The purpose of “Evidence in Practice” is to illustrate the literature search process to obtain evidence that can guide clinical decision making. This article is not a case report. The examination, evaluation, and intervention sections are purposely abbreviated.
A 62-year-old man was referred for outpatient physical therapy 2 weeks after he had fallen and fractured the head of his right radial bone. The fractured radial head was excised the day after the injury, and the patient had been using a sling on his right arm for comfort. The radiology report stated that good bony alignment was observed and that no bony abnormalities were present following surgery. The patient's goal was to return to his job as a courtroom deputy.
The surgeon had referred the patient to me for examination, evaluation, and intervention, noting the need to increase range of motion in the right elbow, but he had specifically written “no passive range of motion exercise” on the consultation form. Upon initial examination, range of motion of the right elbow, as measured with a standard goniometer, as described by Armstrong et al,1 was 75 degrees to 115 degrees of flexion, 10 degrees of supination, and 20 degrees of pronation. Because of the surgeon's note, I was concerned that the absence of interventions designed to increase the range of motion of the right elbow could delay the recovery of function, and I was uncertain why he made this request. I therefore contacted the surgeon to discuss his rationale. He said that he was concerned that passive range of motion following a traumatic injury could lead to heterotopic ossification (the appearance of ectopic bone in the soft tissues around the elbow). I proceeded with the initial physical therapist examination of the patient and the creation of an active exercise program, but I made a mental note to find more information on the relationship between passive exercise and heterotopic ossification.
An orthopedic medicine textbook that I consulted for more information on the relationship between heterotopic ossification and traumatic injuries stated, “Passive stretching or manipulation of the related joint is contraindicated, because it tears more muscle fibers and aggravates the entire process.”2(p484) The textbook author, however, did not provide a reference or evidence for this statement. I learned from another textbook that development of heterotopic ossification is also common following burn injuries, traumatic head injuries, and spinal cord injuries in addition to occurring after traumatic injuries,3 but I did not find an answer to my clinical question. I therefore decided to search for primary, peer-reviewed literature to determine if passive range of motion exercise is contraindicated or if it can be used safely with patients following traumatic injuries.
Database used for search: MEDLINE
I decided to begin my search by accessing www.ncbi.nlm.nih.gov/PubMed, the public, online version of MEDLINE. I chose PubMed because it is available for free and because of the likelihood that I would find information related to my question among the more than 11 million citations from more than 4,600 biomedical journals in the database. The searches were conducted on February 27, 2003.
Keywords for initial search: heterotopic ossification AND passive exercise
I began the search by typing heterotopic ossification in the query box on the main search screen to see how many articles on this condition were available. The search produced 4,257 citations. To narrow my search to the relationship between the intervention and the condition, I added the phrase AND passive range of motion to my initial search term. I capitalized the Boolean term “AND” in order to find articles that were related to both heterotopic ossification and passive range of motion. This search resulted in the citations to 136 articles. I scanned the titles and abstracts of these articles, and I observed that the majority were related to the measurement of joint range of motion and were not related to intervention. I decided to modify the search to by changing passive range of motion to passive exercise. I therefore typed heterotopic ossification AND passive exercise into the query box to narrow the search to articles where passive range of motion was an intervention. The citations to the 3 articles produced by this search are presented in Figure 1.
After reading the abstract of the first citation (Yang et al), a case series of 7 patients, I was encouraged that the authors reported promising results with the use of passive and active exercises following the surgical excision of heterotopic ossification in the elbow region. The subjects in this study, however, were people who already had heterotopic ossification, whereas my patient did not currently have the condition but was at risk for its development. I therefore decided not to obtain the full text of this article.
The title of the second citation (Crawford et al) appeared to relate directly to my clinical question. I decided to obtain a full-text copy of this article from a medical library. The citation and abstract are reproduced below.
