Circulation Journal
Online ISSN : 1347-4820
Print ISSN : 1346-9843
ISSN-L : 1346-9843
Short-Term Change in Gait Speed and Clinical Outcomes in Older Patients With Acute Heart Failure
Shinya TanakaKentaro KamiyaNobuaki HamazakiRyota MatsuzawaKohei NozakiTakeshi NakamuraMasashi YamashitaEmi MaekawaChiharu NodaMinako Yamaoka-TojoAtsuhiko MatsunagaTakashi MasudaJunya Ako
Author information
JOURNALS FREE ACCESS FULL-TEXT HTML Advance online publication

Article ID: CJ-19-0136

Details
Abstract

Background: Evidence for the prognostic value of gait speed is largely based on a single measure at baseline, so we investigated the prognostic significance of change in gait speed in hospitalized older acute heart failure (AHF) patients.

Methods and Results: This retrospective study was performed in a cohort of 388 AHF patients ≥60 years old (mean age: 74.8±7.8 years, 228 men). Routine geriatric assessment included gait speed measurement at baseline and at discharge. The primary outcome of this study was all-cause death. Gait speed increased from 0.74±0.25 m/s to 0.98±0.27 m/s after 13.5±11.0 days. Older age, shorter height and lower hemoglobin level at admission, prior HF admission, and higher baseline gait speed were independently associated with lesser improvement in gait speed. A total of 80 patients died and 137 patients were readmitted for HF over a mean follow-up period of 2.1±1.9 years. In multivariate analyses, change in gait speed showed inverse associations with all-cause death (hazard ratio [HR] per 0.1 m/s increase: 0.83; 95% confidence interval [CI]: 0.73 to 0.95; P=0.006) and with risk of readmission for HF (HR: 0.91; 95% CI: 0.83 to 0.99; P=0.036).

Conclusions: Short-term improvement in gait speed during hospitalization was associated with reduced risks of death and readmission for HF in older patients with AHF.

Heart failure (HF) is a major cause of hospitalization in older adults, and is associated with poor prognosis, reduced quality of life, and increased medical costs.1 Despite treatment according to clinical guidelines, older patients hospitalized for acute HF (AHF) have high morbidity rates, including frequent rehospitalization and death.2,3 Therefore, clinical management and treatment decision-making require accurate prognostic stratification.

Because of their reduced physiological reserves and increased vulnerability, the prognosis of older patients with AHF is dependent not only on HF status and comorbidities but also on condition such as physical function.4 Measurement of gait speed is a simple method of evaluating functional status and obtaining information regarding prognosis in older patients.57 However, evidence for the prognostic value of gait speed is largely based on a single measurement at baseline. Improved physical performance has been shown to be associated with lower rehospitalization and mortality rates in outpatients with cardiovascular disease (CVD), including HF.810 Although hospitalized older patients with AHF show higher rates of frailty and physical functional impairment,11 many older AHF patients experience episodes of improved physical function after recovery from an illness or with medical interventions. HF status is evaluated by combining physical examination results, signs of congestion, body weight change, and net fluid balance to allow clinical decision-making. However, current AHF management strategies rarely address a change in gait speed during hospitalization, and little information is available regarding short-term changes in gait speed and the clinical implications. This study was performed to investigate whether a short-term change in gait speed during hospitalization was associated with clinical outcomes in older AHF patients.

