Factors associated with coronary heart disease in COPD patients and controls.

BACKGROUND: COPD and coronary heart disease (CHD) frequently co-occur, yet which COPD phenotypes are most prone to CHD is poorly understood. The aim of this study was to see whether COPD patients did have a true higher risk for CHD than subjects without COPD, and to examine a range of potential factors associated with CHD in COPD patients and controls.
METHODS: 347 COPD patients and 428 non-COPD controls, were invited for coronary computed tomography angiography (CCTA) and pulmonary CT. Arterial blood gas, bioelectrical impedance and lung function was measured, and a detailed medical history taken. The CCTA was evaluated for significant coronary stenosis and calcium score (CaSc), and emphysema defined as >10% of total area <-950 Hounsfield units.
RESULTS: 12.6% of the COPD patients and 5.7% of the controls had coronary stenosis (p<0.01), whereas 55.9% of the COPD patients had a CaSc>100 compared to 31.6% of the controls (p<0.01). In a multivariable model adjusting for sex, age, body composition, pack-years, CRP, cholesterol/blood pressure lowering medication use and diabetes mellitus, the OR (95% CI) for having significant stenosis was 1.80 (0.86-3.78) in COPD patients compared with controls. In a similar model, the OR (95% CI) for having CaSc>100 was 1.68 (1.12-2.53) in COPD patients compared with controls. Examining the risk of significant stenosis and CaSc>100 among COPD patients, no variable was associated with significant stenosis, whereas male sex [OR 2.85 (1.56-5.21)], age [OR 3.74 (2.42-5.77)], statin use [OR 2.23 (1.23-4.50)] were associated with CaSc>100, after adjusting for body composition, pack-years, C-reactive protein, use of angiotensin converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs), diabetes, emphysema score, GOLD category, exacerbation frequency, eosinophilia, and hypoxemia.
CONCLUSION: COPD patients were more likely to have CHD, but neither emphysema score, lung function, exacerbation frequency, nor hypoxemia predicted presence of either coronary stenosis or CaSc>100.

Background

Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality worldwide and includes phenotypic traits such as emphysema, exacerbation frequency, and respiratory failure. The rate of progression of COPD is heterogeneous, varying greatly between and within afflicted individuals. Many patients exhibit significant comorbidities, among them coronary heart disease (CHD), accompanied by both a high symptom burden and mortality [1,2]. Numerous previous studies have established that CHD is more common in COPD patients than in the general population [3-8]. Cardiovascular disease is the leading cause of death in COPD patients [9].

COPD and CHD share risk factors, especially cigarette smoking [10]. However, it is thought unlikely that the coexistence of COPD and CHD can be explained only by shared risk factors, since reduced lung function has been shown to be a risk factor for CHD independent of smoking [11]. Systemic inflammation has been demonstrated in several studies of COPD patients [12,13], and is known to be associated with atherosclerosis and plaque formation [14]. COPD patients with chronic hypoxemia and increased COPD exacerbation frequency may be especially prone to exhibit systemic inflammation and thus further increased risk for CHD. In addition, COPD patients with predominant emphysema have increased static hyperinflation, which may compromise cardiac function [13].

However, although these factors plausibly connect COPD to CHD, direct evidence that COPD development in itself leads to CHD is scarce. There may be at least two different explanations for this. First, COPD is a heterogenous disease, with large differences in disease manifestations. COPD disease severity is related to lung function decline, yet we know degree of emphysema can vary considerably between to patients with the same lung function. Thus, some patients may be more prone to develop CHD than others, depending on their manifestation of COPD. Second, previous studies have usually relied on self-report of presence of CHD, and not actual visualization of the coronary arteries. Thus, misclassification may also have confounded earlier studies.

In the current study from Western Norway, we performed a combined pulmonary CT scan and coronary CT angiography (CCTA) of our COPD patients and non-COPD controls, together with a thorough medical history, arterial blood gas and lung function measurements. The aim of this study was to see whether COPD patients did have a true higher risk for CHD than subjects without COPD, and to examine a range of potential factors associated with CHD in COPD patients and controls.

Methods

Study population and design

The participants were recruited as a cross-sectional sample from two previous patient-control cohorts: The MicroCOPD study and the follow-up phase of the GeneCOPD study. The MicroCOPD Study was conducted 2012–2015 [15], and included 16 asthma patients whom were excluded from the current study, whereas the GeneCOPD follow-up 2013–2016 only included COPD patients and controls from the first GeneCOPD study in 2003–2004 [16].

