Hindawi

Journal of Cardiac Surgery

Volume 2023, Article ID 3975367, 9 pages
https://doi.org/10.1155/2023/3975367

Review Article
The Association of the Cerebral Protection Strategy with Early
Mortality and Postoperative Stroke in Acute Type A Aortic
Dissection: A Systematic Review and Meta-Analysis

Caius Mustonen1 and Mikko Uimonen 1,2

1
Department of Surgery, Hospital Nova of Central Finland, Jyväskylä, Finland
2
Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland

Correspondence should be addressed to Mikko Uimonen; [email protected]

Received 9 January 2023; Revised 23 April 2023; Accepted 20 May 2023; Published 30 May 2023

Academic Editor: Carlos-A. Mestres

Copyright © 2023 Caius Mustonen and Mikko Uimonen. Tis is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Objective. Te optimal cerebral protection strategy in acute type A aortic dissection (ATAAD) is still without a clear consensus.
Te purpose of this meta-analysis was to compare the outcome of diferent cerebral protection strategies on ATAAD patients.
Materials and Methods. We conducted a systematic review including all studies concerning surgically managed ATAAD patients
between 1.1.2010 and 28.2.2022 and reporting the use of cerebral protection strategies in three large databases (Pubmed, Cochrane
library, and Scopus). Te main outcome events were 30-day mortality and a postoperative stroke rate. Te pooled event rates
adjusted by age, gender, CPB duration, circulatory arrest duration, and total arch reconstruction rate were calculated. Results.
Overall, 39 articles were included covering a total of 16, 876 ATAAD patients. Te estimated adjusted pooled early mortality rate
was 10.1% (95% confdence interval [CI] 9.1–11.3%) in the ACP group, 15.9% (13.3–18.9%) in the RCP group, and 11.6%
(9.2–14.5%) in the HCA group. Compared to the RCP group, ACP and HCA demonstrated lower early mortality (RCP vs. ACP
odds-ratio 1.66 [1.28–2.15], p < 0.001; RCP vs. HCA odds-ratio 1.45 [1.02–2.07], p � 0.039). Te adjusted pooled stroke rate was
9.0% (8.3–9.8%) in the ACP group, 10.5% (9.3–11.7%) in the RCP group, and 9.1% (8.1–10.2%) in the HCA group. Conclusion.
Early mortality might be more common in ATAAD patients treated with RCP compared to ACP and HCA. With regards to
postoperative stroke, the results were inconclusive despite the trending inferiority of RCP compared to the other strategies.

1. Introduction neuroprotective strategies as evidence on the optimal ce-

rebral protection strategy is scarce [2]. Te current guide-
Acute type A aortic dissection (ATAAD) is still associated lines suggest that the addition of either ACP or RCP during
with signifcant mortality and morbidity despite advances in the circulatory arrest is reasonable for ATAAD repair (IIa B)
diagnostics and surgical treatment. ATAAD requires and that circulatory arrest with ACP and moderate or deep
emergency surgery and reconstruction of the diseased aortic hypothermia is reasonable for ATAAD repair during ex-
section. Performing an open distal aortic anastomosis causes tended arch reconstruction (IIa, C) [3].
a substantial risk for ischemic cerebral events during the Since 2010, only four RCTs focusing on cerebral pro-
surgery. Cerebral protection strategies have evolved to tection in aortic arch surgery have been published, only one
mitigate cerebral ischemia during open distal anastomosis. of which [4] focuses on ATAAD patients, while others solely
Te main cerebral protection approaches, antegrade cerebral encompass elective arch reconstructions [5–7]. Terefore,
perfusion (ACP), retrograde cerebral perfusion (RCP), and the evidence on neuroprotective strategies in ATAAD pa-
deep hypothermic circulatory arrest alone (HCA), each have tients is almost exclusive from the retrospective studies and
specifc advantages and pitfalls [1]. Tus far, institutional as institutional series as well as from meta-analysis. Most of
well as surgeon preferences have guided the usage of these studies have heterogeneous patient samples involving
2 Journal of Cardiac Surgery

