Our study aimed to implement the management recommendations derived from our previous research on preoperative investigations [8]. The findings revealed notable pathological results, prompting various modifications in the therapeutic approach including adjusting medication, repairing hiatal hernias, performing diverticulectomy and finally changing the initially planned surgical procedure.

Multiple studies have already investigated various abnormalities in the upper GI tract among bariatric patients [7,8,9,10]. However, despite ongoing research, there is still no consensus or standardized approach to preoperative examinations before bariatric surgery [15].

Moulla et al. examined preoperative EGD in 636 patients, revealing a change in the operative strategy in 1.6% with detection of esophageal adenocarcinomas in three cases (0.5%) [6]. Other studies emphasize the importance of detecting prevalent BE. In a study of 169 patients with a median 7.0 ± 1.5 years follow-up, the LSG group (n = 83) had 3 cases of de novo BE, while the LRYGB group (n = 86) had 1 case (3.6% versus 1.2%). Additionally, the LSG group reported higher prevalence of reflux symptoms and moderate-to-severe reflux esophagitis despite greater proton pump inhibitor use [16].

In our study, precancerous lesions were identified in 5 patients (0.6%), including 4 intestinal metaplasias and 1 ampullary adenoma. BE was observed in 16 patients (1.8%), with no detection of dysplasia requiring further treatment. Among these, BE was identified in three patients who underwent LSG. Initially, LRYGB had been recommended for these patients preoperatively. However, they chose LSG based on personal preference and positive experiences shared by family members or friends who had successful outcomes with LSG. All patients were informed about their increased risk of dysplasia and the potential development of Barrett’s carcinoma. Furthermore, it was advised that these patients undergo regular endoscopic surveillance to monitor for any progression of BE or the development of dysplasia.

According to the Bariatric Outcomes Longitudinal Database, LRYGB is more effective than other weight loss procedures in reducing GERD symptoms [14]. 5-year outcomes of combined data from two randomized clinical trials (SLEEVEPASS and SM-BOSS) revealed that around 8% of patients undergoing SG necessitated conversion to RYGB due to GERD [17].

In our study, we simplified the categorization of reflux esophagitis findings by summing all LA grades up as "reflux esophagitis" for analysis. However, according to the latest Lyon consensus, only LA grades C and D offer evidence of reflux [18]. Applying this refined criterion, we found that only 8 out of 153 LRYGB patients and 1 out of 22 SG patients should have been diagnosed with reflux esophagitis, significantly reducing reflux esophagitis prevalence from 5.2 to 0.9% in RYGB and 14.1% to 4.5% in SG. Despite having LA grade D reflux esophagitis, one patient chose LSG against our recommendation. Surprisingly, a follow-up endoscopy two years later showed grade A reflux esophagitis, suggesting an improvement in GERD-related symptoms. This supports data on the role of LSG in improving GERD symptoms [19, 20].

Our study found a 16.8% prevalence of hiatal hernias detected via EGD and 35.1% via upper GI series. This discrepancy may be attributed to the lower diagnostic accuracy of upper GI series for hiatal hernias [21]. However, our results are consistent with rates reported in other studies ranging between 20 and 40% [21, 22]. Hiatal hernias were more prevalent in patients undergoing LSG compared to those undergoing RYGB with 10.9% and 44.2% (diagnosed via EGD), respectively. Research suggests that hiatal hernias smaller than 2 cm may not be clinically significant [23]. However, as we lacked information on hernia size, we cannot determine the clinical significance of the detected hernias in our study. Intraoperatively, a higher proportion of hiatal hernias diagnosed before RYGB were fixed during surgery (67 out of 80; 83.8%) compared to those diagnosed before SG (32 out of 69; 46.4%), indicating potential overdiagnosis of hiatal hernias before SG. It should be noted that many, if not all, hiatal hernias would likely be identified intraoperatively, which may suggest that preoperative endoscopy and upper GI series do not contribute significantly to a change in therapy for these patients. This observation might imply that a larger number of patients would need to be screened to achieve a meaningful change in therapy. However, the primary strength of preoperative upper endoscopy lies not in the detection of hiatal hernias, but in its ability to diagnose dysplasias, malignancies, and severe reflux esophagitis, which we consider essential before performing bariatric surgery. In our previous study, for example, we identified one case of Barrett’s high-grade dysplasia, two cases of Barrett’s carcinoma, and one case of stomach cancer in asymptomatic patients [8]. In this regard, it is also important to note the significant variation in the costs of upper GI endoscopy, ranging from $3,000 to $6,000 in the United States, compared to approximately 350 CHF at our hospital, which may account for differences in the utilization of upper endoscopy across institutions. The therapeutic benefit of an upper GI series is therefore questionable, given that most relevant hiatal hernias are identified during the upper endoscopy and operative procedure itself. Overall, hiatal hernias were discovered intraoperatively in 21 patients (2.3%) who had not been identified in preoperative investigations. Nevertheless, knowing about a large hiatal hernia preoperatively still can be helpful for assembling the OR team and estimating the duration of the procedure.

