Although preoperative risk assessment for coronary artery bypass grafting (CABG) has been evaluated with multiple predictive models, none have incorporated a low level of cardiorespiratory fitness, which represents one of the strongest predictors of all-cause and cardiovascular mortality in subjects with and without heart disease. The aim of the present study was to evaluate preoperative cardiorespiratory fitness, expressed as METs (1 MET = 3.5 ml O 2 /kg/min), and short-term morbidity and mortality after CABG. The Society of Thoracic Surgeons database was queried for patients who underwent CABG from January 2002 to December 2010 at Beaumont Health Systems. Electronic medical records were reviewed for peak or symptom-limited exercise testing <90 days before CABG. Peak METs were estimated from the achieved treadmill speed, grade, and duration or the cycle ergometer workload, corrected for body weight. Patients who met eligibility criteria (n = 596) were categorized into 2 groups: those with reduced aerobic capacity (<5 METs [n = 78]) and those achieving ≥5 METs (n = 518). Fisher’s exact tests were used to compare preoperative aerobic capacity and short-term postoperative morbidity and mortality between the 2 groups. After adjusting for potential confounding variables, an inverse relation was found between cardiorespiratory fitness and complications after CABG. Specifically, low preoperative cardiorespiratory fitness (<5 METs) was associated with higher operative and 30-day mortality after CABG (p <0.05). In conclusion, these data suggest that preoperative cardiorespiratory fitness provides an independent and additive marker for mortality after CABG.
Preoperative risk assessment for coronary artery bypass grafting (CABG) has been evaluated using multiple predictive models (the European System for Cardiac Operative Risk Evaluation score, the Society of Thoracic Surgeons mortality risk score, the University Health System Consortium risk adjustment model, and the SYNTAX Score), often yielding markedly different levels of risk for any given patient. These findings highlight the need for a more precise delineation of preoperative predictors of short-term complications and prognosis after CABG, including data that may be obtained as part of the preliminary screening for ischemic heart disease. Exercise tolerance or, more specifically, cardiorespiratory fitness (i.e., aerobic capacity) expressed as peak METs (1 MET = 3.5 ml O 2 /kg/min), is one of the strongest and most consistent prognostic markers in subjects with and without coronary heart disease (CHD) and/or other co-morbid conditions (e.g., overweight or obesity, systemic hypertension, type 2 diabetes, metabolic syndrome). Reduced cardiorespiratory fitness levels are also associated with increased morbidity after abdominal aortic aneurysm repair, liver transplantation, noncardiac thoracic surgery, and major abdominal surgery, as well as extended lengths of hospital stay, 30-day readmission rates, and short-term complications after bariatric surgery. Exercise capacity <5 METs generally indicates a less favorable long-term prognosis and higher subsequent mortality, regardless of the underlying extent of CHD. Whether this sign- and symptom-limited fitness demarcation has applicability to risk-stratifying patients who undergo CABG remains unclear. In the present study, we sought to evaluate the relation, if any, between preoperative cardiorespiratory fitness, expressed as peak METs, and associated exercise test responses, on postoperative complications soon after CABG.
Methods
The Society of Thoracic Surgeons General Thoracic Surgery Database was queried for patients who underwent CABG from January 2002 through December 2010 at William Beaumont Hospitals in Royal Oak and Troy, Michigan. Patients who underwent simultaneous interventions such as valve surgery, formal atrial ablative procedures for atrial fibrillation, or aortic surgery were excluded. Eligible patients were queried through William Beaumont Hospital’s electronic medical records for preoperative peak or symptom-limited exercise testing performed <90 days before coronary revascularization surgery. In total, 596 patients were identified who met these criteria. The Institutional Review Board at Beaumont Health System approved the study and retrospective data analysis.
Patients with known or suspected CHD, without significant musculoskeletal or orthopedic limitations, underwent peak or symptom-limited exercise testing using the conventional or modified Bruce treadmill protocol. Laboratory staff members routinely instructed all patients to simply “rest” their palms on the treadmill handrail for balance and support and to avoid tight gripping and/or pressing the hands against the bar to attenuate the potential overestimation of cardiorespiratory fitness (peak METs). Patients who were unable to negotiate treadmill walking (n = 8) underwent cycle ergometer testing (Monark Exercise AB, Vanberg, Sweden) or dual-action arm-leg ergometry using a commercially available ergometer (Airdyne; Schwinn, Vancouver, Washington), using the levers and pedals simultaneously. Heart rate and blood pressure were measured at rest in supine and standing positions, during each 3-minute stage of exercise and throughout a 6-minute recovery. An electrocardiogram was monitored continuously by oscilloscope (Q-Stress II; Quinton, Seattle, Washington), with 3-channel (leads V 1 , V 5 , and aVF) recordings obtained throughout the exercise test and 12-lead electrocardiograms (1 mV/10 mm) recorded at the end of each stage and during peak or maximal exercise. Perception of the intensity of physical effort at submaximal and maximal exercise was obtained using the Borg category scale. Test termination criteria included the following: patient request, volitional fatigue, increasing chest (≥2/4) or leg pain, electrocardiographic abnormalities (≥2-mm ST-segment depression and/or threatening ventricular arrhythmias [≥3 consecutive premature ventricular contractions]), and a hyper- or hypotensive blood pressure response.
