Exercise Training in Patients with Chronic Respiratory Diseases: Are Cardiovascular Comorbidities and Outcomes Taken into Account?—A Systematic Review

Patients with chronic obstructive pulmonary disease (COPD), asthma and interstitial lung diseases (ILD) frequently suffer from cardiovascular comorbidities (CVC). Exercise training is a cornerstone intervention for the management of these conditions, however recommendations on tailoring programmes to patients suffering from respiratory diseases and CVC are scarce. This systematic review aimed to identify the eligibility criteria used to select patients with COPD, asthma or ILD and CVC to exercise programmes; assess the impact of exercise on cardiovascular outcomes; and identify how exercise programmes were tailored to CVC. PubMed, Scopus, Web of Science and Cochrane were searched. Three reviewers extracted the data and two reviewers independently assessed the quality of studies with the Quality Assessment Tool for Quantitative Studies. MetaXL 5.3 was used to calculate the individual and pooled effect sizes (ES). Most studies (58.9%) excluded patients with both stable and unstable CVC. In total, 26/42 studies reported cardiovascular outcomes. Resting heart rate was the most reported outcome measure (n = 13) and a small statistically significant effect (ES = −0.23) of exercise training on resting heart rate of patients with COPD was found. No specific adjustments to exercise prescription were described. Few studies have included patients with CVC. There was a lack of tailoring of exercise programmes and limited effects were found. Future studies should explore the effect of tailored exercise programmes on relevant outcome measures in respiratory patients with CVC.


Background
Chronic obstructive pulmonary disease (COPD), asthma and interstitial lung diseases (ILD) are among the most representative chronic respiratory diseases in the world [1,2]. These diseases affect over 1 billion people and have a significant impact on patients' disability and quality of life (9.5% of the disability-adjusted life years in 2010 [3]), being a leading contributor to disease burden and one of the top causes of death worldwide (over 3 million deaths in 2016) [2][3][4].
Exercise training is a cornerstone intervention in both pulmonary and cardiac rehabilitation [21,22]. It relieves symptoms and improves functionality, exercise tolerance and health-related quality of life in patients with chronic respiratory and cardiovascular diseases [21,22], and therefore might be a promising intervention for the management of patients with these co-occurring conditions. Nevertheless, studies have shown that these effects are usually reduced in patients suffering from chronic respiratory diseases with accompanying cardiovascular comorbidities compared to those without cardiovascular comorbidities [7,19,23]. Moreover, recommendations on how to adjust exercise programmes to co-existing cardiovascular conditions in COPD, asthma and ILD are scarce.
Therefore, in order to inform evidence-based statements, this systematic review aimed to: (i) identify the eligibility criteria in terms of cardiovascular disease that have been used to refer patients with COPD, asthma and ILD for studies investigating the effectiveness of exercise programmes of at least 3 months; (ii) assess the impact of at least 3 months of exercise training on cardiovascular outcomes in these patients; and (iii) identify how the exercise programmes have been tailored to patients' cardiovascular comorbidities.

Search Strategy
This systematic review was reported according to the Preferred Reporting Items for Systematic reviews and Meta-analyses (PRISMA) guidelines [24] and was conducted in two phases. Phase 1 identified the eligibility criteria that have been used to select patients with cardiovascular comorbidities in clinical trials investigating the effectiveness of exercise programmes. Phase 2 assessed the impact of exercise training on cardiovascular outcomes, and identified how the exercise programmes have been tailored to patients' cardiovascular comorbidities.
A systematic literature search was performed in May 2019 on the following electronic databases: PubMed, Scopus, Web of Science and Cochrane. The search terms were limited to titles, abstracts and keywords/MeSH terms. The full search strategy is presented in Appendix A.

