Introduction
One of the most feared complications of hypertensive emergency is acute aortic syndrome (AAS), a spectrum of conditions with the most common being acute aortic dissection. Other variants such as intramural hematomas, penetrating atherosclerotic ulcers, and periaortic hematomas have also been well-described in the literature [1]. The most significant risk factor for the development of an aortic dissection is chronic hypertension, and this poses a unique challenge in the management of hypertensive emergency in AAS [2]. Although relatively rare (estimated annual incidence of 2.9-5.9 per 100,000 [3]), aortic syndromes are associated with substantial mortality and morbidity with recent studies reporting in-hospital mortality estimates of 24% and 11% for type A and type B aortic dissections, respectively [4-6]. It is this high mortality rate and the danger of impending rupture or hypoperfusion injury that drives the need for aggressive management, both medical and surgical. Unsurprisingly, given the close association of AAS with uncontrolled hypertension, neurological complications are not infrequent consequences of the condition and its management. The goal of this article is to provide a comprehensive review of the known neurovascular complications of aortic syndromes with the available management guidelines.
Blood pressure management and hypoperfusion injuries
Cerebral autoregulation is the mechanism by which cerebral vasculature maintains blood flow during changes in cerebral perfusion pressure secondary to changes in blood pressure. The optimal blood pressure range for autoregulation lies between 60 mm Hg and 140 mm Hg, and the autoregulatory curve may be delayed during acute blood pressure changes, leading to serious risk of hyper- and hypo-perfusion insults [7]. Patients who develop global cerebral hypoperfusion, typically iatrogenically secondary to rapid lowering of blood pressure during the acute treatment of AAS, often develop a non-focal hypoactive delirium which can go unrecognized [8]. Figure 1 demonstrates a brain magnetic resonance imaging (MRI) of a patient who presented to the hospital with acute chest pain and was found to have an acute type B aortic dissection. During the management of their dissection, blood pressure was lowered, likely causing the watershed infarcts observed in the MRI of the brain. Unsurprisingly, patients presenting to the emergency department with higher mean arterial pressures (MAPs) and subsequently greater blood pressure drops have higher rates of hypoperfusion injuries [9]. The alarming aspect of higher mortality rates linked to lower minimum blood pressures in patients experiencing neurological symptoms of hypertensive emergency is that efforts to normalize pressure to safeguard the aorta are inadvertently leaving the brain vulnerable to further harm [10]. It is worth noting that these findings did not have a temporal contingency, but solely the minimum blood pressure was shown to have significance-further underlying the need for future studies.
A case series by Blanco et al. [11] showed that global cerebral hypoxic injury was the most common neurologic complication of acute type I aortic dissection. The authors found that it is more than three times as common as focal ischemic injury and likely represents the profound hemodynamic instability that occurs from injury to the great vessels. The risk of global malperfusion injury thus ought to be appreciated twice over when aggressive blood pressure measures are instituted.
The current recommendations for the management of AAS arise from the dogma of anti-impulse therapy. This refers to the attempt to minimize the change in pressure over time, thereby minimizing the risk of continued propagation of the luminal defect. As such, the guidelines suggest maintaining the heart rate to <60 beats per minute and the blood pressure <120/80 mm Hg [1,12]. Beta-blockade is the most effective initial medical management in the emergency setting with either esmolol or labetalol. Calcium channel blockers are a reasonable alternative, though there are no studies comparing their respective efficacies. Vasodilators can be used for additional blood pressure support but only after optimizing negative inotropy first [1].
Ischemic strokes in patients with aortic dissection present a unique management challenge. While general stroke management involves permissive hypertension [13] to maximize perfusion to ischemic areas, this is at odds with the strict blood pressure requirements associated with AAS management. While such management is a challenge, there may be certain medications better suited for such situations. Nitroprusside has been shown to increase intracranial pressure via cerebral vasodilation leading to an increase in cerebral blood flow [14,15]. Similarly, nitroglycerine was also observed to increase intracranial pressures in both animal and human studies through an increase in cerebral vessel capacitance [16,17]. Although these medications should be used cautiously in patients with altered mental status, those with chronic hypertension may benefit from their use as it may help maintain higher cerebral perfusion despite decreased systemic blood pressure.
