Introduction

Cardiopulmonary exercise testing (CPET) is the gold standard method for measuring exercise performance, usually reported as peak oxygen uptake (peak V̇O2). CPET is applied in different populations including healthy subjects, athletes, and patients with various pathological conditions such as heart failure (HF).1, 2 In all these settings, CPET provides relevant information on top of exercise performance and prognosis as regards cardiac, respiratory, and muscle function and limitations.3 However, CPET limited availability as well as the need for trained staff for test supervision and data interpretation make it not accessible as desirable in every setting. Therefore, in clinical practice and in research trials or in large cohort studies, it is common to assess exercise performance and prognosis by simpler tests such as the 6 min walking test (6MWT).

There are conflicting data regarding the extent to which 6MWT represents a metabolically maximal test and about the correlation between peak V̇O2 and the distance walked at 6MWT.4-7 Maximal CPET and 6MWT are in fact two different tests: CPET is a maximal exercise, usually performed with a progressive increase of workload (ramp protocol) aimed at achieving a maximal effort in 8–12 min,8 whereas 6MWT is a constant load test and it is usually considered a submaximal, and therefore believed safer, exercise. Specifically, it is commonly perceived that in more severe patients, a maximal test such as CPET carries a higher risk than a submaximal exercise (e.g. 6MWT). This is confirmed by the discrepancy between the safety measures normally required to perform the two tests (e.g. presence of trained personnel, defibrillator, electrocardiogram monitoring, and presence of a stretcher to handle emergencies).1, 9

The aim of this study is to compare CPET cardiorespiratory parameters with those collected with a portable metabolimeter during 6MWT in a large group of healthy subjects and patients with HF of different severity.

Materials and methods

Anonymized data and materials will be made publicly available at https://zenodo.org/.

One hundred and fifteen HF patients and 40 healthy volunteers participated in the study. Healthy subjects (age 18–80 years) were recruited through word of mouth among hospital employees and their relatives and friends. We excluded athletes or subjects engaged in an intense training programme. All underwent medical history collection and full clinical evaluation including electrocardiogram. None was on treatment with any drugs possibly affecting the cardiorespiratory system. HF patients were recruited at Heart Failure Units of Centro Cardiologico Monzino, IRCCS, and Istituti Clinici Scientifici Maugeri, IRCCS. In all study locations, subjects underwent the same exercise protocol and data analysis, for both CPET and 6MWT. Patients were clinically stable with no recent admissions for worsening HF. Inclusion criteria were as follows: age 18–80 years and New York Heart Association I–III. As part of our routine HF assessment, all patients underwent at least one previous CPET and 6MWT at our laboratory, which confirmed that patients were familiar with the procedures and setting.10, 11 Exclusion criteria were the use of long-term oxygen therapy, previous heart transplantation or left ventricular assist device, neuromuscular co-morbidities affecting the possibility to perform both exercise tests, and concomitant moderate or more severe chronic obstructive pulmonary disease.12 The presence of a permanent pacemaker, implantable cardioverter defibrillator, or cardiac resynchronization therapy was not exclusion criteria. However, we excluded pacemaker-dependent patients with device-induced heart rate (HR).

All patients were on optimal medical therapy with standard HF medications at the highest tolerated dose.

The protocol complies with the World Medical Association Declaration of Helsinki, and it was approved by the Ethics Committee of Centro Cardiologico Monzino, IRCCS, Milan (MEC08-3-032), and of Istituti Clinici Scientifici Maugeri, IRCCS (CE 2204). Informed consent was obtained from all subjects. Data collection was prospective.

