Anticoagulation in emergency general surgery: Who bleeds more? The EAST multicenter trials ACES study

BACKGROUND

While direct oral anticoagulant (DOAC) use is increasing in the Emergency General Surgery (EGS) patient population, our understanding of their bleeding risk in the acute setting remains limited. Therefore, the objective of this study was to determine the prevalence of perioperative bleeding complications in patients using DOACs versus warfarin and AP therapy requiring urgent/emergent EGS procedures (EGSPs).

METHODS

This was a prospective observational trial, conducted between 2019 and 2022, across 21 centers. Inclusion criteria were 18 years or older, DOAC, warfarin/AP use within 24 hours of requiring an urgent/emergent EGSP. Demographics, preoperative, intraoperative, and postoperative data were collected. ANOVA, χ 2 , and multivariable regression models were used to conduct the analysis.

RESULTS

Of the 413 patients enrolled in the study, 261 (63%) reported warfarin/AP use and 152 (37%) reported DOAC use. Appendicitis and cholecystitis were the most frequent indication for operative intervention in the warfarin/AP group (43.4% vs. 25%, p = 0.001). Small bowel obstruction/abdominal wall hernias were the main indication for operative intervention in the DOAC group (44.7% vs. 23.8%, p = 0.001). Intraoperative, postoperative, and perioperative bleeding complications and in-hospital mortality were similar between the two groups. After adjusting for confounders, a history of chemotherapy (odds ratio [OR], 4.3; p = 0.015) and indication for operative intervention including occlusive mesenteric ischemia (OR, 4.27; p = 0.016), nonocclusive mesenteric ischemia (OR, 3.13; p = 0.001), and diverticulitis (OR, 3.72; p = 0.019) were associated with increased perioperative bleeding complications. The need for an intraoperative transfusion (OR, 4.87; p < 0.001), and intraoperative vasopressors (OR, 4.35; p = 0.003) were associated with increased in-hospital mortality.

CONCLUSION

Perioperative bleeding complications and mortality are impacted by the indication for EGSPs and patient's severity of illness rather than a history of DOAC or warfarin/AP use. Therefore, perioperative management should be guided by patient physiology and indication for surgery rather than the concern for recent antiplatelet or anticoagulant use.

LEVEL OF EVIDENCE

Prognostic and Epidemiologic; Level III.

Comparison of a novel, endoscopic chest tube insertion technique versus the standard, open technique performed by novice users in a human cadaver model: a randomized, crossover, assessor-blinded study

Background

The technique of tube thoracostomy has been standardized for years without significant updates. Alternative procedural methods may be beneficial in certain prehospital and inpatient environments with limited resources. We sought to compare the efficacy of chest tube insertion using a novel, endoscopic device (The Reactor™) to standard, open tube thoracostomy.

Methods

Novice users were randomly assigned to pre-specified sequences of six chest tube insertions performed on a human cadaver model in a crossover design, alternating between the Reactor™ and standard technique. All subjects received standardized training in both procedures prior to randomization. Insertion site, which was randomly assigned within each cadaver’s hemithorax, was marked by the investigators; study techniques began with skin incision and ended with tube insertion. Adequacy of tube placement (intrapleural, unkinked, not in fissure) and incision length were recorded by investigators blinded to procedural technique. Insertion time and user-rated difficulty were documented in an unblinded fashion. After completing the study, participants rated various aspects of use of the Reactor™ compared to the standard technique in a survey evaluation.

Results

Sixteen subjects were enrolled (7 medical students, 9 paramedics) and performed 92 chest tube insertions (n = 46 Reactor™, n = 46 standard). The Reactor™ was associated with less frequent appropriate tube positioning (41.3% vs. 73.9%, P = 0.0029), a faster median insertion time (47.3 s, interquartile range 38–63.1 vs. 76.9 s, interquartile range 55.3–106.9, P < 0.0001) and shorter median incision length (28 mm, interquartile range 23–30 vs. 32 mm, interquartile range 26–40, P = 0.0034) compared to the standard technique. Using a 10-point Likert scale (1-easiest, 10-hardest) participants rated the ease of use of the Reactor™ no different from the standard method (3.8 ± 1.9 vs. 4.7 ± 1.9, P = 0.024). The Reactor™ received generally favorable scores for all parameters on the post-participation survey.

Conclusions

In this randomized, assessor-blinded, crossover human cadaver study, chest tube insertion using the Reactor™ device resulted in faster insertion time and shorter incision length, but less frequent appropriate tube placement compared with the standard technique. Additional studies are needed to evaluate the efficacy, safety and potential advantages of this novel device.

Stop the Bleed: Does the Training Work One Month Out?