Another possible intervention for cerebroprotection in AAS is the placement of the patient in a Trendelenburg position. Pathophysiologically, this maneuver may allow for increasing cerebral perfusion while still practicing anti-impulse therapy. Though only a case report, Takamoto et al. [18] found better neurologic outcomes during aortic arch replacement; however, additional research is needed.
The use of transcranial Doppler (TCD) during the titration of blood pressure in the management of aortic syndromes may have potential benefit by monitoring cerebral blood flow in real-time as TCD can detect changes in flow that may indicate hypoperfusion, cerebral edema, or other complications of hypertensive emergency. A study by Ziegler et al. [19], which was done in patients with severe traumatic brain injury, found that TCD may be useful in neuro-prognostication as normal measurements were overall correlated with survival.
Although there is practice-wide consensus on the risks of rapid reduction in blood pressure during a hypertensive crisis, there is a remarkable void in terms of clinical studies that investigated the true outcomes of different blood pressure lowering regimens in patients with AAS. Current literature discusses the theoretical risks through the pathophysiologic mechanisms that can induce cerebral hypoperfusion and disrupt the well-studied cerebral autoregulation system, but randomized control trials that analyze different blood pressure regimens and/or target blood pressure goals are lacking [7].
Thromboembolic strokes and potential intervention
The percentage of patients with AAS and concomitant stroke is estimated to be around 9%-16%, with similar rates found for postoperative repair strokes [20]. Patients with aortic dissections often present with neurologic symptoms that resolve due to transient arterial occlusion [20]. Furthermore, the risk of stroke is unsurprisingly increased with proximity of the dissection to the aortic arch vessels [21]. If a patient presents with simultaneous chest pain and/or acute upper back pain and stroke, without significant electrocardiogram ischemia or acute injury, one should immediately evaluate for AAS. Strokes, when presenting concomitantly with aortic dissections, are most often hemispheric and are right sided in up to 71% of the time [22]. Supra-aortic dissections account for nearly two-thirds of ischemic strokes in aortic dissections with thromboembolism or hypotension accounting for the other third [20]. This contrasts with the perioperative stroke distribution, which is not dominated by supra-aortic dissection or a predominance for one hemisphere over the other [20].
The use of thrombolytic therapy (alteplase, tenecteplase) is largely believed to be contraindicated in patients with AAS. Nevertheless, aortic dissection is not an official contraindication, and the stroke-evaluation pathway does not call for evaluation of a dissection, although some literature suggests it should [23]. Most research regarding the use of thrombolytics in AAS comes from case reports (Table 1) in patients where the dissection was not discovered or was masquerading as myocardial infarction. In the former scenario, overwhelming evidence of high mortality (approximately 70%) rivals mortality rates from aortic dissections in the pre-surgical era [24]. In the latter scenario, complications include cardiac tamponade and cardiac arrest, with the Anglo-Scandinavian Study of Early Thrombosis (ASSET) trial showing particularly high mortality rates in the thrombolytic group [25,26]. While the dogma of “time equals brain” dictates the exact order of imaging studies needed for evaluation of stroke, aortic dissections, if known or incidentally discovered (on computed tomography angiography [CTA] of the neck), should constitute a contraindication for the administration of thrombolytic therapy given the current data.
It is, therefore, of vital importance for neurologists and emergency providers evaluating patients for acute stroke to be attuned to and aware of common aortic dissection symptoms and presentations. Pain, particularly of a stabbing or tearing quality and localizing to the chest, back, or abdomen, should always raise concern [27]. Ischemic peripheral neurologic pain is another red flag that should prompt evaluation for dissection as acute strokes rarely present with pain [25,28]. Nevertheless, a substantial percentage (10%-55%) may not experience any pain, and therefore, other symptoms such as differences in pulse strength or blood pressure between limbs or signs of shock or hypotension must not be overlooked [29].
