Risk factors for anxiety and depression in elderly frail patients based on bedside ultrasound multimodal assessment

Abstract

BACKGROUND

Anxiety and depression represent frequent complications among elderly frail patients that substantially impact functional recovery and quality of life. While bedside ultrasound multimodal assessment might serve as a valuable tool for identifying high-risk patients, the specific parameters underlying its predictive value remain incompletely understood.

AIM

To examine how bedside ultrasound multimodal parameters affect anxiety and depression development in elderly frail patients.

METHODS

This retrospective study enrolled 125 elderly frail patients admitted to our hospital’s geriatrics department between January 2022 and December 2024. Based on bedside ultrasound multimodal assessment severity, participants were categorized into a: (1) Low-risk cohort (mild abnormality, n = 56); and (2) High-risk cohort (moderate-severe abnormality, n = 69). The evaluated ultrasound parameters included diaphragm thickness fraction, rectus femoris cross-sectional area, inferior vena cava collapsibility index, cardiac output, and lung ultrasound score. Anxiety and depression assessment was performed using the Hospital Anxiety and Depression Scale 3 months after admission.

RESULTS

The study cohort included 125 elderly frail patients (average age 76.8 ± 8.5 years, 52.8% males). Anxiety developed in 42 patients (33.6%) and depression in 48 patients (38.4%). Three independent risk factors were identified: (1) Low diaphragm thickness fraction (< 20%) [odds ratio (OR) = 3.68, 95% confidence interval (CI): 1.58-8.57, P = 0.002]; (2) Reduced rectus femoris cross-sectional area (< 4.0 cm²) (OR = 2.95, 95% CI: 1.42-6.13, P = 0.004); and (3) Elevated lung ultrasound score (≥ 12 points) (OR = 2.67, 95% CI: 1.28-5.56, P = 0.009). A clear dose-response relationship was observed (Spearman r = 0.452, P < 0.001).

CONCLUSION

Bedside ultrasound multimodal assessment abnormalities constitute independent predictors of anxiety and depression in elderly frail patients. Early ultrasound-based risk stratification may facilitate targeted psychological screening and timely preventive interventions in this vulnerable population.

Key Words: Elderly frailty; Bedside ultrasound; Multimodal assessment; Anxiety; Depression; Risk factors; Retrospective study

Core Tip: Bedside ultrasound provides fast, non-invasive evaluation for multisystem dysfunction in elderly frail patients. The study shows that diaphragm dysfunction and sarcopenia are independently predictive of anxiety and depression, with a demonstrated dose-response relationship that extends to pulmonary abnormalities diagnosed at the bedside by multimodal ultrasound. The use of early ultrasound-based risk stratification may allow for earlier psychosocial screening and targeted interventions in this high-risk population.

INTRODUCTION

Frailty is a common but complex frail elderly syndrome characterized by fewer body reserves and greater susceptibility to acute morbidity or mortality. According to a systematic review by Collard et al[1], among community-dwelling elderly persons, the prevalence of frailty is 10.7%, and this increases with hospitalization (52.3%)[2]. As the population ages rapidly in China, elderly frail patients have become a major part of the healthcare system, creating significant challenges[2]. In addition to increasing the risk of falls, hospitalization and mortality, frailty has an adverse psychological impact on patients, leading to anxiety and depression[3].

Psychological disorders in elderly frail patients, such as anxiety and depression, can be characterized by excessive worrying, fear, unremitting low mood and loss of interest[4]. Recently, meta-analysis of 25 studies by Soysal et al[3] revealed a pooled prevalence of depression among frail elderly populations as high as 40.4% (95%CI: 36.8%-44.1%), and up to 20%-45% for anxiety states[5]. These psychological sequela lead to a profound impact on patients’ quality of life and social functioning, delay physical rehabilitation and increase healthcare utilization[6,7], lengthen the hospital stay, and increase mortality rates[7].

The pathogenesis of anxiety and depression in frail elderly individuals is likely caused by a combination of neurobiological, psychosocial and physical factors[8]. Neurobiological mechanisms primarily involve functional deficiencies of neurotransmitter systems (serotonin, norepinephrine), chronic inflammation, and oxidative stress[9]. Psychosocial factors include social isolation, loss of independence, and caregiver burden[10]. In recent years, the role of physical function decline in the anxiety and depression pathogenesis has received increasing attention[11].

