Objective: The objective of this study was to evaluate the clinical efficacy and appropriateness of colistin therapy in patients with hematological malignancies. Methods: Age, gender, type of hematologic malignancy, and potential carbapenem-resistant microorganism risk factors were all noted in this retrospective study. In empirical and agent-specific treatment groups, differences in demographic features, risk factors, treatment responses, and side effects were compared. Results: Sixty-three patients were included, 54% were male, and the median age was 49. In the last three months, the hospital ization rate history was 68%, and four patients had a hospitalization history in the ICU. Carbapenem-resistant K. pneumoniae col onization was present in 22 patients (35%). Gram-negative microorganisms were isolated in 34 patients (54%). The carbapenem, quinolone, and colistin resistance rates were 82%, 76%, and 4% respectively. Clinical and microbiological response rates were 60% and 69%. 7 and 28-day mortality rates were 17% and 35%. There was no significant difference in demographic data and comorbid ities in empirical (n=48) and agent-specific (n=15) treatment groups. The rate of carbapenem and glycopeptide treatments before colistin was higher in the empirical treatment group (p = 0.004; p = 0.001). The rate of starting combined antibiotics was higher in the empirical treatment group (p = 0.016). Two of the patients developed renal failure in the first week after treatment. Conclusion: The use of empirical colistin may be unavoidable given the risk considerations. Shortly, colistin-resistant strains may also be a factor affecting treatment success negatively.