- Email: [email protected]
Superior mesenteric artery stenting using embolic protection device for treatment of acute or chronic mesenteric ischemia
From the Society for Vascular Surgery
Superior mesenteric artery stenting using embolic protection device for treatment of acute or chronic mesenteric ischemia Bernardo C. Mendes, MD, Gustavo S. Oderich, MD, Tiziano Tallarita, MD, Karina S. Kanamori, MD, Manju Kalra, MBBS, Randall R. DeMartino, MD, Fahad Shuja, MBBS, and Jill K. Johnstone, MD, Rochester, Minn
ABSTRACT Objective: The objective of the study was to report the feasibility and results of superior mesenteric artery (SMA) stenting using embolic protection devices (EPDs) to treat acute mesenteric ischemia (AMI) and chronic mesenteric ischemia (CMI). Methods: A retrospective review was conducted of consecutive patients who underwent SMA stenting with EPDs from 2007 to 2016. EPDs were used selectively in patients with occlusions, severe calcification, or acute thrombus. A two-wire technique with SpiderFX 0.014-inch filter wire (Medtronic, Minneapolis, Minn) combined with a 0.018-inch wire was used to provide support and to facilitate stenting and EPD retrieval. Presence of macroscopic debris in the EPD was recorded and graded as minor (minimal debris) or major (large thrombus or plaque). End points were technical success, presence of EPD debris, embolization, early morbidity, and mortality. Results: SMA stenting was performed in 179 patients, of whom 65 (36%) had EPDs. The mean age was 73 6 11 years, and 49 were female (75%). Clinical presentation was CMI in 48 patients (74%) and AMI or acute-on-CMI in 17 (26%). Indications for EPD were severe calcification in 22 patients (34%), acute thrombus in 18 (28%), and total occlusion in 16 (25%). Bare-metal stents were used in 33 patients, covered stents in 26, and both types in 6. Adjunctive therapy included thrombolysis in seven patients, thrombectomy in four, and atherectomy in three. Technical success was 100%. There were no instances of filter retention or arterial trauma due to filter manipulation. Distal embolization was noted in four patients (6%), of whom two had AMI. All large emboli were retrieved using catheter aspiration devices, but one small distal embolus was left untreated with no clinical consequences. Two patients had vessel spasm treated by nitroglycerin. Macroscopic debris was noted in 43 patients (66%) and was major in 21 (49%) or minor in 22 (51%). Of the patients with AMI, five (29%) required exploratory laparotomy and four (23%) had bowel resection. Eight additional patients (12%) had early complications (five CMI, three AMI), including cardiac complications, brachial hematoma, acute cholecystitis, and acute respiratory distress syndrome in two patients each. There were no deaths among CMI patients and two early deaths (12%) among those who had AMI. Conclusions: Use of EPDs during SMA stenting is safe and feasible with a two-wire technique. Large macroscopic debris was noted in one-third of the patients when the filter was applied selectively in patients with acute symptoms, occlusions, or severely calcified lesions. Despite the use of EPD, distal embolization occurred in 6% of patients and was successfully treated using catheter aspiration devices. (J Vasc Surg 2018;-:1-8.) Keywords: Chronic mesenteric ischemia; Endovascular treatment; Embolic protection
Endovascular therapy has become the main modality for treatment of atherosclerotic mesenteric arterial disease. Several large series have shown high rates of technical success, symptom improvement, and low morbidity and mortality in patients treated for chronic From the Division of Vascular and Endovascular Surgery, Mayo Clinic. Author conflict of interest: none. Presented at the 2016 Vascular Annual Meeting of the Society for Vascular Surgery, National Harbor, Md, June 8-11, 2016. Correspondence: Gustavo S. Oderich, MD, Gonda Vascular Center, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2018 Published by Elsevier Inc. on behalf of the Society for Vascular Surgery. https://doi.org/10.1016/j.jvs.2017.12.076
mesenteric ischemia (CMI). Patients with acute mesenteric ischemia (AMI) and acute-on-CMI have higher mortality rates and lower technical success, which is related to duration of ischemia, presence of advanced bowel gangrene, and significant thrombus burden within the mesenteric arteries.1-3 In these patients, the superior mesenteric artery (SMA) is generally affected by complex, heavily calcified plaque or associated thrombus, which often requires recanalization of chronic total occlusion or in situ acute-on-chronic thrombotic occlusions. These lesions have higher risk of thromboembolic complications and acute vessel or stent reocclusion, which can lead to bowel gangrene, multisystem organ failure, and death.4 Embolic protection devices (EPDs) have been designed for coronary grafts and carotid arteries but have been successfully used off-label in other vascular beds.5,6 There 1
2
Journal of Vascular Surgery
Mendes et al
---
is little if any controversy that EPDs should be indicated when the consequences of embolization cause irreversible end-organ damage, such as during carotid stenting, which has a known risk of embolic stroke.7 Mesenteric stenting can also cause embolization from catheter manipulations or dislodgment of plaque or thrombus, which may result in occlusion of distal branches, leading to vessel thrombosis and bowel gangrene.4 This is particularly important if the vessel occlusion is not immediately recognized and treated. Despite the relative lack of literature, we have used EPDs selectively in patients with heavily calcified stenosis, occlusions, and acute or subacute presentations. The objective of our study was to analyze the indications, techniques, and outcomes of patients who underwent primary stenting of the SMA using EPDs to treat occlusive mesenteric artery disease.
METHODS This study was approved by the Mayo Clinic Institutional Review Board. All patients consented with research participation. We performed a retrospective review of consecutive patients who underwent SMA stenting with EPDs from 2007 to 2016. The indication for use of an EPD was left to the discretion of the practitioner and was recorded when available on medical records. In general, indications were chronic occlusions, lesions longer than 3 cm with >66% circumferential or luminal calcification, and acute or subacute symptom presentation. EPDs were also used in patients without these anatomic criteria at the discretion of the treating physician. We excluded from the study patients who had mesenteric stenting without embolic protection. Demographics, cardiovascular risk factors, and clinical presentation were recorded. The primary symptom presentation was classified as chronic (>4 weeks), subacute (2-4 weeks), or acute (<2 weeks) on the basis of duration of symptoms. Lesion characteristics were analyzed using preoperative computed tomography angiography whenever possible. Calcified lesions were further analyzed as encompassing <33%, 33% to 66%, or >66% of the vessel circumference in the proximal 1 to 4 cm of the vessel. The length of circumferential calcified plaque, occlusion, and thrombus was also noted. Thrombus was recorded as present when irregular, hypodense filling defects were identified on computed tomography angiography, both occlusive and nonocclusive. Procedural details were noted, including approach (brachial or femoral), stent type, diameter and length, use of dilation before or after the intervention, EPD type and diameter, and presence of intraoperative complications. EPD-related complications were defined as side branch embolization, difficult filter retrieval, improper deployment, and target vessel spasm. Debris in the EPD basket was recorded and usually photographed. This was classified as absent, minor (diminutive or <20% of the filter basket), or major (large debris,
2018
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Retrospective, single-center cohort study Take Home Message: In 65 of 179 patients presenting with mesenteric ischemia, embolic protection devices were used selectively during superior mesenteric artery stenting in cases of occlusion, severe calcification, or acute thrombus with a technical success rate of 100% and macroscopic debris identified in 66% of cases. Recommendation: This study suggests that embolic protection devices should be used during superior mesenteric artery stenting in cases of occlusions, severe calcification, and acute thrombus.
