Optimal brain protection in aortic arch surgery

Abstract

There is considerable debate with regard to the optimal cerebral protection strategy during aortic arch surgery. There are three contemporary techniques in use which include straight deep hypothermic circulatory arrest (DHCA), DHCA with retrograde cerebral perfusion (DHCA + RCP), and moderate hypothermic circulatory arrest with antegrade cerebral perfusion (MHCA + ACP). Appropriate application of these methods ensures appropriate cerebral, myocardial, and visceral protection. Each of these techniques has benefits and drawbacks and ensuring coordinated circulation management strategy is critical to safe performance of aortic arch surgery. In this report, we will review various cannulation strategies, review logistics of hypothermia, and review the relevant literature to outline the strengths and weaknesses of these various cerebral protection strategies.

Introduction

The field of aortic arch surgery, like cardiac surgery, is quite young relative to the rest of medicine. The first attempt at resection of diseased aortic arch and re-establishment of aortic continuity was first performed by Dr. Crafoord in 1944 in the setting of coarctation [1]. Dr. DeBakey and Dr. Cooley reported the first successful repair of an aneurysmal aortic arch in 1955 [2]. This was performed using normothermic cardiopulmonary bypass, cold ischemic cardiac arrest, and high-flow antegrade cerebral perfusion (ACP). Soon thereafter, cannulation strategies started to change as there was a shift from performing cardiac surgery via thoracotomy to sternotomy. Over the next 20 years, the predominant arterial cannulation site changed from the subclavian artery to femoral artery to direct aortic cannulation [3].

Hypothermia was first utilized in cardiac surgery to repair atrial septal defects by Bailey (1952) and independently by Lewis (1953) [4, 5]. Griepp introduced deep hypothermic circulatory arrest (DHCA) for arch surgery in 1975 and others followed soon thereafter [6, 7]. Deep hypothermia in these early series included temperatures from 14 to 18 °C and these early efforts were marred by a 30–40% operative mortality secondary to bleeding, cerebral complications, and cardiac failure. These complications were thought to be associated with the methods of inducing and reversing deep hypothermia and the associated coagulopathy which ensued. In response, Cooley et al. modified their approach and reported their first attempt at moderate hypothermic (24 °C) circulatory arrest in 1982 with a dramatic reduction in operative mortality to 10% [7]. That same year, Lemole et al. first reported use of retrograde cerebral perfusion (RCP) to improve neurologic outcomes in arch surgery [8].

All contemporary strategies for circulation management and cerebral protection techniques utilized during aortic arch surgery can trace their origins back to this 30-year period of innovation. These include deep versus moderate circulatory arrest without or with antegrade or retrograde selective cerebral perfusion and a variety of arterial cannulation strategies. Over the last 40 years, advances in cardioplegia strategies and circulation management have significantly reduced the risk of aortic arch surgery. Having a thorough understanding of the various cerebral protection strategies is essential for safely performing aortic arch surgery. We review the historical and current strategies for cerebral protection during aortic arch surgery, their mechanisms of action, their respective risks and benefits, and their contemporary outcomes.

Arterial cannulation strategies

Multiple options exist for arterial inflow cannulation and the optimal site remains unknown.

Femoral artery cannulation

Femoral artery cannulation is the oldest method of arterial cannulation for arch surgery [9, 10]. Historically, femoral artery cannulation was marred by reports of arterial injury, retroperitoneal bleeding, retrograde dissection, malperfusion, and increased stroke rate [11–13]. The latter was thought to be secondary to retrograde embolization of aortic atheroma [11]. However, recent studies from high-volume centers have demonstrated safety in select elective procedures while in urgent/emergent operations such as in the setting of aortic dissection, femoral artery cannulation is safe for both survival and leading to acceptable neurologic outcomes [9, 14]. Ayyash et al. selectively use femoral artery cannulation in patients who do not have mobile atheroma in the descending thoracic aorta by transesophageal cannulation [9]. In their 2011 study, they reported a 1.8% stroke rate. In the setting of type A dissection, Kamiya et al. reported a 4.5% stroke rate compared to 4.9% in the direct aortic cannulation group (p = 0.86) [15]. Ease of access makes this site of arterial inflow ideal for aortic dissection. In the setting of aortic arch replacement surgery, femoral artery cannulation would not allow ACP without an additional method for cerebral arterial inflow.