Eligibility Criteria and Study Selection
For phase 1, studies were included if they (i) studied adult patients with stable COPD, asthma and/or ILD (i.e., 4 weeks without exacerbations); (ii) implemented at least 12 weeks of exercise training (i.e., endurance and/or strength training) as an intervention [25]; (iii) implemented at least 2 directly supervised exercise sessions per week [26]; (iv) were original prospective quantitative studies; and (v) were written in Portuguese, English, French, Dutch or Spanish languages. Retrospective studies, case studies, case series, abstracts and studies involving alternative modalities of exercise (e.g., yoga, tai chi, qigong) were excluded. After removing duplicates, three reviewers (AM, KQ and AO) assessed all the potential studies identified. Studies were selected based on their titles and abstracts. When the title and abstract were potentially relevant to the purpose of the review, the full text was read carefully to decide on its inclusion. A fourth reviewer (CB) was consulted to solve any disagreements.
For phase 2, studies included in phase 1 that specified the prevalence of cardiovascular comorbidities (i.e., any cardiovascular condition co-existing with the respiratory disease, identified by doing an objective patients' assessment, checking their medical records or ask patients to self-report their comorbidities) in the baseline characteristics of the population under study and/or reported at least one cardiovascular outcome (i.e., heart rate, systolic and diastolic blood pressure, flow-mediated dilation, pulse-wave velocity, intima thickness of arteria carotid, cardiac function and structure, heart rate variability, ECG analysis and blood lipid profile) were included.

Quality Assessment and Data Extraction
Two reviewers (KQ and AO) independently assessed the quality of the studies included in phase 2 with the Quality Assessment Tool for Quantitative Studies, developed by the Effective Public Health Practice Project, Canada [27]. This tool assesses six domains of methodological quality: (i) selection bias; (ii) study design; (iii) confounders; (iv) blinding; (v) data collection methods; and (vi) withdrawals and dropouts [27]. Each domain is rated as "strong", "moderate" or "weak", according to a standardized guide, and the overall rating of the study is determined based on the total number of "strong" and "weak" scores [27].
In phase 1, data regarding the eligibility criteria (i.e., inclusion and exclusion criteria) used to select patients for the study were extracted from all included studies. Afterwards, all conditions that would preclude patients' participation in the exercise programmes, reported either as reasons for inclusion (e.g., absence of severe cardiovascular disease) or exclusion (e.g., presence of severe cardiovascular disease) of these patients, were compiled and reported as exclusion criteria. Additionally, data from the studies included in phase 2 were extracted in a predesigned structured table format comprising the following topics: study (first author, year of publication, country); study design; population (number of participants, diagnosis, age, gender, forced expiratory volume in 1 s (FEV 1 ), forced vital capacity (FVC), diffusing capacity for carbon monoxide (DLCO)); intervention (type and intensity of intervention); duration and frequency (duration of the intervention, duration and frequency of sessions); outcome and outcome measure; and results. For the scope of this review, only cardiovascular outcomes and outcome measures were considered.

Data Analysis and Synthesis
Inter-rater agreement analysis using Cohen's kappa was used to explore the consistency of the quality assessment performed by the two reviewers. The value of Cohen's kappa ranges from 0 to 1 and can be interpreted as slight (≤0.2), fair (0.21-0.4), moderate (0.41-0.6), substantial (0.61-0.8), or almost perfect (≥0.81) agreement [28]. The statistical analysis was performed using IBM SPSS 24.0 (IBM, Armonk, New York, NY, USA).
Whenever possible, effect sizes (ES) were calculated and a meta-analysis was performed. ES were calculated as Cohens' d based on the Pre/Post means and standard deviations or mean differences and standard deviations, according to the formula of Morris [29], and interpreted as small (≥0.2), medium (≥0.5) or large (≥0.8) [30]. Meta-analysis was performed on MetaXL 5.3. Pooled effect estimates were calculated with the inverse variance technique assuming a fixed-effects model. The input data were the Cohen's d value of each study and the respective standard error. The output was the pooled Cohen's d value and corresponding confidence intervals. Homogeneity among the studies was evaluated using Cochran's Q test and the I 2 statistic.

