Article Text


P103 Clinical utility of myocardial perfusion imaging in liver transplant assessment patients
  1. J Leithead,
  2. K Kandiah,
  3. R Steeds,
  4. J W Ferguson
  1. Liver Unit, Queen Elizabeth Hospital, Birmingham, UK


Introduction Coronary artery disease (CAD) is associated with increased short-term morbidity and mortality following liver transplantation. As a result, the AASLD recommends that high risk individuals should undergo CAD evaluation during transplant assessment. However, CAD is often subclinical in these patients and remains a diagnostic challenge. Myocardial perfusion imaging (MPI) is a sensitive predictor of CAD in non liver populations but its clinical utility in this setting remains unclear.

Aim To determine if routine MPI in “high risk” liver transplant assessment patients influences the listing decision, and to determine if a positive MPI identifies patients at risk of an early cardiac event (CE) post transplant.

Method Retrospective study of 623 patients assessed for elective liver transplantation 01/2007–03/2010. The local criteria for CAD evaluation with MPI are; known or history suggestive of CAD, diabetes, smoking history, and peripheral/carotid vascular disease. A CE was defined as myocardial infarction, cardiac arrest, cardiogenic pulmonary oedema or complete heart block (Lee et al 1999) by 90-days post transplant.

Results One hundred and six patients (17%) underwent MPI. 7, 96 and 3 patients had a positive, low risk and indeterminate scan, respectively. The mode of cardiac stress induction did not influence the likelihood of a positive MPI (adenosine 5/76, dobutamine 0/7, exercise 2/20, p=0.657). The only patient factor predictive of a positive scan was known CAD (OR 24.2;95% CI 4.1 to 143.0, p<0.001). The frequency of positive MPI was similar in those who were and were not listed for transplantation (6.9% vs 6.5%, p=0.647). Of the 31 patients not listed, MPI influenced the decision making process in 1 individual. Two hundred and fifty two of the 384 listed patients were transplanted by 03/2010. The patients who had undergone MPI (no:40) were older (58.2 vs 52.6 years, p=0.001) than the non MPI patients, were more likely to be male (80.0 vs 60.8%, p=0.021), and were more likely to have CAD (22.5 vs 1.9%, p<0.001), diabetes (72.5 vs 18.4%, p<0.001), NAFLD (22.5 vs 3.3%, p<0.001), hypertension (32.5 vs 11.4%. p=0.001), and a smoking history (75.0 vs 55.2%, p=0.023). 10 patients (4.7%) had a CE during the specified time period following transplantation. The CE rate (8.8 vs 3.9%, p=0.198) and 3-month mortality rate (10.8 vs 7.7%, p=0.361) were similar for patients who did and did not undergo MPI. Amongst MPI patients a positive scan predicted a CE with a sensitivity of 33.3%, specificity of 90.3% and NPV of 93.3%, and mortality with a sensitivity of 0%, specificity of 87.9% and NPV of 87.9%.

Conclusion MPI is not a clinically useful tool in patients undergoing liver transplant assessment.

Statistics from

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.