Assessment of lower limb muscle strength can predict fall risk in patients with chronic liver disease

Falls are caused by a combination of factors, including loss of lower limb muscle strength (LMS), and associated with declined performance status (PS). Age-related sarcopenia is generally associated with decreased muscle mass and strength of lower limb muscle but without a noticeable loss of those of upper limb or trunk muscle. However, no reports have focused on falls or LMS in chronic liver disease (CLD) patients. This study is the first to analyze the risk factors for falls in patients with CLD, focusing on LMS measurement using the Locomoscan. This study enrolled 315 CLD patients whose LMS was measured. The patients who experienced falls more than 1 year ago or during the observation period were classified as those who experienced falls. We found that risk factors for falls were PS1/2 and decreased LMS (< 0.32 N/kg). The group with sarcopenia had a higher frequency of decreased LMS (54 vs. 26%, p = 0.001) and falls (24 vs. 4.4%, p < 0.001) compared to the non-sarcopenia group. This study found that decreased LMS was an independent risk factor for falls. Assessment of LMS may be used as a better marker associated with the risk of falls in patients with CLD.


Lower limb muscle strength in patients with CLD
The first and second LMS values showed good correlation in the same patients using Locomoscan (r = 0.93, p < 0.001) (Fig. 2a).LMS worsened with increasing age (r = − 0.33, p < 0.001) and ALBI score (r = − 0.29, p < 0.001) and was similar for both male and female patients (Fig. 2b,c).

Risk factors for falls in patients with CLD
We found that 7.0% of CLD patients had experienced falls more than 1 year ago or during the observation period.We calculated the areas under the receiver operating characteristic (ROC) curves of LMS.The cutoff value of LMS for falls was 0.32 N/kg with ROC of 0.81 (0.71-0.91), sensitivity of 0.77, and specificity of 0.77.Patients who experienced falls (n = 22) had a higher frequency of lower LMS than those without falls (n = 293) (82% vs. 26%, p < 0.001) (Table 2).The group with sarcopenia had a higher frequency of falls (24 vs. 4.4%, p < 0.001) and decreased LMS (54 vs. 26%, p = 0.001) compared to the non-sarcopenia group.

Characteristics of the group with decreased lower limb muscle strength
The patients with decreased LMS (n = 94, 30%) were more likely to be over 73 years old, and had with LC, PS1/2, previous episodes of HCC, Child-Pugh grade B, ALBI grade 2b/3, decreased GS, decreased muscle mass in CT/ BIA, sarcopenia, and inability to cross a pedestrian crossing in time, compared to the group without decreased LMS (Table 4).

Risk factors for falls by gender
For the univariate analysis of male patients, the following were significant: age > 73, PS1/2, previous episodes of HCC, BMI < 22, decreased GS, decreased muscle mass in CT/BIA, and decreased LMS.In the multivariate analysis, decreased muscle mass in CT/BIA (OR 13, p = 0.021) and decreased LMS (OR 5.0, p = 0.043) were independent factors (Table 5).www.nature.com/scientificreports/For the univariate analysis of female patients, the following were significant: age > 79, PS1/2, previous episodes of HCC, decreased muscle mass in CT/BIA, inability to cross a pedestrian crossing in time, and decreased LMS.In the multivariate analysis, the inability to cross the pedestrian crossing in time (OR 19, p = 0.043) was an independent factor (Table 5).

