Quadriceps muscle thickness as measured by point-of-care ultrasound is associated with hospital length of stay among hospitalised older patients

Abstract

Background

Predicting hospital length of stay (LOS) can potentially improve healthcare resource allocation. Recent studies suggest that point-of-care ultrasound (POCUS), specifically measurements of muscle thickness (MT), may be valuable in assessing patient outcomes, including LOS. This study investigates the hypothesis that quadriceps MT and echo intensity (EI) can predict patient outcomes, particularly LOS.

Methods

Quadriceps MT and EI were measured using POCUS in patients admitted to a hospital’s acute medical unit. Predictor variables included age, sex, MT, EI and the Charlson Comorbidity Index (CCI). The outcome variable was hospital LOS.

Results

One hundred twenty participants were included (average age 76 ± 7, with 64 women and 56 men). The mean LOS was 27 ± 31 days, and the mean MT was 20 ± 6 mm. Sex-based differences in MT were statistically significant (P = .032). Patients with prolonged LOS over 30 days had lower MT (mean 17 mm vs. 21 mm, P < .0001). One unit increase in MT was significantly associated with ~1.5 fewer days of hospital LOS, and one CCI score increase was associated with almost three more days of hospital LOS. Having low MT significantly increased the odds of staying in the hospital longer than 30 days by more than three times in all models.

Conclusion

Muscle thickness is a strong predictor of hospital LOS, highlighting the potential of POCUS for assessing patient outcomes.

Key Points

• Muscle thickness (MT), as measured by point-of-care ultrasound, can predict hospital length of stay.

• A 1 cm increase in MT was associated with approximately two days of longer hospital stay.

• Point-of-care ultrasound of quadriceps MT can be a valuable tool for evaluating patient outcomes.

Introduction

Hospital length of stay (LOS), defined as the number of days from admission to discharge, is a hospital management and patient outcome indicator directly affecting expenditure, costs and patient satisfaction [1]. Predicting hospital LOS in older adults is critical for optimizing healthcare resource allocation and reducing hospital costs while maintaining patient care quality [2, 3]. Prolonged LOS was associated with increased hospital expenditure, lower availability of acute beds and the risk of infection, further delaying discharge in older patients [4, 5].

Frailty is an increased state of vulnerability to stressors leading to declines in homeostatic reserve and resiliency [6]. Frailty increases the risk of hospitalisation and admission to acute care, while also hospitalisation makes older adults prone to new or worsening frailty [7]. Sarcopenia, the age-related loss of skeletal muscle mass, strength and physical function, is a key component of frailty, [8] and it was associated with prolonged length of hospital stay [9]. The low muscle mass component of sarcopenia is typically assessed using magnetic resonance imaging (MRI), computed tomography (CT), or dual-energy X-ray absorptiometry (DXA), which are not bedside-accessible for hospitalised older patients [10]. However, point-of-care ultrasound (POCUS) has gained attention as a bedside tool for assessing muscle mass by a surrogate measure as muscle thickness (MT) and muscle quality as echo intensity (EI) [11–13]. POCUS-measured MT was significantly correlated with skeletal muscle index [14] and cross-sectional area [15] by CT, appendicular skeletal muscle mass by DXA [16] and lean body mass by bioimpedance analysis [17]. The commonly measured components using POCUS are MT, pennation angle, fascicle length, EI and cross-sectional area [18]. Among these measures, MT was the most common proxy measure for muscle mass in systemic reviews [11, 19]. Echo intensity was proposed for skeletal muscle qualitative changes as fat and fibrous tissue infiltration increases EI in the captured ultrasound image of muscle [20, 21]. Ultrasound measures of muscle mass, particularly in large muscles such as quadriceps femoris, showed a high validity compared to DXA, MRI and CT and a good reliability with high inter- and intra-rater reliability [44]. Another study by Thomaes et al. also found that ultrasound techniques of the rectus femoris muscle are a valid and reliable technique for older cardiac patients compared to CT cross-sectional area [45]. Intra- and inter-observer reliability between two physicians in ICU critically ill patients showed good reliability, ranging from 0.74–0.83 to 0.76–0.81, depending on the measurement sites [46].

