Fluid retention index predicts the 30-day mortality in geriatric care

Abstract

The incidence and medical consequences of dehydration and fluid retention in senior citizens are unclear. The present study used urine sampling to detect renal conservation of water, which is an early sign of dehydration, and assessed its relationship to mortality in elderly patients admitted for acute hospital care. A urine sample was collected from 256 patients (mean age 82 years) and analyzed for color, specific gravity and osmolality. These markers were used to calculate a composite index of fluid retention, which was indicated by urine color ≥ 4, specific gravity ≥ 1.020 and osmolality ≥ 600 mOsmol/kg as suggested from eight previous studies of exercise-induced dehydration, of which one extends to age 69. Concentrated urine consistent with dehydration was present in 39 (16%) of the patients. This finding was relatively more common among those with confusion and/or dementia, but less common in patients with medical disease, and in those taking diuretics daily. Patients with such fluid retention had a higher 30-day mortality when compared to those who were euhydrated (21% versus 8%; p < 0.03). A difference of 10% remained at three months and one year after the admission to hospital. Concentrated urine consistent with fluid retention was found in 16% of the geriatric patients admitted to hospital for acute care. In these patients the mortality within 30 days was almost tripled compared to those who were euhydrated.

Key Words::

• Dehydration
• frail elderly
• Osmolar concentration
• specific gravity
• mortality
• aged

Acknowledgements

The authors are indebted to the Staff of the Departments of Geriatrics at Nacka and Södertälje Hospitals for their assistance with the collection of urine samples. Birgith Olsson and Johanna Pelttari provided both network and support to implement the clinical methods used. Financial support was received from the Stockholm County Council, Sweden (Grant number 2006-0791).

Declaration of interest: The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.