Abstract
Background:
Both dynamic and static progressive (turnbuckle) splints are used to help stretch a contracted elbow capsule to regain motion after elbow trauma. There are advocates of each method, but no comparative data. This prospective randomized controlled trial tested the null hypothesis that there is no difference in improvement of motion and Disabilities of the Arm, Shoulder and Hand (DASH) scores between static progressive and dynamic splinting.
Methods:
Sixty-six patients with posttraumatic elbow stiffness were enrolled in a prospective randomized trial: thirty-five in the static progressive and thirty-one in the dynamic cohort. Elbow function was measured at enrollment and at three, six, and twelve months later. Patients completed the DASH questionnaire at enrollment and at the six and twelve-month evaluation. Three patients asked to be switched to static progressive splinting. The analysis was done according to intention-to-treat principles and with use of mean imputation for missing data.
Results:
There were no significant differences in flexion arc at any time point. Improvement in the arc of flexion (dynamic versus static) averaged 29° versus 28° at three months (p = 0.87), 40° versus 39° at six months (p = 0.72), and 47° versus 49° at twelve months after splinting was initiated (p = 0.71). The average DASH score (dynamic versus static) was 50 versus 45 points at enrollment (p = 0.52), 32 versus 25 points at six months (p < 0.05), and 28 versus 26 points at twelve months after enrollment (p = 0.61).
Conclusions:
Posttraumatic elbow stiffness can improve with exercises and dynamic or static splinting over a period of six to twelve months, and patience is warranted. There were no significant differences in improvement in motion between static progressive and dynamic splinting protocols, and the choice of splinting method can be determined by the patients and their physicians.
Level of Evidence:
Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.
Static progressive and dynamic elbow splints are often used to help stretch a contracted elbow capsule to regain elbow motion after injury1–4. The static progressive splint uses stepwise increases in joint angle that apply a force to contracted tissues that dissipates as the tissues stretch, while the dynamic splint applies a consistent force to the tissues that is maintained as the tissues stretch and additional motion is achieved. Each splint has advocates, but sparse scientific data are available regarding either splint. The literature on splinting is restricted to case series1–5 and case reports6,7, with, in the case series, reported improvements in flexion and extension arc between 25° and 41° after static progressive splinting and approximately 20° after dynamic splinting.
The primary null hypothesis of this clinical trial was that there is no difference between static progressive and dynamic elbow splinting in improvement of the arc of elbow flexion and extension six months after injury or surgery. Secondary study questions addressed differences in flexion arc, forearm rotation, arm-specific disability, and ulnar nerve dysfunction at each of the evaluations.
Participants
The Human Research Committee at our institution approved our protocol. The registration number with ClinicalTrials.gov is NCT01241916. Our inclusion and exclusion criteria match our routine clinical practice. The inclusion criteria were an age of eighteen years or greater, loss of >30° in flexion or extension after an elbow injury or elbow surgery, and failure to have improvement in elbow motion for at least four weeks with regular stretching exercises after acute injury or any time after reconstructive surgery. Exclusion criteria included patients with active infection, heterotopic ossification restricting motion, wound problems, inability to cooperate with a structured rehabilitation protocol, burn-related contractures, primary osteoarthritis, and a total elbow or interposition arthroplasty (either planned or in place).
Randomization
After informed consent, each patient was randomly assigned to a dynamic or static progressive splint on the basis of a random sequence determined by a computerized random-number generator (Windows Excel; Microsoft, Redmond, Washington). Treatment assignment was concealed until informed consent was obtained.
Intervention
We prescribed dynamic splints manufactured by Dynasplint Systems (Severna Park, Maryland) and static progressive splints manufactured by Joint Active Systems (Effingham, Illinois), but alternative splints within these broad classes were accepted according to the preference of insurers and therapists. Insurance coverage of splint wear was not an issue because, in the rare case of delay or denial, the splint companies would make allowances, and—if needed—our research fund covered the costs of the device. The splint was adjusted to fit the patient’s limb by a representative of the manufacturer. Patients were instructed to use the splint according to the standard daily wearing and use protocol. According to these instructions, the static progressive splint is worn three times per day for a thirty-minute period, whereas the dynamic splint is worn for approximately six to eight continuous hours per day or night. Use of the splints was discontinued at the discretion of the patient or if the patient had reached a plateau in active elbow motion (defined as no measurable gains in active range of motion achieved in a thirty-day period measured with a handheld goniometer).