Direct aortic cannulation

Direct aortic cannulation can be instituted quite rapidly without separate incision, which may be necessary in the setting of aortic dissection [15]. As with femoral artery cannulation, there is a risk of end-organ malperfusion especially in the setting of aortic dissection [15]. Its proponents show equal or even superior survival outcomes and equal neurologic outcomes when compared to femoral artery cannulation [16, 17]. Like femoral artery cannulation, it mandates DHCA since it does not allow ACP during arch surgery without an additional arterial inflow.

Axillary/carotid/innominate artery cannulation

Axillary cannulation has more recently become the arterial inflow site of choice during aortic arch surgery [10, 18, 19]. Axillary artery cannulation allows for antegrade flow and avoids manipulation of a potentially diseased aorta thereby preventing potential dislodgement of atheroma to the brain. Axillary artery cannulation can be time consuming as it requires a separate incision and in many high-volume centers a separate graft is sewn onto the artery through which the cannula is placed [19]. Additionally, there have been reports of brachial plexus injury and arm malperfusion with this technique [10]. Carotid artery cannulation has similarly gained recent popularity. Like axillary cannulation, it requires a separate incision and often requires a “chimney graft” approach [20, 21]. Those that use it do so with good outcomes and find it to be less time consuming than axillary cannulation, especially in obese patients in whom axillary cannulation is particularly difficult [20, 21]. Innominate artery cannulation provides similar arterial access while avoiding a separate incision and could avoid some of the local complications associated with axillary cannulation [22]. One of the biggest advantages of axillary, carotid, and innominate artery cannulation is that it allows for easy access for ACP.

Science and logistics of safe hypothermia

Effects of hypothermia and intraoperative neurologic monitoring

In 2013, an expert consensus determined that for the purposes of aortic surgery, hypothermia would be classified as follows: profound hypothermia (< 14 °C), deep hypothermia (14.1–20 °C), moderate hypothermia (20.1–28 °C), and mild hypothermia (28.1–34 °C) [23, 24]. Hypothermia reduces cerebral metabolism of oxygen and glucose which increases the ratio of adenosine triphosphate supply to demand [25, 26]. Hypothermia also decreases temperature-dependent release of excitatory neurotransmitters, decreases free radical and inflammatory cytokine formation and release, and decreases intracellular calcium uptake [26]. Hypothermia can reduce inflammation and cell death associated with cerebral ischemia-reperfusion injury after circulatory arrest [25] and allows for a safe short period of cerebral circulation interruption. On the other hand, hypothermia has significant cerebral and systemic adverse side effects including but not limited to increased blood and plasma viscosity, impaired coagulation, impaired microcirculation, and hyperglycemia [25, 26].

Intraoperative monitoring of cerebral temperatures and neurologic activity optimizes the safety profile of using hypothermia to perform aortic arch surgery. Given that there can be a discrepancy between brain temperature and body temperature depending on the protection strategy, it is recommended that the temperature be measured in two different locations: the nasopharynx for the brain and the bladder or rectum for the body have been well described [25, 27, 28].

Neurological monitoring tools can broadly be broken down into either functional monitoring such as bispectral index (BIS), electroencephalography (EEG), and somatosensory evoked potentials or those aimed at measuring O₂ saturations such as transcranial Doppler, near-infrared spectroscopy (NIRS), and jugular venous oxygen saturations [25, 29]. The advantages and disadvantages are briefly summarized in Table 1. A multimodal approach to neurological monitoring will likely allow for the safest execution of hypothermic circulatory arrest (HCA). Given their ease of use and non-invasive nature, NIRS and BIS are the most often used intraoperative neurologic monitoring tools [25]. NIRS measures the cerebral hemoglobin oxygen saturation (ScrO₂), which is inversely proportional to neurologic damage [25]. Circulatory arrest will result in desaturations of NIRS measurements; surgeons should be cognizant of excessive desaturations or unilateral desaturations which can signify potential neurologic damage [25]. There is no national or international standard for the threshold below which patients are at high risk of neurologic damage [25]. However, multiple groups report that a reasonable threshold is an absolute value of 40% or 25–30% below baseline [30, 31]. BIS serves as a measure of brain activity during cooling, rewarming, and circulatory arrest [25]. BIS decreases in a biphasic manner during cooling and is 0 below 18 °C in the majority of patients [25, 32]. The trajectory of BIS increase varies from patient to patient [32]. BIS can serve as an early warning side of unilateral or bilateral inappropriate brain activity and can alert the surgeon to inappropriate cooling [32]. While it is more cumbersome, EEG provides more detailed insight into the brain’s electric activity. Instead of strictly targeting a “deep hypothermia” temperature threshold, some centers in fact target electrochemical silence prior to instituting circulatory arrest [33, 34]. In some patients, this can be achieved at temperatures as high as 27 °C which allows the patient to benefit from the hypothermia-associated cerebral protection while limiting adverse side effects [34]. In addition to inappropriate EEG spikes, differences in EEG activity between hemispheres throughout the case can also serve as an indicator of inadequate cerebral protection [25, 34].

