Cardiovascular Risk in Patients with Chronic Obstructive Pulmonary Disease: A Systematic Review

Background/Objectives: A comprehensive and up-to-date review on cardiovascular disease (CVD) risk in patients with COPD is needed. Therefore, we aimed to systematically review the risk of a range of CVD in patients with COPD. Methods: We searched three databases (Pubmed, Web of Science, SCOPUS) from inception to September 2023 using terms related to COPD and CVD. Observational studies were included if they (1) were conducted in adults with a diagnosis of COPD based on the GOLD criteria, spirometry, physician diagnosis, or review of electronic health records; (2) reported the risk of CVD, namely of myocardial infarction (MI), ischaemic heart disease (IHD), atrial fibrillation (AF), heart failure, cerebrovascular disease, pulmonary hypertension, and peripheral vascular disease, compared with a control population using a measure of risk. A narrative synthesis was used. Results: Twenty-four studies from 2015 to 2023, mainly from Europe (n = 17), were included. A total of 3,485,392 patients with COPD (43.5–76.0% male; 63.9–73.5 yrs) and 31,480,333 (40.0–55.4% male, 49.3–70.0 yrs) controls were included. A higher risk of CVD in patients with COPD was evident regarding overall CVD, MI, IHD, heart failure, and angina. Higher risks of arrhythmia and AF, stroke, sudden cardiac death/arrest, pulmonary embolism, pulmonary hypertension, and peripheral vascular disease were also found, although based on a small amount of evidence. Conclusions: Patients with COPD have a higher risk of CVD than the general population or matched controls. This review underscores the need for vigilant and close monitoring of cardiovascular risk in individuals with COPD to inform more precise preventive strategies and targeted interventions to enhance their overall management.


Introduction
Chronic obstructive pulmonary disease (COPD) is a heterogeneous lung disease characterized by chronic respiratory symptoms due to abnormalities of the airways and/or alveoli that cause persistent, often progressive, airflow obstruction [1].Worldwide, 391.9 million people aged 30-79 years have COPD [2].In 2019, COPD accounted for 3.3 million deaths and 74.4 million disability-adjusted life years (DALYs) [3].COPD imposes a significant economic and human burden on individuals, healthcare systems, and societies.
The presence of comorbidities often complicates the management of COPD and acute exacerbations, influencing prognosis and requiring specific treatment [1].The presence of comorbidities is linked to reduced health-related quality of life and increased morbidity, heightened susceptibility to hospital admissions, and an elevated risk of mortality [4].
Cardiovascular diseases (CVD) are one of the most important comorbidities of COPD [1,5], yet the overlap between these conditions often goes unrecognized.Research efforts have been made to unravel the key mechanisms behind the bidirectional link between COPD and CVD.Both conditions are related to the same risk factors (e.g., smoking, ageing, and physical inactivity) and to the pathophysiological mechanisms (e.g., arterial stiffness, inflammation, and endothelial dysfunction) [1].Chronic low-grade systemic inflammation is a potential link between the two conditions.Triggers like air pollutants and cigarette smoking increase inflammatory processes.Moreover, the paradoxical effects of cardiac hormones (e.g., brain natriuretic peptide-BNP), released in response to altered myocardial wall stress, further contribute to the intricate relationship between CVD and COPD [6].Lung hyperinflation, hypoxaemia, pulmonary hypertension, and shared genetics are other examples of the pathophysiological mechanisms underlying the association between COPD and CVD [7].
The presence of CVD is associated with poor prognosis, mainly an increased risk of exacerbations, hospitalizations, and mortality [1,5,8,9].Routine cardiovascular assessments, including medical history, physical examination, cardiovascular risk scoring, diagnostic tests (e.g., echocardiogram, chest X-ray, BNP testing, etc.), have been indicated for the early identification of signs of concomitant CVD in patients with COPD [9].Over the past decade, there have been several systematic reviews on the relation between COPD and CVD, with high heterogeneity [10][11][12][13][14][15].The main differences were related to population (stable and exacerbated COPD) and CVD outcomes.The older studies covered an extensive range of CVD outcomes [10,11] and the most recent have focused on specific outcomes (myocardial infarction [12], coronary heart disease [13], stroke [14], and myocardial infarction, cardiovascular death, and stroke) [15]).In addition, some reviews set cut-offs for the sample size (50, 100) [11,15] and limited the outcomes to specific risk measures (e.g., hazard ratio) [13] as inclusion criteria.A comprehensive and up-to-date review of CVD risk in patients with COPD is needed.
Therefore, we aimed to systematically review the risk of a range of CVD in patients with COPD.

