Treating lower extremity malperfusion syndrome in acute type A aortic dissection with endovascular revascularization followed by delayed aortic repair

Objective To assess the outcomes of emergency revascularization with endovascular fenestration/stenting followed by delayed open aortic repair in patients with acute type A aortic dissection with lower extremity (LE) malperfusion syndrome (MPS); that is, necrosis and dysfunction of the lower extremity. Methods From 1996 to 2019, among 760 consecutive acute type A aortic dissection patients 512 patients had no malperfusion syndrome (Non-MPS), whereas 26 patients had LE-MPS with/without renal MPS and underwent endovascular fenestration/stenting, open aortic repair, or both. Patients with coronary, cerebral, mesenteric, and celiac MPS, or managed with thoracic endovascular aortic repair, were excluded (n = 222). All patients with LE-MPS underwent upfront endovascular fenestration/stenting except 1 patient (with signs of rupture) who initially underwent emergency open aortic repair. Results Among the LE-MPS patients, 17 (65%) had LE pain, 15 (58%) had abnormal motor function with 8 (31%) having paralysis, 10 (38%) had LE pallor, 17 (65%) had LE paresthesia, and 20 (77%) had LE pulselessness. Of the 25 patients undergoing upfront endovascular fenestration/stenting, 16 went on to open aortic repair, 3 survived to discharge without aortic repair, and 6 died before aortic repair (3-aortic rupture and 3-organ failure). In-hospital mortality among all patients was significantly higher in the LE-MPS group (31% vs 6.3%; P = .0003). Among those undergoing open aortic repair, postoperative outcomes were similar between groups, including operative mortality (18% vs 6.5%; P = .10). LE-MPS was a significant risk factor for in-hospital mortality (odds ratio, 6.0 [1.9, 19]; P = .002). Conclusions In acute type A aortic dissection, LE-MPS was associated with high in-hospital mortality. Emergency revascularization with endovascular fenestration/stenting followed by delayed open aortic repair may be a reasonable approach.

Conclusions: In acute type A aortic dissection, LE-MPS was associated with high inhospital mortality. Emergency revascularization with endovascular fenestration/ stenting followed by delayed open aortic repair may be a reasonable approach.

CENTRAL MESSAGE
Endovascular revascularization followed by delayed aortic repair had acceptable outcomes in ATAAD patients with lower extremity malperfusion syndrome (necrosis and dysfunction of the lower extremity).

PERSPECTIVE
In-hospital mortality was high amongst patients with lower extremity malperfusion syndrome (necrosis and dysfunction of the lower extremity) in acute type A aortic dissection. Emergency revascularization with endovascular fenestration/stenting followed by delayed open aortic repair had acceptable surgical outcomes and may be a reasonable approach for this disease.

See Commentary on page 111.
Video clip is available online.
Lower extremity (LE) malperfusion occurs in 15% to 40% of acute type A aortic dissection (ATAAD) cases. 1 Prolonged malperfusion can result in malperfusion syndrome (MPS). 1 MPS is a late-stage of malperfusion characterized by tissue necrosis and end-organ dysfunction due to dissection-related aortic branch vessel obstruction and insufficient blood flow to end organs. 2 Likewise, LE-MPS is defined as inadequate blood flow with resultant LE tissue necrosis and sensory and motor dysfunction. Patients with ATAAD with concomitant MPS have a high perioperative mortality between 29% and 89%. 3,4 The optimal surgical management for LE-MPS remains controversial. The conventional management of ATAAD with or without MPS is emergency open aortic repair. 5,6 At the University of Michigan, we have treated LE-MPS patients with initial endovascular revascularization followed by delayed open proximal aortic repair due to multiorgan failure, which could significantly increase operative mortality of upfront emergency open aortic repair. 2,7,8 In patients with MPS, the most critical life-threatening issue influencing outcomes is organ malperfusion, rather than aortic rupture. 2,7 This study aimed to assess the outcomes of emergency revascularization with endovascular fenestration/stenting followed by delayed open aortic repair in ATAAD patients with LE-MPS.