Crawford CM, Varghese G, Mani MM, Neff JR. Heterotopic ossification: are range of motion exercises contraindicated? J Burn Care Rehabil 1986 Jul-Aug;7(4):323-7
The incidence, time of onset, and role of exercise in the progression of heterotopic ossification were documented retrospectively in burn patients. In 12 of 1,066 patients (1.2%) consecutively admitted to a burn center, the abnormal bone formed posteriorly around the elbow joint. The initial signs were localized joint pain and rapid decrease in range of motion, and the average time of onset was 12 weeks after thermal injury. Prior to the diagnosis of heterotopic ossification, all patients were managed with an exercise program of active and active-assisted movements. In patients who persisted with passive and active-assisted range of motion, especially beyond the range of pain-free movements, the ossification progressed to complete ankylosis and required surgical intervention to remove the heterotopic bone. On the other hand, postoperative patients and patients who followed a program of active exercise within the pain-free range gained excellent range of motion. It was concluded that passive stretching of the periarticular structures during the acute phase of heterotopic bone formation is detrimental to the final outcome.
[© 1986 American Burn Association. Abstract reprinted with permission of Lippincott Williams & Wilkins.]
This study was a retrospective review of the medical records of 1,066 people with burns who were admitted to the authors' medical center over a 10-year period. Although these patients were not similar to my patient, I wanted to see whether there was any information that might help me answer my clinical question. Heterotopic ossification around the elbow was identified in 12 patients. According to the article, the patients who were treated with passive range of motion after the diagnosis of heterotopic ossification took longer to recover than the patients who received only active and active-assisted exercises. However, the article did not analyze the author's observations or investigate the relationship between exercise and the development of heterotopic ossification. This article identified all 1,066 patients as performing active exercises before the diagnosis of heterotopic ossification; therefore, the role of passive range of motion as a possible factor in the development of the abnormal bone formation could not be known.
From reading the entire article, I did become aware of 2 animal studies published in the 1980s that suggested a relationship between “forcible movement” and the development of ossifications in rabbits, but not between the application of “gentle movement” and the development of ossifications.4,5 In the description of these 2 animal studies, it appeared that “forcible movement” referred to motion that went beyond the available range of motion and that the passive motion was applied for 5 minutes per day and the joint was then immobilized the rest of the day.
Returning to the original search results, I read the abstract of the third citation (Stover et al) to see if it could provide additional information on the clinical question.
Stover SL, Hataway CJ, Zeiger HE. Heterotopic ossification in spinal cord-injured patients. Arch Phys Med Rehabil 1975 May;56(5):199-204.
Therapy programs during the development of heterotopic ossification in spinal cord injury range widely from complete rest to aggressive exercise programs. The literature is confusing by its multiplicity of recommendations because the basic etiology and pathophysiology are unknown and because some basic differences exist between traumatic myositis ossificans and heterotopic ossification associated with severe neurological impairments. Prospective heterotopic ossification roentgenographic (x-ray) surveys of the hips, knees, shoulders, and elbows were made on 250 consecutive spinal cord-injured patients. Those who had x-ray evidence of early or immature heterotopic ossification or who developed early clinical signs of possible heterotopic ossification were treated with an aggressive program of passive progressive range of motion exercises. Several patients who developed bilateral heterotopic ossification had one side used as their own control. Follow-up x-ray studies and range of motion evaluation suggest that an aggressive range of motion exercise program is indicated for patients who are developing or have heterotopic ossification. There is no evidence that exercise increases inflammation with subsequent ossification, and it frequently causes a pseudarthrosis, permitting adequate functional range of motion.
[© 1975 American Congress of Rehabilitation Medicine and American Academy of Physical Medicine and Rehabilitation. Abstract reprinted with permission of Elsevier Science.]
Although this study was published in 1975 and investigated people with spinal cord injuries, it appeared to reach the opposite conclusion of Crawford et al and suggested that passive range of motion exercises are “indicated for persons who are developing or have heterotopic ossification.”6 For further information, I obtained a full-text copy of this article.
This study described a consecutive series of 250 people with spinal cord injury. Radiographs were taken on admission to the rehabilitation facility. Follow-up radiographs revealed the development of heterotopic ossification around 141 joints in 76 patients. The authors compared the aggressiveness of the therapy program before the transfer to the rehabilitation center with the development of heterotopic ossification and concluded that there was no relationship between the type of physical therapy administered and the development of the condition. In the article, however, the authors did not provide any data to support this conclusion.
The authors did provide a description of the management of one patient with heterotopic ossification of both shoulders. Passive range of motion exercises were administered to one shoulder, and the other shoulder was treated with rest. The authors reported that the shoulder treated with passive exercise demonstrated a greater gain in motion and no increase in the heterotopic ossification compared with the shoulder that did not receive exercise. The description of the cases in this article appeared to support the use of passive exercise after the development of heterotopic ossification, but it was not clear from the articles reviewed if passive exercise can be used with patients at risk for the development of heterotopic ossification. Therefore, I decided to further revise my search by looking at PubMed's Medical Subject Headings.