In the original GeneCOPD study in 2003–2006, only half of the participants had pulmonary CT scans taken. At follow-up in 2013–206, more healthy controls than COPD patients were survivors, and due to cost restraints, not all controls without a baseline CT from 2003–2006 of the lungs were offered one at follow-up. There were 926 potential participants from the two cross-sectional samples, 16 asthma patients were excluded from the MicroCOPD study, 130 subjects had no CT taken, and for 5 subjects we only had a pulmonary CT, either due to allergy against contrast fluids, or finding calculated glomerular filtration rate (GFR) < 30 ml/min/1.73m2. The final study sample included 775 participants, 347 COPD patients and 428 non-COPD controls (Fig 1).

Fig 1

Flow chart of study selection and final CT data availability.

The diagnosis of COPD was based on the Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines [17] and required both spirometrically confirmed airflow obstruction, and a physician’s assessment of the clinical history.

Subjects for whom the calcium score was above 500 or had concurrent coronary artifacts were prevented from assessment of coronary stenosis. In some instances, the export of single DICOM picture files from the IMPAX client was corrupted, and pulmonary CT scans were lost. Table 1 provides an overview of the study population, and numbers of available pulmonary CT measurements, coronary stenosis evaluations and calcium score measurements.

Table 1

Overview of the study population who underwent pulmonary CT, CCTA, and measured CaSc.

	COPD patients	Controls	Sum
Total number of participants	347	428	775
Participants with pulmonary CT	315	391	706
Participants with measured calcium scores	333	424	757
Participants with coronary stenosis evaluation (CCTA)	254	371	625

Both studies were approved by the Norwegian ethical committee (GeneCOPD follow-up by REK-Vest, case number 2010/2015, MicroCOPD by REK-Nord, case number 2011/1307), and all participants gave written informed consent.

Data collection

All patients and controls attended study visits at the outpatient clinic at Department of Thoracic Medicine, Haukeland University Hospital. All participants underwent lung function measurements on a Vitalograph 2160 Spirometer (Vitalograph Ltd, Maids Moreton, UK). Forced vital capacity (FVC) and forced expiration volume in 1 second (FEV1) was measured after bronchodilation. Norwegian pre-bronchodilation reference values were used [18]. Arterial blood was taken and immediately analyzed on an ABL 720 radiometer for analyses of blood gases. Venous blood samples were taken, including hematology, cholesterols, and kidney function (creatinine for calculation of glomerular flow rate). In addition, body composition was assessed with bioelectrical impedance measurements using a Bodystat 1500 to calculate fat mass index (FMI) and fat free mass index (FFMI). Cachexia was defined as having a FFMI < 14 kg/m2 in women and < 17 kg/m2 in men which corresponds to the lower 95% CI in a normal population [19]. Similarly, obesity was defined as FMI > 13.5 kg/m2 in women and > 9.3 kg/m2 in men. Finally, a full medical history was obtained including respiratory symptoms, medication use, COPD exacerbation history in the last 12 months, and known comorbidities including diabetes.

CT protocol

Both the coronary computed tomography angiography (CCTA) and pulmonary CT were performed with a 256-detector row Dual source Flash CT (Siemens®). Initially a scan of the heart region including coronary arteries was performed without contrast to determine calcium score, calculated using Agatstons coronary artery calcium score (CaSc) to describe density and extent of the calcification [20]. The CCTA was not performed if the CaSc was higher than 500 (n = 139, of which 85 were COPD patients). Among 561 subjects with non-significant stenosis, the median (IQR) CaSc was 15 (0–105), whereas among 46 subjects with significant stenosis the median CaSc was 243 (61–391), p-value = 0.0001, Kruskal Wallis test. As defined by Rumberger et al [21], a CaSc > 100 was used as an indicator of a high risk for coronary heart disease.

An interventional cardiologist and a cardiac radiologist evaluated the findings of coronary artery disease using the modified American Heart Association coronary segmentation model [22]. Lumen reduction was analyzed by measuring the diameter of the most stenotic part in the coronary artery. Confirmed coronary stenosis was defined as presence of stenosis (lumen reduction > 50%).

The contrast medium Iomeron® 400 adjusted for body weight was used. Both the inspiratory and expiratory pulmonary scans were obtained with 0.5 mm intervals, and a reconstruction algorithm was added to both the CCTA and pulmonary CT scans. For classification of emphysema on the pulmonary CT scans, 3D Slicer software [23] was used for density mask analysis, where significant emphysema was defined as > 10% of area below the density threshold -950 Hounsfield units (Hu), an accepted threshold validated with histopathology [24,25].

