Introduction

Most rehabilitation programs after hip fracture mainly focus on postoperative range of motion exercises, standing, gait training with progressive weight bearing, and strengthening exercises of the hip extensor and abductor muscles [1]. These conventional programs have been shown to improve the independence of activities of daily living (ADLs) and gait function after hip fracture [2]. However, after rehabilitation hip fracture patients can show balance deficits for 2–3 years [3,4,5], because the standard rehabilitation programs have little effect on balance [6].

The persistence of balance deficits is a serious problem for hip fracture patients since postural instability and fear of falling affect locomotion and ADLs and can lead to falls with potentially severe consequences including fractures and other injuries [7, 8]. Because of negative effects of balance deficits on mobility and ADLs, it is very important to know the factors influencing balance and to identify hip fracture patients with higher risk of poorer balance recovery. Knowledge of these predictors may help physicians to plan more appropriate rehabilitation programs in patients who have greater difficulty in recovering balance to prevent falls and re-fractures in these patients. Indeed, little is known about the factors influencing the rehabilitation of balance and the factors predicting balance levels in hip fracture patients undergoing rehabilitation. The literature on this topic is scant and the data are conflicting [9,10,11].

The primary aim of this retrospective observational study was to assess balance before and after motor rehabilitation in older hip fracture patients. The secondary aim was to identify what factors can predict the balance performance in these patients.

Materials and methods

This retrospective cohort study was conducted at the Istituti Clinici Scientifici Maugeri. It consists of a secondary analysis on data from an Institutional database evaluating the impact of hip diseases on balance, collected in patients aged ≥ 65 years undergoing hip surgery and admitted to our Rehabilitation Unit as in-hospital patients between January 2016 and December 2017. Exclusion criteria were patients aged < 65 years, or referred from other departments outside our district with concomitant acute events, or patients with post-surgery complications that emerged during rehabilitation, and patients who died or were transferred back to acute care during the course of rehabilitation. Patients who did not give informed consent or were not able to walk prior to the fracture or perform basic activities without assistance, or who were not living at home at the time of the fracture, or who had pathological fractures were also excluded. All patients demonstrated, after administration of cognitive-FIM [12], an adequate language comprehension and gave their written informed consent to participate. The Local Review Board of our Institute approved the study protocol. The study was conducted in accordance with the principles of the Declaration of Helsinki.

Mode of assessment

Clinical evaluation of all patients was performed at admission and at the end of rehabilitation by a qualified team of physiatrists and geriatricians with scales of demonstrated reliability, validity and sensitivity [12,13,14] that have been used in previous studies concerning hip fracture recovery [12, 14,15,16,17,18]. Patient demographic characteristics and comprehensive clinical data including orthopedic surgery (internal fixation, bipolar hemi-arthroplasty, total hip replacement arthroplasty, immobilization), time from fracture to admission for rehabilitation (days), body mass index (BMI), and length of stay in hospital (days) (LOS) were also recorded. The following instruments were used for assessment:

The Berg Balance Scale (BBS) was used to measure balance in our hip fracture patients by assessing the performance of functional tasks [15]. The BBS consists of 14 items that require subjects to maintain positions of varying difficulty and perform specific tasks. Each item is scored on a 5-point ordinal scale ranging from 0 (unable to perform) to 4 (normal performance). In this study, we excluded the item “pick up an object from the floor” to avoid excessive hip flexion which may predispose to hip dislocation in arthroplasty patients; therefore, the aggregate score of BBS ranged from 0 to 52.

The Cumulative Illness Rating Scale (CIRS) was used at admission to evaluate comorbidities [14]. In this study, we considered the average severity of all comorbidities (severity index).

The Katz ADL scale was used to assess patients’ functional status at admission [16]. It is a 6-item scale assessing bathing, dressing, toileting, transferring, continence, and feeding. Katz scores range from 0 (severe functional impairment) to 6 (full function).

The Cognitive-Functional Independence Measure (cognitive-FIM) was used to assess the mental status of patients [12]. The cognitive-FIM is a 5-item ordinal scale assessing comprehension, expression, social interaction, problem solving, and memory, with each item scored from 1 (total dependence) to 7 (total independence). The cognitive-FIM total score ranges from 0 to 35.

Hip pain intensity was measured using an 11-point Visual Numeric Scale (VNS) from 0 (no pain) to 10 (intolerable pain) [17].