Discussion
Falls were found in 7.0% of patients with CLD, and decreased LMS, an independent risk factor for falls, was found in 30%.This study examined the association between LMS and falls in patients with CLD.Traditional parameters for sarcopenia in patients with CLD, such as muscle mass and GS measurements alone, may not adequately identify high-risk groups for falls.Moreover, patients with low LMS should be followed up with particular attention to the risk of falls.
Sarcopenia is characterized by low skeletal muscle mass, skeletal muscle weakness, and decreased physical performance 10 .It causes a decline in functional status, impaired mobility, a higher risk of falls, and an increased mortality risk 25 .The prevalence of sarcopenia is reported as 10% in the general elderly population and 39% in cancer patients [26][27][28] .Skeletal muscle mass decreases by 1% per year, so by age 50, there is a 5-10% decrease www.nature.com/scientificreports/compared to age 20, and after age 50, there is a 30-40% decrease 29 .Furthermore, muscle strength declines by 2-5% per year, and the rate of decline increase after the age 50 30,31 .Sarcopenia, which increases with age, was associated with falls, fractures, functional decline, and all-cause mortality in elderly people 32 .In contrast, sarcopenia is also considered as a poor prognostic factor in patients with CLD, even if they are not elderly, and is associated with hospitalization events in patients with primary HCC, including fall fractures 12,13,18 .Therefore, sarcopenia has received increasing attention in recent years among patients with CLD.Falls are associated with declined PS, poor quality of life, and death in the elderly and patients with cirrhosis.In particular, the incidence of falls was 40% per year in those with minimal hepatic encephalopathy compared with 13% in those without 33,34 .In addition, fall trauma occurs in 54-74% and fractures in 6-12% of cases [35][36][37] .Fall accidents have serious mortality consequences.In this study, falls occurred in 7.0% of patients with CLD, despite the fact that many of these patients were young and had good liver function.Preventing falls, which necessitate nursing care and becoming bedridden, may be vital for enhancing healthy life expectancy in patients with CLD.
Few reports have assessed LMS, and the gold standard diagnostic method for decreased LMS and its cutoff values are still unknown.However, falls are caused by a combination of declined LMS, low muscle mass, and imbalance 36,38 .Hence, the decrease in LMS must be accurately assessed.Large multifunctional muscle strength measuring machines used for LMS measurement require a labor-intensive operation, and in clinical settings, measurements are rarely performed, even if the machines are available in the facility.However, due to the complexity and risk of falling during measurement, LMS measurement using simple devices such as handheld dynamometers has not been widely used 39 .In addition, age-related sarcopenia is generally associated with a loss of lower limb muscle mass and LMS but without a noticeable loss of upper limb or trunk muscle mass.The Locomoscan used in this study is an important noninvasive device that solves several LMS measurement problems.Omori et al. reported a high correlation coefficient of 0.82 between the quadriceps trainer, a prototype of the Locomoscan, and knee extension force measurements using the large multifunctional muscle strength measuring machines in healthy populations 21 .Therefore, this study focused on LMS measurement using the Locomoscan to indicate groups at high risk for falls.
This study found that LMS did not correlate strongly with CT PMI, BIA SMI, or regional muscle mass using BIA, unlike GS.Decreased LMS may capture earlier muscle weakness better than the other parameters.The LMS measurement requires more integrated capability, such as instantaneous force and balance, compared to GS, and no direct relationship may have been found.The results also suggest that LMS may not correlate as well as lower limb muscle mass due to overestimation of BIA caused by slight edema in patients with CLD.Moreover, the fact that local muscle mass and quality is hardly related to lower GS/LMS in female patients suggests that small number of cases in this study, the different areas prone to muscle changes depending on gender, and the possibility of other causes of lower GS/LMS; such as instantaneous force and balance are related.Thus, we believe that assessments related to muscle strength, such as LMS, in addition to the assessment of muscle quantity and quality using imaging studies are important as clinically useful markers.
In this study, the risk factors for falls in patients with CLD were PS1/2 and decreased LMS.Moreover, LMS was decreased in 64% of PS1/2 patients.Of the 270 PS0 patients, 65 (24%) had decreased LMS.This indicates that imaging alone cannot accurately predict patients at high risk for falls and that careful muscle strength measurements are important.It is particularly noteworthy that older age was not an independent factor.We believe that muscle assessment in all patients with CLD, regardless of age, is important for improving prognosis.LMS measurement may be useful to provide therapeutic intervention for fall risk before progression to PS1/2.This study had some limitations.First, this study evaluated a small number of cases over a short time.Largescale long-term studies are necessary.Second, we did not assess balance deficits.Reported fall risk factors include muscle weakness and balance deficits 40 .Balance is important for maintaining postural equilibrium and, thus,

Figure 1 .Figure 2 .
Figure 1.Correlation between grip strength (GS) and the other parameters in patients with chronic liver disease.(a) GS and psoas muscle mass index (PMI) using computed tomography (CT), (b) GS and skeletal muscle mass index (SMI) using bioelectrical impedance analysis (BIA), (c) GS and CT value of the multifidus muscle, (d) GS and upper limb muscle mass using BIA in all the patients.(e-h) In male patients.(i-l) in female patients.

Table 2 .
Comparison between the patients who experienced falls and those without falls.

Table 3 .
Risk factors for falls in patients with chronic liver disease.

Table 4 .
Comparison between the patients with decreased lower limb muscle strength and those without.

Table 5 .
Risk factors for falls by gender.