Efficacy of balance training for hip fracture patients: a meta-analysis of randomized controlled trials

Two reviewers performed an electronic literature search for randomized controlled trials (RCTs) or prospective clinical controlled studies comparing the balance training with control in the management of hip fracture. The electronic databases include PubMed, Embase, Web of Science, and the Cochrane Library up to December 2018. No language or date restrictions were applied. The following terms were used as keywords: ((((((((Fractures, Subtrochanteric) OR Subtrochanteric Fractures) OR Fractures, Intertrochanteric) OR Intertrochanteric Fractures) OR Fractures, Trochanteric) OR Trochanteric Fractures) OR Fractures, Hip)) AND (((((Training, Circuit) OR Circuit Training) OR Exercises, Circuit-Based) OR Exercise, Circuit-Based) OR balance training). In addition, further articles were obtained by reviewing references of the selected articles. The detail retrieval process is shown in Fig. 1.

Randomized controlled trials were included if they met the PICOS criteria as follows: Population: patients with hip fracture; Intervention: balance training; Comparator: placebo; Outcomes: overall function, gait speed, lower limb strength, ADLs, performance task scores, and HRQoL; Study design: RCTs or prospective clinical controlled studies.

Two reviewers independently retrieved the relevant data from articles using a standard data extraction form. The extracted data included publication date, authors, study design, inclusion and exclusion criteria, number and demographics of participants, number of included hips, surgical procedure, duration of follow-up, and outcomes. For missing data, such as standard deviations, we tried to get it by contacting the original author first. If it did not work, we calculated missing standard deviations from other available data such as standard errors or the formulas in the Cochrane Handbook for Systematic Reviews of Interventions. Two reviewers extracted the data independently, and any disagreement was discussed until a consensus was reached.

The methodological bias and quality of included studies were assessed by The Cochrane Collaboration’s tool for assessing the risk of bias according to the Cochrane Handbook for Systematic Reviews of Interventions [11]. It is a two-part tool with seven specific domains: sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective outcome reporting, and other sources of bias.

Stata 12.0 (Stata Corp., College Station, TX) was used to perform the meta-analysis. We used standard difference (SMD) and 95% confidence interval (CI) to assess continuous variable outcomes. For dichotomous outcomes, relative risks (RR) with a 95% CI were presented. Heterogeneity between studies was assessed by I² and χ² test. When I² < 50% and P > 0.1, we used a fixed-effects model to evaluate; otherwise, a random-effects was used. In addition, subgroup analysis was performed to explore the source of heterogeneity when heterogeneity existed.

The flowchart for the inclusion of articles is shown in Fig. 1. Initially, a total of 505 studies were searched via the databases and other sources (e.g., references). And 169 of the 505 studies were excluded due to the duplicates by Endnote Software (Version X7, Thompson Reuters, CA, USA). After reading the title and abstract, 327 trials were excluded according to the inclusion criteria. Finally, 9 studies [12‐20] with 872 patients (balance training = 445, control = 427) were included in our meta-analysis (published between 1997 and 2018).

Table 1 shows the detailed characteristics of the trials included. Three studies were originated from the USA, three were from Australia, one from Germany, one from Italy and one from China. Sample size of the included studies ranged from 13 to 120. The type of balance training is different from each other. Duration ranged from 3 weeks to 6 months. Follow-up duration ranged from 1 to 12 months.

Risk of bias summary and risk of bias graph are shown in Fig. 2 and Fig. 3 respectively. A total of three studies were rated as low risk of bias, three studies were qualified as high risk of bias, and the rest of the studies were rated as unclear risk of bias.

Results of this meta-analysis revealed that balance training has a positive role in increasing overall function for hip fracture patients. Moreover, balance training in the balance training group could significantly increase gait, lower limb strength, ADLs, performance task scores, and HRQoL scores than in the control group. Subgroup results have shown that high-frequency training was superior than low-frequency training for increasing overall function.

Only one relevant meta-analysis on the topic has been published [21]. Differences between ours and the previous ones should be noted. Different training frequency should be separately analyzed. Current meta-analysis performed publication bias, subgroup analysis, and subgroup analysis for overall function. Doma et al. [22] indicated that balance training improved walking capacity, balance-specific performance, and functional outcome measures for elderly individuals following total knee arthroplasty.

Our meta-analysis showed that balance training could significantly increase overall function and lower limb strength after hip fracture. Moreover, high-frequency training was recommended. Latham et al. [14] revealed that the use of a home-based functionally oriented exercise program resulted in modest improvement in physical function at 6 months. However, the clinical importance of the balance training remains to be determined.

Previous studies have shown that people have poor functional outcomes after hip fractures [23]. Following fracture, patients are at a high risk of entering a vicious cycle like fear of falling as well as post-fracture pain and muscle weakness [24]. Previous exercise studies using intensive professional supervision and equipment have found a significant capacity for adults with hip fracture to improve after balance training [12, 25]. We included nine RCTs and found that balance training has a positive role in improving overall function, gait speed, and lower limb strength. ADLs were compared between the balance training and control group. Balance training is superior than the control group in terms of the ADLs. Moreover, balance training in the balance training group increased HRQoL scores than in the control group. Combs et al. [26] revealed that balance training could significantly increase health-related quality of life.

Overall, there were several strengths in our research which are as follows: (1) comprehensive retrieval strategy was applied to reduce the risk of publication bias, and (2) we performed sensitivity analysis and subgroup analysis to further increase the robustness of our final results.

Nevertheless, our meta-analysis does have certain limitations which needed to be addressed: (1) training frequency, duration, and follow-up were different, and thus, there is a large heterogeneity for the final outcomes; (2) follow-up duration was relatively short to assess the clinical effects of balance training; thus, long-term effects of balance training was needed; and (3) the detailed blind methods and allocation concealment were not described in some RCTs that may affect the validity of the overall findings.