Introduction
Neurologic injury has been a major concern in cardiac surgery due to the atherosclerotic burden in the vasculature, cross clamping, and the use of extracorporeal circulation, leading to a significant increase in morbidity, mortality, and cost.1-3 Furthermore, disrupting blood flow to the brain will compound on the potential for neurologic complications. The first documented aortic replacement dates back to the 1950s with Denton A. Cooley, Michael DeBakey, and E. Stanley Crawford.4 In both 1951 and 1955, aortic arch replacements with the use of bypass shunts and moderate hypothermia were fraught with failure. Only after DeBakey’s concept of an early form of antegrade cerebral perfusion in 1957, did surgeons successfully enter a new era of aortic surgery. This was followed by Griepp’s introduction of total circulatory arrest under deep hypothermia (up to 14 °C) in 1975.5-7 With these advances, the risk of stroke and mortality with aortic surgery has greatly improved over the past 70 years. The main strategies that exist today to mitigate the risk of brain injury include deep hypothermic circulatory arrest (DHCA), antegrade cerebral perfusion (ACP), retrograde cerebral perfusion (RCP), as well as a hybrid approach.8 The choice in cerebral protection is dependent upon the surgeon’s preference and experience with a specific technique, so it is important to recognize the potential advantages or disadvantages of each strategy.
Brain protection strategies
Dating back to the 1970s, the DHCA strategy works by reducing injury-inducing pathways by lowering the cerebral metabolism.9 The body is usually cooled to temperatures of 20ºC or greater. Longer times of DHCA have been correlated to an increased rate of neurological dysfunction, as described by Yan and colleagues (Fig. 1),10 so there is usually a need for the use of additional cerebral protection strategies (ACP and RCP). RCP requires connecting the arterial tubing to a cannula placed in the superior vena cava (SVC) and then reversing blood flow to the brain in a retrograde fashion. This approach has been successful in reducing both mortality and morbidity.11 An advantage of this strategy is that it provides cold retrograde blood flow to the brain during DHCA and therefore allowing for a decreased risk of embolization.12Moderate hypothermic circulatory arrest (MHCA) with temperatures ranging from 24ºC to 28ºC can be combined with selective ACP.13 Once circulatory arrest is established, one or multiple arch vessels are cannulated and perfused with cold blood in an antegrade fashion. Some studies have suggested that ACP leads to better neurological outcomes when compared with RCP.14The debate between RCP and ACP has been ongoing since the 1990s. Despite the fact that RCP has proven to be effective, many surgeons still prefer using ACP.15 RCP neuroprotective effects are explained by its continuous cerebral cooling, and not from antegrade nutritive blood flow.16 RCP’s efficacy may also be reduced by the presence of venous valves that require an elevated perfusion pressure, which could cause increase intracranial pressure and subsequently worsen neurological outcomes.17 However, RCP does not require exposure or manipulation of the aortic arch vessels, whereas ACP requires insertion of a catheter into the arch vessels, vessel clamping or snaring; which can increase the potential of vessel injury or embolization. Nonetheless, there has not been a clinical study demonstrating that one approach is superior to the other.18 RCP provides improved exposure of the arch vessels, although there is a constant flooding of venous blood return into the operative field, whereas ACP yields a dryer field with catheters potentially obscuring visibility.
Surgical techniques
1. Antegrade cerebral perfusion
Antegrade cerebral perfusion can be safely achieved unilaterally through the innominate or right common carotid artery, as well as combined with the left common carotid artery to achieve bilateral ACP. When comparing unilateral and bilateral ACP, most published series quote use of unilateral ACP only if adequate collateralization is demonstrated through backflow from the contralateral carotid artery or if contralateral perfusion is evident by near-infrared spectroscopy (NIRS).4 Bilateral ACP may be more advantageous when considering existing conditions such as arterial stenosis, stroke, and other vascular anomalies. Many surgeons prefer bilateral ACP to ensure bihemispherical perfusion due to its documented efficacy and safety beyond 40-50 mins of circulatory arrest.19Alternative cannulation strategies can be utilized to establish antegrade cerebral perfusion. A prosthetic graft can be anastomosed to the right subclavian, innominate, or right common carotid artery. Additionally, balloon occlusion perfusion catheters can be utilized to provide antegrade cerebral perfusion without vessel clamping or snarring. Peripheral cannulation of the axillary artery can be performed to establish cardiopulmonary bypass and subsequent ACP with concomitant insertion of a balloon occlusion perfusion catheter into the carotid artery without directly cannulating supra-aortic vessels.4
2. Retrograde cerebral perfusion
Retrograde cerebral perfusion takes a different approach as cold oxygenated blood is delivered retrograde through a cannula placed in the superior vena cava.19 After the patient is cooled to achieve deep hypothermia and circulatory arrest established, RCP can be employed. If a bicaval venous cannulation technique is employed for cardiopulmonary bypass, Y-connectors with limbs are utilized to connect the arterial to the venous lines. These limbs are clamped during the conduct of cardiopulmonary bypass. When beginning to initiate retrograde cerebral perfusion, the SVC cannula is snared and the limb connecting the arterial cannula to the SVC cannula is unclamped.20 This enables cold blood to run retrograde into the SVC to perfuse the brain (Fig. 2). A cardiotomy suction is placed in the open thoracic aorta to suction the blood return from the arch vessels.21 Ice packs are also placed on the scalp of the patient for additional topical brain cooling. RCP, as well as ACP, has been historically performed via a median sternotomy, upper hemisternotomy, or other partial sternal splitting variations. However, currently, RCP is the only cerebral protection strategy that can be employed through a right mini-thoracotomy minimally invasive ascending aorta and hemiarch surgery.22A hybrid approach to these techniques include utilizing 2 separate heater coolers, one for the systemic circulation and one for the cerebral circulation. With this approach, the right carotid artery is exposed in the neck or the innominate artery is exposed from within the chest. Each vessel is directly cannulated and clamped with perfusion coming from a separate tubing connected to the heart lung machine with its own respective heater cooler. Since there are 2 heater coolers, perfusion to the brain can be maintained at 10 degrees C, while perfusion of the body can remain a mild to moderate hypothermia (28 degrees C). This not only allows continuous brain perfusion, but avoids the potential coagulopathy associated with systemic hypothermia.
