(Chest. 2005;128:42-47.)
© 2005
American College of Chest Physicians
Stress Recovery Index for Risk Stratification of Asymptomatic Patients Following Coronary Bypass Surgery*
Riccardo Bigi, MD;
Dario Gregori, MA, PhD;
Lauro Cortigiani, MD;
Paola Colombo, MD and
Cesare Fiorentini, MD
* From the Department of Medical and Surgical Sciences (Dr. Bigi), University School of Medicine and "A. De Gasperis" Foundation, Milan; Department of Public Health and Microbiology (Dr. Gregori), University of Turin, Turin; Cardiovascular Unit (Dr. Cortigiani), Campo di Marte Hospital, Lucca; Cardiothoracic Department (Dr. Colombo), Niguarda Cà Granda Hospital, Milan; and Cardiology (Dr. Fiorentini), Department of Medical and Surgical Sciences, University School of Medicine, "S. Paolo" Hospital, Milan, Italy.
Correspondence to: Riccardo Bigi, MD, Cardiology, Department of Medical and Surgical Sciences, "S. Paolo" Hospital, Via A. di Rudinì 8 - 20142 Milano, Italy; e-mail: Riccardo.Bigi{at}unimi.it
 |
Abstract
|
|---|
Objective: To prospectively assess the prognostic value of the stress recovery index (SRI) following coronary bypass surgery.
Design and patients: Two hundred seventy-eight patients who had undergone coronary bypass surgery and participated in a secondary prevention program were exercise tested and prospectively followed up for a median of 36 months. Cardiac death, nonfatal infarction, and need for further revascularization were target end points. SRI, defined as the difference in absolute values between the area of heart rate-adjusted ST-segment depression during exercise and recovery, was derived in all. Clinical data, resting ejection fraction, and exercise testing data of patients were entered into a sequential Cox model; SRI was entered last. Model validation was performed by bootstrap adjusted by the degree of optimism in estimates. Survival curves were set up using Kaplan-Meier method and compared by the log-rank test.
Results: SRI was the only significant and independent prognostic indicator (hazard ratio, 0.68; 95% confidence interval, 0.53 to 0.89) and increased the prognostic power of the model on top of clinical and exercise testing variables, as demonstrated by the significant (p = 0.01) increase of the area under the receiver operating characteristic curve of the risk function. Survival analysis showed ascending SRI quartiles to identify a significant (p = 0.001) increase in event-free survival.
Conclusions: SRI is of value in predicting outcome after coronary bypass surgery and provides additional prognostic information over clinical and exercise testing data.
Key Words: bypass surgery • coronary artery disease exercise • ECG • risk stratification • testing
|
Introduction
|
|---|
The identification of silent graft disease is a relevant clinical problem in patients who had undergone coronary bypass surgery, especially with the use of venous conduits. Although a negative exercise ECG test result that was positive before the operation generally identifies successful revascularization,1 exercise ECG in asymptomatic patients has poor prognostic value in predicting future cardiac events.2 This causes frequent recourse to stress-imaging modalities.
Several years ago, the heart rate adjustment of ST-segment depression was proposed to improve the accuracy of exercise ECG.34 More recently, the stress recovery index (SRI), based on the comparative analysis of heart rate-adjusted ST-segment modification during exercise and recovery, has been shown to provide significant improvement in the diagnostic and prognostic accuracy of exercise ECG in different clinical settings567 and independently of pharmacologic therapy.8 However, prognostic data on revascularized patients are still lacking. Accordingly, the present study was aimed at prospectively assessing the value of the SRI for the risk stratification of patients following coronary bypass surgery.