Documented exercise test responses included estimated METs, 1-minute recovery heart rate, peak double product, exertional chest pain, significant ST-segment depression (≥1 mm in 2 contiguous leads), percentage of predicted maximum heart rate achieved (220 minus age in years) to estimate maximal heart rate, and exertional hypotension. Peak METs were estimated from the achieved treadmill grade and speed, as well as the duration of the workload attained. If cycle ergometry was performed, peak METs were estimated from the highest achieved workload, expressed as kilogram-meters per minute or watts (1 W = ∼6 kg · m/min), corrected for body weight. Abnormal heart rate recovery was defined as a decrease in the heart rate of ≤12 beats/min at 1 minute into walking recovery. Double product was calculated as peak systolic blood pressure times peak heart rate, divided by 100. Duke treadmill score was calculated using time (minutes) achieved on the conventional Bruce treadmill protocol, angina index, and maximum ST-segment deviation during the test, where Duke treadmill score = exercise time−(5 × ST-segment displacement)−(4 × angina index).
Preoperative risk factors were defined through the Society of Thoracic Surgeons General Thoracic Surgery Database guidelines. A documented history of hypertension was defined as treatment for high blood pressure and/or documentation, on ≥2 previous occasions, of rest systolic blood pressure >140 mm Hg or diastolic pressure >90 mm Hg. Dyslipidemia was listed as a risk factor in patients who were taking prescribed medications for lipid or lipoprotein abnormalities or if total cholesterol was >200 mg/dl, low-density lipoprotein >130 mg/dl, or high-density lipoprotein <40 mg/dl. Chronic obstructive pulmonary disease (COPD) was defined by the percentage of predicted forced expiratory volume at 1 second or use of a bronchodilator before surgery. Preoperative congestive heart failure required documented signs or symptoms within 2 weeks of CABG. Peripheral arterial disease was defined as a history of lower extremity atherosclerotic disease, claudication, amputation, previous vascular surgery, or positive noninvasive testing.
Postoperative events were those defined by the Society of Thoracic Surgeons General Thoracic Surgery Database guidelines. Early mortality was defined as death during the hospitalization or <30 days after CABG. Adverse outcome measure selection was on the basis of clinical judgment. Death at discharge or <30 days after surgery, pneumonia, sternal wound infection, initial ventilation beyond 48 hours, reintubation, atrial or ventricular arrhythmias requiring treatment, reoperation for bleeding, and new central neurologic events were considered adverse postoperative outcomes.
Patients were divided into those with reduced aerobic capacity (<5 METs) and those achieving ≥5 METs during preoperative peak or symptom-limited exercise testing. Univariate comparisons were made of categorical data using Pearson’s chi-square tests as appropriate (expected frequency >5); otherwise, Fisher’s exact tests were used. Baseline continuous characteristics are expressed as mean ± SD or medians with percentiles as appropriate and were evaluated using Wilcoxon’s rank tests. All hypothesis testing was 2 tailed. Multivariate logistic regression was performed for significant adverse outcomes identified by univariate analysis. The baseline variables used were age, gender, peripheral arterial disease, estimated glomerular filtration rate, the ejection fraction, COPD, and congestive heart failure. All analyses were performed using SAS for Windows version 9.2 (SAS Institute Inc., Cary, North Carolina). A p value <0.05 was considered statistically significant.
Results
The study sample had a mean age of 62.8 ± 10 years (490 men [82.2%] and 106 women [17.8%]). The mean body mass index for the entire study population (n = 596) was 29.3 ± 5 kg/m 2 ; average body mass index values for patients with reduced aerobic capacity (<5 METs [n = 78]) and those achieving ≥5 METs (n = 518) were 29.0 ± 6.1 and 29.4 ± 4.8 kg/m 2 , respectively. Figure 1 shows the estimated METs achieved during peak or symptom-limited exercise stress testing in our patient population <90 days before CABG. Cardiorespiratory fitness ranged from 2 to 16 METs, with most patients attaining ≥7 METs. Patients were categorized as those with reduced aerobic capacity (<5 METs [n = 78]) and those achieving ≥5 METs (n = 518). Baseline characteristics of the 2 study groups are listed in Table 1 . Patients with low preoperative cardiorespiratory fitness (<5 METs) were significantly (p <0.05) older; were more often women; had a higher prevalence of COPD, peripheral arterial disease, and congestive heart failure; and had lower ejection fractions.
