Benefits of exercise training and the correlation between aerobic capacity and functional outcomes and quality of life in elderly patients with coronary artery disease

Abstract

Cardiopulmonary exercise training is beneficial to people with coronary artery disease (CAD). Nevertheless, the correlation between aerobic capacity, and functional mobility and quality of life in elderly CAD patients is less addressed. The purpose of the current study is to investigate the beneficial effects of exercise training in elderly people with CAD, integrating exercise stress testing, functional mobility, handgrip strength, and health‐related quality of life. Elderly people with CAD were enrolled from the outpatient clinic of a cardiac rehabilitation unit in a medical center. Participants were assigned to the exercise training group (N = 21) or the usual care group (N = 15). A total of 36 sessions of exercise training, completed in 12 weeks, was prescribed. Echocardiography, exercise stress testing, the 6‐minute walking test, Timed Up and Go test, and handgrip strength testing were performed, and the Short‐Form 36 questionnaire (SF‐36) was administered at baseline and at 12‐week follow‐up. Peak oxygen consumption improved significantly after training. The heart rate recovery improved from 13.90/minute to 16.62/minute after exercise training. Functional mobility and handgrip strength also improved after training. Significant improvements were found in SF‐36 physical function, social function, role limitation due to emotional problems, and mental health domains. A significant correlation between dynamic cardiopulmonary exercise testing parameters, the 6‐minute walking test, Timed Up and Go test, handgrip strength, and SF‐36 physical function and general health domains was also detected. Twelve‐week, 36‐session exercise training, including moderate‐intensity cardiopulmonary exercise training, strengthening exercise, and balance training, is beneficial to elderly patients with CAD, and cardiopulmonary exercise testing parameters correlate well with balance and quality of life.

Introduction

Coronary artery disease (CAD) is a leading cause of mortality worldwide. Many modifiable and nonmodifiable risk factors for CAD are well documented [1]. Physical inactivity is one of the major concerns in healthcare. According to the World Health Organization (WHO), in 2009, the global prevalence of physical inactivity was 17%, which accounted for 6% of the burden of CAD [[2], [3]]. This emphasizes the crucial role of physical activity in health enhancement.

Beneficial effects of exercise and physical activity on CAD are generally established [[4], [5]]. The 6‐minute walk test (6MWT) is a simple method that evaluates the global response of all the systems involved during exercise and represents the submaximal functional capacity [6]. The 6MWT well predicts the functional capacity of people with heart failure and the benefits from cardiac rehabilitation after CAD, with adequate reproducibility [[7], [8], [9]]. However, the correlation between the 6MWT and quality of life was not clearly identified. Besides, exercise has been shown to improve balance and performance of daily activities [10]. The Timed Up and Go (TUG) test is a quick test that can identify population at risk for falls with good sensitivity and specificity [[11], [12]]. However, the correlation between cardiopulmonary fitness and balance improvement is less identified.

Exercise stress testing using a treadmill or bicycle is a widely accepted method for assessing the functional capacity in people with CAD. For persons with disabilities, handgrip exercise testing is an alternative stress method for the evaluation of CAD, albeit with limited sensitivity for the detection of CAD [[13], [14]]. Mroszczyk‐McDonald et al [15] reported improved handgrip strength after cardiac rehabilitation and suggested handgrip strength as a valuable predictor of physical function in elderly people with CAD.

The purpose of this study is to evaluate the benefits of exercise training in older people with CAD, integrating cardiopulmonary exercise testing, functional mobility, handgrip strength, and health‐related quality of life. Furthermore, we observed that cardiopulmonary fitness correlated with the aforementioned parameters in elderly people with CAD.

Methods

Participants

Participants were enrolled from the outpatient clinic of the department of physical medicine and rehabilitation in a medical center. The study population focused on the elderly, which refers to participants aged ≥65 years. Elderly people with CAD who underwent either percutaneous transluminal coronary angioplasty or coronary artery bypass graft surgery 3–6 months earlier and who did not receive any exercise training after the procedures were included in the study. The exclusion criteria were unstable angina, decompensated heart failure, unstable blood pressure control, peripheral arterial occlusive disease with claudication, degenerative osteoarthritis interfering with exercise, and other medical conditions that contraindicate exercise. Participants were then assigned to the exercise training group or the usual care group. General demographic features, including age, sex, height, and body weight, were collected during the interview. The study was approved by the Institutional Review Board of Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan (project number: VGHKS 10‐CT12‐12). Participants gave their informed consent before being enrolled in the study.

