Association of Frailty with Ultrasound Parameters of Diaphragm Function in 153 Elderly Individuals

Introduction

Frailty is a multifaceted syndrome that reflects a decline in the reserve capacity of various physiological systems in older adults, leading to a diminished ability to maintain homeostasis. It serves as a critical indicator of chronic health problems and medical needs in the elderly, revealing significant variations in disease outcomes, rehabilitation effectiveness, and overall quality of life. Epidemiological studies demonstrate a strong correlation between frailty and aging, with prevalence rates notably increasing with age. However, incidence rates of frailty vary widely, ranging from 4.9% to 83.4%, likely due to inconsistent diagnostic criteria across studies [1]. Among the most widely used frailty assessment tools is the Fried frailty phenotype, which evaluates 5 physical components: unintentional weight loss, self-reported exhaustion, weakness (grip strength), slow walking speed, and low physical activity. Individuals meeting 3 or more criteria are considered frail, while those with one or 2 are classified as pre-frail [2]. In contrast, the Frailty Index includes a broader range of deficits, such as comorbidities, cognitive decline, and functional impairments [3]. Among the physiological systems affected by frailty, dysfunction of the respiratory muscles, particularly the diaphragm, has emerged as a significant contributor to adverse outcomes in elderly patients. The diaphragm, as the primary muscle involved in respiration, plays a central role in maintaining ventilatory efficiency and oxygenation. Age-related muscle atrophy, decreased contractility, and chronic comorbidities may impair diaphragm function, compounding the effects of frailty by promoting respiratory insufficiency, fatigue, and reduced physical resilience. Ultrasound assessment of the diaphragm is a non-invasive and increasingly used method for evaluating respiratory muscle function. Key ultrasound parameters include end-expiratory and end-inspiratory diaphragm thickness, diaphragmatic excursion during deep breathing, and diaphragm thickness fraction (DTF), which reflect contractile capacity [4,5].

As the global elderly population expands, frailty has become a major concern in geriatric research. According to the seventh national census in 2020, 13.5% of individuals in China are over 65 years old, a trend mirrored worldwide [6]. Although this study focused on a Chinese hospital population, the findings may have broader relevance due to the global nature of aging and shared frailty-associated challenges. In the United States, elderly patients account for approximately 40.76% of emergency visits [7], and domestic studies indicate that many emergency inpatients are age 61–80 years [8,9]. Frailty is characterized by reduced physiological reserves and increased vulnerability to health stressors, often associated with malnutrition and various chronic conditions [10]. Older adults represent a significant proportion of individuals affected by acute and critical illnesses, with frailty exacerbating their overall prognosis [11]. Research on frailty has primarily focused on community settings, leading to inconsistencies in frailty identification between hospital emergency patients and regular inpatients [12,13]. Early detection and intervention for frailty are crucial, as they can significantly reduce adverse health outcomes. Screening and evaluating frailty in emergency patients enable healthcare personnel to develop personalized treatment plans, optimize resource allocation, and enhance overall quality of life for elderly patients [14].

Furthermore, previous research has highlighted a correlation between respiratory muscle function and frailty in older individuals [15,16]. While some studies have examined diaphragm function in aging, such as Yamada et al [5], who reported age-related reductions in diaphragmatic thickness and inspiratory pressure, there remains limited evidence specifically addressing the relationship between frailty phenotype and diaphragm ultrasound parameters in clinical populations. Ultrasound techniques are often used to assess the cross-sectional area of muscles such as the rectus femoris in critically ill patients, aiding in diagnoses like intensive care unit-acquired weakness (ICU-AW). The diaphragm is particularly vital for respiratory function; its dysfunction can lead to significant complications. Research indicates that inactivity of the diaphragm for prolonged periods can cause muscle atrophy [17], and even short durations of mechanical ventilation can lead to declines in diaphragm strength over time [18]. Despite the relevance of diaphragm function, routine assessments remain underutilized in clinical practice [19]. Therefore, this study aimed to evaluate the association between ultrasound parameters of diaphragm function and frailty as assessed by the Fried frailty scale in a hospital-based cohort of 153 elderly patients.