| Variable | METs <5 (n = 78) | METs ≥5 (n = 518) | p Value |
|---|---|---|---|
| Age (yrs) | 69 ± 10 (71) | 62 ± 10 (62) | <0.0001 |
| Men | 53 (68%) | 437 (84%) | 0.0004 |
| Body mass index (kg/m 2 ) | 29.0 ± 6.1 | 29.4 ± 4.8 | 0.34 |
| Diabetes | 29 (37%) | 180 (35%) | 0.67 |
| Cochran-Gault estimated glomerular filtration rate (ml/min/m 2 ) | 75 ± 27 (71) | 84 ± 31 (81) | 0.002 |
| Last creatinine (mg/dl) | 1.25 ± 1.17 (1.10) | 1.06 ± 0.35 (1.00) | 0.11 |
| End-stage renal disease dialysis | 1 (1.3%) | 2 (0.4%) | 0.34 |
| Hypertension | 65 (83%) | 417 (81%) | 0.55 |
| COPD | 17 (21.8%) | 67 (12.9%) | 0.036 |
| None | 61 (78.2%) | 451 (87.1%) | 0.055 |
| Mild | 11 (14.1%) | 53 (10.2%) | |
| Moderate | 5 (6.4%) | 11 (2.1%) | |
| Severe | 1 (1.3%) | 3 (0.6%) | |
| Peripheral arterial disease | 21 (27%) | 64 (12%) | 0.0006 |
| Cerebrovascular disease | 0 | 6 (1.2%) | 1.00 |
| Hypercholesterolemia or dyslipidemia | 58 (74%) | 406 (78%) | 0.43 |
| Current smokers | 39 (50%) | 225 (43%) | 0.28 |
| Known coronary artery disease | 10 (13%) | 98 (19%) | 0.19 |
| Recent myocardial infarction (<21 days) | 9 (11.5%) | 29 (5.6%) | 0.08 |
| Congestive heart failure | 9 (11.5%) | 24 (4.6%) | 0.028 |
| Cardiogenic shock | 1 (1.3%) | 6 (1.2%) | 1.00 |
| Arrhythmia | 2 (2.6%) | 3 (0.6%) | 0.13 |
| Ejection fraction (%) | 51 ± 11 (50) | 54 ± 11 (55) | 0.021 |
| Time on pump during CABG | 72.9 ± 25.1 (72) | 70.3 ± 21 (70) | 0.33 |
| Number of vessels bypassed | 3.5 ± 1.0 (4.0) | 3.6 ± 1.0 (4.0) | 0.49 |
| CABG status | 0.27 | ||
| Elective | 24 (30.8%) | 194 (37.5%) | p value for emergent vs others = 0.43 |
| Urgent | 53 (68.0%) | 321 (62.0%) | |
| Emergent | 1 (1.3%) | 3 (0.6%) |
Table 2 and Figure 2 provide surgical outcomes data according to estimated METs. Univariate analysis demonstrated statistically significant differences in prolonged postoperative ventilation and operative and 30-day mortality ( Table 3 ). Backward-selection multivariate analysis was performed on prolonged ventilation. Only METs <5 remained in the final model, with an odds ratio of 2.13 (95% confidence interval 1.01 to 4.55, p = 0.047, C-statistic = 0.55). Stepwise selection was used for mortality at 30 days given the number of total events. Models with METs <5 and 1 additional covariate were examined; a model with METs <5, the ejection fraction, and COPD was evaluated, and COPD was not significant. There were not enough events to complete multivariate analyses on sternal wound infection. After adjusting for the ejection fraction, preoperative METs <5 was a statistically significant predictor of operative (odds ratio 6.75, 95% confidence interval 1.73 to 26.4, p = 0.006) and 30-day mortality (odds ratio 5.60, 95% confidence interval 1.50 to 20.8, p = 0.01).
| Variable | METs <5 (n = 78) | METs ≥5 (n = 518) | p Value |
|---|---|---|---|
| Sternal wound infections | 2 (2.6%) | 0 | 0.017 |
| Neuropermanent stroke | 2 (2.6%) | 8 (1.5%) | 0.63 |
| Prolonged ventilation | 10 (12.8%) | 34 (6.6%) | 0.049 |
| Atrial or ventricular arrhythmia | 0 | 15 (2.9%) | 0.24 |
| Reoperation because of bleeding or tamponade | 3 (3.9%) | 8 (1.5%) | 0.16 |
| Operative mortality | 4 (5.1%) | 5 (1.0%) | 0.021 |
| 30-day mortality | 4 (5.1%) | 6 (1.2%) | 0.031 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