Revised search strategy: Medical Subject Headings
According to information on the PubMed Web site, Medical Subject Headings (MeSH) are a controlled vocabulary that the National Library of Medicine uses to index articles. These terms provide a consistent way for MEDLINE (and PubMed) users to retrieve citations that may use different terms to describe the same concepts. Because my searches used keywords that may be used differently in various contexts, I decided to try a search using MeSH terms to see if I could obtain additional information.
Keywords used in the second search: physical therapy techniques AND ossification, heterotopic
The MeSH Database *—accessed from a link on the left side of the main PubMed search screen—allows users to search for MeSH terms. I decided to search for MeSH terms related to physical therapy interventions to see if there were any terms different from those I used in my initial search. This search is illustrated in Figure 2. I typed the term physical therapy in the query box and was presented with a list of MeSH terms related to this topic and their definitions. The second term in this list was Physical Therapy Techniques, which appeared to be a possibility; therefore, I clicked on the link for this term for more information. The MeSH Database then presented subheadings that could be used to narrow the search and identified where the term is included in the MeSH tree, a hierarchical listing of the MeSH vocabulary. The definition included interventions used in the discipline of physical therapy, which appeared appropriate for my purposes because it would include passive range of motion exercises as well as other interventions used by physical therapists. I decided to include this term in the search and clicked on the checkbox next to the term, selected Search Box with AND from the Send dropdown menu, and clicked on the Send to button to include it in the search statement.
I also wanted to see if heterotopic ossification was related to any MeSH terms, so I typed heterotopic ossification into the query box. The MeSH Database suggested Ossification, Heterotopic, which is defined as “the development of bony substance in normally soft structures.” I decided to combine this term with my earlier MeSH term Physical Therapy Techniques using the Boolean term “AND.” I then repeated the procedure I used for Physical Therapy Techniques. This resulted in a statement in the search box (which is located under the query box) that read, “Physical Therapy Techniques” [MESH] AND “Ossification, Heterotopic“[MESH].†
To perform the search, I then clicked on the Search PubMed button. This search resulted in 51 citations, which were presented with the articles most recently added to the MEDLINE database listed first. The results from this search included 2 of the citations obtained in my initial search. The 13 most recent citations from this search (those published since 1997) are presented in Figure 3.
As an alternative to using Physical Therapy Techniques as a MeSH term, I also used the MeSH Database to search for terms related to Passive Exercise. No exact match for the term Passive Exercise was found; however, MeSH terms suggested as possible subject headings included Exercise, Exercise Therapy, and Exercise Movement Techniques. I combined each of these MeSH terms separately with Ossification, Heterotopic to see if any additional studies could be found. These searches resulted in a smaller number of articles than the search using the term Physical Therapy Techniques (Table) and did not produce any additional studies that appeared to be relevant to my question.
To ensure a thorough literature search, I also decided to repeat my searches using another database. I accessed the Cumulative Index of Nursing and Allied Health Literature (CINAHL) through a medical library. I used the same combinations of search terms that I used when searching PubMed and found no additional citations that appeared to answer my clinical question. I therefore decided to return to the search results obtained using Physical Therapy Techniques and Ossification, Heterotopic as subject headings in PubMed to see if these articles provided evidence for my clinical question.
Selection of articles for review:
From the titles and abstracts of articles obtained from the revised search (Fig. 3), citations 4, 7, and 10 appeared to provide additional insight into the clinical question. I therefore attempted to obtain full-text copies of these articles from the medical library. The other articles on the list did not appear to answer the clinical question directly, either because they were exclusively review articles or because they appeared to be primarily related to surgical management. The abstract from citation 4 is printed here.
Linan E, O'Dell MW, Pierce JM. Continuous passive motion in the management of heterotopic ossification in a brain injured patient. Am J Phys Med Rehabil. 2001 Aug;80(8):614-7.
