Background: The benefits and burdens of artificial nutrition (AN) and artificial hydration (AH) in end-of-life care are unclear. We carried out a literature review on the use of AN and AH in the last days of life of cancer patients. Materials and methods: We systematically searched for papers in PubMed, CINAHL, PsycInfo and EMBASE. All English papers published between January 1998 and July 2009 that contained data on frequencies or effects of AN or AH in cancer patients in the last days of life were included. Results: Reported percentages of patients receiving AN or AH in the last week of life varied from 3% to 53% and from 12% to 88%, respectively. Five studies reported on the effects of AH: two found positive effects (less chronic nausea, less physical dehydration signs), two found negative effects (more ascites, more intestinal drainage) and four found also no effects on terminal delirium, thirst, chronic nausea and fluid overload. No study reported on the sole effect of AN. Conclusions: Providing AN or AH to cancer patients who are in the last week of life is a frequent practice. The effects on comfort, symptoms and length of survival seem limited. Further research will contribute to better understanding of this important topic in end-of-life care.

We studied the effect of the Liverpool Care Pathway (LCP) on the documentation of care, symptom burden and communication in three health care settings. Between November 2003 and February 2005 (baseline period), the care was provided as usual. Between February 2005 and February 2006 (intervention period), the LCP was used for all patients for whom the dying phase had started. After death of the patient, a nurse and a relative filled in a questionnaire. In the baseline period, 219 nurses and 130 relatives filled in a questionnaire for 220 deceased patients. In the intervention period, 253 nurses and 139 relatives filled in a questionnaire for 255 deceased patients. The LCP was used for 197 of them. In the intervention period, the documentation of care was significantly more comprehensive compared with the baseline period, whereas the average total symptom burden was significantly lower in the intervention period. LCP use contributes to the quality of documentation and symptom control.