Methods

Study Population

The study population was identified in the Kitasato University Cardiac Rehabilitation Database and consisted of 816 patients ≥60 years old who were admitted for AHF, defined as volume overload and dyspnea at rest or with minimal activity, and evaluated with regard to usual gait speed in routine geriatric assessment between August 2008 and July 2016. Patients who died before assessment of gait speed (n=29), were unable to walk because of unstable medical condition or severe disability (n=110), could not complete admission and/or discharge gait speed assessments (n=248), or had missing outcome data (n=41) were excluded from the analysis. After applying these exclusion criteria, 388 patients were included in the present study. To prevent functional deterioration, all patients participated in the cardiac rehabilitation program for 30–60 min/day, 5 days per week during their hospitalization under the supervision of a physical therapist and nurse using a standardized protocol designed according to the Japanese Circulation Society guidelines for treatment of AHF.12 The supervised rehabilitation program had 2 exercise stages. The first stage consisted of gradual mobilization, including basic activity training, such as sitting up in bed, sit-to-stand motion, and walking within the hospital ward. Once they were clinically stable, the patients proceeded to the second stage, consisting of a gym-based exercise training program: 5 min of stretching, balance training, and resistance training using the patient’s own weight, and 20–40 min of aerobic training using a recumbent ergometer, upright ergometer, or treadmill walking, including warm-up and cool-down periods. For both types of training, the exercise intensity was prescribed at a Borg rating of perceived exertion (RPE) of 11–13 on the Borg RPE scale of 6–20. The study was performed in accordance with the Declaration of Helsinki, and was approved by the Ethics Committee of Kitasato University Hospital.

Data Collection

Data on all variables were collected from electronic medical records. Clinical details on presentation, as well as demographic, biochemical, and echocardiographic data, and information on adverse events after hospital discharge were recorded. The body mass index (BMI) was calculated as body weight (kg) divided by height (m) squared. Simpson’s method was used to estimate the left ventricular ejection fraction (LVEF) from 2D echocardiograms. The estimated glomerular filtration rate (eGFR) was defined according to the formula of the Japanese Society of Nephrology.13 B-type natriuretic peptide (BNP) concentration was determined using a commercial immunoradiometric assay (Shionogi, Osaka, Japan).

Gait Speed Measurement

Usual gait speed was evaluated at hospital admission and discharge. Baseline gait speed assessment was performed once symptoms at rest had subsided and fluid retention had resolved, and a second test was performed at discharge. Both tests were performed only once for all patients. The change in gait speed during the hospitalization period was determined as the absolute difference between test results. To measure usual gait speed, the patients were asked to walk at their usual speed, and were timed over a 10-m walkway. Patients were allowed to use walking aids, such as a cane, during the test. Gait speed in older adults has been shown to have high test–retest reliability with intraclass correlation coefficient >0.9.14

Outcomes

The primary endpoint of this study was all-cause death, and the secondary endpoint was the rate of readmission for HF. The time to the endpoint was calculated as the number of days from the date of gait speed measurement at discharge to the date of the event.

Statistical Analysis

Normally distributed continuous variables are expressed as the means±standard deviation (SD), and non-normally distributed variables are presented as the median and interquartile range (IQR). Categorical variables are expressed as number and percentage. The cohort was divided into 2 groups based on the median baseline gait speed and change in gait speed. Differences between groups were evaluated using the unpaired Student’s t test or Mann-Whitney U-test for continuous variables, and the chi-squared or Fisher’s exact test for categorical variables, as appropriate.

Multivariate linear regression analyses using the backward elimination method were performed to estimate that factors that independently correlated with baseline gait speed and change in gait speed. Baseline variables with P<0.1 in the univariate analyses were entered into the multivariate model. The relationships of a change in gait speed with body weight change and laboratory markers were evaluated using the Spearman rank correlation.

The Get With the Guidelines-Heart Failure (GWTG-HF) risk score (based on race, age, systolic blood pressure, heart rate, blood urea nitrogen, sodium levels, and the presence of chronic obstructive pulmonary disease) and the AHEAD score (based on atrial fibrillation, hemoglobin, age, creatinine, and the presence of diabetes) were calculated for each patient as described previously.15,16 The discrimination and calibration of these risk scores have been well validated in Asian patients with AHF.17,18 Therefore, the GWTG-HF risk score and AHEAD score were used as adjustment variables in the multivariable prognostic model. Survival was evaluated using the Kaplan-Meier method and compared using the log-rank test. The relationship of a change in gait speed with all-cause death and readmission for HF were evaluated by univariate and multivariate Cox regression analyses. Multivariate analyses were adjusted for baseline gait speed, GWTG-HF risk score, and AHEAD score. The associations of a change in gait speed with death and HF readmission risks were examined using a Cox regression model with spline functions with 4 knots at quartiles of the independent variable.

