Evidence in practice.

Clinical question: Is there evidence that bracing could provide adequate stability for a 47-year-old man with a deficient anterior cruciate ligament to resume downhill skiing?

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 47-year-old man was referred by an orthopedic surgeon to our outpatient physical therapy facility for examination and consultation about the possible use of a knee brace. Two months before the referral, the patient twisted his right knee while skiing. The knee became swollen within 2 hours after the injury, and the patient felt that his knee was unstable during ambulation, when descending stairs, and when performing pivoting movements. He saw an orthopedic surgeon who performed a physical examination that yielded a score of 2+ on the Lachman test (indicating greater than 9 mm of excessive anterior tibial translation on the femur (1(p701))) and a negative result on a pivot shift test (indicating no rotational subluxation of the tibia on the femur (1(707-709))). The surgeon also referred the patient for magnetic resonance imaging, which confirmed a partial tear of the right anterior cruciate ligament (ACL). The surgeon discussed the option of reconstruction with the patient, but he indicated that the size of the tear and the functional limitations were not enough to warrant surgery at that time. The patient decided against having the tear surgically repaired because he did not want to have postsurgical pain and disability or make a commitment to the lengthy rehabilitation that would be required.

The patient was initially referred to us to regain full knee range of motion (ROM), increase lower-extremity muscle force, and improve the stability of his knee during activities such as descending stairs and pivoting. Ultimately, the patient wanted to return to recreational skiing with his family and participate in activities, such as baseball, with his young sons. He received physical therapy for 4 weeks, during which he progressed from a program that emphasized active and passive ROM exercises, isometric exercises of quadriceps muscles, and isotonic exercises of the hamstring muscles to a more challenging program designed to improve muscle force and balance. This program included closed kinetic chain strengthening exercises, balance activities on a proprioception board, and continuation of isotonic exercises of the hamstring muscles.

After 1 month of physical therapy intervention, goniometric measures of knee flexion and extension revealed that ROM in the patient's right knee was equal to the unaffected side. A comparison of right and left knee circumferential measures, which were taken with a standard tape measure at the level of the midpatella, indicated no residual swelling. Muscle force of the quadriceps and hamstring muscle groups, as determined by manual muscle testing (MMT) performed using the procedure and grading system described by Hislop and Montgomery, (2) was graded 5/5. MMT is convenient for clinical use because it can be performed quickly and does not require equipment. MMT scores, however, are of questionable reliability (3,4) and lack the sensitivity of instrumented tests. (5,6) Mulroy et al (7) suggested that MMT is most useful for detecting substantial weakness, rather than for identifying small declines in the ability of a large lower-extremity muscle to generate force. We used MMT because instrumented methods, such as hand-held dynamometry and isokinetic dynamometry, were not available to us.

In view of the limitations of MMT, we used functional tests to provide further information about muscle performance. The patient was able to ascend and descend stairs in a step-over-step fashion, pivot, and hop on the right leg. During this testing, however, he reported that he did not feel as stable hopping on his right leg as he did on his left leg. We asked the patient how this injury affected his daily life, and he told us that his knee had "given out" twice when he had been walking and quickly pivoted from his planted, right foot.

The patient returned to the surgeon for a follow-up visit after the 1-month course of physical therapy. At that visit, the surgeon recommended that the patient acquire a functional knee brace (FKB)--a brace designed to improve stability in a patient with an unstable knee due to ligament injury. (8)

The patient returned to our facility for advice about an FKB. His medical insurance would not cover the cost of an FKB, and he wanted to be reasonably certain that he would benefit from wearing this brace before he bought it. We understood his concern and, knowing the difference in price, wondered if a custom-fit knee brace had advantages over an off-the-shelf brace. We decided to search the literature for evidence that bracing improves knee stability during recreational activities and, if so, we wanted to know whether the evidence supported the use of a custom-fit brace rather than an off-the-shelf brace.

* Database used for search: MEDLINE

MEDLINE is the National Library of Medicine's (NLM) comprehensive database of international literature covering every aspect of medicine and health care. MEDLINE contains citations from more than 4,600 biomedical journals. We selected this database because we wanted to access information from medical journals as well as physical therapy and rehabilitation journals. PubMed is an online service of the NLM that provides free public access to MEDLINE. New citations are available in PubMed daily, Tuesday through Saturday. We accessed PubMed through the Internet at www.ncbi.nlm.nih.gov/PubMed. This search was conducted on December 23, 2003.

