Reference/Citation: Rome K, Handoll HH, Ashford R. Interventions for preventing and treating stress fractures and stress reactions of bone of the lower limbs in young adults.
Cochrane Database Syst Rev.20052:CD000450. Update from Gillespie WJ, Grant I. Interventions for preventing and treating stress fractures and stress reactions of bone of the lower limbs in young adults. Cochrane Database Syst Rev.20002: CD000450.
Clinical Question: Do evidence-based interventions exist for the prevention and treatment of stress reactions and stress fractures in young active adults?
Data Sources: This systematic review is an update of the original article, which was published in 2000. The authors conducted a literature review of computerized databases that included the Cochrane Musculoskeletal Injuries Group Specialized Register (April 2004), the Cochrane Central Register of Controlled Trials, MEDLINE (1966 to September 2004), EMBASE (1988 to 2004, week 36), CINAHL (1982 to 2004, September, week 1), Index to Theses (1990 to 2004), and Dissertation Abstracts (1990 to 2004). In addition, the authors searched the Current Controlled Trials at
http://www.controlled-trials.com (June 2004, week 1) and the United Kingdom National Research Registrar at
http://www.update-software.com/national/ (to issue 1, 2004) for current or recently completed studies. They also reviewed the
British Journal of Podiatry, International Journal of Podiatric Biomechanics, Physiotherapy, and the
Australian Journal of Podiatric Medicine for relevant studies. Furthermore, they contacted the Medical Departments of Defense Forces in Europe and North America to identify unpublished or unlisted military studies. Reference lists of all identified studies and Cochrane reviews were also investigated. The computer search strategy included 61 separate entries and included such terms as
stress fractures, stress reactions, shin splints, overuse, athletic injuries, cumulative trauma disorders, running, and
randomized controlled trial. The 3 authors of this updated review independently selected new articles for inclusion. Furthermore, the 12 articles that were included in the original systematic review were also reevaluated to ensure they met the defined inclusion criteria.
Study Selection: To qualify for inclusion, studies had to be randomized or quasirandomized control trials, involve interventions to prevent or treat lower extremity stress reactions and fractures, and include physically active adults (adolescence to middle age) who were involved in athletics or military training. Clinical and radiographic (bone scan or x-ray) evidence of a lower extremity stress reaction or stress fracture was also required for inclusion of treatment-based studies. Specifically, skeletal overuse injuries are considered the result of a cumulative and repetitive process that produces initial microstructural changes or stress reactions that are identified by bone scans or magnetic resonance imaging but not conventional radiographs. If cumulative stresses continue, structural changes are visualized on radiographs and are referred to as stress fractures. In addition, research studies involving the treatment of medial tibial stress syndrome or shin splints were excluded. Desired outcome measures for treatment studies included return to training time, return to normal physical activity, functional performance, quality of life measures, resource management (eg, costs, health care visits, diagnostic procedures), adverse effects, and compliance.
The inclusion criteria for stress fracture prevention studies were similar, except that the authors did not have to provide radiographic evidence of a stress fracture or stress reaction. Prevention studies included a combination of the following outcome measures: occurrence and location of stress fracture, stratification of diagnosis, incidence of other lower limb injuries, complications and adverse effects of prevention techniques, resource management, and compliance with the prevention strategy.
Data Extraction: At least 2 reviewers independently extracted the demographic and outcome data from the newly identified studies, and 1 author verified the data and results from the 12 studies included in the 2000
Cochrane review. Inconsistencies from the original review and data from all new studies were also checked by an additional reviewer. All 3 reviewers then independently evaluated the quality of inclusion studies using a quality scoring scheme (
Table). The categories considered included randomization or group allocation (A), intention-to-treat analysis (B), examiner blinding (C), comparison of experimental and control groups at baseline (D), use of a placebo treatment (E), clearly defined subject inclusion and exclusion criteria (F), and methods of outcome assessments (G). Items A through F were scored from 0 to 2 and item G from 0 to 3, for a total “best” quality assessment score (QAS) of 15. Inconsistencies among reviewers' QAS scores were resolved by discussion and with the aid of a discrepancies form.
Main Results: Search criteria identified 24 new studies since the previous review, 8 of which fulfilled the inclusion criteria. In addition, 4 of the 12 studies included in the original 2000 review were excluded. Three were excluded as a result of insufficient indication of subject or group randomization or quasirandomization, and the fourth excluded study included subjects with the diagnosis of medial tibial stress syndrome. Overall, 16 studies were included.