plaque, or >20% of filter basket). Early outcomes were recorded, including morbidity, mortality, length of intensive care unit and hospital stay, and early reinterventions. Technique. The technique of mesenteric artery stenting has been previously described by our group.8 The left brachial artery approach using surgical exposure with a small incision was used preferentially, but the femoral approach was indicated in patients with a heavily diseased aortic arch and favorable angle of origin in the SMA. A catheter support system was built using a combination of a 6F or 7F 90-cm hydrophilic sheath combined coaxially with a 7F guide catheter and a 5F multipurpose A (MPA) catheter. For complex lesions, such as chronic total occlusions or heavy calcified lesions, the 7F sheath, 7F MPA guide catheter, and 5F MPA catheter were used to engage the stump of the occluded vessel, allowing enough support for passage of a guidewire to the distal SMA (Fig 1). The target lesion was initially crossed using a 0.035-inch soft straight or angled Glidewire (Terumo Interventional Systems, Somerset, NJ), which was exchanged for the interventional wire of choice after confirmation of true lumen access. Our preference was to use a small-profile (0.014- or 0.018inch) stiff guidewire, such as the V-14 or V-18 wires (Boston Scientific, Marlborough, Mass), for most interventions. A covered stent was selected in most patients with proximal SMA lesions that did not involve side branches. Predilation is typically performed for very high grade stenosis and occlusions or in cases in which crossing of the lesion is otherwise difficult. A 320-cm SpiderFX 0.014inch filter wire (Medtronic, Minneapolis, Minn) was used in all except one patient, who had the Accunet Embolic Protection System (Abbott Vascular, Santa Clara, Calif). If a 0.035-inch stent is selected, a two-wire technique is used by combining the 0.014-inch filter wire with a 0.018inch “buddy wire”; the stent is introduced over both wires for better support and to facilitate subsequent retrieval of the EPD.
Journal of Vascular Surgery Volume
-,
Number
Mendes et al
3
-
Fig 1. In cases of superior mesenteric artery (SMA) occlusion, a stiff support system is built with the combination of a 7F sheath, a 7F multipurpose guide catheter, and a 5F multipurpose catheter, which is used to engage the stump of the occluded SMA (A). The lesion is crossed with a hydrophilic Glidewire, followed by a 0.014-inch filter wire and a 0.018-inch buddy wire, deployed through a 0.035-inch catheter (B). The lesion is predilated and subsequently treated with a balloon-expandable stent (C). The completion angiogram after filter retrieval demonstrates a flared, widely patent SMA stent with no distal embolization (D). (Image used with permission of Mayo Foundation for Medical Education and Research. All rights reserved.)
Statistical analysis. Primary end points were technical success (defined as successful stent placement with residual stenosis <30% and filter retrieval), presence of EPD debris, embolization, early (<30 days or in hospital) morbidity, and mortality. Differences were determined by the log-rank test. The Pearson c2 or Fisher exact test was used for analysis of categorical variables. Differences between means were tested with two-sided Student t-test, Wilcoxon rank sum test, or Mann-Whitney test. A P value of < .05 was used to
determine statistical significance. Data are presented as median values or mean 6 standard deviation, as appropriate.
RESULTS Patients. During the study period, 179 patients underwent SMA stenting; 65 patients (36%) in whom EPDs were used were included in the study. There were 49 female (75%) and 16 male patients with a median age of 73 6 11 years. The median Society for Vascular Surgery
4
Mendes et al
Journal of Vascular Surgery ---
2018
Fig 2. Multiple photographs demonstrating the presence of major macroscopic embolic debris in filter baskets of patients treated with superior mesenteric artery (SMA) stenting.
medical comorbidity score was 8 6 5 (range, 0-19). Clinical presentation consisted of CMI in 48 patients (74%) and AMI or acute-on-CMI in 17 (26%). The etiology of acute presentation was thrombotic in all patients. In four patients, a previous endovascular mesenteric revascularization had been performed. Lesion characteristics. Ostial SMA calcification was absent or minimal (<33% of ostial circumference) in 23 patients (35%), moderate (33%-66% of ostial circumference) in 9 patients (14%), and severe (>66% of ostial circumference) in 33 patients (51%). Median length of circumferential calcification was 22 6 17 mm. Occlusion was present in 28 patients (43%), with a median length of 21 6 13 mm. Hypodense, irregular thrombus was present in 43 patients (66%), of which 23 were nonocclusive (54%). Median thrombus length was 15 6 12 mm. The three findings of severe calcification, occlusion, and thrombus were present in 12 patients (18%). This indication for filter use was based on the presence of any one of these three anatomic factors or an acute presentation and occurred in 56 patients (86%). Treatment characteristics. The intervention was performed in an urgent or emergent setting in 14 patients (21%). Brachial access was used in 59 patients (91%), with femoral access used in 3; a hybrid approach (femoral and brachial) was needed in 2 patients, and radial access was used in 1 patient. Lesion predilation with an angioplasty balloon was performed in 33 patients (51%). A single balloon-expandable stent was used in 45 patients (69%), whereas 17 patients had two stents (26%) and three patients had three stents. Stent dimensions
were 5 to 7 mm (6.2 6 0.8 mm) 15 to 40 mm (27 6 10 mm). Postdilation was performed in 37 patients (57%). A bare-metal stent was used in 33 patients (51%) and a covered stent in 26 patients (40%). Six patients had combination of a proximal covered stent with a distal self-expandable bare-metal stent, which was needed to extend across side branches. Since 2014, covered stents were selected in 18 (82%) of 22 patients compared with 14 (33%) of 43 patients treated before 2014 (P < .0001). There was no difference in complications or distal embolization with the use of covered stents. Adjunct procedures were performed in 11 patients (17%) who had acute thrombus. These procedures consisted of intra-arterial thrombolysis in 5 patients, combination of thrombolysis and mechanical thrombectomy in 2 patients, atherectomy in 2 patients, and isolated thrombectomy or a combination of atherectomy and thrombectomy in 1 patient each. Technical success was 97%, with two patients demonstrating residual stenosis despite postdilation. All but one patient had embolic protection with the SpiderFX device; one patient (1.5%) had the Accunet Embolic Protection System early in our experience. The median EPD diameter was 6.0 6 0.8 mm (range, 4-7 mm). Indications for use of EPD were severe calcification in 22 patients (34%), acute thrombus in 18 (28%), and total occlusion in 16 (25%); in nine patients, the indication was based on the physician’s preference. There were no arterial injuries associated with filter deployment, but the EPD was not able to cross the main SMA lesion or did not provide appropriate support for therapy in one patient each. Two patients had vessel spasm successfully
Journal of Vascular Surgery Volume
-,
Number
Mendes et al
5
-
Fig 3. Computed tomography angiography of a patient treated with acute-on-chronic mesenteric ischemia (CMI) and a long proximal superior mesenteric artery (SMA) calcified lesion (A). Two covered stents were successfully deployed with good technical result (B), but angiography demonstrated new occlusion of an ileocolic branch (arrow; C). This was successfully treated using an Export aspiration catheter, with resolution on completion angiography (D). The embolic protection device (EPD) was retrieved with minimal debris; the aspirated material had a “white platelet thrombus” aspect (E).
