Background: We aimed to examine whether inflammatory markers and various risk scoring systems could be utilized to predict the prognosis of breast cancer patients hospitalized in the intensive care unit. Methods: In this retrospective cohort study, we evaluated 71 breast cancer patients followed and treated by the Medical Oncology Department of Medipol University Faculty of Medicine who were admitted to the intensive care unit at any stage of their treatment between 2014 and 2020. In addition to scoring systems developed to assess prognosis in the ICU, we recorded and calculated various inflammation-related markers. All data were compared between groups formed according to survival status. Results: The median age was 58 (range: 33-90) years. Thirty-seven (52.1%) patients had died and 34 (47.9%) patients had survived. Neutrophil lymphocyte ratio (p = 0.021), Modified Glasgow prognostic score (p < 0.001), APACHE II score (p < 0.001) and MPM II (admission) (p < 0.001) were significantly higher in the exitus group than in the survivors. Lymphocyte monocyte ratio (p = 0.030) and prognostic nutritional index (p = 0.004) were significantly higher in the discharged group than in the exitus group. When we evaluated performance of the prognostic scores to predict mortality, we found APACHE II score (AUC: 0.939, 95% CI: 0.888 - 0.990), MPM II-Admission (AUC: 0.936, 95% CI 0.880 - 0.992) and mGPS (AUC: 0.727, 95% CI: 0.600 - 0.854) had the highest area under curve values. Multivariable regression revealed that longer chemotherapy duration (≥2 weeks), having an mGPS score of 2 points, and high MPM-II (≥36 points) were independently associated with mortality. Conclusions: Among the inflammatory markers and scores examined, mGPS and MPM-II were found to be independently associated with mortality in breast cancer patients hospitalized in the intensive care unit. In addition, patients with longer chemotherapy duration had a higher risk of mortality, but this result was limited by various possible confounders.