Statistical analyses

Stata 14 was used for statistical analyses [26]. The Pearson χ2 test was used to compare categorical bivariate associations. Fisher exact test was used to compare variables with cell categories < 5. Multivariable logistic regression was performed to determine independent predictors of coronary stenosis and having a CaSc > 100. The analyses were first employed on the whole study group to compare COPD patients and non-COPD controls, adjusting for sex, age, body composition, smoking, c-reactive protein, use of statins and blood pressure lowering drugs, in addition to presence of diabetes. Afterwards, models were fitted to the COPD group only, to assess whether COPD disease characteristics could predict the presence of either coronary stenosis or a CaSc > 100. First-order interactions between sex, age, smoking and all other variables were tested for the COPD patients. Due to the large number of interactions (n = 66), a p-value of 0.01 was used, for all other analyses a p-value of < 0.05 was considered statistically significant.

Results

Table 2 shows the study characteristics of the COPD and control subjects. Mean age among COPD patients and controls were 69.0 years (SD 7.9) and 64.8 years (SD 8.5) respectively. The COPD patients were older, less likely to have a normal body composition, had smoked more frequently and a larger load, and were more likely to use both cholesterol and blood pressure lowering drugs than control subjects. Among the COPD patients, 11.2% were GOLD category I, 53.5% GOLD category II, 25.9% GOLD category III, and 9.5% GOLD category IV. The COPD patients also had a higher calcium score and emphysema burden, in addition to a higher percentage afflicted with diabetes compared to the control subjects. Whereas 12.6% of COPD patients had confirmed coronary stenosis, only 5.7% of the controls did (p < 0.01). Considering CaSc, 55.9% of the COPD patients had a CaSc value >100, compared with 31.6% of the controls (p < 0.01).

Table 2

Baseline characteristics of the study population in percentages.

	COPD patients	Controls
	(n = 347)	(n = 428)	p*
Sex			0.89
Women	45.8	45.3
Men	54.2	54.7
Age categories (years)			<0.01
40–59.9	12.7	34.1
60–69.9	43.8	42.5
70–90	43.5	23.4
Body composition			<0.01
Normal	53.1	80.4
Cachectic	27.1	7.6
Obese	19.8	11.9
Smoking habits			<0.01
Never	0.3	4.4
Ex	68.6	62.4
Current	31.3	33.2
Pack years per 10 years increase			<0.01
< 20	31.7	57.1
20–40	40.4	33.2
40+	28.0	9.8
Total cholesterol (mmol/L)			0.04
0–4.99	46.2	38.6
5–5.99	29.2	30.9
6–6.99	17.3	24.8
7 +	7.2	5.6
Using statins			<0.01
Yes	32.0	21.8
Using either ACE inhibitors or ARBs			0.35
Yes	16.4	14.0
Having significant coronary stenosis			<0.01
Yes	12.6	5.7
Calcium score (HU)			<0.01
0–100	44.1	68.4
> 100	55.9	31.6
Emphysema as evaluated by CT (> 10% less than 950Hu)			<0.01
Yes	22.9	0.26
C-reactive protein (mg/L)			<0.01
< 5	68.0	90.4
≧ 5	32.0	9.6
Eosinophilia (≥0.3*10^9 cells/L)			<0.01
Yes	36.1	23.8
Diabetes			<0.01
Yes	11.6	5.8

*Chi-square test for categorical variables.

Table 3A and 3B present potential predictors of presence of coronary stenosis or CaSc > 100 in COPD patients and controls respectively. Among a large range of potential predictors, male sex and age were associated with significant coronary stenosis > 50% for both groups (Table 3A). Further, prevalence of CaSc > 100 was associated with male sex, higher age, and statin use for both groups (Table 3B). GOLD category, emphysema score, and COPD exacerbation frequency were not associated with either coronary stenosis or an increased CaSc in the bivariate analyses.

Table 3A

The prevalence of significant (>50% lumen reduction) coronary stenosis among 347 COPD patients and 428 non-COPD controls as evaluated from coronary CT scans.

