Complications associated with arterial catheterization in children

Abstract

Objectives: To examine the prevalence of and risk factors associated with arterial catheterization complications in a large pediatric patient population in an effort to generate hypotheses for future prospective study of arterial catheter placement.

Design: Retrospective cohort study.

Setting: Patients discharged between January 1, 2000, and March 31, 2005, from 33 children's hospitals belonging to the Child Health Corporation of America.

Patients: Patients were 10,394 children identified from the Pediatric Health Information System database. Inclusion criteria included age 1 month to 18 yrs, admitted to a pediatric intensive care unit, received an arterial catheter for monitoring, and hospitalized for >or=1 day following catheter placement.

Interventions: None.

Measurements and main results: We assessed complications as defined by ICD-9 coding associated with arterial catheterization, including thrombosis, embolism, and infection. Complications were reported in 10.3% (1,072) of patients, most frequently infection/inflammation (61.8%), complication of vascular device not otherwise specified (14.9%), mechanical complications (14.1%), and embolic or thrombotic issues (7.5%). Independent predictors of complications associated with arterial catheterization were age (compared with 1-4 months) of 5-11 months (odds ratio [OR] 1.5; 95% confidence interval [CI] 1.25-1.82) or 1-2 yrs (OR 1.39; 95% CI 1.09-1.78), insertion of catheters after the first hospital day and need for cardiac surgery (OR 1.31; 95% CI 1.03-1.68), bone marrow transplantation (OR 1.79; 95% CI 1.19-2.70), and dialysis (OR 1.36; 95% CI 1.05-1.77). There was no association of arterial catheter complications with patient gender, Medicaid status, or presence of coagulopathy or shock.

Conclusions: Complications associated with arterial catheterization are common in critically ill children. Significantly, we were unable to account for the potential confounding effect of central venous catheterization in this study secondary to limitations of ICD-9 coding. This study serves as a hypothesis-generating report of a large pediatric sample and suggests the need to carefully assess arterial catheter-associated complications in a prospective study independent of central venous catheters.