Antibiotic Selection in Ventilator-Associated Pneumonia: Moving Beyond Early- vs. Late-Onset Paradigms?—A Retrospective Cohort Study

Introduction: Ventilator-associated pneumonia (VAP) is associated with increased morbidity and mortality. Early-onset (EOVAP) and late-onset (LOVAP) VAP differ in risk factors, microbiology, and outcomes. Methods: A retrospective cohort study of adult ICU patients admitted to a tertiary hospital between May 2020 and May 2022 compared EOVAP and LOVAP. Results: 194 patients were included (73.7% male), with a median age of 61 years. A total of 70.6% were admitted to the ICU for infection-related respiratory failure, predominantly due to COVID-19 (59.3%). EOVAP represented 21% and LOVAP 79% of cases, with median days to diagnosis of 4 and 11, respectively. Empirical antibiotic therapy was adequate in both groups; de-escalation was possible in 26.8%, and escalation due to treatment failure was required in 13.4%. The most commonly prescribed initial antibiotic was piperacillin–tazobactam (45%). Broad-spectrum antibiotic use did not differ between groups. Amoxicillin–clavulanate resistance was high in both groups (LOVAP 62.1%; EOVAP 48.8%), and piperacillin–tazobactam resistance was more frequent in LOVAP (41.2%). Follow-up cultures showed a high persistence rate (73.2%), with no significant change in resistance patterns. Conclusions: VAP remains common in the ICU, and the COVID-19 pandemic significantly impacted the already complex epidemiology of VAP. High resistance rates to amoxicillin–clavulanate and frequent isolation of MDR pathogens challenge empirical treatment based on onset timing, supporting empiric antibiotic selection guided by local microbiological ecology.