Evaluation
At all time points, patients were evaluated by an independent research fellow not involved in the care of the patients.
Prior to the intervention, demographic information and injury-related medical history were recorded, elbow and forearm motion was measured with a handheld goniometer, and patients completed the Disabilities of the Arm, Shoulder and Hand (DASH) Questionnaire8. The DASH questionnaire was developed by the American Academy of Orthopaedic Surgeons in collaboration with the Council of Musculoskeletal Specialty Societies and the Institute for Work and Health as an outcomes instrument specific to the upper extremity, and is applicable to a wide variety of upper extremity disorders. The questionnaire contains thirty items: twenty-one evaluate difficulty with specific tasks, five evaluate symptoms, and one each evaluates social function, work function, sleep, and confidence. The score is scaled between 0 and 100 points, with higher scores indicating worse function of the upper extremity.
At three months (between eight and sixteen weeks was acceptable), six months (between five and eight months), and twelve months (between ten and eighteen months) of treatment, elbow motion was measured with a handheld goniometer. At the six and twelve-month evaluation, patients completed the DASH questionnaire again.
Subsequent Surgery
Indications for subsequent surgery included infection, nonunion, heterotopic bone blocking motion, symptoms related to implants, and persistent ulnar neuropathy.
Statistical Analysis
The primary study question addressed improvement in elbow flexion and extension arc at the six-month evaluation. A power analysis indicated that a total sample size of forty-eight patients (twenty-four patients in each cohort) would provide 80% power (β = 0.20, α = 0.05) to detect a difference of 10° in improved flexion arc. To account for a possible loss to follow-up of 20% to 25%, we anticipated enrolling thirty patients in each cohort.
Continuous variables (flexion arc, flexion, flexion contracture, forearm rotation, pronation, supination, DASH scores, time between injury or the most recent surgery and enrollment, and number of additional surgeries) were compared between cohorts with use of the Student t test. Differences in dichotomous variables (sex; limb dominance; an occupation involving manual labor versus one involving nonmanual labor; distal humeral fracture versus other injury; elbow fracture-dislocation versus other injury; treatment [original trauma or capsular release for stiffness]; ulnar neuropathy; presence of ipsilateral injury; and completion of the three, six, and twelve-month follow-up) were compared between cohorts with use of the Fisher exact test. In addition, using Spearman correlation, we analyzed the association between the arc of elbow flexion and extension at enrollment and at the six and twelve-month evaluations and the DASH scores at each of these evaluation points, respectively.
This is a so-called pragmatic trial9,10 that compared what happens when patients are prescribed specific splints, no matter how often those splints are used. Pragmatic trials are designed to measure effectiveness (the benefit of treatment in routine clinical practice) rather than efficacy (the benefit of treatment under ideal circumstances). In contrast to explanatory trials, in which compliance with prescriptions is incentivized and tracked, pragmatic trials reflect what happens to the average patient in routine practice when he or she receives a prescription. Pragmatic trials may provide outcomes that are more clinically relevant. Accordingly, patients were analyzed according to an intention-to-treat principle—patients who requested a change to the alternative splinting device or had surgery during the study period were analyzed according to their randomized cohort assignment. Missing data were handled with use of mean imputation as described by Herman et al.11, with the mean of the treatment cohort used as the replacement for the missing observation.
Source of Funding
This study was supported by a grant from Joint Active Systems (Effingham, Illinois). We sought balanced funding from Dynasplint Systems (Severna Park, Maryland), but they declined.