|
Materials and Methods
|
|---|
Study Population and Design
Two hundred seventy-eight consecutive patients who had previously undergone coronary bypass surgery and participated in a secondary prevention program were exercise tested and prospectively enrolled into the study. The surgical procedure had been done with the use of cardiopulmonary bypass with antegrade warm blood cardioplegia for myocardial protection. Following median sternotomy, conventional multivessel bypass grafting had been performed with the use of internal mammary arteries whenever possible. Patients with an acute coronary syndrome following bypass surgery, those unable to perform a symptom-limited exercise test, or those presenting with ECG criteria of left ventricular hypertrophy, left bundle-branch block, ventricular preexcitation, chronic atrial fibrillation, and prognostically relevant comorbidities, as well as those receiving digoxin or with paced cardiac rhythm, were not included. At the time of testing, the median time from surgery was 39 months (range, 18 to 58 months), and all patients were asymptomatic and in stable clinical condition. Following exercise testing, patients were prospectively followed-up for a median time of 36 months (95% confidence interval, 34 to 38 months).
An informed consent to participate in the study was obtained by all patients before enrolling. The study protocol was approved by the local institutional Ethical Committee.
Clinical Data and Follow-up
Hypertension was defined as resting systolic BP > 140 mm Hg, resting diastolic BP > 90 mm Hg, or treatment with antihypertensive drugs.9 Diabetes mellitus was diagnosed according to World Health Organization criteria.10 Hypercholesterolemia was defined as plasma total cholesterol > 6.2 mmol/L,11 or treatment with cholesterol-lowering drugs. Cardioactive drugs were classified as ß-blockers, nondihydropiridinic calcium-antagonists, angiotensin-converting-enzyme inhibitors, and vasodilators (dihydropiridinic calcium-antagonists, nitrates, and 
-adrenergic blockers). Ejection fraction was obtained by two-dimensional echocardiography using the Simpson rule.12 Outcome was determined from the patient interview at the outpatient clinic, hospital chart reviews, and telephone interviews with the patient, a close relative, or the referring physician if necessary. Cardiac death, nonfatal myocardial infarction, and further surgical or nonsurgical revascularization were target end points of the study. Death was defined as cardiac if strictly related to proven cardiac causes (fatal reinfarction, acute heart failure, or malignant arrhythmias) or if sudden and unexpected when occurring outside the hospital. Myocardial infarction was diagnosed based on a combination of symptoms, ECG, and cardiac-enzyme changes.
Exercise ECG Test and SRI Determination
Exercise ECG was performed using an upright, electromagnetically braked, cycle ergometer with 25-W incremental loading every 2 min. The 12-lead ECG was continuously monitored throughout the test for rhythm, rate, and ST-segment changes using the Mason et al13 exercise adaptation. BP was measured by arm-cuff sphygmomanometry during the last 30 s of each work stage. Exercise was continued until chest pain, repetitive arrhythmias, significant conduction abnormalities, ST-segment depression > 0.3 mV, systolic BP > 230 mm Hg, or its drop > 20 mm Hg, or limiting symptoms (dyspnea, dizziness, fatigue, cramp in legs) occurred. After exercise, patients recovered in a sitting position. Total work performed indicated the exercise capacity of the patient. ST-segment deviation in leads without pathologic Q waves, excluding aVR, was measured 60 ms after the J point using the end of P-R segment as reference. ECG response was defined as positive in case of horizontal or downsloping deviation > 0.1 mV in at least two contiguous leads, negative in case of no deviation or upsloping deviation, and nondiagnostic in case of horizontal or downsloping deviation < 0.1 mV. The decision on discontinuing cardioactive drugs before testing and on patient management following testing was independently made by the attending physician, who was unaware of the study aim. Only exercise testing data reported as part of patient care were available as test results.
Details on SRI determination have been extensively described elsewhere.56 Briefly, computer-calculated ST-segment amplitudes were obtained with a time constant of 12 s during exercise and up to 5 min during recovery. At the end of the test, the area subtended to baseline and limited by the ST-segment trend against heart rate during exercise and recovery was calculated in the lead with the greatest ST-segment shift. SRI was defined as the difference in absolute values between the areas defined by ST-segment depression in the heart rate domain during exercise and recovery.