* Initial search terms: anterior cruciate ligament AND (brace OR bracing OR braces)

To begin our search, we typed the phrase anterior cruciate ligament in the query box. Our first inclination was to use the abbreviation "ACL." After some consideration, we decided to spell out the phrase, because we thought that using the abbreviation might not retrieve some important articles. Boolean operators, such as "and" and "or," must be typed in uppercase in PubMed. In the query box following anterior cruciate ligament, we typed AND followed by (brace OR bracing OR braces). We decided that using 3 forms of the word "brace" might retrieve relevant articles that would otherwise be missed if we used only the word "brace." Using the "OR" operator retrieves documents that contain at least one of the specified search terms. The parentheses instructed PubMed to process the search terms inside the parentheses as a unit and then to incorporate those results into the overall search. We clicked the Go button located to the right of the query box. This search retrieved 174 citations. To narrow the number of citations, we decided to set limits on the type of article to be included in the search results.

* Limits: English, Human, Publication Date: 1997-2003

To set limits on our search, we clicked on the Limits link on the Features Bar, located beneath the query box, which displayed a number of options (Fig. 1, page 275). PubMed contains citations published as far back as 1966. We decided to limit our search to articles published in the last 7 years, because we were concerned that braces used in older studies might not be representative of braces manufactured more recently. To the right of the Publication Date/Entrez Date pulldown menu, we typed 1997 in the first box following From and 2003 in the first box following To. We decided to leave the boxes for month and day blank. MEDLINE indexes journals published in approximately 40 languages. We selected English from the Languages pulldown menu to exclude articles published in other languages. We selected Human study group from the Human or Animal pulldown menu because we did not expect animal studies to be helpful in answering our clinical question. We then clicked on the Go button. These actions resulted in 72 citations.

[FIGURE 1 OMITTED]

We wanted to reduce the number of citations further and discussed adding the word "deficiency" to our search. After some consideration, we realized that doing so might exclude pertinent articles using phrases such as "anterior cruciate ligament insufficiency" or "anterior cruciate ligament laxity" instead of "deficiency." As we browsed the article titles in the search results, a new search strategy occurred to us. We noticed that many articles dealt with bracing in people who had undergone ACL reconstruction. We did not feel those studies would be applicable to our patient's bracing decision. Accordingly, we added NOT reconstruction to the initial search terms in the query box, expecting that using the Boolean operator NOT would exclude articles dealing with ACL reconstruction. Our search string was: anterior cruciate ligament AND (brace OR bracing OR braces) NOT reconstruction. This search yielded 28 citations, which seemed to be a reasonable number to review. Figure 2 lists the bibliographic information for these articles.

Figure 2. Citations Retrieved by Search of MEDLINE Using
Keywords "Anterior Cruciate Ligament," "Brace," "Bracing,"
"Braces," and "NOT Reconstruction" and Limited to "English,"
"Human," and Publication Dates 1997-2003.

1. Smith J, Malanga GA, Yu B, An KN. Effects of functional
knee bracing on muscle-firing patterns about the
chronic anterior cruciate ligament-deficient knee. Arch
Phys Med Rehabil. 2003;84:1680-1686.

2. Mallory N, Kelsberg G, Ketchell D. Does a knee brace
decrease recurrent ACL injuries? J Fam Pract. 2003;52:
803-304.

3. Decoster LC, Vailas JC. Functional anterior cruciate
ligament bracing: a survey of current brace prescription
patterns. Orthopedics. 2003;26:701-706.

4. Oakes DA, McAllilster DR. Failure of heat shrinkage
for treatment of a posterior cruciate ligament tear.
Arthroscopy. 2003;19:E1-E4.

5. Kocher MS, Sterett WI, Briggs KK, et al. Effect of functional
bracing on subsequent knee injury in ACL-deficient
professional skiers. J Knee Surg. 2003;16:87-92.

6. Uchio Y, Ochi M, Adachi N, et al. Results of rasping
of meniscal tears with and without anterior cruciate ligament
injury as evaluated by second-look arthroscopy.
Arthroscopy. 2003;19:463-469.

7. Grainger ML, Wilson CH, Augustine SJ, Griffiths HJ.
Radiologic case study. Traumatic dislocation of the knee.
Orthopedics. 2003;26:280, 351-354.

8. Indelli PF, Dillingham MF, Fanton GS, Schurman DJ.
Monopolar thermal treatment of symptomatic anterior
cruciate ligament instability. Clin Orthop.
2003;407:139-147.

9. Beynnon BD, Fleming BC, Churchill DL, Brown D. The
effect of anterior cruciate ligament deficiency and functional
bracing on translation of the tibia relative to the
femur during nonweightbearing and weightbearing. Am
J Sports Med. 2003;31:99-105.

10. Ramsey DK, Wretenberg PF, Lamontagne M, Nemeth
G. Electromyographic and biomechanic analysis of anterior
Cruciate ligament deficiency and functional knee
Bracing. Clin Biomech (Bristol, Avon). 2003;18:28-34.

11. van Glabbeek F, van Riet R, Jansen N, et al. Arthroscopically
assisted reduction and internal fixation of tibial
plateau fractures: report of twenty cases. Acta Orthop
Belg. 2002;68:258-64.

