The timing of open surgical release of post-traumatic elbow stiffness: A systematic review.

BACKGROUND: Open release of post-traumatic elbow stiffness is effective in restoring elbow function, but there is no guideline on the optimal time point of surgical release so far. The purpose of this article was to summarize the current available literature reporting on the timing of open release of post-traumatic elbow stiffness.
METHODS: The PubMed, Cochrane Library, and EMBASE were searched with a set of predefined inclusion and exclusion criteria. Manual searches for references were performed to find potential relevant studies. Two authors separately extracted data from all the articles selected.
RESULTS: 27 articles published between 1989 and 2017 were included with an overall enrollment of 836 patients. We divided all included studies into 3 groups according to the timing of surgical release: group 1 (6-10 months after injury), group 2 (11-20 months after injury), and group 3(>20 months after injury). The mean postoperative Mayo Elbow Performance Score (MEPS) and recurrence rate were similar among the 3 groups; however, the mean gain in arc of motion in group 1 was the highest with the lowest complication rate among the 3 groups.
CONCLUSION: There was a trend toward a shorter waiting time from injury to open arthrolysis from 12 months to 6 months. The shorter waiting period of 6 to 10 months yielded better results. Therefore, early surgical release of stiff elbows is recommended for a shorter rehabilitation time and earlier return to work. LEVEL OF EVIDENCE: Level IV, Systematic Review.

Introduction

Elbow stiffness is generally defined as elbow range of motion (ROM) < 30° to 130°. Morrey et al[ reported that most of our daily activities can be accomplished with 100° of elbow flexion (from 30° to 130°) and 100° of forearm rotation (50° of pronation and 50° of supination). Although a variety of causes may lead to elbow stiffness, elbow stiffness is mainly caused by traumatic causes. Post-traumatic elbow stiffness may result from contracture of the joint capsule and ligaments, muscle contracture, adherence, osteophytes, ectopic ossification, articular surface incongruity, and loose bodies.[

There are 2 major methods to treat elbow stiffness: nonoperative and operative methods. Nonoperative methods include myofascial soft tissue mobilization, joint mobilization, splinting, and so on.[ Operative methods mainly include arthroscopy and open arthrolysis. Due to the complex pathology, arthroscopy often fails to result in complete release and thus open arthrolysis should be the optimal method.[ Generally, if functional ROM is not obtained by nonoperative treatment, surgical release may be taken into consideration.

Operation is usually performed 6 months or more after injury, during which time bone gradually grows to be mature and secondary contracture of soft tissue occurs leading to dissatisfactory outcomes.[ Several studies have reported that early surgical release is both safe and effective.[ There is no guideline on the optimal time point of surgical release so far. Therefore, we initiated a systematic review of studies on the time point of surgical release to compare recurrence rates and obtained ROM between early and late release; thus providing a reference for the optimal time point of surgical release.

Methods

This systematic review is reported on the basis of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines.[ Since it was a systematic review, it did not need ethical approval or patient consent.

Inclusion and exclusion criteria

An article was found eligible when it: (1) met diagnosis standards for post-traumatic elbow stiffness (such as elbow arc of motion <30°–130° after trauma);(2) provided data regarding the time point of surgical release;(3) reported treatment of open surgery and results of post-traumatic elbow stiffness in human adults with a minimal follow-up of 1 year. We excluded articles if they: (1) reported on stiff elbow after burns or brain injury, outcomes of arthroscopic arthrolysis or arthroplasty, or revision operations;(2) were reviews, expert opinions, case reports on 5 or less patients, cadaveric or biomechanical studies, or full texts not available; (3) did not provide related data on outcomes.

Search strategy

We searched the PubMed, Cochrane Library, and EMBASE in September 2016 and updated our search in March 2017 using the following keywords ([elbow] AND (stiff∗ or contracture or ankylos∗) AND (releas∗ or arthrolysis) to include relevant studies as many as possible. In addition, we performed manual searches for references regarding stiff elbow or elbow stiffness to find potential relevant studies. The search was limited to articles written in English.