patients with acute dissections, chronic dissections, and trunk, right subclavian/axillary artery, right common carotid
degenerative aneurysms, even though mortality and stroke artery, left common carotid artery, or left subclavian/axillary
rates have been shown to difer substantially between these artery. In RCP, cerebral perfusion was used via the superior
patient groups [8–12]. Terefore, we conducted a meta- vena cava. In HCA only, no selective cerebral perfusion was
analysis comparing the outcomes of ACP, RCP, and HCA used during systemic circulatory arrest. ACP was further
in surgically treated ATAAD patients during the past and stratifed into unilateral (one-sided cerebral perfusion line)
current decades. and bilateral (two-sided cerebral perfusion via both right
and left-sided arteries) ACP.
2. Methods
We conducted a systematic review and meta-analysis of 2.6. Statistical Methods. A statistical analysis was performed
studies reporting the rates of early mortality and post- using R statistical software version 4.1.2 (R Project for
operative stroke after surgery for ATAAD. Tis review has Statistical Computing). Crude and adjusted pooled event
been reported according to the Preferred Reporting Items rates were calculated. Adjusted event rate per study was
for Systematic reviews and Meta-Analyses (PRISMA) estimated using Poisson regression with the observed event
statement (PRISMA checklist available in Supplementary 1). rate set as a dependent variable and mean age (or median age
if mean not reported), the proportion of males, mean car-
2.1. Search Strategy. We searched PubMed, Cochrane li- diopulmonary bypass time (or median age if mean not
brary, and Scopus on March 1st, 2022 using search phrases reported), mean circulatory arrest time (or median age if
presented in detail in Table 1. Te publication year was mean not reported), and proportion of total aortic arch
limited to 2010 or later. reconstruction set as independent variables. Function “glm”
from “stats” package version 4.1.2 was used. To calculate
adjusted event rates, missing covariate values were replaced
2.2. Inclusion Criteria. All original studies reporting out- using multiple imputations by chained equations using the
comes after surgical treatment for ATAAD with explicitly “mice” function from “mice” package version 3.14.0 [13].
reported cerebral protection strategies and early mortality A meta-analysis using a random intercept logistic re-
and postoperative stroke rates were included. We excluded gression model was conducted, and pooled crude and ad-
studies in which ATAAD patients and other patients were justed outcome event rates and 95% confdence intervals
pooled and studies in which patients with diferent cerebral (CI) were calculated. Te random efects model was used due
protection strategies were pooled, and studies in which only to the high heterogeneity between the studies (I2 > 40%).
pooled outcomes were there were reported. In addition, Inverse variance with a logit transformation was used as
non-English studies were excluded. a meta-analysis method. Odds ratios and 95% CIs for
outcome events between diferent cerebral protection
2.3. Main Outcome Measures. Te main outcome measures strategies were calculated. Te pooled outcome event rates
were early mortality and postoperative stroke. Early mor- were calculated using the function “metaprop” from “meta”
tality was defned as death within 30 days from surgery or package version 5.1 − 1. Lastly, a sensitivity analysis was
death during the primary in-hospital period after surgery. If conducted by systematically removing each study from the
both were reported in a study article, the one indicating meta-analysis one at a time to examine their potential in-
a higher event rate was selected for analysis. Postoperative fuence on the overall efect size estimate.
stroke was defned as a new permanent neurological defcit Te quality of the included studies was assessed using
occurring after the surgery. the Newcastle-Ottawa scale. Tis review has been registered
with PROSPERO (ID: CRD42022359545) and can be
accessed via https://www.crd.york.ac.uk/prospero/display_
2.4. Screening and Data Extraction. Each article was record.php?RecordID=359545.
screened, and the data of included studies were extracted by
a single author. Te following information was gathered:
author, number of patients, mean age, the proportion of 3. Results
males, cerebral protection strategy, postoperative stroke rate,
Te initial search retrieved a total of 2,110 articles. After
early mortality rate, cannulation site, hypothermia tem-
removing the duplicates, 1,817 articles were screened and 91
perature, mean cardiopulmonary bypass time, mean cir-
articles were selected for full-text assessment. Ultimately, 39
culatory arrest time, and proportion of total aortic arch
articles were included in the analysis (Figure 1). Te study
reconstruction. Te cerebral protection strategy was strati-
characteristics are presented in Supplementary 2. Te quality
fed into the following three categories: antegrade cerebral
of the included studies is presented in Supplementary 3. Te
perfusion (ACP), retrograde cerebral perfusion (RCP), and
included studies reported outcomes of 16,876 patients, of
hypothermic circulatory arrest only (HCA).
which 11,355 were operated on patients using ACP, 2,549
using RCP, and 2,972 using HCA (Table 2). Tree of the
2.5. Defnitions of Cerebral Protection Strategies. Te def- studies were prospective, while the others were retrospective
nition of ACP was that cerebral perfusion was used with an studies. Only one of the included studies was a randomized
arterial perfusion line located at the aorta, brachiocephalic controlled trial. In the ACP group, the mean age was lower
Journal of Cardiac Surgery 3

Table 1: Search strategy.

