Endoscopic bariatric surgeries are spreading thanks to their minimally invasive procedure and less risk for the patients [22], and the ESG is facing a promising development thanks to the successful clinical outcomes [14, 23, 24]. Several protocols could be adopted to perform the plications, from transverse monolinear (the C-shaped) [14] to greater curvature compression sutures (Z-shaped coupled with U-shaped) [16], since it has been stated that suture pattern does not influence the outcome in terms of weight loss and comorbidity remission at 12-month follow-up [12]. However, nothing is known in the long term, as well as the different mechanical solicitations that a specific suture pattern may induce within the stomach tissues.

For this reason, this study was aimed at mechanically comparing the effects of different suture patterns on gastric wall in terms of stresses and strains and the volumetric reduction of the stomach, thus providing a quantitative description of the implication of ESG procedures, not based only on a posteriori clinical outcome (BMI, TWL, etc.).

When comparing the capacity variation of the stomach volume after the ESG, the U-shaped achieved the greatest reduction (-80% at baseline and -63% at 15 mmHg of intragastric pressure), while the Z-shaped the lowest, (about -50% at baseline and -26% at 15 mmHg of intragastric pressure) (Table 1). These numbers reflexed also within the pressure–volume curves (Fig. 2), where the more the reduction, the more the pressurisation of the stomach at the same reached volume. This could be directly linked to the potential amount of food that can be introduced in the stomach cavity before feeling satiety, thus the U-shaped revealed the strongest food volume reduction, potentially correlated to a faster and greater patient weight loss. However, in the clinical practise the Z-shaped is usually performed combined with other Z-shaped suture patterns [12] or with a U-shaped path [16]. Thus, the resulting post-ESG stomach would result in a greater stomach reduction. However, the aim of this work was to highlight the differences between these suture patterns from a biomechanical point of view, focussing on the variation in stress/strain distributions and stomach shortness after ESG.

The presence of the sutures constrained the tissues in ESG configurations, which was not able to expand itself, decreasing the elongation strain values (Fig. 3a), especially within the corpus region (from an average of 67% of elongation strain to 57% in the presence of a U-shaped suture pattern). However, the antrum region resulted to be the less influenced stomach region, with an oscillation of ± 2% of elongation strain variation with respect to the pre-surgical model.

When considering the tissue tensile stresses, results from the same inflation volume (1000 ml, Fig. 3b) were compared, revealing the U-shaped pattern as the one that solicitates the most the gastric walls (on average more than three times the average pre-surgical stresses). Indeed, this pattern causes a biaxial loading condition, solicitating both the longitudinal as well as the circumferential direction. Moreover, even if the sutures were applied within a unique region (the corpus), all the three regions faced a significant stress increase, with the antrum the most altered zone (stresses up to eight times greater with respect to the pre-surgical stresses at the same stomach volume). As stated before, a single Z-shaped pattern caused a reduced effect also in terms of stresses, even if it should be considered as a partial but already important effect.

However, the study of this specific pattern was of interest to quantify the stomach shortness, caused especially by this kind of sutures. Stomach shortness is a result of ESG and leads to a faster stomach emptying and an enhanced stomach stiffness along the longitudinal direction. This latter corresponds to the direction of peristalsis movement (active behaviour) during torniodigestion, thus longitudinal rigidity could affect the stomach primary function due to the interaction with the sutures, that often could lead to tissue damage and suture reopening. In order to monitor stomach shortness, the great curvature length (l) or the vertical distance covered by the sutures (e.g. the corpus height, h) could be computed and compared (Fig. 4). In this sense, the Z-shaped pattern may cause the strongest stomach reduction, decreasing l of about 20%, but especially along the vertical direction, with h reduced of 65%. This influence became even more pronounced when the stomach is inflated: being constrained to expand mainly circumferentially rather than longitudinally, when the intragastric pressure reached 15 mmHg the stomach great curvature resulted about one half the pre-operative one, while h only slightly modified as reported in Table 2, from 47.2 mm (baseline) to 51.4 mm (74%). This confirmed the Z-suture hinders the physiological gastric volumetric elongation during digestion, and this effect could be even stronger if then the Z-shaped pattern is coupled with other suture patterns.

Bars of the stomach shortening measured in two ways, by considering the greater curvature and “vertical” length, for each suture pattern analysed

Results also showed that the other sutures patterns as the C-shaped and the U-shaped could cause stomach shortness (Table 2 and Fig. 4), but up to a maximum of about 40%.

Comparisons with previous finite-element analyses were in agreement with these findings [18, 25, 26], such as an increase in the stomach stiffness (expressed as the ratio pressure/volume) by increasing the number of ESG sutures (the U-shaped is also the pattern with the major number of stiches), reducing the average strain that the stomach exhibits. Also, when including the stomach-specific geometry and a higher intragastric pressure [18], the corpus remains the region with smaller strains compared to the other regions. A first attempt was made to analyse endoscopic procedures in a previous work [20], however the geometry was not patient-specific, and the material parameters were related to porcine gastric tissues.

It should be noticed that this study also acknowledges several limitations inherent in the complexity of the problem. These limitations can be seen in the material parameters used to describe the mechanical behaviour of gastric regions (assumed as an average for the entire stomach regions, more details reported in [18]), the adopted boundary conditions applied to the gastroesophageal and gastroduodenal junctions, and the absence of surrounding organs. All these aspects might collectively contribute to an overestimation of the final volume of the inflated stomach.

Being aware of these limitations, this computational analysis highlighted some major aspects that should be evaluated in the clinical practise, not only to fast reach the strongest stomach reduction (hopefully leading to the reduction of patient weight) but also the mechanical and functional aspects that govern the activation of the mechanoreceptors populating gastric wall. Indeed, as stated by Tack et al. [27], the gastric accommodation reflex (a transient relaxation of the proximal stomach during food intake) has been identified as a major factor which controls meal volume through the activation of tension‐sensitive gastric mechanoreceptors, which similarly mediate gastric filling‐related satiation signals. For these reasons, a mechanical description of the solicitation after food intake could help the clinicians in understating in a deeper way the gut–brain axis, helping bariatric patients in loosening a large amount of weight in an easier and sustainable way, maintaining the improvements in the long terms.

Another information that this model could provide concerns the effect exerted by both gastric filling and peristalsis on the sutures. This could help define which characteristics the suture must have to last longer over time, before failing. The choice of suture type falls on the compromise amongst the following factors: proper gastric reduction, major fundus solicitations and contained distensions at the level of the sutures. From the computational analysis, the authors identified as preliminary best choice a “C-shaped” suture pattern, leaving room for further analyses adding peristalsis in the model.