Interventions

In the exercise training group, exercise training included 50 minutes of cardiopulmonary exercise training, 30 minutes of strengthening exercise training, and 30 minutes of balance training. For the cardiopulmonary exercise training, a cycle ergometer was used. The training intensity was determined by the heart rate reserve method as described in the study by Scharff‐Olson et al [16], at 60–80% of the participant's heart rate reserve. The cardiopulmonary exercise training consisted of 10‐minute warm‐up, 30‐minute endurance training, and 10‐minute cool‐down. For strengthening exercise training, participants were given low‐level, progressive resistance exercise using light free weights and weight machines. The participants performed 12–15 repetitions of the exercise, with a resistance of 40–60% of one‐repetition maximum. Balance training focused on exercises related to daily mobility activities, as described in a study by Faber et al [17], which consisted of standing up from a chair, reaching, stepping forward and sideward, heel and toe stands, stepping on and over an obstacle, staircase walking, tandem foot standing, and single‐limb standing. All participants received three sessions of exercise training per week for 12 weeks, which was determined to be a more intense exercise program [18]. The whole exercise training program was monitored by a physician and a physical therapist.

In the usual care group, the participants received medical management and made clinic visits as needed.

Health‐related quality of life evaluation

Health‐related quality of life was evaluated using the Short‐Form 36 questionnaire (SF‐36). SF‐36 contains 36 items and measures eight domains, including physical function, role limitation due to physical problems, bodily pain, general health, vitality, social function, role limitation due to emotional problems, and mental health [19]. Each domain was scored from 0 (lowest level of functioning) to 100 (highest level). Studies suggested the validity and reliability of SF‐36 in the assessment of health perception in people with CAD [[20], [21]]. Participants were asked to complete the questionnaire at baseline and at 12‐week follow‐up.

Handgrip strength measurement

The handgrip strength was measured using a hydraulic hand‐held dynamometer at baseline and at 12‐week follow‐up. Both hands were measured three consecutive times. An average of three measures was used as the handgrip strength value.

Functional mobility and balance assessment

For functional mobility and balance assessment, participants were asked to perform the 6MWT and the TUG test at baseline and at 12‐week follow‐up. During the 6MWT, the participant was instructed to walk on a 30 m long, flat, hard‐surfaced, and straight corridor as far as possible during the 6‐minute period. Resting was allowed during the test. After the test was completed, the total distance walked was recorded [6]. For the TUG test, the participant was instructed to sit on a chair in the starting position; as the test began, the participant stood up from the chair, walked to a point 3 m away, then turned and walked back, and sat down on the chair. The time required to finish the test was recorded. The TUG test is a valid measure that correlates well with balance and can predict safe transfer particularly in the elderly [22]. The TUG test can also be an indicator for functional mobility in elderly population, because balance status is an important factor for good functional mobility.

Cardiopulmonary exercise testing

Aerobic capacity was measured by open‐circuit spirometry at baseline and after 12 weeks of exercise training. All participants were tested using Metamax 3B (Cortex Biophysik GmbH, Leipzig, Germany) consisting of a bicycle ergometer, a gas analyzer, and an electrocardiography monitor. Participants were asked to pedal on an upright bicycle ergometer to assess peak oxygen consumption (VO₂) and anaerobic threshold peak oxygen consumption. Exercise began with an intensity of 0 W workload for a 1‐minute warm‐up, followed by incremental loading using a ramp protocol (10 W/minute) until exhaustion [23]. Blood pressure was measured every minute by a cuff via an automatic blood pressure monitor. Continuous 12‐lead electrocardiography monitoring was used throughout the exercise testing. The exercise endpoint was based on the criteria established by the American College of Sports Medicine [24]. Briefly, exercise was terminated when the heart rate failed to increase with further increases in exercise intensity, peak oxygen consumption reached a plateau with increased workload, respiratory exchange ratio became ≥1.10, or the rate of perceived exertion was >17 on the 6–20 scale. In addition, the absolute indications for terminating exercise testing were followed. The heart rate recovery (HRR) after 1 minute of resting was also recorded. The whole exercise testing was conducted by a physician and a physical therapist, with all the participants completing the test safely.

