U test). Conversely, those whose primary indication was pain were significantly older, with a mean age of 17.9 years versus 13.4 years (p < 0.001, Mann–Whitney U-test). The surgical indications and demographics are summarized in Table 1.

Outcomes and complications

All cases reported excellent (n = 33) or good (n = 10) outcomes, indicating the achievement of surgical goals without residual complication at final review. The final GAS score of 64.2 indicated better than expected outcomes, with a mean improvement by 31.3 points (Figure 7). Specific patient-reported outcomes included improvements in their ability to control a wheelchair, mobilize with a frame and control mobile devices, resulting in significantly enhanced independence.

OPEN IN VIEWER

Figure 7.

Clinical outcome of surgery demonstrating (a) preoperative appearance. (b) postoperative resting posture and (c) function.

Among the 10 patients who experienced complications, it was noted that their preoperative wrist contractures were significantly greater (mean passive extension 71° vs. 35° below neutral, respectively; p < 0.05, Student’s t-test). Five patients encountered significant wound breakdown and dehiscence; two required prolonged vacuum-assisted closure (VAC) therapy and one needed a pedicle flap for wound closure. These patients had all undergone simultaneous release of severe contractures at both the elbow (mean 89°) and wrist (mean –64°). Implant failure occurred in two wrists, with one case involving plate fracture and the other involving screw pull-off, both occurring in the same patient with high spasticity. Subsequently, for the severely contracted wrists, a more extensive radial shortening with release of the wrist flexors is recommended to reduce forces on the plate. Two patients developed transient skin hypersensitivity after repositioning from a severely flexed position, which resolved spontaneously after a few weeks. Finally, one patient suffered a fall and suffered a periprosthetic fracture. Due to this patient requiring treatment for recurrence of an intracranial tumour, surgical intervention was not possible, and the fracture healed spontaneously.

Discussion

This study demonstrated that in appropriately selected patients, the outcomes of wrist arthrodesis are usually favourable. Our primary indication for surgery was the presence of a severe fixed flexion deformity and the absence of voluntary hand control. The key goals for surgery were pain reduction, aesthetic appearance and functional gain, with these being achieved in most patients. However, achieving these results necessitated concurrent surgery on the digits in 39 out of 43 hands. It was observed that patients with severe multi-level contractures experienced a higher rate of complications.

While there is existing literature on the role of wrist arthrodesis in the spastic hand (Van Heest and Strothman, 2009), this study is unique in addressing the importance of patient selection and soft tissue correction of the digits in association with correcting wrist deformity. The available evidence is limited by significant heterogeneity of patient characteristics, underlying pathology, surgical procedure and the need for multi-level surgery. By analysing a subset of patients who required wrist arthrodesis in this study, we aimed to define outcomes in this more severely disabled patient group.

In our series, the typical patient had little active control and preoperatively had an average passive wrist extension of 44° below neutral. For skeletally immature patients and those with less contracture and better control, rebalancing procedures are recommended instead of arthrodesis (Seruya et al., 2016). It is important for the surgeon to recognize the importance of the wrist on the function of the digit extrinsics and how a wrist arthrodesis results in the loss of tenodesis function of the digital tendons. However, in the absence of active control and with fixed flexion contracture, stabilizing the wrist through arthrodesis allows appropriate tensioning of the digits. Our results show promising outcomes of wrist arthrodesis, but further comparative studies are needed to compare similar patient groups undergoing arthrodesis with extensile surgical releases that preserve the wrist joint (Acartürk et al., 2006).

Our wider experience of the surgical management of wrist spasticity allowed us to develop a treatment algorithm to help select suitable patients for wrist arthrodesis or rebalancing procedures:

In skeletally immature patients, we would advocate neurectomy or rebalancing procedures.

The choice of outcome measures in the literature varies significantly; several have utilized the Zancolli classification (Zancolli, 1979) to evaluate extension of the fingers and wrist, but we found the personalized outcome of the GAS allows for the evaluation of multi-level surgery and helps set realistic expectations for patients. Appearance was a significant indication for surgery, particularly in younger patients, while older patients were more likely to present with pain.

While wrist arthrodesis with soft tissue rebalancing generally led to favourable outcomes, there was a notable complication rate, including wound problems, implant failure, periprosthetic fracture and transient skin sensitivity. We observed a correlation between postoperative wound breakdown and the degree of preoperative wrist and elbow contracture, suggesting that release of severe contractures puts neurovascular structures at risk due to stretching. Our current practice includes avoiding postoperative splinting of the elbow in extension and performing sufficient radial shortening to avoid tensioning the volar wrist tissues. Since incorporating this, we have seen no further incidents of wound breakdown.

Patients included in this study primarily presented with spasticity rather than dystonia. Nonetheless, there are studies (Alexander et al., 2000; Hargreaves et al., 2000) that have reported positive outcomes for dyskinetic patients. However, a recent small series (Seok et al., 2021) observed worsening dyskinesia and instances of metal work failure in these patients. Another study (Gatin et al., 2017) focused exclusively on soft tissue surgery for upper extremity contractures, and another (Thevenin-Lemoine et al., 2013) concentrated on a select group of patients undergoing a specific procedure without significant focus on complications related to preoperative contracture status. Our study fills a gap by examining both wrist fusion and digital rebalancing, plus investigating the relationship between preoperative contracture and postoperative complication rates.

It is important to acknowledge the limitations of our study, which is a non-comparative series in a heterogenous population. Standardized outcomes are difficult in this population group, and we found the GAS to be a useful, patient-specific tool. Video and motion analysis outcome measures are not available in our units. Isolation of specific outcomes is difficult when a patient undergoes SEMLS, although we feel SEMLS is a better strategy for patient recovery. We have attempted to minimize these limitations by looking at a defined subset of management of the spastic wrist and evaluating how it impacts on other surgical procedures concurrently required to maximize hand function.

These limitations are similar to those in other publications on wrist spasticity surgery (Hashemi et al., 2021). However, it is hoped that this study provides insight into expected outcomes, especially with the emphasis on digital rebalancing. While most other authors utilize a dorsal plate similar to our approach, alternative techniques, such as volar plates (Bozon et al., 2022) and intra-medullary techniques (Lestienne et al., 2022), have also been described for spastic wrist arthrodesis. A similarly sized study of 41 wrist arthrodesis for spasticity (Van Heest and Strothman, 2009) demonstrated good outcomes but noted a high fracture rate and a significant need for metalwork removal. The other published series have relatively small patient numbers (range 10–20 cases), but generally report improvements in function and appearance (Alexander et al., 2000; Donadio et al., 2016; Hargreaves et al., 2000; Neuhaus et al., 2015; Pomerance and Keenan, 1996; Rayan and Young, 1999; Samade et al., 2022; Vergara-Amador and Franco-Chaparro, 2020).

In conclusion, wrist arthrodesis with concurrent soft tissue rebalancing of the digits is a viable option for improving wrist position, stability and hand function in patients with severe spasticity and fixed flexion contracture. However, careful patient selection and consideration of potential complications are essential.