Sensitivity analysis was performed to address reverse causality after excluding patients who died within the first 6 months of follow-up. To examine the potential effect modification on the association of a change in gait speed with all-cause death, subgroup analyses of change in gait speed were performed in subgroups stratified at the median baseline gait speed. Hazard ratios (HRs) are reported with corresponding 95% confidence intervals (95% CI).

Statistical analyses were performed using SPSS version 23.0 (IBM Corporation, Armonk, NY, USA) and R version 3.2.1 (R Foundation for Statistical Computing, Vienna, Austria). In all analyses, a two-tailed P<0.05 was taken to indicate statistical significance.

Results

Study Population

Table 1 shows the baseline characteristics of the total patient population as well as groups stratified according to baseline gait speed and change in gait speed. The mean age of the study population was 74.8±7.8 years, 58.8% were male, and 47.7% and 36.6% had reduced and preserved ejection fraction, respectively. Baseline measurements of gait speed were obtained at a mean of 3.5±1.8 days (median, 3 days) after hospitalization. Gait speed increased from 0.74±0.25 m/s (median, 0.75 m/s) to 0.98±0.27 m/s (median, 1.02 m/s) after a mean interval of 13.5±11.0 days (median, 10 days) between the 2 measurements (Supplementary Figure 1). The mean improvement in gait speed from admission to discharge was 0.24±0.23 m/s (median, 0.22 m/s). There was no increase in gait speed at follow-up in 38 patients (9.8%), but an increase of more than 0.1 m/s was observed in 279 patients (71.9%). Analyses were performed in subgroups divided according to baseline gait speed, change in gait speed, and baseline characteristics (Supplementary Figure 2). Both slower baseline gait speed and lesser change in gait speed were associated with older age.