Study Selection
The literature search provided a total of 50.970 records. After duplicates removal, 29.756 records were screened for relevant content through title and abstract and 29.248 were excluded. The full text of 508 potentially relevant articles was assessed. From these, 180 articles were included in phase 1 and 42 in phase 2 ( Figure 1).

Quality Assessment
Results of the methodological quality assessment are presented in Table 1. Most of the studies (n = 24; 57.1%) were of weak quality. The agreement between the two reviewers was substantial (k = 0.72; 95%CI = 0.53-0.91; p < 0.001; percentage of agreement = 85.7%).
In total, 1704 patients (65.2% male; data gathered from 34 studies) with a weighted mean age of 65.4 years old and a mean FEV 1 of 53.7% of predicted (data gathered from 36 studies) were enrolled in the included studies.

Results on Cardiovascular Outcomes
Mean systolic blood pressure at daytime

Results on Cardiovascular Outcomes
Mean diastolic blood pressure at nighttime           Data are presented as mean ± standard deviation or median (interquartile range), unless otherwise stated. Legend: 6MWT, 6-min walk test; 12MWT, 12-min walk test; 1RM, one repetition maximum; 10RM, ten repetition maximum; 15RM, fifteen repetition maximum; 95%CI, 95% confidence interval; %pred, percentage predicted; COPD, chronic obstructive pulmonary disease; CPET, cardiopulmonary exercise test; DLCO, diffusing capacity for carbon monoxide; ES, effect size; FEV 1 , forced expiratory volume in 1 s; FVC, forced vital capacity; HDL, high density lipoprotein; HR, heart rate; HR max , maximum heart rate; ILD, interstitial lung disease; IPF, idiopathic pulmonary fibrosis; LDL, low density lipoprotein; mBorg, modified Borg scale; NMES, neuromuscular electrical stimulation; VO 2 max, maximal oxygen uptake; VO 2 peak, peak oxygen uptake; WR, work rate; WR max , maximal work rate; WR peak , peak work rate. Table 3. Characteristics of the studies in patients with asthma included in phase 2 (i.e., studies that specified the prevalence of cardiovascular comorbidities in the baseline characteristics of the population under study and/or reported at least one cardiovascular outcome) (n = 2). Data are presented as mean ± standard deviation, unless otherwise stated. Legend: %pred, percentage predicted; ES, effect size; FEV 1 , forced expiratory volume in 1 s; HDL, high density lipoprotein; HR, heart rate; HR max , maximum heart rate; LDL, low density lipoprotein. Table 4. Characteristics of the studies in patients with ILD included in phase 2 (i.e., studies that specified the prevalence of cardiovascular comorbidities in the baseline characteristics of the population under study and/or reported at least one cardiovascular outcome) (n = 8).   were the outcome measures presenting the larger effects. In patients with COPD, the effects of exercise training programmes on resting heart rate resulted in an overall pooled ES of −0.23 (95% confidence interval −0.33 to −0.13) ( Figure 5). Regarding the exercise programmes, most studies conducted in patients with COPD performed a pulmonary rehabilitation programme [33,54,55,57,59,68,80,90,101,104,106,107,121,123,136,167] (n = 16) or an exercise programme combining aerobic and strength training [38,[43][44][45]54,86,129,130] (n = 8). Sessions were conducted 2-6 times per week and each session lasted from 15 min to 2 h. Programme duration varied between 12 weeks and 18 months. A wide range of intensities was used to prescribe the exercise: 60%-80% of the maximum heart rate, 50%-100% of the peak or maximum oxygen uptake, 50%-125% of the peak or maximum workload, 35%-75% of one-repetition maximum, dyspnoea and perceived exertion levels between 3-6 on the modified Borg scale and 12-16 on the Borg scale. None of the studies specified any adjustments to tailor the exercise programmes to patients' cardiovascular comorbidities. Only one study [108] described adjusting the training programme in different mesocycles in order to improve specific cardiovascular outcomes.