Statistical analysis

Statistical analysis was performed using SAS version 9.3 for Windows (SAS Institute Inc., Cary, NC, USA). An alpha level of 0.05 was accepted as significant for all statistical procedures. All probability values were two tailed. Comparisons of the basic demographic data and characteristics between the exercise group and the usual care group were carried out by Wilcoxon rank‐sum test and by categorical variables, χ ², or Fisher's exact test, as appropriate. The mixed model was used to compare the difference at 12 weeks between the two groups. The difference between the exercise group and the usual care group at baseline and at 12 weeks was analyzed by a linear regression model. A mixed‐model repeated‐measure model was used to analyze the correlation between cardiopulmonary testing, functional mobility, handgrip strength, and SF‐36, interpreted as β regression coefficient. Interactions between intervention (yes/no) and factors (cardiopulmonary testing) were tested to predict functional mobility, handgrip strength, and SF‐36 scores using a mixed‐model repeated‐measures model with an added interaction term (intervention × factors).

Results

Characteristics of the study population

Sixty‐seven participants visiting the outpatient clinic of the Department of Physical Medicine and Rehabilitation in a medical center were assessed for eligibility. Fourteen participants declined to participate, six participants did not meet the inclusion criteria, and three participants could not follow the treatment schedule (Fig. 1). A total of 44 participants with CAD were enrolled in the study, and assigned to the exercise training group (N = 22) and the usual care group (N = 22). In the exercise training group, 21 participants completed the exercise training sessions, and one participant dropped out due to personal reasons. In the usual care group, seven participants were lost to follow‐up at the time of follow‐up at 12 weeks. No major cardiovascular events were recorded in these participants. Finally, 21 participants in the exercise training group and 15 in the usual care group were analyzed. The age of the study population ranged from 65 years to 80 years, with the mean age of 69.4 years, and 78% were male. Characteristics of the enrolled participants are shown in Table 1. No significant differences in baseline characteristics were found between the two groups.

Figure 1.

Flow diagram of enrollment and follow‐up.

Table 1.

Characteristics of the study population.

	Exercise (N = 21)	Usual care (N = 15)	p a
Age, y	69.7 ± 4.5	69.0 ± 4.6	0.5498
Sex
Male	16 (76)	12 (80)	> 0.99
Height, cm	166.7 ± 7.0	165.7 ± 5.0	0.4131
Weight, kg	69.9 ± 9.6	71.6 ± 10.2	0.8222
BMI, kg/m2	25.0 ± 2.8	26.0 ± 3.2	0.2681
Stenosis b	87.4 ± 11.7	69.0 ± 4.6	0.0799
STEMI	15 (71)	10 (67)	> 0.99
Left main	5 (24)	8 (53)	0.0690
Vessels
SVD	5 (24)	0 (0)
DVD	6 (28)	3 (20)
TVD	5 (24)	4 (27)
LM	5 (24)	8 (53)	0.1236
Killip
1	9 (43)	10 (67)
2	9 (43)	1 (7)
3 + 4	3 (14)	4 (26)	0.0594
Type of intervention
PTCA	15 (71)	10 (67)
CABG	6 (29)	5 (23)	> 0.99
HTN	18 (85)	10 (67)	0.2358
DM	10 (47)	4 (27)	0.2036
Smoker	15 (71)	12 (80)	0.7050
Sinus bradycardia	1 (5)	2 (13)	0.5588
Atrial fibrillation	2 (10)	0 (0)	0.5000
ESRD	1 (5)	0 (0)	> 0.99
PAOD	1 (5)	0 (0)	> 0.99
Major CV event	0 (0)	0 (0)	—
Beta blocker	15 (71)	9 (60)	0.4733

The shaded values indicate the re‐calculated p values of the comparisons of continuous variables by Wilcoxon rank‐sum test.

Data are presented as n (%) or mean ± SD.