Bedside ultrasound has emerged as a valuable tool for comprehensive assessment of multiple organ systems in critically ill and frail patients[12]. Point-of-care ultrasound allows rapid, non-invasive evaluation of diaphragm function, muscle mass, cardiac function, volume status, and pulmonary conditions[13]. However, research on the relationship between bedside ultrasound multimodal parameters and psychological outcomes in elderly frail patients is still limited[14,15].

Based on the above background, this retrospective cohort study analyzes the relationship between bedside ultrasound multimodal parameters and the occurrence of anxiety and depression in elderly frail patients, aiming to explore the following scientific questions: (1) Whether specific ultrasound parameters are independent risk factors for anxiety and depression; (2) Whether there is a dose-response relationship between the number of abnormal parameters and psychological distress risk; and (3) Whether different types of ultrasound parameters have different impacts on anxiety and depression occurrence[16].

MATERIALS AND METHODS

Study design

This study employed a retrospective design to explore the impact of bedside ultrasound multimodal parameters on the occurrence of anxiety and depression in elderly frail patients by collecting and analyzing patients’ clinical data. The study protocol was reviewed and approved by Cangzhou Central Hospital (No. 2024-315-02), and all data collection and usage strictly adhered to the relevant provisions of the Declaration of Helsinki. Due to the retrospective study design, patient informed consent was waived, but all patient data were anonymized to ensure patient privacy and safety.

Study population

The study subjects were elderly frail patients hospitalized in the geriatrics department of our hospital from January 2022 to December 2024. A total of 186 elderly frail patients were initially screened for eligibility. Inclusion criteria included: (1) Age ≥ 65 years; (2) Meeting the Fried frailty phenotype criteria [≥ 3 of 5 criteria: Unintentional weight loss (≥ 10 pounds in past year), self-reported exhaustion (≥ 3 days/week), handgrip strength in lowest 20% by gender/body mass index, slow walking speed (lowest 20% by gender/height), and low physical activity (kcal/week in lowest 20%)]; (3) Completed bedside ultrasound multimodal assessment within 48 hours of admission; (4) Clear consciousness upon admission, able to cooperate with psychological assessment; (5) Complete medical records, including detailed past medical history, medication history, and laboratory examination results; and (6) Able to complete 3-month follow-up, including anxiety and depression scale assessment.

Exclusion criteria: (1) Previous history of diagnosed anxiety or depression, or currently taking anxiolytic or antidepressant medications; (2) Severe cognitive impairment (Mini-Mental State Examination score < 18); (3) Complicated with severe organ failure or terminal illness with life expectancy < 6 months; (4) Unable to cooperate with ultrasound examination due to severe obesity or other reasons; (5) History of neurodegenerative diseases such as Parkinson’s disease or dementia; (6) Death or loss to follow-up during the follow-up period; and (7) Incomplete ultrasound data affecting parameter determination. Of the 186 patients screened, 61 were excluded: (1) 18 for pre-existing mood disorders; (2) 15 for severe cognitive impairment; (3) 12 for terminal illness; (4) 9 for inability to complete ultrasound examination; (5) 5 for neurodegenerative diseases; and (6) 2 for incomplete data. According to the above criteria, a total of 125 eligible elderly frail patients were included.

Bedside ultrasound multimodal assessment

The bedside ultrasound multimodal parameters assessed included the following: (1) Diaphragm thickness fraction (DTF): Measured at the zone of apposition using a high-frequency linear probe, calculated as (thickness at end-inspiration - thickness at end-expiration)/thickness at end-expiration × 100%, with DTF < 20% defined as diaphragm dysfunction; (2) Rectus femoris cross-sectional area (RF-CSA): Measured at the junction of the middle and lower third of the thigh using B-mode ultrasound, with RF-CSA < 4.0 cm² defined as sarcopenia; (3) Inferior vena cava collapsibility index (IVC-CI): Calculated as (maximum diameter - minimum diameter)/maximum diameter × 100%, with IVC-CI > 50% defined as hypovolemia; (4) Cardiac output (CO): Calculated using the velocity-time integral at the left ventricular outflow tract, with CO < 4.0 L/minute defined as reduced CO; and (5) Lung ultrasound score (LUS): Evaluated using a 12-zone protocol, with LUS ≥ 12 points defined as significant pulmonary abnormality.

