The incidence of hip fractures is increasing worldwide.¹⁾ Hip fracture is associated with increased hospitalization and rehabilitation costs, a high societal burden given its association with adverse outcomes, including depression and cardiovascular diseases.¹⁾ In Japan, the cost of treating patients with hip fractures leads to economic burdens on society.²⁾ Additionally, older adults with hip fractures have high mortality rates.³⁾ Therefore, effective rehabilitation measures are essential for older adults with hip fractures.

Due to insufficient recovery, many older adults with hip fractures require assistance in their daily activities. Specifically, 20%–60% of older adults with hip fractures require assistance for one year after treatment, despite having been independent in daily life before their injury.⁴⁾ Furthermore, >40% of individuals present with new-onset walking disability six months after a hip fracture,⁵⁾ indicating the insufficiencies of strategies to promote recovery in older patients with fractures.

Several factors, including age, sex, and comorbidities, can hinder recovery in patients with a hip fracture,⁶⁾ with malnutrition being a crucial modifying factor.^7,8) Malnutrition further increases the risk of institutionalization and mortality in older inpatients with hip fractures.⁹⁾ Therefore, the accurate assessment of the nutritional status of older inpatients with hip fractures and the provision of appropriate interventions are essential.

However, the subjective nutritional assessments recommended in inpatient and rehabilitation settings, including the Nutritional Form for the Elderly and Mini Nutritional Assessment Short Form Version 1,¹⁰⁾ are dependent on changes in weight and food intake. Therefore, their ability to accurately assess patients with severe cognitive impairment may be limited. Additionally, these assessments demonstrate interobserver variability.¹¹⁾ While objective nutritional assessments, including the Geriatric Nutritional Risk Index (GNRI),¹²⁾ are dependent on serum albumin concentration, which is measured using blood tests, blood tests are invasive and may not be performed routinely. Therefore, new objective and routine assessment methods are required.

Phase angle, which can be measured using bioelectrical impedance analysis, is associated with cellular health. The phase angle has recently received attention as a noninvasive and objective method for nutritional assessment.^13,14) Poor nutritional status damages cells, thus decreasing the phase angle; hence, the phase angle can reflect nutritional status. In particular, the phase angle is useful for nutritional assessment in older inpatients.¹⁵⁾

Bioelectrical impedance analysis is a simple, low-cost, and reproducible tool,¹⁶⁾ and the phase angle is a versatile and practical method that can be used to assess nutrition. Phase angle may be influenced by certain diseases.^17,18) For example, postoperative edema in patients with hip fractures may affect body composition, which, in turn, influences the phase angle.¹⁹⁾ A previous study demonstrated a lower phase angle in patients with hip fractures admitted to rehabilitation units in Japan compared to that of healthy individuals.²⁰⁾ However, the link between nutritional status and phase angle in older inpatients with hip fractures has not yet been investigated.

Nutritional assessment using bioelectrical impedance analysis may provide useful information for rehabilitation interventions in patients with hip fractures. Therefore, this study assessed the association between the phase angle measured using bioelectrical impedance analysis and malnutrition in older inpatients with hip fractures and ascertained the optimal cutoff phase angle for identifying malnutrition.

The mean±standard deviation age of the patients was 82.4±6.2 years, the mean phase angle was 3.96°±0.76°, and 76 patients (79.2%) had malnutrition (GNRI of ≤98). The patient characteristics are listed in Table 1.

Spearman rank correlation coefficient revealed significant correlations between malnutrition and SMI (r=−0.210, p=0.041) and phase angle (r=−0.29, p=0.004) but not between malnutrition and the other variables—age (r=0.014, p=0.893), sex (r=−0.093, p=0.37), CCI (r=0.046, p=0.654), MMSE (r=−0.097, p=0.366), and mFIM (r=0.051, p=0.619).

Binomial logistic regression confirmed the association of phase angle with malnutrition—crude model (odds ratio [OR]=0.35, 95% confidence interval [CI] 0.17–0.72) and adjusted model (adjusted by age, sex, and SMI; OR=0.31, 95% CI 0.12–0.80) (Table 2). The ROC curve analysis revealed an AUC of 0.71 (95% CI 0.58–0.83, p=0.001). The cutoff phase angle for malnutrition was 3.96° (sensitivity=0.85, specificity=0.63) (Fig. 2).