The Muscle Strength Grading Scale (Oxford Scale) [18] was used to assess hip muscle strength. Scores ranged from 0 to 5 (where 0 = no movement and 5 = muscle contracts against full resistance). For our study, we considered the sum of the strength of flexor and abductor muscles of the hip and strength of quadriceps.

A manual goniometer was used to measure the range of motion (ROM) of the hip joint. The sum of ROM in flexion (0–120°) and abduction (0–45°) was considered [19].

Rehabilitation program

The rehabilitation program consisted of an average of 330 min/week of standard motor rehabilitation (6 days/week) for the entire duration of in-hospital stay and was supported by 150 min/week of occupational therapy in the final 2 weeks of in-hospital stay.

The rehabilitation program was based on lower limb ROM, strengthening exercises (hip abduction, flexion and extension, knee extension and ankle dorsi-plantar flexion in the supine position), bed to chair mobility (bed to chair, chair to toilet, chair to chair), pre-gait (sit to stand and vice-versa, balance in standing position), gait activities (parallel bars, walker, crutches), and ADL training (climbing stairs, bathroom skills).

Each patient’s needs, specific rehabilitation goals and progress/outcomes were discussed at admission and bi-monthly by the rehabilitation team (composed of physicians, physiotherapists, and an occupational therapist). Rehabilitation commenced the day after admission. Patients were discharged when, in the opinion of the rehabilitation team, no further in-hospital improvement with rehabilitation was expected.

Statistical analysis

All statistical analyses were performed with the software application Statistica Version 6. Statistical analysis was performed using the Shapiro–Wilk test, descriptive statistic tests (mean ± SD, percentage), and Student’s t test to examine differences within the group. Spearman’s rank correlation coefficient was used to assess correlations between variables, and correlations < 0.30 were considered weak, from 0.30 to 0.50 moderate, and > 0.50 strong [20]. Backward stepwise multiple regression analyses were performed to identify predictors of outcome measures. Multiple regression analyses on BBS scores were performed and residual distribution and homoscedasticity were checked. Only admission variables that were significant at univariate analysis underwent multivariate analysis. Age, sex, time from surgery to admission for rehabilitation, CIRS severity index, BMI, cognitive-FIM, pain VNS, hip ROM, and hip muscle strength were independent variables investigated in the univariate analysis.

Final score and effectiveness and efficiency of motor rehabilitation on balance performance, as measured by the BBS, were the outcome measures.

Effectiveness was defined as the proportion of potential improvement achieved during rehabilitation. In this study the Effectiveness of motor training on balance of hip fracture patients was calculated with the following formula: [(admission BBS score − discharge BBS score)/(52 − admission BBS)] × 100 [21]. Efficiency of motor rehabilitation training on balance represents the average increase in BBS per day obtained by therapy [21] and was calculated by dividing the total number of BBS points gained during rehabilitation by the number of in-hospital days of patients.

P values < 0.05 were considered as statistically significant.

Results

The study was carried out on 124 eligible hip fracture patients consecutively admitted to our Rehabilitation Unit during the study period. Table 1 reports demographic and clinical characteristics and the final BBS scores of the study population.

Table 1 Demographic and clinical characteristics of study patients (n = 124)
Full size table

In the whole sample of hip fracture patients, the mean BBS score was 8.33 ± 7.23 at admission and 21.79 ± 12.15 at the end of rehabilitation (p < 0.001). The mean gain in BBS score was 13.46 ± 9.31 and it was significant (p < 0.001). The daily gain in BBS score of hip fracture patients was 0.39 ± 0.31 and percentage of improvement in BBS achieved with rehabilitation was 32.28 ± 23.04%. The final levels, daily gain and percentage of improvement in BBS of male hip fracture patients did not differ from those of female patients (p = 0.246, p = 0.924, and p = 0.389, respectively).

Internal Fixation and Hemi-arthroplasty patients were the most representative groups.

Before rehabilitation Internal fixation patients had a longer time from fracture to admission for rehabilitation (p = 0.015) with respect to Hemi-arthroplasty patients because they required more time to achieve a stable fixation. At the end of rehabilitation there were no differences in final levels, daily gain, and percentage of improvement in BBS score between Internal fixation and Hemi-arthroplasty patients (p = 0.978, p = 0.573, p = 0.691, respectively).