12. Lam RY, Ng GY, Chien EP. Does wearing a functional
knee brace affect hamstring reflex time in subjects with
anterior cruciate ligament deficiency during muscle
fatigue? Arch Phys Med Rehabil. 2002;83:1009-1012.

13. Thomson LC, Handoll HH, Cunningham A, Shaw PC.
Physiotherapist-led programmes and interventions for
Rehabilitation of anterior cruciate ligament, medial collateral
Ligament and meniscal injuries of the knee in adults.
Cochrane Database Syst Rev. 2002;(2):CD001354.

14. Solomonov M. Specific functional knee brace produces
"only minor kinematic changes in tibiofemoral joint
motion" of patients with anterior cruciate ligament deficiency.
Clin Biomech (Bristol, Avon). 2001;16:631-633.

15. Niitsu M. Kinematic MR imaging of the knee. Semin
Muskuloskelet Radiol. 2001;5:153-157.

16. Wojtys EM, Huston LJ. "Custom-fit" versus "off-the-shelf"
ACL functional braces. Am J Knee Surg.
2001;14:157-162.

17. Ellen MI, Gilhool JJ, Rogers DP. Nonoperative treatment
of an interosseous ganglion cyst. Am J Phys Med
Rehabil. 2001;80:536-539.

18. Solomonow M, Krogsgaard M. Sensorimotor control
of knee stability: a review. Scand J Med Sci Sports.
2001;11:64-80.

19. Ramsey DK, Lamontagne M, Wretenberg PF, et al.
Assessment of functional knee bracing: an in vivo three-dimensional
Kinematic analysis of the anterior cruciate
Deficient knee. Clin Biomech. 2001;16:61-70.

20. Gotlin RS, Huie G. Anterior cruciate ligament
injuries: operative and rehabilitative options. Phys Med
Rehabil Clin N Am. 2000;11:895-928.

21. Kirkley A, Webster-Bogaert S, Litchfield R, et al. The
effect of bracing on varus gonarthrosis. J Bone Joint
Surg Am. 1999;81:539-548.

22. Beynnon BD, Ryder SH, Konradsen L, et al. The effect
of anterior cruciate ligament trauma and bracing on knee
proprioception. Am J Sports Med. 1999;27:150-155.

23. Scoggin JF 3rd. Commom sports injuries seen by the
primary care physician, part II: lower extremity. Hawaii
Med J. 1998;57:502-505.

24. Wilson LQ, Weltman JY, Martin DE, Weltman A.
Effects of a functional knee brace for ACL insufficiency
during treadmill running. Med Sci Sports Exerc.
1998;30:655-664.

25. Warming T, Jorgensen U. The effect of bracing on
extension strength in patients with ACL insufficiency.
Scand J Med Sci Sports. 1998;8-14-19.

26. Nemeth G, Lamontagne M, Tho KS, Eriksson E. Electromyographic
activity in expert downhill skiers using
functional knee braces after anterior cruciate ligament
injures. Am J Sports Med. 1997;25:635-641.

27. Beynnon BD, Johnson RJ, Fleming BC, et al. The
effect of functional knee bracing on the anterior cruciate
ligament in the weightbearing and nonweightbearing
knee. Am J Sports Med. 1997; 25:353-359.

28. Kramer JF, Dubowitz T, Fowler P, et al. Functional
knee braces and dynamic performance: a review. Clin J
Sport Med. 1997;7:32-39.

* Selection of articles for review: We were most interested in articles reporting on subjects with ACL deficiencies who had been tested under conditions similar to those our patient would experience during his recreational activities. Because we needed to make a recommendation to our patient quickly, we decided to limit the number of articles that we selected. As we browsed the titles, we immediately eliminated 12 articles (Fig. 2: citations 4, 6-8, 11, 13, 15, 17, 18, 20, 21, 23) whose focus was not bracing and ACL insufficiency, 1 citation (3) that reported a survey of brace prescription patterns by physicians, and 1 citation (14) that was a letter to the editor. Citation 2 looked interesting, so we clicked on the yellow rectangular icon to the left of the citation (which indicated that the citation did not include an abstract) to see if we could gain any more information about this paper. The next screen provided a link to the journal article. We started to read the article, but noted that the author was addressing the use of knee braces in patients who had undergone ACL reconstruction. Because our patient had an ACL-deficient knee that was not surgically repaired, we decided against using this paper.

It was difficult to determine the relevance of the remaining articles from their titles. The clickable icon (a small yellow rectangle that, unlike the icon described in the previous paragraph, had 3 black horizontal lines near its top) displayed next to the 13 remaining citations informed us that PubMed provided access to an abstract for each article. We read those abstracts in order to identify the articles most relevant to our clinical question. We selected 3 articles, citations 5, 26, and 28, because the abstracts described test conditions that were either specific to skiing or similar to conditions that our patient would encounter when playing baseball. In addition, we wanted to read the full text of citation 16, because we felt it could be helpful in deciding between a custom-fit and an off-the-shelf brace, a decision that would have to be made if our patient chose to purchase a brace.