treated with nitroglycerin. Macroscopic debris was identified in 43 patients and was considered major in 21 (49%; Fig 2). Major debris in the EPD basket was more common in patients with AMI compared with those with CMI (53% vs 25%; P ¼ .03); in patients with occlusions (50% vs 19%; P ¼ .01); and in patients with concomitant presence of severe calcification, occlusion, and thrombus (67% vs 25%; P ¼ .01). There was a trend toward increased presence of major debris in patients with long ($20 mm) calcified segments, but that did not reach statistical significance (44% vs 24%; P ¼ .08). Presence of major debris in the EPD was not associated with increased embolization rate, laparotomy, bowel resection, morbidity, or mortality. Outcomes. Distal embolization was identified on completion angiography in four patients (6%), two of whom presented with acute symptoms. In one patient, the intervention consisted of recanalization of a thrombotic SMA occlusion with intra-arterial thrombolysis, mechanical thrombectomy using the AngioJet device (Boston Scientific), and stenting with a covered stent. At completion of the procedure, a small side branch demonstrated to-and-fro flow and appeared to have a small embolus, which did not require additional treatment. The patient progressed well postoperatively without complications. The second patient with AMI had an EPD used because of severe calcification; two covered stents were used, and the EPD was found to be packed with atherosclerotic debris. After retrieval, embolic particles were noticed in a branch distal to the ileocolic branch. This was aspirated with an Export catheter (Medtronic) with complete resolution; the embolus had a “white platelet embolus” appearance and was possibly formed distal to the EPD (Fig 3). The patient underwent an exploratory laparotomy on postoperative day 3 because of lactic acidosis and persistent abdominal pain, but no segmental intestinal ischemia was found. The
SMA stent was found to be thrombosed and was recanalized using pharmacomechanical thrombectomy. The patient was ultimately discharged to a skilled nursing facility on postoperative day 13 and died of unrelated causes 2 months after the intervention. One patient with CMI treated by bare-metal stent and EPD for a heavily calcified, occluded SMA had distal embolization to an ileocolic branch that was successfully treated with an Export catheter, with no clinical significance. Finally, a second patient with CMI and distal embolization had an EPD used because of severe ostial calcification, for which atherectomy with the TurboHawk device (Covidien, Plymouth, Minn) was used as an adjunct to stenting. The EPD was noticed to be deployed within a side branch instead of the main SMA trunk; an embolus was identified in the main trunk of the SMA and was successfully treated by catheter aspiration using the Penumbra Indigo system (Penumbra Inc, Alameda, Calif) with resolution of embolus and no clinical consequences. A higher rate of embolization was seen in patients with calcification length >20 mm (15% vs 0; P ¼ .03); occlusion or thrombus >20 mm had similar rates of embolization. No other factors were associated with embolic events. Embolization itself did not result in higher morbidity, requirement for prolonged parenteral nutrition, bowel resection, laparotomy, or mortality. There were three additional intraprocedural complications (4.6%), one branch perforation, one dissection of a replaced hepatic artery, and one SMA dissection at the stented segment. None of these complications were associated with the EPD. The mean hospital stay was 4.3 6 7.2 days and was significantly higher for patients treated for AMI (9.3 6 7.7 days) in comparison to those with CMI (2.6 6 6.2 days; P ¼ .0007). The mean intensive care unit length of stay was 1.1 6 3 days, also significantly higher in
6
Mendes et al
Journal of Vascular Surgery ---
2018
Fig 4. Technique of using two filters for added embolic protection. Computed tomography angiography demonstrates thrombotic occlusion of a calcified superior mesenteric artery (SMA) ostial lesion (A; arrows). After successful crossing of the lesion, two embolic protection devices (EPDs) were deployed within the distal and middle segments of the main trunk of the SMA (B; arrows). The occluded SMA was predilated with a 4-mm angioplasty balloon and stented using a 7- 38-mm iCast covered stent, with good result demonstrated on completion angiography (C). Both filter baskets were successfully retrieved and had a moderate amount of macroscopic atherosclerotic debris (D).
patients with AMI (2.6 6 3.2 days vs 0.6 6 2.9 days; P ¼ .02). Early complications were identified in 21 patients (32%), 12 of whom had major complications (18%). Five patients (29%) with AMI required an exploratory laparotomy, and four (23%) had bowel resection. Additional early major complications (five CMI, three AMI) included cardiac complications, brachial hematoma, acute cholecystitis, and acute respiratory distress syndrome in two patients each. There were two early deaths, both in patients treated for AMI. One patient treated with a bare-metal stent and intra-arterial thrombolysis had clinical improvement and was discharged from the hospital on postoperative day 13, tolerating oral diet, but died on postoperative day 25 of decompensated congestive heart failure. The other patient was an 82-year-old woman treated with two covered stents who also had symptom improvement; on postoperative day 13, the patient was still hospitalized, developed generalized weakness and hypoxic respiratory failure, and ultimately died of acute respiratory distress syndrome. There was no evidence of residual mesenteric ischemia in this patient. The only factor associated with increased number of major complications was acute presentation (53% vs 25%; P ¼ .03). Laparotomy, bowel resection, and mortality were also significantly higher in the AMI group (24% vs 2%, 18% vs 2%, and 12% vs 0%, respectively; P < .02).
DISCUSSION Endovascular treatment has surpassed open surgical bypass as the primary modality for treatment of patients with acute or chronic symptoms of mesenteric ischemia.9 Recent developments in endovascular
technology have expanded the indications to patients with difficult occlusions, long calcified lesions, and fresh acute thrombus. Improvements in perioperative results, technical success, and patency rates have been achieved in the last decade, largely because of better selection of patients and advances in endovascular techniques and medical therapy.10-12 Data in the current study suggest that in the appropriate clinical scenario and when used by experienced operators, EPDs are a relatively safe addition to the inventory of endovascular mesenteric arterial interventions with encouraging outcomes. Distal embolization during mesenteric interventions has not been uniformly reported, and therefore there is a dearth of evidence on its actual incidence. We have used embolic protection selectively on the basis of a previous analysis of outcomes of SMA stenting for CMI without embolic protection. In that study, 8% of patients had major emboli identified on completion angiography. There were two deaths, both among patients with ischemic complications from emboli. Factors independently associated with emboli included recanalizations, severe calcification, lesion length >2 cm, and subacute symptoms. Therefore, we currently recommend use of EPDs selectively for these indications and for patients with AMI.13,14 That study suggested the selective use of EPDs to be appropriate in patients who present with acute thrombus or have one of the anatomic features described before. Subsequently, we reported on intraoperative complications during mesenteric artery stenting, including distal embolization in 6 of 156 patients (4%). Importantly, two of the four deaths in that study resulted from complications of bowel ischemia due to distal embolization, despite abdominal exploration and
Journal of Vascular Surgery Volume
-,
Number
Mendes et al
7
-