BMI = body mass index; CABG = coronary artery bypass graft; CV = cardiovascular; DM = diabetes mellitus; DVD = double‐vessel disease; ESRD = end‐stage renal disease; HTN = hypertension; LM = left main lesion; PAOD = peripheral arterial occlusive disease; PTCA = percutaneous transluminal coronary angioplasty; STEMI = ST elevation myocardial infarction; SVD = single‐vessel disease; TVD = triple‐vessel disease; — = not calculated.

^aComparisons between exercise and usual care groups were carried out by Wilcoxon rank‐sum test and by categorical variables, χ ² or Fisher's exact test, as appropriate.

^bStenosis was calculated by the mean of each most severely stenotic vessel recorded.

Cardiac function and physical fitness improvement

Table 2 shows the changes in cardiovascular parameters and aerobic capacity at 12 weeks. Significant improvements in static [resting and peak heart rate (HR_rest and HR_peak), and blood pressure at peak workload (SBP_peak and DBP_peak] and dynamic [oxygen consumption at the anaerobic threshold (ATMETs) and peak workload (MET_peak)] cardiopulmonary parameters were observed in the exercise training group. Conversely, a significant decrease in aerobic capacity was observed in the usual care group. Significant between‐group differences were found in aerobic capacity, HR_peak, and HRR. The HRR improved from 13.90/minute to 16.62/minute after exercise training (p < 0.05).

Table 2.

Changes in cardiopulmonary testing, functional mobility and handgrip strength, and SF‐36 at 12 weeks.

	Exercise (N = 21)				Usual care (N = 15)				Exercise versus usual care
	Baseline	12 wk	Change	p a	Baseline	12 wk	Change	p a	Difference (SE)	p b
Cardiopulmonary testing
LVEF (%)	54.1 ± 7.6	54.4 ± 7.5	0.3 ± 0.9	0.0875	50.7 ± 7.1	50.9 ± 7.0	0.3 ± 1.5	0.4985	0.1 (0.4)	0.8479
SBPrest (mmHg)	123.4 ± 22.6	120.7 ± 19.7	−2.8 ± 20.1	0.5353	123.9 ± 14.5	125.2 ± 17.0	1.3 ± 11.7	0.6808	−4.0 (5.8)	0.4910
DBPrest (mmHg)	72.0 ± 14.7	71.3 ± 9.0	−0.7 ± 12.0	0.8022	75.9 ± 10.6	72.4 ± 10.5	−3.5 ± 8.5	0.1355	2.8 (3.6)	0.4448
HRrest (beats/min)	74.9 ± 11.6	69.2 ± 10.9	−5.7 ± 10.6	0.0231	71.9 ± 13.1	71.6 ± 11.3	−0.3 ± 7.6	0.8676	−5.3 (3.2)	0.1042
ATMETs	3.0 ± 0.5	3.4 ± 0.6	0.5 ± 0.3	<0.0001	3.1 ± 0.5	3.0 ± 0.5	−0.1 ± 0.2	0.0339	0.6 (0.1)	<0.0001
METpeak	4.0 ± 0.7	4.9 ± 0.9	0.8 ± 0.4	<0.0001	4.2 ± 0.7	4.1 ± 0.7	−0.2 ± 0.2	0.0200	1.0 (0.1)	<0.0001
HRpeak (beats/min)	109.8 ± 13.0	118.4 ± 17.1	8.7 ± 9.2	0.0003	109.3 ± 12.3	108.7 ± 9.7	−0.5 ± 5.5	0.7136	9.2 (2.7)	0.0015
RERpeak	1.13 ± 0.07	1.14 ± 0.06	0.01 ± 0.08	0.4498	1.14 ± 0.07	1.19 ± 0.12	0.05 ± 0.08	0.0274	−0.04 (0.03)	0.1498
SBPpeak (mmHg)	157.4 ± 30.7	167.0 ± 33.5	9.6 ± 18.8	0.0300	154.7 ± 25.3	166.5 ± 22.0	11.9 ± 27.6	0.1182	−2.3 (7.7)	0.7680
DBPpeak (mmHg)	76.8 ± 11.2	85.1 ± 15.6	8.2 ± 11.3	0.0034	80.1 ± 10.4	83.5 ± 9.0	3.3 ± 12.5	0.3190	4.9 (4.0)	0.2286
HRR (beats/min)	13.9 ± 3.5	16.6 ± 4.3	2.7 ± 3.9	0.0047	13.3 ± 2.2	13.5 ± 3.4	0.2 ± 3.1	0.8090	2.5 (1.2)	0.0475
Functional mobility and handgrip strength
6MWT (m)	345.6 ± 63.7	373.3 ± 62.4	27.7 ± 15.2	<0.0001	329.7 ± 47.0	323.9 ± 46.2	−5.8 ± 9.9	0.0393	33.5 (4.5)	<0.0001
TUGT (s)	12.7 ± 1.0	11.5 ± 1.0	−1.3 ± 0.6	<0.0001	13.5 ± 0.7∗	13.5 ± 0.7	0.0 ± 0.5	0.9552	−1.3 (0.2)	<0.0001
GRIP (kg)	11.0 ± 3.7	14.1 ± 4.3	3.2 ± 1.5	<0.0001	10.5 ± 4.8	10.5 ± 4.1	0.0 ± 1.1	0.9115	3.2 (0.5)	<0.0001
SF‐36
Physical	254.7 ± 35.4	283.8 ± 38.4	29.1 ± 29.8	0.0002	236.5 ± 32.7	237.2 ± 32.7	0.7 ± 27.2	0.9256	28.4 (9.7)	0.0061
pf	65.2 ± 11.2	84.7 ± 10.0	19.5 ± 10.3	<0.0001	57.3 ± 13.1	58.7 ± 13.6	1.4 ± 8.8	0.5468	18.1 (2.3)	<0.0001
Mental	278.9 ± 40.7	312.9 ± 35.1	34.0 ± 47.1	0.0035	254.6 ± 46.5	256.8 ± 20.6	2.3 ± 41.7	0.8362	31.7 (15.2)	0.0444