Aortic dissection [Title] AND (circulatory arrest [Title] OR hypotherm∗ [Title] OR
retrograde [Title] OR antegrade [Title] OR selective [Title]) OR (cerebral perfusion
PubMed
[Title] OR cerebral protection [Title] OR brain protection [Title])
Filter year 2010–2022
Aortic dissection AND (circulatory arrest OR hypotherm∗ OR retrograde OR
antegrade OR selective) OR (cerebral perfusion OR cerebral protection OR brain
Cochrane library
protection) in record title - (word variations have been searched)
Filter year 2010–2022
TITLE (“aortic dissection” AND (“circulatory arrest” OR hypotherm∗ OR
retrograde OR antegrade OR selective) OR (“cerebral perfusion” OR “cerebral
protection” OR “brain protection”)) AND (LIMIT-TO (PUBYEAR, 2022) OR
LIMIT-TO (PUBYEAR, 2021) OR LIMIT-TO (PUBYEAR, 2020) OR LIMIT-TO
Scopus (PUBYEAR, 2019) OR LIMIT-TO (PUBYEAR, 2018) OR LIMIT-TO (PUBYEAR,
2017) OR LIMIT-TO (PUBYEAR, 2016) OR LIMIT-TO (PUBYEAR, 2015) OR
LIMIT-TO (PUBYEAR, 2014) OR LIMIT-TO (PUBYEAR, 2013) OR LIMIT-TO
(PUBYEAR, 2012) OR LIMIT-TO (PUBYEAR, 2011) OR LIMIT-TO (PUBYEAR,
2010))
Te search was performed 1st March, 2022.

and the proportion of males was higher than in the other the groups (Figure 5 and Table 3). Te sensitivity analysis
groups. Te circulatory arrest time was shorter in the HCA indicated that the changes in the estimate ranged from
group when compared to the others. Te proportion of total negligible to small (crude stroke rate estimate range of
arch replacement was highest in the ACP group. 8.3–8.8%, tau range of 0.651–0.698, and I2 range of
83.6–86.1%).
Heterogeneity diminished after correcting for covariates
3.1. Early Mortality. Te pooled data contained a total of
in postoperative stroke rate calculations (I2 ≤ 25%, Figure 2).
2,212 early deaths (13.1% of all patients). According to I2
A meta-analysis of the adjusted postoperative stroke rates
values (≥75%) and the funnel plot of crude early mortality
showed a trending superiority of ACP (postoperative stroke
rates (Figure 2), heterogeneity was high between studies. Te
rate 9.0%, 95% CI 8.3–9.8%) in relation to RCP (10.5%, 95%
estimated crude pooled early mortality rate was 10.7% (95%
CI 9.3–11.7%; p � 0.057) (Figure 6 and Table 3). In the
CI 9.1–12.4%) in the whole population. A meta-analysis of
sensitivity analysis, removing any single study had only
the crude early mortality rate did not reveal eligible evidence
a minor efect on the overall estimate (adjusted stroke rate
between cerebral protection strategies (Figure 3 and Table 3).
estimate range of 9.0–9.3%, tau range of 0.192–0.226, and I2
Te sensitivity analysis indicated that no single study had
range of 26.1–38.0%).
a substantial infuence on the efect size estimate, as re-
moving any individual study led to only small changes in the
overall estimate (crude early mortality rate estimate range of 4. Discussion
10.5–11.1%, tau range of 0.54–0.60, and I2 range of
Te present systematic review and meta-analysis of three
83.2–84.9%).
main cerebral protection strategies demonstrate the out-
After correcting for the confounders, heterogeneity
comes of the two major complications after ATAAD sur-
between the studies decreased although residual heteroge-
gery, 30-day mortality, and postoperative stroke. To the best
neity was still high (I2 ≥ 58%, Figure 2). According to the
of our knowledge, this is the frst meta-analysis focusing
adjusted early mortality rates, RCP (early mortality rate
solely on ATAAD patients, and it is the frst to exclude older
15.9%, 95% CI 13.3–18.9%) was inferior to the other groups
studies published before the year 2010. A summary of the
(early mortality rate in ACP 10.1%, 95% CI 9.1–11.3%, and
previous meta-analyses can be founded in Supplementary 4.
p < 0.001 and in HCA 11.6%, 95% CI 9.2–14.5%, and
In this meta-analysis, the observed early mortality and
p � 0.039; Figure 4 and Table 3). Te sensitivity showed no
postoperative stroke rates were 13.1% and 10.0%, re-
major infuence on estimates after removing any single study
spectively, which were slightly lower compared to the early
(adjusted early mortality rate estimate range of 10.2–10.6%,
mortality rate of 17% and postoperative stroke rate of
tau range of 0.307–0.356, and I2 range of 58.1–67.1%).
10–20% previously reported in the three large multicenter
retrospective ATAAD patient registers (IRAD, NORCAAD,
3.2. Postoperative Stroke. A total of 1,695 (10.0% of all pa- and GERAADA) [10–12, 14–16]. One explanation may be
tients) postoperative strokes were observed in the pooled the inclusion of single-arm studies from highly specialized
data. High heterogeneity was observed between studies high-volume centers.
(I2 ≥ 64%, Figure 2). Te pooled postoperative stroke rate In the present meta-analysis, an eligible piece of evidence
estimate was 8.5% (7.0–10.2%) in the whole population, and of a diference between the cerebral protection strategies in
estimates calculated separately in the cerebral protection the crude early mortality or postoperative stroke rate was not
strategy groups did not indicate clear diferences between found. However, the comparison of the adjusted early
4 Journal of Cardiac Surgery