attempted revascularization.4 Other reports have shared a similar experience. In the Cleveland Clinic experience, three of five deaths (60%) resulted from distal embolization.11 Therefore, the potential risk and complications of distal embolization after mesenteric stenting should not be underestimated. The current series represents patients with lesions considered of higher risk for embolic events because of both anatomic characteristics and inclusion of acute or subacute patients, hence the use of embolic protection. Although the rate of embolization was still 6% in this series, we believe that the clinical significance and extent of the embolic event may be lessened by selective use of EPD. Despite the risk of embolization, use of EPDs has not been widely accepted in the mesenteric territory, with only a few reports published. Brown et al15 pioneered the use of such devices in a series of 14 patients treated with mesenteric stents, 13 with embolic protection using temporary balloon occlusion and aspiration in 12 patients (GuardWire; Medtronic) or a filter wire in 1 patient (FilterWire EX; Boston Scientific). In that series, none of the patients had distal embolization, and macroscopic debris was noted in the filter basket on several occasions. Similar to our study, deployment and retrieval of the EPD were safe in all patients without device-related complications. The fact that slightly more than one-third (36%) of the patients treated with SMA stenting in the period received an EPD demonstrates the selective approach in patients with high-risk lesions. Most patients (64%) were treated without an EPD and were not included in this series; although we did not perform a detailed anatomic analysis of these patients, the authors acknowledge that most had low-risk lesions and were not considered candidates for embolic protection. It is difficult to determine the clinical impact of embolization in our study because of the lack of a control group. However, the lack of major clinical events due to embolization (eg, bowel ischemia or gangrene) and the high frequency of potentially harmful macroscopic debris in the filter basket support the selective use of EPDs for the indications already discussed. We found some degree of debris in the basket in two-thirds of the patients, half of which were considered major debris (Fig 2). Notably, patients who present acutely often have fresh, soft thrombus that may be more prone to distal embolization. In some of these cases, thrombus burden required adjunctive thrombolysis, thrombectomy, and use of two filter devices for added protection (Fig 4). This technique should also be considered in patients with occlusions or long, calcified lesions. Finally, EPDs probably are unnecessary in most patients treated by mesenteric stents but should be considered in select patients who have these indications. Despite the use of embolic protection, 6% of our patients had distal embolization, which was manageable by endovascular techniques and did not result in clinical
sequelae. In two patients with acute thrombus, small emboli lodged in a distal ileocolic branch or in a small side branch. Another patient with a chronic lesion had a large embolus lodged in the main trunk of the SMA because the filter was deployed into a side branch, leaving the main artery unprotected. In all cases, the emboli were immediately recognized and successfully treated by catheter aspiration with no clinical sequelae. It is logical to assume that the burden of debris and its clinical significance were tempered by the use of embolic protection. Use of EPDs can be associated with complications, such as arterial dissection or a retained filter, and therefore some technical aspects deserve mention. We selected the SpiderFX filter because of availability over a long 320-cm guidewire and 7-mm diameter. Because added support is needed with difficult lesions, we prefer to use the brachial approach and recommend a “doublewire” technique, in which the filter wire is combined with an 0.018-inch wire of choice. The combination of two wires provides increased support and allows a 0.035-inch stent to be loaded into both wires, facilitating stent deployment and retrieval of the filter basket. After deployment and flaring of the stent, the EPD is retrieved with careful attention to avoid entrapment into the stent. In the event the filter basket is completely full, a second filter can be advanced over the 0.018-inch buddy wire and deployed distally beyond the first filter basket. In that case, emboli that escape from the first basket would be captured by the distal basket. Selection of the stent has also evolved during the last decade. Since our report that has shown superior patency for the covered stent, this has been used more frequently.10 These stents typically have larger profiles requiring a 7F sheath and are available only on a 0.035-inch configuration that directs the use of the double-wire technique, increasing the technical challenge of the procedure. This report has several limitations. The study is not randomized, and we do not have a comparative arm without embolic protection. Therefore, one cannot make final conclusions that indeed EPDs are beneficial. Nonetheless, our results in a higher risk group show no major clinical sequelae from embolization, despite evidence that large macroscopic debris was present in two of three patients treated by mesenteric stenting. Although this is the largest series of patients treated by mesenteric stents with EPDs, the small number of patients and the low rate of events lessen the power of its conclusions. This study focuses on intraoperative and perioperative outcomes; therefore, no long-term results were included in the analysis. It is fair to assume that embolic events in the mesenteric territory would have acute consequences if misdiagnosed, with no particular long-term interference. Finally, there was no costrelated analysis. The use of EPDs increases the cost of the procedure on the order of U.S. $800 to $1500 and
8
Journal of Vascular Surgery
Mendes et al
---
is definitely a factor to be considered for further recommendations; however, the authors believe that the cost is justified in select patients by decreasing the number of complications, including need for laparotomy, bowel resection, and prolonged hospital stay.
CONCLUSIONS This study shows in a select group of high-risk patients that mesenteric stenting with embolic protection is safe, and when it was used for these indications, there were no sequelae from major embolization or ischemic complications. Large macroscopic debris was noted in a significant proportion of patients when embolic protection was used selectively in patients with acute symptoms, occlusions, or severely calcified lesions. Despite embolic protection, distal embolization still occurred in 6% of patients, but this was successfully treated using catheter aspiration devices with no clinical consequences. The authors recommend that EPDs be used selectively in patients with mesenteric artery occlusions, long calcified lesions, and acute presentation.
5.
6.
7.
8.
9.
10.
AUTHOR CONTRIBUTIONS Conception and design: BM, GO Analysis and interpretation: BM, GO, MK, RD, FS, JJ Data collection: BM, GO, TT, KK Writing the article: BM, GO Critical revision of the article: BM, GO, TT, KK, MK, RD, FS, JJ Final approval of the article: BM, GO, TT, KK, MK, RD, FS, JJ Statistical analysis: BM, GO Obtained funding: Not applicable Overall responsibility: GO
11.
12.
13.
14.
REFERENCES 1. Ryer EJ, Kalra M, Oderich GS, Duncan AA, Gloviczki P, Cha S, et al. Revascularization for acute mesenteric ischemia. J Vasc Surg 2012;55:1682-9. 2. Arthurs ZM, Titus J, Bannazadeh M, Eagleton MJ, Srivastava S, Sarac TP, et al. A comparison of endovascular revascularization with traditional therapy for the treatment of acute mesenteric ischemia. J Vasc Surg 2011;53:698-704. 3. Block TA, Acosta S, Bjorck M. Endovascular and open surgery for acute occlusion of the superior mesenteric artery. J Vasc Surg 2010;52:959-66. 4. Oderich GS, Tallarita T, Gloviczki P, Duncan AA, Kalra M, Misra S, et al. Mesenteric artery complications during
15.
2018
angioplasty and stent placement for atherosclerotic chronic mesenteric ischemia. J Vasc Surg 2012;55:1063-71. Mendes BC, Oderich GS, Fleming MD, Misra S, Duncan AA, Kalra M, et al. Clinical significance of embolic events in patients undergoing endovascular femoropopliteal interventions with or without embolic protection devices. J Vasc Surg 2014;59:359-67.e1. Baim DS, Wahr D, George B, Leon MB, Greenberg J, Cutlip DE, et al. Randomized trial of a distal embolic protection device during percutaneous intervention of saphenous vein aorto-coronary bypass grafts. Circulation 2002;105:1285-90. Brott TG, Hobson RW, Howard G, Roubin GS, Clark WM, Brooks W, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363: 11-23. Oderich GS. Mesenteric vascular disease: chronic ischemia. In: Cronenwett JL, Johnston KW, editors. Rutherford’s vascular surgery. 8th ed. Philadelphia: Elsevier; 2014. p. 2. Schermerhorn ML, Giles KA, Hamdan AD, Wyers MC, Pomposelli FB. Mesenteric revascularization: management and outcomes in the United States, 1988-2006. J Vasc Surg 2009;50:341-8. Oderich GS, Erdoes LS, LeSar C, Mendes BC, Gloviczki P, Cha S, et al. Comparison of covered stents versus bare metal stents for treatment of chronic atherosclerotic mesenteric arterial disease. J Vasc Surg 2013;58:1316-23. Sarac TP, Altinet O, Kashyap V, Bena J, Lyden S, Sruvastava S, et al. Endovascular treatment of stenotic and occluded visceral arteries for chronic mesenteric ischemia. J Vasc Surg 2008;47:485-91. Bjorck M, Koelemay M, Acosta S, Goncalves FB, Kolbel T, Kolkman JJ, et al. Editor’s choicedmanagement of the diseases of mesenteric arteries and veins. Eur J Vasc Endovasc Surg 2017;53:460-510. Oderich GS, Bower TC, Sullivan TM, Bjarnason H, Cha S, Gloviczki P. Open versus endovascular revascularization for chronic mesenteric ischemia: risk-stratified outcomes. J Vasc Surg 2009;49:1472-9. Oderich GS, Macedo TA, Malgor RD, Ricotta JJ, Misra S, Duncan AA, et al. Anatomic measurements and factors associated with embolic events during superior mesenteric artery stenting: implications for use of embolic protection devices. Presented at the 37th Annual Symposium of the Society for Clinical Vascular Surgery; Ft. Lauderdale, Fla; March 18-21, 2009. Brown DJ, Schermerhorn ML, Powell RJ, Fillinger MF, Rzucidlo EM, Walsh DB, et al. Mesenteric stenting for chronic mesenteric ischemia. J Vasc Surg 2005;42:268-74.
