Correlation of risk scores with coronary anatomy in non-ST-elevation acute coronary syndrome.

BACKGROUND The literature lacks studies regarding the correlation between risk scores and coronary anatomy in acute coronary syndrome (ACS) OBJECTIVE: Correlate risk scores with the severity of the coronary lesion in ACS with non-ST elevation. METHODS A total of 582 patients were analyzed between July 2004 and October 2006. The correlation between TIMI risk scores and GRACE (hospital and six months) scores was performed for patients with coronary lesion ≥ 50%, using Spearman's non-parametric method. Multiple regression logistics was used to determine the predictive ability of the scores to discriminate to discriminate who will have a coronary lesion ≥ 50%. RESULTS Most subjects were male (319 or 54.8%), mean age of 59.9 (± 10.6) years. A positive correlation was observed between risk scores and coronary lesion ≥ 50% (TIMI r = 0.363 [p < 0.0001]; hospital GRACE r = 0.255 [p < 0.0001]; GRACE at six months r = 0.209 [< 0.0001]). The area under the ROC curve for each score to determine to discriminate who will have a coronary lesion > 50% was: TIMI = 0.704 [CI95% 0.656-0.752; p <0.0001]; hospital GRACE = 0.623 [CI95% 0.573-0.673; p < 0.0001]; GRACE at six months= 0.562 [CI95% 0.510-0.613; p ;= 0.0255]. Comparing the areas under the ROC curve, it was found: TIMI versus hospital GRACE: p = 0.01; TIMI versus GRACE at six months:p < 0.0001; hospital GRACE versus GRACE at six months: p = 0.0461. CONCLUSION Risk scores correlate with the severity of coronary lesions, and the TIMI risk score showed the best predictive ability.