	COPD patients		Controls
	n (%)	p*	n (%)	p*
Sex		0.04		0.02
Women	10 (8.1)		5 (2.8)
Men	22 (16.8)		16 (8.4)
Age categories (years)		0.004†		0.04
40–59.9	1 (2.4)		5 (3.5)
60–69.9	12 (9.8)		8 (4.9)
70–90	19 (21.4)		8 (11.9)
Body composition		0.37		0.33†
Normal	14 (10.1)		19 (6.4)
Cachectic	9 (15.5)		1 (4.0)
Obese	8 (17.0)		0 (0)
Smoking habits		0.27†		0.30†
Never	0 (0)		2 (13.3)
Ex	25 (14.8)		13 (5.7)
Current	7 (8.3)		6 (4.7)
Pack years		0.35		0.07†
< 20	6 (8.3)		11 (5.6)
20–40	14 (14.3)		2 (1.8)
40+	11 (16.2)		3 (11.1)
Total cholesterol (mmol/L)		0.12†		0.02†
0–4.99	16 (14.2)		15 (10.6)
5–5.99	4 (5.6)		3 (2.6)
6–6.99	8 (16.0)		3 (3.3)
7 +	4 (21.1)		0 (0)
Using statins		0.49		0.01
No	21 (11.7)		13 (4.3)
Yes	11 (14.9)		8 (12.1)
Using either ACE inhibitors or ARBs		0.21		0.18
No	25 (11.5)		16 (5.0)
Yes	7 (18.9)		5 (9.6)
C-reactive protein (mg/L)		0.02		0.12†
< 5	16 (9.3)		17 (5.1)
≧ 5	16 (19.8)		4 (11.4)
Diabetes		0.01		0.29†
No	25 (10.9)		19 (5.4)
Yes	7 (29.2)		2 (10.5)
Emphysema as evaluated by CT		0.79
No	21 (11.6)		20 (5.9)
Yes	6 (13.0)		0 (0)
GOLD category		0.85
I/II	22 (12.9)
III/IV	10 (12.1)
> 1 exacerbations last year		0.27†
No	30 (13.8)
Yes	2 (5.7)
Eosinophilia (≥0.3*10^9 cells/L)		0.97
No	21 (12.7)
Yes	11 (12.5)
Blood gases		0.38†
pO2 < 8 kPa	2 (16.7)
pO2 8–9 kPa	3 (6.8)
pO2 > 9 kPa	27 (14.0)
* Chi-square test.
† If cells have less than 5 cases for a comparison, a Fisher’s exact test was chosen.
Table 3B The prevalence of calcium score >100 HU among 347 COPD patients and 428 non-COPD controls as evaluated from coronary CT scans.
	COPD patients		Controls
	n (%)	p*	n (%)	p*
Sex		0.002		<0.001
Women	75 (47.2)		36 (18.7)
Men	111 (63.8)		98 (42.4)
Age categories (years)		<0.001		<0.001
40–59.9	11 (25.0)		18 (12.4)
60–69.9	65 (44.2)		53 (29.4)
70–90	110 (77.5)		63 (63.6)
Body composition		0.17		0.01
Normal	83 (50.6)		93 (27.9)
Cachectic	51 (59.3)		17 (53.1)
Obese	41 (63.1)		20 (40.0)
Smoking habits		0.13†		0.85
Never	0 (0)		6 (31.6)
Ex	135 (58.7)		86 (32.6)
Daily	51 (50.0)		42 (29.8)
Pack years		0.44		0.01
< 20	57 (57.0)		67 (30.3)
20–40	63 (51.2)		36 (28.6)
40+	51 (59.8)		21 (55.3)
Total cholesterol (mmol/L)		0.09		0.01
0–4.99	96 (51.6)		67 (41.4)
5–5.99	50 (50.5)		35 (26.5)
6–6.99	27 (47.4)		26 (24.8)
7 +	13 (52.0)		5 (20.8)
Using statins		<0.001		<0.001
No	112 (48.3)		85 (25.3)
Yes	74 (73.3)		49 (55.7)
Using either ACE inhibitors or ARBs		0.01		0.23
No	147 (52.9)		111 (30.5)
Yes	39 (70.9)		23 (38.3)
C-reactive protein (mg/L)		0.51		0.99
< 5	124 (54.6)		121 (31.6)
≧ 5	62 (58.5)		13 (31.7)
Diabetes		0.66		0.09
No	164 (55.6)		123 (30.7)
Yes	22 (59.5)		11 (47.8)
Emphysema as evaluated by CT		0.10		0.32†
No	127 (54.5)		122 (31.6)
Yes	46 (65.7)		1 (100)
GOLD category		0.11
I/II	113 (52.6)
III/IV	73 (61.9)
> 1 exacerbations last year		0.24
No	161 (57.1)
Yes	23 (47.9)
Eosinophilia (≥0.3*10^9 cells/L)		0.35
No	116 (54.0)
Yes	70 (59.3)
Blood gases		<0.001†
pO2 < 8 kPa	19 (82.6)
pO2 8–9 kPa	43 (68.3)
pO2 > 9 kPa	120 (50.0)
* Chi-square test.
† If cells have less than 5 cases for a comparison, a Fisher’s exact test was chosen.

The unadjusted and adjusted ORs with 95% CI for having coronary stenosis or a CaSc > 100 for COPD patients compared with controls is shown in Fig 2.

Fig 2

The unadjusted and adjusted OR (95% CI) for having coronary stenosis and CaSc > 100 in COPD patients compared with non-COPD controls.

