[Skip to Content]
Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
[Skip to Content Landing]
Article
March 1995

A Comparison of Four Severity-Adjusted Models to Predict Mortality After Coronary Artery Bypass Graft Surgery

Author Affiliations

From the Departments of Surgery (Drs Orr and Maini and Mss Dumas and O'mara) and Critical Care Medicine (Dr Sottile), the Fallon Healthcare System, Worcester, Mass.

Arch Surg. 1995;130(3):301-306. doi:10.1001/archsurg.1995.01430030071015
Abstract

Objective:  To assess the validity of four severity-adjusted models to predict mortality following coronary artery bypass graft surgery by using an independent surgical database.

Design:  A prospective observational study wherein predicted mortality for each patient was obtained by using four different published severity-adjusted models.

Setting:  A university-affiliated teaching community hospital.

Patients:  Eight hundred sixty-eight consecutive patients who underwent coronary artery bypass graft surgery without accompanying valve or aneurysm repair during the period from 1991 to 1993.

Interventions:  None.

Main Outcome Measures:  Predicted mortality rates for each model were obtained by averaging individual patient predictions and were compared with actual mortality rates. We assessed the accuracy of overall prediction for the total series, as well as compared individual patient predictions created by each model. The discrimination of models was assessed with receiver operating characteristic curves and the Hosmer-Lemeshow goodness-of-fit statistic.

Results:  The observed crude mortality rate was 3.7%. The predicted mortality rate ranged from 2.8% to 9.2%, despite relatively good discrimination by the models (area under the receiver operating characteristic curve, 0.70 to 0.74). The individual patient mortality predicted by different models varied by as much as a ninefold difference.

Conclusions:  The currently used coronary artery bypass graft predictive models, although generally accurate, have significant shortcomings and should be used with caution. The predicted mortality rate following coronary artery bypass graft surgery varied by a factor of 3.3 from lowest to highest, making the choice of model a critical factor when assessing outcome. The use of these models for individual patient risk estimations is risky because of the marked discrepancies in individual predictions created by each model.(Arch Surg. 1995;130:301-306)

