Heart rate reactivity, recovery, and endurance of the incremental shuttle walk test in patients prone to heart failure

Abstract

Aims

Few randomized trials assessed the changes over time in the chronotropic heart rate (HR) reactivity (CHR), HR recovery (HRR) and exercise endurance (EE) in response to the incremental shuttle walk test (ISWT). We addressed this issue by analysing the open HOMAGE (Heart OMics in Aging) trial.

Methods

In HOMAGE, 527 patients prone to heart failure were randomized to usual treatment with or without spironolactone (25–50 mg/day). The current sub‐study included 113 controls and 114 patients assigned spironolactone (~70% on beta‐blockers), who all completed the ISWT at baseline and at Months 1 and 9. Within‐group changes over time (follow‐up minus baseline) and between‐group differences at each time point (spironolactone minus control) were analysed by repeated measures ANOVA, unadjusted or adjusted for sex, age and body mass index, and additionally for baseline for testing 1 and 9 month data.

Results

Irrespective of randomization, the resting HR and CHR did not change from baseline to follow‐up, with the exception of a small decrease in the HR immediately post‐exercise (−3.11 b.p.m.) in controls at Month 9. In within‐group analyses, HR decline over the 5 min post‐exercise followed a slightly lower course at the 1 month visit in controls and at the 9 month visits in both groups, but not at the 1 month visit in the spironolactone group. Compared with baseline, EE increased by two to three shuttles at Months 1 and 9 in the spironolactone group but remained unchanged in the control group. In the between‐group analyses, irrespective of adjustment, there were no HR differences at any time point from rest up to 5 min post‐exercise or in EE. Subgroup analyses by sex or categorized by the medians of age, left ventricular ejection fraction or glomerular filtration rate were confirmatory. Combining baseline and Months 1 and 9 data in both treatment groups, the resting HR, CHR and HRR at 1 and 5 min averaged 61.5, 20.0, 9.07 and 13.8 b.p.m. and EE 48.3 shuttles.

Conclusions

Spironolactone on top of usual treatment compared with usual treatment alone did not change resting HR, CHR, HRR and EE in response to ISWT. Beta‐blockade might have concealed the effects of spironolactone. The current findings demonstrate that the ISWT, already used in a wide variety of pathological conditions, is a practical instrument to measure symptom‐limited exercise capacity in patients prone to developing heart failure because of coronary heart disease.

Introduction

The reliable assessment of exercise capacity provides important diagnostic and prognostic information in patients with cardiac¹ and pulmonary disease² and is also widely used to evaluate the efficacy of new therapies. Changes in heart rate (HR) and blood pressure provide important diagnostic and prognostic information.^3,4 Conventionally, symptom-limited, graded, bicycle or treadmill exercise tests are used to determine maximum exercise capacity.⁵ This requires a substantial amount of equipment and is an unfamiliar type of exercise for many patients. The incremental shuttle walk test (ISWT) is an alternative symptom-limited test that requires little equipment and involves a more familiar type of exercise (walking), which has been used to assess the exercise capacity of patients with chronic heart failure (HF).⁶ The ISWT is designed to provoke symptoms and assess maximum exercise capacity.^6,7 The 6 min walking test (6MWT) is designed to assess submaximal exercise capacity.⁷

The chronotropic HR reactivity (CHR) in response to exercise, the delayed HR recovery (HRR) after exercise or exercise endurance (EE) are independent predictors of a worse prognosis, cardiovascular endpoints, post-surgical complications and all-cause mortality in a variety of settings, including older adults,^8–10 patients with pulmonary arterial hypertension,^11–13 patients with obstructive^14–16 or interstitial^17,18 pulmonary disease, cancer patients undergoing lung^19,20 or abdominal²¹ surgery, patients with a history of myocardial infarction,⁷ HF^22–24 or chronic kidney disease²⁵ or patients referred for exercise testing.^26,27 However, few randomized trials assessed the changes over time in CHR, HRR and EE. To address the consistency of the results of exercise testing over time, we analysed CHR, HRR and EE in response to the ISWT in the HOMAGE (Heart OMics in Aging) trial, in which patients at risk of HF were randomized to usual treatment or spironolactone on top of usual treatment.²⁸ The trial design allowed for the assessment of between-group differences in addition to within-group changes over time and to evaluate the reproducibility of the ISWT results as validation in the absence of within-trial validation by state-of-the-art treadmill or bicycle tests.

Methods

Study participants

HOMAGE is a multicentre open-label trial with blinded endpoint evaluation (Registration Number: NCT02556450),²⁸ conducted in nine centres in the United Kingdom, France, Italy, Ireland, Germany and the Netherlands. Each centre had its own recruitment strategies. The protocol was approved by the Greater Manchester Central Research Ethics Committee (Reference Number: 16/NW/0012; EudraCT Number: 2015-000413-48) as well as by each centre's local Ethics Committee. Patients of either sex, aged ≥60 years, were eligible provided that they were at increased risk of developing HF because they already had or were likely to develop coronary heart disease. Additionally, eligible patients had to have a plasma N-terminal pro-brain natriuretic peptide (NT-proBNP) of 125–1000 ng/L or a plasma brain natriuretic peptide (BNP) of 35–280 ng/L. These ranges excluded patients at low HF risk as well as those with advanced disease requiring further investigation and treatment. The main exclusion criteria were an estimated glomerular filtration rate (eGFR)²⁹ of <30 mL/min/1.73 m², serum potassium of >5.0 mmol/L, left ventricular ejection fraction of <45%, atrial fibrillation, a diagnosis of HF prior to randomization and treatment with loop diuretics.

Of the 877 screened patients (Figure 1), 527 were randomized to spironolactone 25–50 mg/day (n = 265) on top of usual treatment or usual treatment alone (n = 262).²⁸ Of all patients randomized and followed up in the HOMAGE trial, 450/527 (85.4%), 324/516 (62.8%) and 400/506 (79.1%) completed the ISWT at baseline and at Months 1 and 9. The current analyses included 227 patients who completed the ISWTs at each of these three time points, the justification being that evaluating the same patients at each time point increases the comparability of the data over time. Of the 227 patients, 113 were randomized to control and 114 to spironolactone. NT-proBNP and high-sensitivity troponin T were assessed by electro-chemiluminescent assays (Roche Diagnostics).