Table 1.

Comparing the advantages and disadvantages of various intraoperative cerebral activity monitoring devices [25, 29]

	Advantages	Disadvantages
Functional monitoring
Electroencephalography (EEG)     ○ Gold standard for regional and global electrocerebral activity monitoring	• Specific regional monitoring • Determines the point of true electrochemical silence (ECS)	• Confounded by certain anesthetics • Complex setup
Bispectral index (BIS)     ○ A simplified measurement of global cerebral activity	• Non-invasive • Simple setup • Correlates well with ECS	• Does not provide laterality unless bilaterally BIS used • Varies patient to patient • Confounded by certain anesthetics
Somatosensory evoked potentials (SEP)     ○ Used to monitor brain stem and cortical activity via peripheral nerve stimulation	• Specific regional monitoring, including brain stem	• May not correlate with true ECS • Complex setup
Oxygenation monitoring
Transcranial Doppler (TCD)     ○ Used to monitor real-time antegrade cerebral flow	• Non-invasive • Inexpensive • Ideal for flow measurement during unilateral antegrade cerebral perfusion	• Operator dependent • No regional cerebral data
Jugular venous O2 (SjO2)     ○ Real-time temperature and venous oxygen saturation	• Simple setup • Detects global metabolic activity	• Invasive • No regional cerebral data
Near-infrared spectroscopy (NIRS)     ○ Real-time non-invasive oxygen saturation	• Detects global and regional metabolic activity • Non-invasive • Simple setup	• No universal definition of desaturation

Considerations during cooling

Independent of the target level of hypothermia or adjunctive cerebral protection measures, the nuances of cooling are critical to optimizing brain and patient protection. It is generally accepted that cooling and rewarming must be performed gradually [7, 35]. Rapid temperature changes can lead to cerebral edema, gas dissociation and emboli, and ischemic and hyperthermic injury [26, 36]. The rate of rewarming both historically and in contemporary series is no more than a 10-degree difference between inflow blood temperature and body and brain temperature [7, 25, 35]. Another factor to consider is the cooling time. Milewski and colleagues found that when targeting deep hypothermia, a cooling time of 45 min leads to EEG silence. As a result, they target this cooling time when EEG monitoring is not present [37].

Acid/base management also impacts cerebral protection. With cooler temperatures, there is increased solubility of gases including carbon dioxide into the blood reducing the partial pressure of carbon dioxide and leading to a relative alkalosis and vasoconstriction [24, 36]. The pH can be maintained and monitored using the alpha stat method or the pH stat method [24, 36]. The pH stat method monitors and adjusts the temperature-corrected pH. The alpha stat method adjusts the pH without taking into account the temperature. As a result, the alpha stat method maintains the cerebral blood flow autoregulation and cerebral metabolism autoregulation which can prevent cerebral edema [24]. The pH stat method actively disrupts this autoregulation leading to increased cerebral blood flow during cooling and potentially better cooling, but this can come at the cost of increased microemboli [24]. Neither of these methods has proven to be more clinically efficacious with regard to neurologic protection [26].

Given the brain’s dependance on oxygen-mediated glucose metabolism, maintenance of euglycemia during hypothermia is critical to cerebral protection [26]. Patients undergoing hypothermic arrest are susceptible to hyperglycemia which can increase availability of excitatory neurotransmitters and lactic acidosis and this can worsen any potential neurological insults [25]. Similarly, hypoglycemia can lead to cerebral damage [38].

Most centers use some sort of pharmacologic cerebral protection such as steroids, barbiturates, statins, mannitol, and magnesium [25]. Inhaled anesthetics have been shown to reduce excitatory neurotransmitter signaling thereby reducing neurologic dysfunction [39]. Barbiturates are often used as adjuncts to the neuroprotection strategy given the theoretical reduction in apoptosis [18]. Propofol has a general antioxidant and anti-inflammatory effect and may reduce postoperative neurologic dysfunction as well [40]. Corticosteroids are widely used as well given their anti-inflammatory effects [41]. Ultimately, while these drugs are used clinically given their theoretical and potential mechanistic beneficial effects, no one drug has proven to reduce the incidence of adverse neurologic events [42].