Study Design
This systematic review was reported according to the preferred reporting in systematic reviews and meta-analysis (PRISMA) guidelines [16].We have registered the review with PROSPERO (registration number: CRD42023491294).

Search Strategy
We searched three databases (Pubmed, Web of Science, SCOPUS) from inception to September 2023.The search strategy included terms related to COPD and CVD.All studies were uploaded into EndNote to remove duplicates.

Eligibility Criteria
Observational studies were included if they met the following criteria: (1) conducted in adults with a diagnosis of COPD based on the GOLD criteria, spirometry, physician diagnosis, or review of electronic health records (diagnosis codes, medications prescribed, healthcare use); (2) reported the risk of CVD compared with a control population (non-COPD population, matched control, general population).Risk had to be reported as a measure of risk (e.g., hazard ratio, rate ratio, risk ratio, odds ratio, incidence rate ratio).The CVD outcomes considered were mortality, hospitalization or morbidity related to myocardial infarction (MI); ischaemic heart disease (IHD); atrial fibrillation (AF); heart failure; cerebrovascular disease; pulmonary hypertension; and peripheral vascular disease.These outcomes were considered as defined in the original studies, but had to be identified through physician diagnosis, clinical measurements, or review of electronic medical records (diagnosis codes, medications prescribed, health care use).We excluded studies that were based on a self-reported diagnosis of COPD or CVD, included an intervention, and used animals.Reviews, non-research letters, abstracts, case reports, conference proceedings, theses, and books were also excluded.

Screening, Selection Process, and Data Extraction
Studies were uploaded to Rayyan software (https://www.rayyan.ai/)for title and abstract independent screening by two reviewers (C.J. and I.F.).Any disagreement was resolved by discussion with a third author (A.S.-S.).Full-text articles were then read independently by two authors (C.J. and A.S.-S.) to identify studies that met the inclusion criteria.The two reviewers used a standardized form to independently extract data from each article, including the author's surname and year of publication, country, study design and setting, diagnostic criteria, population, and CVD risk.

Data Synthesis
Narrative synthesis was used to report the results.This was considered the most appropriate approach given the heterogeneity of data between the included studies.The findings were initially drafted by one researcher (C.J.) and then reviewed by a second researcher (A.S.-S.).

Study Selection
The database search yielded 7173 studies.After removing duplicate results, 4289 articles were screened for relevant content.During title and abstract screening, 4213 articles were excluded.Finally, 76 articles were full-text screened and 56 were excluded.Four additional articles were included through screening of previous reviews.In total, 24 studies were selected for qualitative analysis (Figure 1).

Characteristics of Included Studies
A summary of the included studies is shown in Table 1.The studies are from 2015 to 2023, mainly from Europe (n = 17), with the UK being the most represented country (n = 5).The remaining studies were conducted in North America (n = 5 Canada, n = 1 USA) and Asia (n = 1 Korea).A total of 19 studies were cohorts (15 retrospective and of those 8 casecontrol) and 5 were cross-sectional (1 case-control).The sample size of the studies ranged from 775 to 7,419,791.A total of 3,485,392 patients with COPD (43.5 to 76% male) and 31,480,333 (40 to 55.4% male) controls were included in the studies.Two studies did not
A summary of the risks of the different CVDs, either related to mortality or presence of comorbidity are presented in Tables 2 and 3, respectively, and described below.

Sudden Cardiac Death/Arrest
Patients with COPD had a higher risk of sudden cardiac death (HR 1.93 [33]) and cardiac arrest (HR 1.87-4.06[18]) compared with controls

Pulmonary Embolism
Two studies analyzed the risk of dying from a pulmonary embolism, but the results were non-significant [19,26].Three studies reported results on the risk of pulmonary embolism, but only two showed a higher risk associated with COPD (RR 2.39 and 2.72) [26,35].The same was observed when using odds ratios (OR 2.51 and 5.47) [19,27].

Pulmonary Hypertension
Two studies analyzed the presence of pulmonary hypertension in patients with COPD.Morgan et al. showed that the risk of this condition was higher in patients with COPD, decreasing with age (HR 3.70-27.70)[18].Baty et al. 2013 also found higher odds in patients with COPD (OR 5.60-5.80)[29].

