P1164 CONTEMPORARY RATES, RISK FACTORS AND OUTCOMES OF HEPATIC DYSFUNCTION IN INFANTS ON PARENTERAL NUTRITION THERAPY.

Abstract

Introduction: The efficacious use of Parenteral Nutrition (PN) therapy was initially reported three decades ago. However, shortly thereafter, hepatic dysfunction was recognized as a major clinical problem and frequent indication for termination of PN therapy. Major changes have since occurred in PN including change from 10% to 20% Intralipid emulsions, increased use of Trophamine as the aminoacid source in infants, and advances in tertiary care. The objective of this study was to re-evaluate the contemporary rates, risk factors and outcomes of PN-associated hepatic dysfunction (HD) in infants. Methods: Retrospective study of all infants (age <12-months) that received PN therapy during the 12-month period from September 2002 to August 2003 at a tertiary care Children’s Hospital. Infants on prolonged PN (>2-wks) were reviewed for birth wt, diagnosis, duration and mortality during PN therapy. Medical risk factors were categorized as low birth wt (<1500 g); sepsis (bacteremia, pneumonia, urosepsis); intestinal failure (NEC, gastrointestinal surgery, short gut syndrome); and cardio-respiratory failure (CRF: surgical congenital heart disease and severe pulmonary hypertension). PN-associated HD was defined as rise in conjugated bilirubin to >1.5 mg/dl during PN therapy. Prevalence and resolution of HD were determined. Logistic regression analysis was used to determine the effect of medical risk factors and duration of PN therapy on onset and resolution of HD. Significance wad determined as p <0.05. Results: 508 infants received PN therapy during the 12-m period; and 102 (birth wt 560–3800 g) received PN for a duration >2-wks. The prevalence of HD in subjects receiving prolonged PN therapy (>2-wks) was 37.5% (95% CI: 27.9, 47.4%). Duration of PN therapy for >4-wk was associated with a four-fold risk of HD compared to duration of 2 to 3-wk (OR 4.4; 95% CI: 1.75, 11). Duration of PN was the most significant risk factor for onset of HD (p = 0.0031), along with CRF (p = 0.0042). After controlling for duration on PN, HD was 5.6 times (95% CI: 1.7, 17.8) more likely to resolve in infants with CRF. CRF was the risk factor most associated with mortality while on PN therapy (p = 0.0002). Conclusion: HD is a common complication in ill infants on prolonged PN therapy; however, mortality while on PN therapy most commonly occurred in infants with CRF. Duration of PN therapy was the major risk factor associated with onset, resolution and persistence of HD in infants.