Submitted Aug 21, 2017; accepted Dec 22, 2017.
Superior mesenteric artery stenting using embolic protection device for treatment of acute or chronic mesenteric ischemia Bernardo C. Mendes, MD, Gustavo S. Oderich, MD, Tiziano Tallarita, MD, Karina S. Kanamori, MD, Manju Kalra, MBBS, Randall R. DeMartino, MD, Fahad Shuja, MBBS, and Jill K. Johnstone, MD, Rochester, Minn
ABSTRACT Objective: The objective of the study was to report the feasibility and results of superior mesenteric artery (SMA) stenting using embolic protection devices (EPDs) to treat acute mesenteric ischemia (AMI) and chronic mesenteric ischemia (CMI). Methods: A retrospective review was conducted of consecutive patients who underwent SMA stenting with EPDs from 2007 to 2016. EPDs were used selectively in patients with occlusions, severe calcification, or acute thrombus. A two-wire technique with SpiderFX 0.014-inch filter wire (Medtronic, Minneapolis, Minn) combined with a 0.018-inch wire was used to provide support and to facilitate stenting and EPD retrieval. Presence of macroscopic debris in the EPD was recorded and graded as minor (minimal debris) or major (large thrombus or plaque). End points were technical success, presence of EPD debris, embolization, early morbidity, and mortality. Results: SMA stenting was performed in 179 patients, of whom 65 (36%) had EPDs. The mean age was 73 6 11 years, and 49 were female (75%). Clinical presentation was CMI in 48 patients (74%) and AMI or acute-on-CMI in 17 (26%). Indications for EPD were severe calcification in 22 patients (34%), acute thrombus in 18 (28%), and total occlusion in 16 (25%). Bare-metal stents were used in 33 patients, covered stents in 26, and both types in 6. Adjunctive therapy included thrombolysis in seven patients, thrombectomy in four, and atherectomy in three. Technical success was 100%. There were no instances of filter retention or arterial trauma due to filter manipulation. Distal embolization was noted in four patients (6%), of whom two had AMI. All large emboli were retrieved using catheter aspiration devices, but one small distal embolus was left untreated with no clinical consequences. Two patients had vessel spasm treated by nitroglycerin. Macroscopic debris was noted in 43 patients (66%) and was major in 21 (49%) or minor in 22 (51%). Of the patients with AMI, five (29%) required exploratory laparotomy and four (23%) had bowel resection. Eight additional patients (12%) had early complications (five CMI, three AMI), including cardiac complications, brachial hematoma, acute cholecystitis, and acute respiratory distress syndrome in two patients each. There were no deaths among CMI patients and two early deaths (12%) among those who had AMI. Conclusions: Use of EPDs during SMA stenting is safe and feasible with a two-wire technique. Large macroscopic debris was noted in one-third of the patients when the filter was applied selectively in patients with acute symptoms, occlusions, or severely calcified lesions. Despite the use of EPD, distal embolization occurred in 6% of patients and was successfully treated using catheter aspiration devices. (J Vasc Surg 2018;-:1-8.) Keywords: Chronic mesenteric ischemia; Endovascular treatment; Embolic protection
Endovascular therapy has become the main modality for treatment of atherosclerotic mesenteric arterial disease. Several large series have shown high rates of technical success, symptom improvement, and low morbidity and mortality in patients treated for chronic From the Division of Vascular and Endovascular Surgery, Mayo Clinic. Author conflict of interest: none. Presented at the 2016 Vascular Annual Meeting of the Society for Vascular Surgery, National Harbor, Md, June 8-11, 2016. Correspondence: Gustavo S. Oderich, MD, Gonda Vascular Center, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2018 Published by Elsevier Inc. on behalf of the Society for Vascular Surgery. https://doi.org/10.1016/j.jvs.2017.12.076
mesenteric ischemia (CMI). Patients with acute mesenteric ischemia (AMI) and acute-on-CMI have higher mortality rates and lower technical success, which is related to duration of ischemia, presence of advanced bowel gangrene, and significant thrombus burden within the mesenteric arteries.1-3 In these patients, the superior mesenteric artery (SMA) is generally affected by complex, heavily calcified plaque or associated thrombus, which often requires recanalization of chronic total occlusion or in situ acute-on-chronic thrombotic occlusions. These lesions have higher risk of thromboembolic complications and acute vessel or stent reocclusion, which can lead to bowel gangrene, multisystem organ failure, and death.4 Embolic protection devices (EPDs) have been designed for coronary grafts and carotid arteries but have been successfully used off-label in other vascular beds.5,6 There 1
2
Journal of Vascular Surgery
Mendes et al
---
is little if any controversy that EPDs should be indicated when the consequences of embolization cause irreversible end-organ damage, such as during carotid stenting, which has a known risk of embolic stroke.7 Mesenteric stenting can also cause embolization from catheter manipulations or dislodgment of plaque or thrombus, which may result in occlusion of distal branches, leading to vessel thrombosis and bowel gangrene.4 This is particularly important if the vessel occlusion is not immediately recognized and treated. Despite the relative lack of literature, we have used EPDs selectively in patients with heavily calcified stenosis, occlusions, and acute or subacute presentations. The objective of our study was to analyze the indications, techniques, and outcomes of patients who underwent primary stenting of the SMA using EPDs to treat occlusive mesenteric artery disease.
METHODS This study was approved by the Mayo Clinic Institutional Review Board. All patients consented with research participation. We performed a retrospective review of consecutive patients who underwent SMA stenting with EPDs from 2007 to 2016. The indication for use of an EPD was left to the discretion of the practitioner and was recorded when available on medical records. In general, indications were chronic occlusions, lesions longer than 3 cm with >66% circumferential or luminal calcification, and acute or subacute symptom presentation. EPDs were also used in patients without these anatomic criteria at the discretion of the treating physician. We excluded from the study patients who had mesenteric stenting without embolic protection. Demographics, cardiovascular risk factors, and clinical presentation were recorded. The primary symptom presentation was classified as chronic (>4 weeks), subacute (2-4 weeks), or acute (<2 weeks) on the basis of duration of symptoms. Lesion characteristics were analyzed using preoperative computed tomography angiography whenever possible. Calcified lesions were further analyzed as encompassing <33%, 33% to 66%, or >66% of the vessel circumference in the proximal 1 to 4 cm of the vessel. The length of circumferential calcified plaque, occlusion, and thrombus was also noted. Thrombus was recorded as present when irregular, hypodense filling defects were identified on computed tomography angiography, both occlusive and nonocclusive. Procedural details were noted, including approach (brachial or femoral), stent type, diameter and length, use of dilation before or after the intervention, EPD type and diameter, and presence of intraoperative complications. EPD-related complications were defined as side branch embolization, difficult filter retrieval, improper deployment, and target vessel spasm. Debris in the EPD basket was recorded and usually photographed. This was classified as absent, minor (diminutive or <20% of the filter basket), or major (large debris,
2018
ARTICLE HIGHLIGHTS d
d
d
Type of Research: Retrospective, single-center cohort study Take Home Message: In 65 of 179 patients presenting with mesenteric ischemia, embolic protection devices were used selectively during superior mesenteric artery stenting in cases of occlusion, severe calcification, or acute thrombus with a technical success rate of 100% and macroscopic debris identified in 66% of cases. Recommendation: This study suggests that embolic protection devices should be used during superior mesenteric artery stenting in cases of occlusions, severe calcification, and acute thrombus.
