Effect of an Exercise Intervention on Functional Decline in Very Old Patients During Acute Hospitalizations

Abstract

This randomized clinical trial examines the effectiveness and generalizability of implementing a multicomponent exercise program for treating functional decline associated with acute hospitalization in very old patients.

Older adults are particularly prone to hospitalization hazards, including high risk of functional and cognitive deterioration. One year after discharge, more than half of hospitalized older patients do not recover their preadmission functional levels, associated with increasing risk of nursing home admission and death.¹ We have previously demonstrated that an individualized, multicomponent 5-day exercise program prevented the functional decline associated with acute hospitalization in very old patients.² In this study, therefore, we sought to examine the effectiveness and generalizability in other hospitals across a shorter period of intervention.

Methods

We performed a multicenter randomized clinical trial (RCT) within departments of geriatric medicine in 3 tertiary hospitals in Spain (NCT04600453), approved by the local ethics committee (Pyto2018/7). The trial protocol is available in Supplement 1. Community-dwelling patients acutely hospitalized for medical care (aged ≥75 years, assessed as able to perform specific exercises by physicians) were randomly assigned to the intervention or control (usual care) group within 48 hours of admission (eFigure in Supplment 2). The intervention group received 2 daily 20-minute sessions of exercise over 3 consecutive days. Exercises were adapted from the multicomponent exercise program Vivifrail.³ The morning sessions included individualized supervised progressive resistance, balance, and walking training exercises. The resistance exercises were tailored to the individual’s functional capacity using variable resistance training machines aiming at 2 to 3 sets of 8 to 10 repetitions with a load equivalent to 40% to 60% of the 1-repetition maximum at fast intentional velocity. Balance and gait retraining exercises gradually progressed in difficulty. The evening session consisted of functional unsupervised exercises using light loads, such as knee extension and flexion, and hip abduction using ankle cuffs, and daily walking in the corridor. The control group received routine physiotherapy when needed, promoting mobility. Patients were discharged after clinical stabilization of the cause of the hospital admission and advised to continue performing exercise according to the Vivifrail³ program.

Results

In the first 48 hours after clinical stabilization, 200 patients were randomly assigned after baseline assessment with concealed allocation via a random number sequence (103 control and 97 intervention; 91 [45.5%] women; mean [SD] age, 86.5 [4.4] years; Table 1).

Table 1. Main Demographic, Clinical, Functional, and End Point Data at Baseline by Group.

Variable	Mean (SD)		P value
	Control (n = 103)	Intervention (n = 97)
Demographic and clinical data
Age, y	86.4 (4.41)	85.7 (4.32)	.25
Women, No. (%)	48 (47.5)	41 (42.3)	.27
BMIa	27.1 (5.2)	26.1 (6.2)	.45
Clinical data
No. of diseases	11.4 (4.26)	10.6 (4.37)	.18
CIRS, median (IQR), scoreb	12.6 (5.35)	12.4 (6.24)	.78
MNA-SF, median (IQR), scorec	10.1 (2.46)	10.2 (2.12)	.83
6-m Gait velocity test, m/s	0.39 (0.52)	0.45(0.83)	.16
Primary end-point measures
SPPB scaled	4.81 (2.79)	4.91 (3.11)	.81
Barthel indexe	88.5 (11.4)	87.9 (12.6)	.72
Secondary end-point measures
Mini-Mental State Examinationf	23 (5)	22 (5)	.42
Yesavage Geriatric Depression scaleg	4 (3)	3 (3)	.36
1RM leg press, kg	57.9 (37.6)	55.8 (36.9)	.73
Handgrip, kg	16.3 (6.5)	17.6 (7.4)	.21
Pain (visual analogue scale)h	1.68 (2.60)	1.76 (2.58)	.83

Abbreviations: BMI, body mass index; CIRS, Cumulative Illness Rating Scale; MNA-SF, Mini-Nutritional Assessment (Short Form); 1RM, 1 repetition maximum; SPPB, Short Physical Performance Battery.

^a Calculated as weight in kilograms divided by height in meters squared.

^b CIRS scale evaluates individual body systems, ranging from 0 (best) to 56 (worst).