Risk stratification grouping

According to the number of abnormal ultrasound parameters for each patient, study subjects were divided into two groups: (1) Low-risk group (0-2 abnormal parameters, n = 56); and (2) High-risk group (≥ 3 abnormal parameters, n = 69). The cutoff threshold of ≥ 3 abnormal parameters for defining the high-risk group was determined through receiver operating characteristic curve analysis, which demonstrated that this threshold achieved the best balance between sensitivity (72.5%) and specificity (70.8%) for predicting anxiety/depression (Youden index = 0.433, area-under-curve = 0.768).

Assessment of anxiety and depression

All patients underwent anxiety and depression assessment 3 months after admission. The Hospital Anxiety and Depression Scale (HADS) was used to assess patients’ psychological status. HADS is a widely used screening tool with good reliability and validity in elderly populations. The scale includes 14 items, with 7 items assessing anxiety (HADS-A) and 7 items assessing depression (HADS-D). Each item is scored from 0-3 points. A HADS-A total score ≥ 8 was defined as the presence of anxiety symptoms, and a HADS-D total score ≥ 8 was defined as the presence of depressive symptoms. All scale assessments were completed by trained geriatricians with consistency training conducted before assessment.

Statistical analysis

Data analysis was conducted using SPSS 26.0 with statistical significance set at P < 0.05. Normality was assessed using the Shapiro-Wilk test combined with visual inspection of Q-Q plots and histograms. Continuous variables were described as mean ± SD or median (interquartile range) based on normality testing, while categorical variables were expressed as n (%). Between-group comparisons used t-tests or Mann-Whitney U tests for continuous variables, and χ² or Fisher’s exact tests for categorical variables. Spearman correlation was performed to assess the relationship between ultrasound parameters and psychological scores due to non-normal distribution of several parameters (DTF, LUS) and HADS scores, as confirmed by Shapiro-Wilk tests (P < 0.05). Logistic regression analysis examined associations between ultrasound abnormalities and anxiety/depression, with univariate analysis (P < 0.1) followed by multivariate modeling using the enter method to calculate ORs and 95%CI.

RESULTS

Baseline characteristics of study subjects

A total of 125 elderly frail patients were included, comprising 66 males (52.8%) and 59 females (47.2%), aged 65-92 years, with a mean age of 76.8 ± 8.5 years. There were no significant differences in age and gender between groups (P > 0.05). Low DTF was the most common abnormality (56.0%), followed by reduced RF-CSA (48.8%), elevated LUS (44.0%), elevated IVC-CI (36.0%), and reduced CO (28.8%). The number of abnormal parameters was significantly higher in the high-risk group (3.8 ± 0.9 vs 1.4 ± 0.6, P < 0.001; Table 1).

Table 1 Baseline characteristics of study subjects, n (%)/mean ± SD/mean (interquartile range).

Characteristics	Total (n = 125)	Low-risk (n = 56)	High-risk (n = 69)	P value
Age (years)	76.8 ± 8.5	75.6 ± 8.2	77.8 ± 8.7	0.158
Male gender	66 (52.8)	29 (51.8)	37 (53.6)	0.841
CFS score, median (range)	6 (5-8)	6 (5-7)	6 (5-8)	0.201
Low diaphragm thickness fraction (< 20%)	70 (56.0)	18 (32.1)	52 (75.4)	< 0.001
Reduced rectus femoris cross-sectional area (< 4.0 cm2)	61 (48.8)	15 (26.8)	46 (66.7)	< 0.001
Elevated inferior vena cava collapsibility index (> 50%)	45 (36.0)	12 (21.4)	33 (47.8)	0.002
Reduced cardiac output (< 4.0 L/minute)	36 (28.8)	10 (17.9)	26 (37.7)	0.016
Elevated lung ultrasound score (≥ 12 points)	55 (44.0)	14 (25.0)	41 (59.4)	< 0.001

Overall incidence of anxiety and depression

During the 3-month follow-up, 42 patients (33.6%) developed anxiety and 48 patients (38.4%) developed depression. Combined anxiety and depression occurred in 32 patients (25.6%). HADS-A scores ranged from 2-18 points (mean 6.8 ± 3.6), and HADS-D scores ranged from 2-19 points (mean 7.2 ± 3.8). Among anxiety patients, common symptoms included excessive worry (100%), restlessness (88.1%), and sleep difficulties (83.3%). Among depression patients, common symptoms included loss of interest (100%), fatigue (91.7%), and pessimism (79.2%, Figure 1A).

