Predictive Ability of Inflammatory Markers on In-Hospital Outcomes in Patients Admitted to Coronary Care Unit (MORCOR-TURK INFLAME)

Ömer Kümet,Mehmet Özgeyik,Şahin Topuz,Mustafa Beğenç Taşcanov,Ferhat Dindaş,İrfan Şahin,İbrahim Ersoy,İbrahim Halil Tanboğa
DOI: https://doi.org/10.1177/00033197241273389
2024-08-12
Angiology
Abstract:We investigated the prognostic implications of the systemic immune-inflammatory index (SII), atherogenic index of plasma (AIP), C-reactive protein/albumin ratio (CAR), neutrophil-lymphocyte ratio (NLR), prognostic nutritional index (PNI), and triglyceride/glucose index (TGI) in the MORtality predictors in the CORonary Care Units in TURKey (MORCOR-TURK) population. This is the largest registry of coronary care unit (CCU) patients in Turkey (3157 patients admitted to CCU in 50 different centers). The study population was divided into two according to in-hospital survival status; 137 patients (4.3%) died in-hospital follow-up. A significant correlation was found between death and SII, CAR, NLR, and PNI but not for AIP and TGI in logistic regression. In Model 1 (combining parameters proven to be risk predictors), the -2 log-likelihood ratio was 888.439, Nagelkerke R2 was 0.235, and AUC (area under curve) was 0.814 (95% CI: 0.771-0.858). All other models were constructed by adding each inflammatory marker separately to Model 1. Only Model 3 (CAR + Model 1) had a significantly greater AUC than Model 1 (DeLong P = .01). Our study showed that CAR, but not other inflammatory index, is a significant predictor of in-hospital mortality in CCU patients when added to proven risk predictors.
What problem does this paper attempt to address?