We decided not to pursue citations 10, 12, 19, 22, and 25 because testing conditions were either static, unlike those our patient would encounter, or were not specified in the abstract. We decided not to read the full text of citation 24 because that article measured physiological parameters in subjects who performed treadmill testing to exhaustion, conditions more intense than those under which our patient would be performing. We eliminated citation 27 because the subjects studied did not have ACL injuries.

Citations 1 and 9 looked somewhat interesting to us. After reading the abstracts, however, we decided against reading the entire articles until we reviewed the 4 studies that we mentioned previously. Citation 1 explored the lower-extremity muscle-firing patterns of subjects with ACL-deficient knees when the subjects performed single-leg hop maneuvers. Nemeth et al (citation 26) also examined the lower-extremity muscle-firing patterns of subjects with unstable knees. The study by Nemeth et al was done while subjects were skiing; therefore, we felt that this article would have greater relevance to our patient than citation 1. The testing conditions outlined in the article by Beynnon et al (citation 9) were not entirely clear. The abstract described testing subjects in non-weightbearing and weight-bearing situations, but did not seem to incorporate the more dynamic conditions involving knee rotation that our patient would encounter while skiing and playing baseball. Because we wanted to make a recommendation to our patient relatively quickly, we decided initially to review articles 5, 16, 26, and 28. If the information from these studies was insufficient for making a recommendation to our patient, then we would consider reading citations 1 and 9 for additional evidence to support a clinical decision.

One of the 4 articles that we selected, citation 26, was available through our institution's full-text electronic journal holdings, so we were able to retrieve a copy of that article immediately. We ordered the other 3 articles, citations 5, 16, and 28, through the interlibrary loan services at our institution. We received a copy of each article within 7 business days of our request. Abstracts and discussion of the 4 articles are presented below.

Nemeth G, Lamontagne M, Kam ST, Eriksson E. Electromyographic activity in expert downhill skiers using functional knee braces after anterior cruciate ligament injuries. Am J Sports Med. 1997;25:635-41.

We studied six expert downhill skiers who had sustained anterior cruciate ligament injuries and had different degrees of knee instability. The aim was to measure possible changes in electromyographic activity recorded from lower extremity muscles during downhill skiing in a slalom course without and with a custom-fit brace applied to the injured knee. Surface electrodes were used with an eight-channel telemetric electromyographic system to collect recordings from the vastus medialis, biceps femoris, semimembranosus, semitendinosus, and gastrocnemius medialis muscles from both legs. Without the brace, the eletromyographic activity level of all muscles increased during knee flexion. The biceps femoris muscle was the most activated and reached 50% to 75% of the maximal peak amplitude. With the brace, the electromyographic activity increased in midphase during the upward push for the weight transfer and the peak activity occurred closer to knee flexion in midphase. Also, the uninjured knee was influenced by the brace on the injured leg, a decrease in electromygraphic activity was seen in midphase. Spearman's rank correlation revealed a significant correlation between an increase in biceps femoris activity of the injured leg and increasing knee instability. We suggest that the brace caused an increased afferent input from proprioceptors, resulting in an adaptation of motor control patterns secondarily modifying electromyographic activity and timing.

[[c] 1997 American Orthopaedic Society for Sports Medicine. Abstract reprinted with permission of Sage Publications Inc.]

This article interested us because it studied effects of bracing during skiing, and one of our patient's goals was to return to skiing. Six expert downhill skiers, between 23 and 46 years of age, who sustained an ACL injury participated in this study. Three of the subjects had undergone ACL reconstruction. Lachman test results were positive in 5 of the 6 subjects, and pivot shift test results were positive in 1 of the 6 subjects. Our patient had a score of 2+ on the Lachman test, and results of the pivot shift test were negative. Although he was not an expert skier, our patient's condition was similar to that of some of the participants in this study.

The brace used during testing was a custom-fit Defiance brace. * Electromyographic (EMG) activity of 4 lower-extremity muscle groups was recorded during 2 runs down a slalom course with the brace and 2 runs without the brace. The authors did not state whether the order of test conditions was randomized. Analysis of variance and planned contrasts showed no differences in EMG activity in the injured leg with and without the brace. The authors presented a figure depicting the recorded EMG activity. Referring to that figure, the authors claimed that changes in muscle timing and coordination did occur with the brace but that they were not revealed by the statistical analysis. The authors suggested that the altered patterns of muscle activity could contribute to increased knee stability. This conclusion should be treated with some skepticism, however, given that the results of the data analysis did not support this claim.

