BACKGROUND: Colistin is used as last treatment option for pneumonia associated with multidrug-resistant (MDR) Pseudomonas spp.. Literature about the best administration mode (inhalation versus parenteral treatment) is lacking. METHODS: A retrospective study of 20 intensive care patients with a pneumonia associated with MDR P. aeruginosa receiving colistin sulphomethate sodium (Colistineb(R)) between 2007 and 2009 was performed. A strain was considered multidrug-resistant if it was resistant to at least 6 of the following antibiotics: piperacillin-tazobactam, ceftazidime, cefepime, meropenem, aztreonam, ciprofloxacin, and amikacin. The administration mode, predicted mortality based on the SAPS3 score, SOFA score at onset of the colistin treatment, clinical and microbiological response, and mortality during the episode of the infection were analysed. The non parametric Kruskal-Wallis and Fisher's Exact test were used for statistical analysis of respectively the predicted mortality/SOFA score and mortality rate. RESULTS: Six patients received colistin by inhalation only, 5 were treated only parenterally, and 9 by a combination of both administration modes. All patients received concomitant beta-lactam therapy. The mean predicted mortalities were respectively 72%, 68%, and 69% (p = 0.91). SOFA scores at the onset of the treatment were also comparable (p = 0.87). Clinical response was favorable in all patients receiving colistin by inhalation (6/6) and in 40% (2/5) of the patients receiving colistin parenterally (p = 0.06). In the patients with colistin administered both via inhalation and parenterally, clinical response was favorable in 78% of the patients (7/9) (p = 0.27 as compared to the treatment group receiving colistin only parenterally). When all patients with inhalation therapy were compared to the group without inhalation therapy, a favorable clinical response was present in respectively 87% and 40% (p = 0.06). In none of the patients, the Pseudomonas spp. was eradicated from the follow-up cultures.All patients in the parenterally treated group died. None of the patients receiving colistin by inhalation, and 3 of 9 patients of the combination group eventually died (p = 0.002 and p = 0.03 respectively as compared to the group receiving colistin only parenterally). CONCLUSIONS: Aerosolized colistin could be beneficial as adjunctive treatment for the management of pneumonia due to MDR P. aeruginosa.