Comparing unilateral and bilateral cerebral perfusion during total arch replacement for acute type A aortic dissection Authors: Yumeng Ji, Juntao Qiu, Kai Zhang, Huiyuan Qin, Jian Song, Chenyu Zhou, Shiqi Gao, Bin Hou, Fangfang Cao, Ji Wang, Cuntao Yu, Wei Gao Published: 2024 ABSTRACT AbstractOBJECTIVESTo assess the effects of unilateral versus bilateral antegrade cerebral perfusion (u-ACP vs. b-ACP) on postoperative complications and mid-term follow-up results in Asian patients with acute type A aortic dissection (ATAAD) undergoing total arch replacement (TAR) + the frozen elephant trunk (FET).METHODSClinical baseline data and postoperative complications of 702 ATAAD patients undergoing TAR + FET at China Cardiovascular Centre Fuwai Hospital between January 2019 and December 2022 were collected. Patients were categorized into two groups based on ACP: unilateral (n = 402) and bilateral (n = 300). Propensity scores were matched for 217 pairs of patients. Mid-term and perioperative surgical complications were analysed, and baseline characteristics between the two groups were assessed.RESULTSThe matched cohort revealed comparable 30-day mortality rates between u-ACP and b-ACP: 4.15% and 3.23%, respectively (P = 0.61). Rates of postoperative permanent neurologic dysfunction (PND) were similar across groups (2.76% for u-ACP and 3.23% for b-ACP, P = 0.76). However, u-ACP exhibited a higher incidence of postoperative transient neurologic deficit (TND) at 14.29% compared to 6.91% with b-ACP (P = 0.01). Logistic regression models identified u-ACP and cardiopulmonary bypass (CPB) time as independent risk factors for TND. Mid-term survival rates were comparable between the two groups (5-year survival: 90.29% vs. 93.67%, P = 0.133).CONCLUSIONSBoth u-ACP and b-ACP are effective brain protection techniques for ATAAD patients undergoing TAR. The use of b-ACP significantly reduces the incidence of TND.Keywords: acute type A aortic dissection, antegrade cerebral perfusion, total arch replacement, unilateral and bilateral antegrade cerebral perfusion INTRODUCTION The acute condition known as acute type A aortic dissection (ATAAD) requires urgent surgical intervention and is associated with substantial mortality rates (22%) [1] and neurological complications (5–15%) [2–4]. Prolonged circulatory arrest (CA) time, often necessary during surgical repair, correlates with increased mortality and neurological sequelae, leading to extended hospitalizations and intensive care unit (ICU) stays [5]. Effective cerebral protection strategies during CA are critical as they directly influence the occurrence and prognosis of neurological complications. The introduction of selective cerebral perfusion combined with moderate hypothermic circulatory arrest (MHCA) has become standard practice. However, there remains ongoing debate regarding the optimal use of bilateral antegrade cerebral perfusion (b-ACP) versus unilateral antegrade cerebral perfusion (u-ACP) [6–8]. The choice between u-ACP and b-ACP involves balancing the simplicity and speed of u-ACP against the superior cerebral perfusion capability of b-ACP. Our objective was to evaluate the initial and intermediate outcomes of unilateral versus bilateral cerebral perfusion in ATAAD patients undergoing total arch replacement (TAR) + the frozen elephant trunk (FET). METHODS The study received approval from the ethical review board of Fuwai Hospital of the Chinese Academy of Medical Sciences (No. 2023–2084). Since this study is retrospective, patients’ consent was waived. Study population Between October 2019 and November 2023, a total of 702 patients diagnosed with ATAAD underwent TAR + FET under moderate hypothermia at Fuwai Hospital of the Chinese Academy of Medical Sciences in Beijing, China. Among them, 402 patients underwent u-ACP, while 300 patients received b-ACP. The surgeries were performed by 10 experienced aortic surgeons. These surgeons had performed over 50 surgeries for diagnosed TAAD. Six surgeons consistently utilized u-ACP, while the remaining four utilized b-ACP, irrespective of preoperative cerebral perfusion status. We identified and characterized the cohort, gathering preoperative, operative, and postoperative data from a prospectively maintained database. Surgical indications for TAR were consistent with Sun et al.’s study [9]. Exclusion criteria included patients with chronic aortic dissection and aortic aneurysms (Fig. 1). Permanent neurologic dysfunction (PND) was defined as new-onset postoperative ischaemic or haemorrhagic stroke with computed tomography scan and diagnosed by neurological physician. Transient neurologic deficit (TND), as defined by Ergin’s study [10], encompassed symptoms of confusion, lethargy or agitation for more than 48 h, without radiographic evidence, resolving completely before discharge. ‘Cerebral malperfusion’ referred to dissection involving unilateral or bilateral carotid arteries on imaging, accompanied by preoperative clinical symptoms such as transient syncope, coma, or hypersomnia. Flowchart summary of the study cohort. b-ACP: bilateral antegrade cerebral perfusion; TAR: total arch replacement; u-ACP: unilateral antegrade cerebral perfusion Surgical strategy In surgery, we employed ACP, MHCA (24.1°C–28°C), measured through the nasopharynx, and total cardiopulmonary bypass (CPB). A standardized surgical protocol was followed for ATAAD treatment. CPB was primarily conducted through single axillary artery cannulation or dual cannulation of the femoral and axillary arteries. Intraoperative near-infrared spectroscopy (NIRS) monitored cerebral oxygenation in both brain hemispheres. Right axillary and femoral artery cannulation were employed for CPB. Upon reaching a nasopharyngeal temperature of 32°C, the ascending aorta was clamped, CPB initiated, and cooling continued. Cold-blood cardioplegic solution was directly infused into the left and right coronary ostia following longitudinal opening of the proximal ascending aorta. To prevent cerebral perfusion steal syndrome, lower body circulation was halted by clamping the distal end of the supra-arch vessels when the nasopharyngeal temperature reached 25°C. In the u-ACP group, cerebral perfusion was directed via the right axillary or innominate artery, while b-ACP involved an additional perfusion catheter inserted into the left common carotid artery [11]. Selective cerebral perfusion was maintained at a flow rate of approximately 12–15 ml/kg/min, with a target cerebral perfusion pressure of 60 mmHg. Cerebral saturation was typically maintained at or above 60%. The FET (MicroPort Medical Co., Ltd, Shanghai, China) was inserted, and the distal aortic stump was anastomosed to a four-branched graft (Meadox Hemashield Platinum 4 Branch Graft, Boston Scientific Inc., Marlborough, Mass). Lower body circulation was then restored, and air evacuated from the artificial graft. Subsequently, the left common carotid artery was anastomosed to restore cerebral perfusion, followed by anastomosis of the remaining supra-aortic