Based on the following studies, POCUS-measured muscle mass can be used to predict clinical outcomes such as extended LOS. A pilot study by Salim et al. demonstrated the potential of thigh ultrasound in identifying sarcopenic frail older patients undergoing surgery, revealing that muscle assessments through ultrasound could help identify older adults at risk for prolonged LOS and poor outcomes [22]. Another study by Canales et al. found that preoperative quadriceps MT, measured via POCUS, was a significant predictor of frailty and adverse postoperative outcomes, including prolonged LOS, unplanned admissions to skilled nursing facilities and delirium [23]. Levine et al. reported that lower quadriceps muscle layer thickness, measured using ultrasound, correlated with extended hospital stays in renal transplant patients, reinforcing the utility of ultrasound in predicting extended recovery periods [24].

We hypothesised that quadriceps MT and EI, measured via POCUS, are significant predictors of LOS due to their association with clinical recovery and functional capacity. Our study aimed to:

1. Examine the association between LOS and quadriceps MT and EI as measured by POCUS.

2. Evaluate the impact of MT and EI on prolonged LOS by calculating odds ratios, highlighting the clinical relevance of ultrasound-based muscle assessment.

3. Compare the odds of prolonged LOS due to low MT and high EI with those associated with a high Charlson Comorbidity Index (CCI), a widely used predictor of hospital LOS.

Materials and methods

Participants

Older adult patients were recruited consecutively for a cross-sectional study. One hundred twenty-five patients were assessed for eligibility. Four declined to participate, and one was removed due to steroid use. All subjects provided informed consent before testing.

Inclusion criteria:

We included all patients 65 and older admitted to the acute medicine service at an academic centre from February to September 2022.

Exclusion criteria:

1. Less than 65 years of age.

2. Patients were excluded if an ultrasound could not be performed due to conditions such as fractures that restricted access to the measurement sites (fractures affecting arms and legs, leg oedema, etc.).

3. Patients admitted to an intensive care unit (ICU).

4. Patients with a prior diagnosis of a muscle or neurological disorder affecting muscle mass or quality (e.g. chronic neuromuscular disorders).

5. Having a past diagnosis of a chronic illness that results in cachexia (metastatic cancer, severe malnutrition, etc.).

6. Subjects that are on systemic corticosteroids.

7. All non-ambulatory patients.

8. All patients with end-stage organ failure (end-stage chronic obstructive pulmonary disease (COPD), renal failure requiring dialysis, end-stage heart failure).

At the time of admission, patients’ age and sex were collected at the baseline, along with POCUS measures of quadriceps MT and EI. Length of stay was recorded at discharge. Based on the discharge summary notes, the CCI was calculated on www.mdcalc.com [25].

Consent/ethics

The study protocol was approved by the UBC Research Ethics Board. The IRB number is H21-03838. All patients consented to participate.

Sample size justification

Based on a linear regression model calculation with a medium effect size (R² = 0.07, α = 0.05, power = 0.80), we needed 107 participants. With a 10% dropout rate, we expected approximately 119 participants, rounding up to 120 for the final sample size.

Ultrasound

Ultrasound measurements were performed upon admission to the acute care medicine service after they had agreed to participate in the study. Muscle thickness was measured in B-mode using a POCUS device (GE Vscan Handheld Ultrasound). A single operator (P.T.) captured cross-sectional images of the quadriceps muscles and measured the thickness, as shown in Figure 1. P.T. was an internal medicine resident and was trained in the use of bedside ultrasound by the head of the bedside ultrasound program (S.A.). As per previously described protocols [18, 26] and our previous publications [17, 27], Rectus Femoris and Vastus Intermedius MT was measured in the supine position with knees rested. A measuring tape was used to mark the midpoint between the right greater trochanter and condyle or the right femur (or midpoint between the anterior superior iliac spine and the upper edge of the patella). A linear 9–12 MHz probe was then placed perpendicularly at a 90-degree angle at this midpoint with transmission gel. The operator ensured adequate gel and applied only enough or minimal force to allow adequate contact but avoid muscle compression. The beam angle was oriented as perpendicular as possible to the anterior surface of the femur. The image was then frozen, and the ideal frame was chosen from stored images. Callipers were used to measure the distance from the superficial fascia bordering the rectus femoris and the deep fascia at the interface between the rectus intermedius and the anterior aspect of the femur. Echo intensity was defined by the pixel intensity, measured by (B.F.) importing the captured images to Image J software (Bethesda, MD). Echo intensity was expressed as arbitrary units between a score from Black (0) to White (255).