Material and Methods

ETHICS STATEMENT:

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, China (Approval No: 2022-SR-441). All participants or their legal guardians provided written informed consent prior to participation.

PATIENT SAMPLING AND GROUP ALLOCATION:

This study included 153 elderly patients (64 males and 89 females), aged 60–85 years (mean age: 75.77±2.47 years), who were treated at our hospital between June 1, 2022, and September 1, 2023. Patients were recruited consecutively from inpatient admissions to reduce selection bias and ensure representativeness of the hospital-based elderly population. Only elderly patients were included because frailty predominantly affects this age group and diaphragm dysfunction becomes clinically relevant in older adults [20]. The clinical and demographic characteristics of participants are presented in Table 1.

SAMPLE SIZE:

Sample size estimation was based on a formula for comparing 2 proportions, incorporating a type I error rate (α) of 0.05 and statistical power (1-β) of 0.80:

where Z(α/2) is the standard normal deviate for the desired confidence level, Z(β) is the standard normal deviate for statistical power, p₁ and p₂ are the expected proportions in the frail and non-frail groups, p is the pooled proportion, and c is the ratio of sample sizes between groups. A bilateral α of 0.05 and β of 0.20 were used. Based on previous literature, the incidence rates of frailty were p1=0.95 and p2=0.75. The required sample size was calculated at 139 cases, leading to a target of 153 patients to account for a 10% dropout rate.

INCLUSION AND EXCLUSION CRITERIA:

Inclusion criteria were: age ≥60 years, a life expectancy of at least 6 months as judged by the attending physician, and the ability to communicate and cooperate with study procedures. Patients were excluded if they had significant comorbidities likely to affect diaphragm function or frailty assessment, including stroke, Alzheimer disease, advanced malignancies, chronic obstructive pulmonary disease (COPD), and severe hepatic or renal insufficiency. Additional exclusion criteria were: bedridden status for more than 6 months, acute infections, psychiatric disorders, participation in other clinical trials, or fingertip oxygen saturation (SpO₂) ≤95%, to avoid confounding diaphragm performance due to baseline hypoxemia.

CLINICAL DATA COLLECTION:

Clinical and demographic data were collected on the day of or the day following hospital admission using a structured questionnaire. Data obtained included age, sex, height, weight, and education level. Education was categorized as junior high school and below or senior high school and above. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m²). The number and type of chronic diseases (eg, hypertension, diabetes mellitus, coronary artery disease) were recorded. Fasting venous blood (5 mL) was collected for laboratory testing using standardized procedures. Serum levels of triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), total cholesterol (TC), fasting plasma glucose (FPG), and albumin (ALB) were measured using a Cobas c701 biochemical analyzer (Roche Diagnostics, Germany). White blood cell (WBC) count was assessed with a SYSMEX XN-1000 hematology analyzer (Sysmex Corporation, Japan).

FRAILTY ASSESSMENT:

Frailty status was assessed using the Fried frailty phenotype [21], which includes 5 clinical components: unintentional weight loss, self-reported fatigue, reduced grip strength, slow walking speed, and low physical activity. Each component was scored as 1 if present and 0 if absent. A total score of 0 indicated non-frailty, while a score of 1 or more indicated frailty. Unintentional weight loss was defined as an unplanned loss of ≥5% body weight in the prior year. Exhaustion was evaluated based on responses to the CES-D questionnaire. Grip strength was measured using a Jamar dynamometer with sex- and BMI-adjusted cutoffs. Walking speed was timed over a 4.57-meter distance, and low physical activity was assessed using the Minnesota Leisure Time Activity Questionnaire.

ULTRASONIC EXAMINATION OF DIAPHRAGM:

All ultrasound assessments were performed using a GE LOGIQ e ultrasound machine (GE Healthcare, USA) by 2 trained sonographers with more than 5 years of experience. Patients were examined in a semi-recumbent position at a 30° to 45° incline. To evaluate reproducibility, 15% of all examinations were randomly selected and repeated by a second sonographer, and inter-observer reliability was quantified using intraclass correlation coefficients (ICCs) [22].

