Surgical and endovascular management of acute type A aortic dissection

INTRODUCTION — Acute type A aortic dissection is a surgical emergency, with most operations performed immediately following recognition.

The goals of surgery for acute type A aortic dissection are to prevent rupture and to treat any, or a combination of, acute aortic insufficiency; cardiac tamponade, antegrade propagation into the aortic arch resulting in stroke; retrograde dissection into the aortic root leading to myocardial infarction; or distal malperfusion syndromes leading to visceral or peripheral ischemia.

Surgical management of acute type A aortic dissection has evolved to be more aggressive, using hypothermic circulatory arrest routinely and with expanded indications for aortic arch replacement. Historically, repair involved a clamped ascending aortic replacement with circulatory arrest performed only if the intimal tear extended distally into the aortic arch.

Surgical techniques used to manage acute type A aortic dissection are reviewed. The clinical features, diagnosis, initial evaluation, medical management, and follow-up after repair are reviewed separately. (See "Clinical features and diagnosis of acute aortic dissection" and "Management of acute type A aortic dissection".)

Surgical techniques used to manage acute or chronic type B aortic dissection are reviewed separately. (See "Surgical and endovascular management of acute type B aortic dissection" and "Management of chronic type B aortic dissection", section on 'Dissection repair techniques'.)

ANESTHESIA AND MONITORING — The importance of preoperative coordination between the surgical team, anesthesiologist, perfusion specialist, and nursing staff cannot be overemphasized. Typical monitoring includes, at a minimum, an arterial catheter, a temperature-sensing bladder catheter, central venous line, and intraoperative transesophageal echocardiography. Additional monitoring, which depends in part on the perfusion strategy, may include a second arterial catheter (often an upper extremity line combined with femoral arterial monitoring) and pulmonary artery catheter. (See "Anesthesia for aortic surgery with hypothermia and elective circulatory arrest in adult patients", section on 'Monitoring'.)

Cerebral oximetry is useful for monitoring the cerebral metabolic state during periods of circulatory arrest. Cerebral electroencephalogram is also sometimes used to monitor brain activity prior to and after periods of circulatory arrest. (See "Anesthesia for aortic surgery with hypothermia and elective circulatory arrest in adult patients", section on 'Brain monitors'.)

HYPOTHERMIC CIRCULATORY ARREST — The immediate priorities while commencing repair of acute type A dissection are to establish cardiopulmonary bypass with preferential true lumen flow and perform systemic hypothermia.

Hypothermic circulatory arrest techniques share their origins with the inception of cardiac surgery in the 1950s [1]. These techniques depend on the principle of decreased tissue metabolic demands as body temperature is lowered. Animal studies as well as contemporary clinical studies suggest that circulatory arrest times of up to 45 minutes are safely tolerated during conditions of deep hypothermia (<20 degrees Celsius) [2-4]. The optimal temperature for hypothermic circulatory arrest is debated. The safe duration of hypothermic circulatory arrest may be prolonged with the use of adjunctive cerebral perfusion techniques (antegrade or retrograde). (See 'Antegrade cerebral perfusion' below and 'Retrograde cerebral perfusion' below.)

Cannulation options — During repair of type A aortic dissection, it is essential for the cardiac surgeon to execute a cardiopulmonary bypass cannulation strategy that provides adequate true-lumen perfusion to critical organ systems.

The most common sites for arterial cannulation include the ascending aorta ("central cannulation"), right axillary artery, and the femoral artery. A contemporary review of the Society of Thoracic Surgeons database revealed a fairly equal distribution of cannulation sites for repair of acute type A dissection (direct aortic 19 percent, axillary 27 percent, femoral 36 percent) [5]. Outcomes including procedure-related stroke, need for renal replacement therapy, in-hospital mortality, and long-term survival were similar regardless of approach in one single institution review [6]. Regardless of the specific artery chosen, the cannula must reside within the true lumen and preferentially perfuse the true lumen of the aorta. Peripheral arteries with dissection are typically avoided for cannulation.

The cannulation strategy must also take into account the distal extent of operation (eg, hemiarch versus total arch) and the need for circulatory arrest and concomitant antegrade cerebral perfusion (ACP) and/or retrograde cerebral perfusion (RCP). (See 'Proximal extent of repair' below and 'Cerebral protection' below.)

Ascending aorta — Direct cannulation of the ascending aorta offers a convenient and efficient method that does not require an additional incision or exposure beyond the median sternotomy. It also provides antegrade perfusion to the entire body.

Direct aortic cannulation typically involves a modified Seldinger technique that uses guidance from preoperative computed tomography or intraoperative transesophageal echocardiography to ensure true-lumen cannulation. It can be difficult to place the arterial cannula into the true lumen in certain scenarios (eg, posterior and effaced true lumen, complete circumferential dissection), and experience with this technique is critical to success in the setting of an acute type A dissection. An anatomic schema has been devised for evaluating true-lumen anatomy and potential difficulty of direct aortic cannulation in type A dissection [7]. Three anatomic variants in the ascending aorta in order of increasing difficulty are:

●Anterior true lumen

●Posterior true lumen

●"Free-floating" true lumen from circumferential dissection

A common concern with direct aortic cannulation in this setting is risk of aortic rupture; however, with increasing experience with this technique, there is evidence that this is a rare event [6].