Figure 1 Comparison of different groups. A: Comparison of Hospital Anxiety and Depression Scale scores among different patient groups. Patients with anxiety/depression had significantly higher scores; B: Comparison of functional outcomes between risk groups. The low-risk group significantly outperformed the high-risk group. ^cP < 0.001. ADL: Activities of daily living; HADS: Hospital Anxiety and Depression Scale; HADS-A: Hospital Anxiety and Depression Scale-anxiety; HADS-D: Hospital Anxiety and Depression Scale-depression.

Comparison of functional outcomes between groups

As illustrated in Figure 1B, the low-risk group (0-2 abnormal items) demonstrated consistently superior outcomes across all four indicators compared with the high-risk group (≥ 3 abnormal items), with statistically significant differences observed in each domain (all P < 0.05). The low-risk group achieved higher rates of good functional outcome (71% vs 50%), mobility independence (74% vs 57%), and ADL independence (71% vs 53%), as well as a higher mean Barthel Index score (80 points vs 66 points). These findings suggest that the number of abnormal items at baseline is a meaningful predictor of rehabilitation outcomes, highlighting the clinical importance of early risk stratification in guiding recovery planning.

Univariate analysis of risk factors

Univariate logistic regression showed: (1) Age ≥ 80 years (OR = 2.56, 95%CI: 1.22-5.37, P = 0.013); (2) Low DTF (OR = 3.92, 95%CI: 1.82-8.44, P < 0.001); (3) Reduced RF-CSA (OR = 3.24, 95%CI: 1.58-6.64, P = 0.001); and (4) Elevated LUS (OR = 2.86, 95%CI: 1.42-5.76, P = 0.003) were significant risk factors (Table 2).

Table 2 Univariate analysis of risk factors for anxiety and depression.

Risk factor	Odds ratio	95%CI	P value	Significance
Age ≥ 80 years	2.56	1.22-5.37	0.013	Significant
Female gender	1.95	0.98-3.89	0.058	NS
Low diaphragm thickness fraction (< 20%)	3.92	1.82-8.44	< 0.001	Significant
Reduced rectus femoris cross-sectional area (< 4.0 cm2)	3.24	1.58-6.64	0.001	Significant
Elevated inferior vena cava collapsibility index (> 50%)	1.68	0.82-3.44	0.156	NS
Reduced cardiac output (< 4.0 L/minute)	1.52	0.72-3.21	0.272	NS
Elevated lung ultrasound score (≥ 12 points)	2.86	1.42-5.76	0.003	Significant

NS: Not significant.

Multivariate analysis of risk factors

Three independent risk factors were identified: (1) Low DTF (OR = 3.68, 95%CI: 1.58-8.57, P = 0.002); (2) Reduced RF-CSA (OR = 2.95, 95%CI: 1.42-6.13, P = 0.004); and (3) Elevated LUS (OR = 2.67, 95%CI: 1.28-5.56, P = 0.009). The model showed good fit (Hosmer-Lemeshow P = 0.685) and discriminative ability (C-statistic = 0.768, Figure 2).

Figure 2 Forest plot of multivariate logistic regression analysis. Blue squares indicate significant associations (P < 0.05); gray squares indicate non-significant associations. CO: Cardiac output; DTF: Diaphragm thickness fraction; IVC-CI: Inferior vena cava collapsibility index; LUS: Lung ultrasound score; RF-CSA: Rectus femoris cross-sectional area.