Data are presented as mean ± SD.

ATMETs = metabolic equivalents at the anaerobic threshold; DBP_rest = resting diastolic blood pressure; GRIP = handgrip strength; HR_peak = peak heart rate; HR_rest = resting heart rate; HRR = heart rate recovery after 1 minute of rest; LVEF = left ventricular ejection fraction; MET_peak = peak metabolic equivalent; pf = physical function domain; RER_peak = peak respiratory exchange ratio; SBP_rest = resting systolic blood pressure; SE = standard error; SF‐36 = Short‐Form 36 questionnaire; TUGT = Timed Up and Go test; 6MWT = 6‐minute walk test.

∗ p = 0.0134 when comparing the baseline means between two groups using the independent t test.

^aThe p values were derived from the paired t test, comparing the change of means between baseline and 12‐week training.

^bLinear mixed models were used to compare the differences between two groups.

Functional mobility evaluation and health‐related quality of life assessment

Table 2 demonstrates the results of functional mobility and handgrip strength, and health‐related quality‐of‐life assessment at baseline and at 12 weeks. Significant improvement in functional mobility was found in the exercise training group. The 6MWT results improved significantly from 345.6 m to 373.3 m (p < 0.0001). The TUG test results also improved significantly from 12.7 seconds to 11.5 seconds (p < 0.0001). The handgrip strength improved from 11.0 kg to 14.1 kg after exercise training (p < 0.0001). In the usual care group, a significant decrease was observed in functional mobility, with a reduced walking distance in the 6MWT (p < 0.05). Significant between‐group differences were found in case of all functional mobility measures. Regarding the assessment of quality of life, significant improvements were observed in the physical health and mental health measures of SF‐36, particularly the physical function domain, in the exercise training group. The between‐group differences for both physical health and mental health measures were also significant.

Correlation between aerobic capacity, functional mobility, handgrip strength, and health‐related quality of life

In addition to comparing the between‐group differences to determine the exercise training effect, we found that cardiopulmonary exercise testing correlated with health‐related quality of life, functional mobility, and handgrip strength, in particular those variables with a significant between‐group difference (shown in Table 2). Aerobic capacity (ATMETs and MET_peak) better correlated with functional mobility and handgrip strength in the exercise training group than in the usual care group (Table 3 and Fig. 2A–C). The TUG test best correlated with aerobic capacity. The HRR correlated with handgrip strength. In addition, health‐related quality of life, particularly the physical function domain, correlated with the 6MWT. Aerobic capacity also correlated with physical and mental health measures in the exercise training group, with a particularly significant correlation to physical health component of SF‐36 (Table 4 and Fig. 2D).

