Kidney Function Worsening Is Linked to Parenteral-Nutrition-Dependent Survival in Palliative Care Patients

Background. Parenteral nutrition (PN) is frequently administered in palliative care patients suffering from cachexia. The evidence regarding the use of PN in terminally ill patients is scarce. Routine laboratory parameters might help to decide whether to start or forgo PN, which could decrease overtreatment at the end of life. Kidney failure was frequently associated with survival. However, a relation between kidney function parameters and parenteral nutrition has not been observed thus far. The aim of this retrospective cohort study was to analyze kidney function parameters in palliative care patients under PN, as well as the relation between these parameters and overall survival. Methods. Patients who were admitted to the Department of Palliative Medicine at the Medical University of Vienna were screened for PN treatment. Whether kidney function parameters at baseline or their dynamics over the course of two weeks were associated with survival was assessed with descriptive and interferential statistics. Results. In total, 113 of 443 palliative care patients were administered parenteral nutrition for the first time. The overall survival (OS) for all patients with increased kidney function parameters at baseline was lower (creatinine: hazard ratio (HR) = 1.808, p < 0.001; urea: HR = 1.033, p < 0.001; uric acid HR = 1.055, p = 0.015). No significant increase in creatinine blood levels was observed in the first 2 weeks after the initiation of PN when compared to the non-PN group (p = 0.86). However, if creatinine blood levels increased within the PN group, lower overall survival was found (HR = 2.046, p = 0.007). Conclusion. Increased kidney function parameters, such as creatinine, urea and uric acid, might be used as negative prognostic markers in palliative care patients under PN. Moreover, an increase in creatinine during the administration of parenteral nutrition in the first 2 weeks is linked to worse outcomes. These findings may help future studies to establish objective markers for clinicians to determine whether to start or end PN in palliative cancer patients and decrease potential overtreatment at the end of life.