Correlation between ultrasound parameters and anxiety/depression severity

Spearman correlation analysis revealed significant correlations between ultrasound parameters and HADS scores. DTF showed moderate negative correlations with HADS-A (r = -0.428, P < 0.001), HADS-D (r = -0.395, P < 0.001), and total HADS score (r = -0.438, P < 0.001). RF-CSA also showed negative correlations (HADS-A: r = -0.385; HADS-D: r = -0.412; total: r = -0.418; all P < 0.001). LUS showed positive correlations with psychological distress (HADS-A: r = 0.362; HADS-D: r = 0.345; total: r = 0.378; all P < 0.001). The cumulative number of abnormal parameters showed the strongest correlation with total HADS score (r = 0.452, P < 0.001; Table 3).

Table 3 Correlation between ultrasound parameters and anxiety/depression severity.

Parameter	HADS-anxiety r value	P value	HADS-depression r value	P value	Total r value
Diaphragm thickness fraction (%)	-0.428	< 0.001	-0.395	< 0.001	-0.438
Rectus femoris cross-sectional area (cm2)	-0.385	< 0.001	-0.412	< 0.001	-0.418
Inferior vena cava collapsibility index (%)	0.186	0.038	0.168	0.062	0.192
Cardiac output (L/minute)	-0.152	0.091	-0.178	0.047	-0.172
Lung ultrasound score (points)	0.362	< 0.001	0.345	< 0.001	0.378
Number of abnormal parameters	0.436	< 0.001	0.425	< 0.001	0.452

HADS: Hospital Anxiety and Depression Scale.

Dose-response relationship analysis

Stratified analysis revealed a clear dose-response relationship: Incidence rates were 15.4% (4/26) in the 0-1 abnormal parameter group, 26.7% (8/30) in the 2 parameters group, 48.5% (16/33) in the 3 parameters group, 60.9% (14/23) in the 4 parameters group, and 71.4% (10/14) in the ≥ 5 parameters group (P < 0.001; Figure 3).

Figure 3 Dose-response relationship between abnormal ultrasound parameters and anxiety/depression incidence. The black trend line indicates a linear increase in incidence rate (Spearman r = 0.452, P < 0.001).

Diagnostic performance of prediction models

We employed receiver operating characteristic curve analysis to compare the diagnostic performance of different prediction models. Single parameter models showed area under the curve (AUC) values of 0.682 (DTF), 0.658 (RF-CSA), and 0.635 (LUS). The combination of DTF + RF-CSA improved AUC to 0.728. The three-parameter model (DTF + RF-CSA + LUS) achieved AUC of 0.768 with a sensitivity of 76.2% and specificity of 70.5%. The full five-parameter model showed the highest AUC (0.782), with a sensitivity of 78.6% and specificity of 69.2%. The simplified risk stratification using ≥ 3 abnormal parameters achieved AUC of 0.718, providing a practical clinical screening tool (Table 4).

Table 4 Comparison of diagnostic performance of prediction models.

Model	Area under the curve	95%CI	Sensitivity	Specificity	Positive predictive value/negative predictive value
DTF alone	0.682	0.59-0.77	71.4%	60.3%	53.6%/76.4%
RF-CSA alone	0.658	0.56-0.75	66.7%	62.8%	52.5%/75.4%
LUS alone	0.635	0.54-0.73	61.9%	65.4%	52.0%/74.0%
DTF + RF-CSA	0.728	0.64-0.82	73.8%	66.7%	57.4%/80.8%
DTF + RF-CSA + LUS	0.768	0.68-0.85	76.2%	70.5%	61.5%/82.7%
All 5 parameters	0.782	0.70-0.86	78.6%	69.2%	61.1%/84.2%
≥ 3 abnormal parameters	0.718	0.63-0.81	72.5%	70.8%	60.9%/80.4%

DTF: Diaphragm thickness fraction; LUS: Lung ultrasound score; RF-CSA: Rectus femoris cross-sectional area.

Sensitivity analysis and subgroup analysis

When diagnostic criteria were adjusted to HADS ≥ 11, incidence rates were 52.2% (36/69) in the high-risk group and 26.8% (15/56) in the low-risk group (χ² = 8.12, P = 0.004). Gender-stratified analysis showed significant differences in both males (48.6% vs 24.1%, P = 0.034) and females (56.3% vs 29.6%, P = 0.034). Age-stratified analysis revealed stronger effects in patients ≥ 80 years (OR = 4.82, 95%CI: 1.76-13.18) compared with 65-79 years (OR = 2.28, 95%CI: 0.98-5.31; Table 5).