Data extraction

Two authors (CS and CLY) separately extracted data from articles which had been selected according to the inclusion and exclusion criteria and the final data were selected by their common decisions. The data include: surgical timing, number of patients, gender, mean age, outcome measures, mean follow-up, and general complications.

Statistical analysis

This review has extracted and summarized data from 27 studies involving 836 patients (Table 1). The average time from initial injury to surgery in all patients ranged from 6 to 120 months (Table 2). Early release was defined as release performed at less than 10 months.[ Considering the sum of articles, the sum of patients and surgical timing, we divided all included studies into 3 groups: 6 to 10 months after injury (the first group), 11 to 20 months after injury (the second group) ,>20 months after injury (the third group), as listed in Table 3. It was clear that most studies had different inclusion criteria, outcome measures, and postoperative protocols, thus prohibiting statistical analysis among the different studies. Nonetheless, a general comparison in terms of improvements in arc of motion, complications rate, and recurrence rate were carried out, all of which were crucial parameters in evaluation of surgical treatment.

Table 1

Characteristics of articles included.

Table 2

Functional outcomes of articles included.

Table 3

Summary of different surgical timing.

Results

Process outcomes

We included 27 articles with an overall enrollment of 836 patients (Table 1). The number of patients in the articles ranged from 6 to 91. These articles were published between 1989 and 2017. Figure 1 presents the number of articles screened, assessed, and included in the review. Table 1 summarizes the main characteristics of articles included in the systematic review. There was one retrospective cohort study (level 3) and the rest were case series (level 4), as assessed by the Oxford Centre for Evidence-Based Medicine 2011 Levels of Evidence.[

Figure 1

Flowchart for article selection.

Results of analysis

Range of motion (ROM)

We took the data regarding ROM as main functional outcome (Table 3). The mean preoperative arc of motion was 38.5°, 44.7°, and 50.5° for groups 1 to 3, respectively. The mean postoperative arc of motion was 109.7°, 106.3°, and 108.4° for groups 1 to 3, respectively. The mean gain in arc of motion was 71.2°, 61.6°, and 57.9° for groups 1 to 3, respectively. By comparison, the first group (surgical timing:6–10 months after injury) whose mean preoperative arc of motion was the smallest in 3 groups achieved the highest mean gain in arc of motion, which was 8.6°more than the total mean gain in arc of motion.

Complications

Complications included infection, heterotopic ossification, elbow instability, nerve complications, pain and others and the mean complications rate was 22.9%, 25.8%, and 28.6% for groups 1 to 3, respectively. We classified the complications into major ones (including infection, elbow instability, nerve complications, recurrence, and refracture) and minor ones (including pain and others). The mean rate of major complications was 17.0%, 22.7%, and 21.4% for groups 1 to 3, respectively (Table 4). Group 1 had the lowest major complications rate and the lowest mean complications rate. The mean rate of recurrence was 10.1%, 12.1%, and 9.0% for groups 1 to 3, respectively (Table 4). The mean rate of pain was 0, 1.2%, and 5.1% for groups 1 to 3, respectively (Table 4). Apparently Group 1 had the lowest mean rate of pain.

Table 4

Summary of complications in each group.

Mayo elbow performance score (MEPS)

The mean post-operative MEPS was 92.1, 93.1, and 87.6 for groups 1 to 3, respectively. The mean postoperative MEPS in group 1 was 92.1, which was slightly higher than total mean postoperative MEPS.