Identification of studies via databases and registers

Identification
Studies identified from: Studies removed before
PubMed (n = 1519) screening:
Cochrane library (n = 392) Duplicate records removed
Scopus (n = 199) (n = 293)

Title & abstract screened Studies excluded by title &

(n = 1817) abstract (n = 1726)

Studies sought for retrieval Studies not retrieved

(n = 91) (n = 4)
Screening

Full text assessed for eligibility Studies excluded:

(n = 87) Wrong or mixed patient group
(n = 40)
Cerebral protection strategy
inadequately reported (n = 7)
Wrong outcome variables
(n = 1)
Included

Studies included in meta-

analysis
(n = 39)

Figure 1: PRISMA fow chart of study screening.

mortality rates revealed that RCP might be inferior to ACP in elective aortic surgery, probably due to the meta-analyses
and HCA. In the light of two recent meta-analyses of aortic showing the superiority of cerebral perfusion over HCA [2].
arch surgery patients, this was rather unexpected, as con- Currently, only a few actively publishing centers utilize HCA
tradictory to our fndings, those concluded that HCA may be as their main cerebral protection strategy and their results
associated with higher early mortality and postoperative seem to be in line with the current data [8, 19]. In ATAAD
stroke rates than ACP or RCP [17, 18]. Te discrepancy may surgery, ACP and HCA have been shown to produce equal
relate to the inclusion of elective patients in those previous results regarding early mortality and postoperative stroke
meta-analyses. Compared to ATAAD surgery, cerebral with circulatory arrest durations under 25–30 minutes [20].
perfusion (either antegrade or retrograde) is more regularly Tis duration is considered sufcient for hemiarch re-
utilized in elective aortic arch reconstructions and the placement, which accounts for approximately 80% of pri-
complication rate is also substantially lower in elective de- mary surgeries. More complex arch surgeries require longer
generative aneurysms and chronic dissections [8, 9]. arrest time, and therefore, the use of cerebral perfusion is
Terefore, including elective patients with presumably a necessity. Tis was also seen in the present meta-analysis as
higher rates of cerebral perfusion and concurrently lower the rate of total arch replacement was 48% in the ACP group
complication risks potentially leads to erroneous disfavoring and 15% in the HCA group.
of HCA when considering ATAAD patients. RCP was the least used cerebral protection strategy in
Indeed, according to the results of this meta-analysis, this meta-analysis. Unexpectedly, it had the longest mean
HCA resulted in closely equal rates of early mortality and durations in both cardiopulmonary bypass and circulatory
postoperative stroke as ACP in both crude and adjusted arrest (35 minutes) as well as the lowest circulatory arrest
analyses. Nonetheless, the use of HCA in ATAAD surgery temperatures (20°C) and the lowest frequency of total arch
has been diminishing similarly during the last two decades as reconstructions (6%). Te combination of a low frequency of
 Journal of Cardiac Surgery

Table 2: Patient characteristics.