[IMAGE OMITTED. SEE PDF]

Incremental shuttle walk test

Exercise capacity was measured by the ISWT.⁶ Investigators were asked to conduct a familiarization test for each participant prior to the baseline assessment. As explained in detail in the supporting information, skilled personnel conducted the ISWT using a 10 m course (the shuttle) marked by two cones. The walking speed was determined by bleeps played from a compact disc. After every minute, walking speed increased. There are up to 12 levels of speed and, potentially, 102 shuttles. HR was measured at rest and immediately after the ISWT, and 1, 2, 3 and 5 min after completion of the ISWT. The test was performed at baseline and at Months 1 and 9. CHR was the difference between the HR immediately after the ISWT and the resting HR. Early and late HRR was the maximal HR immediately post-exercise minus the HR at 1 and 5 min post-exercise. An impaired HRR is the difference between the maximal and the 1 min HR of <12 b.p.m.³⁰ EE was assessed by the number of completed shuttles.

Statistical analysis

For database management and statistical analysis, SAS software, Version 9.4 (SAS Institute Inc., Cary, NC, USA), was used. For comparison of means, we used a paired or unpaired t-test, as appropriate, or a Wilcoxon–Mann–Whitney test depending on the distribution. Unpaired and pairwise comparisons of proportions were done by the χ²-statistic and the McNemar test, respectively. The significance was a two-sided α level of ≤0.05. NT-proBNP was logarithmically transformed (base 10) to approximate the normal distribution.

The analyses focused on within-group changes over time (follow-up minus baseline) and on the between-group differences (spironolactone minus placebo) in CHR, HRR and EE. Changes in the ISWT-related variables from baseline to follow-up were given as signed differences and as percentage changes, using the baseline value as the denominator. The differences in the serial HR values during ISWT were implemented by repeated measures ANOVA with time point (within-group changes over time) or time point and treatment (between-group comparisons) as class variables and with the individual patient modelled as a random effect. In sensitivity analyses, the data were stratified by sex and the medians of age, left ventricular ejection fraction and eGFR. For the computation of the intraclass correlation coefficient (ICC), we used a published SAS macro.³¹ ICC values of 0.5–0.6 indicate moderate, 0.7–0.8 strong and >0.8 perfect agreement between two ISWTs.³¹

Results

Patient characteristics

Descriptive data for the 227 analysed HOMAGE patients are shown in Table 1. No patients had a history of hospitalized HF prior to randomization. Most patients were receiving antihypertensive agents (n = 163; 71.8%), lipid-lowering drugs (n = 203; 89.4%), mainly statins (n = 197; 86.8%) and antiplatelet agents (n = 168; 74.0%), and 81 (35.7%) were on treatment with hypoglycaemic agents. Over time, there was no change in the use of antihypertensive drugs in either treatment group (Table S2). At any time during randomized follow-up, only 18 patients (7.93%) were on thiazide diuretics, but 159 (70.0%) were taking beta-blockers. The mean left ventricular ejection fraction was 62.8% (interquartile range: 59.3%–66.9%). At baseline and at the 1 and 9 month visits, 10 (4.41%), 6 (2.64%) and 12 (5.29%) patients used a walking aid: 10 (4.41%) only on one occasion, 6 (2.64%) twice and 2 (0.88%) at each test. Patients, randomized to control or spironolactone, were well balanced with regard to risk factors, clinical characteristics and routine biochemistry (Table 1). The 227 patients included in the present sub-study had broadly similar characteristics compared with the 300 HOMAGE patients not included (Table S1). However, the patients reported here were younger, had a higher eGFR, were less likely to smoke (5.73% vs. 10.3%; P = 0.040) but had a higher prevalence of ischaemic heart disease (79.3% vs. 66.3%; P = 0.001).

Table 1 Baseline characteristics of patients by trial arm.

Characteristic	Control	Spironolactone	P-value
Number with characteristic	113	114
Women	26 (23.0)	23 (20.2)	0.60
Caucasian	112 (99.1)	109 (96.5)	0.50
Current smoking	8 (7.08)	5 (4.39)	0.56
Hypertension	84 (74.3)	87 (76.3)	0.73
Treated hypertension	81 (96.4)	82 (94.3)	0.97
Diabetes	45 (39.8)	42 (36.8)	0.64
Treated diabetes	40 (88.9)	41 (97.6)	0.93
History of coronary artery disease	90 (79.7)	90 (79.0)	0.90
History of myocardial infarction	46 (51.1)	47 (52.2)	0.88
Clinical characteristics
Age (years)	72.4 ± 5.93	72.0 ± 6.16	0.61
BMI (kg/m2)	28.7 ± 4.96	29.6 ± 5.42	0.18
Waist-to-hip ratio	0.97 ± 0.07	0.98 ± 0.07	0.15
Biochemistry
Serum sodium (mmol/L)	139 (138–141)	138 (136–139)	0.40
Serum potassium (mmol/L)	4.3 (4.1–4.6)	4.5 (4.2–4.7)	0.42
eGFR (mL/min/1.73 m2)	72 (61–82)	76 (61–89)	0.074
Plasma hsTnT (ng/L)	12.2 (8.7–17.4)	11.7 (8.4–15.0)	0.064
Plasma NT-proBNP (ng/L)	204 (118–289)	170 (120–331)	0.61

Changes over time on usual treatment

Over the first month (Table 2), the resting HR and the HR immediately post-exercise did not change (P ≥ 0.10), but during the recovery period, the HR decline followed a lower course at Month 1 compared with baseline, reaching significance at 3 and 5 min (P = 0.014). The signed within-group changes (Month 1 minus baseline) in CHR and the early and late HRR were 0.08 (P = 0.94), 0.25 (P = 0.77) and 0.64 b.p.m. (P = 0.52), respectively (Table 3). From baseline to the last follow-up (Table 2), the resting HR did not change (P = 0.55), while HR decreased immediately and 2 and 5 min after exercise (P ≤ 0.037). At last follow-up compared with baseline, the signed within-group changes (Month 9 minus baseline) in CHR and the early and late HRR were −2.51 (P = 0.075), −1.62 (P = 0.24) and −1.12 b.p.m. (P = 0.37), respectively (Table 3). The percentage of control patients with impaired HRR at baseline was 69.0%, and at the 1 and 9 month visits, it was 68.1% (P > 0.99) and 66.4% (P = 0.72). The mean number of completed shuttles on usual treatment was 48.5 at baseline and 48.3 and 49.0 at Months 1 and 9 (Table 2). None of the within-group changes in the number of completed shuttles reached significance (P ≥ 0.61).