Table 3.

Modeling cardiopulmonary testing and SF‐36 as a linear predictor of functional mobility and handgrip strength. ^a

	Exercise						Usual care						Exercise versus usual care b
	6MWT	p	TUGT	p	GRIP	p	6MWT	p	TUGT	p	GRIP	p	6MWT	TUGT	GRIP
	β (SE)		β (SE)		β (SE)		β (SE)		β (SE)		β (SE)		p	p	p
ATMETs	48.0 (7.9)	<0.0001	−1.6 (0.3)	<0.0001	4.6 (0.8)	<0.0001	23.3 (12.6)	0.0865	0.1 (0.3)	0.8360	−0.5 (1.1)	0.6683	0.2255	0.0011	0.0189
METpeak	35.5 (3.3)	<0.0001	−1.1 (0.1)	<0.0001	3.2 (0.4)	<0.0001	27.6 (7.4)	0.0021	−0.4 (0.2)	0.0891	−0.8 (0.8)	0.2903	0.7179	0.0029	0.0016
HRpeak (beats/min)	1.8 (0.4)	<0.0001	−0.1 (0)	<0.0001	0.2 (0)	0.0007	0.2 (0.5)	0.7179	0 (0)	0.3492	0 (0.1)	0.6778	0.1130	0.0048	0.3922
HRR (beats/min)	4.7 (1.1)	0.0005	−0.2 (0)	0.0013	0.5 (0.1)	0.0007	0.1 (1.0)	0.9040	0 (0)	0.3715	0 (0.1)	0.6399	0.0659	0.1257	0.0203
Physical	0.6 (0.1)	<0.0001	0 (0)	0.0026	0 (0)	0.0048	−0.1 (0.1)	0.2960	0 (0)	0.2178	0 (0)	0.0224	0.0004	0.0907	0.2480
pf	1.3 (0.2)	<0.0001	−0.1 (0)	<0.0001	0.1 (0)	<0.0001	0.1 (0.3)	0.7254	0 (0)	0.2921	0 (0)	0.1957	0.0050	0.0025	0.1013
Mental	0.3 (0.1)	0.0310	0 (0)	0.0242	0 (0)	0.0017	−0.1 (0.1)	0.1060	0 (0)	0.7315	0 (0)	0.2458	0.0457	0.1000	0.0834

ATMETs = metabolic equivalents at the anaerobic threshold; GRIP = handgrip strength; HR_peak = peak heart rate; HRR = heart rate recovery after 1 minute of rest; MET_peak = peak metabolic equivalent; pf = physical function domain of SF‐36; SE = standard error; SF‐36 = Short‐Form 36 questionnaire; TUGT = Timed Up and Go test; 6MWT = 6‐minute walk test.

^aOnly the variables with a significant difference from baseline after exercise training (shown in Table 2) are included for further analysis.

^bInteraction p values [intervention (yes/no) × factors (cardiopulmonary testing and SF‐36)].

Figure 2.

Modeling aerobic capacity as a linear predictor of functional measures and physical health. MET_peak correlates better with (A, B) functional mobility (6MWT and TUGT), (C) GRIP, and (D) physical health measure of SF‐36 in the exercise training group than in the usual care group. GRIP = handgrip strength; MET_peak = peak metabolic equivalent; SF‐36 = Short‐Form 36 questionnaire; TUGT = Timed Up and Go test; 6MWT = 6‐minute walk test.

Table 4.

Modeling cardiopulmonary testing as a linear predictor of SF‐36.