Table 1. Baseline Characteristics
  Overall
(n=388)
Baseline gait speed P value Change in gait speed P value
Slow
(n=194)
Fast
(n=194)
Lesser
improvement
(n=194)
Greater
improvement
(n=194)
Age, years 74.8±7.8 76.5±8.4 73.1±6.7 <0.001 76.4±7.8 73.1±7.4 <0.001
Age group, %       <0.001     0.001
 60–69 111 (28.6) 49 (25.3) 62 (32.0)   39 (20.1) 72 (37.1)  
 70–79 168 (43.3) 70 (36.1) 98 (50.5)   91 (46.9) 77 (39.7)  
 ≥80 109 (28.1) 75 (38.7) 34 (17.5)   64 (33.0) 45 (23.2)  
Male, % 228 (58.8) 87 (44.8) 141 (72.7) <0.001 111 (57.2) 117 (60.3) 0.606
Height, cm 158.2±9.4 156.1±10.0 160.3±8.2 <0.001 157.3±9.4 159.1±9.3 0.050
Body weight at admission, kg 58.5±12.4 57.9±14.0 59.2±10.6 0.295 57.4±11.9 59.7±12.8 0.066
BMI, kg/m2 23.3±4.1 23.6±4.7 23.0±3.4 0.156 23.1±4.1 23.4±4.1 0.439
LVEF, % 42.5±16.8 41.6±17.4 43.3±16.1 0.327 42.9±16.5 42.0±17.1 0.589
LVEF group, %       0.084     0.907
 <40 185 (47.7) 100 (51.5) 85 (43.8)   94 (48.5) 91 (46.9)  
 40–49 61 (15.7) 23 (11.9) 38 (19.6)   29 (14.9) 32 (16.5)  
 ≥50 142 (36.6) 71 (36.6) 71 (36.6)   71 (36.6) 71 (36.6)  
NYHA IV at admission, % 235 (60.6) 124 (63.9) 111 (57.2) 0.212 105 (54.1) 130 (67.0) 0.013
SBP, mmHg 136±40 138±42 133±37 0.235 133±38 139±41 0.166
DBP, mmHg 67±15 68±17 66±13 0.147 66±15 68±15 0.298
Heart rate, beats/min 82±21 85±23 80±17 0.020 80±19 85±22 0.023
Atrial fibrillation, % 132 (34.0) 73 (37.6) 59 (30.4) 0.163 74 (38.1) 58 (29.9) 0.108
Ischemic etiology, % 171 (44.1) 73 (37.6) 98 (50.5) 0.014 90 (46.4) 81 (41.8) 0.413
Comorbidities, %
 Hypertension 266 (68.6) 136 (70.1) 130 (67.0) 0.585 127 (65.5) 139 (71.6) 0.229
 Diabetes 175 (45.1) 87 (44.8) 88 (45.4) 1.000 83 (42.8) 92 (47.4) 0.414
 Anemia 220 (56.7) 114 (58.8) 106 (54.6) 0.473 122 (62.9) 98 (50.5) 0.018
 COPD 17 (4.4) 9 (4.6) 8 (4.1) 1.000 7 (3.6) 10 (5.2) 0.621
 Hyperuricemia 122 (31.4) 63 (32.5) 59 (30.4) 0.743 67 (34.5) 55 (28.4) 0.229
 Renal dysfunction 307 (79.1) 157 (80.9) 150 (77.3) 0.454 158 (81.4) 149 (76.8) 0.318
 Prior myocardial infarction 89 (22.9) 37 (19.1) 52 (26.8) 0.091 48 (24.7) 41 (21.1) 0.469
 Prior HF admission 151 (38.9) 77 (39.7) 74 (38.1) 0.835 92 (47.4) 59 (30.4) 0.001
Current smoker, % 45 (11.6) 18 (9.3) 27 (13.9) 0.204 17 (8.8) 28 (14.4) 0.112
Laboratory data at admission
 Hemoglobin, g/dL 12.0±2.3 11.9±2.3 12.2±2.4 0.174 11.7±2.3 12.3±2.3 0.005
 Hematocrit, % 36.6±6.7 36.2±6.7 37.0±6.7 0.269 35.7±6.4 37.6±6.8 0.006
 Albumin, g/dL 3.7±0.4 3.6±0.5 3.7±0.4 0.029 3.7±0.4 3.6±0.5 0.406
 Creatinine, mg/dL 1.1 [0.9, 1.6] 1.1 [0.9, 1.6] 1.1 [0.9, 1.6] 0.875 1.2 [0.9, 1.7] 1.1 [0.9, 1.5] 0.188
 BUN, mg/dL 24 [18, 33] 24 [18, 32] 23 [18, 34] 0.733 24 [18, 35] 23 [17, 31] 0.071
 eGFR, mL/min/1.73 m2 44.8±19.8 43.5±20.4 46.2±19.2 0.173 43.4±20.4 46.2±19.2 0.159
 Sodium, mEq/L 139±5 139±5 140±4 0.079 139±5 139±5 0.867
 Uric acid, mg/dL 6.5±1.8 6.5±2.0 6.4±1.7 0.711 6.5±1.8 6.4±1.8 0.460
 CRP, mg/dL 0.38
[0.15, 1.05]
0.35
[0.16, 0.88]
0.40
[0.15, 1.11]
0.370 0.40
[0.15, 0.89]
0.36
[0.16, 1.16]
0.674
 BNP, pg/mL 761
[413, 1,262]
780
[423, 1,428]
752
[410, 1,140]
0.201 819
[390, 1,264]
738
[444, 1,243]
0.755
GWTG-HF risk score 41±8 41±9 40±8 0.209 41±8 40±9 0.035
AHEAD score 2 [2, 3] 3 [2, 4] 2 [1, 3] 0.058 3 [2, 4] 2 [1, 3] 0.002
Baseline gait speed, m/s 0.74±0.25 0.53±0.13 0.95±0.14 <0.001 0.80±0.27 0.68±0.22 <0.001
Gait speed at discharge, m/s 0.98±0.27 0.84±0.28 1.13±0.18 <0.001 0.87±0.26 1.10±0.23 <0.001
Change in gait speed, m/s 0.24±0.23 0.31±0.25 0.18±0.18 <0.001 0.07±0.11 0.42±0.18 <0.001

Values are expressed as means±SD, n (%), or median (interquartile range). BMI, body mass index; BNP, B-type natriuretic peptide; BUN, blood urea nitrogen; COPD, chronic obstructive pulmonary disease; CRP, C-reactive protein; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; GWTG-HF, Get With the Guidelines-Heart Failure; LVEF, left ventricular ejection fraction; HF, heart failure; NYHA, New York Heart Association; SBP, systolic blood pressure.