Table 2 Changes over time in resting and post-exercise heart rate and walking distance in the control group.

Characteristic	Baseline	Follow-up	P-value	Differences	ICC (95% CI)
Signed (95% CI)	Percentage (95% CI)
Baseline vs. Month 1
Resting HR, b.p.m.	62.1 ± 9.46	60.8 ± 9.10	0.10	−1.26 (−2.78 to 0.26)	−1.23 (−3.51 to 1.06)	0.61 (0.48 to 0.71)
Post-exercise HR
Immediate, b.p.m.	83.1 ± 21.1	81.9 ± 20.2	0.33	−1.18 (−3.56 to 1.20)	0.15 (−3.04 to 3.34)	0.81 (0.73 to 0.86)
1 min, b.p.m.	73.7 ± 14.5	72.2 ± 14.8	0.16	−1.43 (−3.41 to 0.55)	−1.02 (−3.80 to 1.76)	0.74 (0.64 to 0.81)
2 min, b.p.m.	70.6 ± 12.5	69.1 ± 12.5	0.075	−1.55 (−3.26 to 0.16)	−1.46 (−3.91 to 0.99)	0.73 (0.63 to 0.81)
3 min, b.p.m.	69.6 ± 11.9	67.7 ± 11.6	0.014	−1.95 (−3.52 to −0.39)	−2.10 (−4.35 to 0.16)	0.73 (0.64 to 0.81)
5 min, b.p.m.	68.8 ± 11.2	67.0 ± 10.5	0.014	−1.81 (−3.24 to −0.38)	−1.95 (−4.02 to 0.11)	0.74 (0.65 to 0.82)
Shuttle number	48.5 ± 22.1	48.3 ± 22.2	0.83	−0.17 (−1.73 to 1.40)	2.57 (−2.57 to 7.71)	0.93 (0.90 to 0.95)
Baseline vs. Month 9
Resting HR	62.1 ± 9.46	61.5 ± 11.4	0.55	−0.59 (−2.54 to 1.36)	−0.18 (−3.31 to 2.94)	0.50 (0.35 to 0.63)
Post-exercise HR
Immediate	83.1 ± 21.1	80.0 ± 19.9	0.037	−3.11 (−6.02 to −0.19)	−1.72 (−5.26 to 1.82)	0.70 (0.60 to 0.78)
1 min	73.7 ± 14.5	72.2 ± 15.8	0.29	−1.49 (−4.28 to 1.31)	−0.34 (−4.85 to 4.16)	0.51 (0.36 to 0.64)
2 min	70.6 ± 12.5	68.5 ± 11.7	0.016	−2.18 (−3.94 to −0.42)	−2.15 (−4.65 to 0.35)	0.68 (0.57 to 0.77)
3 min	69.6 ± 11.9	68.0 ± 15.1	0.21	−1.59 (−4.11 to 0.92)	−1.39 (−5.07 to 2.29)	0.50 (0.35 to 0.63)
5 min	68.8 ± 11.2	66.8 ± 10.8	0.014	−1.99 (−3.55 to −0.42)	−2.12 (−4.40 to 0.17)	0.70 (0.59 to 0.78)
Shuttle number	48.5 ± 22.1	49.0 ± 22.5	0.61	0.49 (−1.42 to 2.39)	6.69 (−0.01 to 13.4)	0.90 (0.85 to 0.93)

Table 3 Changes over time in the chronotropic heart rate reactivity and recovery in response to exercise.

Characteristic	Baseline	Follow-up	P-value	Differences	ICC (95% CI)
Signed (95% CI)	Percentage (95% CI)
Control group
Baseline vs. 1 month
CHR, b.p.m.	21.1 ± 19.8	21.1 ± 19.3	0.94	0.08 (−1.88 to 2.04)	−13.5 (−57.0 to 29.9)	0.86 (0.80 to 0.90)
HRR vs. maximal HR
1 min, b.p.m.	9.46 ± 11.4	9.72 ± 10.6	0.77	0.25 (−1.43 to 1.93)	15.9 (−27.6 to 59.4)	0.67 (0.55 to 0.76)
5 min, b.p.m.	14.3 ± 14.8	15.0 ± 14.1	0.52	0.64 (−1.33 to 2.61)	9.60 (−57.4 to 76.6)	0.73 (0.64 to 0.81)
Impaired HRR, n (%)	78 (69.0)	77 (68.1)	>0.99	0.88 (−7.15 to 8.92)	…	…
Baseline vs. 9 months
CHR, b.p.m.	21.1 ± 19.8	18.5 ± 20.0	0.075	−2.51 (−5.28 to 0.25)	−14.1 (−50.5 to 22.4)	0.72 (0.62 to 0.80)
HRR vs. maximal HR
1 min, b.p.m.	9.46 ± 11.4	7.84 ± 16.2	0.24	−1.62 (−4.33 to 1.09)	37.0 (−11.5 to 85.5)	0.46 (0.30 to 0.59)
5 min, b.p.m.	14.3 ± 14.8	13.2 ± 14.8	0.37	−1.12 (−3.59 to 1.35)	42.4 (−12.4 to 97.2)	0.60 (0.47 to 0.70)
Impaired HRR, n (%)	78 (69.0)	75 (66.4)	0.72	2.65 (−7.87 to 13.2)	…	…
Spironolactone group
Baseline vs. 1 month
CHR, b.p.m.	19.2 ± 18.9	20.4 ± 19.2	0.34	1.22 (−1.31 to 3.75)	−15.3 (−51.7 to 21.1)	0.74 (0.65 to 0.82)
HRR vs. maximal HR
1 min, b.p.m.	8.64 ± 10.5	8.82 ± 10.1	0.87	0.17 (−1.88 to 2.23)	−53.6 (−109.8 to 2.58)	0.43 (0.27 to 0.56)
5 min, b.p.m.	13.2 ± 13.0	13.3 ± 14.2	0.91	0.13 (−2.05 to 2.32)	−57.5 (−112.4 to −2.50)	0.63 (0.50 to 0.73)
Impaired HRR, n (%)	75 (65.8)	76 (66.7)	>0.99	−0.88 (−12.5 to 10.7)	…	…
Baseline vs. 9 months
CHR, b.p.m.	19.2 ± 18.9	19.5 ± 19.3	0.84	0.30 (−2.72 to 3.31)	9.29 (−39.2 to 57.8)	0.64 (0.52 to 0.74)
HRR vs. maximal HR
1 min, b.p.m.	8.64 ± 10.5	9.95 ± 10.4	0.24	1.31 (−0.89 to 3.50)	−21.9 (−92.2 to 48.3)	0.35 (0.18 to 0.50)
5 min, b.p.m.	13.2 ± 13.0	14.0 ± 14.0	0.55	0.76 (−1.74 to 3.26)	−30.2 (−91.4 to 30.9)	0.50 (0.36 to 0.63)
Impaired HRR, n (%)	75 (65.8)	71 (62.3)	0.64	3.51 (−8.22 to 15.2)	…	…

