To use a large administrative database to determine the mortality, risk factors, and co-morbidities of esophageal variceal bleeding in children.
Retrospective cohort study using Pediatric Health Information System (PHIS) data from 50 tertiary children’s hospitals in the US. ICD codes from 2004 through 2019 identified children 18 years and younger with variceal bleeding and complications. Univariate analyses used the Student t-test for continuous variables (age) and chi-square test for categorical variables (all others). Mixed-effects linear regression was performed for multiple variables.
1902 patients had 3399 encounters for esophageal variceal bleeding. The mortality rate for variceal bleeding was 7.3%, increasing to 8.8% by 6 weeks; any mortality during the study was 20.1%. Transfusion was required in 54.7% of encounters, and 42.6% were admitted to the ICU. Variceal bleeding encounters were complicated by peptic ulcer disease (6.9%), bacteremia (11.4%), acute renal failure (5.1%), mechanical ventilation (18%), ascites (21.3%) and peritonitis (3.3%). Multivariable mixed-effects logistic regression showed that Black race (OR 2.59, p<0.001) or Hispanic ethnicity (OR 2.31, p=0.001), but not sex, household income, or insurance type, were associated with increased mortality. Bacteremia, peritonitis, mechanical ventilation, acute renal failure, and transfusion were associated with higher mortality (OR 2.29, 2.18, 1.93, 6.33, 1.81, respectively; p<0.001, 0.005. 0.011, <0.001, 0.005).
The 6-week mortality rate for variceal bleeding in children is 8.8%. Black or Hispanic children are at higher risk of dying. Serious morbidities associated with variceal hemorrhage impact mortality. These data can inform consideration of prophylactic or therapeutic interventions for children at risk.

Author