Table 2 reports the BBS item scores for the study population. All the BBS items had a significant improvement at the end of rehabilitation (p < 0.001 for all). Standing with one foot in front (0.59 ± 1.08, 012 ± 0.23, and 13.0 ± 125.3) and standing on one foot (0.83 ± 1.15, 017 ± 0.24, and 18.4 ± 26.2) were the BBS items with the lowest score at discharge and the lowest daily gain and percentage of improvement in BBS (Table 2).

Table 2 Berg balance scale items; admission, discharge, efficiency and effectiveness
Full size table

Tables 3 and 4 report the correlations between outcome measures at discharge and clinical and demographic characteristics of the hip fracture patients at admission and at discharge, as assessed by Spearman’s correlation. Table 4 shows that cognitive-FIM (rho 0.57, 0.40 and 0.47, respectively), hip muscles strength (rho 0.62, 0.48 and 0.55, respectively), and Katz index (rho  – 0.67,  – 0.49 and 0.58, respectively) at discharge had moderate-to-strong relationships with final, daily gain and percentage of improvement in BBS.

Table 3 Relationships between outcome measures at discharge and demographic and clinical characteristics at admission in hip fracture patients (n = 124)
Full size table
Table 4 Relationships between outcome measures at discharge  and demographic and clinical characteristics at discharge in hip fracture patients (n = 124)
Full size table

Table 5 shows the results of backward regression analysis performed on hip fracture patients. Only variables at admission that were significant at Spearman’s correlation underwent multivariate analysis. The table shows that Cognitive-FIM was a predictor of final BBS score (beta 0.49, p < 0.001), daily gain in BBS (beta 0.34, p < 0.001) and percentage of improvement in BBS (beta 0.44, p < 0.001). Conversely, hip muscles strength was a predictor of final BBS score (beta 0.32, p = 0.001) and CIRS severity of daily gain in BBS (beta  – 0.29, p = 0.001). The R2 value of the models were respectively 0.39, 0.23, and 0.19.

Table 5 Forward stepwise regression analyses performed in hip fracture patients (n = 124)
Full size table

Discussion

This study analyzed the factors predicting the balance performance in older hip fracture patients undergoing standard motor rehabilitation.

Differently from other studies in the literature [9,10,11], we analyzed in these patients efficiency and effectiveness in BBS, which respectively represent the daily gain and percentage of improvement in balance achieved with rehabilitation [21]. In our opinion, these two parameters can be considered as indicators of the efficacy and appropriateness of a rehabilitation program.

We found that in hip fracture patients at the end of rehabilitation the mean BBS score was 21.79 ± 12.15, while the BBS gain for each day of stay in the Rehabilitation Department was 0.39 and the rate of improvement in the BBS was 32.28%. Moreover, at the end of rehabilitation there were no differences in the final levels, daily gain, and percentage of improvement in BBS score between Internal fixation and Hemi-arthroplasty patients and between male and female hip fracture patients.

To our knowledge, no previous studies in the literature have analyzed the efficiency and effectiveness of motor rehabilitation on balance performance in hip fracture patients; hence it is not possible to compare our findings with other reports.

We used a back multiple regression analysis to identify the factors predicting the balance performance in hip fracture patients. We found that cognitive-FIM at admission was a predictor of both daily gain and percentage of improvement in BBS, while CIRS was a predictor of daily gain only. The relationships among cognitive-FIM, daily gain and percentage of improvement in BBS were positive, while the relationships between CIRS and daily gain in BBS were negative; this suggests that, after rehabilitation, daily gain and percent improvement in balance may be lower in hip fracture patients with cognitive impairment and/or severe comorbidities.

The cognitive impairment interferes with the processes that involve sensory information and motor response necessary to control balance and postural stability [22, 23], but it can also affect patient's participation in the rehabilitation program thus prolonging its length [24]. Moreover, severe diseases related to neurological and musculoskeletal systems can cause joint and muscle damages slowing down the rehabilitation program and thus reducing balance efficiency [9, 25]. In addition, during the course of rehabilitation, the moderate-to-severe comorbidities may be more frequently complicated by adverse clinical events (as chest infections and heart failure) slowing down the rehabilitation course and inducing the same effect as above [26]. Knowledge that balance recovery may be poorer in hip fracture patients with cognitive impairment and severe comorbidities suggests that in these patients rehabilitation programs should include balance task-specific training.

