<b>Multidrug-Resistant Bloodstream Infections in Intensive Care Settings at Nishtar Hospital, Multan: Patterns, Determinants, and Patient Outcomes</b>

Aim: Multi-drug resistant organisms (MDROs) pose a significant threat in intensive care units (ICUs), complicating treatment and worsening outcomes. This study aimed to determine the prevalence, microbiological patterns, risk factors, and clinical outcomes of bloodstream infections (BSIs) caused by multidrug-resistant organisms (MDROs) among patients admitted to intensive care settings (Adult ICU, NICU, PICU, and COVID ICU) at Nishtar Hospital, Multan. Methods: A single-center retrospective observational study was carried out including all consecutive patients with positive blood cultures taken within 48 hours of ICU admission between 1 January 2024 and 31 December 2024. After excluding contaminants and repeat admissions, 635 patients were analyzed. Demographics, comorbidities, device use, recent healthcare exposures, microbiology, empirical antimicrobial therapy, sepsis severity (SOFA), length of stay (LOS) and in-hospital mortality were extracted from electronic and paper records. MDROs were defined according to international interim consensus. Univariate and multivariate logistic regression were used to identify independent risk factors for MDRO-BSI and predictors of in-hospital mortality. Results: Of 635 patients with lab-confirmed BSI, 128 (20.1%) had infections due to at least one MDRO. ESBL-producing Enterobacterales were the most frequent MDRO category (n = 84; 13.2%), followed by MRSA (n = 23; 3.6%), MDR-Pseudomonas spp. (n = 10; 1.6%), and carbapenem-resistant Acinetobacter spp. (n = 7; 1.1%). Independent predictors of MDRO-BSI included: chronic renal impairment (OR 2.1; 95% CI 1.3–3.5), antibiotic exposure within the preceding 90 days (OR 2.7; 95% CI 1.8–4.1), and presence of an indwelling intravascular device at admission (OR 1.9; 95% CI 1.1–3.2). Empirical antibiotic therapy given in ICU was inappropriate (inactive by in-vitro AST) in 46% of MDRO cases vs 14% of non-MDRO cases (p < 0.001). Median LOS was longer for MDRO patients (16 days [IQR 11–25]) than for non-MDRO patients (12 days [IQR 8–19]; p < 0.01). In-hospital mortality did not differ significantly between MDRO and non-MDRO groups (14.8% vs 12.2%; p = 0.47) after adjustment for age, baseline comorbidity burden, and sepsis severity. Conclusion: MDRO-associated BSIs are common in intensive care settings at Nishtar Hospital, Multan, and are associated with higher rates of inappropriate empirical therapy and prolonged hospital stay. Regular local surveillance, early recognition of high-risk patients, device-focused infection prevention, and tailored antimicrobial stewardship strengthened to match ICU epidemiology are recommended to reduce morbidity and optimize antibiotic use.