The authors of 13 studies focused on prevention, and 3 groups evaluated the treatment of stress fractures and reactions. The average number of subjects for prevention and treatment studies, respectively, was 1091 (range = 206 to 3025) and 34 (range = 21 to 60). All 13 prevention studies involved military personnel who performed physical training over a 9-to-14–week period. Quality assessment scores for prevention studies ranged from 4 to 10 (mean score = 7). In 9 prevention studies, the effectiveness of insoles or orthoses was evaluated, and the QAS for these studies ranged from 4 to 9 (mean = 6.2). The investigators in 4 studies assessed “shock-absorbing” insoles or orthoses in shoes or boots versus a control (shoes or boots alone), and an additional 5 groups compared insoles and orthoses against one another. One study's authors also evaluated military training in a modified high-top shoe versus standard military boots (QAS = 8). Two groups assessed the influence of pre-exercise stretching (QAS = 8 and 9, respectively), and one investigated the effects of calcium supplementation (QAS = 10).
In none of the prevention studies were adequate randomization and concealment of treatment before group allocation (item A) accomplished, and the researchers in 3 studies randomized groups (team or platoon) instead of individual participants. Attrition rates exceeded 50% in 2 studies, and missing subjects' data were unaccounted for in the final analysis of 3 studies (item B, intention-to-treat analysis). Also, in only 2 of 13 studies were examiners blinded to group assignment (item C). Radiographic (bone scan or x-ray) evidence for diagnostic confirmation of a stress reaction or fracture was used in 12 studies. The method of diagnosis (item G) was based solely on clinical examination or a self-report questionnaire in 2 studies, and diagnostic methods were not described in 2 studies.
Overall fewer osseous stress injuries were reported in the experimental groups for all 4 studies comparing military personnel in “shock absorbing insoles” with controls (no insoles). However, none of these 4 studies demonstrated a statistically significant reduction in lower extremity overuse osseous injuries. In addition, statistically significant results were reported in only 1 of 5 studies that compared various orthoses and insoles. The authors reported a significant reduction in tibial stress fractures for soldiers wearing custom-made semirigid or soft-foot orthoses versus those wearing standard insoles (relative risk = 0.46, 95% confidence interval = 0.22 to 0.93). In a follow-up study, no significant difference in stress fracture rates was seen between subjects who wore custom-made semirigid orthoses and those who wore biomechanical soft orthoses, thus precluding the ability to identify one best design for stress fracture reduction. No significant stress fracture or lower extremity injury rate differences were seen between the control and experimental groups involved in lower extremity stretching studies. Participants taking calcium supplements did not demonstrate a significant reduction in stress fractures (tibial only) versus controls. The differences among the prevention studies prohibited pooling of the data and subsequent meta-analysis. Authors of all 3 treatment studies investigated the effects of a pneumatic ankle foot orthosis (Aircast Corp, Summit, NJ). Follow-up for outcome measures ranged from 78 days to 6 months. Two studies were conducted with military recruits, and the other was conducted with competitive and recreational athletes (n = 18, age range = 18 to 45 years). Treatment QASs ranged from 7/15 to 11/ 15, with an average score of 9.3/15. Proper randomization (item A) and evaluator blinding (item C) were confirmed in 1 of the 3 treatment studies. Data pooled from all 3 studies reached statistical significance for mean number of days until returning to full activities (weighted mean difference with brace versus without brace = −33.39 days, 95% confidence interval = −44.18 to −22.59 days).
Conclusions: Currently, no solid evidence-based interventions to prevent lower extremity stress reactions or fractures exist. Limited evidence suggests that “shock absorbing” insoles may reduce the overall incidence of lower extremity osseous injuries in military personnel. Unfortunately, research does not support the best design for inserts or footwear modifications. There is also insufficient evidence to determine if pre-performance stretching or calcium supplementation offers added protection from lower extremity osseous overuse injuries. Initial evidence supports the use of a pneumatic brace and early mobilization for the treatment of tibal stress reactions and fractures, but additional studies are required to validate these findings. Further investigation concerning the prevention and treatment of lower extremity stress fractures is needed and would assist researchers in establishing and clarifying evidence-based intervention guidelines. Future randomized control trials that clearly define (ie, provide clinical and radiographic evidence for) the diagnosis of a stress fracture or reaction, implement appropriate randomization, and use intervention and outcome measures (functional and performance measurements) that are appropriate for active adults would assist this ongoing and necessary process.