In this study, the cutoff phase angle for identifying malnutrition in older inpatients with hip fractures was 3.96°. A previous study of inpatients with various diseases reported a mean phase angle on admission of 3.9°±0.9°, with cutoff phase angles for malnutrition of 4.03° and 3.65° for men and women, respectively.²⁷⁾ This study, which included only patients with hip fractures, yielded similar results. Therefore, the cutoff value in this study showed a certain degree of reliability.

Screening for malnutrition is essential to facilitate prompt nutritional intervention in older inpatients with hip fractures.⁹⁾ Nonetheless, although subjective scales using questionnaires allow for convenient nutritional assessment, their reliability remains questionable.¹¹⁾ Although the GNRI is considered a good nutritional assessment index because it incorporates anthropometric factors and serum markers,²⁸⁾ blood tests are required to calculate the GNRI and cannot be routinely performed because of the burden they pose on inpatients. Bioelectrical impedance analysis is safe, reproducible, and easy for inpatients, and the results are independent of the examiner’s level of experience and skill.²⁹⁾ Therefore, our findings overcome the limitations of conventional nutritional assessments and allow for prompt and accurate evaluations.

Nutrition-focused strategies are crucial in the rehabilitation of patients with hip fractures.^7,8) Moreover, routine nutritional assessments and interventions based on these assessments may be beneficial for the recovery of older adults with hip fractures. However, malnutrition is often not assessed or treated in older adult inpatients.³⁰⁾ Therefore, this study, which evaluated a simple method to screen for malnutrition, will help facilitate the development of strategies to promote recovery in older inpatients with hip fractures.

The AUC in this study was 0.71, which was sufficient for diagnostic accuracy, although it was not ideal.^24,25) Therefore, the ability of the phase angle to identify malnutrition is limited. Considering the relatively high sensitivity for discrimination (0.85), it may be possible to more accurately identify patients with malnutrition by performing a detailed nutritional assessment of those who fall below the phase angle cutoff for screening.

This study has a few limitations. First, we used only the GNRI to assess malnutrition; therefore, nutritional assessments should be performed in conjunction with other assessments. In addition, the GNRI does not demonstrate high accuracy in assessing hypernutrition.²⁸⁾ As such, our findings are specific to malnutrition. Nevertheless, the GNRI is a good nutritional indicator²³⁾ which can be used to assess malnutrition in older patients with hip fractures. Another consideration is that the phase angle is affected by age and sex.^31,32) Future studies should consider both age and sex to determine more accurate cutoff values. Second, while we performed bioelectrical impedance analysis using the InBody S10 device according to the manufacturer’s instructions, we could not confirm whether the procedures described in these instructions were followed. Third, as all subjects in this study were postsurgical inpatients and serum albumin levels were affected by surgery, surgery may have influenced the results of this study.³³⁾ In addition, the times between the date of surgery and the bioelectrical impedance analysis and between the date of surgery and the date of the blood tests were not consistent. Performing blood tests and bioelectrical impedance analyses on the same day is likely to yield more accurate results. Nevertheless, the half-life of serum albumin is 20 days,³⁴⁾ and the interval between the dates of blood tests and bioelectrical impedance analysis was approximately 8 days in this study. Moreover, the results showed a strong association between the GNRI calculated from serum albumin and the phase angle regarding their ability to determine malnutrition, confirming a certain degree of reliability in the accuracy of the calculated phase angle cutoff; therefore, the results of this study may be applicable to clinical practice. Finally, selection bias may have resulted from the exclusion of individuals with missing GNRI values or an inability to undergo bioelectrical impedance analysis. Nonetheless, this study proposes a simple and objective malnutrition index for older inpatients with hip fractures, which has potential for clinical applications.

In conclusion, this study investigated the utility of the phase angle as an objective nutritional assessment index in older inpatients with hip fractures. The results suggest that phase angle is a potentially useful screening tool for malnutrition, with a cutoff value of 3.96° in older individuals. Our findings will contribute to the development of rehabilitation strategies for older adult patients with hip fractures. In other words, if the phase angle is indicative of malnutrition, monitoring it during rehabilitation may prevent or reduce the occurrence of malnutrition and promote recovery in older patients with fractures.