Partial ALPPS versus complete ALPPS for staged hepatectomy

Hepatectomy is the most effective treatment for large and/or multiple liver tumors [1]. However, an extensive hepatectomy cannot be performed if there will be an insufficient future liver remnant (FLR) because it may lead to post-hepatectomy liver failure (PHLF). To minimized PHLF, 20 to 25% of liver is needed in healthy livers while 30–35% in diseased livers [2]. In traditional 2-stage hepatectomy, liver hypertrophy can be induced after stage 1 by portal vein embolization (PVE) or portal vein ligation (PVL), and thus the FLR can meet the size requirement for the second stage procedure [3]. However, liver regeneration is slow making the time interval between the 2 stages long [4].

In 2007, Schlitt et al. performed the first “in-situ split” procedure [5], which was then named “Associating Liver Partition and Portal vein ligation for Staged hepatectomy” (ALPPS) by Clavien. ALPPS was performed by separating the future liver remnant and the diseased hemi-liver in the first stage with an in-situ split, in combination with PVL. Schnitzbauer et al. reported that ALPPS could induce rapid hypertrophy of the FLR (median FLR hypertrophy of 74% in 9 days) [5]. However, a serious complication (Clavien-Dindo > grade 3) rate of 44% and a mortality rate of 12% has limited the application of ALPPS [5].

Since the creation of the ALPPS Registry and the adoption of more stringent patient selection criteria, the overall mortality rate has dropped to 9% and serious complication rate to 27% [6]. Patients younger than 60 years old and those with colorectal cancer liver metastases (CRLM) have lower morbidity and mortality rates when they undergo ALPPS, while the prognosis for patients with gallbladder cancer or cholangiocarcinoma is poor. Technical modifications of ALPPS, such as partial ALPPS, tourniquet ALPPS, radiofrequency and microwave ALPPS, and mini-ALPPS have also been shown to be associated with reduced morbidity and mortality [7]. Partial ALPPS was described by Alvarez et al. [8], and consists of dividing the portal vein of the diseased hemi-liver up to middle hepatic vein. Compared with complete ALPPS, partial ALPPS induces comparable FLR hypertrophy with a lower morbidity rate and near zero mortality rate [8, 9]. However, Chan et al. have argued that complete ALPPS induces more rapid FLR hypertrophy than partial ALPPS, while not affecting the perioperative risk in chronic liver diseases [10]. As such, it is difficult to draw a firm conclusion due to the limited power of individual studies.

Thus, the purpose of this study was to conduct a meta-analysis comparing partial ALPPS with complete ALPPS with respect to FLR hypertrophy, the time interval between stages, postoperative complications, and mortality.

The results of this meta-analysis indicated that FLR hypertrophy and time between stages were not different between partial ALPPS and complete ALPPS. However, the postoperative complications rate in the partial ALPPS group was significantly lower than that in the complete ALPPS group, and while the mortality rate in the partial ALPPS group was markedly lower than in the complete ALPPS group the difference did not reach statistical significance.

Although ALPPS has become very popular for achieving fast hypertrophy of the FLR, the high morbidity and mortality of the procedure are important obstacles for its use. Mortality rates of 15–20% have been reported, even by experienced centers [14‐17]. Alvarez et al. suggested that partial parenchymal transection could reduce morbidity without negatively impacting FLR hypertrophy [8]. Petrowsky et al. reported that partial ALPPS was associated with a zero postoperative mortality rate, and lower morbidity than complete ALLPS [9]. Based on clinical and experimental evidence, Linecker et al. reported that at least 50% of the liver needed to be transected in partial ALPPS to trigger equal volume hypertrophy as observed in complete ALPPS [13]. In the articles included in the current meta-analysis, partial ALPPS induced comparable FLR hypertrophy with a lower postoperative complications rate, consistent with prior reports. However, in 2017 Chan et al. reported an opposite result; complete ALPPS induced more rapid FLR hypertrophy than partial ALPPS while not affecting the perioperative risk [10]. Since this study was the only one that included patients with hepatocellular carcinoma (HCC), more studies are required to establish outcomes of ALLPS in patients with HCC.

The mechanism by which ALPPS induces rapid FLR hypertrophy is not clear. There are 2 main hypotheses that have been put forward to account for this phenomenon. One is that the redistribution of portal vein flow causes hypertrophy, which is based on the hypertrophy observed in patients with increased blood flow due to a portal vein embolization (PVE) [18]. Another hypothesis is that growth factors released in response to tissue injury trigger liver regeneration. Experiments using animal models strongly support the hypothesis of growth factors. Injection of plasma from ALPPS animals has been shown to induce rapid liver regeneration in animals that have not undergone ALPPS [19]. In addition, levels of interleukin (IL)-6 and tumor necrosis factor α (TNFα) are elevated after stage 1 hepatectomy in animal models and humans [19]. An interesting study by Langiewicz et al. found a clinical relevance between the Indian hedgehog (Ihh) pathway and ALPPS: liver regeneration was achieved through the hedgehog signaling pathway after both chemical injury and resection. Rapid regeneration, similar to that seen with ALPPS, was observed after administration of recombinant Ihh after PVL, while this phenomenon was blocked by neutralization of Ihh [20]. This suggests that ALPPS activates liver regeneration via the Ihh pathway.

The reason partial ALPPS is safer than complete ALLPS is likely because it is less invasive [21, 22]. Similarly, laparoscopy-assisted ALPPS [23, 24], mini ALPPS [25], and portal vein embolization ALPPS adopt less invasive operative strategies to split the diseased hemi-liver in stage 1, which enhances the tolerability of the stage 2 operation. It is important to understand that the degree of liver partitioning affects FLR hypertrophy, and determines the timing and operative strategy of stage 2. As shown in Table 1, at least 50% liver transection is needed to induce a degree of liver regeneration comparable to complete ALPPS. Time interval between the 2 stages is generally accepted as a factor that determines the surgical difficulty of stage 2; a longer time interval between stages increases the difficulty of the second stage due to greater adhesions [10].

In spite of the potential benefits, a complete ALPPS may not be applied due to difficulty of parenchymal transection in some cases. For example, when a large tumor stretching the middle hepatic vein in right anterior section, it may cause severe venous bleeding during complete ALPPS. A tumor located close to the IVC or caudate also is not feasible for complete ALPPS. In these situations, a p-ALPPS may be preferred and our study may provide essential information.

There are limitations to this study. The included studies were retrospective, and lack the powerful evidence of randomized controlled trials (RCTs), which are still absent with respect to this topic. The indications for ALLPS varied among the studies. In addition, the scarcity of available literature comparing partial and complete ALPPS and the exiguous number of patients hampered the quality of the results.

The results of this meta-analysis indicated that FLR hypertrophy and time between stages were not different between partial ALPPS and complete ALPPS. However, the postoperative complications rate in the partial ALPPS group was significantly lower than that in the complete ALPPS group, and while the mortality rate in the partial ALPPS group was markedly lower than in the complete ALPPS group the difference did not reach statistical significance. More studies are needed to confirm the benefits of partial ALLPS, and indications for the procedure.