Changes over time on spironolactone

Over the first month (Table 4), the resting and the HRs immediately after exercise and during the HRR period up to 5 min post-exercise did not change (P ≥ 0.32). The signed within-group changes (1 month minus baseline) in CHR and in the early and late HRR were 1.22 (P = 0.34), 0.17 (P = 0.87) and 0.13 b.p.m. (P = 0.91), respectively (Table 3). At the last follow-up compared with baseline, the signed within-group changes (Month 9 minus baseline) in CHR and in the early and late HRR were 0.30 (P = 0.84), 1.31 (P = 0.24) and 0.76 b.p.m. (P = 0.55), respectively (Table 3). The percentage of patients assigned spironolactone with an impaired HRR at baseline was 65.8%, and at the 1 and 9 month visits, it was 66.7% (P > 0.99) and 62.3% (P = 0.64; Table 3). The number of completed shuttles averaged 46.4, 48.3 and 49.3 at baseline and at Months 1 and 9, respectively (Table 4). The within-group changes from baseline to the follow-up visits were significant (P ≤ 0.030). The ICCs showed moderate to strong agreement between the baseline and the follow-up data, irrespective of treatment assignment (Tables 2–4).

Table 4 Changes over time in resting and post-exercise heart rate and walking distance in the spironolactone group.

Characteristic	Baseline	Follow-up	P-value	Differences	ICC (95% CI)
Absolute (95% CI)	Percentage (95% CI)
Baseline vs. Month 1
Resting HR, b.p.m.	62.0 ± 9.79	61.8 ± 9.54	0.79	−0.19 (−1.59 to 1.21)	0.44 (−1.75 to 2.63)	0.70 (0.59 to 0.78)
Post-exercise HR
Immediate, b.p.m.	81.2 ± 20.1	82.2 ± 20.0	0.45	1.03 (−1.68 to 3.74)	2.85 (−0.50 to 6.20)	0.74 (0.64 to 0.81)
1 min, b.p.m.	72.5 ± 15.0	73.4 ± 14.8	0.46	0.85 (−1.44 to 3.15)	2.51 (−0.35 to 5.37)	0.66 (0.54 to 0.75)
2 min, b.p.m.	69.9 ± 13.7	70.4 ± 12.5	0.59	0.48 (−1.29 to 2.25)	1.86 (−0.61 to 4.32)	0.74 (0.64 to 0.81)
3 min, b.p.m.	68.8 ± 12.8	69.1 ± 11.9	0.73	0.29 (−1.38 to 1.97)	1.44 (−0.86 to 3.73)	0.74 (0.64 to 0.81)
5 min, b.p.m.	68.0 ± 12.1	68.9 ± 12.0	0.32	0.90 (−0.87 to 2.66)	2.27 (−0.27 to 4.82)	0.69 (0.58 to 0.77)
Shuttle number	46.4 ± 21.7	48.3 ± 23.1	0.025	1.93 (0.25 to 3.61)	7.35 (1.53 to 13.2)	0.92 (0.88 to 0.94)
Baseline vs. Month 9
Resting HR, b.p.m.	62.0 ± 9.79	60.8 ± 9.38	0.12	−1.20 (−2.71 to 0.31)	−1.09 (−3.41 to 1.23)	0.64 (0.51 to 0.73)
Post-exercise HR
Immediate, b.p.m.	81.2 ± 20.1	80.3 ± 19.0	0.58	−0.90 (−4.12 to 2.32)	1.31 (−2.71 to 5.33)	0.61 (0.48 to 0.71)
1 min, b.p.m.	72.5 ± 15.0	70.3 ± 12.4	0.091	−2.21 (−4.78 to 0.36)	−0.89 (−4.29 to 2.50)	0.49 (0.33 to 0.61)
2 min, b.p.m.	69.9 ± 13.7	67.7 ± 11.1	0.042	−2.19 (−4.30 to −0.08)	−1.47 (−4.27 to 1.34)	0.57 (0.44 to 0.68)
3 min, b.p.m.	68.8 ± 12.8	66.6 ± 10.0	0.021	−2.17 (−4.00 to −0.34)	−1.68 (−4.19 to 0.82)	0.62 (0.49 to 0.72)
5 min, b.p.m.	68.0 ± 12.1	66.3 ± 9.69	0.076	−1.66 (−3.50 to 0.18)	−1.01 (−3.56 to 1.54)	0.59 (0.45 to 0.69)
Shuttle number	46.4 ± 21.7	49.3 ± 25.3	0.030	2.89 (0.28 to 5.50)	13.5 (3.89 to 23.2)	0.82 (0.75 to 0.87)

Between-group differences

Figure 2 shows the time course of HR from rest to 5 min post-exercise. There were no significant between-group differences, irrespective of the time point in the trial (P ≥ 0.11). Subgroup analyses stratified for sex (Figure S1) or median age (<71 vs. ≥71 years; Figure S2), left ventricular ejection fraction (<63% vs. ≥63%; Figure S3), eGFR (<73 vs. ≥73 mL/min/1.73 m²; Figure S4) or use of beta-blockers (Figure S5) produced results similar to those shown in Figure 2 without significant subgroup-by-time point interactions (P ≥ 0.059). However, at baseline, in women (Figure S1), age < 71 years (Figure S2) or eGFR < 73 mL/min/1.73 m² (Figure S4), HR followed a slightly higher course in the control group compared with the spironolactone group (P ≤ 0.033). Unadjusted and adjusted analyses (Table 5) did not reveal any between-group differences at baseline or Months 1 and 9.