Blood pressure monitoring during all phases of the operation remains critical to optimizing patient safety. Many centers recommend bilateral, versus unilateral, radial arterial line monitoring especially when ACP is used [18]. Intraoperatively, both hypotension and hypertension during ACP have been found to be detrimental [18]. Postoperatively, no set blood pressure management guideline exists; however, our practice has been to maintain a mean arterial pressure of 80 mmHg to 100 mmHg in hopes of optimizing cerebral blood flow.

Outcomes endpoints

Survival and neurologic dysfunction have been the mainstay in reporting outcomes after aortic arch surgery since its infancy [6, 7]. However, in 1994, Dr. Griepp first proposed the separating of neurologic dysfunction into temporary neurologic dysfunction (TND) and permanent neurologic dysfunction (PND) [43]. TND refers to events such as temporary delirium, transient ischemic attacks, and other brain-associated adverse symptoms that resolve. PNDs refer to stroke or coma which do not resolve. This classification has been adopted by most high-volume centers. More recently, Leshnower et al. drew attention to subclinical diffusion-weighted images (DWI) on postoperative magnetic resonance imaging (MRI) as part of their Deep Hypothermia With Retrograde Cerebral Perfusion Versus Moderate Hypothermia With Antegrade Cerebral Perfusion for Arch Surgery (DRAMA) trial. Their finding of 70% of patients in their small study having DWIs indicates that there is clearly a high subclinical neurologic insult that is incurred during aortic arch surgery [44].

Deep hypothermic circulatory arrest

Straight DHCA is the “oldest” means of systemic and cerebral protection in aortic arch surgery; it was popularized by Griepp and has been used for the last 45 years [6]. Isolated DHCA in the modern day involves cannulating the patient and cooling the patient gradually with or without topical cooling of the head to 14–20 °C then turning off the cardiopulmonary bypass circuit. The site of arterial cannulation can be the femoral artery, axillary artery, or direct aortic. Venous cannulation can be the right atrium or the femoral vein. Advantages of straight DHCA include a dry and motionless operative field and lack of aortic cross clamp, thus preventing clamp-induced cerebral embolism [45]. Disadvantages include exposing the brain to reperfusion injury, extra time on the cardiopulmonary bypass machine to cool and rewarm, and hypothermia-associated coagulopathy [45].

Damberg et al. recently reported their outcomes with isolated DHCA [35]. They reported a rate of TND, PND, and early mortality of 5.1%, 2%, and 2.9% respectively in a series of 613 elective and non-elective aortic operations [35]. They identified acute type A dissection, redo status, and descending aortic pathology as risk factors for mortality within 1 year. Interestingly, DHCA time was not found to be a risk factor for 1-year mortality on multivariable analysis [35]. The majority of these patients had less than 40-min circulatory arrest time and the authors found DHCA time greater than 50 min was associated with increased stroke rates [35]. Given this, these results cannot be generalized for those patients who require a longer HCA time.

Retrograde cerebral perfusion

Continuous RCP as an adjunct to DHCA was first reported by Ueda et al. [46]. It has the potential to provide additional assistance cooling the brain during circulatory arrest, flush out possible emboli, and possibly provide nutritive flow [45]. This requires a more complicated cardiopulmonary bypass circuit. The typical cardiopulmonary bypass circuit and a bridge between the venous and arterial tubing to achieve both venous drainage and arterial inflow are required. Venous cannulation, often central, is performed via a right atrial cannula and separate superior venous cannula (SVC) after SVC isolation with Rummel tourniquet. This SVC cannula is then connected to the venous and arterial tubing via the separate bridge. The arterial cannula can be placed in the femoral artery, axillary artery, or directly in the aorta. This setup allows the SVC cannula to be able to participate in venous drainage while cooling and RCP during circulatory arrest. RCP flow rates should target 100–400 mL/min to a pressure of 25–30 mmHg to minimize cerebral edema [37, 44, 47]. It is not uncommon for debris to be expelled from the carotid arteries upon initiation of RCP and this debris should be scavenged if possible, to prevent descending thoracic aorta embolism.