We report a man admitted to inpatient rehabilitation 6 wk after traumatic brain injury, who presented with bilateral knee heterotopic ossification. In addition to conventional physical therapy, we applied a continuous passive motion device during 4 wk increasing the range of motion of the knees. On the basis of the limited current literature and this case, we suggest that the use of continuous passive motion devices for heterotopic ossification may be effective and safe and should be the subject of further study.
[© 2001 Association of Academic Physiatrists. Abstract reprinted with permission of Lippincott Williams & Wilkins.]
This recently published case report described the use of continuous passive motion by a 27-year-old man with heterotopic ossification of both knees. Although this report did not describe the use of passive exercise before the development of the condition—and although causality cannot be inferred from a case report—this article did provide a description of using passive motion up to 210 minutes every day for 1 month as part of the plan of care.
From the discussion section of this article, I learned of another, more recent animal study that assessed the relationship of passive motion and heterotopic ossification in rabbits.7 Published in 1996, this study focused on the use of continuous passive motion devices. No additional development of heterotopic ossification was noted among rabbits that already had ossifications. The findings of this study suggest that it may not be the passive exercise itself that is responsible for the development of the condition, rather it may be how the exercise is administered (eg, gradual compared with forceful) that is responsible.
I then turned to the next article that appeared to be useful in answering my clinical question (citation 7). This article was not available in the medical library, and I did not automatically attempt to obtain a full-text copy from interlibrary loan because it was written in French and, therefore, would not provide a prompt answer to the clinical question. The abstract of this clinical trial was available in English, however, and I decided to read it to see if the article was worth pursuing further. The abstract is reprinted below.
Rafai M, Largab A, Cohen D, Trafeh M. [Pure posterior luxation of the elbow in adults: immobilization or early mobilization. A randomized prospective study of 50 cases] Chir Main. 1999;18(4):272-8.
INTRODUCTION: Pure posterior dislocation of the elbow is frequent in young subjects. The objective of treatment must be to reduce the dislocation and avoid complications, the most frequent being stiffness, but also elbow instability. The objective of this prospective study was to evaluate the functional and anatomical characteristics of two treatment modalities: plaster immobilization and early mobilization. MATERIAL AND METHODS: 50 cases of pure posterior dislocation of the elbow were included in a prospective study and randomized to two groups: Group I: twenty six cases were treated by reduction under general anaesthesia and plaster immobilization for three weeks, followed by rehabilitation. Group II: twenty four cases were treated by reduction under general anaesthesia, followed by early mobilization. RESULTS: We evaluated our results in terms of loss of amplitude of elbow movement (particularly extension), stiffness, instability, relapses, pain and ossification. This study demonstrated better recovery of elbow function in patients treated by early mobilization: 96% of good results with recovery of normal extension in group II versus 81% of cases in group I. Stiffness was observed in 19% of patients in group I versus 4% in group II; this difference was very significant. Comparison of pain revealed no significant difference and no relapses, instability or ossifications were observed in either of the two groups. DISCUSSION AND CONCLUSION: Early mobilization is superior to plaster immobilization, as it allows recovery of better quality elbow function without inducing instability or recurrence.
[© 1999 Éditions scientifiques et médicales Elseiver SAS. Abstract reprinted with permission of Éditions scientifiques et médicales Elseiver.]
The abstract reported that subjects who were treated with mobilization following subluxation injuries of the elbow achieved greater improvements in the measured outcomes than the subjects treated with immobilization. No cases of development of heterotopic ossification were reported in any of the 50 subjects in the mobilization and immobilization groups. Unfortunately, the abstract did not describe the mobilization intervention, and it is not known if the authors' use of this term included passive techniques. The frequency, duration, and intensity of the mobilization were not provided. In addition, the abstract did not provide information on the subjects other than their elbow injuries, making it more difficult to generalize the results to my patient. If I had more time before my patient's next visit, which was scheduled for the next day, I would have attempted to obtain a full-text copy of this article and have it translated to see if it could provide answers to these concerns.
I then turned to the last citation that appeared to be useful in answering my clinical question (citation 10). The abstract is reprinted below.
Ellerin BE, Helfet D, Parikh S, Hotchkiss RN, Levin N, Nisce L, Nori D, Moni J. Current therapy in the management of heterotopic ossification of the elbow: a review with case studies. Am J Phys Med Rehabil. 1999 May-Jun;78(3):259-71.
