Cardiovascular disease is common in patients with chronic obstructive pulmonary disease (COPD) but often remains unrecognized. Ischemic electrocardiographic (ECG) changes are associated with a higher risk of dying from coronary heart disease but have never been systematically evaluated in COPD. Also, their relation to clinical outcome has not been studied. We aimed to determine the frequency of ischemic ECG changes and its relevance in relation to clinical outcome and predictors of impaired survival in patients with COPD. Clinical characteristics, pulmonary function, and co-morbidities were assessed in 536 patients with COPD during baseline assessment of a comprehensive pulmonary rehabilitation program. Moreover, electrocardiograms at rest were obtained in all patients. All electrocardiograms were scored independently by 2 cardiologists using the Minnesota scoring system. Major or minor Q or QS pattern, ST junction and segment depression, T-wave items, or left bundle branch block were considered ischemic ECG changes. One hundred thirteen patients (21%) had ischemic ECG changes. Moreover, 42 of 293 patients (14%) without self-reported cardiovascular co-morbidities had ischemic ECG changes. In addition, patients with ischemic ECG changes had higher dyspnea grades (Modified Medical Research Council (mMRC) 2.9 ± 1.1 vs 2.6 ± 1.1, p = 0.032), worse exercise performance (6-minute walking distance 387 ± 126 vs 425 ± 126 m, p = 0.004), more systemic inflammation (high-sensitivity C-reactive protein 11.2 ± 16.2 vs 7.9 ± 10.7 mmol/l, p = 0.01), higher scores on the Charlson Co-morbidity Index (1.8 ± 0.9 vs 1.5 ± 0.8 points), and higher scores BODE (5.3 ± 3.7 vs 4.5 ± 3.4 points, p = 0.033) and on ADO indexes (5.2 ± 1.7 vs 4.8 ± 1.7 points, p = 0.029) compared to patients without ischemic ECG changes, whereas forced expiratory volume in the first second was similar (40.8 ± 15.2% vs 42.6% ± 15.9%, p = 0.30). In conclusion, ischemic ECG changes are common in patients with COPD and associated with poor clinical outcome irrespective of forced expiratory volume in the first second. These results suggest an important role for cardiovascular disease in impaired survival in these patients.
Patients with chronic obstructive pulmonary disease (COPD) have an increased risk for cardiovascular morbidity and mortality. This risk is related to disease severity and the presence of low-grade systemic inflammation but independent of smoking status or other well-known risk factors for atherosclerosis. In patients with stable COPD cardiovascular co-morbidity often remains undiagnosed and untreated. Also, the impact of cardiovascular disease on clinical outcomes in COPD is currently unclear. In this study we systematically coded electrocardiograms of patients with stable but disabling COPD using Minnesota criteria. The first aim was to determine the frequency of ischemic-like electrocardiographic (ECG) changes (ischemic electrocardiograms). In addition, differences in clinical outcomes (pulmonary function, body composition, exercise capacity, dyspnea, systemic inflammation, co-morbidities) between patients with and without ischemic ECG findings were studied. Relations of ischemic ECG findings to clinical predictors of impaired survival in COPD (updated BODE and ADO indexes ) were investigated.
Methods
Data were collected and analyzed retrospectively in 543 consecutive patients with COPD. These retrospective analyses were exempt from institutional review board approval because of the use of de-identified pre-existing data. Patients were entering a comprehensive pulmonary rehabilitation program from January 2005 through March 2007 at the dedicated pulmonary rehabilitation center of CIRO+ in Horn, The Netherlands. Patients were referred by respiratory physicians in southeast Netherlands. To avoid the influence of concurrent infection on clinical outcomes, 7 patients with high-sensitivity C-reactive protein (hs-CRP) >100 mmol/L were excluded. Smoking status, smoked pack-years, and use of long-term oxygen therapy were recorded. Furthermore, patients’ current pharmacologic cardiovascular treatment (diuretics, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, β blockers, calcium antagonists, nitrates, digitalis, antiarrhythmics, statins, antiaggregants, or coumarins) and pulmonary treatment (short-acting β 2 -agonists, long-acting β 2 -agonists, short-acting anticholinergics, long-acting anticholinergics, inhaled corticosteroids, oral corticosteroids, theophylline, and leukotrienes) were registered.