The controls had reference odds of 1, depicted by the stippled line. Adjustment was made for sex, age, body composition, pack years smoked, C-reactive protein, statin use, use of ACE inhibitors or ARBs, and diabetes. The COPD patients had an OR of 2.4 (1.35–4.27) for having coronary stenosis on the coronary CT scans compared with the controls, however after adjustment the association was reduced, and did not reach statistical significance (OR 1.80 (0.86–3.78, Fig 2 & Table 4A).

Table 4A

The odds ratios (95% CI) for having significant coronary stenosis on CT angiography (n = 574).

	OR	(95% CI)	p
Study category
Controls	1
COPD	1.80	(0.86–3.78)	0.12
Sex
Women	1
Men	2.51	(1.24–5.09)	0.01
Age per 10 years increase	1.95	(1.28–2.96)	0.002
Body composition
Normal	1
Cachectic	1.04	(0.43–2.49)	0.93
Obese	0.75	(0.30–1.84)	0.53
Pack years per 10 years increase	1.04	(0.86–1.25)	0.07
C-reactive protein (mg/L)
< 5	1
≧ 5	2.31	(1.11–4.77)	0.03
Using statins
No	1
Yes	1.02	(0.49–2.11)	0.96
Using either ACE inhibitors or ARBs
No	1
Yes	1.56	(0.69–3.52)	0.29
Diabetes
No	1
Yes	1.97	(0.78–4.95)	0.15
Table 4B The odds ratios (95% CI) for having calcium score > 100 Hu on CT angiography (n = 775).
	OR	(95% CI)	p
Study category
Controls	1
COPD	1.68	(1.12–2.53)	0.014
Sex
Women	1
Men	2.89	(1.97–4.22)	<0.001
Age per 10 years increase	3.43	(2.65–4.46)	<0.001
Body composition
Normal	1
Cachectic	1.01	(0.60–1.69)	0.98
Obese	1.29	(0.78–2.12)	0.32
Pack years per 10 years increase	1.02	(0.91–1.15)	0.70
C-reactive protein (mg/L)
< 5	1
≧ 5	1.23	(0.77–1.98)	0.39
Using statins
No	1
Yes	2.60	(1.69–3.98)	<0.001
Using either ACE inhibitors or ARBs
No	1
Yes	1.52	(0.92–2.51)	0.11
Diabetes
No	1
Yes	0.71	(0.36–1.38)	0.31

The COPD patients did have a statistically significant increased risk for having a CaSc >100, with an OR (95% CI) of 1.68 (1.12–2.53), Fig 2 & Table 4B. Male sex and higher age were associated with both coronary stenosis (Table 4A) and CaSc > 100 (Table 4B). CRP indicating systemic inflammation was predicative for significant coronary stenosis, but not CaSc > 100.

Looking at predictors among COPD patients, no variables were significantly associated with coronary stenosis (Table 5A). Considering the chance of finding CaSc > 100 in the COPD patients, male sex, higher age and statin use were associated at a statistically significant level (Table 5B). Neither of the COPD characteristics predicted the chance for having coronary stenosis or CaSc as judged by CCTA in these models, and neither of the first-order interactions tested were statistically significant (data not shown).

Table 5A

The odds ratios (95% CI) for having significant coronary stenosis on CT angiography among COPD patients (n = 205).