Several retrospective studies have compared aortic cannulation with other cannulation sites in acute type A aortic dissection. One study involving 235 patients reported no significant differences between direct aortic versus femoral cannulation for the rate of stroke (11.7 versus 7.2 percent, respectively) or 30-day mortality (20.2 versus 16.9 percent, respectively) [8]. Similar findings were reported in another review (stroke: 4.9 versus 4.5 percent, respectively; early mortality: 14 versus 23 percent, respectively) [6,9].

Axillary artery — The right axillary artery offers several advantages as a cannulation site for acute type A aortic dissection. The right axillary artery is typically devoid of severe atherosclerotic disease and is often spared from the dissection process.

Axillary artery cannulation facilitates convenient ACP during periods of circulatory arrest. However, because the right arm will be hyperperfused, it is critical to monitor bilateral arterial pressures (ie, right radial artery line and another functioning left arterial line [radial or femoral]).

Drawbacks to using the axillary artery for cannulation include the increased time to expose the axillary artery and prepare for cannulation, which typically requires sewing on an 8 mm sidearm graft. Furthermore, the axillary artery is intimately invested in the brachial plexus (figure 1), and therefore meticulous dissection must be performed to avoid nerve injury. Vascular injury is also a possibility. A common concern with right axillary artery cannulation is whether it is safe to use if the innominate artery is involved with the dissection. Preferential false lumen flow may occur if the dissection flap partially or completely occludes the orifice of the innominate artery. A review comparing right axillary cannulation and nonaxillary cannulation in this circumstance reported no difference in stroke rate (8.3 versus 8.8 percent, respectively) or in-hospital mortality (9.5 versus 10.8 percent, respectively) [10]. Reported stroke rates for axillary cannulation during acute type A dissection repairs have varied ranging from 1.8 to 14 percent [10-13].

Femoral artery — Femoral artery cannulation offers an efficient alternative as an arterial cannulation site. All cardiac surgeons are facile with femoral exposure and preparation of this artery for cardiopulmonary bypass.

The obvious drawback to femoral artery cannulation is that it provides retrograde rather than antegrade arterial perfusion during cardiopulmonary bypass. Significant atheromatous disease of the descending thoracic aorta or abdominal aorta represents a relative contraindication to femoral artery cannulation due to an increased risk of embolic showering and stroke. Furthermore, complicated dissection flaps with multiple reentry points throughout the aorta may result in inconsistent true-lumen perfusion leading to visceral or cerebral malperfusion. Nevertheless, concerns regarding retrograde perfusion are probably overstated in this population, likely related to low rates of significant atherosclerotic disease.

In appropriately selected patients (ie, younger patients without atherosclerotic disease), femoral cannulation appears safe. In a review of 79 patients with acute type A dissection who received femoral cannulation, the rates of malperfusion (2.5 percent) and stroke (8.1 percent) were deemed acceptable [11]. In a retrospective review comparing femoral (198 patients) and axillary (107 patients) artery cannulation for acute type A dissection patients, early mortality (19 versus 16 percent, respectively) and stroke rates (17 versus 14, respectively) were similar [12].

Others — Several less common arterial cannulation strategies deserve mention.

A ventricular transapical cannulation strategy has been described for type A dissection [14]. This strategy ensures true lumen perfusion; however, a drawback of this technique is the inability to cross-clamp the aorta prior to circulatory arrest.

Other sites include the innominate or subclavian arteries. Lastly, in cases where no suitable cannulation site can be identified, the aorta can be transected and cannulated directly, the so-called "Samurai" technique [15]. This allows for prompt hypothermia and circulatory arrest to reconstruct the aorta.

Cerebral protection — Adjuncts to hypothermic circulatory arrest, namely ACP and/or RCP, are commonplace. Adjunctive cerebral perfusion is applicable and probably carries greater importance since many patients with acute type A aortic dissection present with neurologic injury/stroke or subclinical neurologic malperfusion from aortic arch and/or cerebral vessel dissection. As such, hypothermic circulatory arrest without a cerebral perfusion strategy for aortic arch surgery has largely been abandoned in many aortic surgery centers.

The use of cerebral perfusion techniques during hypothermic circulatory arrest offer several advantages, including extension of the "safe" duration of circulatory arrest and uniform and continued cooling of the brain, and they help to flush the cerebral circulation.

The optimal route of cerebral perfusion (antegrade, retrograde) for hypothermic circulatory arrest during aortic arch surgery is debated.