Peripheral Vascular Disease
Both the risk of death from peripheral vascular disease (HR 1.32 [36]) and the risk of having PVD (HR 1.42-7.70[18]) were higher in patients with COPD.In patients with COPD, the odds of having PVD was found to be 1.85-2.80[29,36].
Figure 2 presents a summary of risks related to mortality and comorbidity.

Peripheral Vascular Disease
Both the risk of death from peripheral vascular disease (HR 1.32 [36]) and the risk of having PVD (HR 1.42-7.70[18]) were higher in patients with COPD.In patients with COPD, the odds of having PVD was found to be 1.85-2.80[29,36].
Figure 2 presents a summary of risks related to mortality and comorbidity.

Discussion
Most studies that have examined the risk of CVD in patients with COPD suggest that those with COPD have a higher risk than the general population or matched controls.This review has also highlighted the heterogeneity in the criteria used to diagnose COPD among the published studies.

CVD
Eight studies analyzed the risk of CVD (either as cause of death or as a comorbidity) [17,19,23,24,26,32,39] and seven found a higher risk in patients with COPD compared with the general population/matched controls.The only study that did not show a significant

Discussion
Most studies that have examined the risk of CVD in patients with COPD suggest that those with COPD have a higher risk than the general population or matched controls.This review has also highlighted the heterogeneity in the criteria used to diagnose COPD among the published studies.

CVD
Eight studies analyzed the risk of CVD (either as cause of death or as a comorbidity) [17,19,23,24,26,32,39] and seven found a higher risk in patients with COPD compared with the general population/matched controls.The only study that did not show a significant association was the study conducted in an Asian population and this difference may be due to the characteristics of the population studied and the specific epidemiological features [41].Further research needs to be conducted to better understand the link between COPD and CVD [42], mainly in regions that are less represented in the collected evidence, such as Asia, South America, and Africa.

MI, IHD, and Angina
COPD has been associated with a higher risk of MI [17,19,23,25,26,28,31,35], IHD [17,23,29,30,35,36,[38][39][40], and angina [18,19,23,26,35].The only study that did not find a significant association between MI (death or event) was Rodriguez et al., which may be in part explained by the exclusion of patients with a history of coronary heart disease [25].This evidence is in line with a previous review [12].These consistent findings across multiple studies suggest a robust association between COPD and an increased susceptibility to CVD, underscoring the need for vigilant monitoring of cardiovascular health as well as cardiopulmonary risk in individuals with COPD [9,43].Assessment of blood eosinophil count together with commonly used forced expiratory volume in 1 s (FEV1) can be of value for monitoring patients with COPD.Different incidences of non-fatal MI and cardiovascular death across distinct pharmacological treatments were found, which were more pronounced with increasing blood eosinophil [44].Future research could delve deeper into the underlying mechanisms of this association to inform more precise preventive strategies and targeted interventions to enhance the overall management of COPD patients [42].

Heart Failure
Evidence is robust showing that COPD is associated with an increased risk of heart failure [17][18][19]23,25,26,30,[35][36][37][38]40], with most studies showing more than a two-fold higher risk compared to controls.This has been a topic of interest in recent opinion reviews [43,45,46].The EURObservational Research Programme Heart Failure Long-Term Registry found that up to 19% of people hospitalized with heart failure also had diagnosed COPD [47].To reduce symptoms, delay progression, and improve prognosis, it is essential to better screen and diagnose these two coexisting conditions and when present, to establish a management strategy that addresses both simultaneously.This is especially true given the rising mortality rates and the significant negative impact of each disease on quality of life and functional status [48].

Other CVD Outcomes
Seven studies have shown the link between stroke and COPD [17][18][19]23,26,35,36], with consistent risk measures.These results are aligned with a recent systematic review on the topic, which showed pooled odds for stroke risk of 1.40 (95% CI 1.24-1.59)and for stroke mortality of 1.20 (95% CI 1.13-1.27)[14].Evidence from other CVD outcomes is still scarce, with too few studies to allow a proper synthesis of the associated risk.Future observational cohort studies should gather more information regarding these outcomes.