plaque, or >20% of filter basket). Early outcomes were recorded, including morbidity, mortality, length of intensive care unit and hospital stay, and early reinterventions. Technique. The technique of mesenteric artery stenting has been previously described by our group.8 The left brachial artery approach using surgical exposure with a small incision was used preferentially, but the femoral approach was indicated in patients with a heavily diseased aortic arch and favorable angle of origin in the SMA. A catheter support system was built using a combination of a 6F or 7F 90-cm hydrophilic sheath combined coaxially with a 7F guide catheter and a 5F multipurpose A (MPA) catheter. For complex lesions, such as chronic total occlusions or heavy calcified lesions, the 7F sheath, 7F MPA guide catheter, and 5F MPA catheter were used to engage the stump of the occluded vessel, allowing enough support for passage of a guidewire to the distal SMA (Fig 1). The target lesion was initially crossed using a 0.035-inch soft straight or angled Glidewire (Terumo Interventional Systems, Somerset, NJ), which was exchanged for the interventional wire of choice after confirmation of true lumen access. Our preference was to use a small-profile (0.014- or 0.018inch) stiff guidewire, such as the V-14 or V-18 wires (Boston Scientific, Marlborough, Mass), for most interventions. A covered stent was selected in most patients with proximal SMA lesions that did not involve side branches. Predilation is typically performed for very high grade stenosis and occlusions or in cases in which crossing of the lesion is otherwise difficult. A 320-cm SpiderFX 0.014inch filter wire (Medtronic, Minneapolis, Minn) was used in all except one patient, who had the Accunet Embolic Protection System (Abbott Vascular, Santa Clara, Calif). If a 0.035-inch stent is selected, a two-wire technique is used by combining the 0.014-inch filter wire with a 0.018inch “buddy wire”; the stent is introduced over both wires for better support and to facilitate subsequent retrieval of the EPD.
Journal of Vascular Surgery Volume
-,
Number
Mendes et al
3
-
Fig 1. In cases of superior mesenteric artery (SMA) occlusion, a stiff support system is built with the combination of a 7F sheath, a 7F multipurpose guide catheter, and a 5F multipurpose catheter, which is used to engage the stump of the occluded SMA (A). The lesion is crossed with a hydrophilic Glidewire, followed by a 0.014-inch filter wire and a 0.018-inch buddy wire, deployed through a 0.035-inch catheter (B). The lesion is predilated and subsequently treated with a balloon-expandable stent (C). The completion angiogram after filter retrieval demonstrates a flared, widely patent SMA stent with no distal embolization (D). (Image used with permission of Mayo Foundation for Medical Education and Research. All rights reserved.)
Statistical analysis. Primary end points were technical success (defined as successful stent placement with residual stenosis <30% and filter retrieval), presence of EPD debris, embolization, early (<30 days or in hospital) morbidity, and mortality. Differences were determined by the log-rank test. The Pearson c2 or Fisher exact test was used for analysis of categorical variables. Differences between means were tested with two-sided Student t-test, Wilcoxon rank sum test, or Mann-Whitney test. A P value of < .05 was used to
determine statistical significance. Data are presented as median values or mean 6 standard deviation, as appropriate.
RESULTS Patients. During the study period, 179 patients underwent SMA stenting; 65 patients (36%) in whom EPDs were used were included in the study. There were 49 female (75%) and 16 male patients with a median age of 73 6 11 years. The median Society for Vascular Surgery
4
Mendes et al
Journal of Vascular Surgery ---
2018
Fig 2. Multiple photographs demonstrating the presence of major macroscopic embolic debris in filter baskets of patients treated with superior mesenteric artery (SMA) stenting.
medical comorbidity score was 8 6 5 (range, 0-19). Clinical presentation consisted of CMI in 48 patients (74%) and AMI or acute-on-CMI in 17 (26%). The etiology of acute presentation was thrombotic in all patients. In four patients, a previous endovascular mesenteric revascularization had been performed. Lesion characteristics. Ostial SMA calcification was absent or minimal (<33% of ostial circumference) in 23 patients (35%), moderate (33%-66% of ostial circumference) in 9 patients (14%), and severe (>66% of ostial circumference) in 33 patients (51%). Median length of circumferential calcification was 22 6 17 mm. Occlusion was present in 28 patients (43%), with a median length of 21 6 13 mm. Hypodense, irregular thrombus was present in 43 patients (66%), of which 23 were nonocclusive (54%). Median thrombus length was 15 6 12 mm. The three findings of severe calcification, occlusion, and thrombus were present in 12 patients (18%). This indication for filter use was based on the presence of any one of these three anatomic factors or an acute presentation and occurred in 56 patients (86%). Treatment characteristics. The intervention was performed in an urgent or emergent setting in 14 patients (21%). Brachial access was used in 59 patients (91%), with femoral access used in 3; a hybrid approach (femoral and brachial) was needed in 2 patients, and radial access was used in 1 patient. Lesion predilation with an angioplasty balloon was performed in 33 patients (51%). A single balloon-expandable stent was used in 45 patients (69%), whereas 17 patients had two stents (26%) and three patients had three stents. Stent dimensions
were 5 to 7 mm (6.2 6 0.8 mm) 15 to 40 mm (27 6 10 mm). Postdilation was performed in 37 patients (57%). A bare-metal stent was used in 33 patients (51%) and a covered stent in 26 patients (40%). Six patients had combination of a proximal covered stent with a distal self-expandable bare-metal stent, which was needed to extend across side branches. Since 2014, covered stents were selected in 18 (82%) of 22 patients compared with 14 (33%) of 43 patients treated before 2014 (P < .0001). There was no difference in complications or distal embolization with the use of covered stents. Adjunct procedures were performed in 11 patients (17%) who had acute thrombus. These procedures consisted of intra-arterial thrombolysis in 5 patients, combination of thrombolysis and mechanical thrombectomy in 2 patients, atherectomy in 2 patients, and isolated thrombectomy or a combination of atherectomy and thrombectomy in 1 patient each. Technical success was 97%, with two patients demonstrating residual stenosis despite postdilation. All but one patient had embolic protection with the SpiderFX device; one patient (1.5%) had the Accunet Embolic Protection System early in our experience. The median EPD diameter was 6.0 6 0.8 mm (range, 4-7 mm). Indications for use of EPD were severe calcification in 22 patients (34%), acute thrombus in 18 (28%), and total occlusion in 16 (25%); in nine patients, the indication was based on the physician’s preference. There were no arterial injuries associated with filter deployment, but the EPD was not able to cross the main SMA lesion or did not provide appropriate support for therapy in one patient each. Two patients had vessel spasm successfully
Journal of Vascular Surgery Volume
-,
Number
Mendes et al
5
-
Fig 3. Computed tomography angiography of a patient treated with acute-on-chronic mesenteric ischemia (CMI) and a long proximal superior mesenteric artery (SMA) calcified lesion (A). Two covered stents were successfully deployed with good technical result (B), but angiography demonstrated new occlusion of an ileocolic branch (arrow; C). This was successfully treated using an Export aspiration catheter, with resolution on completion angiography (D). The embolic protection device (EPD) was retrieved with minimal debris; the aspirated material had a “white platelet thrombus” aspect (E).