Antegrade cerebral perfusion — ACP delivers cold, oxygenated blood to the cerebral circulation to assist in cooling as well as providing nutritive blood flow.

When an axillary artery cannulation strategy is used, ACP is easily accomplished (see 'Axillary artery' above). It is important to monitor bilateral cerebral O₂ saturation during circulatory arrest techniques, since clamping both the innominate and left common carotid artery is often necessary to provide adequate antegrade flow when perfusing from the right axillary artery (figure 2). In the case of an incomplete circle of Willis, unilateral ACP from the right axillary artery will provide incomplete left-sided cerebral protection; adding a left common carotid artery ACP catheter can be considered. If an alternative cannulation strategy is used (eg, femoral, aortic), balloon-tip catheters can be inserted directly into the innominate (unilateral) and left common carotid artery (bilateral). Some cardiac surgeons advocate perfusing the left subclavian artery as well; this is especially pertinent for extensive aortic arch repairs and when a dominant left vertebral artery is present. Direct cerebral vessel cannulation provides a simple and effective means of ACP; however, drawbacks include the possibility of vessel injury from dissection and the inconvenience of a cluttered surgical field during the open distal anastomosis. Representative studies illustrating various outcomes for ACP include the following:

●A retrospective review compared outcomes in 203 patients with acute type A dissection who underwent arch replacement [16]. No significant differences were demonstrated for 30-day mortality for bilateral ACP versus unilateral ACP (11.6 versus 20.7 percent) or permanent neurologic disability (8.4 versus 16.9 percent).

●In a review of a 342 patients undergoing unilateral selective ACP with moderate hypothermia (median 25.9 Celsius) for emergency type A dissection repair, in-hospital mortality was 11.7 percent, and a stroke rate of 7.3 percent was observed [17]. No difference was reported in the outcomes of hemiarch versus transverse arch repairs.

●In a study of 453 type A dissection patients using unilateral or bilateral ACP with mild hypothermia (28 Celsius), stroke occurred in 6 percent of patients, and the 30-day mortality rate was 7 percent [18]. Most patients received a hemiarch ascending replacement, but nearly 25 percent had a transverse arch repair.

Retrograde cerebral perfusion — RCP is a rapid and simple technique to perform that involves cerebral perfusion of cooled, oxygenated blood typically from the superior vena cava (figure 3).

Debate exists whether RCP techniques provide nutritive blood flow to the brain; however, dark, deoxygenated blood is often observed draining from the cerebral vessels during periods of circulatory arrest with RCP, providing indirect evidence that oxygen extraction is occurring. Because of concern over lack of nutritive flow, a commonly held belief is that RCP techniques should be limited to shorter circulatory arrest times. In a study of prolonged (over 50 minutes) deep hypothermic circulatory arrest using RCP for aortic arch aneurysm repair, the stroke rate was 2 percent and mortality was 8 percent, which was similar to other reports from shorter-duration procedures [19]. In a multi-institutional study that compared RCP with ACP for acute type A dissection, the observed difference in rate of early mortality (14 versus 19 percent) or stroke (21.8 versus 14.3 percent) was not statistically significant [12].

RCP also has the (theoretical) advantage of flushing embolic debris from the cerebral circulation. Support of this notion is provided by a study that compared RCP with ACP in patients undergoing hemiarch repair for aneurysmal disease, which reported significantly lower levels of postoperative brain magnetic resonance imaging lesions in the RCP compared with ACP group (45 versus 100 percent) [20].

SURGICAL REPAIR

General conduct — Once cardiopulmonary bypass and systemic hypothermia are achieved, the left ventricle is typically vented and a decision is made regarding cross-clamping the aorta prior to circulatory arrest (figure 4). The ascending aorta can generally be cross-clamped safely; however, a large component of intramural hematoma or significant ascending aortic atheromatous disease might preclude safe cross-clamping, in which case the patient is cooled to circulatory arrest, the pump is turned off, and then the aorta is resected and reconstructed.

When clamping is selected, the ascending aorta is typically clamped early during the cooling process, and cardioplegia is administered. Great care is taken to adequately protect the right ventricle, and topical cooling and intermittent doses of retrograde and/or antegrade cardioplegia are given. In general, retrograde cardioplegia is used with supplemental direct ostial antegrade cardioplegia, which is safe and effective, except under extraordinary circumstances of extensive aortic root and coronary destruction. Measurement of myocardial septal temperature is sometimes used and kept below 14°C.

The proximal aorta is then typically prepared while cooling occurs. The aortic valve is inspected for any abnormalities, and the integrity of the coronary arteries is assessed. The proximal extent of the operation is determined (ie, supracoronary ascending replacement versus aortic root replacement), as well as need for aortic valve replacement (see 'Proximal extent of repair' below). At minimum, an aortic valve resuspension is performed. The proximal extent of the dissection is also identified, and any intimal tears extending into the aortic sinus segment are noted. Felt "neo-media" can be fashioned at this point in time if a supracoronary ascending replacement is chosen [21].