References
1.
Ellwood PM.  Outcomes management (The Shattuck Lecture) . N Engl J Med . 1988;318:1549-1556.Article
2.
Williams SV, Nash DB, Goldfarb N.  Differences in mortality from coronary artery bypass graft surgery at five teaching hospitals . JAMA . 1991;266:810-815.Article
3.
O'Connor GT, Plume SK, Olmstead EM, et al.  A regional prospective study of in-hospital mortality associated with coronary artery bypass grafting . JAMA . 1991;266:803-809.Article
4.
Kouchoukos NT, Ebert PA, Grover FL, Lindesmith GG.  Report of the ad hoc committee on risk factors for coronary artery bypass surgery . Ann Thorac Surg . 1988;45:348-349.Article
5.
Topol EJ, Califf RM.  Scorecard cardiovascular medicine . Ann Intern Med . 1994; 120:65-70.Article
6.
Loop FD, Berrettoni JN, Pichard A, et al.  Selection of the candidate for myocardial revascularization: a profile of high risk based on multivariate analysis . J Thorac Cardiovasc Surg . 1975:69:40-51.
7.
Kennedy JW, Kaiser GC, Fisher LD, et al.  Multivariate discriminant analysis of the clinical and angiographic predictors of operative mortality from the Collaborative Study in Coronary Artery Surgery (CASS) . J Thorac Cardiovasc Surg . 1980;80:876-887.
8.
Pierpont GL, Kruse M, Ewald S, Weir EK.  Practical problems in assessing risk for coronary artery bypass grafting . J Thorac Cardiovasc Surg . 1985;89:673-682.
9.
Parsonnet V. Dean D, Bernstein AD.  A method for uniform stratification of risk for evaluating the results of surgery in acquired adult heart disease . Circulation . 1989;79( (suppl I) ):I3-I12.
10.
Higgins TL, Estafanous FG, Loop FD, et al.  Stratification of morbidity and mortality outcome by preoperative risk factors in coronary artery bypass patients: a clinical severity score . JAMA . 1992;267:2344-2348.Article
11.
O'Connor GT, Plume SK, Olmstead EM, et al.  Multivariate prediction of in-hospital mortality associated with coronary artery bypass graft surgery . Circulation . 1992;85:2110-2118.Article
12.
Hannan EL, Kilburn H Jr, Racz M, Shields EP, Chassin MR.  Improving the outcomes of coronary artery bypass surgery in New York State . JAMA . 1994;271:761-766.Article
13.
Edwards FH, Clark RE, Schwartz M.  Coronary artery bypass grafting: the Society of Thoracic Surgeons National Database experience . Ann Thorac Surg . 1994;57:12-19.Article
14.
Poses RM, Cebul RD, Collins M, Fager SS.  The importance of disease prevalence in transporting clinical prediction rules . Ann Intern Med . 1986;105:586-591.Article
15.
Wasson JH, Sox HC, Neff RK, Goldman L.  Clinical prediction rules: applications and methodologic standards . N Engl J Med . 1985;313:793-799.Article
16.
Nashef SAM, Carey F, Silcock MM, Oomen PK, Levy RD, Jones MT.  Risk stratification for open heart surgery: trial of the Parsonnet system in a British hospital . BMJ . 1992;305:1066-1067.Article
17.
Junod FL, Harlan BJ, Payne J, et al.  Preoperative risk assessment in cardiac surgery: comparison of predicted and observed results . Ann Thorac Surg . 1987; 43:59-64.Article
18.
Lemeshow S, Hosmer DW Jr.  A review of goodness of fit statistics for use in the development of logistic regression models . Am J Epidemiol . 1982;115:92-106.
19.
Katz D, Foxman B.  How well do prediction equations predict? using receiver operating characteristic curves and accuracy curves to compare validity and generalizability . Epidemiology . 1993;4:319-326.Article
20.
Hanley JA, McNeil BJ.  The meaning and use of the area under a receiver operating characteristic (ROC) curve . Radiology . 1982;143:29-36.Article
21.
Charlson M, Ales KL, Simon R, MacKenzie CR.  Why predictive indexes perform less well in validation studies: is it magic or methods? Arch Intern Med . 1987;147:2155-2161.Article
22.
Diamond D.  Future imperfect: the limitations of clinical prediction models and the limits of clinical prediction . J Am Coll Cardiol . 1989;14:12A-22A.
23.
Marshall G, Grover FL, Henderson WG, Hammermeister KE.  Assessment of predictive models for binary outcomes: an empirical approach using operative death from cardiac surgery . Stat Med. 1994;13:1501-1511.Article
24.
Heckerling PS, Conant RC, Tape TG, Wigton RS.  Reproducibility of predictor variables from a validated clinical rule . Med Decis Making . 1992;12:280-285.Article
25.
Teres D, Lemeshow S.  Why severity models should be used with caution . Crit Care Clin . 1994;10:93-110.
26.
Rocca B, Martin C, Viviand X, Bidet P, Saint-Gilles HL, Chevalier A.  Comparison of four severity scores in patients with head trauma . J Trauma . 1989;29:299-305.Article
27.
Lemeshow S, Teres D, Avrunin JS, Pastides H.  A comparison of methods to predict mortality of intensive care unit patients . Crit Care Med . 1987;15:715-722.Article
28.
Lemeshow S, Teres D, Klar J, Avrunin JS, Gehlbach SH, Rapoport J.  Mortality probability models (MPM II) based on an international cohort of intensive care unit patients . JAMA . 1993;270:2478-2486.Article
29.
Hannan EL, Kilburn H Jr, O'Donnell JF, et al.  Adult open heart surgery in New York State: an analysis of risk factors and hospital mortality rates . JAMA . 1990; 264;2768-2774.
30.
Teres D, Steingrub S. Severity scoring for the cardiovascular patient. In: Dantzer D, Kvetan V, eds. The Critically III Cardiac Patient. Philadelphia, Pa: JB Lippincott. In press.
×