Interestingly, all subjects in this study reported feeling more stable and confident when wearing the brace. The researchers suggested that brace wearers demonstrated beneficial motor control patterns in response to afferent input from knee proprioceptors and from the brace-skin-bone interface to the central nervous system. Given that this study was limited to 6 subjects (which therefore limited the generalizability of the results to our patient) and given that the results were not statistically significant, we wanted further information to determine if bracing our patient's knee might enhance his function.

The second article (Kocher et al) we read was another study involving skiers with ACL-deficient knees. Kocher et al studied 180 subjects, considerably more than Nemeth et al. We hoped that this larger number of subjects would strengthen the conclusions that Kocher et al discussed.

Kocher MS, Sterett WI, Briggs KK, Zurakowski D, Steadman JR. Effect of functional bracing on subsequent knee injury in ACL-deficient professional skiers. J Knee Surg. 2003 Apr;16(2):87-92.

The effect of functional bracing on subsequent knee injury in anterior cruciate ligament (ACL) deficient professional skiers was evaluated. A cohort of 180 ACL-deficient skiers was identified from a knee screening of 9410 professional skiers from 1991-1997. An ACL-deficient knee was defined by an abnormal examination (Lachman or pivot-shift) and a [greater than or equal to] 5-mm KT-1000 manual maximum difference. The dependent variable was subsequent knee injury, which occurred in 12 knees. A significantly higher proportion of injuries occurred in nonbraced skiers compared with braced skiers (P=.005). The risk ratio for subsequent knee injury comparing nonbraced with braced skiers was 6.4 (13% and 2%, respectively). Univariate analysis revealed no significant effects of the other covariates. Logistic regression identified bracing status (P<.01; odds ratio=8) and KT-1000 manual maximum difference (P=.02; odds ratio=1.3) as significant multivariate risk factors for subsequent knee injury, controlling for covariates.

[[c] 2003 Slack Inc. Abstract reprinted with permission of Slack Inc.]

This study took place from 1991 to 1997 at a major ski resort where a total of 9,410 employees who were professional skiers underwent annual screening exams for their knees. Of the 9,410 subjects screened, 180 skiers had an ACL-deficient knee, as defined in the abstract. Their mean age was 38.6 years. Because our 47-year-old patient had an ACL-deficient knee with abnormal Lachman test results, we felt that he was similar in age and knee pathology to the subjects studied in this article. The study participants skied a minimum of 80 days per season, which was substantially more than our patient expected to ski.

After skiers were determined to have an ACL-deficient knee, the researchers conferred with a physician, who considered individual preference and willingness to wear an FKB in making a bracing decision. Based on physician recommendations, 101 skiers received a custom-fit knee brace, and 79 skiers did not receive a brace.

The braced and unbraced skiers did not differ in age, occupation, or results of Lachman or pivot shift tests. The braced skiers, however demonstrated greater laxity on instrumented testing than nonbraced skiers. Any subsequent knee injury, defined as an injury that prevented the person from working, was traced through workers' compensation claims. The results of the study showed that the skiers who were not braced had a higher proportion of subsequent knee injuries than the braced skiers did. Two of the 101 braced skiers and 10 of the 79 unbraced skiers sustained subsequent knee injuries, which corresponded to a risk of subsequent knee injury that was 6.4 times higher for nonbraced skiers. Despite the fact that instrumented testing indicated greater laxity in the group that was braced, the group that was unbraced had more knee injuries.

Our patient hoped that wearing a brace during skiing would decrease his chances of reinjuring his knee. The results of this article support this notion. Based on their findings, the authors recommended bracing for professional skiers with ACL deficiencies. There were at least 2 reasons why we questioned the appropriateness of this recommendation for our patient. First, our patient would be skiing much less than the professional skiers in the study. A brace may not be indicated for the limited amount of skiing that he planned to do. Second, because the skiers were not randomly assigned to bracing groups, we were concerned that factors other than wearing an FKB could account for the results of this study. Patient preference and willingness to wear a brace played a role in physician recommendations about bracing. This may have created some inequity between the subject groups because those who were not willing to wear a brace could have been greater risk takers. That risk-taking nature may have lead some people to engage in skiing behavior that was more likely to result in knee injury. Because we were concerned about the applicability of the findings to our patient and the manner in which study participants were assigned to bracing groups, we wanted to see if there was more evidence to support the use of an FKB with a person with an ACL-deficient knee in situations that were similar to our patient's recreational activities.

The third article we read was a review of 12 studies that compared performance of subjects with ACL injuries on a range of functional tests under braced and unbraced conditions. This article appealed to us because many of the studies reviewed had relatively large numbers of subjects and considered functional situations similar to those that our patient would experience during his recreational activities.

Kramer JF, Dubowitz T, Fowler P, Schachter C, Birmingham T. Functional knee braces and dynamic performance: a review. Clin J Sport Med. 1997;7:32-9.