^c The Mini-Nutritional Assessment (Short-Form) ranges from normal nutritional status (12-14 points), risk of malnutrition (8-11 points), or malnourished (0-7 points) (<17 points).

^d The SPPB scale ranges from 0 (worst) to 12 (best).

^e The Barthel Index ranges from 0 (severe functional dependence) to 100 (functional independence).

^f The Mini-Mental State Examination ranges from 0 (worst) to 30 (best).

^g The Yesavage Geriatric Depression Scale ranges from 0 (best) to 15 (worst).

^h Visual analog scale ranges from 0 (best) to 10 (worst).

The exercise intervention program provided a significant benefit over usual care in blinded primary outcome, the Short Physical Performance battery (SPPB). At discharge, the exercise group had a mean increase of 0.93 points (95% CI, 0.39-1.48 points) on the SPPB scale over the usual care group (Table 2). There was a worsening over hospitalization in the Barthel Index over time of −11.0 points (95% CI, −14.6 to −7.44 points), with a trend for attenuated declines in the intervention group: −6.49 points; 95% CI, −10.20 to −2.73; P = .09. Furthermore, there was a significant improvement in subjective pain perception associated with the exercise intervention on the visual analogic scale (VAS) scale (−1.21 points; 95% CI, −2.09 to −0.33; P = .01), whereas no such trend was found in the control group (Table 2). No adverse effects associated with the prescribed exercises were recorded apart from occasional muscle pains during the first days.

Table 2. Results of Primary and Secondary End Points by Group^a.

Variableb	Group		Between-group difference (95% CI)	P value between groups
	Control (n = 103)	Intervention (n = 97)
Primary end point
Change in functional capacity
SPPB scale	0.46 (0.09 to 0.84)	1.39 (1.00 to 1.79)	0.93 (0.39 to 1.48)	.001
Barthel Index	−11.0 (−14.6 to −7.44)	−6.49 (−10.2 to −2.73)	4.55 (−0.65 to 9.75)	.09
Secondary end points
MMSE	0.73 (0.07 to 1.40)	1.09 (0.39 to 1.80)	0.36 (−0.60 to 1.33)	.47
GDS	−0.25 (−0.66 to 0.17)	−0.78 (−1.22 to −0.33)	−0.53 (−1.14 to 0.08)	.09
Handgrip, kg	0.15 (−0.46 to 0.77)	1.04 (0.41 to 1.68)	0.89 (0.00 to 1.77)	.05
Pain (visual analogue scale)	0.28 (−0.33 to 0.90)	−0.93 (−1.56 to −0.29)	−1.21 (−2.09 to −0.33)	.01
Length of stay in hospital, median (IQR), d	7 (3)	7 (3)	0.21 (−0.72 to 1.56)	.65

Abbreviations: GDS, Yesavage Geriatric Depression Scale; MMSE, Mini Mental State Examination; SPPB, Short Physical Performance Battery.

^a All data were derived from linear mixed-effects model and included all randomized patients. For each group, data are expressed as change from baseline (admission) to discharge, determined by the time coefficients (95% CI) of the model. The between-group difference was determined with time x group interaction coefficient.

^b Explanations of scales used are given in the footnotes to Table 1.

Discussion

An individualized intervention of multicomponent exercise is feasible and attenuates the disability associated with hospitalization, extending findings of our initial RCT in which an acute decline in ADLs was prevented in a similar cohort.² The results of this multicenter RCT highlight factors which may have contributed to the lower effectiveness compared with our previous exercise study.² In the initial RCT,² the improvement in the SPPB was 2.2 (95% CI, −2.09 to 0.33) points, compared with this RCT (SPPB, 0.93 points; 95% CI, 0.39-1.48). Benefits for Barthel Index and cognition were also attenuated/nonsignificant. Factors that could have reduced the efficacy of the intervention include the reduction of the number of training days (from 5 days to 3), and a greater complexity and number of baseline diseases (11 vs 9) per person. Improvement in activities of daily living in older or frail patients requires a more intensive approach than merely promoting mobility,⁴ and the amount of days implementing this exercise is important when optimizing the benefits in hospitalized older adults.^5,6

Supplement 1.

Trial Protocol

Supplement 2.

eFigure. Study Flow Diagram.

Supplement 3.

Data Sharing Statement