Introduction

There is increase evidence that abnormal heart rate recovery (HRR) after the 6-min walking test (6MWT) is associated with morbidity and mortality in various respiratory diseases [1,2,3]. However, the literature is scarce regarding the relationship between HRR after the 6MWT and prognosis in patients with chronic obstructive pulmonary disease (COPD) [4, 5]. In chronic respiratory entities, abnormal autonomic cardiac response could be consequence of a lower parasympathetic activation and/or increased sympathetic tone, being widely accepted that parasympathetic activations plays a protective role [6]. Consequently, it is possible that numerous factors present in COPD patients namely hypoxemia, dynamic hyperinflation, systemic inflammation, and medication could affect the autonomic cardiac response to exercise with adverse consequence in diseases prognosis [7].

The two main objectives of our study were to identify the determinants of heart rate recovery at 1 min (HRR1) after completion of a 6MWT and to establish whether abnormal HRR1 predicts acute exacerbations (AECOPD).

Methods

We performed a prospective multicentre study that included the follow sites: Discipline of Physiotherapy, Faculty of Health Sciences, University of Sydney; Fundación Neumológica Colombiana, Bogotá; ELEGI/Colt laboratory, University of Edinburgh; Centre for Inflammation Research, The Queen’s Medical Research Institute; 1st Department of Respiratory Medicine, National and Kapodistrian University of Athens; and Hospital Clinic in Barcelona. The protocol was accepted by the local ethics committees at each study site.

The main inclusion criterion were: (1) COPD patients with stable disease (2 months without exacerbations) and, (2) optimized medication according to GOLD guidelines [8]. Patients were excluded if they had an unstable cardiovascular disease, such as heart failure or coronary artery disease. Patients with pace maker, using medication that affect heart rate recovery (i.e., Β-Blockers, calcium antagonist, etc.), or cardiac arrhythmia (i.e., atrial fibrilation) were also excluded. Active participation in a Pulmonary Rehabilitation program during the last 12 months was another exclusion criterion.

Patients were followed up for 12 months after the execution of the 6MWT. An investigator recorded the patients’ vital status and the frequency of AECOPD (with or without hospital admission). AECOPD was defined according to the GOLD definition as acute events characterized by a worsening of the patient’s respiratory symptoms that is beyond normal day-to-day variations and leaded to a change in medication [8].

The 6MWT was performed according to American Thoracic Society [9]; however, the test was modified by recording heart rate at the end of the 6MWT and at 1 min after completion of the test with the patient seated.

Results are presented as the mean and standard deviation (SD) for normally distributed variables or median and percentile 25–75 (P25-75) for skewed numerical variables. Univariate and multivariate linear and logistic regression analyses were performed in order to evaluate the determinants of abnormal heart rate recovery and the factors associated with AECOPD, respectively. Receiver Operating Characteristic (ROC) analysis was performed for the evaluation of the performance of HRR1 in the prediction of AECOPD and in order to determine the best cut-off point for HRR1 to predict AECOPD [10]. Afterwards, Kaplan–Meier analysis was used to assess the differences in the times to first AECOPD during follow-up period among subjects according to dichotomous classification of HRR1 (above or below the best cut-off point). The log-rank test determined statistical significance. Calculations were done with SPSS/PC (version 22, SPSS Inc., Chicago, IL, USA). A p value of <0.05 was considered significant.

Results

A total of 101 COPD patients were evaluated. Subject characteristics are listed in Table 1. After univariate and multivariate lineal regression analyses, 6MWT and DLco were the only independent determinants of HRR1 after 6MWT (Table 2). In Table 3, univariate and multivariate regression analyses showed that HRR1 remained an independent predictor of the frequency for AECOPD over the follow-up period. In ROC analysis, HRR1 presented an area under the curve (AUC) of 0.703 (95% CI 0.604–0.801) for the prediction of AECOPD. An HRR1 equal or less than 14 beats appears as the best cut-off point to predict AECOPD. Kaplan–Meier curves evaluating the time to first AECOPD according to HRR1 values are presented in Fig. 1. Patients with low HRR1 presented increased risk of AECOPD at 12 months post-6MWT assessment compared to those patients with a high HRR1 response (p = 0.004, log-rank test). Accordingly, patients with low HRR1 showed a mean exacerbation of 1.5 (1.7) at 12 months. In contrast, patients with HRR > 14 beats had a 0.5 (1) exacerbations during the same follow-up period.

Table 1 Anthropometric and functional characteristics of the COPD patients
Full size table
Table 2 Univariate and multivariate linear regression analysis evaluating the determinants of HRR1
Full size table
Table 3 Univariate and multivariate binary logistic regression analysis evaluating the effect of HRR1 and significant confounders on AECOPD during 12-months follow-up
Full size table
Fig. 1
figure 1

The Kaplan–Meier analysis showed that HRR1 ≤ 14 beats is associated with significantly shorter time to first AECOPD than HRR1 > 14 (p = 0.004 by the log-rank test)

Full size image

Discussion

The current study demonstrates that HRR1 is a clinical biomarker with a significant predictive capacity for AECOPD. To our knowledge, this is the first study aimed at identifying whether the abnormal HR response after the 6MWT predicts main outcomes in COPD. We also found that the 6-min walking distance and the diffusion capacity for carbon monoxide are the main determinants of HRR1 [1, 7].

Previous studies examining the prognostic value of the HRR1 after the 6MWT in respiratory patients showed that a HRR1 below 13 or 16 beats was associated with poorer survival in patients with pulmonary fibrosis and pulmonary hypertension, respectively [1, 2]. Our study shows that patients with HRR1 greater than 14 beats had a very low likelihood of AECOPD over the follow-up period.

In regards to the autonomic abnormal cardiac response in COPD, more than ten years ago Laccase et al. [11] demonstrated an association between HRR1 after a maximal exercise test and mortality. However, HRR1 after the 6MWT in COPD as a prognostic factor has never been reported. This is a tangible clinical finding given that the 6MWT is a widely used, simple test to apply into the clinical scenario [12].

Although the relationship between autonomic cardiac dysfunction and COPD is not fully clarified, our findings provide evidence that chronic complex diseases, such as COPD, are associated with autonomic dysfunction and sympathetic overactivation [13]. Moreover, a recent published study indicates that heart rate variability (HRV) at rest (as expression of autonomic imbalance) during AECOPD might increase the risk of sudden death [14]. However, although the relationship between HRV and HRR1 has been partially explored in COPD [15], these markers of autonomic function could be potentially modulated as consequence of physical exercise training into pulmonary rehabilitation programs [16, 17].

In conclusion, HRR1 after the 6MWT could be a potential predictor of AECOPD in COPD patients. Research is needed to examine the physiological mechanisms linking the delayed HRR and frequency of AECOPD in COPD patients.