With thrombolytic therapy being a poor treatment option, mechanical thrombectomy (MT) is often the only option in patients with stroke secondary to large vessel occlusion in the setting of AAS. Although the evidence for such interventions is limited, aortic dissection does not constitute a contraindication to MT. Kehara et al. [30] postulate that neurovascular intervention is a reasonable option for postoperative large territory infarct with either thrombectomy or stenting, with hemorrhagic transformation being the most major adverse event. While there have been case reports of successful MT and endovascular recanalization in cases of ischemic stroke due to acute type A aortic dissection (ATAAD) [31], the risk of further vascular dissection and dislodgement of thrombotic material remains a concern for centers that perform these catheter-based approaches [32]. Nevertheless, accumulating evidence in endovascular neuro-intervention suggests that, despite the risks, MT offers an acceptable safety profile with improved outcomes [33,34]. Arterial access also presents a major hurdle in these situations as femoral access is largely impossible with a dissection and carotid access may involve a complicated closure [30,35]. Transbrachial approach may be reasonable in emergency situations though the sparsity of available data means it should be evaluated on a case-by-case manner [36].
Spinal cord infarction
While spinal cord ischemia and infarction are rare complications of AAS, they can have devastating consequences. Symptoms may include complete loss of motor function below the area of the lesion and near complete loss of sensory function at and below the level of the lesion, with only proprioception and vibratory sense retained [36]. During acute blood pressure control of a patient with AAS, a patient may develop these signs when the blood pressure falls below a certain level and resolve if the blood pressure is allowed to relax to a higher level, which may be higher than the recommended treatment goal of systolic blood pressure (SBP) <120 mm Hg. Quick recognition is paramount and comanagement of these patients often requires consultation with neurology and/or neurosurgery as a lumbar cerebrospinal fluid (CSF) drain may be indicated. It is estimated that aortic pathology accounts for nearly 16% of all anterior spinal cord infarcts (ASCI) [37] and around 1% of patients with ATAAD will have ASCI [21]. Interestingly, the overwhelming majority of cases are iatrogenic in the setting of operative repair [38] with DeBakey type I and type III aortic dissections being the highest risk for spinal cord injury. Rates of ASCI were roughly 1 in 130 (0.74) in repair of aortic dissection or ruptured aortic aneurysm and 1 in 600 (0.16%) in those undergoing repair of unruptured aortic aneurysm [39].
Elshony et al. [40] conducted a systemic literature review of 67 cases of aortic dissection with ASCI, in which only 48% of patients had an initial presentation of typical chest pain radiating to the back. This correlates with a study by Gaul et al. [20] demonstrating typical chest pain in only two-thirds of cases of aortic dissection with spinal cord injury. They also found the following neurological distribution of symptoms: anterior cord syndrome (46%), pure motor symptoms (39%), complete cord syndrome (7%), pure sensory (3%), Cauda equina syndrome (3%), and Brown-Séquard syndrome (1.5%). Notably, 27% of all cases had bowel/bladder dysfunction. Notably, initial severity of neurological symptoms and lack of early symptom improvement were poor predictors of outcome. Radiographic findings did not strongly correlate with outcomes [40].
While physical examination and history are crucial to identify possible spinal cord ischemia, MRI with gadolinium remains the gold standard for neuroimaging, although computed tomography (CT) imaging may be used in hemodynamically unstable patients or those who cannot undergo MRI. While spinal cord swelling and pencil-like hyperintensities on T2 weighted imaging may be observed, it is now standard to evaluate all possible cases of ASCI with diffusion weighted imaging, as conventional MRI may miss ASCI in the acute setting [41-45]. Figure 2 shows a spine MRI of a patient found to have acute ischemia in their spinal cord at the level of T12-L1 after undergoing thoracic endovascular aortic repair and multiple vascular stents. Postoperatively, the patient exhibited lower extremity weakness and urinary retention, prompting an MRI of the lumbar spine to be performed.