Once systemic cooling has occurred for at least 45 minutes, circulatory arrest is typically commenced, and the distal anastomosis is usually performed first in an "open" fashion (hemiarch, partial arch, or total arch replacement) (see 'Distal extent of repair' below). Felt "neo-media" is often used, especially in the setting of a hemiarch replacement. Great care is taken to reinforce any "problem areas" of the distal anastomosis with pledgeted sutures while under circulatory arrest. Many cardiac surgeons prefer to reinforce the posterior aspect of the suture line, as this can be extremely difficult to repair following cross-clamp removal. After completion of the distal extent, vigorous de-airing maneuvers are performed in preparation for reestablishment of cardiopulmonary bypass. The patient is then systemically rewarmed, and typically the proximal extent of operation is completed. (See 'Proximal extent of repair' below.)

Immediate concerns after weaning from cardiopulmonary bypass following acute type A dissection repair include assessment of operative result and hemorrhage. Any Dacron graft kinks should be addressed, and the status of aortic valve competency should be assessed on transesophageal echocardiography. Significant hemorrhage from coagulopathy is common following type A dissection repair. Adjuncts to traditional resuscitation with blood product administration include prothrombin complex concentrate and recombinant factor VII administration, which are safe and effective following cardiac surgery and may result in fewer blood product transfusions when compared with traditional fresh frozen plasma-based resuscitation [22-25]. The safety and efficacy of factor VII use has specifically been reported in thoracic aortic surgery [26]. Additional study is needed to determine if routine use of these adjuncts, along with administration of autologous blood and other adjunctive therapies, is warranted during acute type A dissection. (See "Blood management and anticoagulation for cardiopulmonary bypass".)

Proximal extent of repair — The proximal extent of operation (ie, supracoronary ascending replacement versus aortic root replacement) is determined after inspecting the aortic valve for any abnormalities and assessing the integrity of the coronary arteries. Most type A dissections are repaired with a supracoronary anastomosis near the level of the sinotubular junction. Data from the Society of Thoracic Surgeons (STS) database show that about 25 percent of repairs involve a root replacement, and only 2.5 percent were of the valve-sparing variety (Yakoub or David repair) (figure 5) [5]. (See "Overview of open surgical repair of the thoracic aorta", section on 'Involvement of the aortic root'.)

There are a variety of reasons that aortic root replacement has been avoided, such as the possibility of aortic valve replacement (if a valve-sparing approach not used), concerns regarding coronary button integrity, and increased operative time/complexity. However, root replacement is safe during acute type A dissection repair. In a matched cohort study, early mortality rates were similar for root replacement compared with a supracoronary proximal anastomosis, and freedom from proximal reintervention was significantly higher in the root replacement cohort at seven years (96 versus 80 percent) [27].

In general, aortic root replacement is typically limited to the following scenarios during repair of type A aortic dissection:

●Intimal tear extending into the sinus segment/destruction of the root complex

●Aortic root aneurysm (>4.5 cm)

●Marfan syndrome or other connective tissue disorder

●A young patient (<50 years old) with bicuspid aortic valve

Most patients who present with acute type A aortic dissection have a completely normal aortic valve. Therefore, if root replacement is indicated, strong consideration should be given for a valve-sparing aortic root replacement. However, a very specific skillset and level of experience is needed to successfully perform valve-sparing aortic root replacement in the setting of type A dissection. Surgeons should have an established comfort level with valve-sparing aortic root replacement on an "elective basis" prior to undertaking this highly technical operation during acute type A dissection repair. Despite excellent outcomes with valve-sparing root replacement during acute type A dissection repair, several words of caution must be emphasized. In the acute dissection patient, hematoma near the pulmonary artery can render the infra-annular dissection quite hazardous. Furthermore, great care is needed during coronary button reconstruction if the dissection extends near the coronary ostia. Theoretically, all type A dissection patients could be considered for a valve-sparing approach, as the dissection never extends below the aortic annulus; however, it is possible that the intimal tear can involve the aortic leaflet commissures. Extreme caution should be taken when contemplating a valve-sparing approach in this scenario, as reconstruction of the valve can be hazardous. If the surgeon is not sufficiently experienced or undertaking valve-sparing aortic root replacement is deemed to be potential hazardous, a Bentall procedure should be performed.

Several series have shown excellent outcomes with valve-sparing root replacement in acute type A dissection patients.

●In a review comparing valved conduits (Bentall procedure) with valve-sparing aortic root replacement, there were no differences in early outcomes or need for aortic valve replacement at three years follow-up [28].

●In a report of a 16-year experience of valve-sparing root replacement in acute type A aortic dissection, operative mortality was very low (3 percent), and freedom from aortic insufficiency at 10 years was 95 percent [29].

●Finally, in a review of 109 patients receiving a valve-sparing root replacement during type A repair, in-hospital mortality was 11 percent, and valve-related reintervention rate at eight years was 13 percent [30].