Table 5 Sensitivity analysis and subgroup analysis.

Analysis	Subgroup	High-risk	Low-risk
Hospital Anxiety and Depression Scale ≥ 11 criteria	Overall (n = 125)	52.2% (36/69)	26.8% (15/56)
Gender-stratified	Male (n = 66)	48.6% (18/37)	24.1% (7/29)
Gender-stratified	Female (n = 59)	56.3% (18/32)	29.6% (8/27)
Age-stratified	≥ 80 years	OR = 4.82	95%CI: 1.76-13.18
Age-stratified	65-79 years	OR = 2.28	95%CI: 0.98-5.31

OR: Odds ratio.

DISCUSSION

This retrospective study sheds new light onto the association between multimodal parameters from bedside ultrasound and the occurrence of anxiety/depression in elderly frail patients, helping to unravel complex pathophysiological mechanisms accounting for the high burden of psychological complications in this vulnerable population. These results are of major importance in clinical practice and indicate the necessity of a comprehensive assessment of geriatric patients that combines both physical and psychological evaluation.

Our results confirm and expand findings about the association between physical frailty and psychological outcomes. This was supported by the finding of Makizako et al[17], who found that physical frailty is a strong predictor of incident depressive symptoms with a hazard ratio of 2.73, whereas Arts et al[11] found that low-grade inflammation provided a basis from which possible depression in the frail elderly could develop. However, these studies used clinical frailty scales or functional assessments almost exclusively. We developed an ultrasound method that offers fast, quantitative organ-level data in a manner that may lead to earlier diagnosis. Our dose response relationship, with an incidence of 15.4% in those with 0-1 abnormalities and increasing to 71.4 for those with ≥ 5 abnormalities, demonstrates new evidence that cumulative organ dysfunction is an important driver of psychological vulnerability. This extends beyond Lahousse et al[18], which demonstrated that irrespective of frailty severity, pathophysiological markers of specific physiological traits are independently predictive of mood disorder.

It is a two-way street between functional outcomes and psychological distress. We provide evidence supporting three potentially mechanistic pathways, but the proposed pathophysiological mechanisms are speculative and should be confirmed in future prospective mechanistic studies. First, a common upstream cause of multiple organ dysfunction detected by ultrasound, which further dysregulates neurobiological pathways, predisposing to anxiety/depression and vice versa, may simultaneously predict deterioration of physical function, leading to poor recovery. As part of the physiological insult, chronic hypoxia, inflammation and metabolic derangements that accompany organ dysfunction may provide a substrate for damage to both physical and mental health trajectories in an independent pathway. Second, poor functional recovery may itself act as a mediator; physical disability commonly results in loss of independence, social isolation and psychological distress. Third, anxiety and depression can both decrease rehabilitation participation and motivation, forming a vicious cycle[19].

This observed association with anxiety and depression and bedside ultrasound abnormalities can be explained by several interlinked pathophysiological pathways that remain to be directly tested. Low DTF indicates diaphragm dysfunction, which may play a role in anxiety and depression through multiple mechanisms[20]. First, impaired diaphragm function causes dyspnea and decreased exercise tolerance, both major anxiety-provoking factors. Second, diaphragm weakness has been associated with sleep-disordered breathing[19] which in turn correlates with both anxiety and depression[21]. Third, diaphragm dysfunction may hinder vagal tone and lead to psychological disorder-related autonomic dysregulation by impairing parasympathetic activity.

Sarcopenia caused by decreased RF-CSA links psychological distress via divergent pathways[22]. There are two main factors that compound these issues, one short term and the other long term. Second, sarcopenia is related to systemic inflammation and high proinflammatory cytokines (interleukin-6, tumor necrosis factor-alpha, C-reactive protein), which may contribute to depression pathogenesis[23]. Third, skeletal muscle itself is an endocrine organ, and muscle loss may reduce levels of beneficial myokines, such as brain-derived neurotrophic factor, which plays a key role in mood.