The group showing greater improvement were younger, had lower gait speed at admission, lower percentages of anemia and prior hospitalization for HF, and higher heart rate, hemoglobin, hematocrit, and proportion of New York Heart Association (NYHA) functional class IV at admission compared with the group showing lesser improvement (Table 1). The group with greater improvement had slightly lower risk than the group with lesser improvement as indicated by the GWTG-HF risk score and AHEAD score. Supplementary Table 1 shows the baseline characteristics of the patients included in the study (n=388) compared with those excluded from the study (n=428). Patients included in the analysis were younger and had higher body weight, BMI, hemoglobin, hematocrit, albumin, eGFR, and sodium, and lower creatinine, blood urea nitrogen, C-reactive protein, GWTG-HF risk score, and AHEAD score, as well as lower prevalence of diabetes, anemia, and prior hospitalization for HF.

Factors Correlated With Gait Speed

Supplementary Table 2 and Table 2 show the results of univariate and multivariate linear regression analyses for predicting baseline gait speed and change in gait speed. The results indicated that older age, female sex, higher body weight on admission, NYHA functional class IV on admission, higher heart rate, and lower albumin and sodium levels were independently associated with slow baseline gait speed. In addition, older age, shorter height, lower hemoglobin level, prior hospitalization for HF, and higher gait speed on admission were independently associated with lesser improvement in gait speed during hospitalization.

Table 2. Multivariable Linear Modeling of Factors Independently Correlated With Gait Speed
Variables B coefficient* Standardized β* t value* P value
Baseline gait speed
 Age −0.008 −0.256 −5.218 <0.001
 Male 0.185 0.364 7.161 <0.001
 Body weight at admission −0.003 −0.150 −2.828 0.005
 NYHA IV at admission −0.059 −0.114 −2.496 0.013
 Heart rate at admission −0.001 −0.114 −2.465 0.014
 Albumin at admission 0.096 0.168 3.675 <0.001
 Sodium at admission 0.008 0.154 3.363 0.001
Change in gait speed
 Age −0.008 −0.287 −5.990 <0.001
 Height 0.004 0.170 3.596 <0.001
 Prior HF admission −0.055 −0.119 −2.775 0.006
 Hemoglobin at admission 0.011 0.115 2.541 0.011
 Baseline gait speed −0.438 −0.481 −10.706 <0.001

*A factor with a negative coefficient was associated with a faster gait speed and greater improvement in gait speed. Abbreviations as in Table 1.

Clinical Course

As shown in Supplementary Table 3, the mean change in body weight for the whole study population was −3.5±3.4 kg between admission and discharge, the mean 6-min walking distance (6MWD) at discharge was 356±122 m, and the median hospital stay was 17 days. At discharge, 87.9% of the patients were prescribed angiotensin-converting enzyme inhibitors or angiotensin-receptor blockers, 76.8% received β-blockers, and 84.8% received diuretics. The correlations of a change in gait speed with changes in body weight and laboratory markers are shown in Supplementary Table 4. Change in gait speed showed significant correlations with change in body weight (r=−0.104, P=0.041), albumin (r=0.106, P=0.037), and eGFR (r=0.101, P=0.046). Changes in C-reactive protein and BNP showed tendencies towards a relationship with a change in gait speed, although these were not statistically significant (r=−0.094, P=0.066; r=−0.095, P=0.075, respectively).

Prognostic Value of Change in Gait Speed

A total of 80 (20.6%) patients died and 137 (35.3%) patients were readmitted because of HF over a mean follow-up period of 2.1±1.9 years. The Kaplan-Meier survival curves indicated significant associations of greater improvement in gait speed with both lower all-cause mortality and HF readmission rates (log-rank: P=0.006 and P=0.033, respectively) (Figure 1).

Figure 1.

(A,B) Kaplan-Meier curves according to the degree of improvement in gait speed (greater vs. lesser than the median gait speed change). HF, heart failure.