Distal extent of repair — Most cardiac surgeons elect to open the aortic arch to examine for an entry tear and "stabilize" the inner curvature. In review of the STS database, nearly 80 percent of type A dissection repairs involved circulatory arrest and open distal aortic anastomosis [5]. Some debate exists regarding the need for transverse arch replacement at the time of initial repair, though transverse arch replacement has been shown to be safe in these patients [31,32]. In a report of arch replacement when intimal tears were present in the arch, in-hospital mortality was 15.4 percent and the rate of stroke was 20.5 percent, which did not differ significantly from those who received a hemiarch repair [31].

Several groups have advocated for routine antegrade thoracic endovascular aortic repair (TEVAR; ie, frozen elephant trunk) at the time of repair of DeBakey 1 type A dissection (figure 6), with published mortality rates of 4.2 to 6.6 percent and stroke rates of 0.9 to 4.2 percent [33,34]. A theoretical advantage of this approach is promotion of positive aortic remodeling and decreased rates of arch and descending thoracic aorta aneurysmal degeneration. This approach comes at the cost of an increased risk of spinal cord ischemia with thoracic aortic coverage and also increases the operative complexity. All told, it remains an area of active debate whether routine antegrade TEVAR at the time of type A dissection repair is useful or overly aggressive and unnecessary. At a minimum, strong consideration should be made for transverse arch replacement in the setting of intimal tear within the arch as well as preexisting aortic arch aneurysm.

Managing malperfusion — Malperfusion syndrome is defined as regional ischemia within an organ system. Acute type A aortic dissection associated with malperfusion is an especially devastating presentation of this disease process, complicating 11 to 34 percent of acute type A dissections [35-38].

Possible sites of malperfusion include the coronary, cerebral, spinal, visceral, and extremity vascular beds, presenting as electrocardiogram changes, stroke, paresis, abdominal pain, renal insufficiency, or a pulse deficit, respectively. Mortality of type A aortic dissection associated with malperfusion is high (about 30 percent) and increases dramatically with multiple sites of malperfusion (eg, three systems, 43.4 percent 30-day mortality [35]) [37].

End-organ ischemia can occur by several mechanisms during acute dissection. Dynamic obstruction (figure 7) occurs when a branch of the aorta is occluded by the dissection flap leading to malperfusion. This is in contrast to static obstruction (figure 8), where the aortic dissection extends into the branch vessel and leads to ischemia by thrombus formation or occlusion of the true lumen [39].

Specific vascular beds

Cerebral — Cerebral malperfusion (stroke or coma) complicates around 4 to 6 percent of acute type A dissection [40,41]. Patients presenting with acute type A dissection complicated by stroke represent a significant challenge to the cardiac surgeon. Despite being associated with a significantly increased risk for perioperative death, many of these patients can be salvaged with emergency repair, including those who present with coma. Expedient repair and minimization of "door-to-brain time" should be offered in many of these patients (even in the face of delays up to 24 hours) who seem otherwise salvageable based on comorbidities, age, or preoperative status. It is unclear if an optimal perfusion strategy or distal extent of repair (hemiarch versus transverse arch) can be recommended for these patients. Such decisions many times are influenced by the clinical presentation and acuity.

Involvement of the carotid artery in type A dissection is a risk factor for cerebral malperfusion. In a review of 1444 patients with acute type A dissection, 279 (19 percent) had unilateral common carotid artery involvement, and 161 (11 percent) had bilateral common carotid artery dissection [42]. A preoperative neurologic deficit was significantly more common in patients with common carotid artery dissection (23 versus 3 percent). Interestingly, patients with common carotid artery dissection were significantly more likely to present with concurrent distal dissection in the descending thoracic (86 versus 61 percent) and iliac artery (58 versus 30 percent) and also had higher rates of corresponding noncerebral organ malperfusion (38 versus 22 percent). This argues that involvement of the common carotid artery in the dissection is a marker for a more "aggressive" acute type A dissection. Careful discussion with the family or other caretakers is warranted before intervention is undertaken in patients with cerebral malperfusion or definitive stroke. (See "Management of acute type A aortic dissection", section on 'Decision for comfort care'.)

Small observational studies have specifically examined outcomes following surgical therapy in this patient population.

●One group retrospectively reviewed 16 patients who underwent type A dissection repair that was complicated by preoperative neurologic injury. Hospital mortality was 43.7 percent for the cerebral malperfusion group, compared with 17.0 percent for the noncerebral malperfusion group. Multivariate analysis determined that preoperative neurologic injury was the sole risk factor for hospital mortality in the study [43].

●In a review of the International Registry for Acute Aortic Dissections (IRAD) data of 87 (4.7 percent) patients with acute type A complicated by stroke, as well as 54 (2.9 percent) with coma, pharmacologic therapy only (anti-impulse therapy as palliative management) was offered in 24.1 percent of stroke patients and 33.3 percent of coma patients [40]. Of those who underwent emergency repair, in-hospital mortality rates were 27.0 percent for the stroke cohort and 44.4 percent for coma.

