Journal Logo

SURGICAL PERSPECTIVES

Demise of “Hilar En Bloc Resection by No-touch Technique” as Surgery for Perihilar Cholangiocarcinoma

Dissociation Between Theory and Practice

Nagino, Masato MD, PhD; Clavien, Pierre-Alain MD, PhD

Author Information
doi: 10.1097/SLA.0000000000004986
  • Free

In 1994, Neuhaus et al from Germany developed a novel approach for perihilar cholangiocarcinoma (PHC). This surgery, named “extended bile duct resection,” was a “no-touch technique” combining total hepatectomy, pancreatoduodenectomy, extended lymphadenectomy, and liver transplantation.1 Extended bile duct resection can ideally offer en bloc eradication of the entire biliary tree, without touching the region of the hepatoduodenal ligament to avoid cancer cell spread. This superextended surgery was performed in 16 PHC patients between March 1992 and November 1998.2,3 However, it was eventually concluded that extended bile duct resection is not recommended due to the increased morbidity/mortality and unsatisfactory survival outcome.3

In 1999, Neuhaus et al also reported a survival advantage of PHC patients who underwent combined right trisectionectomy and portal vein resection (H145678-B-PV, according to the New World Terminology4), later termed “hilar en bloc resection.”2 In this hepatectomy with preemptive routine portal vein resection, the extrahepatic bile ducts were resected en bloc with the portal bifurcation, the right hepatic artery, and liver segments 1 and 4 to 8; thus, this “no-touch technique” can avoid surgical preparation of the hilar vessels in the vicinity of the tumor. If remnant liver volume is sufficient, such surgery may be attractive, because it is seemingly logical and usually not very technically demanding. However, the issue is that the volume of the left lateral sector is often small, occasionally too small, and scarcely ever enough,5 which frequently leads to posthepatectomy liver failure. A much better 5-year survival of 61% was reported in PHC patients who underwent hilar en bloc resection, but caution should be exercised because this value was calculated with only 14 patients after excluding 9 patients who died of postoperative complications (n = 4) and/or had R1 resection (n = 9); this survival analysis gave a “raised bottom” value, which was premature for evaluating the clinical significance of a surgical procedure. Notably, histologically confirmed tumor invasion into a resected portal vein was found in only 5 (22%) of 23 patients who underwent hilar en bloc resection.

Thereafter, in 2012, they conducted a retrospective study comparing 50 PHC patients who underwent hilar en bloc resection with other 50 PHC patients who underwent conventional standard hepatectomy, and again claimed oncological superiority of hilar en bloc resection.6 However, the mortality of hilar en bloc resection was high, and statistical significance in survival was derived only after excluding perioperative deaths. The latest report from Campus Charité, Universitätsmedizin Berlin,7 where hilar en bloc resection has been inherited as a routine procedure, showed that 90-day mortality of 145 right-sided hepatectomies (H156784’, n = 6 and H145678, n = 139; of these, 120 underwent hilar en bloc resection)4 for PHC was still as high as 18%. In addition, no convincing data indicating the oncological superiority of hilar en bloc resection were given. Molina et al from Spain also pursued hilar en bloc resection due to oncological reason, but surgical outcome was poor: the reported mortality after right-sided hepatectomy (H15678, n = 7; H145678, n = 9)4 was 25% (4/16).8 According to the present first author's experience at Nagoya University Hospital, 90-day mortality of all right-sided hepatectomies for PHC was only 1.8% (5/283 between 2007 and 2020, unpublished data), including 0% (0/39) in H145678-B-PV, 3.0% (1/33) in H145678-B, 3.6% (2/56) in H15678-B-PV, and 1.3% (2/155) in H15678-B.4 Surgical mortality (90-day) should be less than 5%, because high mortality remarkably spoils oncological positive effect, if any.

In 2009, Hirano et al from Japan showed that “no-touch resection” with right-sided hepatectomy (H15678, n = 24; H145678, n = 1)4 and routine portal vein resection for hilar malignancy was feasible with acceptable mortality of 4% (1/25).9 Afterwards, they reported that “no-touch resection” had a positive impact on survival.10 Nevertheless, this procedure was discarded more than 5 years ago, due to disappointing oncological effects (personal communication with professor Hirano from Hokkaido University, Sapporo, Japan). At present, all centers in Japan and most other centers in the world do not use hilar en bloc resection. So far, the present first author has never used hilar en bloc resection due to unreliability of evidence, while having experience with more than 1000 resections of PHC.11,12 The present author's consistent policy is that vascular resections should be performed only when the vessel adheres to and cannot be freed from the tumor, and that even if invasion is suspected preoperatively, the vessel is not resected when the vessel can be freed from the tumor without difficulty.12,13 Resections under this policy have offered better short- and long-term outcomes, as reported previously.12,13

