88. New onset central diabetes insipidus in pediatric scoliosis fusions utilizing tranexamic acid

Abstract

<h3>BACKGROUND CONTEXT</h3> Central diabetes insipidus (CDI) is a polyuric disease characterized by renal excretion of large volumes of dilute urine, plasma volume contraction, hypernatremia and serum hyperosmolality. CDI occurs frequently after pituitary surgery and traumatic brain injury, and has been rarely reported after significant spinal cord trauma. CDI occurring in pediatric surgery during spinal fusion has only been sporadically reported and, to date, we found only one reported case of CDI with a documented suppressed vasopressin level that developed during spine fusion for idiopathic scoliosis in a pediatric patient. CDI presenting during anesthesia has been related to medications commonly used in anesthesia such as propofol, dexmedetomidine, sevoflurane, ketamine and opioids. Our scoliosis surgery protocol uses total intravenous anesthesia (TIVA) with infusions of remifentanil, propofol, ketamine and tranexamic acid (TXA) (loading dose 10 mg/kg, maintenance at 1 mg/kg/hour, later increased to 5 mg/kg/hour). Since increasing the TXA dosing, we have documented 9 cases of intraoperative CDI marked by sudden development of polyuria. We did not have any documented cases of CDI prior to TXA use or at the lower dose of 1mg/kg/hour. <h3>PURPOSE</h3> To document the incidence of CDI in pediatric patients undergoing posterior scoliosis utilizing TXA. <h3>STUDY DESIGN/SETTING</h3> Retrospective chart review. <h3>PATIENT SAMPLE</h3> The study included 599 posterior scoliosis surgery cases performed by three different surgeons between March 2012 and November 2017 at a tertiary pediatric hospital. <h3>OUTCOME MEASURES</h3> Cessation of polyuria and correction of sodium elevation. There were no cases of CDI postoperatively. <h3>METHODS</h3> Retrospective chart review. <h3>RESULTS</h3> Of 599 posterior spine fusions, 9 cases were diagnosed with introperative CDI. Five of nine of there curves were idiopathic and females predominanted. Mean urine output prior to TXA infusion was 1.1 ml/kg/hour and increased to 4.2 ml/kg/hour within 1-3.5 hours of the start of TXA infusion. Concomitantly, mean plasma sodium increased from 137 ± 3.2 mEq/L to 142 ± 3.5 mEq/L. All responded to vasopressin infusion with TXA continuation orcessation of TXA alone. No cases of postoperative CDI were seen. Group 1 (TXA at 1 mg/kg/hr) had 0/363 cases. Group 2 (TXA at 2-5mg/kg/hr) had 9/236 cases. A Fisher's exact test for comparing the Group 1 rate (0%) vs Group 2 rate (3.8%) had a significant p-value of <0.001. <h3>CONCLUSIONS</h3> This is the first reported series of intraoperative CDI development in pediatric deformity correction surgery utilizing TXA infusion. The evidence suggests a dose-dependent association between TXA and CDI. Surgeons and anesthesiologists need to be aware that TXA in pediatric scoliosis surgery can lead to intraoperative CDI. Rapid treatment with vasopressin or cessation of infusion can prevent potential sequelae. <h3>FDA DEVICE/DRUG STATUS</h3> Tranexamic Acid (Approved for this indication)