Abstract Rationale During the first COVID-19 wave, prolonged invasive mechanical ventilation (IMV), high device utilization, and staff pressure increased the risk of ventilator-associated pneumonia (VAP) and cross-contamination. We implemented a practical disinfection bundle with targeted surveillance of expiratory valves to quantify contamination and describe outcomes during surge conditions. Methods Retrospective observational study in the ICU of a 122-bed public hospital (Babahoyo, Ecuador), April-November 2020. All adults with COVID-19 receiving ventilatory support were included (N = 129). Variables included ventilatory modality (IMV vs. high-flow), ventilator model and operating hours, and in-hospital outcome. Post-use microbiologic surveillance of ventilator valves was performed in a convenience sample of units; cultures targeted expiratory valves given aerosol retro-flow risk. The primary endpoint was contamination of expiratory valves with carbapenemase-producing organisms. Secondary endpoints were in-hospital mortality and bundle feasibility (post-disinfection cultures). Descriptive statistics and chi-square tests compared outcomes across ventilator models. Results Of 129 patients, 114 (88.4%) required IMV and 15 (11.6%) high-flow. Overall mortality was 79.1% (102/129); mortality among IMV patients was 79.8% (91/114) versus 73.3% (11/15) on high-flow. Device utilization totaled 43,909 operating hours across models (Evita 2, 34,393 h; INSPIRATION ICU, 6,730 h; Siaretron 4000, 2,322 h; Trilogy 201, 464 h). Outcomes by model among IMV users were: Evita 2, 74 deceased/20 alive; INSPIRATION ICU, 13/1; Siaretron, 4/1; Trilogy, 0/1; differences were not significant (χ² = 5.694, df = 4, p = 0.223). Microbiologic surveillance sampled 14 ventilators; 3/14 (21.4%) expiratory valves were positive for carbapenemase-producing Klebsiella pneumoniae (KPC). All positives were from Evita 2 units; paired clinical cultures confirmed KPC in corresponding patients. After chemical disinfection consistent with manufacturer specifications, repeat random cultures from the ventilator fleet were negative, supporting the feasibility and effectiveness of immediate post-use processing. Conclusions During the pandemic surge, expiratory valve contamination with KPC was documented in one in five sampled ventilators, highlighting a plausible fomite pathway for VAP/MDRO transmission. A simple disinfection bundle centered on rapid post-use processing and targeted valve surveillance was feasible and associated with culture negativity after reprocessing. ICUs should prioritize: (1) explicit valve-focused steps within cleaning and sterilization SOPs; (2) situational surveillance criteria during outbreaks; and (3) brand-agnostic training covering valve architecture and safe disassembly. These findings are applicable across resource-constrained settings facing high ventilator throughput and prolonged IMV exposure. This abstract is funded by: NO FUNDING
Briones-Zamora et al. (Fri,) studied this question.