vessels to the branches of the graft [12]. Statistical analysis Propensity score matching (PSM) was employed to minimize the impact of confounding variables and selection bias in our study, which includes age, sex, creatinine, operation time, CPB time, cross-clamp time, and CA time (Supplementary Fig. 1). The matching was performed using the nearest neighbour algorithm with a 1:1 ratio and a calliper width of 0.2 standard deviations of the logit of the propensity score to minimize bias. Standardized mean differences (SMDs) were calculated for covariates, with SMDs below 0.1 indicating acceptable balance between groups. Patient characteristics were described using appropriate measures such as median (interquartile range), mean ± SD, or median for quantitative variables, based on the results of the Shapiro–Wilk normality test. Variables that did not meet the normal distribution criteria were analysed with non-parametric methods. Specifically, the chi-square test was used to evaluate associations between categorical variables and Fisher’s exact test was used as a more accurate alternative if the sample size was too small (n < 40 or expectation < 1). The Kruskal–Wallis test, followed by Dunn’s post-hoc tests for multiple comparisons, was applied to assess differences between groups for continuous variables. We employed mean imputation as handling missing data strategy. Data analysis was conducted using the industry-standard software package SPSS (IBM Corp, Chicago, IL, USA). We assessed the assumption of proportional hazards, and Kaplan–Meier curves were used to compare outcomes between the u-ACP and b-ACP groups. A significance level of P < 0.05 was considered statistically significant. RESULTS Preoperative data The study encompassed 702 patients, of whom 540 (76.9%) were male. The average age of the patients was 50.27 years. b-ACP was utilized in 300 patients, constituting 42.74% of the cohort, while u-ACP was performed in 402 patients, comprising 57.26%. The incidence of preoperative poor cerebral perfusion is 5.27%. There were no instances of conversion from u-ACP to b-ACP during surgery. Both u-ACP and b-ACP groups demonstrated similar demographics and preoperative characteristics. However, the u-ACP group exhibited a significantly higher creatinine level (87.3 vs. 76.54 µmol/L, P = 0.001). Following PSM, the two groups showed high comparability (Table 1). Table 1:Preoperative characteristicsPreoperative characteristicsAllMatchedVariablesTotal (n = 702)u-ACP (n = 402)b-ACP (n = 300) P-valueu-ACP (n = 217)b-ACP (n = 217) P-valueAge (years), mean ± SD50.27 ± 11.3349.82 ± 11.3850.8 ± 11.280.06550.3 ± 10.7750.39 ± 11.280.68Males, n (%)540 (76.9%)316 (78.6%)224 (74.7%)0.22160 (73.73%)172 (79.2%)0.17Smoking, n (%)243 (34.6)138 (34.3%)105 (35%)0.85383 (38.25%)81 (37.33%)0.84Alcohol, n (%)46 (6.6%)29 (7.2%)17 (5.7%)0.41919 (8.76%)15 (6.91%)0.48Hypertension, n (%)569 (81.1%)319 (79.4%)250 (83.3%)0.183177 (81.57%)183 (84.33%)0.44Diabetes mellitus (%)39 (5.6%)24 (6.0%)15 (5%)0.57912 (5.53%)10 (4.61%)0.66CAD, n (%)69 (9.8%)32 (8%)37 (12.3%)0.05218 (8.29%)24 (11.06%)0.32Shock, n (%)6 (0.85%)4 (0.99%)2 (0.66%)0.342 (0.92%)1 (0.46%)0.46Tamponade (%)7 (0.99%)5 (1.24%)2 (0.66%)0.1762 (0.92%)00.23Cerebral malperfusion, n (%)37 (5.27%)21 (5.22%)16 (5.33%)0.3713 (5.99%)10 (4.61%)0.24Ischaemic stroke, n (%)32 (4.6%)16 (4.0%)16 (5.3%)0.39510 (4.61%)13 (5.99%)0.52Cerebral haemorrhage, n (%)6 (0.9%)3 (0.7%)3 (1%)0.7043 (1.38%)1 (0.46%)0.62Marfan syndrome, n (%)22 (3.1%)12 (3%)10 (3.3%)0.79312 (2.8%)5 (2.3%)0.56Pericardial effusion, n (%)133 (19%)73 (18.2%)60 (20%)0.8242 (19.35%)45 (20.74%)0.9Creatinine (µmol/L), mean ± SD82.68 (66.57, 103)87.3 (68.42, 106)76.54 (63.6, 99.39) 0.001* 87.53 ± 28.0288.5 ± 36.290.83Ejection fraction (%), mean ± SD60.32 ± 4.9560.12 ± 4.6659.87 ± 5.260.58560.87 ± 4.7560.04 ± 5.570.26Haemoglobin (g/L), mean ± SD130.73 ± 18.51131.84 ± 18.01129.4 ± 19.040.225131.16 ± 18.52130.06 ± 18.980.83Platelets (×109), mean ± SD192.88 ± 75.19186.15 ± 70.02200.98 ± 80.410.183189.49 ± 74.26202.73 ± 83.980.28Glucose (mmol/L), median (IQR)6.85 (5.88, 7.93)7.01 (6.09, 8.09)6.68 (5.76, 7.68)0.0747.2 ± 2.316.82 ± 1.520.36BUN (mmol/L), median (IQR)6.68 (5.36, 8.53)6.82 (5.5, 8.62)6.57 (5.29, 8.43)0.7276.87 ± 2.257.31 ± 3.050.08Albumin (g/L), mean ± SD38.44 ± 4.5538.87 ± 4.6637.92 ± 4.370.05538.78 ± 4.0638.32 ± 4.330.06Lactate (mmol/L), median (IQR)1.5 (1.1, 2.14)1.5 (1.1, 2.06)1.5 (1.07, 2.32)0.6081.7 ± 0.941.79 ± 0.940.42Onset consultation time (hours), median (IQR)10 (5, 18)10 (5, 16)10 (5, 16)0.65710 (5, 18)10 (5.38, 20)0.91Onset surgery time (hours) malperfusion25 (10.5, 42)25.2 (10, 38)25 (10.5, 43)0.6725 (10, 40)25 (11, 43)0.87 Coronary malperfusion16 (2.27%)10 (2.48%)6 (2%)0.424 (1.84%)3 (1.38%)0.38 Visceral malperfusion75 (10.68%)41 (10.2%)34 (11.33%)0.3420 (9.21%)22 (10.14)0.78 Limb malperfusion40 (5.7%)23 (5.72%)17 (5.67%)0.6514 (6.45%)10 (4.6%)0.24CAD: coronary artery disease; BUN: blood urea nitrogen; SD: standard deviation; IQR: interquartile range; u-ACP: unilateral antegrade cerebral perfusion; b-ACP: bilateral antegrade cerebral perfusion; Bold font: p-values less than 0.05. Preoperative characteristics CAD: coronary artery disease; BUN: blood urea nitrogen; SD: standard deviation; IQR: interquartile range; u-ACP: unilateral antegrade cerebral perfusion; b-ACP: bilateral antegrade cerebral perfusion; Bold font: p-values less than 0.05. Intraoperative data Cross-clamp time, nasopharyngeal temperature, rectal temperature, cooling time, units of transfused platelets, and cerebral perfusion flow differed prior to matching. Using the PSM method, 217 matched pairs of patients were derived, effectively addressing most discrepancies. Although a statistical difference persisted in nasopharyngeal and rectal temperatures between the two groups post PSM (P = 0.01), the absolute temperature differences were minimal and lacked practical significance (24.52 ± 1.56°C vs. 25.91 ± 1.81°C, 26.63 ± 1.93°C vs. 27.7 ± 1.64°C; Table 2). Table 2:Perioperative characteristicsPerioperative characteristicsAllMatchedVariablesTotal (n = 702)u-ACP (n = 402)b-ACP (n = 300) P-valueu-ACP (n = 217)b-ACP (n = 217) P-valueBentall operation, n (%)166 (23.6%)97 (24.1%)69 (23%)0.37739 (17.97%)47 (21.66%)0.63Wheat operation, n (%)13 (1.9%)5 (1.2%)8 (2.7%)0.3764 (1.84%)5 (2.3%)0.34David operation, n (%)30 (4.3%)20 (5.0%)10 (3.3%)0.56511 (5.07%)9 (4.15%)0.28CABG, n (%)165 (23.5%)92 (22.9%)73 (24.3%)0.65545 (20.74%)48 (22.12%)0.73CPB Cannulation, n (%)0.1320.16 Axillary, n (%)338 (48.15%)150 (39.8%)188 (62.67%)80 (36.87%)89 (41.01%) Femoral + axillary, n (%)464 (51.85%)233 (57.96%)231 (77%)137 (63.13%)128 (58.99%)CPB time (min), mean ± SD189 ± 69.77198.17 ± 64.09181.66 ± 79.750.152197.48 ± 79.84195.95 ± 70.270.87Cross-clamp time (min), mean ± SD118.47 ± 51.69126.14 ± 52.9113.16 ± 48.06 0.016* 121.49 ± 43.37120.98 ± 46.40.54Circulatory arrest time (min), mean ± SD16.42 ± 8.518.84 ± 9.1914.22 ± 6.830.11116.72 ± 5.6216.21 ± 6.920.77Nasopharyngeal temperature (°C), mean ± SD24.74 ± 1.724.53 ± 1.5125.14 ± 1.98 0.005* 24.52 ± 1.5625.91 ± 1.81 0.01* Rectal temperature (°C), mean ± SD26.68 ± 1.8126.41 ± 1.8027.21 ± 1.73 0.001* 26.63 ± 1.9327.7 ± 1.64 0.01* Cooling time (min), mean ± SD103.35 ± 40.62107.21 ± 42.0895.64 ± 3.41 0.012* 92.19 ± 32.9389.06 ± 37.210.12Rewarming time (min), mean ± SD103.04 ± 57.57102.28 ± 46.13104.56 ± 75.590.745103.83 ± 51.21100.4 ± 65.580.52Operation duration (min), mean ± SD422 ± 119.04430.17 ± 107.58407.43 ± 138.340.075412.01 ± 127.82400.93 ± 126.50.64Transfused RBC units (n), median (IQR)4 (0, 6)4 (0, 6)2 (0, 6)0.1881 (1, 1)1 (1, 1)0.44Transfused platelets units (n), median (IQR)0 (0, 2)0 (0, 2)0 (0, 1.75) 0.004* 0 (0, 2)0 (0, 2)0.18Cerebral perfusion time (min), mean ± SD26.5 ± 8.9527.93 ± 9.1424.7 ± 8.36 0.001* 25.3 ± 8.1624.24 ± 6.830.79Cerebral perfusion flow (ml/min), median (IQR)10 (8, 12.5)8 (5, 10)13 (10, 15) 0.001* 7.67 ± 2.6912.43 ± 2.8 0.01* CABG: coronary artery bypass graft; LCC: left common carotid; CPB: cardiopulmonary bypass; SD: standard deviation; IQR: interquartile range; Bold font: p-values less than 0.05. Perioperative characteristics CABG: coronary artery bypass graft; LCC: left common carotid; CPB: cardiopulmonary bypass; SD: standard deviation; IQR: interquartile range; Bold font: p-values less than 0.05. Postoperative outcomes Before matching, there were no significant disparities in the primary postoperative outcomes, except for TND (13.4% vs. 8%, P = 0.023; Table 3). After PSM, a notable difference in TND persisted between the two cohorts (14.29% vs. 6.91%, P = 0.01). There was no statistically significant difference observed in PND between the two groups both before and after PSM (Table 3). Logistic regression analysis showed that the use of b-ACP could significantly reduce the risk of TND compared to u-ACP (OR = 0.365, 95% CI [0.221, 0.604], P = 0.001; Table 4). Regarding 30-day mortality (Table 3), encompassing deaths occurring either in-hospital or post-discharge, no significant difference was observed between the two cohorts (4.15% vs. 3.23%, P = 0.61). CPB time (OR = 1.013, 95% CI [1.005, 1.021], P = 0.002) and pericardial effusion (OR = 2.394, 95% CI [1.23, 4.67], P = 0.01) were significantly associated with 30-day mortality (Supplementary Table 2). Table 3:Postoperative outcomesPostoperative outcomesAllMatchedVariablesTotal (n = 702)u-ACP (n = 402)b-ACP (n = 300) P-valueu-ACP (n = 217)b-ACP (n = 217) P-valueMechanical ventilation time (H), median (IQR)18 (12, 43)19 (13, 55)18 (11, 37)0.17418 (13, 48)18.5 (12, 37.25)0.51ICU time (H), median (IQR)91 (62, 144)90 (63, 159.88)91.94 (62, 139.54)0.9190 (63.13, 155.58)93.92 (61.94, 138.63)0.45Reintubate, n (%)12 (1.7%)6 (1.5%)6 (2%)0.6084 (1.84%)4 (1.84%)1Readmission to ICU, n (%)34 (4.8%)22 (5.5%)12 (4%)0.36911 (5.07%)12 (5.53%)0.83Second thoracotomy, n (%)21 (3%)14 (3.5%)7 (2.3%)0.3776 (2.76%)6 (2.76%)1TND, n (%)78 (11.1%)54 (13.4%)24 (8%) 0.023* 31 (14.29%)15 (6.91%) 0.01* PND, n (%)22 (3.1%)15 (3.7%)7 (2.4%)0.4876 (2.76%)7 (3.23%)0.46 Postoperative ischaemic stroke, n (%)19 (2.7%)14 (3.5%)5 (1.7%)0.1426 (2.76%)5 (2.3%)0.76 Postoperative cerebral haemorrhage, n (%)3 (0.4%)1 (0.2%)2 (0.7%)0.57902 (0.92%)0.5 Paraplegia, n (%)17 (2.4%)10 (2.5%)7 (2.3%)0.895*7 (3.23%)5 (2.3%)0.56 Hemiplegia, n (%)6 (0.9%)5 (1.2%)1 (0.3%)0.2462 (0.92%)1 (0.46%)0.56 Paresis, n (%)7 (1%)3 (0.7%)4 (1.3%)0.46804 (1.84%)0.12IABP, n (%)8 (1.1%)4 (1.0%)4 (1.3%)0.7291 (0.46%)4 (1.84%)0.37ECMO, n (%)10 (1.4%)5 (1.2%)5 (1.7%)0.7512 (0.92%)4 (1.84%)0.69Tracheotomy, n (%)6 (0.9%)2 (0.4%)4 (1.3%)0.412 (0.92%)4 (1.84%)0.69Renal insufficiency, n (%)169 (24.1%)100 (25%)69 (23%)0.56550 (23.04%)56 (25.81%)0.5Haemodialysis, n (%)50 (7.1%)32 (8.0%)18 (6%)0.31813 (5.99%)15 (6.91%)0.7Liver insufficiency, n (%)265 (37.7%)164 (40.8)101 (33.7%)0.05489 (41.01%)73 (33.64%)0.1130-day mortality, n (%)32 (4.4%)21 (5.2%)11 (3.7%)0.2289 (4.15%)7 (3.23%)0.61H: hour; ICU: intensive care unit; TND: transient neurologic deficit; PND: permanent neurologic dysfunction; IABP: intra-aortic balloon pump; ECMO: extracorporeal membrane oxygenation; SD: standard deviation; IQR: interquartile range; Bold font : p-values less than 0.05. Postoperative outcomes H: hour; ICU: intensive care unit; TND: transient neurologic deficit; PND: permanent neurologic dysfunction; IABP: intra-aortic balloon pump; ECMO: extracorporeal membrane oxygenation; SD: standard deviation; IQR: interquartile range; Bold font : p-values less than 0.05. Table 4:Risk factors for TNDVariablesUnivariateMultivariateOR (95% CI) P-valueOR (95% CI) P-valueAges1.02 (0.995–1.04)0.1270.987 (0.97–1.004)0.125Males1.29 (0.72–2.34)0.3940.66 (0.37–1.18)0.167Smoking1.28 (0.79–2.08)0.314Alcohol2.07 (0.96–4.46)0.065Hypertension1.90 (0.92–3.91)0.081Diabetes mellitus1.49 (0.61–3.68)0.385CAD0.50 (0.08–1.62)0.334Cerebral malperfusion1.56 (0.48–3.79)0.772Ischaemic stroke1.12 (0.39–3.37)0.798Cerebral haemorrhage1.61 (0.19–13.94)0.667Marfan syndrome0 (0–∞)0.998Pericardial effusion0.97 (0.60–1.57)0.893Creatinine1.009 (1.003–1.014) 0.002* 1.004 (0.998–1.011)0.153Ejection fraction1.01 (0.96–1.06)0.645Haemoglobin0.992 (0.979–1.004)0.187Platelets0.997 (0.993–1.001)0.101Glucose1.02 (0.96–1.07)0.535Albumin1.01 (0.96–1.06)0.592Lactate0.996 (0.93–1.07)0.907CABG1.63 (0.978–2.719)0.061CPB time1.004 (1.002–1.006) <0.001* 1.007 (1.001–1.014) 0.024* Cross-clamp time1.005 (1.001–1.009) 0.025* 0.996 (0.987–1.005)0.332Circulatory arrest time1.02 (0.994–1.05)0.1380.99 (0.961–1.021)0.531b-ACP0.56 (0.338–0.93) 0.025* 0.365 (0.221–0.604) <0.001* Nasopharyngeal temperature0.96 (0.84–1.09)0.565Rectal temperature0.97 (0.85–1.10)0.639Cooling time1.006 (1.001–1.012) 0.03* 0.996 (0.984–1.009)0.541Rewarming time1.004 (1.001–1.007) 0.02* 0.995 (0.988–1.003)0.254OR: odds ratio; CI: confidence interval; CAD: coronary artery disease; CABG: coronary artery bypass graft; CPB: cardiopulmonary bypass; b-ACP: bilateral antegrade cerebral perfusion; Bold font : p-values less than 0.05. Risk factors for TND OR: odds ratio; CI: confidence interval; CAD: coronary artery disease; CABG: coronary artery bypass graft; CPB: cardiopulmonary bypass; b-ACP: bilateral antegrade cerebral perfusion; Bold font : p-values less than 0.05. Follow-up The median follow-up time was 3 years, ranging from 1 to 5 years. The completion rate of follow-up is 100%. Follow-up was conducted by personal communication and clinic visit. During this period, a total of 24 deaths occurred. The mid-term survival rate appeared higher in the b-ACP group, although the difference was not statistically significant (5-year survival: 90.8% vs. 93.67%, P = 0.133; Fig. 2). In the u-ACP group, eight patients experienced cerebrovascular events during follow-up, compared to four patients in the b-ACP group. Follow-up data. Survival (Kaplan–Meier analysis) of all patients with acute type A aortic dissection (ATAAD) undergoing total arch replacement using unilateral or bilateral antegrade cerebral perfusion (u-ACP or b-ACP) DISCUSSION In the present study, we included 702 patients