Indexed and abstracted in Science Citation Index Expanded and in Journal Citation Reports /Science Edition Bratisl Med J 2023; 124 (4) 277 – 279 DOI: 10.4149/BLL_2023_042 CLINICAL STUDY C-reactive protein-to-serum albumin ratio as a marker of prognosis in adult intensive care population Ibrahim KARAGOZ 1 , Bahri OZER 2 , Ilker ITAL 1 , Mustafa TURKOGLU 1 , Akin DISIKIRIK 1 , Songul PELTEK OZER 3 Abant Izzet Baysal University Hospital, Department of General Surgery, Golkoy, Bolu, Turkey. bahriozer@hotmail.com ABSTRACT BACKGROUND: Patients in intensive care unit (ICU) require close follow up and clinical attention due to variability in the course of their underlying morbidities. The estimation of prognosis in these subjects has an utmost importance. Recent studies showed that C-reactive protein-to-serum albumin ratio (CAR) could be a reliable marker of inflammation in certain conditions. We aimed to compare CAR levels of deceased patients to those in survived subjects treated in ICU. PATIENTS AND METHODS: We retrospectively analyzed the data of adult patients. CAR was simply calculated by dividing the levels of CRP by those of serum albumin. Patients were grouped either as deceased or survived according to the prognosis. The data of the survived and deceased ICU subjects were compared. RESULTS: A total of 208 subjects, 101 deceased and 107 survived, were enrolled in the study. Median CAR levels of the deceased and survived subjects were 49.5 (3–153 %) and 11 (0.2–119 %), respectively (p < 0.001). CAR was significantly correlated with PDW (r = 0.24; p < 0.001) and serum creatinine (r = 0.27; p < 0.001) levels. In ROC analysis, CAR values higher than 30.2 % have 72 % sensitivity and 70 % specificity in predicting mortality in ICU population (AUC: 0.74; p < 0.001; 95% CI: 0.67–0.81). CONCLUSIONS: We suggest that CAR levels of the subjects in ICU should be evaluated during medical care. Increased CAR levels should alert physicians for a worse outcome in those subjects (Tab. 1, Fig. 1, Ref. 21). Text in PDF www.elis.sk KEY WORDS: C-reactive protein-to-serum albumin ratio, intensive care, mortality. 1 Abant Izzet Baysal University Hospital, Department of Anesthesiology, Bolu, Turkey, 2 Abant Izzet Baysal University Hospital, Department of General Surgery, Bolu, Turkey, and 3 Izzet Baysal Training and Research Hospital, Department of Pathology, Bolu, Turkey Address for correspondence: Bahri OZER, MD, Abant Izzet Baysal Univer- sity Hospital, Department of General Surgery, 14280, Golkoy, Bolu, Turkey. Phone: +903742534656, Fax: +903742534615 Introduction Patients in intensive care unit (ICU) require close follow up and clinical attention due to variability in the course of their underlying morbidities (1). The estimation of the prognosis in these subjects has an utmost importance. For this purpose, various prognostic tools have been developed, such as APACHE II and SOFA scores. However, these scores are not easy to assess while novel prognostic tools to determine prognosis in ICU population are still a necessity. Recent studies showed that C-reactive protein-to-serum al- bumin ratio (CAR) could be a reliable marker of inflammation in certain conditions. For instance, the Care Time study suggested CAR as a valuable predictor of diabetic kidney injury in subjects with type 2 diabetes mellitus (2). Moreover, Liu et al found that CAR could predict the outcome in patients with coronary arterial disease (3). Both type 2 diabetes mellitus and coronary heart dis- ease are characterized with an increased burden of inflammation. Subjects in ICU have also increased inflammatory markers in se- rum. Therefore, we hypothesized that CAR could be associated with the prognosis in ICU population. We aimed to compare CAR levels of deceased patients to those in survived subjects treated in ICU. Patients and methods After obtaining approval from the local ethics committee (as of 13th of July in 2021, approval No: 2021/192), we retrospectively analyzed the data of the adult patients in ICU of Abant Izzet Bay- sal University hospital between January 2021 and January 2022. Patients with sepsis, rheumatologic conditions, pregnancy, and under 18 years of age were excluded from the study. General characteristics of the subjects, namely age, gender, duration of stay in ICU, hematological parameters (white blood cell count (WBC), neutrophil count (neu), lymphocyte count (lym), hemoglobin (Hb), hematocrit (Htc), erythrocyte distribu- tion width (RDW), platelet count (PLT), mean platelet volume (MPV), platelet distribution width (PDW)), fasting glucose, se- rum albumin, creatinine, and C-reactive protein (CRP) were re- corded after being obtained from patients’ files and institutional database. CAR was simply calculated by dividing the levels of CRP to those of serum albumin. Data about comorbidities, and positive urine or blood culture findings were also recorded. Pa- tients were grouped either as deceased or survived according to