Can effluent hyaluronic acid or creatine kinase predict sinusoidal injury severity after cold ischemia?

It is well recognized that current selection criteria used to assess liver grafts before implantation are inaccurate and correlate poorly with graft outcome. A bench or laboratory-based test that could indicate the extent of liver injury immediately before implantation would be a valuable adjunct to clinical assessment. Hyaluronic acid (HA) and creatine kinase (BB component; CK-BB) levels in the caval effluent after liver perfusion have been suggested as indicators of preservation injury. Our objective was to investigate the relevance of preserved liver effluent HA and CK-BB as a predictor of early graft function. Perfused liver effluent HA and CK-BB levels were measured. Graft function was measured in terms of peak serum aspartate transaminase and its level on day 5 postoperatively as well as peak bilirubin level and prothrombin time. The cold ischemia time (CIT) was recorded. Statistical comparisons were made among HA level, CK-BB level, CIT, and graft function parameters. The study was conducted at The Liver and Hepatobiliary Unit, Queen Elizabeth Hospital, Birmingham, United Kingdom. Fifty patients undergoing OLT were studied. HA level was measured in 50 patients and CK-BB level in 30 patients. The main outcome measures were graft function and graft outcome. The graft function data are grouped according to effluent HA levels above or below 400 micrograms/L. Thirteen patients (26%) had a level below 400 micrograms/L and the remaining 37 (74%) were above this threshold. There were no significant differences between the groups for these indicators of graft function. There was no difference between the 2 groups for CIT. The overall median HA level was 1212 micrograms/L (range 39-4000 micrograms/L). The median total CK activity in the perfusate was 302 IU/L (range 118-1155 IU/L). The proportion of CK-BB activity from this total was 146 IU/L (8-641 IU/L), or 48% of the total CK activity. In a multiple regression analysis with CK-BB activity as the dependent variable, there was no demonstrable numerical relationship to graft function. In a separate multiple regression analysis similar results were obtained for HA. We conclude that the level of HA or CK-BB levels should not be used in determining the suitability for implantation of a harvested hepatic allograft.