Discussion
Improvements to cerebral protection strategies have enabled longer circulatory arrest times when compared to DHCA alone, which limits the safe circulatory arrest time to only 30 minutes.18 By incorporating ACP and RCP into the cerebral protection strategy, stroke and mortality rates have been reduced. Although using MHCA with ACP has been the primary neuroprotective strategy in many high-volume aortic centers worldwide, the question of which strategy provides better outcomes in ascending aorta, hemiarch or arch replacement remains an area of significant controversy and debate. Thus, several studies have attempted to answer which cerebral protection strategy offers the best patient outcomes (Table 1). In a prospective study, Abdelgawad and colleagues demonstrate excellent results utilizing MHCA with ACP. Although, these results were comparable to those of DHCA with RCP. ACP was correlated with both shorter clamp and ischemic times with no statistically significant difference in mortality and stroke rates.23 Other studies have also advocated for ACP as the strategy of choice. In a 2008 study, Apostolakis and colleagues investigated outcomes in 48 patients undergoing urgent aortic surgery for type A dissection. Their results concluded that ACP had a lower rate of stroke, earlier extubation, shorter ICU stay, and reduced costs.24 Wiedemann and colleagues analyze 329 patients undergoing aortic surgery for type A dissection and also found ACP to be associated with better short and long-term survival rates.25However, several studies suggest RCP with DHCA to be equivalent, if not better, to ACP with MHCA. In a prospective, randomized study of 20 patients undergoing hemiarch replacement, Leshnower and colleagues used magnetic resonance imaging (MRI) to evaluate the incidence of brain injury. Despite no significant difference in clinically evident neurologic injury between patient who had MHCA with ACP and DHCA with RCP patients, the former had a higher incidence of radiographic neurologic injury (n=9,100% vs n=5, 45%; p=0.01).26Moreover, an STS database analysis of 7353 patients found that retrograde cerebral perfusion demonstrated a low risk of acute stroke in patients who underwent type A dissection repair.27The literature is further obscured by multiple studies advocating one technique over the other. A large Society of Thoracic Surgeons database analysis evaluated 7830 patients undergoing hemiarch or total arch replacement from 2014 to 2016. The study excluded aortic dissections from the analysis. It concluded that in patients requiring more than 30 minutes of circulatory arrest, optimal strategies for cerebral protection include deep hypothermia with either antegrade or retrograde cerebral perfusion, or moderate hypothermia with antegrade cerebral perfusion.28 This conclusion essentially suggests that either strategy is acceptable as long as some form of cerebral perfusion is used. Furthermore, in a large study of 4,128 patients undergoing ascending aortic repair for Type A dissection in Japan, RCP had similar results in both mortality and neurologic outcomes when compared to ACP.29Recently, minimally invasive surgical techniques have been applied to replacement of the ascending aorta and hemiarch at experienced centers. The majority of minimally invasive procedures reported are performed via an upper hemisternotomy (Fig. 3).30-32 In a mini-thoracotomy approach, cerebral perfusion is exclusively performed via RCP.33 The initial outcomes of the first reported minimally invasive, right mini-thoracotomy replacement of the ascending aorta and aortic valve in 20 patients was excellent; with no strokes, reoperations for bleeding, conversions to sternotomy or 30-day mortality.34 In 2018, Lamelas and colleagues further described their experience with a sternal-sparing replacement of the ascending aorta with or without a concomitant aortic valve replacement via a right mini-thoracotomy in 74 patients. A propensity score-matched analysis was performed between 63 patients undergoing a sternotomy and an equivalent number of patients undergoing a minimally invasive non-sternotomy approach. The 30-day mortality was 3.2 % in both groups, while stroke incidence was 0% in the minimally invasive group and 1% in the sternotomy group. Despite a longer circulatory arrest times in the minimally invasive group, stroke rate, as well as the transfusion requirements, ventilation times, as well as the ICU and hospital stays were less than the sternotomy group.35
Conclusions
The literature regarding the preferential use of antegrade versus retrograde cerebral perfusion during aortic surgery remains controversial. Analysis of many large databases in thoracic surgery have yielded excellent results for both cerebral protection strategies. The minimally invasive hemisternotomy or right mini-thoracotomy approach to replacement of the ascending aorta has also demonstrated excellent results. Future studies will be needed to compare sternotomy and minimally invasive approaches utilizing both antegrade and retrograde cerebral perfusion techniques.