Recruitment and Participant Flow
Between October 2003 and May 2008, all sixty-six patients who were invited enrolled in the trial: thirty-one patients were assigned to the dynamic splinting cohort and thirty-five to the static progressive splinting cohort. There were three protocol violations in which the patients were enrolled earlier than four weeks after injury. Fifty-nine patients completed the three-month evaluation (twenty-nine [94%] in the dynamic cohort and thirty [86%] in the static progressive cohort), fifty-two completed the six-month evaluation (twenty-four [77%] in the dynamic cohort and twenty-eight [80%] in the static progressive cohort), and forty-nine patients completed the twelve-month evaluation (twenty-one [68%] in the dynamic cohort and twenty-eight [80%] in the static progressive cohort).
Three-month follow-up evaluations for three patients (two in the static splinting cohort and one in the dynamic splinting cohort), six-month follow-up evaluations for nine patients (four in the static progressive splinting cohort and five in the dynamic splinting cohort), and twelve-month follow-up evaluations for six patients (five in the static progressive splinting cohort and one in the dynamic splinting cohort) occurred just outside the acceptable range for these evaluations, but the data were used as it was thought to be more accurate than mean imputation.
There were no differences in missed follow-ups among cohorts at three (p = 0.43), six (p = 1.00), and twelve (p = 0.28) months.
Baseline Data
Prior to the initiation of splint use, there were no differences between the dynamic and static progressive splinting cohorts in the type of treatment prior to enrollment (treatment of the acute trauma versus subsequent surgery for stiffness or nonunion), presence of an injury of the ipsilateral upper extremity, number of additional surgical procedures, time between initiation of splint use and closed reduction or last surgery for treatment of the initial trauma or the most recent surgery for release or nonunion, occupation involving manual labor versus nonmanual labor, type of injury, limb dominance, sex, and age (see Appendix).
Although it was not mandated in the trial design, all patients in the dynamic splinting cohort used a splint manufactured by Dynasplint Systems, and patients in the static progressive splinting cohort used a splint from Joint Active Systems.
Subsequent Treatment
Three patients (10%) in the dynamic splinting cohort requested a change to a static progressive splint prior to the three-month evaluation because of discomfort and pain using the dynamic splint.
Two patients in the static progressive splinting cohort had an elbow infection that required irrigation and debridement with partial removal of implants at two and three weeks after enrollment.
Between the three and six-month evaluation, one patient in the dynamic splinting cohort who had changed to a static progressive splint had surgery to remove heterotopic bone that blocked elbow flexion. In the static progressive splinting cohort, there were five patients who had a total of eight additional surgical procedures during this time interval: three patients had a contracture release with excision of heterotopic bone that interfered with motion, one patient had an ulnar nerve release, and one patient had four surgical procedures for an infected nonunion of a fracture.
Between the six and twelve-month evaluations, two patients in the dynamic splinting cohort had an elbow contracture release, with excision of heterotopic bone and an ulnar nerve release in one and with excision of a proximal radioulnar synostosis in the other. In the static progressive splinting cohort, five patients had additional elbow surgery between six and twelve months: three patients had a contracture release, with excision of heterotopic bone in one patient, release of the ulnar nerve in one patient, and with exploration of a suspected ulnar nonunion (that was found to be healed) and capsular release in the third patient; one patient had surgery for symptoms related to implants; and one patient had a release of a stiff long finger during this period. One patient had an ununited fracture of the radial head after plate and screw fixation, but declined additional surgery.
There were no significant differences in the number of subsequent surgical procedures between patients in the dynamic versus static progressive splinting cohort at the three (p = 0.18), six (p = 0.14), and twelve-month (p = 0.96) evaluations.
Outcomes
There were no differences in the average arc of elbow flexion and extension and the improvement in the average flexion and extension arc between the two cohorts at any evaluation time (Table I and Fig. 1). A so-called functional arc (≥130° of flexion and ≤30° of a block to extension) was seen in six patients in the static progressive splinting cohort and in four patients in the dynamic splinting cohort at the time of the three-month follow-up; in eight and ten patients, respectively, at six months; and in nineteen and nine patients, respectively, at twelve months. At enrollment and at the twelve-month follow-up evaluation, there were no differences in the average DASH scores between cohorts. At six months, the patients in the static progressive splinting cohort had a better average DASH score (p < 0.05) (Fig. 2). There was not a significant correlation between the average arc of elbow flexion and extension with the average DASH score at enrollment (r = −0.220, p = 0.08) and at six months (r = −0.217, p = 0.08). There was a significant association between the average elbow flexion and extension arc and the average DASH score at twelve months (r = −0.421, p < 0.001).
