Mechanisms Underlying the Association of Chronic Obstructive Pulmonary Disease With Heart Failure

Objectives The purposes of this study were to determine why chronic obstructive pulmonary disease (COPD) is associated with heart failure (HF). Specific objectives included whether COPD is associated with myocardial fibrosis, whether myocardial fibrosis is associated with hospitalization for HF and death in COPD, and whether COPD and smoking are associated with myocardial inflammation. Background COPD is associated with HF independent of shared risk factors. The underlying pathophysiological mechanism is unknown. Methods A prospective, multicenter, longitudinal cohort study of 572 patients undergoing cardiac magnetic resonance (CMR), including 450 patients with COPD and 122 age- and sex-matched patients with a median: 726 days (interquartile range: 492 to 1,160 days) follow-up. Multivariate analysis was used to examine the relationship between COPD and myocardial fibrosis, measured using cardiac magnetic resonance (CMR). Cox regression analysis was used to examine the relationship between myocardial fibrosis and outcomes; the primary endpoint was composite of hospitalizations for HF or all-cause mortality; secondary endpoints included hospitalizations for HF and all-cause mortality. Fifteen patients with COPD, 15 current smokers, and 15 healthy volunteers underwent evaluation for myocardial inflammation, including ultrasmall superparamagnetic particles of iron oxide CMR. Results COPD was independently associated with myocardial fibrosis (p < 0.001). Myocardial fibrosis was independently associated with the primary outcome (hazard ratio [HR]: 1.14; 95% confidence interval [CI]: 1.08 to 1.20; p < 0.001), hospitalization for HF (HR: 1.25 [95% CI: 1.14 to 1.36]); p < 0.001), and all-cause mortality. Myocardial fibrosis was associated with outcome measurements more strongly than any other variable. Acute and stable COPD were associated with myocardial inflammation. Conclusions The associations between COPD, myocardial inflammation and myocardial fibrosis, and the independent prognostic value of myocardial fibrosis elucidate a potential pathophysiological link between COPD and HF.

C hronic obstructive pulmonary disease (COPD) is associated with left ventricular heart failure (HF), independent of shared risk factors, and HF independently contributes to mortality in patients with COPD (1).
The pathophysiological mechanisms underlying the association remain unclear.
Myocardial fibrosis, driven by myocardial inflammation occurring as part of systemic inflammation, is a widely held hypothesis, but evidence for this is lacking, and other mechanisms have been proposed (2)(3)(4).
Elucidating the underlying pathophysiology would enable improved risk stratification and development and evaluation of therapies targeting the pathophysiology.
This multicenter study aimed to determine why COPD is associated with HF. Specifically, it aimed to answer the following questions. 1) Is COPD associated with myocardial fibrosis? 2) Is myocardial fibrosis associated with hospitalization for HF and death in patients with COPD? 3) Are COPD and smoking associated with myocardial inflammation?

METHODS
The study was approved by an ethics committee at each site, and all participants provided written informed consent.   Values are median (interquartile range) or mean AE SD depending on distribution. a n ¼ 435; b n ¼ 425; c n ¼ 393 (it was not possible to calculate ECV in 57 patients because the native or post-contrast T1 maps were not acquired or because same-day hematocrit was not available); d n ¼ 449; e n ¼ 348; f n ¼ 100.
BSA ¼ body surface area; CMR ¼ cardiac magnetic resonance; COPD ¼ chronic obstructive pulmonary disease; ECV ¼ extracellular volume; EDV ¼ end diastolic volume; EF ¼ ejection fraction; eGFR ¼ estimated glomerular  Values are mean AE SD depending on distribution. a n ¼ 514, per 1 ml/min per 1.73 m 2 increase. b n ¼ 515 consisting of 393 patients with COPD and 122 non-COPD patients, as shown in Table 1.
CMR ¼ cardiac magnetic resonance; other abbreviations as in Table 1.   Table 6). Similarly, myocardial fibrosis remained independently associated with the composite outcome when RV ejection fraction was included in the model (Supplemental Table 7).   Table 8).
Acute COPD was associated with elevated C-reactive protein and white cell counts, which were lower in patients with stable COPD but did not return to normal (Supplemental Table 9). Two patients demonstrated evidence of chronic myocardial    There were no significant differences between acute and stable COPD. Indicators of myocardial edema were inconsistent; there were no differences in myocardial T2 relaxation time between acute and stable COPD, although myocardial T1 was higher during the acute than during the stable period.
Myocardial capillary permeability was numerically higher during acute COPD, but the differences were not statistically significant. with COPD ( Figure 2). Myocardial R2* behavior over time (p ¼ 0.015) (Supplemental Table 10) was significantly higher in smokers than in healthy volunteers but differences in R2*:R1 ratios were not statistically significant (Supplemental Table 9, Figure 2). There was no evidence of myocardial edema or increased capillary permeability.
There were significant relationships between the severity of myocardial inflammation and biochemical The reported prevalence of HF in patients with COPD ranges between approximately 7% and 31%, but the broad "type" of HF with which COPD associates (i.e., reduced vs. preserved ejection fraction [EF]) has been poorly characterized (11). In the current study, COPD was associated with lower LVEF than matched patients without COPD, but in most of the patients, LVEF was within the accepted normal range or at most mildly reduced, and LVEF was not independently associated with hospitalization for HF or death, thus suggesting that the predominant type of HF associated with COPD is HF with preserved EF.
Although large epidemiological studies have demonstrated that the association between COPD and HF is independent of common risk factors such as ischemic heart disease, diabetes, and hypertension, the mechanism responsible for the association has  (13,20,21   USPIO are phagocytosed by active cardiac macrophages, which are widely considered to be part of tissue inflammatory response, but they are not specific to the initiating injury. Thus, this study is unable to determine the cause of the inflammatory response.

CONCLUSIONS
The associations among COPD, myocardial inflammation and myocardial fibrosis, and the independent prognostic value of myocardial fibrosis elucidate a potential pathophysiological link between COPD and HF. The findings have the potential to improve patient risk stratification and to be the basis for developing new therapeutic strategies for patients with COPD.
ACKNOWLEDGMENTS The authors recognize the support received from AMAG Pharmaceuticals, which provided the USPIO (ferumoxytol) as part of a research agreement. The authors also thank Siemens for access to Work in Progress sequences. The sponsors, AMAG Pharmaceuticals, Siemens, and the funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation or approval of the manuscript; and decision to submit the manuscript for publication.

FUNDING SUPPORT AND AUTHOR DISCLOSURES
The study was supported by a research grant from Guerbet. Guerbet had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation or approval of the manuscript; and the decision to submit the