Antegrade cerebral perfusion Retrograde cerebral perfusion Hypothermic circulatory arrest
(n � 11, 355) (n � 2, 549) only (n � 2, 972)
Age, mean∗ 56 61 61
Male, n (%) 7, 845 (69) 1, 647 (65) 1, 915 (64)
Cardiopulmonary bypass time, minutes, mean∗ 190 201 185
Circulatory arrest time, minutes, mean∗ 32 35 24
Hypothermia temperature, mean∗ 24 20 25
Total arch reconstructed, n (%) 5, 551 (49) 153 (6) 447 (15)
∗
Estimated as a patient count weighted mean of reported mean/median values of included studies.
5
6 Journal of Cardiac Surgery

0.0
0.2
0.2
Standard Error

Standard Error
0.6 0.4
0.6
1.0
0.8
1.4 1.0

-5 -4 -3 -2 -1 0 -4 -3 -2 -1 0
Logit Transformed Proportion Logit Transformed Proportion
(a) (b)
0.0
0.2
0.2
Standard Error

Standard Error
0.6 0.4
0.6
1.0
0.8
1.4 1.0
-5 -4 -3 -2 -1 0 -4 -3 -2 -1
Logit Transformed Proportion Logit Transformed Proportion
(c) (d)

Figure 2: Funnel plots to assess the risk for publication bias and systematic heterogeneity of proportions of outcome events. Black diagonal
lines show 95% confdence interval, and the dark grey area shows 99% confdence interval. (a) � Crude early mortality rate, (b) � adjusted
early mortality rate, (c) � crude stroke rate, and (d) � adjusted stroke rate.

Observed Pooled estimate

Cerebral protection strategy Patients Estimate [95% CI]
early deaths for early mortality

Antegrade cerebral perfusion

1346 11355 0.103 [0.086; 0.122]
Heterogeneity: I2 = 85%, p < 0.01
Hypothermic circulatory arrest only
491 2972 0.095 [0.046; 0.185]
Heterogeneity: I2 = 83%, p < 0.01
Retrograde cerebral perfusion
375 2549 0.128 [0.096; 0.168]
Heterogeneity: I2 = 75%, p < 0.01

Overall 2212 16876 0.107 [0.091; 0.124]

0 0.1 0.2 0.3 0.4

Figure 3: Comparison of crude pooled early mortality by using the cerebral protection strategy.

Table 3: Pairwise comparisons between cerebral protection strategies.

HCA vs. RCP vs. RCP vs.
ACP OR p ACP OR p HCA OR p
(95% CI) (95% CI) (95% CI)
Early mortality
Crude 1.17 (0.68–1.99) 0.574 1.20 (0.75–1.92) 0.440 1.03 (0.54–1.97) 0.928
Adjusted 1.14 (0.85–1.54) 0.377 1.66 (1.28–2.15) <0.001 1.45 (1.02–2.07) 0.039
Postoperative stroke
Crude 1.22 (0.67–2.21) 0.515 1.02 (0.59–1.75) 0.956 0.83 (0.40–1.73) 0.625
Adjusted 0.99 (0.81–1.20) 0.894 1.20 (1.00–1.44) 0.057 1.21 (0.95–1.55) 0.122
ACP � antegrade cerebral perfusion, HCA � hypothermic circulatory arrest, RCP � retrograde cerebral perfusion, OR � odds-ratio, and CI � confdence
interval.
Journal of Cardiac Surgery 7

Adjusted Pooled estimate

Cerebral protection strategy Patients Estimate [95% CI]
early deaths for early mortality

Antegrade cerebral perfusion

1268 11355 0.102 [0.091; 0.114]
Heterogeneity: I2 = 71%, p < 0.01
Hypothermic circulatory arrest only
409 2972 0.113 [0.089; 0.142]
Heterogeneity: I2 = 64%, p < 0.01
Retrograde cerebral perfusion
406 2549 0.156 [0.128; 0.187]
Heterogeneity: I2 = 62%, p < 0.01

Overall 2083 16876 0.111 [0.100; 0.122]

0.05 0.1 0.15 0.2 0.25 0.3

Figure 4: Comparison of adjusted pooled early mortality by using the cerebral protection strategy.