In this study, there were no differences with regard to improvement in motion between dynamic and static progressive splinting (an average improvement of 40° with dynamic splinting and 39° with static progressive splinting at six months and 47° versus 49°, respectively, at twelve months). These results compare favorably with, or are better than, results reported in case series on static progressive splinting, which had average improvements from 25° to 41°, and 20° with dynamic splinting1–5. The type of splint used had no consistent influence on measures of function or disability. Although the major gains in elbow motion were made during the first few months after trauma or surgery, patients continued to have improvement until six to twelve months after the injury or surgery, indicating that patience and persistence with nonoperative treatment of posttraumatic elbow stiffness are warranted.
Among the sixty-six patients enrolled in the trial, none had subsequent surgery for capsular release alone. Additional surgery was indicated for other conditions including heterotopic bone that blocked motion (which had developed subsequent to enrollment), fracture nonunion, ulnar neuropathy, infection, or prominent hardware. If no heterotopic bone, implant, or malunion is interfering with elbow motion, posttraumatic elbow stiffness appears to respond to stretching exercises over time and it is unusual for a patient with capsular contracture alone to elect operative treatment. This is an interesting secondary finding of the investigation that merits additional study.
Three patients who were prescribed a dynamic splint requested a change to a static progressive splint because of pain and discomfort with splint use. This is consistent, albeit weak, support for the opinion of some authorities that wearing a dynamic splint is more demanding and uncomfortable than wearing a static progressive splint12.
Our clinical experience and preliminary research from other studies suggest that regaining elbow motion after injury is related in part to comfort, confidence, optimism, and self-efficacy in the performance of elbow stretching exercises that, because they are uncomfortable, often trigger a protective response that can hinder effective stretching. The findings of the current study that elbow motion can improve for more than six to twelve months, and patients with capsular contracture alone uncommonly request surgery when given time to stretch their elbows back to a range that is functional for them, support the concept that a contracted elbow capsule can be stretched if patients are given the appropriate time and support to develop the proper mindset for effective stretching.
This hypothesis merits additional study. This possibility is particularly important because health providers trying to help patients with stretching exercises often caution them “to work to pain but not beyond,” or “to be wary of experiencing too much pain as it can cause inflammation,” both of which are debatable concepts that may reinforce a maladaptive, overprotective mindset. In addition, the time threshold for offering treatment for elbow stiffness appears to have shortened from six to three months after injury, both of which are relatively arbitrary and seem to overemphasize the role of technology and surgical treatment when what is needed is to encourage the patient’s active role in his or her recovery to optimize elbow stretching exercises. This is conjecture and hypothesis generation for the next set of studies.
The present study should be interpreted in light of its shortcomings, including evaluation of fewer than 80% of the enrolled patients at planned follow-up times, which downgrades this randomized trial to Level II, according to the Oxford Centre for Evidence-Based Medicine Levels of Evidence for therapeutic studies13. The number of patients lost to follow-up is typical for investigations in patients with musculoskeletal injury, who are notoriously difficult to maintain in the study protocol12. There were no significant differences in the number of patients lost to follow-up between the two cohorts. Other limitations were the protocol violations in which enrollment and evaluation points were slightly outside the planned range of acceptable times. It was decided to use the data from these time points rather than replacing them with use of mean imputation as the original data were thought to be more representative; however, we did not plan this prior to initiating the trial and it represents an opportunity for the introduction of bias. We recommend that investigators performing prospective trials in patients with musculoskeletal trauma plan how they will handle data gathered outside acceptable follow-up times prior to initiating the trial. Another shortcoming is the lack of a control group of patients who did not use any type of splint to assist with elbow stretching exercises. Therefore, we cannot be sure that either splint was better than self-assisted elbow stretching exercises. In addition, the inclusion criteria and timing of the splint prescription may be specific to our practice. Furthermore, we evaluated patients with the DASH questionnaire only—a general but widely used arm-specific disability measure—rather than an elbow-specific measure. On the other hand, evidence suggests that general measures correlate strongly with more specific measures such as the Patient-Rated Elbow Evaluation (r = 0.89)14. Use of the more widely used DASH instrument has many advantages. It would be interesting to document satisfaction with the splinting devices in future studies—something we did not record. Finally, this study is a so-called pragmatic trial9,10. This is not thought to be a limitation. The advantage of a pragmatic trial over an explanatory trial is that it measures the benefit of treatment in routine clinical practice rather than under ideal circumstances. The outcomes reflect daily practice and are therefore considered more clinically relevant. On the other hand, some may argue that if there would be a difference in outcomes between the two splints under ideal circumstances, a physician may be more rigorous in his efforts to improve patients’ compliance with splint use.