Management of ASCI in the setting of AAS can be quite challenging as low blood pressure is the inciting factor. Regarding AAS, a blood pressure goal of 120/80 mm Hg and heart rate less than 60 beats per minute are usually targeted, while in ASCI, avoiding systemic hypotension is essential, with maintenance of at least an MAP of 85-90 mm Hg [46]. While there are no studies demonstrating the optimal blood pressure goal to maintain in AAS with spinal cord ischemia, both extreme hypotension and hypertension should be avoided. Erbel [47] looked at 4,167 patients with ATAAD and 2,071 patients with type B dissection and outcomes based on presenting SBP. In both the type A and type B aortic dissection groups, rates of spinal cord ischemia were higher in the SBP <80 or >150 mm Hg groups compared to 81-100 and 101-150 mm Hg. Given the lack of evidence for optimal blood pressure goals in aortic dissection complicated by ASCI, definitive management requires additional research. In the setting of worsening neurological function, liberation of the standard blood pressure goal of 120/80 mm Hg may be reasonable.
Surgical management remains the gold standard for treatment of AAS with spinal cord ischemia in those with good preoperative risk stratification. While specific surgical techniques are discussed elsewhere, the use of a lumbar CSF drain may help prevent further spinal cord ischemia in those undergoing open or endoscopic thoracic, thoracoabdominal aortic aneurysm, and thoracic endovascular aortic repair surgery [48,49]. While optimal intracranial pressure goals for treatment of ASCI are unknown, empiric recommendations of 10-15 mm Hg appear to be a reasonable target to optimize spinal cord perfusion during thoracic aortic surgery [50].
Nonetheless, early diagnosis and prompt treatment of AAS complicated by ASCI is crucial to prevent worsening neurological deficits, death, and increase the chance of meaningful neurological recovery. Still, mortality rates remain high for AAS with rates for type A aortic dissection near 50% after 14 days and type B aortic dissection near 10% around 30 days [1].
Cervicocephalic insufficiency/dissection extension
Common carotid artery (CCA) pathology is estimated to occur in up to 30% of ATAAD and is associated with a significantly increased risk for both preoperative neurological symptoms and irreversible cerebral malperfusion leading to permanent neurological deficits [51]. Additionally, brain malperfusion is an independent risk factor for early mortality in ATAAD [52,53]. The sequelae of neurological complications stem from compression of the true lumen of the CCA by either the false lumen or intussusception by the dissection flap itself resulting in malperfusion and may be accompanied by thromboembolic phenomena off the false lumen [54]. Patients with ATAAD CCA involvement on presentation had higher rates of stroke (18.6% vs. 8.1%) and lower in-hospital survival (83.7% vs 93.4%) [55]. Preoperative supraortic branch involvement was also an independent predictor of both postoperative risk of cerebral malperfusion and early mortality [48].
Presenting symptoms of ATAAD with CCA involvement are generally right hemisphere predominant, owing to increased hydraulic stress in the right lateral wall of the aorta during expansion of the hematoma [56]. Nearly 19% of patients with ATAAD present with syncope and while the etiology may be due to cerebral malperfusion associated with CCA pathology, cardiac tamponade is also a cause [57]. Additionally, 3%-6% of patients with ATAAD presented with neurological complications in the form of stroke or coma as their predominant symptom [58,59].
There is no gold standard for diagnosing carotid dissection secondary to ATAAD, and often both CT/MRI and ultrasound are used together to assess anatomic defects and flow characteristics. A systemic review from 2009 found the sensitivity and specificity for cervicocephalic dissection to be similar for both magnetic resonance angiography and CTA [60]. Notably, carotid duplex detects abnormalities in the range of 68%-95% of cases and therefore should not be the first line if suspicion for CCA dissection is high [61,62].