Electrocardiograms at resting were obtained in all patients. ECG data were recorded on analog tape and included the conventional 12 leads (Mac1600, GE Healthcare, Munich, Germany). Paper tracings were produced and scored independently by 2 cardiologists (N.H.M.K.U.-L. and A.P.G.) according to the Minnesota Code ECG rules. In case of disagreement consensus was achieved in conjunction. Ischemia on electrocardiogram was defined as ≥1 of the following: major or minor Q or QS pattern, ST junction and segment depression, T-wave items, or left bundle branch block. These patterns are often seen as indications of silent myocardial ischemia. The Minnesota Codes used in this report are listed in Table 1 .
| ECG Pattern | Minnesota Codes | All (n = 536) |
|---|---|---|
| Major Q or QS pattern | 1.1 | 23 (4.3%) |
| Minor Q or QS pattern | 1.2 + 1.3 | 15 (2.8%) |
| ST junction (J) and segment depression | 4.1 + 4.2 | 19 (3.5%) |
| T-wave items | 5.1–5.3 | 47 (8.8%) |
| Left bundle branch block | 7.1 | 11 (2.1%) |
| Ischemia | all above | 113 (21%) |
| Right bundle branch block | 7.2 | 35 (6.5%) |
| Atrium fibrillation or flutter | 8.3 | 13 (2.4%) |
| Atrioventricular conduction defect, grade 1 | 6.3 | 5 (0.9%) |
| Left-axis deviation | 2.1 | 52 (9.7%) |
| High-amplitude R waves | 3.1 + 3.2 | 8 (1.5%) |
In every patient the Charlson Co-morbidity Index was calculated. Apart from cardiovascular features that are included in the Charlson Co-morbidity Index, namely myocardial infarction, chronic heart failure, cerebrovascular disease, and peripheral arterial disease, other cardiovascular co-morbidities were registered: hypertension, arrhythmia, angina pectoris, coronary heart disease as defined by coronary surgery or percutaneous coronary intervention, and abdominal aortic aneurysm repair. “Any cardiovascular co-morbidity” was defined as ≥1 of these cardiovascular co-morbidities.
Lung function parameters (forced expiratory volume in the first second and forced vital capacity) were collected using standardized postbronchodilator spirometry. Carbon monoxide transfer factor was determined with a single breath-hold method (MasterScreen Body, Carefusion, Höchberg, Germany). Degree of airflow limitation was defined according to the classification of the Global Initiative for Chronic Obstructive Lung Disease.
Arterial oxygen tension at rest and arterial carbon dioxide tension were analyzed with a blood gas analyzer (ABL 330, Radiometer, Copenhagen, Denmark). Hs-CRP was assessed in duplicate (COBAS-MIRA, Radiometer) on a venous blood sample collected from all patients in the fasted state.
Functional exercise capacity was measured with the 6-minute walking distance performed according to published guidelines including a practice walk.
Body composition was assessed by body mass index calculated as body weight divided by height squared and fat-free mass index was calculated as fat-free mass divided by height ; the latter was measured by bioelectrical impedance (Bodystat, Douglas, United Kingdom).
Self-perceived dyspnea in relation to physical disability was assessed using the modified Medical Research Council dyspnea scale.
The updated BODE and ADO indexes are predictors of 3-year mortality in patients with COPD. The updated BODE index is based on scores on body mass index, forced expiratory volume in the first second, modified Medical Research Council dyspnea scale, and 6-minute walking distance. The ADO index is based on age, forced expiratory volume in the first second, and modified Medical Research Council dyspnea scale. A higher score means a worse prognosis.