[IMAGE OMITTED. SEE PDF]

Table 5 Heart rate reactivity, recovery and exercise endurance by randomization group.

Variable	Control	Difference (95% confidence interval)
Control (n = 113)	Spironolactone (n = 114)	Unadjusted	P-value	Adjusted	P-value
CHR, b.p.m.
Baseline	21.1 ± 19.8	19.2 ± 18.9	−1.86 (−6.93 to 3.21)	0.47	−2.03 (−6.87 to 2.82)	0.41
Month 1	21.1 ± 19.3	20.4 ± 19.2	−0.72 (−5.76 to 4.32)	0.78	0.58 (−2.44 to 3.59)	0.71
Month 9	18.5 ± 20.0	19.5 ± 19.3	0.95 (−4.19 to 6.10)	0.72	2.11 (−1.67 to 5.90)	0.27
Early HRR, b.p.m.
Baseline	9.46 ± 11.4	8.64 ± 10.5	−0.82 (−3.68 to 2.05)	0.57	−0.84 (−3.64 to 1.97)	0.56
Month 1	9.72 ± 10.6	8.82 ± 10.1	−0.89 (−3.61 to 1.81)	0.51	−0.45 (−2.73 to 1.83)	0.70
Month 9	7.84 ± 16.2	9.95 ± 10.4	2.11 (−1.45 to −5.67)	0.24	2.51 (−0.71 to 5.73)	0.13
Late HRR, b.p.m.
Baseline	14.3 ± 14.8	13.2 ± 13.0	−1.12 (−4.76 to 2.52)	0.55	−1.13 (−4.69 to 2.43)	0.53
Month 1	15.0 ± 14.1	13.3 ± 14.2	−1.62 (−5.32 to 2.08)	0.39	−0.85 (−3.58 to 1.88)	0.54
Month 9	13.2 ± 14.8	14.0 ± 14.0	0.76 (−3.01 to 4.53)	0.69	1.24 (−1.92 to 4.41)	0.44
EE, n
Baseline	48.5 ± 22.1	46.4 ± 21.7	−2.11 (−7.84 to 3.63)	0.47	−1.93 (−6.88 to 3.02)	0.44
Month 1	48.3 ± 22.2	48.3 ± 23.1	−0.01 (−5.94 to 5.92)	>0.99	2.15 (−0.10 to 4.40)	0.061
Month 9	49.0 ± 22.5	49.3 ± 25.3	0.30 (−5.96 to 6.56)	0.93	2.49 (−0.75 to 5.74)	0.13

Discussion

The HOMAGE trial offered the opportunity to assess in patients randomized to usual treatment with or without spironolactone the within-group changes over time and the between-group differences in the ISWT-related key variables, that is, CHR, HRR and EE. In the within-group analyses, the main findings were as follows: First, irrespective of randomization, resting HR and CHR did not change from baseline to follow-up, with the exception of a small decrease in HR immediately post-exercise (−3.11 b.p.m.) in controls at Month 9. HR decline over the 5 min post-exercise followed a slightly lower course at the 1 month visit in control patients and at the 9 month visits in both groups, but not at the 1 month visit in the spironolactone group. Finally, compared with baseline, EE increased by two to three shuttles at Months 1 and 9 in the spironolactone group but remained unchanged in the control group. In unadjusted between-group analyses, there were no HR differences at any time point from rest up to 5 min post-exercise. In subgroups dichotomized by sex or the medians of age or eGFR, at baseline, HR followed a slightly higher course in the controls compared with patients assigned spironolactone. At the 9 month visit, this was still the case in women and participants aged <71 years, but the time point-by-subgroup interactions were not significant (P-values 0.40 and 0.059 for sex and age group, respectively).

Heart rate responses

Exercise increases sympathetic tone via circulating adrenaline and the neural release of noradrenaline. The initial HR decline within 30 s post-exercise is predominantly mediated by vagal reactivation, with sympathetic withdrawal playing a lesser role. However, starting from 2 min post-exercise, the decline in HR is mainly associated with sympathetic withdrawal.³⁰

The literature does not provide a consistent definition of impaired post-exercise HRR.³⁰ The causes of the inconsistency are the variability in exercise protocols, varying ways to characterize HRR, the widely diverse characteristics of the examined individuals and differences in endpoint definitions and follow-up duration. As reviewed elsewhere,³⁰ across studies, the definition of a deficient HRR ranged from 12 to 30 b.p.m. at the first minute post-exercise and from 22 to 42 b.p.m. at the second minute. In the current study, we applied the most commonly applied definition, that is, a difference between the post-exercise HR and the 1 min post-exercise HR of <12 b.p.m. HOMAGE included patients likely to have or have coronary heart disease as a cause of subsequent HF.²⁸ Given the limited number of primary endpoints in the whole trial (control vs. spironolactone: n = 11 vs. 9; P = 0.50)²⁸ over the 9 month follow-up, the prediction of adverse health outcomes was not within the scope of the current subgroup analysis. However, multiple studies,^32–35 albeit not all,³⁶ using classic treadmill exercise protocols,^32,35 treadmill exercise with nuclear myocardial perfusion imaging³⁶ or treadmill echocardiographic exercise^33,36 demonstrated the accuracy of an impaired HRR in the prediction of mortality^32,33,36 or in the association of HRR with coronary heart disease^34,35 or high-risk features on myocardial perfusion imaging³⁴ over and beyond other risk indicators.

Reproducibility or repeatability refers to the probability of getting the same results when a variable is measured under similar conditions by different methods, by different observers applying the same method or after a short interval that does not include biological or pharmacologically induced variability. In our current study, we did not assess reproducibility in the proper sense of the word, but rather within-group changes over time and the between-group differences in these changes. However, two studies described HRR reproducibility.^37,38 In a retrospective study of 90 patients undergoing treadmill exercise testing twice at an interval of 18 weeks or less, none of the abnormal HRR definitions provided more than 55% concordance between tests.³⁷ However, a second study applying the same treadmill protocol demonstrated that resting HR (ICC = 0.92), CHR (ICC = 0.88) and HRR measured from 1 to 5 min post-exercise are reproducible in healthy adults when tests are repeated after 1 week and 1 month.³⁸ Our 1 month observations are in line with the second study referred to above,³⁸ and the ICC at Month 9 still indicates moderate (ICC = 0.5–0.6) or strong (ICC = 0.7–0.8) reproducibility, thereby providing a surrogate validation of the ISWT in the absence of state-of-the-art validation by treadmill or bicycle exercise tests. Given that there were no or only minimal within- or between-group differences in CHR and HRR, our current study extends the above observations to 9 months in patients with underlying coronary heart disease who are therefore prone to HF.

