Serum sodium concentration predicting mortality in patients with aneurysmal subarachnoid hemorrhage

•The impact of hypernatremia on surgical patients with aneurysmal subarachnoid hemorrhage remains inconsistent.•We apply multivariable logistic regression and propensity score matching to minimize the impact of confounding bias.•In this patient group, hypernatremia is independently linked to higher risks of all-cause mortality and inferior course. To date, inconsistent evidence exists on the role of hypernatremia in patients with aneurysmal subarachnoid hemorrhage (aSAH) who underwent surgical clipping. We aimed to investigate the association between serum sodium and mortality in these patients. A cohort study was performed to include adult patients with aSAH who underwent surgical clipping in a university hospital. The primary outcome was follow-up mortality. Propensity score matching (PSM) was used for matching patients’ baseline characteristics. Net reclassification improvement (NRI), and integrated discrimination improvement (IDI) were calculated to assess and compare the reclassification and discrimination capacity of different models. Trends in serum sodium over time were detected by the ordinary least squares model. Of 618 aSAH patients with surgical clipping during the study period, normal serum sodium was observed in 467 patients (75.6 %), and admission hypernatremia was noted in 151 patients (24.4 %). After adjustment with multivariate regression analysis, patients with hypernatremia had significantly higher odds for follow-up mortality (aOR: 2.86, 95 % CI: 1.54 to 5.30; P = 0.001). PSM analysis observed similar results (aOR: 2.38, 95 % CI: 1.29 to 4.55; P = 0.009). The incorporation of serum sodium during hospitalization markedly enhanced the IDI (P