Fluid intake and mortality: drinking in the data

Abstract

Due to the scale of chronic kidney disease (CKD) [1], management of patients is delivered from a broad range of settings. Research into CKD is often focused towards interventions relevant to the later stages of disease seen in secondary care and can fail to acknowledge the large quantities of clinical input delivered in a community setting. Furthermore, while medical students are still taught a hierarchy of ‘conservative, medical, then surgical treatments’, guidelines for the management of CKD have focused on patient identification and pharmacological treatments [2] with scant data on public health interventions that may help ameliorate the growing problem of this condition at a ‘pre-health-care’ level. In this issue, Strippoli et al. [3] address the received wisdom that drinking increasing amounts of fluid each day is associated with health benefits. This long-standing belief, repeated even in current governmental advice [4], is believed to originate from an opinion statement, made in the absence of robust epidemiological data, by the 1945 US Food and Nutrition Board of the National Research Council [5] recommending a daily fluid intake of 2.5 L. Using information onestimated daily fluid intake obtained as part of the Blue Mountains Eye Study (1992–1994), the authors considered the effect of increased fluid intake (importantly estimated from both fluids and prepared food as stated in the 1945 guidance) in relation to mortality and loss of estimated glomerular filtration rate (eGFR). In a population of 3858 patients with a median follow-up period of 13.1 years, greater volumes of fluid intake were not associated with reduced risk for death when fluid consumption was considered as a continuous variable {hazard ratio for death 0.99 [95% confidence interval (95% CI) 0.98–1.01] per 250 mL increase in total daily fluid intake}, nor when upper and lower quartiles of fluid intake were compared. A similar result was seen in the 1479 patients with multiple measurements of renal function. Here, a non-significant increase in eGFR was associated with increasing daily fluid intake [0.06 mL/min/1.73 m 2 for every 250 mL increase in daily fluid take (95% CI �0.03 to 0.14)].