Objective: The purpose of the present review was to examine current experimental research on the effectiveness of functional knee braces (FKBs) used by patients with anterior cruciate ligament injury during dynamic performance tests. Data sources: Twelve studies published in peer-reviewed journals and listed in the Excerpta Medica system were reviewed. Study selection: All studies compared braced and unbraced tests performed by the same subjects, using tests characterized by weight bearing/axial loading. Data extraction: Studies were reviewed independently by three investigators. Data synthesis: Tests included one-leg hop, figure-of-eight run, stair climbing, walking, cutting, agility runs, straight running, and bicycle ergometry. Experimental situations were classified as follows: (a) maximal effort tests, which compared overall measures of performance such as the distance hopped and the time to run a specific distance; and (b) matched submaximal effort tests, which compared specific variables such as electromyography, range of motion, ground reaction forces, and energy costs. Bracing was found to be advantageous in three of the 16 maximal effort situations, disadvantageous in two and of no measurable effect in 11. Six of the 10 matched effort situations reported differences in the criterion measurements when braced, while four reported no differences. Conclusions: If FKB prescription is to be based solely on empirical evidence of efficacy from performance tests, then further investigation is required to provide this evidence. Future research needs to examine the subjective and psychological aspects of FKB usage along with the results of objective performance tests.

[[c] 1997 Lippincott Williams & Wilkins. Abstract reprinted with permission of Lippincott Williams & Wilkins.]

Reading this review article informed us about several studies relevant to our clinical question. Initially, we were concerned that this 1997 article, which reviewed papers published between 1982 and 1992, would not adequately represent the performance of subjects using braces that were designed more recently. On the other hand, after reading this article, we appreciated the many factors related to injury, bracing, and surgical and rehabilitative interventions that could change the effectiveness of an FKB. We relied on the summaries in the article, which saved us considerable time and allowed us to make a recommendation to the patient much more quickly than if we had read the full text of each article reviewed.

This article was not, however, a systematic review, which is a "summary of the medical literature that uses explicit methods to systematically search, critically appraise, and synthesize the world literature on a specific issue." (9(p 133)) This paper did not specify the search process for the included articles or the rationale for selection of articles to be reviewed. Because the literature search and selection process were not systematic, the authors may have missed some important articles on the topic, the articles may have been summarized incorrectly, and interpretations of findings may have been subject to the authors' bias. (10) Although the abstract indicates that 3 investigators independently reviewed each study, no further detail is provided about the method used to review each study or extent of agreement among the reviewers. Mindful of the limitations of this paper, we decided that the advantages of reading this literature review outweighed the disadvantages.

In reading the review article, we found that bracing was reported to provide some benefit in several experimental situations. Mishra et al (11) examined the effectiveness of 4 custom-fit FKBs in 42 patients with isolated anterior knee laxity (8 had undergone ACL reconstruction). On the one-leg-hop test, patients who demonstrated a low hop index (the ratio of the distance hopped on the injured leg to the distance hopped on uninjured leg multiplied by 100) when unbraced improved their performance by approximately 50% when braced. Marans et al (12) tested 10 people with ACL injuries (none had undergone reconstructive surgery and all had positive pivot-shift tests) using no brace, 3 custom-fit braces, and 3 off-the-shelf braces. When subjects wore 2 of the custom-fit braces, they ran faster figure-of-eight drills and cutting maneuvers than when they performed the same tests without a brace. Cook et al (13) used a custom-fit FKB to test 14 athletes with non-reconstructed ACL injuries. Straight running times were faster when athletes were braced than when they were unbraced.

Among the detrimental effects of FKB use were slower performance times and increased energy costs. In the paper by Marans et al, (12) the time subjects took to complete a 40-m, straight-ahead run was longer when they used one off-the-shelf brace and one custom-fit brace compared with their time when they did not use a brace. This finding conflicts with the report of Cook et al (13) that straight running times improved when subjects wore a brace. Houston and Goemans (14) tested 7 men with instability in the ACL, the medial collateral ligament, or both, each wearing his own custom-fit or off-the-shelf FKB. This study revealed that blood lactate concentrations were 41% higher 1 minute after subjects completed a 15-minute stationary bicycle endurance ride for the braced condition. Zetterlund et al (15) tested 10 men with non-reconstructed ACL ruptures using their own custom-fit brace. Heart rate and voluntary oxygen consumption (V[O.sub.2]) during 6-minute horizontal treadmill runs were higher for braced trials than for unbraced trials.

Although performance on functional tests was of interest to us, our patient was a recreational athlete and, therefore, more concerned about knee stability than optimal performance. In 5 of the 12 studies reviewed, patients with bracing reportedly felt more stable, had fewer episodes of "giving way," and believed that wearing a brace enhanced their performance. The authors of the review article commented that, although wearing a brace might increase motivation and confidence in some people, it also might lead to a false sense of security and lead them to engage in activities with substantial risk of injury. Another issue raised by the authors was the trade-off between stabilization and performance: braces that provide substantial mechanical stability might adversely affect mobility and performance. In view of the limited evidence that FKB use improves performance on functional tests, the authors of the review article recommended against using an FKB in isolation, but rather in conjunction with activity modification and an ongoing exercise program.