In 2020, Bednarsch et al from Germany described left- and right-sided hepatectomy with hilar en bloc resection as a routine procedure for PHC.14 Left-sided resections included H1234 (n = 8), H12345’8’ (n = 24), and H123458 (n = 4),4 and right-sided resections included H15678 (n = 8), H156784’ (n = 17), and H145678 (n = 20).4 The present authors were somewhat surprised because hilar en bloc resection combined with left-sided hepatectomy was found technically demanding, while Neuhaus, an original proponent of the no-touch technique, never performed left-sided hilar en bloc resection. Unexpectedly, mortality was significantly worse after right-sided hilar en bloc resection than left-sided hilar en bloc resection (15.6% = 7/45 vs 8.3% = 3/36, P = 0.003). Median survival was poorer in right-sided resection than in left-sided resection (38 months vs 45 months), although the difference was not statistically significant (P = 0.159). Their challenging spirit is commendable, and the data showed that left-sided hilar en bloc resection is feasible and can offer comparable long-term survival. However, the authors failed to demonstrate oncological benefit of hilar en bloc resection itself. Furthermore, recent evidence has shown that right-sided resection is surgically associated with poorer outcomes than left-sided resection.15

Quite recently, Muller et al reported surgical and oncological results of benchmark PHC study where 24 expert centers around the world participated (13 from Europe, 8 from Asia, and 3 from the USA).15 Benchmark (= low-risk) patients were defined as patients who underwent standard major hepatectomy (H15678, H145678, H1234, or H123458)4 without vascular resection and who did not have major comorbidities. However, patients who underwent routine portal vein resection (hilar en bloc resection) as part of their surgical protocol from 5 participating centers were included. All patients underwent hepatectomy between 2014 and 2018. The total number of hepatectomized patients was 1829, of which 708 (39%) qualified as benchmark cases. Even in benchmark patients, the 90-day mortality after routine portal vein resection was as high as 26.6% (17/64), which was significantly higher than that of right-sided hepatectomy without portal vein resection (7.6% = 23/304, P < 0.001). Furthermore, median survival was significantly poorer in the former than in the latter (27 months vs 54 months, P < 0.001). In addition, both survival times were significantly worse than the median survival of 61 months in left-sided hepatectomy (n = 340) (vs with routine portal vein resection, P < 0.001; vs without routine portal vein resection, P = 0.04). These results strongly indicate no oncological superiority of right-sided hepatectomy, including hilar en bloc resection, over left-sided hepatectomy.16

Wide tumor-free margins and no dissection of tumor-bearing areas are basic rules used in oncological surgery for malignancy. Seemingly, hilar en bloc resection by the no-touch technique follows this central dogma, but this procedure looks good on paper only, similar to desk theory. In the 1980s, Nakao, who was a great pioneer of extended surgery for pancreatic cancer,17 aggressively performed pancreatoduodenectomy or total pancreatectomy combined with routine portal vein resection, but later he discarded this prophylactic en bloc vascular resection because patient survival did not improve (personal communication with emeritus professor Nakao from Nagoya University, Nagoya, Japan). This is another good example of dissociation between theory and practice. Although randomized controlled study may be necessary for final conclusions, such study is unmistakably impossible to do, due to rarity of the disease, low resectability, and ethical reasons associated with the high surgical mortality. Carefully performed novel benchmark studies may provide high level evidence and help in adjusting therapeutic paradigms.18

On the same token, Associating Liver Partition with Portal vein ligation for Staged hepatectomy (ALPPS) was proposed as a promising novel technique for PHC through extended right-sided hepatectomy. Although this 2-stage approach appeared straightforwardly feasible, the initial results were highly vexing, for example with the early analysis of the international ALPPS registry reporting on a scary postoperative mortality rate of 48%, obviously preventing any conclusion about putative oncologic benefits.19 The comparative 28% mortality rate with standard 1-stage approaches, originating from 2 well-established HPB centers, even raised the role of surgery at all for PHC in their cohort. The dismal “off the chart” initial results with the ALPPS registry were due to a combination of inexperience with a new surgical technique, for example, in the vast majority of centers it was the first and only ALPPS procedure for PHC, and a violation of basic surgical principles including complex biliary reconstructions during the first stage and with more than half of the extended resections performed in the presence of biliary infection or bilirubin levels ≥100 μmol/L or both. In clear retrospect, this initial failure of ALPPS for PHC relates rather to the circumstances than the technique.20 Indeed, a more recent analysis of 40 modified ALPPS procedures for PHC from nine expert centers, better respecting a non touch approach, disclosed a postoperative mortality rate of 7.7%, that is, within the benchmark cut-off of 8% mortality15 but so far, no demonstrable conclusive oncologic benefits (personal communication with professor Balci from Ankara University, Ankara, Turkey). Although ALPPS remains in no way a routine approach for PHC, current available data do not exclude a role in highly selective scenarios of Bismuth IV tumors requiring extended right-sided hepatectomy, but more convincing data is requested before promoting this approach.