treated with nitroglycerin. Macroscopic debris was identified in 43 patients and was considered major in 21 (49%; Fig 2). Major debris in the EPD basket was more common in patients with AMI compared with those with CMI (53% vs 25%; P ¼ .03); in patients with occlusions (50% vs 19%; P ¼ .01); and in patients with concomitant presence of severe calcification, occlusion, and thrombus (67% vs 25%; P ¼ .01). There was a trend toward increased presence of major debris in patients with long ($20 mm) calcified segments, but that did not reach statistical significance (44% vs 24%; P ¼ .08). Presence of major debris in the EPD was not associated with increased embolization rate, laparotomy, bowel resection, morbidity, or mortality. Outcomes. Distal embolization was identified on completion angiography in four patients (6%), two of whom presented with acute symptoms. In one patient, the intervention consisted of recanalization of a thrombotic SMA occlusion with intra-arterial thrombolysis, mechanical thrombectomy using the AngioJet device (Boston Scientific), and stenting with a covered stent. At completion of the procedure, a small side branch demonstrated to-and-fro flow and appeared to have a small embolus, which did not require additional treatment. The patient progressed well postoperatively without complications. The second patient with AMI had an EPD used because of severe calcification; two covered stents were used, and the EPD was found to be packed with atherosclerotic debris. After retrieval, embolic particles were noticed in a branch distal to the ileocolic branch. This was aspirated with an Export catheter (Medtronic) with complete resolution; the embolus had a “white platelet embolus” appearance and was possibly formed distal to the EPD (Fig 3). The patient underwent an exploratory laparotomy on postoperative day 3 because of lactic acidosis and persistent abdominal pain, but no segmental intestinal ischemia was found. The
SMA stent was found to be thrombosed and was recanalized using pharmacomechanical thrombectomy. The patient was ultimately discharged to a skilled nursing facility on postoperative day 13 and died of unrelated causes 2 months after the intervention. One patient with CMI treated by bare-metal stent and EPD for a heavily calcified, occluded SMA had distal embolization to an ileocolic branch that was successfully treated with an Export catheter, with no clinical significance. Finally, a second patient with CMI and distal embolization had an EPD used because of severe ostial calcification, for which atherectomy with the TurboHawk device (Covidien, Plymouth, Minn) was used as an adjunct to stenting. The EPD was noticed to be deployed within a side branch instead of the main SMA trunk; an embolus was identified in the main trunk of the SMA and was successfully treated by catheter aspiration using the Penumbra Indigo system (Penumbra Inc, Alameda, Calif) with resolution of embolus and no clinical consequences. A higher rate of embolization was seen in patients with calcification length >20 mm (15% vs 0; P ¼ .03); occlusion or thrombus >20 mm had similar rates of embolization. No other factors were associated with embolic events. Embolization itself did not result in higher morbidity, requirement for prolonged parenteral nutrition, bowel resection, laparotomy, or mortality. There were three additional intraprocedural complications (4.6%), one branch perforation, one dissection of a replaced hepatic artery, and one SMA dissection at the stented segment. None of these complications were associated with the EPD. The mean hospital stay was 4.3 6 7.2 days and was significantly higher for patients treated for AMI (9.3 6 7.7 days) in comparison to those with CMI (2.6 6 6.2 days; P ¼ .0007). The mean intensive care unit length of stay was 1.1 6 3 days, also significantly higher in
6
Mendes et al
Journal of Vascular Surgery ---
2018
Fig 4. Technique of using two filters for added embolic protection. Computed tomography angiography demonstrates thrombotic occlusion of a calcified superior mesenteric artery (SMA) ostial lesion (A; arrows). After successful crossing of the lesion, two embolic protection devices (EPDs) were deployed within the distal and middle segments of the main trunk of the SMA (B; arrows). The occluded SMA was predilated with a 4-mm angioplasty balloon and stented using a 7- 38-mm iCast covered stent, with good result demonstrated on completion angiography (C). Both filter baskets were successfully retrieved and had a moderate amount of macroscopic atherosclerotic debris (D).
patients with AMI (2.6 6 3.2 days vs 0.6 6 2.9 days; P ¼ .02). Early complications were identified in 21 patients (32%), 12 of whom had major complications (18%). Five patients (29%) with AMI required an exploratory laparotomy, and four (23%) had bowel resection. Additional early major complications (five CMI, three AMI) included cardiac complications, brachial hematoma, acute cholecystitis, and acute respiratory distress syndrome in two patients each. There were two early deaths, both in patients treated for AMI. One patient treated with a bare-metal stent and intra-arterial thrombolysis had clinical improvement and was discharged from the hospital on postoperative day 13, tolerating oral diet, but died on postoperative day 25 of decompensated congestive heart failure. The other patient was an 82-year-old woman treated with two covered stents who also had symptom improvement; on postoperative day 13, the patient was still hospitalized, developed generalized weakness and hypoxic respiratory failure, and ultimately died of acute respiratory distress syndrome. There was no evidence of residual mesenteric ischemia in this patient. The only factor associated with increased number of major complications was acute presentation (53% vs 25%; P ¼ .03). Laparotomy, bowel resection, and mortality were also significantly higher in the AMI group (24% vs 2%, 18% vs 2%, and 12% vs 0%, respectively; P < .02).
DISCUSSION Endovascular treatment has surpassed open surgical bypass as the primary modality for treatment of patients with acute or chronic symptoms of mesenteric ischemia.9 Recent developments in endovascular
technology have expanded the indications to patients with difficult occlusions, long calcified lesions, and fresh acute thrombus. Improvements in perioperative results, technical success, and patency rates have been achieved in the last decade, largely because of better selection of patients and advances in endovascular techniques and medical therapy.10-12 Data in the current study suggest that in the appropriate clinical scenario and when used by experienced operators, EPDs are a relatively safe addition to the inventory of endovascular mesenteric arterial interventions with encouraging outcomes. Distal embolization during mesenteric interventions has not been uniformly reported, and therefore there is a dearth of evidence on its actual incidence. We have used embolic protection selectively on the basis of a previous analysis of outcomes of SMA stenting for CMI without embolic protection. In that study, 8% of patients had major emboli identified on completion angiography. There were two deaths, both among patients with ischemic complications from emboli. Factors independently associated with emboli included recanalizations, severe calcification, lesion length >2 cm, and subacute symptoms. Therefore, we currently recommend use of EPDs selectively for these indications and for patients with AMI.13,14 That study suggested the selective use of EPDs to be appropriate in patients who present with acute thrombus or have one of the anatomic features described before. Subsequently, we reported on intraoperative complications during mesenteric artery stenting, including distal embolization in 6 of 156 patients (4%). Importantly, two of the four deaths in that study resulted from complications of bowel ischemia due to distal embolization, despite abdominal exploration and
Journal of Vascular Surgery Volume
-,
Number
Mendes et al
7
-
