Reconstructing topography and extent of injury to the superior mesenteric artery plexus in right colectomy with extended D3 mesenterectomy: a composite multimodal 3-dimensional analysis

Three different specimen groups underwent diverse methodologies of anatomical analysis. All specimen groups represented the same anatomical area—The D3 volume (level 3 of lymph node dissection) for the right colon, defined as mesenteric fat enclosing the superior mesenteric vessels between the following borders [15, 16]:

The study included 21 specimens. Group I contained five specimens (3 females; age 80–93 years), which underwent anatomical dissection. Group II contained seven specimens (5 females, age 71–86 years), where two were randomly selected to first undergo NanoCT scanning. All specimens ultimately underwent serial histological sectioning. Another two specimens were selected for 3D histology reconstruction. Group III contained nine postoperative specimens (5 females, age 55–69 years), where all directly underwent serial histological sectioning.

Dimensions of Group II specimens had a mean of 5.83 cm (± SD 0.76) in length and 2.57 cm (± SD 0.51) in width. The dimensions of Group III specimens had a mean of 6.48 cm in length (± SD 2.07) and 2.65 cm (± SD 0.54) in width.

A review of the literature shows that a clear definition of the SMA vascular sheath is still missing [17, 25‐30]. The results of anatomical dissection in this study reveal only one surgical dissection plane. This plane separates the SMA adventitia from a paravascular sheath as seen on our histological slides (Fig. 4). In this way, it separates the contents of the SMA adventitia (lymphatics, intrinsic nerves, and vasa vasorum) from those of the paravascular SMA sheath, which in turn supply the mesentery and the gut [18, 31]. Moreover, the SMA paravascular sheath thickness (Group I and II results) decreases caudally [32], still allowing the identification of this plane, consequently maximizing the extent of lymphadenectomy.

With the current trend to increase the extent of mesenterectomy in cancer surgery [33‐35], concern has been raised about postoperative bowel function in patients where the SMAP may have been injured [8, 36]. It is also significant to point out that the extent of this injury depends not only on the morphology of the plexus but also on the surgical technique deployed. The results of this study clearly show that the SMAP is inherent to the paravascular sheath of the SMA, and not in the periphery of the SMV. These results also represent the first attempt to define the extent of SMAP injury in the context of a clinical trial, following a clearly defined surgical technique for performing right colectomy for cancer [16, 37]. This procedure involves a more radical (extended) D3 mesenterectomy than the traditional D3 right colectomy, namely operating within the arterial sheath of the SMA. The results of this study are therefore not transferable to the traditional mesenterectomy where these nerves are not damaged. It is also relevant to point out that this nerve plexus analysis has been performed at the most cranial level of the D3 dissection (5 mm proximal to the origin of the middle colic artery and where the potential for injury to the plexus is maximized), using a representative histopathological slide to establish nerve number and surface area removed. This singular slide actually represents the transectional injury to the SMAP, where our comparisons reveal a considerable injury (59% nerve fibers and 49% nerve area). Analysis of histological slides also lets us conclude that a relatively fewer number of nerve fibers, although larger in diameter, are left behind in the patient.

A literature search on postoperative intractable diarrhea after SMAP injury identified seven relevant articles [1, 2, 7, 32, 38‐40]. The great majority of these publications concern surgery for pancreatic cancer and entail an excisional injury to the SMA paravascular sheath, including nerve ganglions [2, 41], between the origin of the SMA and the inferior border of the pancreas. Needless to say, the pancreatic head or the whole pancreas has been removed. Furthermore, the disruption of the bowel continuity during pancreas surgery also transects the bowel’s intrinsic nerve plexus, implying intrinsic denervation. These articles are not randomized controlled trials and contain small numbers of patients, rendering the level of the evidence low. A newer publication on the subject [5] compares two groups of 37 patients operated for pancreatic cancer in a randomized setting where the intervention was either right half dissection of the SMAP or SMAP preservation. The results seem to imply that diarrhea was more often present in the right half dissection of the SMAP (70.3%) when compared to preservation (48.6%), but still not significant. Perhaps more important is the fact that almost 60% of all these patients had recurrence. A previous study by Inoue et al. [41] also presents similar results.