Table 3 shows the results of Cox regression analyses for all-cause death and readmission for HF. Change in gait speed showed significant inverse correlations with all-cause death (adjusted HR for each 0.1 m/s increase in gait speed change: 0.83; 95% CI: 0.73–0.95; P=0.006) and with HF readmission risk (adjusted HR: 0.91; 95% CI: 0.83–0.99; P=0.036) even after adjusting for baseline gait speed, GWTG-HF risk score, and AHEAD score. The association of change in gait speed with all-cause death and HF readmission risks are shown in Figure 2. Mortality rate and HF readmission risk decreased with greater improvement in gait speed.

Table 3. Association Between Change in Gait Speed and Clinical Outcomes in Cox Regression Analyses
  Univariate analyses Multivariate analyses
HR 95% CI P value HR 95% CI P value
All-cause death 0.82 0.73–0.92 <0.001 0.83 0.73–0.95 0.006
HF readmission 0.90 0.83–0.97 0.009 0.91 0.83–0.99 0.036
All-cause death or HF readmission 0.88 0.82–0.95 <0.001 0.90 0.83–0.97 0.008

Multivariate analyses: adjusted for GWTG-HF risk score, AHEAD score, gait speed at admission. Hazard ratios (HR) were calculated for increasing 0.1 m/s of gait speed change. CI, confidence interval. Other abbreviations as in Table 1.

Figure 2.

Short-term change in gait speed and clinical outcomes. Short-term improvement in gait speed was shown to be associated with reductions in all-cause death (A) and rate of readmission for heart failure (HF) (B).

Sensitivity analysis excluding patients who died within 6 months of follow-up did not noticeably affect the estimated association between a change in gait speed and all-cause death (adjusted HR: 0.80; 95% CI: 0.68–0.93: P=0.003). The study population was divided into subgroups according to baseline gait speed, and analyses indicated that a change in gait speed was significantly and inversely associated with all-cause death in the slow baseline walking group (adjusted HR: 0.80; 95% CI: 0.65–0.97; P=0.026). The change in gait speed showed a similar tendency in the fast baseline walking group, but the association was not statistically significant (adjusted HR: 0.82; 95% CI: 0.68–1.00; P=0.052).

Discussion

Two-thirds of the older AHF patients included in the present study showed an improvement in their usual gait speed by more than 0.1 m/s during hospitalization. Multiple factors were associated with a lack of improvement in gait speed, including older age, shorter height, history of prior HF admission, and lower hemoglobin level. Short-term improvement in gait speed showed independent correlations with reduced risks of all-cause death and readmission for HF in older AHF patients. The associations between a change in gait speed and all-cause death were similar between patients with slow or fast baseline gait speed. Taken together, these observations suggested that assessment of a short-term change in gait speed is useful for risk stratification of older AHF patients.

Consistent with the results presented here, improvements in measures of physical performance, such as gait speed, Short Physical Performance Battery, and 6MWD, are reported to be associated with reduced rates of rehospitalization and death not only in older subjects but also in overall patients with CVD, including HF.810,19 Although changes in physical performance have largely been examined for prognostic significance based on mid- or long-term (3–6 months or 1–3 years) improvements in outpatients, short-term (mean interval, 15.2±8 days) improvement in 6MWD has been reported as independently associated with improved survival in patients with CHF.20 The 6MWD has been widely adopted in both clinical and research settings for assessment of exercise capacity, prognosis, and response to therapeutic interventions in patients with HF.2023 Gait speed shows a strong positive correlation with 6MWD, and the ability of gait speed to predict prognosis were reported to be comparable to those of 6MWD in an all-CVD cohort and across various diseases, including HF.6 To our knowledge, this is the first report regarding the association between a short-term change in gait speed and clinical outcomes in older AHF patients.