Strengths and Limitations
This review benefits from the use of a comprehensive search strategy that included three bibliographic databases.Our review used more robust diagnostic criteria for both COPD and CVD, thus excluding several studies previously presented in prior reviews.Nevertheless, the risk for each CVD was summarized as presented in the original studies, but significant heterogeneity in CVD definitions may exist across these studies, which were not accounted for.Patients with COPD were mostly selected based on diagnosis codes using internationally standardized classifications, but considering that COPD is frequently underdiagnosed, more symptomatic/severe COPD are likely to be over-represented in the original studies included.In addition, some studies may have excluded a proportion of patients with COPD from their analyses by using stricter criteria such as a specific age range (excluding either younger or older patients) or by requiring spirometry data, which is not always available.Validation and comparison of the accuracy of CVD and COPD case definitions [49,50] in distinct databases can be a strategy to overcome these limitations in the future [51].The severity of COPD and COPD phenotypes may also play a role in the development of CV comorbidities as has been previously suggested [52].However, this could not be determined.Only the study from Ingebrigtsen presented the risk in relation to GOLD grades.With the increasing number of COPD diagnoses based on spirometry [53], future studies may shed light on this unanswered research question [42].Another limitation is that most studies did not adjust the risk of CV outcomes for patients' smoking status.This is probably because this information was not available in several databases.However, six studies included this adjustment, three of which used primary care databases and three published since 2020.This shows that the availability and quality of reporting of this information is likely to be improving, which will allow future observational studies to better adjust for possible confounders and clarify the role of COPD in the development of CVD.It also highlights the potential of using large datasets that include EMR from primary and secondary care [54].In addition, we did not account for potential duplication of findings, as many studies were the sole representatives from a country or derived from the same database but assessed distinct outcomes.However, it is important to note that in studies conducted in the UK and Canada, despite the majority utilizing different databases and/or index periods and focusing on distinct mortality and comorbidity risks, some overlap in findings may still exist.This will need to be better considered in future systematic reviews including meta-analysis with sensitivity analysis.As this was primarily a qualitative synthesis, with the aim of summarizing all existing evidence on the topic, the risk of bias of the included studies was not assessed.

Clinical Implications and Research Future Directions
To address the coexistence of COPD and CVD a comprehensive and integrated care model is advocated, encompassing primary prevention, screening in primary care settings, and fostering of multidisciplinary collaboration (pulmonologists, cardiologists, and general practitioners) in both primary and secondary care [6,42].CVD early detection can be performed through minimal checkup (e.g., medical history, physical examination, blood tests, cardiovascular risk scoring) and, if further investigation is needed, through predictive biomarkers (e.g., BNP) and imaging procedures, such as echocardiography and coronary computed tomography angiography [9,55].Clinical decision support systems are also recommended as a pivotal strategy [6].These collaborative efforts are crucial for stratifying cardiovascular risk in COPD patients ensuring the delivery of appropriate treatment.The higher CVD risk emphasizes also the burden of cardiopulmonary risk in patients living with COPD [43,56].Recognizing that exacerbations pose a significant threat to both pulmonary and CV health, there is an imperative need for proactive identification and targeted treatment of COPD individuals at risk of exacerbation.It would be of benefit if, in the near future, we could quantify and categorize levels of cardiopulmonary risk in patients with COPD.Well-designed retrospective and longitudinal real-world studies including electronic health records from primary and secondary care are needed to comprehensively address this research question [42].The potential of large datasets will probably enhance the quality and comprehensiveness of the data collected.As diagnosis codes are associated with some misclassification, future studies may consider the presence of COPD only when related outcomes are assessed (e.g., FEV1).As data availability on smoking habits and COPD severity improves, future studies should provide clearer insights into their correlation with CVD.This review focused on the risk of CVD in COPD in a unidirectional way.Yet, as COPD also adversely affects the prognosis of CVD, future reviews could include both angles of this bidirectional relationship.

Conclusions
Most studies which have investigated the risk of CVD in patients with COPD suggest that those with COPD have a higher risk than the general population or matched controls.This review underscores the need for vigilant and close monitoring of cardiovascular and cardiopulmonary risk in individuals with COPD to inform more precise preventive strategies and targeted interventions to enhance their overall management.The retrospective design of most studies, coupled with limited or no adjustments for confounding factors, impairs the drawing of definitive conclusions regarding CV risk in patients with COPD.Additional well-designed prospective studies to comprehensively address this question are needed.

Table 1 .
Characteristics and results of included studies.

Table 2 .
Measures of mortality risk in patients with COPD.

Table 3 .
Measures of comorbidities risk in patients with COPD.