Abstract
AIM AND BACKGROUND: Antimicrobial sensitivity (AMS) reports are often available after 72 hours of identification of gram-negative (GN) hospital-acquired infection (HAI). Prediction of carbapenem-resistant infection (CRI) among GN strains is important even before AMS reports are available, for judicious use of empirical antibiotics. We aimed to study the predictors of CRI in patients with HAI. MATERIALS AND METHODS: We conducted a single-center prospective observational study between April 2023 and September 2024 on patients of GN sepsis with HAI. The use of empirical carbapenem antibiotics, organ dysfunction scores, the modified nutritional risk in critically ill (mNUTRIC) score, blood-count-derived inflammation indices, type of HAI, AMS reports, and in-hospital mortality were noted. RESULTS: A total of 935 sepsis patients with HAI were screened, and there were 195 patients with GN infection. Among the 195 patients, 145 (74.4%) had CRI and 50 (25.6%) had non-CRI. Multivariable logistic regression revealed that the length of intensive care unit (ICU) stay before the day of HAI (p = 0.009, adjusted odds ratio (OR) 1.155, 95% confidence interval (CI) 1.037-1.286), presence of ventilator-associated pneumonia (VAP) (p-value < 0.001, adjusted OR 4.170, 95% CI: 1.858-9.361), empirical carbapenem antibiotics before the day of HAI (p-value = 0.004, adjusted OR 3.164, 95% CI: 1.439-6.957), and septic shock on the day of HAI (p-value 0.012, adjusted OR 4.162, 95% CI: 1.366-12.677) were the independent risk factors of CRI. CONCLUSION: In GN sepsis patients with HAI, respiratory infection (VAP), length of ICU stay prior to HAI, septic shock, and empirical carbapenem antibiotic administration are risk factors of CRI. HOW TO CITE THIS ARTICLE: Mareguddi AB, Chaudhuri S, Shanmukhappa SM, Parampalli V, Bhatt MT, Fernandes R, et al. The Novel "RISC" Score as a Risk-prediction Model of Carbapenem-resistant Hospital-acquired Infections in Adult Sepsis Patients - A Prospective Observational Study. Indian J Crit Care Med 2025;29(4):352-362.