A standardized suprapubic bottom-to-up approach in robotic right colectomy: technical and oncological advances for complete mesocolic excision (CME)

Minimally invasive resection concepts for right colectomy (RC) must fulfil the same oncologic criteria as compared with an open approach which includes “no-touch isolation technique”, ligation of the vascular pedicles at their origin, oncologic lymphadenectomy and “distal and radial clearance” of the neoplasm from resection margins [1].

Although still inconclusive as based predominantely on retrospective studies [2] en bloc resection of the lymphatic drainage of the malignant neoplasia, comprising associated lymph nodes performing complete mesocolic excision (CME) in combination with central vessel ligation (CVL) has been repetitively demonstrated to bear the potential of improving overall and disease specific survival following surgery for CRC [3‐7].

Three key aspects for CME plus CVL found a consensus as a gold standard in RC: development and separation of the right retrocolic connecting fascia to develop an unhurt mesocolon as a sound cluster, CVL with dissection of the ileocolic vessels at their origin to optimize the vertical lymph node dissection for regional control and achievement of an appropriate length of colon to remove pericolic lymph nodes, maximizing the collection of longitudinal lymph nodes [8]. However, in contrast to total mesorectal excision (TME) for rectal cancer, routine implementation of CME in RC has yet not been achieved to a comparable level [9].

This is despite the demonstrated oncological improvement and the principles underlying CME being anatomically logical. The complex embryonic rotation around the mesenteric root and subsequent folding (Fig. 1a) implicates a set of planes that comprises layers and surfaces relevant for RC. In this context, critical planes for retrocolic dissection are the fascia of Toldt (right retrocolic fascia) and in upper medial continuity with Fredet’s fascia (fascia preduodenopancreatica) up to the superior mesenteric vein (SMV). These fusion structures are ventrally delineated by the posterior layer of the ascending mesocolon (the meso-fascial interface) and dorso-laterally, by the prerenal fascia, representing the posterior parietal peritoneum covering the retroperitoneum (the retro-fascial interface) [8]. Laterally the white line of Toldt marks the access to Toldt’ fascia (Fig. 1b).

Lately, first reports of the concepts of robotic right colectomy (RRC) incorporating CME and CVL came up mainly following the adaption or modification of the medial to lateral approach as established in laparoscopic techniques [10‐13]. In a first study reported by Petz and colleagues a suprapubic approach was utilized [14]. Here we propose a standardized four-step procedure of RRC implementing suprapubic port positioning as introduced lately [15] with a consequent down-to-up approach latter with improved implementation of CME and CVL.

In our experience, the novel suprapubic access in combination with the down-to-up approach facilitates technically improved CME and CVL in robotic right colectomy if compared to the “classical” MTL strategy. The number of retrieved lymph nodes was significantly increased by the here presented BTU-approach of RRC when compared with MTL. Although operating on rather elderly patients, we demonstrated the procedure in this early experience to be safe observing moderate morbidity comparable between both groups.

Recently, RRC has been praised as an alternative to classic laparoscopic RC [12, 18]. CME allows for resection along the unimpaired contour delineating the dorsal mesentery and the anatomo-embryological fusion fascias and is therefore vital for a true radical R0 resection, as the mesocolon contains the whole potential routes for local neoplastic spread through lympho-vascular, neuro-perineural and fibro-fatty tissues [3, 9, 19‐21]. CVL facilitates a comprehensive lymph node harvest along the supplying vessels paralleled by a relevant effect on regional recurrence and systemic dissemination [8, 22]. Lateral access is applied for open CME, during which the right colon is mobilized lateral to medial firstly along Toldt’s fascia; then, the visceral fascia is sharply dissected right sided along superior mesenteric vessels in order to expose and dissect the right colon-feeding vessels. In contrast, the laparoscopic CME to date applies a medial approach, during which the visceral peritoneum along the root of the right mesocolon is cut and lymph nodes lining up the surgical trunk are dissected firstly, followed by the seek and separation of Toldt fascia and ligation of the central vessels. Latter technique was applied here in the MTL-group.

Several laparoscopic concepts were presented to adapt the anatomical defaults of an oncological RC primarily proposed for open RC [3, 23]. Feng et al. proposed a hybrid medial approach of laparoscopic right colectomy prospectively compared to a completely medial approach. The hybrid approach is based on a first up-to-down dissection (separation of the gastrocolic ligament and dissection of the middle colic vessels and Henle’s trunk) combined with a subsequent classical medial-to lateral down-to-top approach. The complete medial approach on the other hand, involved a down-to-up approach in every step, including the dissection of the middle colic vessels and the Henle trunk as well as dissection of the inferior edge of the pancreas. Like in all medial to lateral access variants, authors pointed out that in any case, surgical and oncological safety are critically depending on the precise initial identification of the anatomical planes after dissection of the medial and transverse mesocolic peritoneum [24]. Matsuda et al. also implemented a variation of a cranio-to-caudal approach in their concept of laparoscopic right meso-colectomy, noting that lymph node dissection around the middle colic vessels is technically demanding. Authors stressed the necessity of detailed knowledge of the embryonic developed structures and planes to dissect the retro-mesocolic space [25, 26]. These and other reports illustrate the challenge of minimally invasive RC implying CME and CVL to meet both patient and oncological safety demanding meticulous knowledge of the anatomy to unveil the correct planes.