DIAPHRAGMATIC EXCURSION:

A 2–4 MHz convex probe was placed below the right costal margin along the midclavicular line. B-mode imaging was used to visualize the diaphragm, followed by M-mode scanning to capture the motion curve during deep breathing. Diaphragm displacement (excursion) was recorded over 3 consistent respiratory cycles, and the average value was used for analysis.

DIAPHRAGM THICKNESS AND CONTRACTION:

A 7–12 MHz linear probe was used to measure diaphragm thickness at the 8^th to 10^th intercostal spaces along the anterior axillary line. Images were frozen at the end of expiration and at peak inspiration to determine end-expiratory (TEE) and end-inspiratory (TEI) diaphragm thickness. The diaphragm thickness fraction (DTF) was calculated using the formula: DTF (%)=[(TEI-TEE)/TEE]×100. Measurements were performed once per subject due to resource constraints, but standard protocols were followed to minimize variability.

STATISTICAL ANALYSIS:

All data were analyzed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA). Continuous variables were expressed as mean±standard deviation (SD), and group comparisons were performed using independent-sample t-tests. Categorical variables were summarized as counts and percentages and compared using chi-square (χ²) tests. Pearson correlation was employed to assess the relationship between frailty scores and diaphragm parameters such as DTF and diaphragmatic excursion. Multivariate logistic regression was used to identify independent predictors of frailty, with frailty status (yes/no) as the dependent variable. Independent variables included age, sex, number of chronic diseases, serum ALB level, DTF, and diaphragm excursion. Results were presented as odds ratios (ORs) with 95% confidence intervals (CIs). Receiver operating characteristic (ROC) curve analysis was used to evaluate the predictive value of diaphragm ultrasound parameters for frailty. The outcome variable was frailty status (binary: frail=1, non-frail=0). Independent predictors tested included diaphragm thickness fraction (DTF) and diaphragmatic excursion during deep breathing. The area under the curve (AUC), sensitivity, and specificity were calculated using the Youden index to determine optimal cutoff values. An AUC >0.75 was considered indicative of good diagnostic performance. A two-tailed P value of <0.05 was considered statistically significant.

Results

BASELINE CHARACTERISTICS OF PATIENTS:

The study included a total of 153 elderly patients. Based on the Fried frailty scale, participants were classified into a non-frailty group (n=36) and a frailty group (n=117) (Figure 1). Statistical analysis showed significant differences in sex (χ2=4.32, P=0.037), age distribution (χ2=13.11, P=0.001), and number of chronic diseases (χ2=35.26, P<0.001) between the 2 groups. The frailty group included a higher proportion of females (74/117, 63.2%) than the non-frailty group (15/36, 41.7%). Most patients in the non-frailty group were aged 60–70 years (n=20), whereas the majority in the frailty group were aged 76–85 years (n=57). Regarding comorbidities, 62 patients in the frailty group had ≥4 chronic conditions, compared to only 6 patients in the non-frailty group. No significant differences were found in education level (χ2=1.11, P=0.292) or BMI (t=0.15, P=0.882) between the 2 groups. Mean BMI was 23.14±2.21 in the non-frailty group and 23.21±2.30 in the frailty group (Table 1).

COMPARISON OF LABORATORY PARAMETERS:

As shown in Table 2, hemoglobin levels were similar between non-frail and frail patients (119.85±8.71 g/L vs 120.04±8.94 g/L, t=0.112, P=0.911). White blood cell counts and fasting plasma glucose levels did not differ significantly between groups (P>0.05). However, serum albumin levels were significantly lower in the frailty group (34.56±3.54 g/L) compared to the non-frailty group (39.64±3.58 g/L), with t=6.031 and P<0.001. There were no significant group differences in triglycerides, total cholesterol, HDL-C, or LDL-C. These findings suggest that hypoalbuminemia is associated with frailty, potentially reflecting malnutrition or systemic inflammation.