Study and Country
Studies conducted in patients with asthma performed either an exercise programme combining aerobic and strength training for 3 months [189] or aerobic training for 6 months [186]. Sessions occurred 3 times/week, for 30 min each, at an intensity of 60%-80% of the maximum heart rate. No specific adjustments to improve specific cardiovascular outcomes were reported.

Discussion
To the best of the authors' knowledge, this is the first comprehensive overview of the scientific literature summarizing (i) the eligibility criteria in terms of cardiovascular disease used to select patients with chronic respiratory disease to exercise training studies, (ii) the impact of at least 3 months of exercise training on cardiovascular outcomes, and (iii) adjustments made to tailor exercise training prescription to patients with cardiovascular comorbidities. It was found that (i) in the majority of the studies (58.9%) patients with cardiovascular comorbidities were excluded a priori, (ii) there is limited evidence about the impact of exercise training on cardiovascular outcomes in patients with chronic respiratory diseases, and (iii) none of the studies explicitly mentioned how to tailor exercise training modalities in light of cardiovascular comorbidities.
A large diversity was found regarding the cardiovascular conditions that are used as exclusion criteria in exercise-related research. Interestingly, the majority of the exclusion criteria reported (34/45) are not considered contraindications to exercise training. Indeed, just a minority of the studies (18.3%) excluded only patients with acute/unstable cardiovascular disease that contraindicated participation in exercise training. Most studies excluded patients with both stable and unstable cardiovascular comorbidities, although at least 20%-50% of the patients with COPD, asthma or ILD present cardiovascular comorbidities [5,7,18,20,210,211]. Thus, by excluding patients with cardiovascular comorbidities or any other comorbidity that does not present any contraindication to perform exercise training, translation of knowledge to clinical practice can only be done for a subset, or sometimes even a minority, of patients. This finding might have a far-reaching consequence, namely that current knowledge (including clinical guidelines) is disease-centred and, thus, inadequate to sufficiently support/guide clinicians on how to prescribe exercise for patients with chronic respiratory diseases and multiple chronic conditions [13,212]. Furthermore, some of the criteria reported (e.g., cardiovascular disease) were too vague to allow understanding of which conditions were really excluded and over 20% of the included studies did not report any information concerning to eligibility criteria, even though this is key information to ensure clarity and transparency of the research [213].
Exercise training programmes in patients with cardiovascular comorbidities resulted in significant improvements in general reported outcomes, namely symptoms, functionality, exercise capacity, muscle strength and health-related quality of life, comparable to the ones usually found in respiratory patients [22]. However, regarding cardiovascular outcomes, in the majority of the studies (71.4%) only small to moderate effects were found, with the larger effects being reported for heart rate variability measurements (ES = [−0.78; 2.64]) and blood lipid profile (ES = [−2.31; 0.62]). Additionally, a small but significant overall effect of exercise training programmes on resting heart rate of patients with COPD was found. These results are yet not inferior to the ones previously reported for patients with cardiovascular diseases, in whom beneficial effects of exercise training have been found for heart rate variability and heart rate recovery [214], and inconsistent but significant and modest effects have been reported for arterial blood pressure and blood lipid profile [214][215][216][217]. We hypothesized that several reasons might be contributing to the limited effects found. First, most studies have not reported any specific adjustments in the exercise prescription to tailor the programme to patients' cardiovascular comorbidities, although it is plausible that they have made some adjustments without specific reporting in the published paper. It is known that cardiovascular conditions require specific considerations when formulating the exercise plan [22], and different recommendations exist based on the prevalent cardiovascular disease (e.g., coronary artery disease, congestive heart failure, peripheral arterial disease, pulmonary arterial hypertension) and its severity [218]. Indeed, it is mandatory to tailor exercise duration, frequency, mode, intensity and monitoring to patients' specificities and needs, clinical conditions, cardiovascular phenotype (risk