Morbidity and Duration of ICU Stay after Cardiac Surgery* A Model for Preoperative Risk Assessment Kenneth

J Tuman,

M.D., F.C.C.P.; Robert J McCarthy, Pharm.D.;

Robert J March, M.D., F.C.C.P.; Hassan Najafi, M.D., RC.C.P.; and

Anthony D. lvankovich, M.D., F.C.C.P.

Although risk factors for mortality after cardiac surgery have been identified, there is DO widely applicable method for readily determining risk of postoperative morbidity based on preoperative severity of illness. The goal of this study was to develop a model for stratifying the risk of serious morbidity after adult cardiac surgery using readily available and objective clinical data. After univariate analysis of risk factors in 3,156 operations, 11 variables were identi6ed as important predictors by logistic regression (LR) analysis and used to construct an additive model to calculate the probability of serious morbidity. Reliable correlation was found between a simpli6ed additive model for clinical use and the LR model. The clinical and logistic models were then tested prospectively in 394 patients and demonstrated a pattern of increasing morbidity with as-

cending scores similar to that predicted by the reference group. Increasing clinical risk score was also associated with a greater frequency of individual complications as well as prolongation of leu stay. This study demonstrates that it is feasible to design a simple method to stratify the risk of serious morbidity after adult cardiac surgery. With further prospective multicenter re60ement and testing, such a model is likely to be useful for adjusting severity of illness when reporting outcome statistics as well as planning resource utilization. (Cheat 1992; 102:36-44)

rove me nt s in medical treatment and recent I mptrends in interventional cardiology have altered

of results emerging from multiple centers, it is imperative to identify appropriate markers of severity of illness and patient mix prior to surgery This is essential if we are to minimize distortion of apparent rates of hospital morbidity and mortality and avoid false conclusions regarding quality of medical care within given institutions. Outcome data have also become important to payers of health care, such as the Health Care Finance Administration (HCFA)8and insurance carriers, since cost of surgery is highly inHuenced by complexity and intensity of complications as well as duration of ICU and total postoperative stay The model developed by HCFA uses data from Medicare billing and employs partial severity adjustment based on a few preoperative variables. The model is seriously deficient in its ability to predict mortality and morbidity? Other risk stratification methods have been devised to predict operative mortality reasonably well, but they are generally too complicated to be practical and often require data that are not readily available in all cardiac surgical centers. Scores quantifying ventricular dysfunction or extent of coronary atherosclerosis are valuable but are often too complex for widespread clinical application to risk assessment. Therefore, a practical method is needed to reliably estimate morbidity and mortality among patients undergoing cardiac surgery, using readily available preoperative markers of severity of illness. This study has identified

the profile of patients referred for cardiac surgery. The proportion of high-risk patients has increased dramatically with greater numbers of elderly patients, those requiring reoperation, and patients with serious associated illnesses presenting for cardiac surgery. 1 While several studies have used multivariate statistical methods to predict the risk of death after cardiac surgery;"? there is a paucity of information regarding preoperative predictors of serious postoperative complications. In many instances, operative mortality does not correlate with complication rates and length of hospital stay. Therefore, mortality alone is not an adequate marker of quality of care or cost effectiveness. For example, if death occurs intraoperatively or soon after cardiac surgery, costs will be substantially lower than if a patient dies after a protracted postoperative course requiring intensive support of the cardiopulmonary, renal, or neurologic systems. Furthermore, irrespective of cost, these morbidities are important indicators not only of quality of care but of quality of life after cardiac operations. With scrutiny

*From the

Departments of Anesthesiology (Drs. Tuman, McCarthy, and Ivankovich) and Cardiovascular and Thoracic Surgery (Drs. March and Najafi), Rusb-Presbyterian-St. Luke's Medical Center, Chicago. Manuscript received October 12, 1991; revision accepted February

24.

Reprint requests: Dr. Tuman, Department of Anesthesiology/739 ]elke, Rush-Presbyterian-St. Lukes Medical Center, Chicago 60612

38

APACHE= acute physiologic and chronic health evaluation; CABG= coronary artery bypass graft; CHF = congestive heart failure; HCFA = Health Care Finance Administration

Morbidity and Durationof leuStay after C8rdiac Surgery(Tuman .t aI)

a practical method of estimating perioperative risk in patients undergoing coronary revascularization or valvular surgery based on a universally available clinical risk scoring system. METHODS

Development of Clinical Risk Score

This study was approved by the institutional Human Investigation Committee and is based on the analysis of prospectively collected data on 3,156 consecutive adult patients undergoing cardiac surgery. Cardiac surgery combined with ventricular aneurysmectomy, endocardial mapping, carotid endarterectomy, or intracardiac procedures other than valvular surgery were excluded to enhance the homogeneity of the patient population. Preoperative patient demographics and factors previously demonstrated to be associated with greater mortality and morbidity after cardiac surgery were selected for examinations? (Table 1).The preoperative risk evaluation of each patient was performed prospectively during the preoperative visit. Left ventricular dysfunction was diagnosed based on ejection fraction of less than 0.35, and a history of myocardial infarction (MI) required diagnostic electrocardiographic changes on the baseline tracing or that the patient had been informed by a cardiologist of a definite infarction. V nstable angina was defined as recent crescendo pains unresponsive to conventional medical therapy, or occurring at night or at rest. Congestive heart failure (CHF) (typical roentgenographic changes with either rales or an 83 gallop), previous cardiac surgery, emergency procedure (need for operation because of acute hemodynamic compromise, unstable angina unresponsive to maximal pharmacologic and mechanical support, or complications of angioplasty with arterial injury), arrhythmias requiring preoperative therapy, renal dysfunction (preoperative creatinine concentration > 1.4 mg/dl), diabetes mellitus (glucose intolerance requiring therapy with oral hypoglycemic agents or insulin), chronic pulmonary disease (resulting in functional disability and/or requiring long-term bronchodilator therapy and/or FEV. of less than 75 percent of predicted), and known preoperative cerebrovascular disease were recorded. Other perioperative factors that were noted included age, gender, presence of cardiac cachexia (body mass index [8MI] < 15 kg/m t), obesity (BMI >30 kg/mt), and pulmonary hypertension (mean pulmonary artery pressure ~25 percent of systemic values). Morbidity was defined as the presence of one or more of the following categories of complications: cardiac-perioperative myocardial infarction (new and persistent postoperative electrocardiographic changes and an increase in cardiac specific enzymes) or low cardiac output syndrome (cardiac index