ULTRASOUND ASSESSMENT OF DIAPHRAGM FUNCTION:

The results of diaphragm ultrasound evaluations are presented in Table 3 and Figure 2. No statistically significant differences were observed in end-inspiratory diaphragm thickness (0.24±0.08 mm vs 0.25±0.06 mm, t=0.805, P=0.423) or end-expiratory thickness (0.18±0.04 mm vs 0.15±0.09 mm, t=1.938, P=0.055). In contrast, diaphragmatic excursion during deep breathing was significantly reduced in the frailty group (49.38±2.95 mm) compared to the non-frailty group (56.94±2.41 mm), with t=13.996, P<0.001. Additionally, diaphragm thickness fraction (DTF), an indicator of contractile function, was significantly lower in the frailty group (68.12±5.42%) than in the non-frailty group (74.31±6.25%), t=5.775, P<0.001. These results suggest impaired diaphragm mobility and contraction capacity among frail elderly patients.

CORRELATION BETWEEN DIAPHRAGM FUNCTION AND FRAILTY SCORE:

Pearson correlation analysis showed a moderately strong positive correlation between diaphragm contraction (DTF) and the total frailty score, with r=0.672 and P<0.001 (Figure 3). This finding supports the hypothesis that declining diaphragm function is associated with increasing frailty severity.

MULTIVARIATE LOGISTIC REGRESSION ANALYSIS:

A multivariate logistic regression model was applied to identify independent predictors of frailty, with frailty status coded as a binary dependent variable (frail=1, non-frail=0). The model incorporated sex, age, number of chronic diseases, serum albumin (ALB), diaphragm thickness fraction (DTF), and diaphragmatic excursion as independent variables. As shown in Tables 4 and 5, older age emerged as a significant predictor of frailty (OR=42.777, 95% CI: 23.344–78.386, P<0.001), as did the number of chronic diseases (OR=1.493, 95% CI: 1.098–2.031, P=0.011). Lower DTF was also significantly associated with frailty (OR=1.613, 95% CI: 1.459–1.782, P<0.001), as was reduced diaphragmatic excursion during deep breathing (OR=1.453, 95% CI: 0.342–1.983, P<0.001). Serum albumin demonstrated a borderline association (P=0.073), whereas sex was not a statistically significant factor (P=0.850). These findings indicate that both demographic and diaphragm-related functional measures contribute meaningfully to frailty risk in elderly patients.

DIAGNOSTIC VALUE OF DIAPHRAGM ULTRASOUND PARAMETERS:

ROC curve analysis was performed to evaluate the diagnostic accuracy of DTF in identifying frailty. As shown in Figure 4, the area under the curve (AUC) was 0.846 (95% CI: 0.7602–0.9311), indicating strong discriminatory capacity. The optimal DTF cutoff value was determined using the Youden index, with a sensitivity of 81.32% and specificity of 78.33%. These findings suggest that diaphragm ultrasound measurements, particularly DTF, have clinical utility in screening elderly patients for frailty. However, external validation in multicenter cohorts is required before such measures can be adopted in routine geriatric assessments.

Discussion

Frailty is an important clinical syndrome characterized by diminished strength and endurance, reduced multi-system physiological reserve, and increased vulnerability to adverse health outcomes such as disability and mortality [23]. Recent findings indicate that frailty is particularly prevalent among the elderly, with incidence rates reaching 20.0% in those over age 65 and escalating to 45.1% among individuals over 80 [24]. The impact of frailty on the quality of life for older adults cannot be overstated, as it has been linked to increased risks of falls, cognitive decline, dementia, and higher mortality rates, which can range from 15% to 50% compared to non-frail peers [25]. Considering the ongoing economic development and improvements in living standards, global life expectancy is increasing. In China, for instance, data from the National Bureau of Statistics [26] reveal that as of 2016, the population aged 65 and older constituted 10.85% of over 1.38 billion people, with a life expectancy at birth of 76 years. As the population ages, the incidence of geriatric conditions is rising, drawing significant national attention to elderly health. The frailty associated with aging has become a focal point in geriatric research. Studies have shown that frailty significantly heightens the risk of non-Alzheimer dementia [27], emphasizing the need for targeted interventions.

