Parenteral nutrition-induced hepatobiliary dysfunction in infants and prepubertal rabbits

We analyzed clinical, biochemical, and histologic parameters of ten infants with parenteral nutrition-induced hepatobiliary dysfunction. The data were compared with the results of a rabbit model. All infants were born prematu rely with low birth weight. Their clinical diagnoses were necrotizing enter ocolitis (6), gastroschisis (1), intrauterine volvulus (1), and lung hypopl asia (2). All required total (TPN) or partial parenteral nutrition for at l east 8 weeks. All had repeated episodes of infections or sepsis. A rise in bilirubin and aminotransferase levels occurred after a minimum of 5 weeks; peak bilirubin levels ranged from 4 to 14 mg% and aminotransferases from 40 to 140 IU/I. One child later developed gallstones. Liver biopsies after 1 to 24 months showed fibrosis, bile-duct proliferation, cholestasis, and hyd ropic degeneration. All of the above-mentioned clinical factors have been a ccused of causing the observed biochemical and histologic changes. In our r abbit model we were able to produce almost identical symptoms by TPN alone: gallbladder distension, sludge, and stones developed after 1-4 weeks of TP N as well as uncharacteristic changes in aminotransferases and bilirubin af ter 4 weeks. Liver histology revealed severe hydropic degeneration of zone 3 as early as 1 week after beginning TPN. A rise of fibrosis and bile-duct proliferation after 1 to 4 weeks of infusion was statistically significant. Cholestasis, as was observed in the infants, could not be detected. In our model, all alterations observed could be attributed exclusively to TPN. We therefore assume that TPN was the true cause of the dysfunction. In a seco nd experimental series infusions were reduced to 80% PN and free access to lab chow. These animals produced normal feces, indicating physiologic enter al stimulation. They developed the same degenerative and proliferative hist ologic changes, whereas gallbladder distension, sludge, and stones were not noted. We conclude that: (1) The TPN solution itself is responsible for th e histologic changes in the liver, which is supported by the fact that hydr opic degeneration of zone 3 is typical of a direct toxic effect; and (2) Co mplete enteral starvation with an absence of enteral stimulation causes dis ease of the lower biliary tract.