Comparative Outcomes of Open Mesenteric Bypass after Prior Failed Endovascular or Open Mesenteric Revascularization for Acute and Chronic Mesenteric Ischemia
Christopher Jacobs, Salvatore Scali, Thomas Huber, Martin Back, Dean Arnaoutakis, Michol Cooper, Scott Berceli, Gilbert Upchurch, Kristina Giles
Univeristy of Florida, Gainesville, FL
Background: Open mesenteric bypass(OMB) for acute or chronic mesenteric ischemia(AMI/CMI) can be technically demanding and is associated with significant postoperative morbidity and mortality. Increasingly, patients present after failed endovascular intervention or OMB with recurrent symptoms; however, little is known about outcomes with repeat OMB after prior failed mesenteric revascularization. The purpose of this analysis was to analyze outcomes of OMB after prior failed open or endovascular mesenteric artery revascularization.
Methods: A single center retrospective analysis was performed on all AMI/CMI patients undergoing OMB from 2002-2018. To determine the effect of prior failed mesenteric revascularization, primary(pOMB) and ‘redo’ mesenteric bypass(rOMB: defined as OMB after failed mesenteric stent and/or OMB) were compared. The primary end-point was in-hospital mortality. Secondary outcomes included complications, restenosis, freedom from re-intervention, and long-term survival. Kaplan-Meier life tables were utilized to estimate freedom from secondary end-points and multivariable Cox proportional hazards modeling was employed to identify predictors of survival.
Results: A total of 189 OMB procedures(pOMB, n=133[70%]; rOMB, n=56[30%]) were reviewed. rOMB patients were more often female(70% vs. 56%;p=.07), had lower incidence of preoperative smoking(current/prior) (41% vs. 56%;p=.05), but no other demographic, BMI or comorbidity differences. Compared to pOMB, rOMB was more frequently performed for CMI indication(70% vs. 52%)(p=.02). rOMB was performed for failed OMB rather than failed stent in 32%(n=18 of 56; 28%-CMI; 41%-AMI). Antegrade bypass configuration was employed in a majority of cases(rOMB-88% vs. pOMB-83%;p=.41) with single SMA target(vs. SMA and celiac) most frequent(rOMB-77% vs. pOMB-62%;p=.14). Conduits utilized between redo and pOMB had an apparent trend: Prosthetic(75% vs. 74%), autologous femoral vein(25% vs. 18%), cryo/bovine(0% vs. 8%)(P-trend=.08). No other procedure-related differences were identified(Table).
Complications(63% overall), LOS(21±23days), discharge to rehab versus home(34% among survivors) and in-hospital death(22% overall) were similar regardless of redo status for the overall group and for AMI/CMI independently. Redo status did not significantly increase the need for bowel resection at the index operation for AMI(AMI 47% vs. 44%;p=.81) or the need for secondary operative procedures for AMI/CMI(41% vs. 52%;p=.18).
rOMB was independently associated with higher risk of restenosis/occlusion at 1 and 3-years, respectively: rOMB-88±6%, 79±8%;pOMB-97±2%, 93±4%;p=.03) and lower freedom from re-intervention(1,3-year:rOMB-88±5%, 84±6%;pOMB-95±3%, 95±3%;p=.06)(Figure). Notably, rOMB was not an independent predictor of mortality and in fact was associated with a trend toward improved overall 1 and 5-year survival(rOMB-75±6%, 63±7% vs. pOMB-70±4%, 42±5%;p=.07). Predictors of mortality included chronic kidney disease(OR 1.7,1.1-2.5;p=.02) and CHF(1.8,1.1-3.1;p=0.3) while rOMB was protective(0.7,0.4-1;p=.05).
Conclusions: Patients presenting with recurrent AMI/CMI after prior failed endovascular or OMB can anticipate similar outcomes compared to primary OMB subjects. Conduit choice and configuration can be selectively applied depending on anatomic features and surgeon preference to achieve similar outcomes. Re-intervention rates are higher after rOMB, highlighting the need for implementation of surveillance protocols to optimize long-term durability.
| Table. Comparison for Redo vs. Primary Open Mesenteric Bypass for Acute and Chronic Mesenteric Ischemia | |||
| Feature, % (No.) | Redo N=56 | Primary N=133 | p-value |
| Age | 63±12 | 65±11 | .24 |
| Female sex | 70%(39) | 56%(74) | .07 |
| Preop BMI | 24±5 | 23±5 | .49 |
| Smoking (current/former) | 41%(23) | 56%(75) | .05 |
| Dyslipidemia | 61%(34) | 47%(62) | .08 |
| Preoperative Statin | 55%(31) | 36%(48) | .02 |
| Preoperative Clopidogrel | 36%(20) | 18%(24) | .01 |
| *No difference in frequency of CAD, HTN, DM, CKD Stage ≥ 3, CVOD, PAD, CHF between groups | |||
| Presentation and Outcomes | |||
| CMI (vs AMI) | 70%(39) | 52%(69) | .02 |
| Hospital transfer (data for CMI only) | 30%(12) | 15%(10) | .05 |
| EBL(mL) | 650±570 | 810±1130 | .38 |
| Crystalloid (mL) | 3400±1400 | 3350±1850 | .86 |
| Intraoperative packed red cells(units) | 0.9±1.2 | 1.6±.2.2 | .06 |
| Cell scavenger auto-transfusion(mL) | 100±250 | 130±450 | .70 |
| LOS, days(±SD) | 20±20 | 22±24 | .59 |
| Any Complication occurrence | 55%(31) | 66%(88) | .16 |
| In-hospital Death | 16%(9) | 25%(33) | .19 |
| 30-Day Death | 11%(6) | 16%(21) | .36 |
| F/u time, months (±SD) | 27±40 | 19±27 | .14 |
| Acute Mesenteric Ischemia Patients | N=17 | N=64 | |
| LOS, days(±SD) | 19±14 | 27±29 | .29 |
| Any Complication occurrence | 65%(11) | 78%(50) | .25 |
| In-hospital Death | 29%(5) | 38%(24) | .54 |
| 30-Day Death | 29%(5) | 23%(15) | .61 |
| Chronic Mesenteric Ischemia Patients | N=39 | N=69 | |
| LOS, days(±SD) | 21±23 | 18±17 | .49 |
| Any Complication occurrence | 51%(20) | 55%(38) | .70 |
| In-hospital Death | 10%(4) | 13%(9) | .67 |
| 30-Day Death | 3%(1) | 9%(6) | .21 |
Back to 2020 Posters