Discussion

The surgical timing is vital to the treatment and rehabilitation of elbow stiffness, but the surgical timing is controversial. There is no guideline on the optimal surgical time point for release so far, which is usually dependent on surgeons’ experience. Surgical release is commonly indicated when nonoperative treatment fails to improve ROM after a period. It is emphasized that a delay of at least 12 or 6 months is necessary to ensure that the maturation of bone occurs,[ which leads to a low risk of recurrence. However, during the waiting period the stiff elbow gives rise to significant inconvenience to a patient's daily life. During this period the ligaments and capsules undergo fibrosis, and articular cartilage degrades and muscles atrophies, contributing to secondary joint arthrosis. Secondary joint arthrosis, in turn, will aggravate elbow stiffness.[ The ulnar nerve will be damaged by ischemia and compression due to constant flexion of the elbow or pressure caused by HO.[

Generally, when nonoperative treatment fails to obtain functional ROM, surgical release may be taken into consideration. But the optimal time point of surgical release is difficult to determine. By studying many articles, we found that there was a trend toward a shorter waiting time from injury to open arthrolysis from 12 months to 6 months. Due to a high incidence of recurrence, surgical excision of HO is usually delayed for 12 to 24 months (from injury to operation) until the maturation of bone occurs.[ Most of these articles were published before 2000. As time went by, many authors whose articles had been published after 2000 advised that nonoperative treatment such as physiotherapy, dynamic splinting, and static progressive splinting should be continued for at least 6 months before operative treatment.[ Apparently, the waiting period from injury to operation was shorter than before.

From Table 3, it was clear that the first group (surgical timing: 6–10 months) achieved the highest mean gain in arc of motion, which was 8.6°more than the total mean gain in arc of motion. From Tables 3 and 4, group 1 had the lowest major complications rate, the lowest mean complications rate, and the lowest mean rate of pain. But the mean rate of recurrence in group 1 was not the lowest. We thought that there were 2 major factors contributing to this result. Firstly, the slight recurrent HO seen on plainfilms made a contribution. At the least 12 months follow-up, the small degrees of HO did not affect the function of elbow and the final ROM was considerably improved.[ Secondly, the number of patients in the first group were small, which could improve the ratio. So we concluded that the shorter waiting period of 6 to 10 months yielded better results. Therefore, early surgical release of stiff elbows is recommended for a shorter rehabilitation time and earlier return to work.

There is no definite evidence indicating that the surgical delay means less recurrence and better function. Rex et al[ have reported that they believe that it is most beneficial to excise HO in the first 6 months following injury and it is unnecessary to wait for the maturity of HO before excision. Viola and Hanel[ performed early excision, at a mean waiting period of 23 weeks, in 15 elbows with post-traumatic elbow stiffness, and they found no recurrence. Yang et al[ analyzed the results of early surgical management (mean, 6 months) in 7 patients. Their results were satisfactory. Almost full range of motion and complete functional ability following surgery were recovered in 6 of the 7 patients. Koh et al[ reported that delayed surgery is associated with less favorable results. Early improvement in ROM is significant for early rehabilitation and the recovery of elbow function.[ Early release will bring significant convenience and less pain to patients’ daily life. Early release will accelerate the treatment time and recovery period, which will enable patients to return to daily life and work as soon as possible.

Many factors affect postoperative outcomes. Physical therapy and constant passive exercise are useful to minimize the loss of ROM acquired intraoperatively. The rigorous rehabilitation exercises are beneficial to prevent the recurrence of joint adhesion, thus leading to good results. Patients’ compliance with postoperative rehabilitation program is important for favorable postoperative outcomes.

There are some limitations in this review. Firstly, the number of patients among 3 groups differs a lot; thus probably leading to the high rate of recurrence in the first group, which affected our analysis. Secondly, the level of evidence in articles included is low and the general strength of evidence is level IV. Thirdly, although all articles included were selected strictly according to inclusion and exclusion criteria, it was unhomogeneous in surgical techniques, the degree of stiffness in elbows, postoperative rehabilitation protocols, and the follow-up time among studies, thus limiting other specific statistical analysis.

Conclusions

In summary, there was a trend toward a shorter waiting time from injury to open arthrolysis from 12 months to 6 months. The shorter waiting period of 6 to 10 months yielded better results. Therefore, early surgical release of stiff elbows is recommended for a shorter rehabilitation time and earlier return to work. More and larger prospective studies are needed to determine the most beneficial surgical timing.