Statistical analysis

Data analysis was performed using R version 2023.12.1. All descriptive variables were expressed in mean (standard deviation [SD]). We assessed the distribution of all continuous variables and found that all except LOS were normally distributed; however, we presented all as mean (SD) for consistency and ease of interpretation, supported by the central limit theorem given our sample size (n = 120). Clinical relevance and existing literature drove the choice of 30 days as a cut-off for defining prolonged LOS [28, 29], a threshold commonly used in internal medicine to differentiate typical from extended hospitalisations. Comparisons of sex (male vs. female) and LOS (<30 vs. ≥30 days) were performed using the t-test. Muscle thickness was classified as high and low based on the median value of 20 mm. Charlson Comorbidity Index was classified as high and low based on the mean value (mean CCI = 6, low CCI ≤ 6, high CCI > 6). Correlations between age, MT, EI, LOS and CCI were assessed using Pearson’s correlation coefficient. Univariate and multivariate linear and logistic regression analyses were performed to evaluate the associations between variables. A P-value of <.05 was considered statistically significant.

Results

Table 1 shows the basic characteristics of the study participants. A total of 120 participants were included in the data analysis. The study included 120 patients: 40 under cardiology, 25 under a geriatrician, 14 managed by a family medicine practitioner or hospitalist and 38 under a general internal medicine physician. The most common admitting diagnoses were cardiovascular diseases (n = 58), followed by cerebrovascular diseases and neurological conditions (n = 19), and falls and fractures (n = 18).

The mean (standard deviation (SD)) age was 76 (7) years. The mean (SD) LOS was 27 (31). The mean quadriceps MT was 20 (5) mm, and EI was 68 (26) a.u. We found statistically significant differences between females (n = 64) and males (n = 56) in mean age (females 78 (7) vs. males 74 (6), P < .001), MT (females 19 (5) vs. males 21 (5), P = .032) and EI (females 76 (29) vs. males 58 (18), P < .001) measures. Muscle thickness was higher in patients with shorter hospital stays (mean [SD] = 21 [6] mm) than in those with prolonged LOS (mean [SD] = 17 [4] mm). The mean (SD) CCI was 6 (2).

Table 2 shows the bivariate Pearson correlation between age, MT, EI, LOS and CCI. All variables were correlated significantly with age. Muscle thickness had a statistically significant negative weak correlation (R = −0.35, P < .001) with age, while EI (R = 0.25, P < .01) and LOS (R = 0.18, P < .01) had a positive weak correlation with age. Charlson Comorbidity Index (R = −0.19, P < .01) correlated negatively with MT, but the correlation was less than LOS’s (R = −0.29, P < .01).

Table 3 shows the results of univariate and multivariate linear regression analysis.

In both univariate and multivariate linear regression analysis, one unit increase in MT was associated with approximately 1.5 fewer days of hospital LOS, and the results were statistically significant. In Model 4, one CCI score increase was associated with 2.7 more days of hospital LOS.

Table 4 shows the association between low MT, high EI, high CCI and prolonged LOS in hospitalised older adults. Low MT significantly increased the odds of staying in the hospital longer than 30 days by more than threefold in all models. However, high EI was associated with decreased odds of prolonged LOS, and the results of the different models were not statistically significant. Higher CCI scores increased the odds of prolonged LOS by approximately twofold in all models. We adjusted for EI in the low MT model to assess whether the association with prolonged LOS remained significant after accounting for muscle quality, given EI’s correlation with muscle strength and quality.

Discussion

Interpretation of results

Our study demonstrates that quadriceps MT, measured by POCUS, is a significant predictor of hospital LOS in older adults. Each mm increase in MT was associated with approximately 1.5 fewer days of hospital stay, underscoring the importance of muscle health in clinical recovery. The association between MT and LOS remained significant even after adjusting for the CCI.

Notably, although high EI was associated with reduced odds of prolonged LOS, this finding lacked statistical significance. This suggests that muscle quality, as measured by EI, may have a complex relationship with LOS that requires further investigation. Our study used non-corrected EI measures, while others [30, 31] used equations by Young et al. to correct EI based on subcutaneous fat thickness (SFT). Unfortunately, the SFT measure was not available in our study. Corrected EI predicted hospital-associated complications [30], and muscle function significantly compared to non-corrected EI [31].

Our study used the median value (20 mm) for a cut-off point for classification. In younger renal transplant patients (mean age 50), where the cut-off point was 20 mm (20th percentile), the low MT group still had longer LOS and a higher proportion of LOS over 14 days [24]. Low MT increases the risk of extended hospital stay, and prolonged hospitalisation, especially in the ICU, increases the risk of muscle wasting, as shown by decreased quadriceps MT over time [32].