The timing of aortic repair in acute type A dissection complicated by neurologic compromise has been examined. A theoretical concern exists for worsened outcomes if surgery is delayed in these patients (eg, due to transfer from referring facilities, delays in diagnosis). In one review of 41 patients with acute type A dissection and stroke, the time to surgery was significantly associated with lack of neurologic improvement on logistic regression (odds ratio [OR] 14.6, 95% CI 2.7-8.5) [44]. In a retrospective review of 27 patients with acute type A dissection presenting with coma, 21 patients received immediate operation within five hours of onset of symptoms [45]. In-hospital mortality was 14 percent in the immediate group versus 67 percent in the delayed group (mean 34 hours from time of onset of symptoms). However, in a retrospective review of 50 patients with acute type A dissection complicated by stroke, the time to operation was not a good predictor of neurologic outcome or death [46]. The strongest predictor for mortality in this study was coma at presentation (OR 14.12, 95% CI 3.58-55.73). Of note, most patients in this study were classified as receiving an operation between 6 to 24 hours of symptom onset (64 percent of patients in the neurologic injury group). The authors concluded that time from symptom onset to definitive repair should not be used as a contraindication for proceeding with expedient repair.

Coronary — An occasional patient with acute type A aortic dissection will present with right coronary artery dissection and resultant myocardial ischemia. This typically presents with characteristic electrocardiogram changes (inferior ST and T wave changes), regional wall motion abnormalities, arrhythmias, and/or cardiogenic shock.

Prompt revascularization and myocardial protection are paramount in these patients. Typically, a segment of great saphenous vein will be harvested, and a right coronary artery bypass will be performed at the onset of cardiopulmonary bypass. Thereafter, cardioplegia can be given down the right coronary artery, and attention is then turned to ascending aortic repair. Often, the right coronary artery will later be ligated at the ostium in this situation.

Visceral — Perhaps the most lethal manifestation of malperfusion syndrome in acute type A dissection is visceral ischemia. This presents in a variety of ways, including abdominal pain, peritoneal signs, lactic acidosis, and acute kidney injury. Computed tomographic angiography may also show hypoperfusion of abdominal organs or abrupt lack of contrast in an artery. While "malperfusion first" appears to be a viable strategy with comparable results to immediate definitive surgery, there remains considerable debate over which might be preferred. Additional data are needed to definitively support one treatment paradigm over the other. Individual institutional factors relating to availability of advanced endovascular interventions also factors into the management decisions for these patients.

In-hospital mortality rates for type A aortic dissection complicated by visceral malperfusion range from 34 to 63 percent [40,47-49]. However, in a series of 82 patients with acute type A complicated by malperfusion who underwent immediate definitive aortic repair, in-hospital mortality was overall low and similar to those who had no malperfusion syndrome (13.4 percent that 8.5 percent) [50].

For patients presenting with visceral malperfusion syndrome and no evidence of cardiac tamponade, coronary ischemia, or acute aortic insufficiency, the previously held dictum of immediate proximal aortic operation (open surgery as described above) has been challenged. A "malperfusion first" strategy proceeds with initial endovascular management to restore visceral flow using thoracic endovascular aortic repair (TEVAR) or fenestration of the descending thoracic aorta, followed by interval definitive aortic repair. This concept was initially pioneered by the group at the University of Michigan, who reported a 20-year experience of 82 type A dissection patients [51]. In-hospital mortality was 39 percent in the visceral malperfusion group, and in the last decade of their analysis, no patients suffered aortic rupture while awaiting open surgery. Among patients who survived to definitive open repair, postoperative mortality was not significantly different from those who received an open operation who presented with no malperfusion syndrome (2.1 versus 7.5 percent). In another institution, among 10 patients who underwent TEVAR followed by delayed aortic replacement, mortality was 30 percent [49]. In their algorithm, patients with abdominal malperfusion and no signs of cardiac tamponade, ongoing chest pain, or rupture underwent TEVAR followed by intensive care unit admission with serial exam and lactate assessment. If abdominal pain resolved and lactate was normalized after a 24-hour observation period, definitive proximal aortic repair was performed. Three patients in the "TEVAR-first" group died prior to receiving definitive proximal aortic surgery.

Extremity — Lower extremity malperfusion presents as a cool, pulseless leg. In the past, initial revascularization with femoral-femoral or axillary-femoral bypass to restore flow to the affected extremity followed by definitive central aortic repair was supported. Contemporary practice reserves revascularization for cases in which malperfusion persists following central aortic repair, which is performed immediately and usually improves the malperfusion syndrome. Depending on the extent and duration of ischemia, prophylactic fasciotomies should be considered to reduce the risk of compartment syndrome. (See "Acute compartment syndrome of the extremities" and "Lower extremity fasciotomy techniques", section on 'Prevention'.)

