We are pleased that our findings1 are supported by the recent publication by Feyssa et al.,2 and we are especially gratified by their support for the observation that a Model for End-Stage Liver Disease score > 15 before transjugular intrahepatic portosystemic shunt (TIPS) insertion is a predictor of poor outcomes for transplant recipients. An important difference is that recurrent hepatitis C was the predominant disease in their series of 26 patients, whereas cholestatic diseases were predominant in our series.
We agree that the numbers in both series are too small for robust conclusions, but the clear message from our publication1 is that the outcomes for transplant recipients with a pre-TIPS Model for End-Stage Liver Disease score > 15 are extremely poor, and retransplantation should be considered because TIPS placement is unlikely to be beneficial.
The rate of TIPS insertion in transplant recipients is 0.6% at our institutions, so prospective studies would be extremely difficult because of the scarcity of study patients, even though we perform more than 200 transplants per year. Larger multicenter or multinational studies may also prove to be impractical because of the heterogeneous nature of the transplant recipient population and the wide variations in practice [eg, the higher rate of TIPS insertion (4%) in Feyssa et al.'s series2].