Statistical Analysis
Continuous variables are presented as median with the corresponding interquartile difference. Categorical variables are presented as absolute number with corresponding percentages. Univariate odds ratios refer to the effect of being in the highest as compared to the lowest quartile for continuous variables or in the category with the highest observed frequency for categorical variables, respectively. The individual effect of clinical data, resting ejection fraction, and exercise testing results on survival was evaluated by Cox proportional-hazards regression analysis. Proportional hazard assumption was checked by plotting Schoenfeld results against fitted time and varying coefficients and with the Grambsh and Therneau test.14 In order to assess whether SRI added prognostic information to routinely obtained information, clinical data were entered first (model 1), resting ejection fraction and exercise testing data second (model 2), and SRI last (model 3). Separate models were developed for each group of variables. Analysis was performed with all variables of a given model plus all variables from the antecedent models identified as independent. All variables were entered into the model without any transformation or cutting-off. Nonlinearity was assessed by Wald test comparing higher-order models with that including only linear terms. In case of nonlinearity, a restrictive cubic spline15 was used to model a nonlinear effect of the covariate. Selection criteria was the Akaike information criterion16 applied backward for each model. Models were cross-validated by bootstrap technique (200 runs).17 Discrimination index D (the higher the better) and the Somer concordance index Dxy (the closer to one in absolute value the better), representing the concordance between predicted and observed outcome adjusted for data censoring, were obtained. Multivariate hazard ratios are presented with their 95% confidence intervals (CIs). Areas under the receiver operating characteristic curve18 of the estimated cumulative hazard functions were compared to provide evidence of a significant increase in predictive accuracy of the model after the addition of SRI.
Cumulative survival curves as a function of time by quartiles of SRI were generated with the Kaplan-Meier method and compared by the log-rank test. Estimated percentage event rates were derived from the Kaplan-Meier estimates to take censoring of the data into account. The statistical significance was settled at a p value < 0.05. A statistical package (S-PLUS, release 2000; Insightful Corporation; Seattle, WA) were used for analysis.
|
Results
|
|---|
Clinical Characteristics and Exercise Testing Results
The clinical characteristics of the study population and exercise testing results are summarized in Table 1
. Ninety-seven patients (35%) were tested who were receiving ß-blockers, 14 patients (5%) were tested receiving calcium antagonists, 92 patients (33%) were tested receiving angiotensin-converting enzyme inhibitors, and 39 patients (14%) were tested receiving nitrates. One hundred thirty-three patients (48%) were receiving statins.
No complications were observed. The exercise test was stopped because of muscular fatigue in 225 patients (81%), orthopedic limitation in 4 patients (1.4%), chest pain in 14 patients (5%), ST-segment depression 
0.3 mV in 8 patients (3%), excessive increase in BP in 14 patients (5%), decrease in BP in 3 patients (1%), and significant arrhythmias in 10 patients (3.6%). Test results were positive according to standard ECG criteria in 123 patients (44%) and were negative in 155 patients (56%) [p < 0.01].
Prediction of Outcome
Follow-up information was available in all patients. During a median follow-up time of 36 months (first quartile, 25 months; third quartile, 49 months), 8 patients (2.8%) died of a proven cardiac cause (four fatal reinfarctions and one irreversible heart failure) or suddenly, 15 patients (5.4%) had a nonfatal reinfarction, and 15 patients (5.4%) underwent further revascularization procedures. Clinical and exercise testing data stratified according the occurrence of target events are reported in Table 1. The results of multivariate analysis are summarized in Table 2
: after adjusting for the most predictive clinical (model 1) and exercise testing (model 2) variables, SRI (hazard ratio, 0.68; 95% CI, 0.53 to 0.89) remained the only variable significantly and independently correlated to the outcome (model 3). Furthermore, the area under receiver operating characteristic curve of the risk function increased from 0.65 (95% CI, 0.56 to 0.74) to 0.77 (95% CI, 0.68 to 0.85) [p = 0.01]) after the addition of SRI, thus demonstrating its significant contribution to the accuracy of the model.
