Exercise capacity and mortality – a follow-up study of 3033 subjects referred to clinical exercise testing

Figures & data

Table 1. Characteristics of the study population (n = 3033) according to all-cause mortality events.

All-cause mortality event	Men		p value	Women		p value
	Yes (n = 370)	No (n = 1422)		Yes (n = 155)	No (n = 1086)
Age (yr)	56.9 (9.6)	48.8 (9.8)	<0.001	59.3 (9.0)	51.4 (8.4)	<0.001
BMI (kg·m^−2)	27.1 (4.3)	26.6 (3.7)	0.018	27.2 (4.5)	26.8 (4.4)	0.316
Death due to CVD, n (%)	186 (50.3)			57 (36.8)
Diagnosed diseases and interventions
°Diabetes, n (%)^a	70 (18.9)	205 (14.4)	0.032	21 (13.5)	136 (12.5)	0.719
°Hypertension, n (%)^a	162 (43.8)	535 (37.6)	0.030	70 (45.2)	466 (42.9)	0.597
°CHD, n (%)^a	189 (51.1)	467 (32.8)	<0.001	63 (40.6)	274 (25.2)	<0.001
°History of myocardial infarction, n (%)^b	32 (8.6)	32 (2.3)	<0.001	4 (2.6)	12 (1.1)	0.129
°Heart failure, n (%)^b	11 (3.0)	12 (0.8)	0.003	8 (5.2)	7 (0.6)	<0.001
°Coronary bypass surgery/PCI, n (%)^b	18 (4.9)	19 (1.3)	<0.001	0 (0.0)	3 (0.3)
Medication used during exercise test
°β-Blockers, n (%)^c	90 (24.3)	154 (10.8)	<0.001	26 (16.8)	158 (14.5)	0.466

Values are means (SD) unless otherwise stated. Numbers do not match due to missing data. CVD: cardiovascular disease; CHD: coronary heart disease; PCI: percutaneous coronary intervention.

^aInformation after the baseline.

^bInformation at the baseline.

^cA patient was not included in users of β-blockers if there was a cessation of medication at least for one day.

Table 2. Baseline resting and exercise characteristics of the study population (n = 3033) according to all-cause mortality events.

	Men			Women
All-cause mortality event	Yes (n = 370)	No (n = 1422)	p value	Yes (n = 155)	No (n = 1086)	p value
Maximal exercise workload^a (W)	126.5 (41.9)	165.2 (42.1)	<0.001	82.9 (23.4)	100.1 (24.2)	<0.001
Ischemic ST-depression during exercise, n (%)	109 (29.5)	171 (12.0)	<0.001	29 (18.7)	77 (7.1)	<0.001
Resting heart rate (beats/min)	71.7 (14.9)	71.1 (14.0)	0.517	75.1 (14.4)	75.6 (13.5)	0.677
Resting systolic BP (mmHg)	148.2 (23.3)	138.1 (18.0)	<0.001	149.1 (21.6)	139.6 (20.3)	<0.001
Resting diastolic BP (mmHg)	89.4 (10.1)	86.9 (9.8)	<0.001	88.5 (9.2)	86.1 (9.8)	0.006
Maximal heart rate (beats/min)	145.3 (23.8)	163.5 (20.5)	<0.001	148.3 (20.1)	159.7 (17.2)	<0.001
Maximal systolic BP (mmHg)	202.7 (27.0)	205.9 (22.2)	0.032	199.9 (22.0)	197.8 (22.8)	0.284
Maximal diastolic BP (mmHg)	99.8 (13.2)	97.5 (11.8)	0.001	100.0 (11.1)	98.0 (11.5)	0.047
Heart rate reserve^b (beats/min)	73.7 (23.7)	92.5 (21.9)	<0.001	73.2 (19.3)	84.1 (17.8)	<0.001
Heart rate recovery^c (beats/min)	52.3 (16.6)	64.1 (15.5)	<0.001	53.9 (14.6)	61.4 (13.5)	<0.001
Systolic BP increase^d (mmHg)	54.5 (26.3)	67.8 (22.0)	<0.001	50.7 (19.2)	58.2 (19.6)	<0.001
Systolic BP recovery^e (mmHg)	46.8 (25.0)	58.4 (21.1)	<0.001	48.0 (21.7)	57.0 (20.4)	<0.001

Values are means (SD) unless otherwise stated. Numbers do not match due to missing data.

^aExercise workload was estimated as the mean of last four minutes of exercise.