To sum up, all evidence currently indicates no surgical and oncological benefit of hilar en bloc resection based on the no-touch theory. Thus, it is now time to abandon this “excessive” surgery.

REFERENCES

1. Neuhaus P, Blumhardt G. Extended bile duct resection: a new oncological approach to the treatment of central bile duct carcinomas. Langenbeck's Arch Surg 1994; 379:123–128.
2. Neuhaus P, Jonas S, Bechstein WO, et al. Extended resections for hilar cholangiocarcinoma. Ann Surg 1999; 230:808–818.
3. Seehofer D, Thelen A, Neumann UP, et al. Extended bile duct resection and liver transplantation in patients with hilar cholangiocarcinoma: long-term results. Liver Transpl 2009; 15:1499–1507.
4. Nagino M, DeMatteo R, Lang H, et al. Proposal of a new comprehensive notation for hepatectomy: the “New World” terminology. Ann Surg 2021; 274:1–3.
5. Ito J, Komada T, Suzuki K, et al. Evaluation of segment 4 portal vein embolization added to right portal vein for right hepatic trisectionectomy: a retrospective propensity score-matched study. J Hepatobiliary Pancreat Sci 2020; 27:299–306.
6. Neuhaus P, Thelen A, Jonas S, et al. Oncological superiority of hilar en bloc resection for the treatment of hilar cholangiocarcinoma. Ann Surg Oncol 2012; 19:1602–1608.
7. Benzing C, Krenzien F, Wokfsberger A, et al. A tailored approach in lymph node-positive perihilar cholangiocarcinoma. Langenbech's Arch Surg. (in press). doi: 10.1007/s00423-021-02154-4.
8. Molina V, Sampson J, Ferrer J, et al. Surgical treatment of perihilar cholangiocarcinoma: early results of en bloc portal vein resection. Langenbeck's Arch Surg 2017; 402:95–104.
9. Hirano S, Kondo S, Tanaka E, et al. No-touch resection of hilar malignancies with right hepatectomy and routine portal reconstruction. J Hepatobiliary Pancreat Surg 2009; 16:502–507.
10. Tamoto E, Hirano S, Tsuchikawa T, et al. Portal vein resection using the no-touch technique with a hepatectomy for hilar cholangiocarcinoma. HPB (Oxford) 2014; 16:56–61.
11. Nagino M. Value of ALPPS in surgery for Klatskin tumours. Br J Surg 2019; 106:1574–1575.
12. Nagino M, Ebata T, Yokoyama Y, et al. Evolution of surgical treatment for perihilar cholangiocaricnoma: a single-center 34-year experience. Ann Surg 2013; 258:129–140.
13. Mizuno T, Ebata T, Yokoyama Y, et al. Combined vascular resection for locally advanced perihilar cholangiocarcinoma. Ann Surg. (in press) doi:10.1097/SLA.0000000000004322.
14. Bednarsch J, Czigany Z, Lurje I, et al. Left- versus right-sided hepatectomy with hilar en-bloc resection in perihilar cholangiocarcinoma. HPB (Oxford) 2020; 22:437–444.
15. Mueller M, Breuer E, Mizuno T, et al. Perihilar cholangiocarcinoma: Novel benchmark values for surgical and oncological outcomes from 24 expert centers. (Submitted to ESA - November issue Annals of Surgery).
16. Nagino M, Ebata T, Mizuno T. Oncological superiority of right-sided hepatectomy over left-sided hepatectomy as surgery for perihilar cholangiocarcinoma: Truth or biased view? Ann Surg 2021; 274:31–32.
17. Nakao A, Ichihara T, Nonami T, et al. Clinicopathologic and immunohistochemical studies of intrapancreatic development of carcinoma of the head of the pancreas. Ann Surg 1989; 209:181–187.
18. Gero D, Muller X, Staiger RD, et al. How to establish benchmarks for surgical outcome? : A checklist based on an international expert delphi consensus. Ann Surg. (in press) doi:10.107/SLA.0000000000003931.
19. Olthof PB, Coelen RJS, Wiggers JK, et al. High mortality after ALPPS for perihilar cholangiocarcinoma: case-control analysis including the first series from the international ALPPS registry. HPB (Oxford) 2017; 19:381–387.
20. Lang H, de Santibanes E, Clavien PA. Outcome of ALPPS for perihilar cholangiocarcinoma: case-control analysis including the first series from the international ALPPS registry. HPB (Oxford) 2017; 19:379–380.
Keywords:

hilar en bloc resection; no-touch technique; perihilar cholangiocarcinoma

Copyright © 2021 Wolters Kluwer Health, Inc. All rights reserved.