Hypophosphatemia in critically ill patients undergoing Sustained Low-Efficiency Dialysis with standard dialysis solutions

Background In patients admitted to the Intensive Care Unit (ICU), Kidney Replacement Therapy (KRT) is an important risk factor for hypophosphataemia. However, studies addressing the development of hypophosphatemia during prolonged intermittent KRT modalities are lacking. Thus, we evaluated the incidence of hypophosphatemia during Sustained Low-Efficiency Dialysis (SLED) in ICU patients; we also examined the determinants of post-SLED serum phosphate level (s-P) and the relation between s-P and phosphate supplementation and ICU mortality. Methods We conducted a retrospective analysis on a cohort of critically ill patients with severe renal failure and KRT need, who underwent at least three consecutive SLED sessions at 24-72 h time intervals with daily monitoring of s-P concentration. SLED with Regional Citrate Anticoagulation (RCA) was performed with either conventional dialysis machines or continuous-KRT monitors and standard dialysis solutions. When deemed necessary by the attending physician, intravenous phosphate supplementation was provided by sodium glycerophosphate pentahydrate. We used mixed-effect models to examine the determinants of s-P and Cox proportional hazards regression models with time-varying covariates to examine the adjusted relation between s-P, intravenous phosphate supplementation and ICU mortality. Results We included 65 patients [mean age 68 years (SD 10.0); mean Acute Physiology and Chronic Health Evaluation II score 25 (range 9-40)] who underwent 195 SLED sessions. The mean s-P before the start of the first SLED session (baseline s-P) was 5.6 +/- 2.1 mg/dL (range 1.5-12.3). Serum phosphate levels at the end of each SLED decreased with increasing age, SLED duration and number of SLED sessions (P < .05 for all). The frequency of hypophosphatemia increased after the first through the third SLED session (P = .012). Intravenous phosphate supplementation was scheduled after 12/45 (26.7%) SLED sessions complicated by hypophosphataemia. The overall ICU mortality was 23.1% (15/65). In Cox regression models, after adjusting for potential confounders and for current s-P, intravenous phosphate supplementation was associated with a decrease in ICU mortality [adjusted hazard ratio: 0.24 (95% confidence interval: 0.06 to 0.89; P = 0.033)]. Conclusions Hypophosphatemia is a frequent complication in critically ill patients undergoing SLED with standard dialysis solutions, that worsens with increasing SLED treatment intensity. In patients undergoing daily SLED, phosphate supplementation is strongly associated with reduced ICU mortality.