Both Nemeth et al and Kramer et al commented on the subjects' perceptions of greater knee stability when wearing a brace. Those reports led us to believe that our patient needed to carefully consider how much he perceived the need for external support to his knee. We still wanted more information regarding the effects of bracing on undesirable tibial translation. In addition, we knew that if our patient decided to purchase a brace, he would have to choose between custom-fit and off-the-shelf models. We felt the following article would provide us with helpful information on both of these issues.

Wojtys, EM, Huston LJ. "Custom-Fit" versus "Off-the-Shelf" ACL functional braces. Am J Knee Surg. 2001;4:157-62.

Many sports medicine practitioners believe "custom-fit" functional braces are superior in performance to "off-the-shelf" braces for anterior cruciate ligament (ACL)-deficient knees. However, this is not well substantiated. This study compares a Donjoy custom-fit ACL brace (CE 2000), Donjoy off-the-shelf brace (Goldpoint), and an athletic taping technique to determine their role in our clinical practice. Five patients (3 men and 2 women) with isolated, unilateral, chronic ACL tears with an average age of 27 years (range: 19-35) years were used to evaluate these three restraint systems. Anterior tibial laxity, quadriceps and hamstrings strength, endurance, standing long jump, brace migration with exercise, and pattern of muscle response to forced anterior tibial displacement were studied. Each patient was tested without a brace and then in each of the three test conditions (custom brace, off-the-shelf brace, and tape), with the order of testing randomized. The Donjoy custom-fit ACL functional brace did not reduce anterior laxity or improve standing long jump, muscle strength, endurance, or muscle response times significantly more than the off-the-shelf ACL brace. Both braces improved anterior stability over knee taping when the knee muscles were contracted under the low forces used in this study. After 1 hour of exercise, brace migration was significantly greater (P=.03) for the CE-2000 custom brace (18.6 mm) than for the Goldpoint off-the-shelf brace (4.5 mm). There appears to be no advantage to the more expensive custom-fit knee brace over the off-the-shelf brace.

[[c] 2001 Slack Inc. Abstract reprinted with permission of Slack Inc.]

We were interested in this article because we wanted to present evidence to our patient regarding the potential benefits of custom-fit braces versus off-the-shelf knee braces, if he decided to invest in an FKB. All subjects in the study had documented isolated ACL tears with no concurrent meniscal or chondral bone pathology. Although our patient was 20 years older than the mean age of the subjects in this group, we felt that the information regarding knee stability and function discussed in this article might apply to him. The tests administered in this study that were of the greatest interest to us were measures of anterior tibial translation, muscle force and endurance, and brace migration. Anterior tibial translation was measured by positioning the subject's lower extremity in a specially designed device while applying a force to the posterior aspect of the leg. Subjects demonstrated less anterior tibial translation when wearing the custom-fit and off-the-shelf braces than when they were unbraced; however, there was no difference in the amount of translation between the 2 brace types. That finding suggested to us that an FKB might provide additional control of anterior tibial translation for our patient. We were somewhat disappointed, however in the method used to measure stabilization, because it was a static test situation. We were uncertain whether the braces tested would reduce anterior tibial translation during activities involving rotation.

From the results of isokinetic testing, the authors concluded that neither the custom-fit brace nor the off-the-shelf brace affected hamstring and quadriceps muscle force and endurance. That finding suggested that wearing a brace would not compromise our patient's muscle force and endurance. Interestingly, the results of the brace migration test did not support investment in a custom-fit FKB. It was interesting to note that the custom-fit brace slipped significantly more during functional testing than the off-the-shelf brace did.

* Clinical decision: We felt that we had sufficient information from the 4 articles that we read to help the patient make an informed decision about whether he should purchase a knee brace. Three of the articles offered some support for the notion that wearing a brace improves knee stability in people with ACL injury. Based on the results of their EMG study of expert downhill skiers, Nemeth et al suggested that bracing could contribute to knee stability by increasing afferent input from proprioceptors. In the study by Wojtys and Huston, knee bracing reduced the amount of anterior tibial translation that occurred when a force was applied to the posterior aspect of the leg in a static test situation. Kocher et al stated that nonbraced skiers with ACL-deficient knees experienced a significantly higher proportion of injuries than braced skiers with ACL-deficient knees.

Although the evidence was not overwhelming, these 3 studies offered some indication that our patient might have greater knee stability if he wore a knee brace. The most compelling information in these articles were the observations of Nemeth et al and Kramer et al that many study participants reported their knees felt more stable when they wore a brace. We wanted to be careful not to overstate those findings, but we did wonder whether our patient might feel more confident in his recreational activities if a knee brace provided him with a sense of greater stability and less worry about "giving way."