Heterotopic ossification, or the appearance of ectopic bone in para-articular soft tissues after surgery, immobilization, or trauma, complicates the surgical and physiatric management of injured joints. The chief symptoms of heterotopic ossification are joint and muscle pain and a compromised range of motion. Current therapies for prevention or treatment of heterotopic ossification include surgery, physical therapy, radiation therapy, and medical management. Unlike heterotopic ossification of the hip, heterotopic ossification of the elbow has not been extensively investigated, leaving its optimal management ill-defined. To remedy this deficiency, we review risk factors, clinical anatomy, physical findings, proposed mechanisms, and current practice for treatment and prevention of heterotopic ossification. We then consider and draw conclusions from four cases of elbow injury treated at our institutions (three complicated by heterotopic ossification) in which treatment included surgery, radiation therapy, physical therapy, and medical therapy. We summarize our institutional practices and conclude with a call for a randomized clinical trial to better define optimal management of heterotopic ossification of the elbow.
[© 1999 Association of Academic Physiatrists. Abstract reprinted with permission of Lippincott Williams & Wilkins.]
I obtained a full-text copy of this review article because it included 4 case reports. The review of the literature appeared to be recent and tended to support the belief that development of heterotopic ossification of the elbow is multifactorial; however, because of the nonsystematic nature of the review, I could not be sure that the authors' search was comprehensive. The 4 case reports provided a description of the medical management of the condition. One of the patients used a continuous passive motion device after the development of heterotopic ossification about the elbow, without complications. None of the cases, however, described specific physical therapy interventions prior to the onset of heterotopic ossification.
I did not obtain a clear answer from my literature searches on whether passive range of motion exercises are contraindicated following traumatic injuries. There appeared to be some recent evidence from case reports and animal studies that passive range of motion, performed gradually over a long period of time, helps improve range of motion without worsening the heterotopic ossifications already present. In contrast, other animal studies suggest the administration of passive exercise with high force that exceeds the available range of motion might aggravate the condition. I therefore decided that gentle passive exercise may be considered as an intervention option following a traumatic injury because the evidence suggested that application of passive exercise that was not forceful was associated with a lower risk for the development and progression of heterotopic ossification.
Before the patient's second visit to outpatient physical therapy, I discussed the results of the literature search with the patient's surgeon. The surgeon agreed with my interpretation and agreed with my plan of care that included passive range of motion exercise. I informed the surgeon that I would monitor the patient's joint range of motion and would contact him if range of motion decreased or did not improve because, according to the articles by Crawford et al and Ellerin et al, these are some of the signs of the development of heterotopic ossification.
At the patient's second visit, I informed him of my discussion with the surgeon and the summary of my literature search. I advised the patient that the literature did not provide conclusive evidence that passive exercise is responsible for the development of abnormal bone formation about the elbow following an injury similar to his. The inclusion of passive exercise, however, might assist in the return of elbow range of motion.
The patient appreciated being informed about the risks and benefits associated with this physical therapy intervention. Based on his goal of returning to work as a courtroom deputy as soon as possible, he permitted me to include passive exercise in conjunction with the active exercise he was already performing in his plan of care. We decided to utilize a continuous passive motion machine for elbow flexion and extension during his visits in the clinic, but we decided against obtaining a unit for home use because he was concerned about the set-up of the unit. At the third visit, the patient's wife attended the clinic, so we could instruct her in the performance of passive range of motion exercises for elbow flexion, extension, supination, and pronation and to avoid forceful movements and prolonged stretching at the end range of motion. With the combination of clinic visits and the home exercise, the patient was scheduled to perform over 1 hour of passive exercise daily. At each clinic visit, range of motion was assessed with a goniometer in order to monitor for changes that may suggest the onset of any complications from the passive exercise.
↵* As this article was being prepared for publication, the National Library of Medicine announced that the MeSH Browser (see “Evidence in Practice,” January 2003, p. 68) was being replaced with the MeSH Database. According to a press release by NLM, the new database “provides additional search and display features.” This article reflects the new search procedures for the MeSH Database. For additional information on the features of the new database, see www.nlm.nih.gov/pubs/techbull/ma03/ma03_mesh.html.
↵† The acronym MESH is included in the search and is enclosed with brackets to limit the search to articles with records that have this term in the MeSH field. When words are enclosed in quotation marks, PubMed treats them as a single term and not as separate words.
- Physical Therapy