METHODS
This study was approved by the Institutional Review Board at Michigan Medicine (HUM 001118517), a waiver of informed consent was obtained, and it was in compliance with Health Insurance Portability and Accountability Act regulations.

Data Collection
Data from 1996 to 2019 was retrieved from the ATAAD registry at Michigan Medicine and supplemented with data from the Society of Thoracic Surgeons Michigan Medicine Cardiac Surgery Data Warehouse to identify the study cohort and determine pre-, intra-, and postoperative characteristics. These data were further supplemented with a retrospective medical record review. Information about survival was collected from the National Death Index Database through June 30, 2020. 9 LE-MPS was diagnosed by clinical symptoms, including pulselessness, pain, motor or sensory deficit of the lower extremity; abnormal lab values (ie, elevated lactate, creatine kinase [CK], CKMB, and myoglobin) indicating tissue ischemia and necrosis; and radiographic evidence (computed tomography angiogram) of dynamic or static obstruction of arterial flow to the lower extremities. All patients with LE-MPS were confirmed to have muscle tissue necrosis from malperfusion, including serology (eg, CK, CKMB, myoglobin, and lactate) and clinical exam. Hemodynamically stable patients with LE-MPS underwent upfront endovascular fenestration/ stenting before open aortic repair. Patients were then allowed to recover from lactic acidosis, shock, rhabdomyolysis, fasciotomy or amputation if needed, and acute respiratory distress syndrome (ARDS) before open proximal aortic repair.
Our technique for endovascular fenestration/stenting has been previously described. 3,10,11 This technique includes angiographic evaluation of the various vascular territories, including the LE and subsequent fenestration of the dissection flap with a 16-mm diameter balloon, aortic true lumen stenting with a 16-to 18-mm diameter self-expanding stent if the true lumen remains collapsed, and/or branch vessel fenestration/stenting if a gradient>15 mm Hg persists between the aortic root or ascending aorta and a branch vessel. 12 Details of endovascular intervention, including levels of aortic fenestration was detailed in Table E1.

Patient Selection
Between August 1996 and August 2019, a total of 760 patients presented with an ATAAD at our institution.

Statistical Analysis
Data are presented as median (25%, 75%) for continuous data and n (%) for categorical data. Univariate comparisons between the groups were performed using Wilcoxon rank-sum tests for continuous data and c 2 tests for categorical data. Logistic regression models were used to calculate the odds ratio (OR) of significant factors for in-hospital mortality adjusting age, sex, cardiogenic shock, acute renal failure, renal MPS, and LE-MPS. These variables were chosen based on their clinical relevance and our previous studies. 7,11 Due to small sample size, a Firth correction model was performed. The Kaplan-Meier method with log-rank testing was used to describe survival over time. Statistical calculations were performed using SAS version 9.4 (SAS Institute Inc).

Preoperative Demographic Data
Compared with the non-MPS group, the LE-MPS group had a significantly higher proportion of acute renal failure (42% vs 3.5%), renal malperfusion (31% vs 0%), and spinal cord malperfusion (7.7% vs 0%). The median time from admission to open aortic repair was longer in the LE-MPS group compared with the non-MPS group (1 vs 0 days; P <.0001). Otherwise, preoperative comorbidities were similar between LE-MPS and non-MPS groups ( Table 1).

Abbreviations and Acronyms
ARDS ¼ acute respiratory distress syndrome repair (3 aortic rupture and 3 organ failure) (see Table 2). Additionally, 3 patients survived to discharge without aortic repair for the following reasons: lack of patient's interest in proceeding with open aortic repair, previous aortic root and ascending aorta replacement, and poor surgical candidacy for open repair. Those 3 patients have all survived for more than 2 years after discharge. Maximum serum lactate level was significantly higher in patients who died due to aortic rupture or organ failure compared with patients who survived endovascular fenestration/stenting (6.0 mmol/L vs 2.0 mmol/L; P ¼ .02) (see Table 2).