The effect of SRI on outcome, estimated by restricted cubic spline, is reported in Figure 1
: progressively negative values are associated with increasing risk of adverse events, while positive values predict a favorable long-term outcome; the flex point of the curve corresponds to a SRI value of approximately – 15 mV x beats per minute. Finally, the effect of SRI on survival was analyzed using the product-limit Kaplan Meier method: ascending SRI quartiles were able to identify a significant (p = 0.001) increase in event-free survival (Fig 2
).
|
Discussion
|
|---|
The results of this study demonstrate that SRI based on comparative analysis of heart rate-adjusted ST-segment depression during exercise and recovery phase has significant and independent value for risk stratification of asymptomatic patients following coronary bypass surgery and provides an additional prognostic contribution to clinical and standard exercise ECG information. This is in keeping with and expands upon previous observations in patients with a recent myocardial infarction6 and in those with arterial hypertension.7
Periodic assessment of myocardial ischemia is generally done in the late phase following bypass surgery to assist in evaluation and treatment of chronic established coronary artery disease. Even though exercise ECG testing can provide more useful information at distance than shortly after operation,2 its accuracy remains largely influenced by the clinical likelihood of coronary disease progression, as in the presence of typical ischemia symptoms, or with clinical conditions which act as promoters of atherogenesis, such as diabetes mellitus, hemodialysis, or immunosuppressive therapy. In asymptomatic patients, a number of limitations can reduce the accuracy of exercise ECG. In particular, reliability of ST-segment changes are significantly affected by frequent resting abnormalities, making accessory testing parameters, such as symptom status, hemodynamic response, and exercise capacity, more trustworthy for test interpretation.
The SRI approach has been shown to significantly increase the diagnostic5 and prognostic67 value of ST-segment analysis even in the presence of cardioactive therapy.8 The physiologic rationale of comparative stress recovery adjustment of ST-segment depression grounds on earlier observations.192021 In addition, the SRI has two main advantages: to combine information on total amount and relative rate of development and resolution of normalized ST-segment depression, and to be independent on the achievement of a critical threshold of ST depression. These characteristics give the SRI superior prognostic ability as compared to standard ST-segment analysis in previous studies67 dealing with different patient populations, where conventional analysis of exercise ECG has limited accuracy. The addition of standard exercise test information failed to improve the predicting capacity of the statistical model in this study, with no testing variable being selected at multivariate analysis. Most likely explanations of this finding are the known unsatisfactory prognostic ability of standard exercise ECG interpretation and the substantial preservation of left ventricular function in this population making its prognostic importance at rest and during exercise less relevant than usually observed.
The results of the present study reinforce and expand previous findings in patients with chronic coronary disease in the late phase after bypass surgery. This has the potentially relevant implication of reducing the recourse to the more expensive and time-consuming imaging techniques and to provide a first-line, low-cost, and generally available screening tool that may be of help for the risk stratification of patients and the optimization of human and economic resources.
|
Footnotes
|
|---|
Abbreviations: CI = confidence interval; SRI = stress recovery index
Received for publication November 9, 2004.
