Albumin cobalt binding assay to rule out acute coronary syndrome

The purpose of this study was to validate the Albumin Cobalt Binding (ACB) assay at the Seattle Veterans Affairs (VA) Hospital to determine if it would provide an earlier rule-out of acute coronary syndrome (ACS) in patients, compared to current use of cardiac injury markers. This study compares the distribution of ischemia modified albumin (IMA) values of our patient population to those provided by the kit manufacturer. IMA values were determined photometrically on a Roche Modular Analytical System on 200 subjects: 69 subjects not experiencing chest pain (normals), 78 subjects presenting to the emergency room (ER) with chest pain whose initial and subsequent troponin results were negative (non-converters), and 53 subjects presenting to the ER with chest pain whose initial troponin result was negative but subsequent troponin results were positive (converters). Based on the relationships between IMA values in the initial samples from the non-converters and converters, we constructed a ROC curve to identify an optimum IMA rule-out value. The IMA values (mean+/-SD) for the normals, non-converters, and converters were 89+/-7.1, 100+/-13.9, and 126+/-14.1 U/ml, respectively, and each mean was statistically different from the means of the other groups. The ROC curve comparing converters and non-converters showed an area of 0.89 (p <0.001) compared to the line of identity. An IMA cut-off of 97 U/ml gives a 98% sensitivity and 45% specificity and may be the best decision point to differentiate between these groups in our population. Nine of 78 non-converters were classified as having unstable angina. In conclusion, the ACB assay has a strong negative predictive value and sensitivity in our population for predicting the troponin results at 6 to 24 hr post-presentation. Because ACB results may be facility- and instrument-dependent, each facility should conduct an independent ROC analysis to determine the optimal IMA rule-out level. The ACB assay, when used in conjunction with cardiac injury markers and assessment of unstable angina, holds promise in reducing inappropriate low-risk hospital admissions and improving the clinical management of patients with chest pain.