Postprocedure/Operative Outcomes
Among all patients, the LE-MPS group had significantly higher in-hospital mortality after endovascular fenestration/ stenting or open aortic repair (31% vs 6.3%; P ¼ .0003), but other postintervention outcomes, including atrial fibrillation, new-onset renal failure, paraplegia, among others were similar between groups (Table 3).
Patients with LE-MPS who successfully underwent initial endovascular stenting/fenestration followed by an open aortic repair had significantly longer postoperative lengths of stay compared with non-MPS patients (14 vs 10 days; P ¼ .047). Otherwise, there were no significant differences in outcomes, including new-onset paraplegia, stroke, in-hospital mortality, and operative mortality, between groups after open aortic repair (Table 4). Among the LE-MPS group, 6 patients (23%) underwent an LE fasciotomy and 0 patients underwent LE amputation. Among all ATAAD patients (both LE-MPS and non-MPS), LE-MPS was a significant risk factor for in-hospital mortality (OR, 6.0; 95% CI 1.9-19; P ¼ .002) as was cardiogenic shock (OR, 5.2; 95% CI, 2.3-12.1; P ¼ .0001) ( Table 5).

DISCUSSION
In this study, the patients with LE-MPS had a significantly higher overall in-hospital mortality (31%) compared with patients without LE-MPS (6%). In patients with LE-MPS who were treated with emergency LE revascularization and recovered from MPS, postoperative outcomes and long-term survival were similar to the patients without LE-MPS (Figure 3, Video Abstract and Video 1).
There has been a confusion of malperfusion and MPS in the literature. We define malperfusion as compromised blood flow to end organs, the cause of MPS, and MPS is Initial open aorta repair + delayed endovascular fenestration/stenting (n = 1) the consequence of prolonged malperfusion; that is, the necrosis and dysfunction of the end organs from end organ malperfusion, 7,13 and in this specific study, the end-organ was the LEs. MPS frequently is complicated with multiorgan failure and metabolic acidosis. The difference between malperfusion and MPS is similar to the difference of "bacteremia and sepsis (septic syndrome) or "HIV and AIDS." 14 Malperfusion of the LE was not an indication for emergency endovascular fenestration/stenting and delayed open aortic repair, such as loss of femoral artery pulse but with normal function of the LE. However, MPS was an indication for emergency endovascular fenestration/stenting, such as loss of femoral pulse with LE motor or sensory deficit, elevated CK or serum lactate level, and radiographic evidence of dynamic or static obstruction of iliac or femoral arteries. All 26 patients with LE-MPS in our study had clinical evidence of LE malperfusion and subsequent necrosis and dysfunction of LE. For ATAAD patients with LE malperfusion, we all are in agreement that those patients should be treated with emergency open aortic repair. However, for ATAAD patients with LE-MPS (necrosis and dysfunction of LE), same as mesenteric MPS, the optimal management remains controversial. The conventional wisdom is still an emergency open aortic repair to resolve LE malperfusion and prevent aortic rupture. 5,6 At the University of Michigan, we have treated patients with LE-MPS with initial endovascular revascularization followed by delayed open proximal aortic repair due to extremely high operative mortality (80%-90%) of emergency open aortic repair in patients with preoperative MPS. 2,7,8 We believe that in those patients, expeditious open aortic repair can resolve only dynamic malperfusion of the LE but cannot resolve MPS (ie, necrosis of the LE that has already happened in patients) and its complications, such as organ failure and metabolic acidosis. Instead, upfront emergency aortic repair can worsen the LE-MPS due to the persistent static malperfusion to the LE during the open aortic repair and massive inflammatory reaction of the body to cardiopulmonary bypass and hypothermic circulatory arrest. A recent study from the Cleveland Clinic showed that 30% of patients with LE malperfusion need additional revascularization for ongoing extremity ischemia after open aortic repair for ATAAD. 15 The open aortic repair only resolved lower extremity malperfusion in 70% of the patients. Endovascular fenestration/stenting can resolve both static and dynamic LE malperfusion. 7,13 During endovascular fenestration/ stenting, we measured the blood pressure in the femoral artery and ascending aorta to confirm the LE malperfusion was resolved for every patient. Because of the necrosis of LE, those patients could quickly develop multiorgan failure after resolution of malperfusion due to ischemia/reperfusion injury, namely acute renal failure (42% in patients with LE-MPS in this study), ARDS, severe metabolic acidosis, and hyperkalemia that could result in arrhythmia and asystole. Therefore, we recommend delayed open   .0003 In the LE-MPS group, any complications after IR procedures or OR were recorded as outcomes. In the non-MPS group, any complications after OR were recorded as outcomes.
