Reply to Letter to the Editor: “The HIBA Index for ALPPS, Preliminary Results to Interpret With Caution”
We would like to thank Olthof et al for their comments on our article.1 As stated by the Amsterdam's group, we agree that hepatobiliary scintigraphy (HBS) is of paramount importance to assess resectability before the second stage of Associating Liver Partition and Portal vein occlusion for Staged hepatectomy (ALPPS). Recent studies have shown how liver volume increase is not always paralleled by an equivalent increase of liver function in staged hepatectomies.2 In our study, we compared for the first time 3 different interstage HBS parameters between patients who developed or not posthepatectomy liver failure (PHLF) after completion of ALPPS second stage. Interestingly, those patients with PHLF had significantly lower values of function than those without, regardless of future liver remnant (FLR) volume. We also tried to establish the safest cut-offs for these 3 parameters (FLR-C, FLR-F, and HIBA index) to avoid PHLF, but obviously, as stated in our article and pointed out by the authors, the small number of patients limited any firm conclusions. Nonetheless, our real intent was first, to demonstrate that FLR-C (ie, the proportion of total liver function produced by the FLR) alone represents an incomplete information since it does not reflect the underlying liver function. Secondly, that the cut-off of 2.69%/min/m2 proposed for major hepatectomies by De Graaf et al3 may not be adequate when used for patients submitted to ALPPS. This is because of the fact that when the anterior projection is used to calculate FLR function (FLR-F)4 this may lead to a significant overestimation of FLR-F if the FLR is represented by left liver segments.5 In ALPPS this overestimation is even increased as right posterior segments are displaced more dorsally than “normal” livers, secondary to full mobilization of the liver in the first stage, and also to the enlarged FLR hypertrophy. This was reflected in our study by a median FLR-C 18.9% larger than SPECT (up to 33%), when only the anterior projection was used. The resulting median FLR-F was 3.21%/min/m2 compared with a median value of 1.93%/min/m2 obtained by the combination of Gmean and SPECT. For this reason, in our opinion, all efforts should be directed towards standardizing the method to enable comparisons among different centers.
Last but not least, we introduced a new index, the so-called HIBA (Hospital Italiano de Buenos Aires) index,1 able to measure sectorial liver function between ALPPS stages. The HIBA index is a simplification of the Ekman algorithm6 but differently from De Graaf et al,3 we did not retain body surface area to calculate total liver uptake and we used Gmean (data from anterior and posterior projections) and SPECT/CT 3-dimensional analysis to define sectorial liver function. We agree with the authors that the 15% value of HIBA index, above which clinically significant PHLF should be avoided, has to be taken with caution as well as its diagnostic accuracy, due to the small number of patients included and to the low incidence of PHLF in the described population. In this sense, a proposal for retrospective external validation of the HIBA index has been submitted and is under consideration by the ALPPS registry's scientific committee. Indeed, we strongly believe that the use of a functional test such as HBS should become a standard of care to be included in the decision-making algorithm for proceeding to the second stage of ALPPS. Meanwhile, we should remain open minded as new frontiers for regional function tests are behind the door and are not enemies.