	Exercise						Usual care						Exercise versus usual care a
	Physical	p	pf	p	Mental	p	Physical	p	pf	p	Mental	p	Physical	pf	Mental
	β (SE)		β (SE)		β (SE)		β (SE)		β (SE)		β (SE)		p	p	p
ATMETs	28.9 (10.3)	0.0110	14.9 (3.2)	0.0001	20.0 (11.1)	0.0860	8.6 (15.3)	0.5811	−9.4 (6.0)	0.1400	−1.5 (11.5)	0.9001	0.3127	0.0004	0.3682
METpeak	28.1 (5.4)	<0.0001	11.7 (1.9)	<0.0001	16.7 (7.0)	0.0252	1.4 (10.3)	0.8906	0.6 (4.1)	0.8850	−5.2 (7.4)	0.4907	0.0307	0.0012	0.1136
HRpeak (beats/min)	1.0 (0.4)	0.0181	0.5 (0.1)	0.0005	0.7 (0.4)	0.1216	0.2 (0.7)	0.7945	−0.1 (0.3)	0.8433	−0.1 (0.6)	0.8427	0.3339	0.1161	0.2645
HRR (beats/min)	0.7 (0.6)	0.2710	1.4 (0.5)	0.0110	1.7 (1.6)	0.2893	−0.2 (2.0)	0.9339	0.2 (0.7)	0.812	−0.1 (1.5)	0.9525	0.0646	0.3450	0.5458

ATMETs = metabolic equivalents at the anaerobic threshold; HR_peak = peak heart rate; HRR = heart rate recovery after 1 minute of rest; MET_peak = peak metabolic equivalent; pf = physical function domain of SF‐36; SE = standard error; SF‐36 = Short‐Form 36 questionnaire.

^aInteraction p values [intervention (yes/no) × factors (cardiopulmonary testing and SF‐36)].

Discussion

Our study results suggest that 12 weeks of exercise training, including moderate‐intensity cardiopulmonary exercise training, strengthening exercise, and balance training, improves aerobic capacity, functional mobility, and health‐related quality of life in elderly people with CAD. Moreover, peak oxygen consumption more consistently predicts balance, handgrip strength, and physical health in elderly patients with CAD. Thus, cardiopulmonary exercise testing provides information on more parameters than just aerobic capacity. It can also be used to predict balance status and quality of life in elderly CAD patients.

Physical activity in the elderly population has been shown to maintain good quality of life and, more importantly, to have beneficial effects on the primary and secondary prevention of CAD [25]. The importance of cardiac rehabilitation is emphasized in the general CAD population, specifically in the elderly CAD population [[4], [26]]. In our study, after 12weeks of moderate‐intensity exercise training, no significant change was found in left ventricular ejection fraction. However, significant improvements was observed in anaerobic threshold and peak oxygen consumption (15.4% and 20.6%, respectively), and the degree of aerobic exercise capacity improvement is comparable to that reported in previous studies [[27], [28]]. The usual care group, by contrast, has significantly declined aerobic capacity.

The HRR immediately after exercise is reflective of aerobic capacity, and a delayed HRR, defined as a reduction of the heart rate by <12 beats within the 1^st minute after peak exercise, is an independent predictor of mortality, especially in the older population [[29], [30], [31]]. The effect of cardiac rehabilitation on HRR has been demonstrated in both CAD and chronic heart failure populations [[32], [33]]. In our study, the HRR improved from 13.9 ± 3.45 bpm to 16.62 ± 4.28 bpm after 12 weeks of exercise training, an improvement that was less than that reported in the abovementioned studies. One possible explanation may be that the included age group was not quite comparable to studies above mentioned. The included population in our study was older, who have declined parasympathetic activity [34].

Cardiac rehabilitation has beneficial effects on health‐related quality of life and psychological status improvement in elderly CAD population [[5], [21]]. Subjective evaluation of quality of life after cardiac rehabilitation using SF‐36 provides good validity and reliability in the assessment of health perception in people with CAD [20]. In our study, except for the bodily pain domain, scores of all seven other domains of SF‐36 increase after 12 weeks of exercise training (data not shown), especially in the physical function domain. The bodily pain may be related, in part, to underlying musculoskeletal problems or other medical comorbidity. In addition, our results correlated with the objective cardiopulmonary exercise testing. Dynamic parameters of cardiopulmonary exercise testing, including the ATMETs and MET_peak, correlated with physical and mental health measures of SF‐36, particularly more significantly with the physical health measure in the exercise training group.