All HF patients were evaluated by left ventricular ejection fraction (LVEF) (Simpson biplane method) by cardiac ultrasound13 and underwent N-terminal pro-brain natriuretic peptide (NT-proBNP) or brain natriuretic peptide (BNP) measurements. BNP values were converted in NT-proBNP equivalent using 6.25 as correction factor (n = 35).14

Cardiopulmonary exercise test

Cardiopulmonary exercise tests were usually performed in the early afternoon. All CPETs were performed by means of a stationary ergospirometer (Quark PFT, COSMED, Rome, Italy) using an electronically braked cycle ergometer. The progressively increasing workload exercise protocol (ramp) was set to achieve peak exercise in ~10 min.8 In the absence of clinical events, CPET was interrupted when the subjects stated that they had reached maximal effort. We performed a breath-by-breath analysis of expiratory gases and ventilation (V̇E). V̇E vs. carbon dioxide production (V̇E/V̇CO2) slope was calculated as the slope of the linear relationship between V̇E and V̇CO2 from 1 min after the beginning of the loaded exercise to the end of the isocapnic buffering period.3 The respiratory exchange ratio (RER) was measured as V̇CO2/V̇O2, and we use 1.05 as a cut-off value to define a maximal exercise.15 CPETs were conducted on a different day from 6MWT.

Six-minute walking test

The 6MWTs were performed between one and two working days from the CPET and at the same time of the day of CPET using a dedicated hospital corridor. The metabolic values during the 6MWT were collected and assessed using a wearable ergospirometer (K5, COSMED).16 As per standard procedure, the K5 ergospirometer was calibrated every day following factory instructions.17, 18 Breath-by-breath measurements of V̇O2, V̇E, and V̇CO2 were recorded while the subjects were performing exercises.16, 18 HR was monitored through an HR monitor (Polar T31, Polar Electro Oy, Kempele, Finland).

Moreover, all participants were asked to score the degree of fatigue at the beginning and at the end of the exercise using a modified Borg symptom score ranging from 0 (no symptoms) to 10 (worst symptoms) points.19

We performed a standard 6MWT in all participants collecting the usual parameters (total distance walked measured in metres, Borg scale, HR, and haemoglobin oxygen saturation (SpO2) at the beginning and at the end of the 6MWT)9 on top of cardiorespiratory parameters. We instructed subjects to walk at regular pace as far as they could from end to end during the test. Every 60 s, subjects were encouraged with a standard sentence also mentioning the elapsed time.9

Figure 1 shows a subject performing the 6MWT with the K5 equipment (upper panel) and an example of breath-by-breath data collected (V̇O2 and V̇E).

Example of a subject performing a 6 min walking test wearing a portable metabolimeter (K5, COSMED, Rome, Italy) (upper panel). In the lower panel are shown oxygen uptake (V̇O2) and ventilation (V̇E) traces of the same subject during a complete 6 min walking test.

K5 wearable ergospirometer has been extensively used and validated.17, 18

Statistical analysis

Cardiopulmonary exercise test data are reported as average over 20 s or slopes as appropriate.3 As proposed by Wasserman et al.,20 patients were divided into three groups according to peak V̇O2: <12, 12–16, and >16 mL/kg/min.

Oxygen uptake during 6MWT (6MWT-V̇O2) was calculated and expressed both as mL/kg/min and as a per cent of the peak V̇O2 obtained at CPET. 6MWT cardiorespiratory parameters are the average of the last 60 s of exercise.

Data were recorded breath by breath. To account for erratic breaths, we cleaned outliers as follows: data were removed if they deviated above the 75th percentile or below the 25th percentile more than two times the 25–75th percentile delta. The analysis was performed within each test; considering all 6MWTs, the breaths removed for V̇O2 were 1.90% of the recorded breaths. A similar percentage of breaths were removed for the other analysed variables.

Normally distributed data, expressed as mean ± standard deviation, were examined by Student's t-test to compare patients and controls. For non-normally distributed parameters, data are expressed as the median and inter-quartile range. Trends across severity groups were assessed by analysis of covariance.

The associations between 6MWT and CPET parameters were evaluated with linear regression.

Analyses were carried out with the SAS statistical package v. 9.4 (SAS Institute Inc., Cary, NC, USA), and all tests were two sided. P < 0.05 was considered statistically significant.