Observed Pooled estimate

Cerebral protection strategy Patients Estimate [95% CI]
postoperative strokes for postoperative strokes

Antegrade cerebral perfusion

1067 11355 0.082 [0.064; 0.103]
Heterogeneity: I2 = 88%, p < 0.01
Hypothermic circulatory arrest only
379 2972 0.102 [0.065; 0.156]
Heterogeneity: I2 = 77%, p < 0.01
Retrograde cerebral perfusion
249 2549 0.090 [0.069; 0.117]
Heterogeneity: I2 = 64%, p < 0.01

Overall 1695 16876 0.085 [0.070; 0.102]

0 0.05 0.1 0.15 0.2

Figure 5: Comparison of crude pooled postoperative stroke by using the cerebral protection strategy.

Adjusted Pooled estimate

Cerebral protection strategy Patients Estimate [95% CI]
postoperative strokes for postoperative strokes

Antegrade cerebral perfusion

1024 11355 0.090 [0.083; 0.098]
Heterogeneity: I2 = 29%, p = 0.04
Hypothermic circulatory arrest only
275 2972 0.093 [0.083; 0.104]
Heterogeneity: I2 = 0%, p = 0.89
Retrograde cerebral perfusion
271 2549 0.107 [0.096; 0.120]
Heterogeneity: I2 = 0%, p = 0.77

Overall 1570 16876 0.093 [0.087; 0.100]

0 0.05 0.1 0.15 0.2

Figure 6: Comparison of adjusted pooled postoperative stroke by using the cerebral protection strategy.

total arch operations and relatively long average durations of chronic dissections even though the emergency status and
circulatory arrest for hemiarch surgeries may indicate that it therefore the possibility to plan the operation in advance as
was used in less experienced aortic centers although this well as risk status of the patients have substantial diferences.
hypothesis could not be further assessed in this meta- Te diferences in patient groups refect in the outcomes of
analysis. However, it has been shown that the cerebral surgery as early mortality and postoperative stroke rate
capillary oxygen supply using RCP is insufcient which may located around 17% and 10–20% in ATAAD [10–12, 14–16]
also contribute to the inferior outcomes of RCP [21]. compared to 1–5% early mortality and 1–5% stroke rate of
Te newly founded guidelines on ATAAD surgery ad- degenerative aneurysms and chronic dissections [8, 9, 22].
vocate using cerebral perfusion in ATAAD patients [3]. With such fundamental issues, the validity of these guide-
However, these recommendations are based on studies that lines should be critically assessed. To patch this faw, well-
include, besides the ATAAD patients, a signifcant pro- designed randomized controlled studies with sufcient
portion of patients with degenerative aneurysms and statistical power focusing solely on ATAAD patients are still
8 Journal of Cardiac Surgery

