EURO-SCORING FOR RISK STRATIFICATION IN CABG SURGERY - 10 YEARS EXPERIENCE

Abstract

Aim of the study was to evaluate patients pre operatively with Euroscoring System to judge the end point,
hospital mortality in CABG done in last ten years by a single surgeon. From June 1989 to June 1999 the data
of 262 cases of CABG done was collected on the data sheet each case was scored pre operatively with
Euroscoring System. The different groups were made from this system, Group 1 Score 0-1, Group 2 Score 3-
4, Group 3 Score 5-6, Group 4 Score 7-8, Group 5 > 9, Another Grouping was Group 1 score 0-5, and Group
2 score 6-10 and Group 3 score 11-15. Pre op data was collected and analyzed by SPSS Version 7.5, The End
point was hospital mortality. In this group of 262 patients, the age range was 25-77 with the mean 52.41 years,
248 (94.7%) were male and 14 (5.3%) were female. In this whole cohort of pas patients 227 (86.6%) were
having stable angina pectoris and 35 (13.4%) were having unstable angina. Pre op angina status was Class I
in 5 (1.9%), Class II 88 (33,6%), Class III 132 (50.4%) and Class IV were 37 (14.1%). There were 116 (44.3%)
hypertensive, 56 (21.4%) were diabetics and 9 (3.4%) were obese. Recent myocardial Infarction was there in
9 (14%) of cases, the old non Q- wave infarction was present in 18 (6.9%) of cases and Q- wave infarction was
present in 42 (16%) of cases. Pre op Ejection fraction was good in (EF>50%) in 204 (77.9%) cases, Fair EF
30-49%) in 50 (19.1%) cases, poor EF<30%) in 8 (3.1%) cases. Pre op support of intra-aortic balloon
counterpulsation (IABP) was used in 3 (1.1%), pre operative ventilation was done in 1 (0.4%), Inotropic
support was present in 4 (1.5%), and vasodilators were given in 14 (5.3%) cases. Elective surgery was done in
92%, urgent in 4%, emergency in 3% and salvage surgery was done in 1% of cases. Mortality in Euroscore
Group I (Score 0-1) was 3.1%, in Group II (Score 2-3) 9.4%, in Group III (Score 4-5 ) 19%, in Group IV
(Score 6-7) 25% and in Group VI ( Score >9) the mortality was 60%. The Euroscore from 0-5 was having
6.1% mortality, the score from 6-10 was having 20% and the score from 11-15 was having 80% mortality. On
Logistic regression overall predictive accuracy of Euroscoring is very good (90%), Predictive accuracy, 37%
of deaths could be explained on the existing variables positive predictive value is 19.05% and negative
predictive value is 99.17%. The predictive accuracy of Euroscoring changes with various risk groups. In low
risk Groups (Score 0-5) and (Score 6-10) Euroscore predicts survival more accurately. In high risk Group
(11-15) Euroscoring better predicts mortality rather than survival. The factors included in permutations of
Euroscore explain only 37% of the observed mortality. It is noted that the observed mortality is consistently
higher than that predicted by logistic regression. Euroscoring is a good tool of risk stratification to predict the
out come but not ideally suited to our clinical circumstances. Though we have documented an overall
predictive accuracy of 92%, it is limited in its usefulness because it does not take into consideration certain
risk factors found to be important in our patient population. In addition, the relative weight assigned to
various risk factors in scoring needs to be readjusted for our patient population in the light of observations
made on our patient population. There is a need to develop a scoring system of our own which could be used
for better prediction of outcomes in our clinical circumstances