With these techniques, two recent large single-center series from Lau et al. and Milewski et al. have reported a rate of TND, PND, and early mortality of 3–4%, 1–3%, and 3–4% respectively [47, 37]. Both studies reported outcomes on patients undergoing proximal aortic and proximal arch or total arch replacement. Milewski et al. did not include emergent or aortic dissection patients in their study. Lau et al. performed a risk factor analysis and showed that increased age, prior myocardial infarction, longer cardiopulmonary bypass time, and intraoperative blood transfusion were predictors of early mortality. Prior stroke and redo status were risk factors for postoperative stroke [47]. Lau et al. also propensity score matched the patients with circulatory arrest times greater than 50 with those who had circulatory arrest times less than 50 min. In this matched sub-analysis, they did not find a significant difference in rates of TND, PND, or early mortality. These results suggest that RCP lengthens the safe duration of circulatory arrest with respect to cerebral protection.

Antegrade cerebral perfusion

DeBakey was the first to use ACP in arch surgery in 1955, but it fell out of favor with the advent of DHCA [2]. ACP was then re-introduced by Frist and Bachet in the late 1980s and early 1990s but it was really Dr. Kazui who popularized today’s ACP [48–50]. Optimal techniques regarding ACP are still being debated, and these debates center around optimal flow rate, cerebral perfusion temperature, circulatory arrest temperature, and unilateral versus bilateral ACP [51]. Advantages of ACP include that it can provide potentially nutritive cerebral perfusion, ensure constant cooling, and allow less intense hypothermia potentially obviating the systemic risks of deep hypothermia [45]. Cannulation strategy for ACP can include axillary or innominate artery and right atrial cannulation through which both systemic cooling and unilateral selective ACP can be performed. Alternatively, cardiopulmonary bypass cooling can be instituted via direct aortic or femoral artery cannulation with separate cannulation through the axillary or innominate artery for ACP. Cannulating the left carotid artery allows for bilateral ACP. Bilateral ACP has the theoretical advantage of providing more complete nutritive flow, at the potential increased risk of showering emboli from vessel manipulation [25, 52, 53]. The majority of studies comparing unilateral ACP versus bilateral ACP have not shown a difference in postoperative PND, TND, or early mortality [25, 52, 53]. The decision will likely be patient specific and based on presence of carotid obstruction and patency of the circle of Willis [25, 52, 53].

Kazui et al. popularized a fixed 10 mL/kg/min flow rate during ACP while others use a pressure-based flow rate targeting blood pressure around 75 mmHg [49, 51]. Inadequate flow may cause ischemic damage, especially if the circulatory arrest time is long while excessive flow can cause cerebral edema [24]. Natural cerebral autoregulation occurs between 60 and 160 mmHg and ensuring adequate pressure to allow this autoregulation while minimizing excessive pressure to prevent cerebral edema will likely allow safe ACP [24, 54].

ACP has been performed in the context of deep, moderate, and even mild hypothermia [51, 55]. Preventza et al. compared DHCA with ACP versus moderate hypothermic circulatory arrest (MHCA) with ACP in cases requiring arch intervention and a circulatory arrest time > 30 min. They reported an operative mortality and PND rate of 16% and 12% respectively for the DHCA + ACP group, and 12% and 6% respectively for the MHCA + ACP group which were not statistically significant [55]. However, the MHCA group had lower incidence of reoperation for bleeding and lower incidence of prolonged ventilation [55]. Additionally, in a sub-analysis, they showed that DHCA + ACP led to higher stroke rates for patients requiring >45-min circulatory arrest (17% versus 7%, p = 0.041) [55]. Leshnower et al. did a similar comparison but just in the setting of acute type A dissection and did not show a difference in mortality, TND, or PND; however, the MHCA + ACP group had shorter postoperative mechanical ventilation times (1.3 days versus 1.9 days, p = 0.05) [56]. Ahmad et al. recently published their results using ACP with mild to moderate hypothermia in aortic arch surgery. They reported a TND rate, PND rate, and 30-day mortality of 6%, 2%, and 6% respectively and also compared unilateral versus bilateral ACP and found no difference between the two strategies in the incidence of TND or PND [51]. These findings were corroborated by Tong et al. [57]. While ACP as an adjunct to DHCA, MHCA, or mild HCA results in excellent cerebral protection, prolonged circulatory arrest time at higher temperature may affect the viscera adversely. Ahmad et al. do not recommend ACP times greater than 60 min in order to protect visceral organs from prolonged mild HCA [51], while Di Eusanio et al. showed that ACP times greater than 90 min during strict MHCA (22–26 °C) do not adversely affect mortality or neurologic outcomes [58].