factors and diseases), fitness level, medication intake (beta blockers, statins, glinides, sulfonylurea), abnormal responses to exercise (myocardial ischemia, atrial fibrillation, ventricular tachycardia) and rehabilitation goals [1,[218][219][220][221]. Moreover, the impact of exercise training relies on this proper tailoring of the exercise programme, since it has been shown that different exercise prescriptions result in significant differences in clinical outcomes [221]. Future studies should therefore assess the impact of exercise programmes specifically tailored to patients with co-occurring respiratory disease and cardiovascular comorbidities [7] and report the intervention in detail. Second, guidelines for cardiac rehabilitation from the leading scientific societies recommend that exercise should progress from moderate to vigorous intensity, three times per week [21]. Nevertheless, in some of the included studies, patients exercised at lower intensities and/or fewer times per week, which might have also contributed to the relative lack of effects since the minimum dose of exercise for cardiovascular benefits (>150 min/week of endurance training, energy expenditure 1000-2000 kcal/week) might have not been reached [218,222]. Third, some of the included studies only used strength training in their exercise programmes. Indeed, strength training has been recommended in patients with cardiovascular diseases, but as an adjunct to aerobic training, the last being a core component in these patients' rehabilitation [21]. From these observations, it became clear that current exercise prescription to patients with COPD, asthma or ILD with cardiovascular comorbidities is far from optimal and deserves significant reconsideration. Nonetheless, digital support on how to prescribe exercise in these patients in accordance to all the different clinical guidelines for different cardiovascular diseases is available, and thus could be used to support health professionals [218]. Lastly, most studies including patients with cardiovascular comorbidities only focused on the assessment of resting heart rate. Although this is a relevant outcome measure and results from meta-analysis in patients with COPD favour intervention, recommendations for patients with cardiovascular diseases advocate a more comprehensive assessment, including outcomes such as arterial blood pressure, blood lipid profile or echocardiography, that are also more in line with the aims of rehabilitation in these patients [223,224]. Therefore, outcomes should be better targeted to patients' cardiovascular comorbidities [19].
Besides the known prevalence and increased risk of morbidity and mortality that cardiovascular comorbidities impose on patients with chronic respiratory diseases [7,211], only three studies [196,207,209] (all conducted in the last 5 years) included patients with cardiovascular comorbidities and assessed cardiovascular outcome measures. This denotes the current gap in the literature regarding exercise programmes and emphasises the need for specific studies focusing on cardiovascular outcomes in these patients.
This systematic review has a number of limitations that need to be acknowledged. First, as it was anticipated that a large number of studies would be found, only exercise programmes lasting at least 12 weeks were included, which might have led to the loss of other relevant studies. Nevertheless, 12 weeks has been recommended as the minimum exercise duration required to reach benefits in patients with cardiovascular disease [25]. Second, as only few studies including patients with ILD were found, all types of ILD were grouped, although different types of ILD present different characteristics and possibly different cardiovascular comorbidities and responses to exercise training programmes. Third, most of the included studies were of weak quality. Nonetheless, since in exercise interventions blinding of participants is impossible and patients are usually referred by physicians to ensure their safety, it was virtually impossible to ensure strong quality in the quality assessment tool used.

Conclusions
Although a large number of studies explored the effects of at least 3 months of exercise training in patients with chronic respiratory diseases, only few included patients with cardiovascular comorbidities. Limited effects of the exercise programmes were found on cardiovascular outcome measures, possibly due to the lack of tailoring of the exercise training prescription and comprehensiveness of the cardiovascular outcome measures. Future studies focusing on patients with combined respiratory and cardiovascular diseases and exploring the effects of exercise programmes specifically tailored to these patients are needed to bridge the gap in the literature. Funding: This research received no external funding.