Laparoscopic repeat liver resection

Abstract Recurrence of liver cancers inside the liver are often treated with liver resection (LR). However, increased risks of complications and conversion during operation were reported in laparoscopic repeat LR (LRLR). The indication is still controversial. One multi‐institutional propensity score matching analysis of LRLR vs open repeat LR for hepatocellular carcinoma, two propensity score matching analyses for colorectal metastases, and two meta‐analyses including hepatocellular carcinoma, intrahepatic cholangiocarcinoma, metastases, and other tumors have been reported to date. LRLR was reported with better to comparable short‐term and similar long‐term outcomes. Furthermore, the shorter operation time and the smaller amount of intraoperative bleeding for LRLR was reported for the patients who had undergone laparoscopic rather than open LR as an earlier procedure. The speculations are presented, that complete dissection of adhesion can be dodged and laparoscopic minor repeated LR can minimize the liver functional deterioration in cirrhotic patients. LRLR, as a powerful local therapy, could contribute to the long‐term outcomes of those with deteriorated liver function. However, the procedure is now in its developing stage worldwide and further accumulation of experiences and evaluation are needed.

deformity of the liver and structures make the identification of lesions and the important structures problematic. Easy bleeding from the liver capsule causes a suboptimal surgical field during the dissection of adhesion. 17 Former surgical histories can cause these changes which increase the risks of complications and conversions during LRLR.

| FE ATURE S OF LLR
LLR is reported to be beneficial for patients with chronically injured liver. [18][19][20] It can minimize the damages to collateral vessels as well as liver parenchyma through its minimal laparotomy, mobilization, and compression. Postoperative ascites and liver failure are reported to be reduced. 21  In a similar framework, direct approach for working space in LRLR, after minimal dissection of adhesion, can be enabled especially in small LRLR. [25][26][27] Several reports have shown that time and blood loss in LLR were similar in primary and repeat settings, 27,28 although there are usually large differences between primary and repeat open procedures. Furthermore, the shorter operation time and the smaller intraoperative bleeding for LRLR was reported, when it was applied to the patients who had previously undergone LLR rather than open LR. 29

| S TUD IE S OF LRLR AND OUR PROPENS IT Y SCORE MATCHING ANALYS IS FOR H CC PATIENTS
There is no randomized control trial for LRLR vs open repeat LR (ORLR), although the number of reports is increasing. Our multiinstitutional propensity score matching analysis of LRLR vs ORLR for HCC, 30 two propensity score matching analyses for colorectal metastases, 31,32 and two meta-analyses that included cases of HCC, intrahepatic cholangiocarcinoma, metastases, and other tumors 33,34 have been published to date (Table 1). These studies showed that LRLR reduced bleeding, had less or similar morbidity, and shortened or similar length of stay with the equivalent long-term outcomes.
The magnified view and strained adhesion by pneumoperitoneum in LLR can facilitate meticulous dissection 35 and also laparoscopic approach can make complete dissection of adhesion unnecessary as mentioned above. 24,25 Contrary to the LR for metastases, minor LR for the fibrotic liver with poor functional reserve and collaterals is frequently applied for HCC. The advantages of LLR are reported especially for the HCC patients' management during the long history with repeat oncogenesis. 35,36 We conducted the first multi-institutional propensity score matching analysis of LRLR vs ORLR for HCC 30 with 1582 repeat LR cases at 42 high-volume centers around the world. It showed that LRLR was not inferior to ORLR in short-and long-term outcomes and LRLR is feasible for selected patients. The analysis was performed on an intention-to-treat basis, and the conversion rate was 3.8%.
This low conversion rate may derive from the selection. This study showed that LRLR was generally applied to patients of poor performance status with poor liver function but with favorable factors related to tumors and surgical procedures. Our analysis also revealed notable differences between centers in the number and percentage Our study with propensity score matching showed that LRLR results in less blood loss, a longer operation time, and similar long-term outcomes. With the exception of morbidity and hospital stay, our data were comparable to previous reports. Decreased morbidity is considered as one of the advantages of LLR for HCC patients. However, our patients after matching have a favorable liver function, and thus, the influence of LLR on morbidity might be lower. Also, the differences in hospital stay between centers and/or areas, possibly due to insurance systems and hospitalization practices, were large. This might be the reason why there is no difference in hospital stay.
The number and percentage of LRLRs for HCC differed greatly between centers in our study. The number of LRLRs per center ranged from 0 to 67 (median 10). LRLR accounted for 41.0% of all repeat LR cases and from 0% to 100% (median 57.1%) of the cases undertaken at each center. Also, no correlation was found TA B L E 1 Summary of previous reports of LRLR (propensity score matching analyses & meta-analyses) Therefore, we believe that this procedure is still in its developing stage worldwide. Among our own experience of 33 repeat and 12 three-or-more-time repeat LLR cases, there were three cases with anatomical resection or resections exposing major vessels after previous anatomical resection who developed bile leakage and >30 days hospital stay. Anatomical alterations on major vessels with scars and adhesions may have big influences on later resections also exposing them. Evaluations of such setting of LRLR should be required after the accumulation of more experiences in this area.
Nevertheless, our international multicenter propensity score analysis showed that neither short-nor long-term outcomes of LRLR are inferior to those of ORLR. A large-scale study conducted after further establishment of the procedure and greater accumulation of experience is needed to confirm the role of LRLR.

D I SCLOS U R E
Conflict of Interest: The author declares no conflicts of interest related to this publication.
Author Contributions: Morise Z. collected the data and wrote this paper.