To our knowledge, this is the first attempt to systematically evaluate the specific cost of complications in bariatric surgery. While it appears logical that more significant complications generate higher perioperative treatment costs, more detailed insights are essential to allow more precise economic modeling when evaluating the financial impact of postoperative complications. A detailed cost data of complications may also offer insights into the source of expenses when perioperative complications occur. In addition, if the structure of the cost data is well understood in the clinical context, it may be used as a tool to minimize the financial impact of complications.

Our data show the baseline cost outcomes of patients without complications, which are in line with previous publications from our group [5, 8]. Interesting observations can be found when closely evaluating the itemized costs as outlined in Table 3: operating room and anesthesia costs are similar for patients with and without a Clavien I complication, but this post is slightly higher for patients with a Clavien II and significantly higher for Clavien III complication. It appears logical that a Clavien III complication incurs higher costs for the operating room on average because, as per its definition, this class of complication requires a re-intervention. However, this should not be the case for patients suffering from a Clavien II complication.  While not statistically significant – also due to the relatively small cohort of patients with a Clavien II complication - we found higher OR costs for patients with a Clavien II complication when compared to patients with or without a Clavien I complication. In this cohort, the cost increases most likely stem from the initial surgery as no re-intervention was conducted. This can be caused either by a prolonged procedure or by higher usage of instruments during the initial surgery. It seems reasonable to assume that a deviation from the standard surgical path (standard length case and/or standard set of instruments) might put the patients at a higher risk of perioperative complications. Still, this dataset does not allow the exact identification of cost drivers in each itemized segment and this comparison is not adequately powered. As such, there might also be a bias since longer procedures can be more complex in patients with high BMIs and other conditions that may lead to longer OR times or higher risk of postoperative complications. It would be interesting to explore further if cost-driving events in the OR can be used as predictors for perioperative complications and as such, if warning systems could be developed depending on the course of the surgical procedures. However, this dataset is not sufficiently deep for such analyses.

A closer look at the itemized costs of the patients with a Clavien I complication show comparable costs to the cohort without complications. However, we found an increase - although not statistically significant - in imaging and laboratory costs. This observation seems reasonable in the clinical context as a Clavien I complication is a deviation from the standard clinical path not requiring specific treatment. As a matter of fact, most of these patients did not have a real medical problem, bur received additional imaging or laboratory work prescribed by an inexperienced resident. As a teaching institution, residents are primarily responsible for the post-operative patients at night and over weekends and have the freedom to order these tests. Under these circumstances, we often observe a rather generous use sometimes without clear medical indication. However, as per definition these patients need to be classified as a Clavien I and it easily explains increased costs for imaging and laboratory. Another logical observation is the absence of intensive care unit costs for patients with or without a Clavien I and Clavien II complication, but the presence of intensive care unit costs when a re-intervention is needed. All the above mentioned clinically relevant consistencies suggest that this set of cost data is robust and can be used as a solid baseline for economic modeling.

Overall, Clavien I complications within the 18 days of robotic bariatric surgery seem to incur a moderate cost increase of about 3.6%, but the further increase of about 36% and 164% for Clavien II and III complications are rather exponential. It appears evident that any complication requiring treatment is not only devastating for the patient, but also an economic challenge to the healthcare system. As such, there is room for investments to avoid complications and from an economic perspective. As this is a first dataset quantifying the economic impact of complications in bariatric surgery, it can be used as a guide to show the extent of upfront investments that can be afforded to improve outcomes. While these investments might include a wide range of measures including surgeon training, patient preparation, and others, we want to focus on the investment in new surgical technologies. Albeit the absence of high quality and high level of evidence supporting robotics for a widespread application in bariatric surgery, this technology finds an increasing adoption, also driven by the desire to reduce perioperative complications of bariatric surgery. As an example, a few groups have published the positive impact of surgical robotics for gastric bypass surgery regarding procedure times per BMI and a reduction in severe complications such as intestinal leaks [8,14-16]. One study showed a cost reduction with a robotic approach versus laparoscopy for gastric bypass surgery due to a reduction in complications and the use of fewer surgical staplers [8]. This dataset shows that a Clavien III complication costs about USD 33,000. Therefore, a 1% reduction in a postoperative re-intervention rate would save USD 33,000 in 100 patients, and this money can be spent on new surgical technologies such as robotics.

Currently, the da Vinci Surgical System is the most widely used robot and its published costs have been listed above [7]. While these costs may fluctuate depending on the exact surgical approach and the instruments required, this post has been calculated to be USD 1,582.91 in a previous academic publication for gastric bypass surgery in 2011 [8]. Applying this cost structure, it becomes obvious, that a 1% reduction in re-interventions (saving about USD 33,000) would not be sufficient to counterbalance this upfront investment, but a significantly greater reduction of Clavien III complications would be required to financially justify the use of the da Vinci Surgical System. While this is a very simplified model, it outlines the practical applicability of this dataset and the extent of financial considerations that have to be made before widespread clinical adoption.

Still, several shortcomings of this analysis have to be mentioned: first, this dataset is mostly based on economic data and many conclusions require more data breadth and depth. In addition, this analysis includes a limited number of subjects in the healthcare environment of Switzerland. As such, the data might not be applicable in other geographic regions or might not be the same when looking at larger cohorts. Also, this is a retrospective analysis with all commonly known sources of bias. Lastly, the cost of robotic surgery is currently under development: Intuitive Surgical as well as new competitors address the cost issue with alternative models including flat-fee payments and lease options [17, 18]. Still, this is a unique approach for analyzing the bariatric cohort and might be a decent source of reflection for both clinicians and economists.