POSTOPERATIVE CARE — Once the patient is admitted to the surgical intensive care unit, the focus shifts to continued resuscitation and support of end-organs, hemodynamic monitoring, continued assessment for hemorrhage, and early neurologic exam.

Fastidious control of blood pressure is crucial in the first hours following type A dissection repair, as significant hypertension can result in hemorrhage. If a thoracic endovascular aortic repair was used with significant coverage (ie, frozen elephant trunk) of the descending thoracic aorta, blood pressure must be tailored to protect against spinal cord injury as well as postoperative hemorrhage. Rates of spinal cord injury following frozen elephant trunk repair for acute type A dissection repair range from 1 to 4 percent [34,52-55]. Therefore, the clinical team must be vigilant for early detection of proximal lower extremity neurologic dysfunction following these types of repairs and, if detected, must immediately raise blood pressure goals and if not already in place, insert a lumbar drain to allow cerebrospinal fluid drainage and optimization of cerebral perfusion pressure. [56]

Survivors of type A dissection repair require lifelong aortic surveillance, as most patients have residual dissection in the distal aorta, which can progressively dilate over time. Patients typically receive annual cross-sectional imaging (computed tomographic angiography). (See "Management of acute type A aortic dissection", section on 'Serial imaging'.)

REINTERVENTION FOLLOWING REPAIR — Repeat surgery is required in 20 to 40 percent of patients at 10-year follow-up, most commonly due to aneurysmal degeneration of residual aorta, aortic regurgitation, or aortic pseudoaneurysm and/or graft infection [32,57-60]. Higher reoperation rates generally reflect surgery among those with Marfan syndrome or other genetically mediated conditions and younger patients. In one small review, Marfan patients underwent 3.2 procedures per patient compared with 1.5 for non-Marfan patients [61]. In a study of from the International Registry of Acute Aortic Dissections (IRAD), the initial dissection was type A in 97 of 204 recurrent dissection [62]. Patients with recurrent aortic dissection (all) were more likely to have Marfan syndrome (21.5 versus 3.1 percent), but not bicuspid aortic valve (3.6 versus 3.2 percent). (See "Management of Marfan syndrome and related disorders".)

Proximal aortic repair — Reintervention may be needed to manage residual aortic root dilation, aortic valve degeneration or failure of prior valve repair or resuspension leading to aortic regurgitation. In one review, other indications for reoperation included enlargement of the false lumen in the residual aorta, suture dehiscence, and pseudoaneurysm formation at the proximal or distal aortic graft anastomosis, or at the coronary button anastomosis in patients who previously underwent a Bentall procedure [61].

In a review of 302 patients with type A aortic dissection, 124 patients had an aortic root >4 cm that was preserved (68 percent) [63]. At 5, 10, and 15 years, the cumulative incidence of proximal aortic reoperation was 5, 9, and 25 percent, respectively. Most patients required reoperation for aortic root dilation or aortic insufficiency. An aortic root diameter >45 mm and age were independently associated with proximal reoperation on multivariate analysis.

In another review of 104 patients with prior type A aortic dissection repair, progression of aortic dilation was seen in 91 patients (87 percent), aortic regurgitation in 21 (20 percent), and pseudoaneurysm in 15 (14 percent) [64]. Proximal reoperation was more frequent in patients who had a supracoronary ascending aorta replacement. Patients with an unresected intimal tear and distal extension of dissection flap experienced a higher rate of aortic arch and thoracoabdominal aorta redo procedures. In-hospital mortality after redo surgery was 7.7 percent.

In a review of 716 patients, 585 (81.7 percent) had undergone root repair and 131 (18.3 percent) root replacement [65]. The cumulative incidence of reoperation at 1, 5, and 10 years was similar after propensity score matching (root repair: 0, 1.4, 3.4 percent, respectively; root replacement: 0.8, 3.8, 8.6 percent, respectively). Sinus of Valsalva diameter ≥45 mm was a significant risk factor for proximal aortic reoperation in multivariate analysis. Survival was similar between the groups.

Distal aortic repair — The natural history of the distal aorta was evaluated in 89 surgical survivors of type A dissection. Serial computed tomographic (CT) angiography identified a median expansion rate of approximately 1 mm/year. Male sex, an initial descending aorta diameter of greater than 4 cm, or an initial diameter of less than 4 cm with a patent false lumen were identified as predictive of more rapid descending aorta expansion [66]. The overall risk of reoperation was 16 percent at 10 years.

The occurrence of a residual aortic dissection requiring subsequent surgical repair was evaluated in a report of 58 patients who were discharged from the hospital and then followed with serial magnetic resonance imaging [60]. Residual distal dissection was present in 45 patients (78 percent). The annual aortic expansion rate of this segment was 0.37 cm overall and significantly higher at 0.56 cm in the absence of thrombus in the false lumen. During a seven-year follow-up, 16 patients (28 percent) underwent reoperation because of progressive dilation of the untreated aortic segment. Residual dissection was confirmed in all but one, and 13 had no thrombosis in the false lumen.