	OR	(95% CI)	p
Sex
Women	1
Men	2.72	(0.97–7.56)	0.06
Age per 10 years increase	1.74	(0.94–3.23)	0.08
Body composition
Normal	1
Cachectic	2.71	(0.79–9.34)	0.11
Obese	1.92	(0.63–5.84)	0.25
Pack years per 10 years increase	0.97	(0.75–1.26)	0.82
Using statins
No	1
Yes	0.60	(0.21–1.71)	0.34
Using either ACE inhibitors or ARBs
No	1
Yes	2.64	(0.80–8.71)	0.11
C-reactive protein (mg/L)
< 5	1
≧ 5	2.19	(0.80–5.97)	0.13
Diabetes
No	1
Yes	2.49	(0.67–9.25)	0.17
Percent area with < 950 Hu density (emphysemascore) per 1% increase
GOLD category	1.02	(0.95–1.08)	0.62
I/II	1
III/IV	0.57	(0.18–1.82)	0.35
> 1 exacerbations the last year
No	1
Yes	0.39	(0.07–2.26)	0.29
Eosinophilia (≥0.3*10^9 cells/L)
No	1
Yes	0.96	(0.36–2.56)	0.93
Respiratory failure (pO2 < 8 kPa)
No	1
Yes	2.15	(0.33–14.21)	0.43
Table 5B The odds ratios (95% CI) for having calcium score > 100 Hu on CT angiography among COPD patients (n = 271).
	OR	(95% CI)	p
Sex
Women	1
Men	2.79	(1.51–5.14)	0.001
Age per 10 years increase	3.69	(2.39–5.70)	<0.001
Body composition
Normal	1
Cachectic	0.70	(0.32–1.49)	0.35
Obese	1.35	(0.64–2.85)	0.43
Pack years per 10 years increase	0.96	(0.81–1.13)	0.60
Using statins
No	1
Yes	2.42	(1.27–4.63)	<0.01
Using either ACE inhibitors or ARBs
No	1
Yes	2.29	(1.01–5.19)	0.047
C-reactive protein (mg/L)
< 5	1
≧ 5	1.05	(0.55–1.99)	0.89
Diabetes
No	1
Yes	0.65	(0.26–1.64)	0.36
Percent area with < 950 Hu density (emphysemascore) per 1% increase	1.02	(0.98–1.06)	0.44
GOLD category
I/II	1
III/IV	1.16	(0.56–2.39)	0.69
> 1 exacerbations the last year
No	1
Yes	1.30	(0.55–3.08)	0.56
Eosinophilia (≥0.3*10^9 cells/L)
No	1
Yes	0.85	(0.47–1.54)	0.59
Respiratory failure (pO2 < 8 kPa)
No	1
Yes	2.87	(0.70–11.84)	0.15

Discussion

In this study of both COPD patients and non-COPD controls who underwent pulmonary CT imaging and CCTA, we confirmed the increased presence of coronary artery disease in COPD patients. After adjustment for potential confounders, COPD patients had significantly higher odds for having an increased calcium score. However, in the same multivariable analyses, age, male sex, and CRP ≥5 was significantly associated with having significant coronary stenosis, and age, male sex and statin use for increased calcium score. Among COPD patients only, neither of the COPD phenotypic traits GOLD category, presence and severity of respiratory failure, or CT evaluated emphysema, predicted the chance of finding stenosis on CCTA or CaSc > 100.

Although the coexistence of COPD and CHD has been established in several large studies of different study designs [4,5,27], there has been concern regarding misclassification, both due to COPD having many different manifestations, but also since establishing CHD with certainty requires somewhat invasive procedures. To date, few studies have performed coronary angiography in COPD patients. In a North American study of 351 COPD patients and 122 patients with interstitial lung disease (ILD) undergoing evaluation for lung transplantation, 60% had CHD of those in which coronary angiography was performed [6]. Interestingly, the prevalence of CHD was similar in COPD and ILD patients, although the smoking load was threefold higher among COPD patients.

In another study from Brazil, the prevalence of CHD was 88% among 101 COPD patients compared with 45% among 109 non-COPD controls [28]. In this study smoking habits varied greatly between study groups, and the prevalence of CHD in both groups were quite high. This may be due to a selection of patients under suspicion of heart disease, as is a likely cause of a high CHD prevalence in the largest study to date; a retrospective cohort study of 26,137 patients who underwent cardiac intervention and revascularization in Alberta, Canada [29]. In this registry study without spirometry data, the prevalence of CHD was 87% in COPD patients and 89% among non-COPD patients, however with more severe CHD among the COPD patients. In none of the three mentioned studies was emphysema evaluated with pulmonary CT.

Another sign of CHD is an increased coronary artery calcium score (CaSc) [30]. Just as with the few studies using coronary angiography, the few previous studies using CaSc have not yielded consistent findings. In a South-Korean study of 4905 men undergoing regular health measurements, lower FVC and FEV1 was correlated with a higher CaSc [31]. However, in a general population sample from North America including both sexes and more comprehensive smoking characterization, neither FEV1 nor emphysema were correlated with CaSc [32]. In an Italian lung cancer screening trial among 1,159 smokers, the lung function in the study samples were lower than in the general population studies, however, no association between emphysema or FEV1 and CaSc was found [33]. In this study non-gated low-dose CT with 5 mm slice thickness was used, which could have underestimated the CaSc.

Two studies have specifically examined COPD patients. In a matched case-control study of 81 COPD patients and 81 non-COPD controls from Switzerland, no difference in CaSc between groups were seen [34]. And in a sub-analysis from the ECLIPSE cohort, CaSc was found to be significantly higher among 672 COPD patients than 270 controls after adjustment for age, sex, ethnicity and pack-year smoking history [35]. Among COPD patients, neither FEV1, GOLD category, nor exacerbation frequency was significantly related to CaSc. However, CaSc was calculated from pulmonary CT, not as part of a coronary CaSc evaluation [35].

