(431) Vitamin K Prescribing Trends Among Critically Ill Children With Sepsis: A Retrospective Cohort Study

Introduction: Vitamin K has been anecdotally prescribed for sepsis-related coagulopathy. Yet, no reports have described a vitamin K deficiency in this population nor do recommendations exist to support its application in sepsis. We sought to characterize the prescribing trends of vitamin K among critically ill children hospitalized for sepsis using a large administrative registry.

Methods: We performed a multicenter retrospective cohort study using the Pediatric Health Information System registry including children 0-17 years of age hospitalized for sepsis in the pediatric intensive care unit of 48 referral centers from 2016–2022. The primary outcome was vitamin K prescribing rates assessed chronologically by calendar year and regionally by hospital center. Descriptive data included demographics, clinical trajectory, vitamin K deficiency, clinically relevant bleeding (estimated by red blood cell [RBC] transfusions), and comorbid hepatic insufficiency. Descriptive statistics included Wilcoxon rank-sum, student’s t, and chi-square tests. Chronologic trends were assessed using JoinPoint regression.

Results: Of the 31,221 encounters studied, 4,539 (14.6%) were prescribed vitamin K (median center-specific prescribing rate: 14.2%; interquartile range [IQR]: 8.8, 21). We noted a linear chronologic prescribing trend from 13% in 2016 to 14.5% in 2022 (+0.14%/year, R^2 = 0.1). Those prescribed vitamin K had greater rates of hepatic dysfunction (20.5% vs 3.1%), postoperative care (38.9% vs 18.3%), mechanical ventilation (71.5% vs 43.2%), central venous catheterization (45.6% vs 24.5%), mortality (17.1% vs 4.4%), RBC transfusion (26.5% vs. 11.2%), and length of stay (median 16 [IQR:8,33] vs 8 [4,15] days) (all P< 0.001). Vitamin K deficiency was documented in only 1.3% of encounters who received the drug and 0.2% for the entire study cohort.

Conclusions: In this multicenter cohort study, vitamin k prescribing was common among critically ill children with sepsis despite the absence of documented vitamin deficiency. Prospective observational data are needed to estimate the presence of vitamin k deficiency in this population that would inform clinical efficacy trials of vitamin K as a means to mitigate clinically relevant bleeding and support this indication.