While the numerous approaches for surgical management of CCA dissection in the setting of ATAAD are discussed elsewhere, there is no surgical consensus and repair strategies are usually made at the discretion of the surgeon on a case-by-case basis [63,64]. In regard to immediate medical management, all patients with transient ischemic attack (TIA)/stroke symptoms should be evaluated with standard vascular head and neck stroke imaging to evaluate potential candidacy for reperfusion therapy, as discussed previously. While there is no literature that discusses blood pressure goals in cases of carotid dissection due to ATAAD, systemic hypotension should be avoided as it is a predictor for poor outcomes in cases of both carotid dissection and ATAAD [47,65,66].
It is well established that in the first six months of non-aortic arch CCA dissection, recurrence of ischemic neurological events is high without medical therapy [67], and patient outcomes were vastly improved on either antiplatelet or anticoagulation as shown in the Cervical Artery Dissection in Stroke Study (CADISS) trial [68]. Charlton-Ouw et al. [55] analyzed 43 cases of CCA dissection from ATAAD over a 10-year period and concluded that most patients should be trialed on antithrombotic therapy over surgical intervention after arch repair. Notably, they found that degree of stenosis or false lumen extension was not predictive of stroke on presentation or rate of additional ischemic events after arch repair. Given these findings, the authors recommended anti-thrombotic therapy over surgical repair for most cases of CCA dissection from ATAAD. Given the sparsity of evidence, the use of anti-thrombotic therapy should be evaluated on a case-by-case basis until further research is performed.
Unlike in the anterior neck vessels, extension of an AAS into the vertebral arteries is uncommon. A study by Witsch et al. [69] found that patients with vertebral artery and carotid artery dissections have an increased future risk of aortic dissection. In cases of vertebral artery dissection and AAS, the question of anticoagulation becomes more complex. The guidelines for management of acute aortic dissections are missing recommendations for the use and timing of anticoagulation. Several case reports exist to suggest that anticoagulation is safe in both types of aortic dissection and even carries benefit for the residual false lumen [70,71]. Intramural hematomas and peri-aortic hematomas, on the other hand, are likely contraindications to anticoagulation and experts believe resolution of the hematoma is prudent prior to starting anticoagulation [72,73]. In those cases, it would be reasonable to pursue an endovascular repair to prevent the risk of dissection propagation and stroke.
Surgical repair
Surgical management plays a crucial role in the treatment of AAS, aiming to prevent catastrophic complications such as aortic rupture or organ malperfusion. Despite advancements in medical therapy, open surgical repair remains the gold standard for AAS management, particularly in cases where endovascular repair is not feasible or contraindicated. The primary goals of surgery are to resect the diseased aortic segment, restore aortic continuity, preserve vital organ perfusion, and reduce the risk of aortic rupture and tamponade [52].
While open surgical repair is important to improve outcomes in AAS, accessing the aorta directly can lead to neurologic complications, including stroke. The incidence of stroke following open surgical repair of AAS varies but is reported to be around 5% to 30% [74].
In repair of type A dissections, a literature review of the Society of Thoracic Surgeons Adult Cardiac Surgery Database evaluated the effects of cannulation strategy, temperature, cerebral protection techniques, and repair techniques on cerebral perfusion as aortic arch repair requires hypothermic circulatory arrest which puts the brain at highest risk of insult during this portion. Ghoreishi et al. [74] found that 13% of patients who had repair of type A aortic dissections had postoperative stroke. Upon adjustment, factors such as femoral cannulation site (odds ratio, 0.60; P<0.001) and total arch replacement (odds ratio, 1.30; P=0.013) were identified as being associated with the incidence of postoperative stroke whereas retrograde cerebral profusion was found to have reduced risk of postoperative stroke (odds ratio, 0.75; P=0.008).
Intraoperatively, one technique, known as antegrade cerebral perfusion (ACP), allows continuous perfusion of the brain while the ascending aorta is clamped, reducing the risk of cerebral ischemia which often involves cannulating the innominate artery, right axillary artery, or the subclavian artery [75]. Retrograde cerebral perfusion is achieved by perfusing the brain via the femoral artery maintaining cerebral perfusion during aortic arch repair [75]. There is mixed evidence on which cerebral perfusion technique is superior. Further studies are needed to compare the long-term neurologic outcomes of different cerebral perfusion strategies and to identify the optimal strategy for individual patients based on their specific clinical characteristics.