These programs have been shown in the literature to be superior to conventional motor rehabilitation in improving balance, lower limb strength, ADLs, and the quality of life in older patients after hip fracture [6, 27, 28] and to produce benefits that last for at least 12 months [6].

Moreover, the associations among outcome measures, cognitive impairment and comorbidity suggest: (i) to activate in these patients measures that can prevent the onset and worsening of cognitive impairment [29], and (ii) to promptly treat the adverse clinical events in patients with severe comorbidities [30].

The multivariate analysis also showed that the cognitive-FIM and hip muscles strength of lower limb after surgery were important predictors of final BBS levels. Relationships among final BBS score, cognitive impairment and hip muscle strength were positive, indicating that after rehabilitation the final balance levels may be lower in hip fracture patients with cognitive impairment and/or poor hip muscles strength.

Our findings on cognitive function are in agreement with Ariza-Vega et al. [11], who showed that cognitive impairment is an independent predictor of mobility in hip fracture patients. Conversely, our finding on muscle strength is not confirmed by the literature in hip patients after surgery.

In any case, muscle strength is a factor that influences balance because it enables the motor responses necessary to control balance and postural stability [22, 31].

In this study, we considered the strength of flexor and abductor muscles of the hip and strength of quadriceps. The sum of the strength of all these muscles was an important predictor of final BBS score: this suggests that standard motor rehabilitation programs should be integrated with muscle-strengthening exercises in patients who have poor muscle strength in the lower limb. Indeed, muscle-strengthening programs have been shown to concurrently improve both hip strength and balance [32, 33].   Sylliaas et al. [32] showed that a 3-month high-intensity strength training improved both strength and balance in home-dwelling hip fracture patients [32]. However, balance task-specific training improved both strength and balance as well [6, 27].

The study also analyzed the factors which at the end of rehabilitation were associated with outcome measures. Spearman’s correlation pointed out that cognitive-FIM, hip muscles strength, and Katz index at discharge had moderate-strong relationships with final, daily gain and percent improvement in BBS, while ROM at discharge had a moderate relationship with the outcome measures.

Studies on this issue analyzed only the BBS score at the end of rehabilitation and found that male, sex, increased comorbidity, cognitive impairment, older age and a longer hospitalization were variables associated to a worse balance [9, 11]. Our study confirms the associations between cognitive impairment and worse balance at the end of rehabilitation reported by Ariza-Wega et al. [11] and points out that also poor hip muscles strength and higher ADL disability are associated to worse balance. These associations confirm the need to activate, early after the onset of hip fracture, measures to prevent and counteract cognitive impairment [29], ADL disability, and lower limb muscle strength deficits [29, 31].

Finally, the study assessed the final levels, daily gain and percentage of improvement achieved with rehabilitation in single items of BBS of hip fracture patients and found a significant improvement in all the items of BBS. However, the scores were lower in two items: "on standing with one foot in front" and "standing on one foot". These items assess balance in highly unstable conditions and their correct execution requires higher levels of cognitive function and muscle strength thus explaining their lower daily gain and percentage of improvement in hip fracture patients.

Our study has some limitations. The ROM assessment considered only the hip ROM in flexion and abduction, while muscle assessment considered only the strength of flexor and abductor muscles of the hip and strength of quadriceps, which in other studies were significant determinants of performance on static and dynamic balance tests [34, 35]. Muscle strength assessment was performed with manual and not instrumental tests, which have a higher level of accuracy and reliability [36].

We did not analyze all the possible factors that may have influenced balance [10, 11]. Patients were admitted to a specific rehabilitation program and therefore the results may not apply to other programs with different entry criteria. Finally, the study was not population-based (patients were referred from general hospitals) and therefore it does not represent all hip fracture patients.

In conclusion, we show that in hip fracture patients, cognitive function, comorbidities and hip strength are important predictors of balance. Knowledge of these findings can be useful for physicians to identify patients needing more specific rehabilitation programs (Fig. 1) and to plan a new tailored treatment approach for balance.

Fig. 1
figure 1

Flow chart of the program proposed to guide physicians in providing the appropriate level of rehabilitation to hip fracture patients

Full size image