Numerous studies have shown a significant correlation between sarcopenia and prolonged LOS [9, 22, 33, 34]. Low muscle mass is a key component in the diagnosis of sarcopenia, and low MT is often used as an indicator of low muscle mass [35].

Potential benefits of shorter hospital stay and cost-effectiveness of point-of-care ultrasound

Shorter hospital stays benefit hospitals by increasing hospital income through a higher hospital turnover rate [37], and they can save money through LOS reduction programs [38]. For patients, more extended hospitalisation was associated with increased risk and cost associated with infection, adverse drug reactions, and more antibiotic use, elevating the risk of antibiotic resistance [37, 39].

When POCUS was used in the diagnostic processes, it reduced the LOS (by 0.31 days, P = .14), reduced the total hospitalisation cost significantly (P < .001) by $4329, per day hospitalisation cost by $751 (P < .001), and total and per day radiologist cost significantly ($123 and $35 respectively, both P < .001) [40]. The use of portable bedside ultrasound resulted in the reduction of hospital LOS by 860 days over 3-year period and the cost reduction of €170 per day in a retrospective economic analysis study, and noted that the implementation of bedside ultrasound was cost effective [41]. Studies evaluating cardiac conditions using handheld ultrasound compared to standard echocardiographic device resulted in reductions in total cost around 30%, reduced workload and faster diagnosis [42, 43].

Compared to imaging modalities like CT scans, POCUS offers greater convenience and accessibility due to its portability, cost-effectiveness, objectivity, absence of radiation exposure and minimal training requirements [22, 36].

Mechanistic considerations

Several mechanisms may explain the relationship between MT and LOS, some supported by current literature and others hypothesised based on broader knowledge in the field. First, decreased muscle mass is associated with reduced functional capacity, which may lead to prolonged hospitalisation and increased risk of adverse outcomes. The following possible mechanism would be that the low muscle mass may indicate poor dietary status, impacting recovery and LOS. Teixeira et al. showed that low muscle mass was a predictor of malnutrition and prolonged hospital stay in patients with acute exacerbation of COPD [9]. This suggests that muscle mass might serve as a marker of overall nutritional reserve, potentially influencing recovery from acute illness. Lastly, reduced muscle mass is a key component of frailty, associated with poor outcomes in hospitalised older adults. Canales et al. found that preoperative quadriceps MT, measured via POCUS, was a significant predictor of frailty and adverse postoperative outcomes, including prolonged LOS [23].

Charlson Comorbidity Index

Multimorbidity, or the coexistence of chronic conditions, is prevalent in older adults [47]. Among other factors to predict LOS, the most commonly used indicator is the number of comorbid conditions. Several studies have used the CCI to assess LOS, which was positively correlated with LOS in older adult populations [48–52]. Higher CCI scores are associated with increased LOS due to the complexity and severity of comorbid conditions [48, 49]. We included CCI in our analysis as a covariate since it has shown strong associations with LOS in the previous literature. In the cases where we are unable to identify comorbid conditions when the patient is unresponsive, demented, or past medical history cannot be found, POCUS could also be used instead of CCI.

We found that having low MT was associated with three times more odds of prolonged LOS, while having high CCI was associated with twice the odds of staying in the hospital for more than 30 days. Although further research is warranted, these findings suggest that MT could serve as a rapid and valuable admission metric for identifying patients at risk of prolonged LOS.

Study limitations

It is a single-centre study, so the study population may differ from that of the general population, limiting the generalisability of the findings. Also, the institution-specific factors may influence the findings. The cross-sectional design limits our ability to establish causality between MT and LOS. While we adjusted for basic characteristics, such as age, biological sex and CCI, we did not account for other potential confounders, such as nutritional status or physical activity level. Since this was a proof-of-concept study in an acute medical setting, not a laboratory setting, many standard measures, such as height and weight, are unavailable due to patient illness, inability to stand and dehydration. Researchers use different cut-off points for quadriceps MT since there is no established standardisation and cut-off points for MT and EI. This could explain the inconsistency in the literature. Further research is needed to evaluate whether quadriceps muscle can be a valid measure for low muscle mass.

Conclusion

In conclusion, our study underscores the importance of quadriceps MT, assessed via POCUS, as a predictor of hospital LOS in older adults. Each mm increase in MT was associated with approximately 1.5 fewer days of hospital stay. Low MT significantly increased the odds of staying in the hospital longer than 30 days by more than threefold in all models.

Declaration of Conflicts of Interest:

None.

Declaration of Sources of Funding:

None.

References