All statistics were performed using SPSS 17.0 (SPSS, Inc., Chicago, Illinois). Results are expressed as mean ± SD unless indicated otherwise. Independent-samples Student’s t test was used to investigate differences in clinical outcomes between patients with and without ischemic electrocardiogram. Differences between Global Initiative for Chronic Obstructive Lung Disease stages were investigated by 1-way analysis of variances using Fisher’s least significant difference post hoc test. Differences in the proportion of patients with ischemic electrocardiogram between different Global Initiative for Chronic Obstructive Lung Disease stages were assessed using chi-square test. Relation between ischemic electrocardiogram and functional outcome parameters adjusted for forced expiratory volume in the first second was assessed by performing logistic regression analysis with the presence or absence of ischemic electrocardiogram and forced expiratory volume in the first second as predictors and functional outcome parameters as dependent variables. All dependent variables were entered as continuous variables by a forced-entry method. Colinearity analysis of the variables ischemic electrocardiogram and forced expiratory volume in the first second revealed a variance inflation factor <10 indicating absence of colinearity. A p value <0.05 was considered statistically significant. For planned stratification and comparison to a population study, age was divided into 2 categories that included most patients: 55 to 64 and 65 to 74 years. Other age groups were not calculated because of small numbers of subjects and misbalance in gender.
Results
Baseline characteristics of 536 elderly patients with moderate to very severe COPD are listed in Table 2 . On average patients had normal body composition, impaired functional exercise performance, and had moderate to severe dyspnea during daily life. Almost 1/3 of patients had hs-CRP >10 mmol/L, 1/4 were current smokers, 42% had ≥1 self-reported co-morbidity, and >1/2 of patients was on ≥1 cardiovascular drug.
| Men | 330 (62%) |
| Age (years) | 63.7 ± 9.4 |
| Body mass index (kg/m 2 ) | 25.1 ± 5.1 |
| Fat-free mass index (kg/m 2 ) | 16.3 ± 2.3 |
| Pack-years | 39.8 ± 19.6 |
| Current smokers | 139 (26%) |
| Long-term oxygen therapy | 50 (9%) |
| Forced expiratory volume in first second (percent predicted) | 42.2 ± 15.7 |
| Ratio of forced expiratory volume in first second to forced vital capacity (%) | 41.0 ± 11.5 |
| Diffusion lung capacity for carbon monoxide (percent predicted) | 50.9 ± 17.9 |
| 6-minute walking distance (meters) | 417 ± 127 |
| Partial pressure of carbon dioxide—arterial carbon dioxide tension (kPa) | 5.4 ± 0.8 |
| Partial pressure of oxygen—arterial oxygen tension (kPa) | 9.2 ± 1.4 |
| High-sensitive C-reactive protein (mmol/L) | 8.6 ± 12.1 |
| High-sensitive C-reactive protein <10 mmol/L | 386 (72%) |
| High-sensitive C-reactive protein ≥10 mmol/L | 150 (28%) |
| Modified Medical Research Council dyspnea scale (points) | 2.7 ± 1.1 |
| Charlson Co-morbidity Index (points) | 1.6 ± 0.9 |
| ≥2 points on Charlson Co-morbidity index | 225 (42%) |
| Any cardiovascular co-morbidity | 243 (45%) |
| Use of cardiovascular medication | 303 (57%) |
Frequencies of different ECG changes are listed in Table 1 . In total 21% of patients had an ischemic electrocardiogram. The proportion of patients with ischemic electrocardiogram was similar after stratification for Global Initiative for Chronic Obstructive Lung Disease stages: 23 of 124 patients (19%) in stage 1/2, 39 of 176 patients (22%) in stage 3, and 51 of 236 patients (22%) in stage 4. Frequency of ischemic electrocardiogram tended to be higher in women but the difference was not significant ( Table 3 ).