with ATAAD undergoing TAR + FET, divided into two cohorts according to cerebral perfusion. Our observation had no significance in postoperative PND and surgical mortality between the two groups. In addition, there was no discernible difference in mid-term survival. Nevertheless, the occurrence of TND was reduced in the b-ACP group. Our objective was to evaluate the independent effect of cerebral perfusion strategy on endpoint events. Therefore, we not only matched preoperative characteristics but also intraoperative characteristics, including CPB time. The 30-day death rates for the u-ACP and b-ACP cerebral perfusion groups were similar (4.15% vs. 3.23%, P = 0.61). The survival at 5 years was 93.67% in b-ACP and 90.29% in u-ACP. The survival rates were higher in the b-ACP group, but no statistically significant differences existed. As shown in Fig. 2, there is a clear trend between the u-ACP group and the b-ACP group, and there may be a significant difference in survival rates in a few years. We will follow up on the results. According to multivariable logistic regression analysis, CPB time (OR = 1.013, P = 0.002) and pericardial effusion (OR = 2.394, P = 0.01) are independent risk factors for 30-day death. Patients with preoperative ATAAD who have the accumulation of a significant amount of pericardial effusion pose a risk of pericardial tamponade, leading to a poor prognosis [13]. The incidence rates of PND in the two groups of this study were 3.23% and 2.76% (u-ACP vs. b-ACP), respectively. In u-ACP, there were four left brain new-onset postoperative stroke, two right-sided, and nine both sides. In b-ACP, there were three right-sided and two both sides stroke. Logistic regression in our study indicated that CPB time was a possible risk factor for PND (OR = 1.004, P = 0.003; Supplementary Table 1). Piperata et al.’s study [14] included 646 individuals from three centres who underwent surgical repair of ATAAD using ACP. The b-ACP cohort showed a notably higher incidence of PND (14% vs. 4%). We suspect that may be caused by the CPB time (170 min vs. 195 min, P < 0.001). To mitigate such disparities, we conducted propensity matching on intraoperative characteristics. Following PSM, a significant difference in the incidence of TND between the two groups was observed (14.29% vs. 6.91%, P = 0.01). A multivariable logistic regression analysis confirmed our findings (OR = 0.365, P < 0.001; Table 4). We hypothesize that the advantages of b-ACP in reducing PND outcomes may not be fully realized due to the relatively short duration of CA at our centre [15, 16]. Nevertheless, b-ACP appears to offer superior cerebral protection against TND, potentially through mechanisms involving the circle of Willis [17]. The pathophysiological basis for this association warrants further investigation, especially in patients undergoing surgery with u-ACP via the right axillary artery, where dependence on the circle of Willis for cerebral blood supply to the left hemisphere is crucial [18]. In our series, all patients underwent MHCA at 24.74 ± 1.7°C. In our previous study [19], which included 1018 patients undergoing TAR, the cohort was divided into two groups: MHCA (>24°C, n = 438) and DHCA (≤24°C, n = 580). The DHCA (deep hypothermic circulatory arrest) group required more RBC and platelet transfusions and experienced significantly higher blood loss (175.6 ± 45.6 vs. 166.8 ± 49.8 ml, P = 0.003). Additionally, their CPB times were generally longer. Our previous research has consistently