In elderly people with CAD, the 6MWT is a valid and reliable measure for the assessment of functional ability in Phases II and III cardiac rehabilitation [[9], [35]]. In our study, after 12 weeks of exercise training, the improvement in 6MWT results was from 345.62 m to 373.33 m, which was comparable to the minimal clinically important difference of 25 m determined in people with CAD [36]. The correlation between aerobic capacity, the 6MWT, and quality of life was less studied. Verrill et al [8] suggested no apparent correlation between the 6MWT and quality of life scores. In our study, we found that the improved aerobic capacity, particularly dynamic parameters such as ATMETs and MET_peak, correlated with the 6MWT, but not to a significant level. The improved performance in the 6MWT correlated significantly with physical and mental health measures of SF‐36. However, the TUG test better correlated with aerobic capacity.

Exercise can improve balance and daily activity performance, and the TUG test is a feasible and easy‐to‐perform test recommended for the screening of balance and risk of falling down in older population [10]. A cutoff value of >14 seconds is considered a good predictor for risk of falls in community‐dwelling, frail older adults without known neurologic diseases [11]. Few studies determined the correlation between aerobic capacity and the TUG test after exercise training. Gjellesvik et al [37] reported a significant increase in peak oxygen consumption and functional ability after treadmill training in people with chronic stroke. In our study results, significant improvement in the TUG test results was observed in the exercise training group. Moreover, the improved TUG test results significantly correlated with dynamic cardiopulmonary exercise testing parameters, ATMETs, and MET_peak. However, a low correlation was found between the TUG test and SF‐36. To our knowledge, in few studies, cardiopulmonary fitness correlated with functional mobility and balance in the elderly CAD population. The elderly and those with CAD are reported to have a higher risk of falls [38]. Based on our study results, we suggested that exercise training for the elderly CAD patients not only promotes better cardiopulmonary fitness, but also improves balance status in this elderly population who are vulnerable to falls.

Handgrip strength is a valid measure to estimate the physical functional status in the elderly [39]. Our study revealed significantly increased handgrip strength after 12 weeks of exercise training. However, the handgrip strength measured in our study group was generally weaker, improving from 10.95 kg to 14.14 kg in the exercise training group. There is a significant correlation between handgrip strength, aerobic capacity, and the HRR. Handgrip strength is predictive of the general health status, particularly in aging adults, for prediction of premature mortality, development of disability, and risk of complications or prolonged length of stay during hospitalization [40]. The study by Wallymahmed et al [41] suggested a positive correlation between aerobic capacity and handgrip strength in patients with type 1 diabetes. Ades et al [42] also reported that handgrip strength was a good predictor of peak aerobic capacity in elderly CAD patients with a self‐reported disability. A lower handgrip strength was also reported to predict the failure to improve cardiopulmonary fitness in cardiac rehabilitation [43]. Thus, exercise training in elderly CAD patients should not only focus on aerobic exercise training, but also emphasize the importance of strengthening exercise in promoting better cardiopulmonary fitness. The HRR is another measure that represents aerobic capacity. A literature review showed that there are not many reports discussing the correlation between handgrip strength and the HRR, and further analysis should be performed. Furthermore, in our study, handgrip strength. Furthermore, in our study, handgrip strength showed no significant correlation with physical and mental health measures of SF‐36.

The abovementioned results suggest that using the 6MWT solely as a functional mobility assessment tool may not well correlate with aerobic capacity. The use of the TUG test and handgrip strength in addition may better predict aerobic capacity in elderly people with CAD. We modeled aerobic capacity as a linear predictor of functional measures and physical health, as shown in Fig. 2. The cutoff point for prognostic evaluation should be further investigated.

The limitation of our present study is that we focused on a specific population, the elderly people with CAD. Further stratification with age‐matched comparison is needed, because cardiopulmonary fitness and functional ability may vary in different age groups of the elderly population.

Conclusion

In summary, 12‐week, 36‐session exercise training, including moderate‐intensity cardiopulmonary exercise training, strengthening exercise, and balance training, is beneficial for elderly people with CAD. The benefits pertain not only to increased cardiopulmonary fitness that may further prevent another major coronary event, but also to improved functional mobility, balance, and quality of life. Furthermore, the improvement in aerobic capacity correlates well with balance and perception of physical health in elderly people with CAD. This addresses the important concept that appropriate exercise training in elderly CAD patients not only shows cardiovascular benefits, but also possibly reduces the risk of falls in this vulnerable population.

Acknowledgments

This work was supported by Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan (grant number VGHKS‐100‐062).