Based on the papers we read, the effect of bracing on performance of functional and athletic activities is equivocal. Wearing a knee brace did not appear to be of either great benefit or great detriment to performance. Nemeth et al reported no difference in ski-run times with and without a brace, and Wojtys et al showed shorter one-legged standing long jump distances with a brace. We did not think that those results or the findings reported in the review article by Kramer et al, which addressed physical performance, should strongly affect the patient's decision one way or the other, given his modest goals.

Based on the suggestion in many of these studies that FKB wear improved knee stability and decreased the incidence of subsequent knee injury, and the anecdotal information regarding subjects' perceptions of increased stability in several of the studies, we decided to recommend that our patient order an FKB. Wojtys and Huston concluded that a custom-fit brace offered no advantage over an off-the-shelf brace. We were cautious in accepting that conclusion, given that only 5 patients participated in their study. A sample size that small could make it difficult for statistical analyses to reveal significant differences. Considering that the custom-fit brace slipped more than the off-the-shelf brace and the difference in price (custom-fit, approximately $1,200; off-the-shelf, approximately $600), we decided to recommend an off-the-shelf model.

The review article by Kramer et al raised other valuable points that we planned to share with the patient. For instance, it was important to explain that bracing does not replace the need to continue an exercise regimen aimed at maximizing muscle performance during functional activities. We decided to design a sport-specific home program for the patient that included strengthening exercises, balance activities, and graduated functional activities, such as figure-of-eight runs. Also, we cautioned the patient against developing a false sense of security while wearing the brace. We advised him to avoid activities that might put him at risk for further injury. For example, although we suggested that this patient can return to skiing, we recommended that he ski conservatively (eg, avoid icy conditions and moguls).

* dj Orthopedics Inc, 2985 Scott St, Vista, CA 92083.

References

(1) Magee DJ. Orthopedic Physical Assessment. 4th ed. Philadelphia, Pa: Saunders; 2002:chap 12.

(2) Hislop HJ, Montgomery J. Daniels and Worthingham's Muscle Testing: Techniques of Manual Examination. 6th ed. Philadelphia, Pa: WB Saunders Co; 1995.

(3) Wadsworth CT, Krishnan R, Sear M, et al. Intrarater reliability of manual muscle testing and hand-held dynametric muscle testing. Phys Ther. 1987;67:1342-1347.

(4) Frese E, Brown M, Norton BJ. Clinical reliability of manual muscle testing: middle trapezius and gluteus medius muscles. Phys Ther. 1987;67:1072-1076.

(5) Bohannon RW. Manual muscle test scores and dynamometer test scores of knee extension strength. Arch Phys Med Rehabil. 1986;67:390-392.

(6) Bohannon RW. Measuring knee extensor muscle strength. Am J Phys Med Rehabil. 2001;80:13-18.

(7) Mulroy SJ, Lassen KD, Chambers SH, Perry J. The ability of male and female clinicians to effectively test knee extension strength using manual muscle testing. J Orthop Sports Phys Ther. 1997;26:192-199.

(8) Styf J. The effects of functional knee bracing on muscle function and performance. Sports Med. 1999;28:77-81.

(9) Sackett D, Straus SE, Richardsom WS, et al. Evidence-Based Medicine: How to Practice and Teach EBM. 2nd ed. New York, NY: Churchill Livingstone Inc; 2000.

(10) Law M, ed. Evidence-Based Rehabilitation: A Guide to Practice. Thorofare, NJ: Slack Inc; 2002.

(11) Mishra DK, Daniel DM, Stone ML. The use of functional knee braces in the control of pathologic anterior knee laxity. Clin Orthop. 1989;241:213-220.

(12) Marans HJ, Jackson RW, Piccinin J, et al. Functional testing of braces for anterior cruciate ligament-deficient knees. Can J Surg. 1991;34:167-172.

(13) Cook FF, Tibone JE, Redfern FC. A dynamic analysis of a functional brace for anterior cruciate ligament insufficiency. Am J Sports Med. 1989;17:519-524.

(14) Houston ME, Goemans PH. Leg muscle performance of athletes with and without knee support braces. Arch Phys Med Rehabil. 1982;63:431-432.

(15) Zetterlund AE, Serfass RC, Hunter RE. The effect of wearing the complete Lenox Hill Derotation Brace on energy expenditure during horizontal treadmill running at 161 meters per minute. Am J Sports Med. 1986;14:73-76.

Ellen Wetherbee, PT, MEd, OCS, is Assistant Professor and Academie Coordinator of Clinical Education, Department of Physical Therapy, University of Hartford, West Hartford, Conn.

Geraldine L Pellecchia, PT, PhD, is Associate Professor of Physical Therapy, Physical Therapy Department, University of Connecticut, Storrs, Conn.