needed before recommending against the usage of HCA as Conflicts of Interest

a sole cerebral protection strategy in ATAAD surgery.
Te authors declare that they have no conficts of interest.
4.1. Strengths and Limitations. Te main strength of this
Authors’ Contributions
meta-analysis was the focus on ATAAD patients, which
provided a more accurate estimation of outcomes in these Caius Mustonen and Mikko Uimonen were responsible for
patients when compared to the previous meta-analyses with the conceptualization, methodology, project administration,
heterogeneous patient samples. All studies reporting early data curation, roles or writing the original draft, reviewing,
mortality and postoperative stroke rates after surgery for and editing. Mikko Uimonen was also responsible for formal
ATAAD between 2010 and 2022 were included, and no study analysis and visualization.
was excluded by design or study setting. Tis provided a high
number of patients in each cerebral protection strategy Supplementary Materials
group with an urge for a pragmatic approach and modern
treatment practices. As a counterpart, the inclusion of both Supplementary 1: the Preferred Reporting Items for Sys-
observational and experimental studies may cause bias in the tematic reviews and Meta-Analyses (PRISMA) checklist.
outcome event estimates due to selection bias and other Supplementary 2: characteristics of the included studies.
artifcial determinants of surgical decision-making in ex- Supplementary 3: quality of the included studies by using the
perimental studies. However, the inclusion of all studies Newcastle-Ottawa scale. Supplementary 4: published prior
enables catching the whole spectrum of various clinical meta-analyses on cerebral protection with HCA, ACP,
conditions related to diferent cerebral protection strategies and RCP. (Supplementary Materials)
during the surgery for ATAAD patients. Furthermore, along
with the crude estimation, the covariate-corrected estima- References
tion of outcome event rates provided further insight into
outcomes related to cerebral protection strategies. To the [1] M. P. Falasa, G. J. Arnaoutakis, G. M. Janelle, and
T. M. Beaver, “Neuromonitoring and neuroprotection ad-
best of the authors’ knowledge, no such adjustments have
vances for aortic arch surgery,” JTCVS Techniques, vol. 7,
been made in the previous meta-analyses on the feld. With p. 11, 2021.
the more or less inevitable diferences in patient charac- [2] R. De Paulis, M. Czerny, L. Weltert et al., “Current trends in
teristics and surgical techniques used, correction for known cannulation and neuroprotection during surgery of the aortic
covariates related to the outcomes improves the precision of arch in Europe,” European Journal of Cardio-Toracic Surgery,
estimation and thereby improves the usability of the fnd- vol. 47, no. 5, pp. 917–923, 2015.
ings. Tis was also seen in the decreased heterogeneity [3] S. C. Malaisrie, W. Y. Szeto, M. Halas et al., “2021 the
between the studies after correcting the outcome event rates American Association for Toracic Surgery expert consensus
for the covariates. However, it should be acknowledged that document: surgical treatment of acute type A aortic dissec-
in addition to the covariates used in the adjustments, there tion,” Te Journal of Toracic and Cardiovascular Surgery,
are also other factors that may have afected the outcomes. vol. 162, no. 3, pp. 735–758.e2, 2021.
[4] X. Sun, H. Yang, X. Li et al., “Randomized controlled trial of
Since the raw data of the included studies were not available
moderate hypothermia versus deep hypothermia anesthesia
and the data on the covariates were gathered from the study on brain injury during Stanford A aortic dissection surgery,”
articles, other confounding variables than those we con- Heart and Vessels, vol. 33, no. 1, pp. 66–71, 2018.
trolled could not be accounted for in the adjusted analysis, [5] L. G. Svensson, E. H. Blackstone, C. Apperson-Hansen et al.,
leading to residual confounding within the estimates. “Implications from neurologic assessment of brain protection
Terefore, the analyses are mostly exploratory in nature. for total arch replacement from a randomized trial,” Te
Journal of Toracic and Cardiovascular Surgery, vol. 150, no. 5,
5. Conclusions pp. 1140–1147.e11, 2015.
[6] O. V. Kamenskaya, A. S. Klinkova, A. M. Chernyavsky,
Early mortality might be more common in ATAAD patients V. V. Lomivorotov, I. O. Meshkov, and A. M. Karaskov,
treated with RCP compared to ACP and HCA. With regards “Deep hypothermic circulatory arrest vs. Antegrade cerebral
to postoperative stroke, the results were inconclusive despite perfusion in cerebral protection during the surgical treatment
the trending inferiority of RCP compared to the other of chronic dissection of the ascending and arch aorta,” Journal
of Extra-Corporeal Technology, vol. 49, no. 1, p. 16, 2017.
strategies. A well-designed randomized controlled trial with
[7] B. G. Leshnower, S. Rangaraju, J. W. Allen, A. Y. Stringer,
high statistical power or largely registered studies solely with T. G. Gleason, and E. P. Chen, “Deep hypothermia with
ATAAD patients is still needed before declaring the supe- retrograde cerebral perfusion versus moderate hypothermia
riority or inferiority of cerebral protection strategies. with antegrade cerebral perfusion for arch surgery,” Te
Annals of Toracic Surgery, vol. 107, no. 4, pp. 1104–1110,
Data Availability 2019.
[8] B. A. Ziganshin, B. G. Rajbanshi, M. Tranquilli, H. Fang,
Te data underlying this article are included in the sup- J. A. Rizzo, and J. A. Elefteriades, “Straight deep hypothermic
plementary materials. circulatory arrest for cerebral protection during aortic arch
Journal of Cardiac Surgery 9

