Indocyanine green clearance as a predictor of outcome in liver transplant patients

Introduction

Invasive measurement of indocyanine green clearance (ICG) has been shown to predict the outcome after liver transplantation in a small series of patients. A non-invasive technique using ICG plasma disappearance rate (PDR) has recently been shown to correlate well with invasive techniques. We report our experience of ICG PDR in a large cohort of liver transplant patients.

Methods

The ICG PDR was measured by the LiMON (PULSION Medical Systems) finger probe after injection of 0.25 mg/kg ICG in 79 patients after orthotopic liver transplantation. Clinical and laboratory markers of overall organ and liver graft function were recorded at the time of ICG measurement. Associations with duration of ICU admission, early graft failure and 3-month survival were also analysed. Values are presented as the median (interquartile range).

Results

Twenty-four out of 79(30%) were transplanted for acute liver failure (ALF). Eighty-one grafts were implanted, two patients undergoing retransplantation for primary graft failure. Twenty-one (27%) patients died, five of whom had ALF. Variables for the whole group at the time of ICG clearance were: APACHE II 17 (11.5–22), SAPS 46 (33–63), lactate 2 mmol/l (1.2–3.5), INR 1.4 (1.2–1.7), AST 563 IU/l (158–1268) and bilirubin 75 mmol/l (27–128).

Correlations were observed between the ICG PDR and SAPS (r = -0.45, P < 0.001), APACHE II (r = -0.23, P = 0.04), bilirubin (r = -0.57, P = < 0.001) and length of ICU stay (r = -0.32, P = 0.014). No relationship was seen with typical markers of graft function (lactate, AST and INR). In grafts implanted into patients with chronic liver disease (n = 57), the ICG correlated with SAPS and bilirubin (r = -0.55, P < 0.001; r = -0.62, P < 0.001 respectively). However, for those with ALF, only the bilirubin had an association with ICG PDR (r = -0.44, P = 0.029). The median ICG PDR in survivors was significantly higher than in those who died; 15 (10.2–21)%/min vs 9.8 (5–16.3)%/min (P = 0.01). No significant difference was seen between survivors or nonsurvivors in the ALF group, although similar significant differences were observed between survivors and nonsurvivors in the chronic liver disease group. In those whose death was directly attributable to poor graft function, the median ICG PDR was lower; 4.8 vs 10.5%/min (NS). Using a receiver-operator characteristic curve, a cutoff value for ICG-PDR of 9.5%/min identified graft failure (requiring retransplan-tation) or patient death with a sensitivity of 78% and specificity 50%.

Conclusion

Non-invasive ICG PDR is a useful marker of outcome in patients undergoing orthotopic liver transplantation. The multifactorial nature of ICG clearance is demonstrated by strong correlations with composite measures of organ function.