The utilization of upper GI series and esophageal manometry is subject to debate. In our series, no alterations to the planned surgical procedure were made based on findings in the upper GI series.

During long-term follow-up, the preoperative manometric data of the esophageal body can be predictive of the development of postoperative esophageal dilation, stasis and aggravation or new onset of symptoms [24]. Patients with asymptomatic, compensated GERD but low-pressure LES are at high risk to develop GERD symptoms postoperatively which has to be taken into account when planning LSG [25].

Our data revealed that screening patients with esophageal manometry, whether symptomatic or not, resulted in pathological findings in 37 patients (25.3%), leading to a procedural change in 7 (4.8%) patients. When symptoms were present, the detection rate increased to 41.5%. These findings align with other studies suggesting that performing manometry before bariatric surgery in patients with esophageal symptoms, such as heartburn, regurgitation, dysphagia, and non-cardiac chest pain, may indicate abnormal esophageal motility [26].

It has been suggested that LRYGB may be the preferred procedure for patients with motility disorders, as reduced LES pressure following LSG could exacerbate GERD, thereby complicating esophageal motility disorders [27,28,29]. Achalasia has already been linked to unfavorable outcomes following bariatric surgery, and its management prior to surgery appears to ameliorate the postoperative course [29]. In this context, it is important to consider that the progression of achalasia could require Heller myotomy with Dor fundoplication, which is not feasible after LSG [30]. Our findings indicate that maintaining the practice of routine manometry before LSG mainly yields advantages for symptomatic patients. However, it seems reasonable to extend the necessity of manometry in any patient with symptoms suggestive of severe motility disorder before any bariatric procedure [26]. While manometry did not reveal significant value in asymptomatic patients scheduled for LSG or LRYGB, neither upper GI series nor manometry influenced surgical decisions for those undergoing LRYGB.

Prioritizing cost-effectiveness as a rationale for deciding upon specific preoperative examinations is recommended by some authors [5, 31]. While cost considerations are undoubtedly important, it is crucial to acknowledge that certain findings can profoundly impact a patient’s life, even if they occur rarely. In our case, findings, such as precancerous lesions, can have a significant and lasting effect, particularly if they are located in the gastric remnant. The challenging endoscopical access associated with LRYGB could potentially carry the risk of progression to carcinoma. Furthermore, the timely diagnosis of severe GERD, BE, and severe motility disorders before LSG can prevent patients from requiring a later conversion to LRYGB [32, 33].

Building upon the findings of our previous study [8], which indicated a lack of clinical relevance in performing upper GI series and manometry, particularly when LRYGB is planned, the current study provides further evidence supporting the safety of foregoing upper GI series in all patients. Hence, we suggest to conduct manometry exclusively in symptomatic patients, aligning with the existing ASMBS guidelines that already advocate a similar approach for upper GI endoscopy [15]. The lack of a universal consensus highlights the necessity for additional research and development in this field to establish more standardized and evidence-based protocols for the preoperative assessment in bariatric surgery.

Our study’s limitations stem primarily from its retrospective nature, which poses constraints on data collection and introduces potential biases. The generalizability of our findings is limited due to the lack of international consensus on standardized preoperative investigations for bariatric surgery, compounded by variations arising from individual surgeon preferences and healthcare systems. Additionally, differences in healthcare costs and regional disparities further complicate the approach to preoperative assessments. Diagnostic findings may be operator-dependent and vary due to personal interpretations, and differences between clinics. Moreover, the definition of clinically significant findings lacks clarity and relies on subjective surgeon interpretation.