OBJECTIVE We present our experience with five cases of pandrug-resistant Pseudomonas aeruginosa ventilator-associated pneumonia (VAP) and analysis of risk factors. METHODS Case-control study in a 15-bed intensive care unit (ICU). METHODS The study included 5 cases and 20 controls. Each case patient was matched to four contemporary controls according to gender, prior hospital admissions, hospitalization duration, ICU admission cause, Acute Physiology and Chronic Health Evaluation (APACHE) II and Sequential Organ Function Assessment (SOFA) scores on ICU admission, and length of ICU stay, and mechanical ventilation duration until first VAP episode by a multidrug-resistant bacterium. RESULTS Recorded variables included age, gender, daily APACHE II and SOFA scores, patient medication, treatment interventions, positive cultures and corresponding antibiograms, occurrence of infection, sepsis, and septic shock, other ICU-associated morbidity, length of ICU stay and mechanical ventilation, and patient outcome. Healthcare worker and environmental cultures, and a hand-disinfection survey were performed. Pandrug-resistant P. aeruginosa isolates belonged to the same genotype and were bla (VIM-1)-like gene positive. The outbreak resolved following reinforcement of infection-control measures (September 27). The sole independent predictor for pandrug-resistant P. aeruginosa VAP was combined use of carbapenem for more than 20 days and colistin use for and more than 13 days (odds ratio 76.0; 95% confidence interval 3.7-1487.6). An additional risk factor was more than 78 open suctioning procedures during 6-26 September (odds ratio 16.0; 95% confidence interval 1.4-185.4). CONCLUSIONS Prolonged carbapenem-colistin use predisposes to VAP by pandrug-resistant P. aeruginosa. Cross-transmission may be facilitated by open suctioning.