Artificial hydration therapy, laboratory findings, and fluid balance in terminally ill patients with abdominal malignancies.

To explore the association between hydration volume and laboratory findings, and between calculated fluid balance and changes in clinical signs of dehydration and fluid retention in terminally ill cancer patients, a secondary analysis of a large multicenter, prospective, observational study was performed. The study enrolled 125 abdominal cancer patients who received laboratory examinations in the last week before death. Patients were classified into two groups: the hydration group (n = 44), who received 1L or more of artificial hydration per day both 1 and 3 weeks before death, and the nonhydration group (n = 81). The mean albumin level 1 week before death was significantly lower in the hydration group than in the nonhydration group, and the interaction between hydration group and decrease in the albumin level was statistically significant after adjusting multiple covariates (from 2.8 +/- 0.68 mg/dL 3 weeks before death to 2.4 +/- 0.56 mg/dL 24 hours before death in the hydration group vs. a decrease of 2.8 +/- 0.53 to 2.6+ /- 0.45 mg/dL in the nonhydration group, P = 0.015). There was no significant difference between the groups in the mean blood urea nitrogen/creatinine, sodium, or potassium levels 1 week before death. Among 53 patients who had oral fluid intake of less than 500 mL/day throughout the last 3 weeks and completed a fluid balance study, the median of calculated fluid balance was -400 mL/day 3 weeks before death, -521 mL/day 1 week before death, and -421 mL/day 24 hours before death. Calculated fluid balances did not significantly differ between the patients with deterioration of dehydration signs, edema, ascites, and pleural effusion during the final 3 weeks and those without. These data suggest that active artificial hydration might result in hypoalbuminemia, with no clear beneficial effects on normalizing blood urea nitrogen/creatinine, sodium, or potassium levels. Fluid balance did not significantly correlate with changes in dehydration-and fluid retention-signs. Calculated fluid balance is not an appropriate alternative to direct monitoring of patient symptoms. More studies are needed to determine the clinical efficacy of artificial hydration for terminally ill cancer patients.