Exercise endurance

An assessment of exercise capacity is widely used to grade the severity of chronic HF. Until recently, the 6MWT, a measure of submaximal EE, has most frequently been used for the assessment of interventions in HF patients,³⁹ but results have often been disappointing.⁴⁰ The ISWT is designed to provoke symptoms, such as breathlessness, and assess maximum symptom-limited exercise capacity.^7,41 In a systematic review including 13 studies in patients with chronic lung disease and 8 in patients with cardiac disease,⁴² the correlations between distance covered in the ISWT and peak oxygen consumption ranged from 0.67 to 0.95 (P < 0.01). The ICCs for test–retest reliability ranged from 0.76 to 0.99. Moreover, the ISWT was responsive to interventions including pulmonary rehabilitation and bronchodilator administration. The minimum clinically important difference in the distance covered in patients with lung disease was approximately 48 m.^42,43 For cardiac rehabilitation, the minimum clinically important difference was 70 m, but smaller estimates may apply for those with comorbid lung disease (39 m), obesity (29 m) or depression (52 m).^42,43

In the current study, the ISWT improved with spironolactone at 9 months (P = 0.030), but in between-group analyses, significance was lost, irrespective of adjustment. The ISWT was used as the primary endpoint in a single-centre trial including 76 men comparing testosterone substitution to placebo over 12 months. Testosterone improved ISWT by 25 m (P = 0.006) as well as symptoms.⁴⁴ EE did not improve with spironolactone compared with placebo in older people with reduced functional status.⁴⁵ The Aldosterone Receptor Blockade in Diastolic Heart Failure Trial (ALDO-DHF) also failed to show an improvement in bicycle exercise capacity with spironolactone compared with placebo in a population similar to HOMAGE.⁴⁶ However, another placebo-controlled trial suggested an improvement in treadmill exercise capacity with spironolactone.⁴⁷ By and large, the small effects of spironolactone on the ISWT responses in HOMAGE are fairly consistent with the majority of trials of mineralocorticoid receptor antagonists in patients with cardiovascular disease.

Study limitations

This study has several limitations. First, we excluded patients who did not complete the three ISWTs, that is, at baseline and at Months 1 and 9. The rationale for this decision was the primary focus of our analyses of the change over time in CHR, HRR and EE. Not having the same patients at each time interval during the course of the trial would have precluded direct comparisons between the short- and long-term within- and between-group changes in these variables. Second, approximately 70% of patients were on treatment with beta-blockers, which have a negative chronotropic effect on the resting HR and the HR responses to exercise.^48,49 Beta-blockade might have masked potential changes in CHR, HRR or EE produced by spironolactone. Third, women were underrepresented in our study, as in many other trials. Few patients had musculoskeletal dysfunction requiring the use of a walking aid during the ISWT. These observations limit the generalizability of the current findings. Fourth, the HOMAGE trial protocol did not include a comparison of the exercise responses to ISWT with state-of-the-art endurance tests by bicycle or treadmill exercise.⁵ Finally, given the relatively small sample size and the short follow-up, we could not relate the ISWT results to clinical events. However, as reviewed in detail above,^32–35 a multitude of studies showed that HRR and EE predict mortality and cardiovascular endpoints.

Conclusions

Spironolactone on top of usual treatment compared with usual treatment alone does not change resting HR, CHR, HRR or EE in response to the ISWT. Beta-blockade might have concealed the effects of spironolactone. However, the current findings demonstrate that the ISWT, already used in a variety of pathological conditions ranging from apparently healthy individuals^8,9 to patients with advanced pulmonary,^11–18 malignant,^19–21 cardiovascular^7,22–24 or renal²⁵ disease, is a practical instrument to measure symptom-limited exercise capacity in patients with comorbidities, such as coronary heart disease. In these patients, ISWT might replace the more complex state-of-the-art symptom-limited bicycle or treadmill exercise tests when such diagnostic resources are not readily available or when complex exercise testing is not indicated or difficult to repeat at short time intervals.

Acknowledgements

The authors are indebted to the many investigators who were involved in HOMAGE. Their names are listed in the supporting information. This article was submitted for publication on their behalf.

Funding

HOMAGE was funded by the European Union Seventh Framework Program. OMRON Healthcare, Co., Ltd., Kyoto, Japan, provided a non-binding grant to the Non-Profit Research Association Alliance for the Promotion of Preventive Medicine (APPREMED), Mechelen, Belgium.

Conflict of interest statement

None of the authors reported a conflict of interest.

References

von Haehling S, Arzt M, Doehner W, Edelmann F, Evertz R, Ebner N, et al. Improving exercise capacity and quality of life using non‐invasive heart failure treatments: Evidence from clinical trials. Eur J Heart Fail 2021;23:92‐113. doi:10.1002/ejhf.1838

Singh R, Aggarwal D, Dutta K, Jaggi S, Sodhi MK, Saini V. Assessment of the feasibility of 1‐min sit‐to‐stand test in evaluating functional exercise capacity in interstitial lung disease patients. J Exerc Rehabil. 2023;19:363‐369. doi:10.12965/jer.2346418.209

Nishiyama Y, Morita H, Harada H, Katoh A, Adachi H, Koga A, et al. Systolic blood pressure to exercise as predictor of mortality in patients with chronic heart failure. Int Heart J 2010;51:111‐115. doi:10.1536/ihj.51.111

Carneiro HA, Song RJ, Lee J, Schwartz B, Vasan RS, Xanthakis V. Association of blood pressure and heart rate responses to submaximal exercise with incident heart failure: The Framingham Heart Study. J Am Heart Assoc 2021;10: [eLocator: e019460]. doi:10.1161/JAHA.120.019460

Balady GJ, Arena R, Sietsema K, Myers J, Coke L, Fletcher GF, et al. Clinician's guide to cardiopulmonary exercise testing in adults: A scientific statement from the American Heart Association. Circulation 2010;122:191‐225. doi:10.1161/CIR.0b013e3181e52e69