For aortic arch repair, cerebral oximetry, electroencephalography (EEG), and bispectral index can be used to monitor cerebral perfusion intraoperatively. In a study of 71 cases by Keenan et al. [76], intraoperative EEG was used during hypothermic circulatory arrest with ACP and showed that 45% of patients had abrupt loss of electrocerebral activity suggestive of cerebral ischemia. The EEG activity was then restored with initiation of unilateral ACP in all but one patient who required bilateral ACP. This finding highlights the importance of careful monitoring and management of cerebral perfusion during aortic surgeries to prevent neurological complications.
Another tool to be used in the intraoperative setting is TCD, which, as mentioned earlier, could have applications for neuroprognostication and the medical management of aortic dissection patients. In a study by Ghazy et al. [77], nine patients underwent aortic arch surgery with TCD for monitoring and optimizing cerebral perfusion. The study identified inadequate cross-filling via TCD monitoring and successfully optimized cerebral perfusion intraoperatively. One patient with extensive air bubbles experienced a postoperative stroke. This research offers valuable insights into using TCD for monitoring cerebral perfusion during aortic arch surgery, confirming the adequacy of cerebral perfusion strategies and highlighting the need for optimization.
In repair of type B dissections, neurologic complications are also a significant concern. The Study of Thoracic Aortic Type B Dissection Using Endoluminal Repair trial (STABLE) is a prospective, nonrandomized, multicenter study that evaluated the safety and performance of endovascular treatment for patients with complicated type B aortic dissections. In the study, the incidence of neurologic complications ranged from 2.5% to 7.5% and included stroke, TIA, paraplegia, and paraparesis, which all occurred within 30 days of repair [78]. For the patients who experienced stroke, imaging was performed which showed a distribution of infarcts that were likely related to procedural or technique-related factors [78].
Regarding spinal cord ischemia during aortic dissection repair, there is limited evidence to suggest specific risk factors; however, anatomical variation of cord vasculature is likely a driving factor. The great radicular artery of Adamkiewicz (GRA) is the largest radicular artery and typically arises from the left posterior intercostal artery somewhere between the 9th and 12th intercostal arteries. Anatomical variations of the GRA can make predicting spinal cord ischemia challenging, especially in the perioperative setting where cross-clamping the aorta at the same spinal cord level may have drastically different neurological outcomes for patients. Other commonly obstructed spinal perfusing arteries in AAS include intercostal arteries, lumbar arteries, and thoracic radicular arteries [79].
Outcomes of patients undergoing surgical or endovascular repair of AAS complicated by ASCI are mainly limited to case reports. However, the literature review by Elshony et al. [40] found that out of the 19 patients who underwent open surgical repair (12 type A aortic dissection, 4 type B aortic dissection, 3 unknown classification), 13/19 (68%) had good recovery, 4/19 (21%) had residual neurological deficits, and 2/19 died (11%). Additionally, they found 6 patients who had lumbar CSF drains placed only (type B aortic dissection), 5/6 (83%) had instant recovery of neurologic function and the remaining 1 patient had residual deficits due to a delay of lumbar CSF drain placement. Another 25/67 total cases were treated conservatively with medical management, with only 12/15 (48%) of those patients able to walk again [40].
In summary, AAS and its medical and surgical management can lead to multiple neurologic complications impacting cerebral and spinal perfusion. Quick recognition is crucial so alterations in blood pressure goals and/or interventions can be instituted rapidly to offset the complication rate. This highlights the need for further research to optimize monitoring and improve neurologic outcomes. Tools such as EEG and TCD show promise in assessing and optimizing cerebral perfusion intraoperatively and future studies should focus on comparing the effectiveness of these tools in developing strategies to minimize risk of neurologic complications during AAS repair.