Comparative studies

The majority of studies currently comparing these three cerebral protection methods are single center and retrospective in nature. Kayatta et al. provide a summary of recent single-center comparative studies in their review [24]. The advent of the Society of Thoracic Surgeons (STS) database has allowed for more large-scale comparative studies. Itagaki et al. recently reviewed the outcomes of arch surgery patients in the STS National Database from 2014 to 2016 and separated patients based on straight HCA, ACP, and RCP [59]. The median nadir temperatures were 19.9 °C, 22.6 °C, and 20.1 °C respectively and they found that the early mortality was lowest for the RCP group at 4.4% compared to the ACP group at 6.1% and straight HCA group at 8.5% [59]. Similarly, the PND rate was lowest for the RCP group at 4.6% compared to the ACP group at 6.2% and no cerebral perfusion group at 7.6% [59]. The study also showed that the longer the time of circulatory arrest, the greater the risk of the combined outcome of early mortality and PND [59]. Straight HCA times greater than 20 min led to increased risk of the combined outcome [59]. A significant increase in the risk of the combined outcome did not occur in the ACP or RCP group until at least 50 min of circulatory arrest [59]. A recent network meta-analysis reviewing 68 studies and nearly 27,000 patients corroborated these findings and found that RCP was protective of postoperative stroke (odds ratio 0.69, 95% confidence interval 0.57–0.82) and operative mortality (odds ratio 0.66, 95% confidence interval 0.56–0.78) when compared to straight DHCA [53]. Similarly ACP was protective of postoperative stroke (odds ratio 0.62, 95% confidence interval 0.51–0.75) and operative mortality (odds ratio 0.63, 95% confidence interval 0.51–0.76) when compared to straight DHCA [53]. In this study, there was no difference between ACP and RCP [53].

To date, there have been only 3 prospective randomized controlled trials comparing different cerebral protection techniques and all 3 studies have limited number of patients. Svensson et al. randomized 30 patients into 3 groups: straight DHCA, DHCA + RCP, DHCA + ACP [60]. Nearly all patients, 96%, (27 patients) had TND in the first week and all but 9% (3 patients) had resolution by 3 weeks postoperatively [60]. Ultimately, they concluded that neither RCP nor ACP had additive benefit over straight DHCA [60]. Okita et al. randomized 60 patients to either DHCA + RCP or DHCA + ACP [61]. While there was no difference in early mortality or PND, the RCP group had a higher incidence of TND compared to the ACP group (33% versus 13%, p = 0.05) [61]. Neurocognitive testing was similar between the two groups [61]. Most recently, Leshnower et al. randomized 20 patients to either DHCA + RCP or MHCA + ACP. Each patient had a postoperative head MRI as well as underwent neuro testing. There was no difference in early mortality, PND, TND, or neurocognitive scores [44]. However, 100% (9/9) MHCA + ACP patients had diffusion-weighted lesions on MRI versus 45% (5/11) in the DHCA + RCP group [44].

Conclusion

The incidence and nature of neurologic morbidity during aortic arch surgery is multifactorial and the improvement in outcomes over the last few decades likely reflects a better understanding of all aforementioned variables. Thus, no study will truly reflect the impact of any single intervention such as cannulation strategy or degree of hypothermia. However, certain tenets seem to remain true during comparative studies and single-center series. Both RCP + DHCA and ACP + MHCA seem to have improved neurological outcomes compared to use of isolated DHCA [59]. DHCA should generally be used for procedures requiring less than 50 min, ideally less than 20 min, of circulatory arrest [35, 59]. RCP and ACP can extend the circulatory arrest time to 50 min and even potentially up to 90 min in the case of ACP [51, 58, 59]. However, regardless of the circulatory arrest time, in most hands, both RCP and ACP will lead to a reduced mortality and neurologic deficit rate when compared to DHCA alone [53]. Expert consensus presented at the STS 2020 meeting recommended that if the circulatory arrest time is less than 30 min, ACP or RCP should be used, but that there is no consensus between DHCA + RCP and MHCA + ACP. If HCA time is greater than 30 min, ACP is recommended, but there is no consensus between DHCA + ACP and MHCA + ACP. Recent prospective randomized controlled trial data demonstrates that the incidence of subclinical neurologic events is likely much higher than currently estimated [44]. Further prospective studies with postoperative imaging are needed to truly determine the optimal cerebral protection strategy.

Funding

None.

Declarations

Ethical approval

This review complied with all institutional and national ethical guidelines for a review article. All reported data are from previously published studies and no new patient information or data was reported.

Informed consent

This study was analyzing previously published data and no new patient data was reported; as a result, patient consent was deemed waived.

Conflict of interest

The authors declare no competing interests.