When compared with the classical laparoscopic medial to lateral approach the here proposed robotic concept may adopt the requirements for oncological optimization in RC by switching from a search of the planes (Toldt’s / Fredet’s fascia) towards a consequent utilization of the planes as dissection guide from the beginning (white line of Toldt, Fig. 1a) prior to any vessel ligation keeping the “envelope” of the resection specimen intact. In addition, initial mobilization in combination with a head down angulation of patient positioning (Trendelenburg) facilitates CME in caudal parts, that otherwise would be difficult to reach with the suprapubic port positioning. Furthermore, the consequent bottom-to-up approach, permits subsequent dissection of the retro- mesocolic space along the fascia of Toldt and ventral to the duodenum as well as the pancreas head exposing all the essential planes for CME and CVL prior to vessel dissection and ligation. The initial complete mobilization of the mesocolon eases the identification of the superior mesenteric vessels. The suprapubic approach facilitates the dissection right lateral to the superior mesenteric vessels and along a dorsally mobilized and subsequently controllable mesocolon under optimized vision enabling the identification and dissection of the branches to the right colon in all their variations. Excessive tension on the ileocolic vessels is avoided by obtaining exposure through a local step-by-step traction-counter-traction strategy. This may reduce the known risk factors of iatrogenic vessel injuries in CME and CVL [27] and possibly simplifies the procedure especially in obese patients who represent a challenge with conditions including missing the space for tension on ileocolic vessels to adequately expose the medial plane of the ascending mesocolon in the classical medial to lateral approach. In addition, the true bottom-to-up alignment of the procedure allows continuous visual orientation during ligation of vessels to the right of the superior mesenteric vessels in the course of CME by deploying the mesenteric root and middle colic artery as targeting structures.

The mean lymph node yield of 16.3 in our MTL group of patients is comparable to those observed by other authors for classic laparoscopic right colectomy [1] and robotic assisted right colectomy [28, 29] if performed with MTL strategy. The mean lymph node retrieval in the BTU-cohort of 40.2 is superior to the range of levels reported for classic laparoscopic as well as for open techniques. The improved lymph node yield in robotic RC if compared to historic laparoscopic and open techniques is in accordance with other reports [29, 30]. Our levels of lymph node harvest were comparable to very recent early series of BTU-RRC [31]. We were able to reduce the number of procedures with lymph node harvest below the critical number of 12 [17] from 29% to zero with the BTU-approach. In light of improved technical approach and oncological achievements represented by increased lymph node yield we demonstrated an improvement with the bottom-to-up strategy in RRC over the medial to lateral path which may be even superior to open and laparoscopic strategies.

Operating time seems rather high in this series; however, it is comparable to other studies on RRC [32‐34]. Mean skin-to-skin time is explainable by the early experience with this novel procedure. It is in the range of other reports on early stages of adopting techniques in minimally invasive techniques in RC [25, 33]. We operated in elderly patients with ASA scores manly ranging at 2 to 3 that decided us to prolong the hospital observation time of our patients leading to a comparatively mean length of hospital stay. The observations of other authors, comparing young and elderly patients with robotic colorectal surgery in general [35] plus our increasing experience with the procedure should shorten hospital stay over time.

This study has some limitation particularly concerning comparison of the effectiveness of lymph node harvest. For one, it is limited by the small number of patients especially in the MTL group. Consequentely statistical analyses on lymph node numbers between the two groups, although performed by conservative testing with respect to small groups size, has to be interpreted with caution. Further the fact, that all patients but one in the MTL group were operated in the initial phase of our robotic colorectal surgery program, may play a role for the lymph node count distinctions. Although differences were pronounced concerning lymph node harvest and the fact that other variables like operating time, hospital stay and complication profiles were comparable between groups, one can not exclude a bias for lymph node yield in favour of BTU-operated patients based on the learning curve.

With the suprapubic approach, the utilization of robotic systems may not just target a simplification of the minimally invasive procedure of RC [1]. The here proposed standardized robotic four-step suprapubic approach with down-to-up mesocolic mobilization and subsequent CME plus CVL demonstrated to be safe even in elderly patients. It may bear the potential of exceeding a minimally invasive technique of RC from the stage of “being easier than laparoscopy” to an oncological advanced concept. Robotic systems as used by us and other groups may provide the technical requirements to combine advantages of both open and minimally invasive surgical concept in oncologic RC. The preliminary clinical results in this study need to be proven in a multi-center randomized setting in larger cohorts and on a long-term basis.