The Stop the Bleed initiative empowers and trains citizens as immediate responders, to recognize and control severe hemorrhage. We sought to determine the retention of short-term knowledge and ability to apply a Combat Application Tourniquet (CAT) in 10 nonmedical personnel. A standard "Stop the Bleed" (Bleeding Control) course was taught including CAT application. Posttraining performance was assessed at 30 days using a standardized mannequin with a traumatic below-knee amputation. Technique, time, pitfalls, and feedback were all recorded. No participant had placed a CAT before the initial class. After the initial class, self-report by a Likert scale survey revealed an increased confidence in tourniquet application from 2.4 pretraining to 4.7 posttraining. At 30 days, confidence decreased to 3.4 before testing. Six of 10 were successful at tourniquet placement. Completion time was 77.75 seconds (43-157 seconds). Successful participants reported a confidence level of 4.7 versus those unsuccessful at 3.3. The "Stop the Bleed" initiative teaches lifesaving skills to the public through a short training course. This information regarding the training of nonmedical personnel may assist in strengthening training efforts for the public. Further investigations are needed to characterize skill degradation and retention over time.

Stop the Bleed: Does the Training Work One Month Out?

The Stop the Bleed initiative empowers and trains citizens as immediate responders, to recognize and control severe hemorrhage. We sought to determine the retention of short-term knowledge and ability to apply a Combat Application Tourniquet (CAT) in 10 nonmedical personnel. A standard “Stop the Bleed” (Bleeding Control) course was taught including CAT application. Posttraining performance was assessed at 30 days using a standardized mannequin with a traumatic below-knee amputation. Technique, time, pitfalls, and feedback were all recorded. No participant had placed a CAT before the initial class. After the initial class, self-report by a Likert scale survey revealed an increased confidence in tourniquet application from 2.4 pretraining to 4.7 posttraining. At 30 days, confidence decreased to 3.4 before testing. Six of 10 were successful at tourniquet placement. Completion time was 77.75 seconds (43–157 seconds). Successful participants reported a confidence level of 4.7 versus those unsuccessful at 3.3. The “Stop the Bleed” initiative teaches lifesaving skills to the public through a short training course. This information regarding the training of nonmedical personnel may assist in strengthening training efforts for the public. Further investigations are needed to characterize skill degradation and retention over time.

Resuscitative endovascular balloon occlusion of the aorta for non-traumatic intra-abdominal hemorrhage

BACKGROUND

Hemorrhagic shock is the second leading cause of death in blunt trauma and a significant cause of mortality in non-trauma patients. The increased use of resuscitative endovascular balloon occlusion of the aorta (REBOA) as a bridge to definitive control for massive hemorrhage has provided promising results in the trauma population. We describe an extension of this procedure to our hemodynamically unstable non-trauma patients.

METHODS

This is a retrospective review of patients requiring REBOA for end stage non-traumatic abdominal hemorrhage from our tertiary care facility. After excluding patients with trauma, supradiaphragmatic bleed and thoracic/abdominal aortic aneurysms, demographics, etiology of bleed, REBOA placement specifics, complications and outcomes were reviewed.

RESULTS

From August 2013 to August 2016, 11 patients were identified requiring REBOA placement for hemodynamic instability from non-traumatic abdominal hemorrhage. Average patient age was 54.9 (SD 15.2). Sixty-four percent suffered cardiac arrest prior to REBOA, with mean shock index of 1.29. Average time from diagnosis of shock (MAP ≤ 65) or signs of bleeding to placement of REBOA was 177 min. The leading etiologies of hemorrhage were ruptured visceral aneurysm and massive upper gastrointestinal bleed. REBOA was placed by both acute care and vascular surgeons. The procedure was mainly completed in the operating room in 82% of the patients and at the bedside in 18%. One patient expired before operative repair. Definitive surgical control of the source of bleeding was obtained by open surgical approach (n = 6) and combined surgical and endovascular approach (n = 4). In-hospital survival was 64%. There were no local complications related to REBOA placement.

CONCLUSION

Similar to the trauma population, REBOA is an adjunctive technique for proximal control of bleeding as well as resuscitation in end stage non-traumatic intra-abdominal hemorrhage. We propose an algorithmic approach to REBOA use in this population and a larger prospective review is necessary to determine both the timing of REBOA placement and which non-traumatic patients may benefit from this technique.

LEVEL OF EVIDENCE

V.

STUDY TYPE

Brief report.

"Second-look" laparotomy: warranted, or contributor to excessive open abdomens?

INTRODUCTION

The overuse of temporary abdominal closure and second look (SL) laparotomy in emergency general surgery (EGS) cases has been questioned in the recent literature. In an effort to hopefully decrease the number of open abdomen (OA) patients, we hypothesize that reviewing our cases, many of these SL patients could be managed with single-stage operative therapy and thus decrease the number of OA patients.

