Natraemia, chloraemia, kalaemia, magnesemia, phosphatemia, ionized calcaemia, azotaemia, albuminaemia, osmolarity, lactataemia, arterial gases and thing haematocrit were measured immediately before and at 6, 12, 24, 36 and 48 hours after starting the treatment. The total volume of fluid administered and the evolution of ICP were recorded during the study period (48 hours). Episodes of ICH, modifications on the control CT (bleeding, herniation or brain swelling), osmotherapy and/or barbiturate use, transfusion, vasopressor use, time to achieve more than 50% of goal calories of enteral nutrition, duration of mechanical ventilation, length of ICU stay and mortality rate were also recorded in the ICU. Safety was assessed by recording adverse events.

DefinitionsStrong ion difference (SID) was defined as (Na+ + K+ + Ca2+ + Mg2+) – (Cl- + lactate) mEq/L [22]. Hyperchloraemic metabolic acidosis was defined as SID below 40 mEq/L associated with chloraemia above 108 mmol/L according to local laboratory normal ranges.EndpointsThe primary endpoint was the occurrence of hyperchloraemic metabolic acidosis within 48 hours. The secondary outcomes were electrolyte status, ICP, rate of ICH episodes, volume of intravenous fluid, duration of vasopressor therapy, duration of mechanical ventilation, length of ICU stay and death in the ICU.Statistical analysisTo the best of our knowledge, the incidence of hyperchloraemic acidosis in brain-injured patients has not been documented to date. We have thus performed a post hoc analysis of the chloraemia values collected in a study of TBI patients with ICH receiving HSS [11].

We found a 65% incidence of hyperchloraemia within the first four days in the ICU before any HSS infusion. The sample size needed to detect a 45% decrease in the incidence of hyperchloraemic acidosis, assuming a basal rate of 65% in a two-sided test performed with a statistical power of 85% and an �� risk of 0.05, was 20 patients in each group in this pilot study. Taking into account exclusions, and in an attempt to keep the power of the study, 42 patients (21 patients in each group) were included.The full analysis set (FAS) of patients was the primary population used for statistical analysis of efficacy (per-protocol analysis) and was defined as all randomised patients treated with the study drug who did not receive forbidden therapy (HSS infusion).

All randomised patients (the intention-to-treat (ITT) population) were analysed for the primary outcome and safety variables.We first verified that in all patients the incidence of hyperchloraemic acidosis at 48 hours was significantly decreased in the balanced group compared with the control group using Fisher’s exact test. Six patients experienced hyperchloraemic acidosis Dacomitinib prior to inclusion (four in the saline group and two in the balanced group). We therefore decided a posteriori to perform two complementary sensitivity analyses.