Accepted for publication January 24, 2005.
|
References
|
|---|
- McConahay, DR, Valdes, M, McCallister, BD, et al (1977) Accuracy of treadmill testing in assessment of direct myocardial revascularization. Circulation 56,548-552[Abstract/Free Full Text]
- Krone, RJ, Hardison, RM, Chaitman, BR, et al Risk stratification after successful coronary revascularization: the lack of a role for routine exercise testing. J Am Coll Cardiol 2001;38,136-142[Abstract/Free Full Text]
- Elamin, MS, Mary, DASG, Smith, DR, et al Prediction of severity of coronary artery disease using slope of submaximal ST-segment/heart rate relationship. Cardiovasc Res 1980;14,681-691[ISI][Medline]
- Kligfield, P, Okin, PM, Ameisen, O, et al Correlation of the exercise ST/HR slope with anatomic and radionuclide cineangiographic findings in stable angina pectoris. Am J Cardiol 1985;56,418-421[CrossRef][ISI][Medline]
- Bigi, R, Maffi, M, Occhi, G, et al Improvement in identification of multivessel disease after acute myocardial infarction following stress-recovery analysis of ST depression in the heart rate domain during exercise. Eur Heart J 1994;15,1240-1246[Abstract/Free Full Text]
- Bigi, R, Cortigiani, L, Gregori, D, et al Exercise versus recover electrocadiography to predict mortality in patients with uncomplicated myocardial infarction. Eur Heart J 2004;25,558-564[Abstract/Free Full Text]
- Bigi, R, Cortigiani, L, Gregori, D, et al Exercise versus recovery electrocardiography for predicting outcome in hypertensive patients with chest pain. J Hypertens 2004;22,1-7
- Bigi, R, Curti, G, Sponzilli, C, et al Assessment of multivessel coronary artery disease by means of stress-recovery st/hr index in postinfarction patients on ß-blocker therapy. Ann Noninv Electrocardiol 1999;4,60-66[CrossRef]
- The fifth report on the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure. Arch Intern Med 1993; 153:154–183
- World Health Organization.. Diabetes mellitus: report of a WHO Study Group. No. 727. 1985 World Health Organization. Geneva, Switzerland:
- Sacks, FM, Pfeffer, MA, Moye, LA, et al The effect of pravastatin on coronary events after myocardial infarction in patients with average cholesterol levels. N Engl J Med 1996;335,1001-1009[Abstract/Free Full Text]
- Rogers, EW, Feigenbaum, H, Weyman, AE Echocardiography for quantitation of cardiac chambers. Yu, PN Goodwin, JF eds. Progress in cardiology 1979;Vol 8,807 Lea & Febiger. Philadelphia, PA:
- Mason, RE, Likar, I, Biern, RO, et al Multiple lead exercise electrocardiography: experience in 107 normal subjects and 67 patients with angina pectoris and comparison with coronary cinearteriography in 84 patients. Circulation 1967;36,517-525[Abstract/Free Full Text]
- Grambsh, P, Therneau, T Proportional hazard tests and diagnostics based on weighted residuals. Biometrika 1994;81,515-526[Abstract/Free Full Text]
- Hastie, T, Tibshirani, R Generalized additive models. 1990 Chapman & Hall. London, UK:
- Akaike, H Information theory as an extension of the maximum likelihood principle. Petrov, BN Csaki, F eds. Second International Symposium on Information Theory 1973,267-281 Akademiai Kiado. Budapest, Hungary:
- Efron, B, Tibshirani, R, Robert, J An introduction to the bootstrap. 1993 Chapman & Hall. London, UK:
- Hanley, JA, MCNeil, BJ The meaning and the use of the area under a receiver operating characteristic (ROC) curve. Radiology 1982;143,29-36[Abstract/Free Full Text]
- Detry, JMR, Piette, F, Brasseur, LA Haemodynamic determinants of exercise ST-segment depression in coronary patients. Circulation 1970;42,593-599[Abstract/Free Full Text]
- Parker, JO, Chiong, MA, West, RO, et al Sequential alteration in myocardial lactate metabolism, ST segments and left ventricular function during angina induced by atrial pacing. Circulation 1969;40,113-131[Abstract/Free Full Text]
- Homans, DC, Sublett, E, Dai, XZ, et al Persistence of regional left ventricular dysfunction after exercise-induced myocardial ischemia. J Clin Invest 1986;77,66-73[ISI][Medline]
TOP
Abstract
Introduction