^bHeart rate reserve was defined as a difference in heart rate between rest and maximum.

^cHeart rate recovery was defined as a difference in heart rate between maximum and four minutes of recovery.

^dSystolic blood pressure increase was defined as a difference in systolic blood pressure between rest and maximum.

^eSystolic blood pressure recovery was defined as a difference in systolic blood pressure between maximum and four minutes of recovery.

Figure 1. All-cause and CVD deaths are presented by Cox multivariate proportional hazard regression models according to achieved maximal workload (watts) during maximal exercise testing. (a) shows all-cause mortality in men and (b) in women. (c) shows CVD mortality in men and (d) in women. The cut-off values concerning CVD mortality for men were <125 W (n = 274), 125–162.49 W (n = 441), 162.5–189.12 W (n = 415) and ≥189.12 W (n = 378) and for women <81 W (n = 206), 81–99.99 W (n = 264), 100–112.49 W (n = 285) and ≥112.5 W (n = 311) according to quartiles. The cut-off values for all-cause mortality among men were <125 W (n = 348), 125–161.99 W (n = 485), 162–187.49 W (n = 404) and ≥187.5 W (n = 446) and in women <81.0 W (n = 240), 81–99.99 W (n = 293), 100–112.49 W (n = 304) and ≥112.50 W (n = 319).

Table 3. Sex-specific adjusted hazard ratios and their 95% CI for the association of exercise capacity during exercise stress test with CVD and all-cause mortality among the study population (n = 3033).

	Univariate hazard ratio (95% CI)	p-value	Multivariable-adjusted hazard ratio (95% CI)^b^,c	p-value
CVD mortality
Men
Maximal workload				<0.001^d
Q1, workload of <81.00/125 W^a	21.88 (9.55–50.14)	<0.001	4.24 (1.67–10.76)	0.002
Q2, 125.00–161.99 W^a	9.38 (4.06–21.68)	<0.001	4.04 (1.70–9.63)	0.002
Q3, 162.00–187.49 W^a	3.25 (1.31–8.04)	0.011	2.22 (0.89–5.53)	0.087
Q4, >187,49 W	reference		reference
Women
Maximal workload				<0.001^d
Q1, workload of <81.00 W^a	21.45 (5.10–90.21)	<0.001	5.41 (1.22–24.00)	0.027
Q2, 81–99.99 W^a	12.14 (2.84–51.94)	0.001	5.10 (1.16–22.40)	0.031
Q3, 100–112.49 W^a	3.82 (0.79–18.41)	0.094	2.41 (0.50–11.74)	0.275
Q4, >112.49W^a	reference		reference
All-cause mortality
Men
Maximal workload				<0.001^d
Q1, workload <125 W^a	10.16 (6.65–15.50)	<0.001	4.01 (2.50–6.44)	<0.001
Q2, workload 125-161.99 W^a	4.51 (2.92–6.96)	<0.001	2.78 (1.78–4.35)	<0.001
Q3, workload 162-187.49 W^a	2.24 (1.38–3.62)	0.001	1.87 (1.15–3.03)	0.011
Q4, >187,49 W	reference		reference
Women
Maximal workload				<0.001^d
Q1, workload of <81.00 W^a	6.37 (3.56, 11.40)	<0.001	2.25 (1.20–4.25)	0.012
Q2, 81–99.99 W^a	4.15 (2.29, 7.52)	<0.001	2.12 (1.15–3.92)	0.016
Q3, 100–112.49 W^a	2.00 (1.05, 3.83)	0.036	1.38 (0.72–2.67)	0.331
Q4, >112.49 W^a	Reference		Reference

Maximal exercise capacity is forced into the model as quartiles.

^aFor men the first quartile indicates exercise workload of <125.00 W; the second quartile, 125.00–161.99 W; the third quartile, 162.00–187.49 W; and the fourth quartile, > 187.49 W. For women the first quartile indicates exercise workload of <81.00 W; the second quartile, 81.00–99.99 W; the third quartile, 100.00–112.49 W; and the fourth quartile, > 112.49 W. The fourth quartile was used as a reference group.

^bFor men the full model was adjusted for age, BMI, maximal heart rate (beats/min), resting systolic BP (mmHg), CHD (yes vs. no), history of myocardial infarction (yes vs. no).

^cFor women the full model was adjusted for age, BMI, heart rate reserve defined as a difference in HR between rest and maximum (beats/min).

^dp-value for linear trend.