Endograft complications — Although perioperative outcomes are improved for endovascular compared with open surgery, complications related to the endograft placement such as endoleak or device migration may indicate the need for reintervention, which occurs in up to one-third of patients. (See "Endoleak following endovascular aortic repair", section on 'Endoleak management after TEVAR'.)

FUTURE ENDOVASCULAR DIRECTIONS — Thoracic endovascular aortic repair (TEVAR) therapy has seen considerable technologic improvement as well as increasing applications in more proximal segments of the aorta. Several single and dual branch arch endografts are currently being investigated in clinical trials. (See "Endovascular devices for thoracic aortic repair", section on 'Advanced devices'.)

In addition to frozen elephant trunk repair of the distal aorta in Debakey 1 type A aortic dissection and TEVAR as part of a "malperfusion first" strategy for managing visceral malperfusion, there is also enthusiasm for TEVAR in managing the ascending aorta, particularly in patients at prohibitive risk for open surgical repair. (See 'Distal extent of repair' above and 'Visceral' above.)

Reports have described the feasibility of using existing TEVAR grafts to address ascending aortic pathology, though techniques remain limited at this time [67-69]. In a review from the Cleveland Clinic, 24 of 53 patients were turned down for type A dissection repair due to very high or prohibitive surgical risk and underwent CT scanning [70]. Nineteen had entry tears that were deemed coverable by a TEVAR device (located between the sinotubular junction and innominate artery or distal to the left subclavian artery). In a another review of 167 patients with type A aortic dissection, CT scans showed that 68 percent had an entry tear that was in a coverable zone of the ascending aorta with sufficient proximal and distal landing zone [71]. Despite these observations that many patients with acute type A dissection have anatomy potentially amenable to endovascular repair, significant technological device improvements are needed as well as operator experience prior to this becoming a realistic treatment option. As these technologies evolve, there is no question they will be applied to acute type A dissection patients, especially those with advanced age or severe comorbidities.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Aortic dissection and other acute aortic syndromes".)

SUMMARY AND RECOMMENDATIONS

●Type A aortic dissection – Acute type A aortic dissection (figure 6) is a surgical emergency for patients who are deemed suitable operative candidates. The goals of surgery for type A aortic dissection are to prevent aortic rupture and to treat any or a combination of acute aortic insufficiency; cardiac tamponade, retrograde dissection into the aortic root leading to myocardial infarction; antegrade propagation into the aortic arch resulting in stroke; or distal malperfusion syndromes. (See 'Surgical repair' above.)

●General principles – Repair of acute type A aortic dissection requires cardiopulmonary bypass and hypothermic circulatory arrest with adjunctive cerebral perfusion is common. The cardiac surgeon selects a cannulation strategy that provides adequate true-lumen perfusion to critical organ systems while avoiding peripheral vessels involved in the dissection. Outcomes for the three main cannulation strategies (ascending aorta, axillary artery, femoral artery) are similar. The optimal route of cerebral perfusion (antegrade versus retrograde) during hypothermic circulatory arrest is debated, as is the optimal temperature. (See 'Hypothermic circulatory arrest' above.)

●Extent of repair – After cardiopulmonary bypass and systemic hypothermia are established, the proximal extent of the dissection is evaluated to determine whether supracoronary ascending aortic replacement or aortic root replacement (typically valve sparing) is more appropriate. Then the distal extent of the dissection is evaluated to determine whether hemiarch, partial arch, or total arch replacement is needed. Whether antegrade descending thoracic endovascular aortic repair (ie, frozen elephant trunk) should be performed at the time of DeBakey 1 type A acute aortic dissection repair is debated. (See 'Proximal extent of repair' above and 'Distal extent of repair' above.)

●Managing malperfusion – Possible sites of malperfusion, defined as regional ischemia within an organ system, include the coronary, cerebral, visceral, spinal, and lower extremity vascular beds. For patients with coronary or cerebral malperfusion, expedient surgical repair is paramount. For patients with visceral malperfusion in the absence of cerebral or coronary malperfusion, whether to prioritize the malperfusion using a "malperfusion first" strategy is unknown. For peripheral malperfusion, central aortic repair may be more expeditious for restoring extremity blood flow; however, if the duration of ischemia is prolonged, extremity fasciotomy may be necessary. (See 'Managing malperfusion' above.)

●Reoperation – Survivors of type A dissection repair require lifelong aortic surveillance, typically annual cross-sectional imaging. Reoperation is required in 20 to 40 percent of patients at 10 years typically related to aneurysmal degeneration of residual aorta or aortic regurgitation, among other reasons. (See "Management of acute type A aortic dissection", section on 'Long-term management' and 'Reintervention following repair' above.) [32]