Values are presented as median (interquartile range) for continuous variables and number/total number (%) for categorical variables. P value<.05 is statistically significant. MI, Myocardial infarction; CVA, cerebrovascular accident; LOS, length of stay. endovascular procedure to correct the malperfusion is much more tolerable than an open aortic operation on cardiopulmonary bypass. Endovascular fenestration/stenting resolves the malperfusion with minimal surgical trauma to salvage the living/borderline tissue in the leg as much as possible.
After reperfusion, the limb could be preserved and the patient can recover. However, if a patient has an obviously dead leg due to prolonged malperfusion, amputation should be performed. Delayed open aortic repair in ATAAD patients after upfront endovascular revascularization could place patients at risk of aortic rupture. 7 In this study, the median time from admission to open aortic repair was 24 hours longer in the LE-MPS group compared with the non-MPS group. Six patients died before open aortic surgery. Three of them died from organ failure after all malperfusion was resolved with fenestration/stenting, which achieved similar results as open aortic repair but with much less trauma and influence on those patients. The other 3 patients died from aortic rupture that could have been prevented by open aortic repair (Table E2). The maximum serum lactate level was 6 mmol/L and CK level was>5000, indicating severe ischemia and necrosis of the lower extremity. Their operative mortality would be 33% to 89%. 3,15 In this whole cohort, 3out of 25 (12%) patients died from aortic rupture, which was still much lower than the operative mortality of emergency open aortic repair. Most of our aortic ruptures happened during the first decade. During the second decade, as we gained more experience of medically managing ATAAD patients, only 4% of patients had aortic rupture in all patients we managed with upfront fenestration/stenting. 7 The risk of dying from multiorgan failure in patients with MPS is 6 times higher than dying from aortic rupture. 7 Nevertheless, the aortic rupture was higher in this cohort of patients with LE-MPS compared with mesenteric malperfusion syndrome. 11 We should be more cautious for patients with isolated LE-MPS with or without renal malperfusion and repair the dissected proximal aorta in those patients whenever we think the patients can tolerate cardiopulmonary bypass and circulatory arrest without being on extracorporeal membrane oxygenation postoperatively.
Our in-hospital mortality of ATAAD patients with LE-MPS was comparable to the study from the Cleveland Clinic with an in-hospital mortality of 33% in ATAAD patients who underwent emergency open aortic repair and  the LE-MPS group, our findings indicated that emergency upfront endovascular revascularization improved postoperative outcomes, especially renal failure requiring hemodialysis and permanent strokes. Long-term survival of patients with LE-MPS following discharge from the hospital alive was similar to the patients without LE-MPS. Taken together, our strategy of treating patients with ATAAD with emergency endovascular revascularization by fenestration/stenting followed by delayed open aortic repair produced acceptable perioperative and long-term outcomes. This study has limitations as a single center, retrospective study. There was no control group of LE-MPS patients treated with immediate open aortic repair. The International Registry of Acute Aortic Dissection (IRAD) data were not used as control group due to the varying definitions of malperfusion and MPS and the inability to distinguish a uniform control group. Our study was designed as a descriptive study to report the outcomes of ATAAD patients with LE-MPS treated with endovascular fenestration/stenting followed by delayed open aortic repair. The sample size was small and could yield type II error.

CONCLUSIONS
Outcomes were favorable in stable ATAAD patients with LE-MPS treated with emergency revascularization via endovascular fenestration/stenting followed by delayed open aortic repair. Our strategy in this sick patient population may be a reasonable approach.