Patient and public involvement

Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

Results

A total number of 155 subjects were enrolled (66 ± 11 years; male 77%), of whom 40 were healthy (59 ± 8 years; male 67%, body mass index 25.1 ± 3.4 kg/m2) and 115 were HF patients (69 ± 10 years; male 80%, body mass index 26.2 ± 4.3 kg/m2; P < 0.01 for age and gender distribution vs. healthy subjects). One healthy subject and nine patients were active smokers.

Heart failure patients had an average LVEF of 34.6 ± 12.0% and a median NT-proBNP of 1994 pg/mL [733–5329]. Specifically, 27 patients (23%) had an LVEF > 40% (HF with preserved or middle range ejection fraction), and 88 (77%) had HF with reduced LVEF. Beta-blocker therapy was present in 105 patients (91%), angiotensin-converting enzyme inhibitors or angiotensin receptor blockers in 66 (57%), angiotensin receptor–neprilysin inhibitor in 40 (35%), mineralocorticoid antagonist in 69 (60%), diuretic in 104 (90%), anticoagulants in 36 (31%), antiplatelet agents in 29 (25%), and digitalis in 6 (5%). Healthy subjects were not taking any medication.

Both CPET and 6MWT were performed without untoward events in all cases. Table 1 shows cardiopulmonary variables at CPET and at 6MWT of HF and healthy subjects. HF patients were stratified by peak V̇O2: Group 1, <12 mL/kg/min (n = 45); Group 2, 12–16 mL/kg/min (n = 44); and Group 3, >16 mL/kg/min (n = 26). Groups characteristics were as follows: (i) Group 1, 71.5 ± 9.4 years, female gender 10 (22%), LVEF 34.5 ± 12.7%, NT-proBNP 3262 [1196–8799], and 91% of patients received beta-blockers; (ii) Group 2, 69.5 ± 7.4 years, female gender 8 (18%), LVEF 33.3 ± 12.1%, NT-proBNP 1668 [618–3821], and 91% of patients received beta-blockers; and (iii) Group 3, 62.2 ± 13 years, female gender 5 (19%), LVEF 35.6 ± 10.6%, NT-proBNP 1994 [729–5607], and 96% of patients received beta-blockers. According to exercise limitation severity, patients showed higher V̇E/V̇CO2 slope and lower peak workload and V̇O2/work values (Table 2).

