Comparison of Postoperative Bleed Rates and Location of Bleed Between Vessel Sealing Devices after Laparoscopic Sleeve Gastrectomy

Introduction

The gold standard in treatment for patients with obesity continues to be bariatric surgery. The laparoscopic sleeve gastrectomy (SG) is the most commonly performed bariatric procedure. ¹ Compared with other bariatric procedures it generally has a safer surgical profile, but it is not without complications. One of the most common complications is postoperative bleeding (POB), with rates ranging from 1%–9% ² in the literature. It is associated with significant postoperative morbidity including reoperation, readmission, and need for transfusion. ³ Various strategies exist to prevent POB in SG, including staple-line reinforcement, staple line oversewing, and application of fibrin sealants. No one study has found one method to be more effective than another.^2,4

Energy-based vessel sealing devices (VSD) are used to facilitate dissection and hemostasis, enabling surgeons to perform surgery efficiently and safely with decreased blood loss. ⁵ At our institution, two VSD, Thunderbeat^® (Olympus) and Maryland LigaSure™ (Covidien) are utilized for intraoperative dissection. The LigaSure™ device combines physical compression with radiofrequency energy to produce a modulated current that denatures proteins and elastin in vessel walls, enabling it to seal vessels up to 7 mm in diameter. ⁶ The Thunderbeat^® has this same max vessel sealing diameter, however it has the ability to provide dual integration of both bipolar and ultrasonic energies to create reliable vessel seals. ⁷ Both of these VSD are commonly used in laparoscopic abdominal surgery. The surgeon’s choice of which device to use typically comes down to availability at the institution, familiarity with using the instrument, and preference.

The purpose of this study was to evaluate POB in patients undergoing laparoscopic SG when using either the Thunderbeat^® or the Maryland LigaSure™. Primary outcomes included POB as defined by the Metabolic and Bariatric Surgery Accreditation Quality Improvement Program (MBSAQIP) registry between the two VSD. Secondary outcomes included location of bleed, re-admission, re-operation, transfusion requirements, and overall safety.

Methods

A retrospective, single center, review of all patients who underwent primary SG from July 1, 2013, to August 31, 2022, was performed using our MBSAQIP registry. The start date of 2013 was chosen as this is when both VSD became available for use at our institution. Patients that underwent conversion to SG from another bariatric procedure or patients undergoing primary robotic-assisted SG were excluded. Basic patient demographic characteristics included age at time of surgery, sex, race, and preoperative BMI. From this cohort, patients were categorized into two groups based on the VSD, either Maryland LigaSure™ or Thunderbeat^®, used during their SG. This categorization was performed by manually confirming patients’ operative notes to ensure each SG was placed into the correct VSD group. Venous thromboembolism (VTE) prophylaxis included early ambulation, sequential compression devices (SCDs), and subcutaneous heparin 5000 units TID (July 2013—June 2021) or weight based Lovenox dosing (July 2021 onwards).

Then, from this cohort, the MBSAQIP registry was reviewed for “postoperative bleeding,” which is defined as bleeding of the gastrointestinal (GI) tract within 30 days postop and readmission, reoperation, intervention, any blood transfusion, or a critical blood pressure drop and/or increase in heart rate that results in transfer to higher level of care or administration of pressors. Any patient’s chart who was found to have a POB per our MBSAQIP registry was further reviewed to assess secondary outcomes, including: if POB required readmission or reoperation, time to reoperation, the location of the bleed, any transfusion requirements, and the patient’s overall outcome.

SG were performed using the same general technique. The greater curvature was mobilized using either VSD. Starting approximately 5 cm from the pylorus, the sleeve is constructed over a 40 Fr Bougie with serial firings of the Endo-GIA 60 black stapler loads, followed by purple loads buttressed with bioabsorbable material (e.g., Seamguard) to decrease postoperative bleeding. Heparin is given preoperatively (on induction) and postoperative VTE prophylaxis was given as stated above, starting postoperative day zero.

Categorical variables were expressed as percentages and continuous variables as weighted mean and standard deviation. Analysis was performed using chi-square test for categorical data and independent sample t-test for continuous data. A P value <0.05 was considered statistically significant. Institutional Review Board approval was obtained.

Results

Discussion

A total of 8157 patients were identified in our MBASQIP registry during the study interval. Of those, 37 (.45%) patients were identified as having a POB. POBs are a well-studied complication with significant health care associated costs, patient morbidity, and mortality. Therefore, understanding what may put patients at higher risk of POB is important to consider when performing a SG. There are various techniques of staple line reinforcement to prevent POBs in SG, which has included the staple line oversewing with running absorbable suture, buttressing of the staple line with either biosynthetic or biological tissue, or application of fibrin glues along staple line.^2,4,8 However, the utilization of varying VSD specifically during a SG has not been examined to the same degree as the techniques stated above.

Literature around reinforcement of the staple line for bleeding prevention is favorable.