Pulz C, Diniz RV, Alves ANF, Tebexreni AS, Carvalho AC, de Paola ÂAV, et al. Incremental shuttle and six‐minute walking tests in the assessment of the functional capacity in chronic heart failure. Can J Cardiol 2008;24:131‐135. doi:10.1016/s0828‐282x(08)70569‐5

Lim HJ, Jee SJ, Lee MM. Comparison of incremental shuttle walk test, 6‐minute walking test, and cardiopulmonary exercise stress test in patients with myocardial infarction. Med Sci Monit 2024;28: [eLocator: e938140]. doi:10.12659/MSM.938140

Girotra S, Kitzman DW, Kop WJ, Stein PK, Gottdiener JS, Mukamal KJ. Heart rate response to a timed walk and cardiovascular outcomes in older adults: The Cardiovascular Health Study. Cardiology 2012;122:69‐75. doi:10.1159/000338736

Georgiopoulou VV, Kalogeropoulos A, Chowdhury R, Binongo JNG, Bibbins‐Domingo K, Rodondi N, et al. Exercise capacity, heart failure risk, and mortality in older adults: The Health ABC Study. Am J Prev Med 2017;52:144‐153. doi:10.1016/j.amepre.2016.08.041

Oppewal A, Hilgenkamp TIM, van Wijck R, Evenhuis HM. Heart rate recovery after the 10‐m incremental shuttle walking test in older adults with intellectual disabilities. Res Dev Disabil 2014;35:696‐704. doi:10.1016/j.ridd.2013.12.006

Billings CG, Hurdman JA, Condliffe R, Elliot CA, Smith IA, Austin M, et al. Incremental shuttle walk test distance and autonomic dysfunction predict survival in pulmonary arterial hypertension. J Heart Lung Transplant 2017;36:871‐879. doi:10.1016/j.healun.2017.04.008

Marsico A, Dal Corso S, De Carvalho EF, Arakelian V, Phillips S, Stirbulov R, et al. A more effective alternative to the 6‐minute walk test for the assessment of functional capacity in patients with pulmonary hypertension. Eur J Phys Rehabil Med 2021;57:645‐652. doi:10.23736/S1973‐9087.21.06561‐8

Minai OA, Nguyen Q, Mummadi S, Walker E, McCarthy K, Dweik RA. Heart rate recovery is an important predictor of outcomes in patients with connective tissue‐associated pulmonary hypertension. Pulm Circ 2015;5:565‐576. doi:10.1086/682432

Chen SY, Huang CK, Wu CL, Peng HC, Yu CJ, Chien JY. Prognostic value of the post‐exercise heart rate recovery and BHDE‐index in chronic obstructive pulmonary disease. BMC Pulm Med 2023;23:263. doi:10.1186/s12890‐023‐02557‐7

Zhao D, Abbasi A, Casaburi R, Adami A, Tiller NB, Yuan W, et al. Identifying a heart rate recovery criterion after a 6‐minute walk test in COPD. Int J Chron Obstruct Pulmon Dis 2021;16:2545‐2560. doi:10.2147/COPD.S311572

Fermont JM, Masconi KL, Jensen MT, Ferrari R, Di Lorenzo VAP, Marott JM, et al. Biomarkers and clinical outcomes in COPD: A systematic review and meta‐analysis. Thorax 2024;74:439‐446. doi:10.1136/thoraxjnl‐2018‐211855

Imamuras S, Inagaki T, Abe M, Terada J, Kawasaki T, Nagashima K, et al. Impaired dynamic response of oxygen saturation during the 6‐min walk test is associated with mortality in chronic fibrosing interstitial pneumonia. Respir Care 2023;68:356‐365. doi:10.4187/respcare.10231

Swigris JJ, Swick J, Wamboldt FS, Sprunger D, du Bois R, Fischer A, et al. Heart rate recovery after 6‐min walk test predicts survival in patients with idiopathic pulmonary fibrosis. Chest 2009;136:841‐848. doi:10.1378/chest.09‐0211

Voorn MJJ, Franssen RFW, Verlinden JMWF, Bootsma GP, de Ruysscher K, Bongers BC, et al. Associations between pretreatment physical performance tests and treatment complications in patients with non‐small cell lung cancer: A systematic review. Crit Rev Oncol Hematol 2021;158: [eLocator: 103207]. doi:10.1016/j.critrevonc.2020.103207

Ha D, Choi H, Zell K, Raymond DP, Stephans K, Wang XF, et al. Association of impaired heart rate recovery with cardiopulmonary complications after lung cancer surgery. J Thorac Cardiovasc Surg 2015;149:1168‐1173. doi:10.1016/j.jtcvs.2014.11.037

Argillander TE, Heil TC, Meils RJF, van Duijvendijk P, Klaase JM, van Munster BC. Preoperative physical performance as predictor of postoperative outcomes in patients aged 65 and older scheduled for major abdominal cancer surgery. Eur J Surg Oncol 2022;48:570‐581. doi:10.1016/j.ejso.2021.09.019

Zweerink A, van der Lingen ALCJ, Handoko ML, van Rossum AC, Allaart CP. Chronotropic incompetence in chronic heart failure: A state‐of‐the‐art review. Circ Heart Fail 2018;11: [eLocator: e004969]. doi:10.1161/CIRCHEARTFAILURE.118.004969

Cahalin LP, Forman DE, Chase P, Guazzi M, Myers J, Bensimhon D, et al. The prognostic significance of heart rate recovery is not dependent upon maximal effort in patients with heart failure. Int J Cardiol 2013;168:1496‐1501. doi:10.1016/j.ijcard.2012.12.102

Ferreira JP, Metra M, Anker SD, Dickstein K, Lang CC, Ng L, et al. Clinical correlates and outcome associated with changes in 6‐minute walking distance in patients with heart failure: Findings from the BIOSTAT‐CHF study. Eur J Heart Fail 2019;21:218‐226. doi:10.1002/ejhf.1380

Watson EL, Major RW, Wilkinson TJ, Greening NJ, Gould DW, Barratt J, et al. The association of muscle size, strength and exercise capacity with all‐cause mortality in non‐dialysis‐dependent CKD patients. Clin Physiol Funct Imaging 2020;40:399‐406. doi:10.1111/cpf.12655