Table 1. Metabolic data during cardiopulmonary exercise test and during 6 min walking test in healthy subjects and heart failure patients Healthy subjects (n = 40) Patients (n = 115) n Mean SD n Mean SD P Cardiopulmonary exercise test Peak V̇O2 (mL/min) 40 2047 581 115 1008 296 <0.001 Peak V̇O2 (mL/kg/min) 40 28.1 7.4 115 13.5 3.5 <0.001 V̇E/V̇CO2 slope 40 27.2 4.0 115 37.8 9.4 <0.001 Peak VE (L/min) 40 81.1 23.1 115 47.1 13.3 <0.001 Peak RER 40 1.18 0.10 115 1.09 0.12 <0.001 V̇O2/work 40 10.1 1.1 92 8.5 1.7 <0.001 Rest HR (b.p.m.) 35 74 11 113 64 9 <0.001 Peak HR (b.p.m.) 40 154 16 116 100 24 <0.001 Peak work (W) 40 172 53 115 78 26 <0.001 6 min walking test Basal V̇O2 6MWT (L/min) 40 428 97 115 508 140 0.001 V̇O2 6MWT (L/min) 40 1410 317 115 959 270 <0.001 V̇O2 6MWT (mL/kg/min) 40 19.4 3.9 115 12.8 3.2 <0.001 V̇O2 6MWT (% peak V̇O2) 40 72% 20% 114 98% 20% <0.001 V̇E 6MWT (L/min) 40 36.1 8.8 115 33.1 9.4 0.079 VT 6MWT (L) 40 1.5 0.4 114 1.2 0.3 <0.001 V̇E/V̇CO2 6MWT 40 32.0 3.1 115 45.0 7.8 <0.001 Basal HR 6MWH (b.p.m.) 40 80.9 15.1 114 69.9 10.7 <0.001 HR 6MWT (b.p.m.) 40 108.3 19.0 115 87.2 16.7 <0.001 PetO2 6MWT (mmHg) 40 104.9 3.6 115 111.1 5.4 <0.001 PetCO2 6MWT(mmHg) 40 39.0 3.0 114 30.6 4.4 <0.001 V̇CO2 6MWT (mL/min) 40 1140 262 115 753 214 <0.001 RER 6MWT 40 1.03 0.06 115 1.07 0.14 0.014 Distance 6MWT (m) 40 498 55 114 390 90 <0.001 SpO2 basal 6MWT (%) 40 97.9 0.9 113 97.1 1.5 0.001 SpO2 stop 6MWT (%) 40 97.1 1.4 113 96.1 2.7 0.025 Borg scale 40 2 [1–3.25] 115 3 [1–4.8] 0.094 6MWT, 6 min walking test; HR, heart rate; Peak, peak exercise at cardiopulmonary exercise test; PetCO2, end-tidal carbon dioxide pressure; PetO2, end-tidal oxygen pressure; RER, respiratory gas exchange; SpO2, haemoglobin oxygen saturation; V̇CO2, expired CO2 volume; V̇E, ventilation; V̇O2, oxygen uptake; VT, tidal volume. Table 2. Metabolic data during cardiopulmonary exercise test and during 6 min walking test in the three groups of heart failure patients Group 1 V̇O2 < 12 Group 2 V̇O2 12–16 Group 3 V̇O2 > 16 Bonferroni n Mean SD n Mean SD n Mean SD P for trend ANOVA g1 vs. g2 g2 vs. g3 g1 vs. g3 Cardiopulmonary exercise test Peak V̇O2 (mL/min) 45 776 160 44 1030 201 26 1367 230 <0.001 <0.001 <0.001 <0.001 <0.001 Peak V̇O2 (mL/kg/min) 45 10.44 1.11 44 13.54 0.96 26 18.80 2.68 <0.001 <0.001 <0.001 <0.001 <0.001 V̇E/V̇CO2 slope 44 40.9 10.4 44 38.1 8.7 26 31.6 5.3 <0.001 <0.001 0.410 0.010 <0.001 Peak V̇E (L/min) 45 39.5 9.6 44 47.4 12.2 26 59.0 11.6 <0.001 <0.001 0.003 <0.001 <0.001 Peak RER 45 1.09 0.14 44 1.06 0.08 26 1.11 0.11 0.732 0.224 V̇O2/work 30 7.64 1.71 44 8.33 1.62 24 9.59 0.86 <0.001 <0.001 0.190 0.005 <0.001 Rest HR (b.p.m.) 44 64 8 44 65 9 24 64 11 0.623 0.706 Peak HR (b.p.m.) 45 92 21 44 100 23 26 111 21 0.005 0.002 0.255 0.121 0.002 Peak work (W) 45 60 19 44 79 18 26 105 22 <0.001 0.00 <0.001 <0.001 <0.001 6 min walking test Rest V̇O2 6MWT (L/min) 45 519 128 44 513 139 26 482 161 0.320 0.545 V̇O2 6MWT (L/min) 45 835 225 44 975 267 26 1147 237 <0.001 <0.001 0.024 0.016 <0.001 V̇O2 6MWT (mL/kg/min) 45 11.3 2.5 44 12.7 3.1 26 15.6 2.4 <0.001 <0.001 0.043 <0.001 <0.001 V̇O2 6MWT (% peak V̇O2) 45 109% 0.3 43 94% 0.2 26 84% 0.1 <0.001 <0.001 0.004 0.213 <0.001 V̇E 6MWT (L) 45 32.5 9.9 44 33.0 9.8 26 34.3 7.9 0.439 0.728 VT 6MWT 44 1.1 0.3 44 1.2 0.3 26 1.2 0.3 0.321 0.357 V̇E/V̇CO2 6MWT 45 49.6 6.8 44 43.9 6.9 26 38.8 5.9 <0.001 <0.001 <0.001 0.007 <0.001 Rest HR 6MWH (b.p.m.) 44 71 10 44 70 11 26 69 12 0.598 0.864