| Electrocardiogram Without Ischemic Changes (n = 423) | Ischemic Electrocardiogram (n = 113) | p Value | |
|---|---|---|---|
| Age (years) | 63.3 ± 9.3 | 64.9 ± 9.5 | 0.120 |
| Men | 269 (64%) | 61 (54%) | 0.062 |
| Body mass index (kg/m 2 ) | 25.0 ± 5.1 | 25.3 ± 5.1 | 0.610 |
| Fat-free mass index (kg/m 2 ) | 16.3 ± 2.3 | 16.3 ± 2.2 | 0.980 |
| Pack-years | 40.3 ± 19.9 | 38.1 ± 18.5 | 0.280 |
| Current smoker | 116 (27%) | 23 (20%) | 0.038 |
| Long-term oxygen therapy | 41 (10%) | 9 (8%) | 0.575 |
| Forced expiratory volume in 1 second (percent predicted) | 42.6 ± 15.9 | 40.8 ± 15.2 | 0.300 |
| Ratio of forced expiratory volume in 1 second to forced vital capacity (%) | 41.2 ± 11.4 | 40.4 ± 11.7 | 0.540 |
| Diffusion lung capacity for carbon monoxide (percent predicted) | 50.57 ± 17.5 | 52.26 ± 19.3 | 0.800 |
| 6-minute walking distance (meters) | 425 ± 126 | 387 ± 126 | 0.004 |
| 6-minute walking distance <350 m | 109 (26%) | 40 (35%) | 0.042 |
| Partial pressure of carbon dioxide—arterial carbon dioxide tension (kPa) | 5.4 ± 0.8 | 5.5 ± 0.9 | 0.340 |
| Partial pressure of oxygen—arterial oxygen tension (kPa) | 9.3 ± 1.4 | 9.2 ± 1.4 | 0.960 |
| High-sensitive C-reactive protein (mmol/L) | 7.9 ± 10.7 | 11.2 ± 16.2 | 0.010 |
| High-sensitive C-reactive protein <10 mmol/L | 314 (74%) | 72 (64%) | 0.027 |
| High-sensitive C-reactive protein ≥10 mmol/L | 109 (26%) | 41 (36%) | |
| Modified Medical Research Council dyspnea scale (points) | 2.6 ± 1.1 | 2.9 ± 1.1 | 0.032 |
| Update BODE index (points) | 4.5 ± 3.4 | 5.3 ± 3.7 | 0.033 |
| ADO index (points) | 4.8 ± 1.7 | 5.2 ± 1.7 | 0.029 |
| Charlson Co-morbidity Index (points) | 1.5 ± 0.8 | 1.8 ± 0.9 | 0.002 |
| Charlson Co-morbidity Index ≥2 | 165 (39%) | 60 (53%) | 0.007 |
| Myocardial infarction | 28 (7%) | 25 (22%) | <0.001 |
| Cerebrovascular disease | 24 (6%) | 14 (12%) | 0.012 |
| Chronic heart failure | 15 (4%) | 13 (12%) | 0.001 |
| Peripheral arterial disease | 40 (10%) | 16 (14%) | 0.147 |
| Operated abdominal aortic aneurysm | 10 (2%) | 8 (7%) | 0.013 |
| Invasively treated coronary disease ⁎ | 34 (8%) | 25 (22%) | <0.001 |
| Angina pectoris | 18 (4%) | 6 (5%) | 0.630 |
| Treated hypertension | 51 (12%) | 22 (20%) | 0.041 |
| Atrial fibrillation or other arrhythmia | 47 (10%) | 21 (19%) | 0.034 |
| Any cardiovascular co-morbidity | 172 (41%) | 71 (63%) | <0.001 |
| Use of cardiovascular medication | 219 (52%) | 84 (74%) | <0.001 |
⁎ Percutaneous transluminal coronary angioplasty or coronary artery bypass grafting.
The proportion of ischemic electrocardiograms was higher in subjects with self-reported cardiovascular co-morbidity. They had more frequently a major Q or QS pattern and a left bundle branch block. There were no differences in minor Q or QS patterns or ST-segment or T-wave items. Also, no differences were seen in other nonischemic ECG items in subjects with and without reported cardiovascular co-morbidity except for atrial fibrillation, which was (if reported) included as a cardiovascular co-morbidity ( on-line supplement Table S1 ). Patients with ischemic electrocardiogram had higher scores on the Charlson Co-morbidity Index and its cardiovascular parts, more frequently had hypertension and atrial fibrillation, and were treated more frequently for coronary disease or abdominal aortic aneurysm ( Table 3 ).
The proportion of patients with ischemic electrocardiogram was similar after stratification for age in groups 55 to 64 and 65 to 74 years of age. Within the 2 age groups the frequency of ischemic electrocardiogram tended to be higher in women but did not reach statistical significance ( Figure 1 ).