The current study, with superior characterization of the presence of coronary heart disease [36], adds weight to the notion that COPD patients indeed have more CHD than non-COPD patients, and more than smoking history alone can explain. However, it appears that COPD severity in terms of lower FEV1, presence of respiratory failure, or having a history of frequent exacerbations did not present an increased risk for having CHD, much the same as was seen in the ECLIPSE cohort. Neither did increased smoking in terms of pack-years present a clear risk for CHD or having a higher CRP, among COPD patients. However, CRP ≥5 was associated with having significant coronary stenosis, but not CaSc >100, among the study population. CRP is only an indirect measurement of systemic inflammation, yet the findings in this study was in concordance with the works of Jenny et al and Lin et al [37,38].

Use of cholesterol lowering drugs was significantly associated with CaSc >100 in our study. This is possibly because of primary or secondary prophylaxis treatment, but it is also worth considering increased CaSc after treatment with cholesterol lowering drugs such as statins as a representation of plaque repair and stabilization [39].

Thus, if COPD severity is not predictive of increased risk for CHD even though COPD patients as a group have more CHD, what could that mean? One possibility is a risk factor being at play early in the course of disease, linked to the start or development of COPD, rather than its progression. It is already known that smokers who quit still have a higher rate of decline in FEV1 after smoking, once COPD is established, thus a similar phenomenon may be happening here. Which factors these may be are obviously unknown and not captured by the current study, but it is tempting to suggest factors of known relevance to CHD, and thus somehow linked to inflammation.

Second, it also means that clinicians need to keep in mind that having COPD, regardless of disease severity, carries with it an increased risk of CHD. This awareness is important to avoid oversight of symptoms like dyspnea and vague chest discomfort, which can easily be interpreted as symptoms caused by the known disease COPD.

The analyses in this study are associated with some limitations. First, causality cannot be established from a cross-sectional study. Second, study sample size may have been too low, and there is a chance for type II errors. Thus, we urge caution in interpreting all negative associations as proof of a non-causal relationship. Third, all of the subjects included in the study did not do a complete set of the CT scans for different reasons. 56 subjects were registered with no pulmonary CT because of a technical failure in transferring the images from the radiological database. However, this is unlikely to have affected the results because of the random nature of these subject’s exclusion. Fourth, the study participants, mainly recruited from previous studies and from the thoracic outpatient clinic of Haukeland University Hospital, may have had a more severe disease burden than the COPD patients in the general population. However, all GOLD stages were represented. Fifth, CaSc is a surrogate marker for atherosclerosis in the coronary arteries, associated with increased risk for coronary artery stenosis and CHD. A CaSc of zero does not fully exclude the presence of significant coronary stenosis in patients with chest pain syndrome [40], a state that can be difficult to distinguish in a patient with COPD and CHD. CaSc also identifies coronary artery lesions containing calcium. However, identifying non-calcified lesions is somewhat limited [41]. It is possible that a CaSc of zero might underestimate the overall coronary lesion burden, but the majority of the patients included in our study underwent both CaSc and CCTA for coronary stenosis evaluation.

In conclusion, the current study confirms the added risk for CHD among COPD patients, but despite the extensive characterization of the study cohort, did not identify specific phenotypes at risk. Rather it points to the importance of having a high degree of suspicion for CHD even in COPD patients with mild lung function impairment.

10 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Med vennlig hilsen/ Best regards!

Kjell Torén, MD, PhD

Academic Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Svedsen and colleagues analyse the presence of coronary artery plaques and calcium score, respectively, in patients with COPD compared with controls without CPOD in a Norwegian cohort, with focus of analysis on potential risk factors for these outcomes among the patients with COPD. As outlined nicely in the paper, this topic is of interest given the large co-occurrence of cardiovascular disease and COPD and the relevance for morbidity and prognosis. Reading the paper, I had some comments and suggestions for strengthening the paper:

- A general comment is that the analysis in fact focuses on a range of potential predictors of the outcomes (stenosis and calcium score) - not only COPD phenotype. Suggest that is reflected in the title ('Factors associated with' or 'Predictors of ...' instead of COPD phenotype), especially as many of the findings not pertain closely to phenotype, rather factors more broadly. It also takes a while in Methods before the phenotype variables are introduced. Please revise (and clarify) the aim accordingly (now the aim is rather vague and differ in Abstract and the main text).

- Need to be clearer on the limitation in terms of sample size / power. Many well established risk factors for cardiovascular disease (CVD) did not "fall out" in the analysis -- however, this is not to be interpreted as these factors are not risk factors in COPD patients (COPD patients are "spared") - when, in fact, it is the other way around (not implying this is stated by the authors, it is not), but important that this limitation is made clear. Insufficient power likely to explain the absence of a lot of expected associations? - that also is a prolem for the relevance of the analysis as a whole, but the analysis is still valuable as it is well done and could be combined with other datasets for meta-analyses

- Factors in the models: please state how adjustment factors as well as factors to evaluate in the model were selected.

