Impact of preoperative hyponatraemia on paediatric perioperative mortality.

Abstract

Preoperative hyponatraemia is an independent risk factor for postoperative mortality in adults. To our knowledge, this has not been investigated in children. Using data from the 2014 and 2015 data sets of the American College of Surgeons National Surgical Quality Improvement Program-Pediatric (NSQIP-P), we conducted a retrospective study of children undergoing surgery. The primary outcome was 30-day all-cause mortality. The secondary outcomes of interest were postoperative seizure within 30 days and prolonged length of stay. To identify any independent association between preoperative hyponatraemia, defined as mild (serum sodium of 131-135 mEq L^-1) or severe (≤130 mEq L^-1), and death, postoperative seizures, or prolonged length of stay, multivariable logistic regression models were generated. A total of 152 894 patients were identified, and of these 35 291 were included in the final analysis. Preoperative hyponatraemia was present in 5422 patients or 15.4% of the final cohort. There were 432 (0.80%) deaths at 30 days. Compared with patients with a normal preoperative sodium concentration, those with mild (P=0.003; odds ratio [OR]: 1.59; 95% confidence interval [CI]: 1.17-2.18) and severe (P=0.002; OR: 2.16; 95% CI: 1.32-3.54) hyponatraemia had increased rates of death, after adjusting for co-morbidity and procedural complexity. Both mild (P<0.001; OR: 1.53; 95% CI: 1.42-1.65) and severe (P<0.001; OR: 1.51; 95% CI: 1.19-1.93) hyponatraemia were independently associated with prolonged length of stay, after adjusting for relevant co-variates. This retrospective analysis identified an association between preoperative hyponatraemia and perioperative mortality and length of stay in paediatric patients.