In our opinion, this study supports the following concepts: (1) There is no difference in the improvement in motion with dynamic and static progressive splinting as tools to assist stretching exercises for posttraumatic elbow stiffness, (2) patients with pure capsular contracture do not generally request surgical treatment, and (3) meaningful gains in elbow motion can occur more than six to twelve months after injury or surgery. Future studies should address the role of a splint compared with active, self-assisted stretching exercises alone as well as the impact of the patient’s mindset on motion, function, and disability after elbow injury. On the basis of the best available scientific evidence, patients who elect to use a splint as a tool to assist with elbow stretching can select either a dynamic or a static progressive splint.
A table showing baseline data is available with the online version of this article as a data supplement at jbjs.org.
Gelinas
JJ;
Faber
KJ;
Patterson
SD;
King
GJ. The effectiveness of turnbuckle splinting for elbow contractures. J Bone Joint Surg Br.
2000;82:74-8.
Green
DP;
McCoy
H. Turnbuckle orthotic correction of elbow-flexion contractures after acute injuries. J Bone Joint Surg Am.
1979;61:1092-5.
Bonutti
PM;
Windau
JE;
Ables
BA;
Miller
BG. Static progressive stretch to reestablish elbow range of motion. Clin Orthop Relat Res.
1994;303:128-34.
Doornberg
JN;
Ring
D;
Jupiter
JB. Static progressive splinting for posttraumatic elbow stiffness. J Orthop Trauma.
2006;20:400-4.
Shewring
DJ;
Beaudet
M;
Carvell
JE. Reversed dynamic slings: results of use in the treatment of post-traumatic flexion contractures of the elbow. Injury.
1991;22:400-2.
Dickson
RA. Reversed dynamic slings. A new concept in the treatment of post-traumatic elbow flexion contractures. Injury.
1976;8:35-8.
Richard
R;
Shanesy
CP
3rd;
Miller
SF. Dynamic versus static splints: a prospective case for sustained stress. J Burn Care Rehabil.
1995;16(
3 Pt 1):284-7.
Hudak
PL;
Amadio
PC;
Bombardier
C. Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand) [corrected]. The Upper Extremity Collaborative Group (UECG) Am J Ind Med.
1996;29:602-8.
Roland
M;
Torgerson
DJ. What are pragmatic trials?BMJ.
1998;316:285.
Manchikanti
L;
Hirsch
JA;
Smith
HS. Evidence-based medicine, systematic reviews, and guidelines in interventional pain management: Part 2: Randomized controlled trials. Pain Physician.
2008;11:717-73.
Herman
A;
Botser
IB;
Tenenbaum
S;
Chechick
A. Intention-to-treat analysis and accounting for missing data in orthopaedic randomized clinical trials. J Bone Joint Surg Am.
2009;91:2137-43.
O’Driscoll
SW;
Giori
NJ. Continuous passive motion (CPM): theory and principles of clinical application. J Rehabil Res Dev.
2000;37:179-88.
MacDermid
JC. Outcome evaluation in patients with elbow pathology: issues in instrument development and evaluation. J Hand Ther.
2001;14:105-14.
Disclosure: One or more of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of an aspect of this work. In addition, one or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, one or more of the authors has had another relationship, or has engaged in another activity, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.