- Intro: states that COPD severity is graded based on lung function -- not sure that statement is entirely correct any more, please revise

- Aim: as mentioned above, make the aims similar and specific - the aim in the paper is not good - "... whether different COPD patients had an increased risk..." needs revision in terms of scientific contents and language (surely different patients will have different risks, it would be surprising if all COPD patients had the same risk, etc) - evaluate factors associated with... ?

- Methods: please state which patients that were included in the used databases - now it only says that 16 asthma patients were included (which was confusing at the first read); inclusion and exclusion criteria. Suggest to move the data on actually included number of participants to the start of Results.

- Why was the CaSc threshold set to 100 for the outcome?

- Tables: no need to include both the "yes" and the "no" categories for binary variables

- Methods and Results: how were the models specified to begin with (in terms of factors evaluated and adjusted for) and how were the final factors to report determined?

- How many had a SaSc >500 (and were excluded from CTA)? Potential influence on findings?

- Table 4: "for the chance of finding..." - please revise and specify

- One wonders how good CaSc is as outcome - how was the correlation within the outcomes (CaSc with stenosis)? Would be good with more data on that, including in Methods as rationale for the cut-off and use of that outcome

- Please revise the sentence "Since this is a cross-sectional study, we can only speculate", as well as the sentences with "which can easily be interpred as", and "makes it uncertain to establish"

Reviewer #2: The main purpose of this study was to investigate it if different COPD characteristics or phenotypes are associated with an increased risk of coronary heart disease. The conceptualization is really interesting and the overall planning of the study is sound and clear. I have no major concerns but a number of minor comments mainly about design and analyses.

1. Design and data collection

The design and data collection could be more clearly described. The title says case control study, but is it really? The study has two parts, the first investigating the prevalence of coronary atherosclerosis in patients with COPD and non-COPD controls and the second investigating the associations of specific COPD characteristics with the dependent variable of CHD. However, it is not like the groups are defined from the outcome CHD and compared about the risk factors and it is unclear how the controls were chosen. The authors refer to the original papers of the involved cohorts, but it is unclear from the present paper if matching of cases and controls have been performed. As for the second part, it is unclear if the independent variables are collected at the same time as the outcome variable or not. The paper would benefit a lot from a flow chart clarifying the design, timing, the contributions from different other cohorts, and the attritions.

2. Comparison of CHD in COPD and non-COPD

The first major conclusion is that COPD patients are more likely to have objective measures of CHD. This is not surprising as the COPD patients are older, have more pack years and more diabetes. Have you considered comparing the groups by propensity score matching?

3. Logistic regression

In the second part, logistic regression uses different COPD characteristics as independent variables.

How were the explanation variables chosen? A priori based on subject knowledge matter or based on unadjusted analyses?

I find the research question very relevant, but it would have been even more interesting with more COPD characteristics as markers for systemic inflammation or other phenoptypes. Was chronic bronchitis or eosinophilic COPD investigated as independent variables? Emphysema was defined as 10% of the lungs or not, what happens if the analyses are repeated using emphysema a continuous score as independent variable? And was both hypoxemia and hypercapnia analysed? ACO? Non-smoking COPD? COPD with early onset or exposure during childhood/premature birth?

It would have been of great interest if the analyses did not only adjust for different characteristics/phenotypes, but also included stratification and interaction analyses to see if there is an effect modification by phenotype, ie- if factors associated with CHD differs between different phenotypes.

4. Power analysis

The second major conclusion is that no specific COPD characteristics were independently associated with a higher risk for CHD. Could this be a type 2 error? Was there a power analysis performed to ensure a reasonable number of participants?

5. Attrition

Was there any attrition analysis?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

27 Feb 2022

We have uploaded a word document with our point - to point responses to the two reviewers.

Submitted filename: Response letter draft.docx

Click here for additional data file.

7 Mar 2022

Factors associated with coronary heart disease in COPD patients and controls

PONE-D-21-35460R1

Dear Dr. Eagan,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Kjell Torén, MD, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Dear dr Eagan!

I think you and your co-authors satisfactory have addressed the comments by the reviewers. When you do not agree, you have argued quite convincingly for your sake.

Hence, the manuscript is accepted.

29 Mar 2022

PONE-D-21-35460R1

Factors associated with coronary heart disease in COPD patients and controls

Dear Dr. Eagan:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Kjell Torén

Academic Editor

PLOS ONE