Objectives: The aims of the present evaluation were to determine whether the elevations in liver enzymes observed in phase-I trials are more common in subjects being hospitalised than in ambulatory subjects and to assess the relevance of these elevations. We therefore investigated the effect of hospitalisation on liver enzyme levels in subjects of all phase-I trials of sufficient length performed at the two Human Pharmacology Centres of Boehringer Ingelheim, located in Biberach and Ingelheim, Germany, over a 10-year period.
Methods: The evaluation was based on 29 phase-I trials conducted between 1987 and 1996. These trials consisted of at least 4 days of observation of 220 subjects on placebo treatment in 273 trial participations. The mean changes (transformed into reference ranges) in the liver enzymes alkaline phosphatase, gamma glutamyl transpeptidase, aspartate aminotransferase, alanine aminotransferase and the AST/ALT ratio of hospitalised subjects were compared with those of ambulatory subjects.
Results: A small but statistically significant increase in the mean of the parameters GGT, AST and ALT during phase-I trials was observed in hospitalised subjects compared with ambulatory subjects. The increases were: 8% of the reference range width for GGT, 8% for AST and 17% for the ALT, although the differences were small in absolute terms. A relevant increase (defined as an increase to above the reference range or by at least 50% of the reference range) was more common in hospitalised subjects.
Conclusion: Hospitalisation, although it is believed to protect subjects from confounding environmental factors, can itself be associated with increases in liver enzyme levels. The definition of a relevant increase can be used to identify subjects who need further evaluation. The cause of the increase in liver enzyme levels remains unclear.