(852) Incidence and Risk Factors for Hyponatremia in Pediatric Patients Admitted With TBI

Introduction: More than 50,000 children are hospitalized with traumatic brain injury (TBI) each year in the United States. Morbidity in these pts is due to axonal injury, intracranial bleeding and vasogenic, cytotoxic edema. Sodium dysregulation is a potentially avoidable cause and has been associated with worse neurological outcomes in adults. Target sodium levels are unclear and have been described with reference to use of hyperosmolar therapy in TBI guidelines however there are limited studies looking at incidence and risk factors for low serum sodium in these pts.

Methods: Single-center retrospective study, looking at pts admitted with TBI to PICU at Oishei Children’s hospital, Buffalo from 2014 – 2019. ICD codes for intracranial injury were used for screening and pts with < 24h stay, existing intracranial lesions, CHF and BPD were excluded. Hyponatremia was defined as Na < 135 meq/L during first 14 days of hospital stay.

Results: 115 pts were reviewed and they were equally distributed across age groups. Mechanisms of injury in order of frequency were - fall from height, struck by blunt force, suspected child abuse and pedestrian struck by motor vehicle. Initial neuroimaging findings showed - subdural hemorrhage was seen most often (48 % pts), followed by intraparenchymal, epidural, subarachnoid and intraparenchymal hemorrhage respectively.
13 % pts had hyponatremia and extradural, subdural hemorrhage, presence of cranial fractures were found to be associated with increased risk. None of the mechanisms of injury were found to be associated with low sodium. 3 pts died and 2 of them had hyponatremia at least once during their course. Median LOS, discharged to rehab facility, frequency of G-tube placement was similar between normal and low serum sodium groups.

Conclusions: Incidence of hyponatremia in our cohort was lower compared to previous studies. This could be explained by recent widespread use of isotonic fluids like N.S in these pts. The association with certain neuroimaging findings could be reflective of the severity of injury. Some potential confounders like type, volume of iv fluids and nutrition, were not reviewed as their timings of initiation were not well documented. Further prospective studies are needed to look at other risk factors and neurological outcomes in these pts.