METHODS

This is a retrospective review of prospectively collected data from Jun 2013-Jun 2014, evaluating EGS patients managed with an OA who required bowel resection in either index or SL laparotomy. Demographics, clinical variables, complications and mortality were collected. Fisher's exact t test was used for statistical analysis.

RESULTS

During this time frame, 96 patients were managed with OA and 59 patients required a bowel resection. 55 (57%) of those required one bowel resection at the index operation with 4 (4.2%) only requiring one bowel resection at the second operation. In the patients requiring bowel resections, 18 (30%) required a resection at SL. At SL laparotomy, resection was required for questionably viable bowel at the index operation 60% (11), whereas 39% (7) had normal appearing bowel. Indications for resection at SL laparotomy included evolution of existing ischemia, new onset ischemia, staple line revision, and "other". 23 patients (39%) were hemodynamically unstable, contributing to the need for temporary abdominal closure. In the multivariate analysis, preoperative shock was the only predictor of need for further resection. Complications and mortality were similar in both groups.

CONCLUSION

Almost one-fifth of the patients undergoing SL laparotomy for open abdomen required bowel resections, with 6.8% of those having normal appearing bowel at index operation, therefore in select EGS patients, SL laparotomy is a reasonable strategy.

Resuscitative Endovascular Balloon Occlusion of the Aorta: Pushing Care Forward

BACKGROUND

Resuscitative endovascular balloon occlusion of the aorta (REBOA), used to temporize noncompressible and junctional hemorrhage, may be deployable to the forward environment. Our hypothesis was that nonsurgeon physicians and high-level military medical technicians would be able to learn the theory and insertion of REBOA.

METHODS

US Army Special Operations Command medical personnel without prior endovascular experience were included. All participants received didactic instruction of the Basic Endovascular Skills for Trauma Course™ together, with individual evaluation of technical skills. A pretest and a posttest were administered to assess comprehension.

RESULTS

Four members of US Army Special Operations Command-two nonsurgeon physicians, one physician assistant, and one Special Operations Combat Medic-were included. REBOA procedural times moving from trial 1 to trial 6 decreased significantly from 186 ± 18.7 seconds to 83 ± 10.3 seconds (ρ < .0001). All participants demonstrated safe REBOA insertion and verbalized the indications for REBOA insertion and removal through all trials. All five procedural tasks were performed correctly by each participant. Comprehension and knowledge between the pretest and posttest improved significantly from 67.6 ± 7.3% to 81.3 ± 8.1% (ρ = .039).

CONCLUSION

This study demonstrates that nonsurgeon and nonphysician providers can learn the steps required for REBOA after arterial access is established. Although insertion is relatively straightforward, the inability to gain arterial access percutaneously is prohibitive in providers without a surgical skillset and should be the focus of further training.

Hydrophobically modified chitosan gauze: a novel topical hemostat

BACKGROUND

Currently, the standard of care for treating severe hemorrhage in a military setting is Combat Gauze (CG). Previous work has shown that hydrophobically modified chitosan (hm-C) has significant hemostatic capability relative to its native chitosan counterpart. This work aims to evaluate gauze coated in hm-C relative to CG as well as ChitoGauze (ChG) in a lethal in vivo hemorrhage model.

METHODS

Twelve Yorkshire swine were randomized to receive either hm-C gauze (n = 4), ChG (n = 4), or CG (n = 4). A standard hemorrhage model was used in which animals underwent a splenectomy before a 6-mm punch arterial puncture of the femoral artery. Thirty seconds of free bleeding was allowed before dressings were applied and compressed for 3 min. Baseline mean arterial pressure was preserved via fluid resuscitation. Experiments were conducted for 3 h after which any surviving animal was euthanized.

RESULTS

hm-C gauze was found to be at least equivalent to both CG and ChG in terms of overall survival (100% versus 75%), number of dressing used (6 versus 7), and duration of hemostasis (3 h versus 2.25 h). Total post-treatment blood loss was lower in the hm-C gauze treatment group (4.7 mL/kg) when compared to CG (13.4 mL/kg) and ChG (12.1 mL/kg) groups.

CONCLUSIONS

hm-C gauze outperformed both CG and ChG in a lethal hemorrhage model but without statistical significance for key endpoints. Future comparison of hm-C gauze to CG and ChG will be performed on a hypothermic, coagulopathic model that should allow for outcome significance to be differentiated under small treatment groups.

Penetrating fetal trauma with late complications: a case report

Survival after direct penetrating trauma to a fetus is very unusual. Our case below illustrates nonoperative management of a penetrating thoracoabdominal gunshot injury to the retroperitoneum with a late complication arising 15 years after the initial incident.