Our study found a frailty incidence rate of 76.47% among 153 hospitalized elderly patients, aligning closely with the 76.40% observed by Papageorgiou et al [28]. This notably high incidence may be attributed to the generally poorer health conditions of hospitalized patients, who often contend with multiple comorbidities, in contrast to community-dwelling populations. Nutritional intake, physical inactivity, and social isolation during hospitalization may also contribute to increased frailty risk. The association between frailty and muscle dysfunction is critical, as both conditions are highly prevalent among the elderly and share modifiable risk factors and prognostic relevance. Muscle dysfunction, particularly sarcopenia, is considered a precursor to frailty and can be objectively assessed through imaging modalities such as ultrasound [29,30]. Importantly, our findings extend this concept to the respiratory system, showing that diaphragm function – specifically diaphragm thickness fraction (DTF) and excursion – is significantly associated with frailty status. This is consistent with the work of Pegorari et al, who reported a significant correlation between reduced respiratory function and increased frailty in community-dwelling older adults, highlighting that inspiratory muscle weakness can precede overt physical decline [31]. Similarly, Trevisan et al demonstrated that decreased peak expiratory flow was both cross-sectionally and longitudinally associated with frailty progression, further supporting respiratory function as a reliable early indicator of frailty [32]. In our study, Pearson correlation analysis revealed a statistically significant association between diaphragm contraction and frailty scores (r=0.672, P<0.001), supporting the growing consensus that respiratory muscle performance is integral to frailty pathophysiology.

Furthermore, in line with findings by Antoniu et al, who explored frailty as a functional limitation in patients with advanced COPD, our results indicate that diminished diaphragm movement and contractility reflect underlying sarcopenia and ventilatory inefficiency in frail individuals [33]. While their study focused on COPD, our data suggest similar respiratory vulnerability in non-COPD elderly individuals with high frailty scores.

In addition, our study observed significantly lower serum albumin levels in the frailty group. Albumin is a widely recognized marker of both nutritional status and systemic inflammation – 2 key contributors to frailty. Hypoalbuminemia may indicate diminished protein reserves, chronic inflammation, or both, and should be routinely assessed in geriatric settings.

Our logistic regression analysis further confirmed that older age, multimorbidity, lower DTF, and reduced diaphragm excursion were independent predictors of frailty. These findings are consistent with prior research linking chronic illness burden and muscle function impairment to adverse geriatric outcomes [34–36]. The ROC curve analysis revealed that DTF demonstrated strong diagnostic value for frailty, with an AUC of 0.846, sensitivity of 81.32%, and specificity of 78.33%. This suggests that quantitative diaphragm assessments could be integrated into frailty screening protocols in high-risk elderly populations.

While these findings are promising, several limitations must be acknowledged. First, the sample size, although sufficient for internal analysis, limits generalizability and statistical power for subgroup comparisons. Second, the study employed a cross-sectional case-control design, which precludes causal inferences. Third, although diaphragm ultrasound is reliable, measurements were performed at a single time point by 2 operators, which may have introduced operator variability despite standardized protocols. Fourth, relevant confounders such as physical activity, nutritional intake, respiratory effort, and undiagnosed neuromuscular disease were not comprehensively evaluated. Fifth, our frailty assessment relied exclusively on the Fried phenotype, without inclusion of cognitive or psychosocial dimensions. Lastly, the lack of significant differences in biochemical markers aside from albumin may reflect sample size limitations or non-differential physiological profiles.

Conclusions

This study demonstrates a high prevalence of frailty among hospitalized elderly patients and identifies diaphragm dysfunction – specifically, reduced thickness fraction and excursion – as significantly associated with frailty status. Diaphragm ultrasonography may serve as a useful adjunct in the clinical assessment of frailty. However, these findings are based on a single-center cohort and should be interpreted with caution. Further multicenter, prospective studies are needed to confirm the diagnostic utility, establish causality, and evaluate the role of diaphragm ultrasound in routine geriatric screening.