Pulmonary lesions identified with lung ultrasound may be a risk factor for psychological complications through direct and indirect mechanisms[24]. LUS scores are generally higher whenever significant pulmonary congestion, interstitial edema or atelectasis occur, all of which affect gas exchange and lead to poor tissue oxygenation. Chronic hypoxemia was associated with altered cognition, mood and anxiety[25]. In addition, pulmonary complications also require oxygen therapy, long hospitalization, and intensive monitoring, which cause psychological stress and decreased quality of life.

Identifying bedside ultrasound parameters as independent predictors of anxiety and depression has significant clinical implications. Geriatric protocols currently aim at evaluation and therapy of physical function, with little systematic work on psychological complications[26]. Our results indicate that multimodal assessment of routine bedside ultrasound could be a useful tool in identifying patients at risk for psychological distress and early intervention mechanisms.

Investigations that report a dose–response relationship between the number of abnormal ultrasound parameters and severity of psychological distress provide additional support to the notion that cumulative organ dysfunction, rather than isolated abnormalities, is responsible for neuropsychiatric outcomes[27]. This finding is consistent with new data emerging from frailty, where multi-system impairment appears to be linked to faster aging and higher susceptibility or fragility and several related complications. Early identification of high-risk patients during the hospitalization course may also increase the judicious deployment of mental health resources and aid in targeted preventive strategies[28].

There are immediate clinical implications for our findings in the geriatric setting. Bedside ultrasound-based multimodal assessment provides a non-invasive screening tool that is available without further laboratory assessments. We propose a three-tier approach to risk stratification based on the number of abnormal parameters: (1) Low-risk (0-2 abnormal parameters); (2) Intermediate-risk (3 abnormal parameters); and (3) High-risk (≥ 4 abnormal parameters), and corresponding intervention intensities are then offered. This should include psychologic screening periodically (2 weeks, 1 month and 3 months postadmission) in high-risk patients, possibly followed by preventive interventions[29]. Such an approach could be easily integrated into current geriatric care pathways, with limited excess resource implications.

The general incidence of anxiety and depression found in this study is consistent with previous studies capturing rates between 25% (2) and 50% depending on assessment timepoint, diagnostic criteria, and population[30]. Nevertheless, most previous studies have been conducted in Western populations and there is sparse evidence specifically for the Chinese elderly, where disparities in cultural dimensions, healthcare delivery patterns and risk factor profiles potentially exist. The high prevalence of diaphragm dysfunction and sarcopenia in our study population mirror the physical manifestations of frailty documented in older patients with Chinese ancestry.

These results point to several specific directions for future research. First, a prospective longitudinal cohort study with monthly ultrasound and psychological measures over 12 months would provide a more thorough understanding of the temporal patterning of psychological distress and assist in defining appropriate intervention windows. Second, a randomized controlled trial that assesses whether targeted respiratory muscle training is effective in decreasing the incidence of anxiety among patients with low DTF would provide key evidence for causality and clinical guidelines. Third, mediation pathways linking inflammatory cytokines (interleukin-6, tumor necrosis factor-alpha), myokine (e.g. irisin), neurotrophic factors (brain-derived neurotrophic factor) and stress hormones (e.g., cortisol) should be combined with ultrasound characteristics and psychological assessments[31,32].

This study has several limitations. The retrospective design limits our ability to infer causal relationships. Second, the single-center setting may confine generalizability. Finally, the HADS scale, while valid in multiple countries and cultures, does not include some psychological problems. Moreover, we did not evaluate trends in ultrasound findings over time that might have prognostic value. This fifth source of bias is the absence of a systematic record regarding follow-up treatments (such as anxiolytics, antidepressants, rehabilitation frequency and culinary treatment), an important fact that suggests that there could be more cross-influencing variables for psychological outcomes. Sixth, the role of potential confounding factors (e.g., social support and coping strategies) was not systematically evaluated[33-36].

CONCLUSION

Bedside ultrasound multimodal assessment abnormalities constitute independent predictors of anxiety and depression in elderly frail patients. The combination of DTF, RF-CSA, and LUS provides optimal predictive performance, supporting integration of multimodal ultrasound assessment into routine geriatric care for early identification of high-risk patients.