attempted revascularization.4 Other reports have shared a similar experience. In the Cleveland Clinic experience, three of five deaths (60%) resulted from distal embolization.11 Therefore, the potential risk and complications of distal embolization after mesenteric stenting should not be underestimated. The current series represents patients with lesions considered of higher risk for embolic events because of both anatomic characteristics and inclusion of acute or subacute patients, hence the use of embolic protection. Although the rate of embolization was still 6% in this series, we believe that the clinical significance and extent of the embolic event may be lessened by selective use of EPD. Despite the risk of embolization, use of EPDs has not been widely accepted in the mesenteric territory, with only a few reports published. Brown et al15 pioneered the use of such devices in a series of 14 patients treated with mesenteric stents, 13 with embolic protection using temporary balloon occlusion and aspiration in 12 patients (GuardWire; Medtronic) or a filter wire in 1 patient (FilterWire EX; Boston Scientific). In that series, none of the patients had distal embolization, and macroscopic debris was noted in the filter basket on several occasions. Similar to our study, deployment and retrieval of the EPD were safe in all patients without device-related complications. The fact that slightly more than one-third (36%) of the patients treated with SMA stenting in the period received an EPD demonstrates the selective approach in patients with high-risk lesions. Most patients (64%) were treated without an EPD and were not included in this series; although we did not perform a detailed anatomic analysis of these patients, the authors acknowledge that most had low-risk lesions and were not considered candidates for embolic protection. It is difficult to determine the clinical impact of embolization in our study because of the lack of a control group. However, the lack of major clinical events due to embolization (eg, bowel ischemia or gangrene) and the high frequency of potentially harmful macroscopic debris in the filter basket support the selective use of EPDs for the indications already discussed. We found some degree of debris in the basket in two-thirds of the patients, half of which were considered major debris (Fig 2). Notably, patients who present acutely often have fresh, soft thrombus that may be more prone to distal embolization. In some of these cases, thrombus burden required adjunctive thrombolysis, thrombectomy, and use of two filter devices for added protection (Fig 4). This technique should also be considered in patients with occlusions or long, calcified lesions. Finally, EPDs probably are unnecessary in most patients treated by mesenteric stents but should be considered in select patients who have these indications. Despite the use of embolic protection, 6% of our patients had distal embolization, which was manageable by endovascular techniques and did not result in clinical
sequelae. In two patients with acute thrombus, small emboli lodged in a distal ileocolic branch or in a small side branch. Another patient with a chronic lesion had a large embolus lodged in the main trunk of the SMA because the filter was deployed into a side branch, leaving the main artery unprotected. In all cases, the emboli were immediately recognized and successfully treated by catheter aspiration with no clinical sequelae. It is logical to assume that the burden of debris and its clinical significance were tempered by the use of embolic protection. Use of EPDs can be associated with complications, such as arterial dissection or a retained filter, and therefore some technical aspects deserve mention. We selected the SpiderFX filter because of availability over a long 320-cm guidewire and 7-mm diameter. Because added support is needed with difficult lesions, we prefer to use the brachial approach and recommend a “doublewire” technique, in which the filter wire is combined with an 0.018-inch wire of choice. The combination of two wires provides increased support and allows a 0.035-inch stent to be loaded into both wires, facilitating stent deployment and retrieval of the filter basket. After deployment and flaring of the stent, the EPD is retrieved with careful attention to avoid entrapment into the stent. In the event the filter basket is completely full, a second filter can be advanced over the 0.018-inch buddy wire and deployed distally beyond the first filter basket. In that case, emboli that escape from the first basket would be captured by the distal basket. Selection of the stent has also evolved during the last decade. Since our report that has shown superior patency for the covered stent, this has been used more frequently.10 These stents typically have larger profiles requiring a 7F sheath and are available only on a 0.035-inch configuration that directs the use of the double-wire technique, increasing the technical challenge of the procedure. This report has several limitations. The study is not randomized, and we do not have a comparative arm without embolic protection. Therefore, one cannot make final conclusions that indeed EPDs are beneficial. Nonetheless, our results in a higher risk group show no major clinical sequelae from embolization, despite evidence that large macroscopic debris was present in two of three patients treated by mesenteric stenting. Although this is the largest series of patients treated by mesenteric stents with EPDs, the small number of patients and the low rate of events lessen the power of its conclusions. This study focuses on intraoperative and perioperative outcomes; therefore, no long-term results were included in the analysis. It is fair to assume that embolic events in the mesenteric territory would have acute consequences if misdiagnosed, with no particular long-term interference. Finally, there was no costrelated analysis. The use of EPDs increases the cost of the procedure on the order of U.S. $800 to $1500 and
8
Journal of Vascular Surgery
Mendes et al
---
is definitely a factor to be considered for further recommendations; however, the authors believe that the cost is justified in select patients by decreasing the number of complications, including need for laparotomy, bowel resection, and prolonged hospital stay.
CONCLUSIONS This study shows in a select group of high-risk patients that mesenteric stenting with embolic protection is safe, and when it was used for these indications, there were no sequelae from major embolization or ischemic complications. Large macroscopic debris was noted in a significant proportion of patients when embolic protection was used selectively in patients with acute symptoms, occlusions, or severely calcified lesions. Despite embolic protection, distal embolization still occurred in 6% of patients, but this was successfully treated using catheter aspiration devices with no clinical consequences. The authors recommend that EPDs be used selectively in patients with mesenteric artery occlusions, long calcified lesions, and acute presentation.
5.
6.
7.
8.
9.
10.
AUTHOR CONTRIBUTIONS Conception and design: BM, GO Analysis and interpretation: BM, GO, MK, RD, FS, JJ Data collection: BM, GO, TT, KK Writing the article: BM, GO Critical revision of the article: BM, GO, TT, KK, MK, RD, FS, JJ Final approval of the article: BM, GO, TT, KK, MK, RD, FS, JJ Statistical analysis: BM, GO Obtained funding: Not applicable Overall responsibility: GO
11.
12.
13.
14.
REFERENCES 1. Ryer EJ, Kalra M, Oderich GS, Duncan AA, Gloviczki P, Cha S, et al. Revascularization for acute mesenteric ischemia. J Vasc Surg 2012;55:1682-9. 2. Arthurs ZM, Titus J, Bannazadeh M, Eagleton MJ, Srivastava S, Sarac TP, et al. A comparison of endovascular revascularization with traditional therapy for the treatment of acute mesenteric ischemia. J Vasc Surg 2011;53:698-704. 3. Block TA, Acosta S, Bjorck M. Endovascular and open surgery for acute occlusion of the superior mesenteric artery. J Vasc Surg 2010;52:959-66. 4. Oderich GS, Tallarita T, Gloviczki P, Duncan AA, Kalra M, Misra S, et al. Mesenteric artery complications during
15.
2018
angioplasty and stent placement for atherosclerotic chronic mesenteric ischemia. J Vasc Surg 2012;55:1063-71. Mendes BC, Oderich GS, Fleming MD, Misra S, Duncan AA, Kalra M, et al. Clinical significance of embolic events in patients undergoing endovascular femoropopliteal interventions with or without embolic protection devices. J Vasc Surg 2014;59:359-67.e1. Baim DS, Wahr D, George B, Leon MB, Greenberg J, Cutlip DE, et al. Randomized trial of a distal embolic protection device during percutaneous intervention of saphenous vein aorto-coronary bypass grafts. Circulation 2002;105:1285-90. Brott TG, Hobson RW, Howard G, Roubin GS, Clark WM, Brooks W, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363: 11-23. Oderich GS. Mesenteric vascular disease: chronic ischemia. In: Cronenwett JL, Johnston KW, editors. Rutherford’s vascular surgery. 8th ed. Philadelphia: Elsevier; 2014. p. 2. Schermerhorn ML, Giles KA, Hamdan AD, Wyers MC, Pomposelli FB. Mesenteric revascularization: management and outcomes in the United States, 1988-2006. J Vasc Surg 2009;50:341-8. Oderich GS, Erdoes LS, LeSar C, Mendes BC, Gloviczki P, Cha S, et al. Comparison of covered stents versus bare metal stents for treatment of chronic atherosclerotic mesenteric arterial disease. J Vasc Surg 2013;58:1316-23. Sarac TP, Altinet O, Kashyap V, Bena J, Lyden S, Sruvastava S, et al. Endovascular treatment of stenotic and occluded visceral arteries for chronic mesenteric ischemia. J Vasc Surg 2008;47:485-91. Bjorck M, Koelemay M, Acosta S, Goncalves FB, Kolbel T, Kolkman JJ, et al. Editor’s choicedmanagement of the diseases of mesenteric arteries and veins. Eur J Vasc Endovasc Surg 2017;53:460-510. Oderich GS, Bower TC, Sullivan TM, Bjarnason H, Cha S, Gloviczki P. Open versus endovascular revascularization for chronic mesenteric ischemia: risk-stratified outcomes. J Vasc Surg 2009;49:1472-9. Oderich GS, Macedo TA, Malgor RD, Ricotta JJ, Misra S, Duncan AA, et al. Anatomic measurements and factors associated with embolic events during superior mesenteric artery stenting: implications for use of embolic protection devices. Presented at the 37th Annual Symposium of the Society for Clinical Vascular Surgery; Ft. Lauderdale, Fla; March 18-21, 2009. Brown DJ, Schermerhorn ML, Powell RJ, Fillinger MF, Rzucidlo EM, Walsh DB, et al. Mesenteric stenting for chronic mesenteric ischemia. J Vasc Surg 2005;42:268-74.
Submitted Aug 21, 2017; accepted Dec 22, 2017.