surgery: safe and efective,” Te Journal of Toracic and

Cardiovascular Surgery, vol. 148, no. 3, pp. 888–900, 2014.
[9] O. Preventza, J. S. Coselli, A. Garcia et al., “Moderate hy-
pothermia at warmer temperatures is safe in elective proximal
and total arch surgery: results in 665 patients,” Te Journal of
Toracic and Cardiovascular Surgery, vol. 153, no. 5,
pp. 1011–1018, 2017.
[10] L. O. Conzelmann, E. Weigang, U. Mehlhorn et al., “Mortality
in patients with acute aortic dissection type A: analysis of pre-
and intraoperative risk factors from the German Registry for
Acute Aortic Dissection Type A (GERAADA),” European
Journal of Cardio-Toracic Surgery, vol. 49, no. 2, pp. e44–e52,
2016.
[11] R. A. Chemtob, S. Fuglsang, A. Geirsson et al., “Stroke in acute
type A aortic dissection: the nordic consortium for acute type
A aortic dissection (NORCAAD),” European Journal of
Cardio-Toracic Surgery, vol. 58, no. 5, pp. 1027–1034, 2020.
[12] A. Evangelista, E. M. Isselbacher, E. Bossone et al., “Insights
from the international registry of acute aortic dissection: a 20-
year experience of collaborative clinical research,” Circulation,
vol. 137, no. 17, pp. 1846–1860, 2018.
[13] S. van Buuren and K. Groothuis-Oudshoorn, “Mice: multi-
variate imputation by chained equations in R,” Journal of
Statistical Software, vol. 45, no. 3, pp. 1–67, 2011.
[14] L. O. Conzelmann, I. Hofmann, M. Blettner et al., “Analysis
of risk factors for neurological dysfunction in patients with
acute aortic dissection type A: data from the German Registry
for Acute Aortic Dissection Type A (GERAADA),” European
Journal of Cardio-Toracic Surgery, vol. 42, no. 3, pp. 557–565,
2012.
[15] A. Geirsson, K. Shioda, C. Olsson et al., “Diferential out-
comes of open and clamp-on distal anastomosis techniques in
acute type A aortic dissection,” Te Journal of Toracic and
Cardiovascular Surgery, vol. 157, no. 5, pp. 1750–1758, 2019.
[16] M. Di Eusanio, H. J. Patel, C. A. Nienaber et al., “Patients with
type A acute aortic dissection presenting with major brain
injury: should we operate on them?” Te Journal of Toracic
and Cardiovascular Surgery, vol. 145, no. 3, pp. S213–S221.e1,
2013.
[17] I. Hameed, M. Rahouma, F. M. Khan et al., “Cerebral pro-
tection strategies in aortic arch surgery: a network meta-
analysis,” Te Journal of Toracic and Cardiovascular Sur-
gery, vol. 159, no. 1, pp. 18–31, 2020.
[18] D. Abjigitova, K. M. Veen, G. van Tussenbroek et al., “Ce-
rebral protection in aortic arch surgery: systematic review and
meta-analysis,” Interactive Cardiovascular and Toracic
Surgery, vol. 35, no. 3, 2022.
[19] M. Jormalainen, P. Raivio, C. Mustonen et al., “Direct aortic
versus peripheral arterial cannulation in surgery for type A
aortic dissection,” Te Annals of Toracic Surgery, vol. 110,
no. 4, pp. 1251–1258, 2020.
[20] T. Krüger, E. Weigang, I. Hofmann, M. Blettner, and
H. Aebert, “Cerebral protection during surgery for acute
aortic dissection type A,” Circulation, vol. 124, no. 4,
pp. 434–443, 2011.
[21] M. P. Ehrlich, C. Hagl, J. N. McCullough et al., “Retrograde
cerebral perfusion provides negligible fow through brain
capillaries in the pig,” Te Journal of Toracic and Cardio-
vascular Surgery, vol. 122, no. 2, pp. 331–338, 2001.
[22] Y. Iba, K. Minatoya, H. Matsuda et al., “Contemporary open
aortic arch repair with selective cerebral perfusion in the era of
endovascular aortic repair,” Te Journal of Toracic and
Cardiovascular Surgery, vol. 145, no. 3, pp. S72–S77, 2013.