Johnson NP, Goldberger JJ. Prognostic value of late heart rate recovery after treadmill exercise. Am J Cardiol 2012;110:45‐49. doi:10.1016/j.amjcard.2012.02.046

Messinger‐Rapport B, Snader CEP, Blackstone EH, Yu D, Lauer MS. Value of exercise capacity and heart rate recovery in older people. J Am Geriatr Soc 2003;51:63‐68. doi:10.1034/j.1601‐5215.2002.51011.x

Cleland JGF, Ferreira JP, Mariottoni B, Pellicori P, Cuthbert J, Verdonschot JAJ, et al. The effect of spironolactone on cardiovascular function and markers of fibrosis in people at increased risk of developing heart failure: The heart ‘OMics’ in AGEing (HOMAGE) randomized clinical trial. Eur Heart J 2021;42:684‐696. doi:10.1093/eurheartj/ehaa758

Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF III, Feldman HI, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604‐612. doi:10.7326/0003‐4819‐150‐9‐200905050‐00006

Okutucu S, Ugur N, Aytemir K, Oto A. Heart rate recovery: A practical clinical indicator of abnormal cardiac autonomic function. Exper Rev Cardiovasc Ther 2011;9:1417‐1430. doi:10.1586/erc.11.149

Lu L, Shara N, North East SAS Users Group Conference (NESUG). Reliability analysis: Calculate and comparing intra‐class correlation coefficients (ICC) in SAS. 2007. https://www.lexjansen.com/nesug/nesug07/sa/sa13.pdf. Accessed 08 April 2024

Nishime EO, Cole CR, Blackstone EH, Pashkow FJ, Lauer MS. Heart rate recovery and treadmill exercise score as predictors of mortality in patients referred for exercise ECG. JAMA 2000;284:1392‐1398. doi:10.1001/jama.284.11.1392

Vivekananthan DP, Blackstone EH, Pothier CE, Lauer MS. Heart rate recovery after exercise is a predictor of mortality, independent of the angiographic severity of coronary disease. J Am Coll Cardiol 2003;42:831‐838. doi:10.1016/S0735‐1097(03)00833‐7

Gera N, Taillon LA, Ward RP. Usefulness of abnormal heart rate recovery on exercise stress testing predicts high‐risk findings on single‐photon emission computed tomography myocardial perfusion imaging in men. Am J Cardiol 2009;103:611‐614. doi:10.1016/j.amjcard.2008.11.004

Ghaffari S, Kazemi B, Aliakbarzadeh P. Abnormal heart rate recovery after exercise predicts coronary artery disease severity. Cardiol J 2011;18:47‐54.

Chen MS, Blackstone EH, Pothier CE, Lauer MS. Heart rate recovery and impact of myocardial revascularization on long‐term mortality. Circulation 2004;110:2851‐2857. doi:10.1161/01.CIR.0000147539.39775.F4

Yawn BP, Ammar KA, Thomas R, Wollan PC. Test retest reproducibility of heart rate recovery after treadmill exercise. Ann Fam Med 2003;1:236‐241. doi:10.1370/afm.37

Tulumen E, Khalilayeva I, Aytemir K, Kaya EB, Deveci OS, Aksoy H, et al. The reproducibility of heart rate recovery after treadmill exercise. Ann Noninvasive Electrocardiol 2021;16:365‐372. doi:10.1111/j.1542‐474X.2011.00464.x

Huang J, Yu Z, Wu Y, He X, Zhao J, He J, et al. Prognostic significance of blood pressure at rest and after performing the six‐minute walk test in patients with acute heart failure. Am J Hypertens 2023;37:199‐206. doi:10.1093//ajh/hpad115

Olsson LG, Swedberg K, Clark AL, Witte KK, Cleland JGF. Six minute corridor walk test as an outcome measure for the assessment of treatment in randomized, blinded intervention trials of chronic heart failure: A systematic review. Eur Heart J 2005;26:778‐793. doi:10.1093/eurheartj/ehi162

Morales FJ, Montemayor T, Martinez A. Shuttle versus six‐minute walk test in the prediction of outcome in chronic heart failure. Int J Cardiol 2000;76:101‐105. doi:10.1016/s0167‐5273(00)00393‐4

Parreira VF, Janaudis‐Ferreira T, Evans RA, Mathur S, Goldstein RS, Brooks D. Measurement properties of the incremental shuttle walk test: A systematic review. Chest 2014;145:1357‐1369. doi:10.1378/chest.13‐2071

Singh SJ, Jones PW, Evans R, Morgan MDL. Minimum clinically improvement for the incremental shuttle walking test. Thorax 2008;63:775‐777. doi:10.1136/thx.2007.081208

Malkin CJ, Pugh PJ, West JN, van Beek EJR, Jones TH, Channer KS. Testosterone therapy in men with moderate severity heart failure: A double‐blind randomized placebo controlled trial. Eur Heart J 2006;27:57‐64. doi:10.1093/eurheartj/ehi443

Burton LA, Sumukadas D, Witham MD, Struthers AD, McMurdo MET. Effect of spironolactone on physical performance in older people with self‐reported physical disability. Am J Med 2013;126:590‐597. doi:10.1016/j.amjmed.2012.11.032

Edelmann F, Wachter R, Schmidt AG, Kraigher‐Krainer E, Colantonio C, Kamke W, et al. Effect of spironolactone on diastolic function and exercise capacity in patients with heart failure with preserved ejection fraction. The Aldo‐HF randomized controlled clinical trial. JAMA 2013;309:781‐791. doi:10.1001/jama.2013.905

Kosmala W, Rojek M, Przewlocka‐Kosmala M, Wright L, Mysiak A, Marwick TH. Effect of aldosterone antagonism on exercise tolerance in heart failure with preserved ejection fraction. J Am Coll Cardiol 2016;68:1823‐1834. doi:10.1016/j.jacc.2016.07.763

Jackson G, Atkinson L, Oram S. Reassessment of failed beta‐blocker treatment in angina pectoris by peak‐exercise heart rate measurements. Br Med J 1975;3:616‐618. doi:10.1136/bmj.3.5984.616

Nyberg G. Effect of beta‐adrenoceptor blockers on heart rate and blood pressure in dynamic and isometric exercise. Drugs 1976;11:185‐195. doi:10.2165/00003495‐197600111‐00037