Background: Prior studies indicate that an elevated creatinine kinase (CK)-MB imparts poor prognosis in patients with acute coronary syndrome despite a normal troponin. Its prognosis in the undifferentiated chest pain observation unit (CPU) population remains undefined.
Objective: To compare rates and predictors of 30-day adverse cardiac events in 2 cohorts (CK ±/MB+ vs. normal [CK ±/MB-]) in low-moderate-risk CPU patients.
Methods: Consecutive CPU patients were followed in a retrospective cohort study for primary outcome (acute coronary syndrome, percutaneous transluminal coronary angioplasty, coronary artery bypass graft, abnormal stress test, cardiac hospitalization, or death within 30 days) by using standardized chart reviews and national death registry. Exclusions were: those aged 30 years or younger, positive troponin, ischemic electrocardiogram, hemodynamic instability, heart failure, or dialysis.
Results: Between January 2006 and April 2009, 2979 patients were eligible, of which 350 excluded and 2629 analyzed. MB+ compared with normal patients were more likely to be: older (mean, 53.4 ± 14 vs. 51.5 ± 12 years; P = 0.04); male (71% vs. 40%; P = 0.01); renal insufficient (5% vs. 2%; P = 0.01); hypertensive (50% vs. 44%; P = 0.04); dyslipidemic (44% vs. 33%; P = 0.01) obese (55% vs. 43%; P = 0.01); and with known coronary artery disease (14% vs. 5%; P < 0.01). Composite adverse events were 213 (8%) and did not significantly differ for either initial MB+ vs. normal (9.1%, 8.0%; odds ratio, 1.1, 0.7-1.9) or serial MB+ vs. normal (7.5%, 7.4%; odds ratio, 1.0, 0.5-1.8). In a multiple logistic regression model, male sex, diabetes, and prior CAD predicted adverse events, whereas CK-MB along with race, hypertension, smoking, dyslipidemia, family history, and obesity did not.
Conclusions: Elevated CK-MB does not add value to serial troponin testing in low-moderate-risk CPU patients.