Berger et al. published an analysis of the 2012–2014 MBSAQIP registry which showed a statistically significant benefit to prevention of postoperative bleeding in SG patients who received staple line reinforcement with or without additional staple line oversewing. Further analysis, focusing on surgeon experience level, showed no association between intraoperative techniques and bleeding, suggesting surgeon experience may be more important than type of reinforcement used. ⁹ A more recently published analysis of MBSAQIP data from 2015–2019 by Aboueisha et al., also looked at bleeding complications with various staple line reinforcements and found that both buttressing and oversewing can decrease bleeding rates. ¹⁰ Not all studies show benefit to all forms of staple line reinforcement. Wang et al. performed a meta-analysis of seven randomized controlled trials, which found no significant difference in postoperative bleeding when oversewing was performed during SG. ¹¹ This study has a much smaller sample size compared with the prior two analyses which may explain the lack of a demonstrable difference.

None of the above studies address the use of VSDs and their effects on POB. The literature has firmly been focusing on staple reinforcement in POB prevention, and not surgical dissection instrument choice. To the best of our knowledge, this study involves the largest series of patients undergoing SG in which the differences between the LigaSure™ and Thunderbeat^® devices are examined. The Thunderbeat^® device was used more frequently in our study, but overall, there was no significant difference in overall POB rate between the devices. Velotti et al., performed a similar retrospective review of patients undergoing SG comparing outcomes between the LigaSure™ and the Harmonic Ace™, a VSD that functions similarly to the Thunderbeat^®. There was no significant difference found between overall bleeding rate, as presented in our study. ¹² While our study demonstrated no significant difference in overall POB rate, we did find that a POB was more likely to require reoperation when performed with the LigaSure™. There was no difference in reoperation rate reported in Velotti et al.’s review.

When a literature review is broadened beyond bariatric surgical procedures, there are papers directly comparing these VSDs. In the field of otorhinolaryngology, a prospective, randomized controlled trial of 28 patients was performed by Kuipers et al. In this study, the Thunderbeat^® combined with standard instrumentation was compared against standard instrumentation alone while performing a total laryngectomy and neck dissection. They found no difference in major bleed postoperatively, and additionally had less intraoperative blood loss and an overall shorter operative time. ¹³ Perhaps more comparable to bariatric surgery, Milsom et al. performed a randomized controlled trial of 60 patients undergoing left colectomy, assessing the clinical performance between the Thunderbeat^® and the Maryland LigaSure™ VSDs. Again, no difference in bleed rate was shown between the devices. ¹⁴

The incidence of POB after SG ranges from 1%–9%, and most commonly is the result of staple line bleeding. Other locations should also be investigated, including the cut edge of the mesentery/omentum, the short gastric vessels, the spleen, and the trocar sites on the abdominal wall.^2,15 In our study, the most common overall locations were the cut edge of the mesentery/omentum (10/24) and staple line (8/24). However, this location appeared to have a significant difference between the two VSD. POBs were found more commonly at the cut edge of mesentery/omentum in the LigaSure™ group, and more commonly at the staple line in the Thunderbeat^® group. In any reoperation for bleeding, all possible areas of the abdominal cavity should be thoroughly explored to find the source of bleeding. Knowing that there may be a propensity for a bleed to be more commonly found in one location when using a specific VSD can assist in surgical decision making during the index case, encouraging extra time to ensure hemostasis in those areas prior to completion.

Other examined outcomes in our cohort did not show significant differences. Time to reoperation and transfusion requirements were not statistically different among all groups. This suggests that regardless of instrument used, a bleed will likely reveal itself and require intervention in a similar time period. Transfusion requirements were not significantly different between the 2 groups, whether comparing the operative or nonoperative subgroups. This is likely due to the fact that the transfusion protocol remained the same at the institution of study during this time. One patient had multiple reoperations; however, the additional operations were secondary to a massive acute PMVT and not related to the initial POB. This patient was in the LigaSure™ group. One mortality occurred; this patient was in the Thunderbeat^® group. There was no statistical difference in mortality between the two VSD groups.

There are several limitations to this study. First, our study is retrospective and relies on previously collected data, and analysis is limited to the information that was available. It is possible that our analysis only captures significant bleeding events and may not capture patients who had clinically insignificant POB. The retrospective nature also accounts for the unequal number of cases in each VSD group or cohort. Second, our study includes 7 surgeons of varying levels of skill and experience which was not controlled for. Third, VSD choice was based on preference alone, there was no randomization involved. Fourth, staple line reinforcement with Seamguard was used as part of the general standard surgical technique, however, with the nature of the MBSAQIP registry, this would not capture any additional small reinforcement (i.e., clips or single suture).

Our findings suggest that while both the Thunderbeat^® and LigaSure™ VSD are safe to use in SG, the Thunderbeat^® is comparatively more hemostatic than the LigaSure™ as it results in fewer reoperative POB events. Additionally, POB from LigaSure™ are more likely to arise in the cut edge of mesentery, while POB from Thunderbeat^® are more likely to be located in the gastric staple line.