We postulate that the oblique trajectory of the SMAP nerves demonstrated here leads to a more extensive, if not complete, injury due to a longitudinal excision of the SMAP performed at surgery. The potential level of damage to the SMAP, along the border of the SMA, is even more clear while observing the 3D-printed model on Fig. 8a. It is also worth noting that it was easier to measure SMAP trajectory angle with the physical 3D-printed models [42]. The functional outcome after the extended D3 mesenterectomy has been investigated and described in previous articles by Thorsen et al. When referring to the small bowel motility in these articles, it is essential to point out that the nerve injury caused by this specific surgical procedure exclusively involves the extrinsic nerves (SMAP), not the intrinsic (Enteric Nervous System), since the small bowel continuity is preserved. The amount of nerve injury described in this study can explain the “rapid small bowel transit” shown in Thorsen’s motility study performed three weeks after surgery [13]. More specifically, it manifests through a short period with frequent, loose stools with a maximum point 9 days after the operation. This stool pattern differs significantly from the stool pattern in patients, undergoing right colectomy with a traditional mesenterectomy without nerve damage [43]. Despite the extensive damage, a fast normalization process begins shortly after surgery, probably reflecting the bowel's extraordinary plasticity. A slow normalization over months follows. Six months after the operation, the small bowel transit time shows a clear tendency toward normalization [13]. In the long-term follow-up, the patients who have undergone the extended D3 mesenterectomy still present a slightly increased stool frequency and reduced consistency compared to those who have undergone the traditional mesenterectomy. However, this does not affect the long-term quality-of-life scores, which are equal in both groups [9, 14, 44].

The previously described interarterial nerve fibers between branches of the SMA [29], adaptation of the remaining nerves, as well as reinnervation [45] can provide an explanation for the normalization of postoperative bowel function that occurs during the first two years after surgery. Moreover, Thorsen et al. [14] have demonstrated that individual anatomical traits in patients, such as branching patterns of jejunal arteries, can have an impact on bowel function after surgery. Namely, having no jejunal artery proximal to the origin of the middle colic artery led to more diarrhea in patients operated with right colectomy and extended D3 mesenterectomy. In comparison with SMAP injury in pancreas surgery, the level of SMAP injury, while performing right colectomy with extended D3 mesenterectomy, lies distal to the inferior border of the pancreas and stretches to 1 cm below the origin of the ICA. With right colectomy, neither ganglions nor pancreatic tissue are removed, providing another possible explanation for the difference in clinical presentation between the two procedures.

The strength of this article is the multimethodological, macro/micro, 2D/3D, destructive/nondestructive approaches that allowed us to perform a composite interpretation. Characterizing the architecture of the SMAP was achieved using different technologies, namely 3D histology, nanoCT reconstructions, 3D printing, and supplemented with traditional anatomical dissection. While anatomical dissection showed nerve patterns and limits within the D3 area, histological imaging allowed for nerve count and area quantification. NanoCT imaging showed plexus trajectory and angulation. For these reasons, we deem it as important to combine both destructive and nondestructive imaging techniques to assure accurate anatomical information in future investigations. The number of anatomical (12) and clinical (9) specimens can be deemed sufficient when compared to other similar articles. Methodologically, it is important to remark that accurate measurements can be affected by different degrees of tissue shrinkage after the use of different fixation or contrast media. While the extension of shrinkage varies according to tissue properties, one must also consider a degree of tissue retraction after surgical excision (prior to fixation). Overall, we can generalize that our specimens retracted between 4 and 6% in all dimensions after excision and fixation [46].

When 3D histological reconstructions are in concern, a weakness is found in the manual alignment process [47]. The true trajectory of nerve fibers, namely their oblique direction along the SMA, was first noted on nanoCT scans, after 3D histology models had already been reconstructed. This can be explained by the fact that 3D histology reconstruction involves manual alignment and segmentation steps, both time-consuming and user-dependent processes.

This article documents how the SMAP network is contained within the paravascular sheath of the SMA. The extent of injury caused to the SMAP, while performing right colectomy with extended D3 mesenterectomy, was 48% in nerve area and 59% in nerve count. This knowledge, in turn, provides a better insight into the differences between excision and transection injury of the SMAP. Additionally, the 35°–55° inclination range of the SMAP fibers can imply an even larger injury of the plexus when a longitudinal excision is performed. Finally, the implementation of multiple research methodologies allowed for a unique and comprehensive understanding of plexus topography and microanatomy.