The potential mechanism underlying the association between gait speed change and prognosis in HF patients remains unclear. Gait speed reflects the performance of the cardiorespiratory, nervous, and musculoskeletal systems, and is associated with mobility and exercise capacity in performing activities of daily living. Regular physical activity has been shown to have positive effects on a number of risk factors, including inflammation, autonomic function, vascular endothelial function, arterial stiffness, and myocardial blood flow.24,25 These factors probably each play a role in the mechanism underlying the inverse association between gait speed improvement and reduced adverse outcome risk. Our results revealed a negative correlation of a change in gait speed with a change in body weight during the period of hospitalization. A greater degree of weight loss during hospitalization is associated with higher urine output and dyspnea relief in AHF patients,26 both of which are traditionally included among the primary treatment goals and represent important endpoints in clinical trials. Taken together, these data and those of the present study suggested that short-term improvement in gait speed can be used as an indicator of improved HF status through medical interventions. Short-term improvement in gait speed may also represent a subclinical indicator of physiological reserve, and act as an indicator of resilience or the ability to improve or recover from stressful events.

The results of the present study have implications for both clinical practice and the design of future clinical studies in older AHF patients. Gait speed can be measured quickly, easily, and repeatedly without incurring significant costs and does not require a large amount of space. The current criteria require measurement of gait speed to identify sarcopenia and frailty,27,28 and its measurement has been recommended in clinical practice as a possible vital sign and for clinical trials as a functional outcome in older adults.29,30 The results of the present study support the inclusion of routine gait speed measurement and determination of its changes in clinical practice. The association between a change in gait speed and survival was stronger among patients at higher risk (i.e., those with slow baseline walking speed). In contrast, the change in gait speed showed a similar tendency in the fast baseline walking group, but the association was not statistically significant. In this study, the change in the gait speed of baseline fast walkers was lower than that of baseline slow walkers. The association between a change in gait speed and death may be attenuated by the baseline gait speed. In addition, the small sample size and limited number of clinical events in the baseline fast walker subgroup may have reduced the statistical power to detect such an association, and further studies in larger cohorts are required. The assessment of gait speed and its changes during the period of hospitalization may be useful for patient evaluation and for directing treatment of HF. Interventions, such as cardiac rehabilitation, may also be used to target gait speed for treatment. As gait speed is strongly associated with mobility and inversely associated with risks of adverse events in older patients with HF, clinicians should make efforts to maintain and improve gait speed even in baseline fast walkers. In our cohort, lower hemoglobin levels were associated with a smaller degree of improvement in gait speed, and anemia is common in patients with HF, particularly older patients admitted to hospital.31 These observations suggested that older AHF patients with lower hemoglobin levels may require additional treatment because anemia is associated with advanced symptoms, poor functional status, and increased risks of both HF rehospitalization and death.32

Study Limitations

This study had several limitations, including the small size of the patient population, the single-center nature of the study, and that only Asian patients with AHF were included in the analysis. In addition, the small number of events might increase the risk of type I error. A substantial number of patients could not complete baseline and/or discharge gait speed assessments, and they were excluded from the analysis. The patients excluded from the study were older and had greater severity of HF compared with those included in the analysis, which may have introduced bias. This was also a retrospective study, and therefore the accuracy of some variables was dependent on the accuracy of medical records. Further studies in larger cohorts and in other populations, including patients with more severe HF, are required to validate short-term change in gait speed during the period of hospitalization. Finally, multivariate analysis may mitigate bias after adjustment for pre-existing prognostic factors, which have been shown to stratify early post-discharge and long-term mortality risks.18,33 However, unmeasured and unadjusted factors, such as pre-admission physical activity and other complications, including infection, paralysis, cognitive function, and physical disabilities, and changes in baseline variables, all of which may have an effect on gait speed, leave residual bias, and the results must be replicated in future studies.

Conclusions

Change in gait speed during the period of hospitalization was inversely associated with risks of HF readmission and death in older AHF patients. The results presented here suggested that measurement of the short-term change in gait speed may be useful as a simple risk stratification tool in the clinical setting.

Conflict of Interest

The authors have no conflicts of interest to disclose.

Sources of Funding

This study was supported by the Grant for Clinical and Epidemiologic Research of the Joint Project of Japan Heart Foundation and the Japanese Society of Cardiovascular Disease Prevention Sponsored by AstraZeneca, and Grant for Clinical Research (Medical Profession) of the Japanese Circulation Society.

Disclosures

None.

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-19-0136

References
 
© 2019 THE JAPANESE CIRCULATION SOCIETY
feedback
Top