Infections due to multidrug-resistant (MDR) Gram-negative bacteria have been increasing and they are an important cause of nosocomial morbidity and mortality, especially in immunocompromised patients. In order to determine efficacy and safety of colistin (colistimethate sodium) use in the treatment of MDR Pseudomonas aeruginosa sensitive to colistin, a comparison of renal function, other toxicities, and outcome of therapy was done between a group of patients treated with colistin and patients treated with other antipseudomonas drugs as control group. A group of 26 patients that was hospitalized in our institution between February 2002 and December 2006 and treated with intravenous colistin for an infection caused by MDR P.aeruginosa was compared in a matched-pair analysis to a group of 26 patients treated with other antipseudomonas drugs. Patients were 52% male and 48% female; mean age was 37 years (range 17–63). All of the patients were treated for haematological malignancy, most received intensive chemotherapy regimens (44%), 19% received allogeneic and 31% autologous transplants. Groups of patients did not differ in age, sex, disease, or kind of treatment received. All of the patients in both groups had clinical signs of sepsis; in 69% of patients from colistin group and 84% from control group P.aeruginosa was isolated from blood, and in 27% and 12% it was isolated from skin lesions that had clinical presentation of echtyma gangrenosum, respectively. Patients treated with colistin received 3 MU of colistin every 8 hours for a mean (± SD) duration of 12.5 (± 5.4) days. Due to nature of their disease, and severity of infections, all of the patients received more than two other possibly nephrotoxic drugs; in colistin group 4 other concomitant drugs, on control group 3; most frequently vancomycin, cefepime, amikacine, garamycine and amphotericine B deoxycholate was used. Of 26 patients receiving colistin, 76.9% of patients had the drug discontinued after successful clearance of infection, while in control group 65.4% of patients had the drug discontinued due to same reason. Only one patient had displayed neurological toxicity (Jacksons attack with secondary generalisation), but the drug was not discontinued, dose was modified, patient had no further attacks. There was no statistically significant difference in the level of serum creatinine, creatinine clearance (calculated), or potassium levels between prior to therapy and after treatment discontinuation between groups. One patient treated with colistin developed renal failure and was subjected to continuous venovenous hemodiafiltration; of note is that at the time colistin introduction patient already had impaired renal function. In one patient drug was discontinued due to suspected allergic reaction. No other adverse events of colistin therapy were noted. Colistin is an effective antimicrobial drug for the treatment of severe infections caused by MDR P.aeruginosa in haematological patients. The safety profile observed is acceptable in these severe life-threatening infections, in matched-pair analysis it did not display greater toxicity than other antipseudomonas drugs. Further studies are needed to better address the treatment of MDR P. aeruginosa, naimely the optimal dose and schedule, also route of administration of colistin, as well as drug-to-drug interactions.