shown that MHCA offers more advantages compared to DHCA. According to the latest guidelines [20], for complex aortic arch surgeries, the combination of ACP with moderate hypothermia (24.1°C–28°C) is recommended. In our current study, b-ACP was associated with significantly higher temperatures compared to u-ACP (25.91 ± 1.81°C vs. 24.52 ± 1.56°C; P = 0.01). While a 1.4°C difference may not necessarily lead to clinically different outcomes, it is noteworthy that this difference in minimum temperature correlated with a shorter duration of CPB time (197.48 ± 79.84 min vs. 195.95 ± 70.27 min) in the b-ACP group. This is a retrospective observational study, so it has inherent limitations. For example, the definition of TND may be subject to subjective influence. Additionally, it is crucial to acknowledge that this study is based on data from a single centre and does not encompass a multicentre experience. Variations in outcomes among different centres and surgeons may exist. In order to compare the physiological benefits of unilateral and bilateral approaches, intraoperative data were matched, potentially influencing the respective advantages of each approach during surgery. CONCLUSION Overall, both u-ACP and b-ACP effectively safeguarded the brain in cases of ATAAD, leading to positive short-term outcomes and medium-term survival rates. The use of b-ACP significantly reduces the incidence of TND. Supplementary Material Glossary ABBREVIATIONS Asian patients with acute type A aortic dissection Bilateral antegrade cerebral perfusion Circulatory arrest Confidence interval Cardiopulmonary bypass Frozen elephant trunk Moderate hypothermic circulatory arrest Odds ratio Permanent neurologic dysfunction Propensity score matching Standard deviation Total arch replacement Transient neurologic deficit Unilateral antegrade cerebral perfusion Contributor Information Yumeng Ji, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Juntao Qiu, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Kai Zhang, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Huiyuan Qin, Division of Cardiology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China. Jian Song, Department of Cardiovascular Surgery, Shandong Provincial Hospital Affiliated to Shandong University, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, China. Chenyu Zhou, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Shiqi Gao, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Bin Hou, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Fangfang Cao, Department of Intensive Care Unit, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Ji Wang, Department of Intensive Care Unit, Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Cuntao Yu, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. Wei Gao, Department of Vascular Surgery, Fuwai Hospital, National Center for Cardiovascular Diseases, National Clinical Research Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China. SUPPLEMENTARY MATERIAL Supplementary material is available at ICVTS online. FUNDING This study was supported by the National High-Level Hospital Clinical Research Funding of Fuwai Hospital (Nos. 2022-GSP-GG-24; 2023-GSP-GG-17) and the Beijing Municipal Natural Science Foundation (No. 7224341). Conflict of interest: all authors disclosed no relevant relationships. DATA AVAILABILITY The data underlying this paper cannot be shared publicly because of relevant data protection regulations. However, the data will be shared on reasonable request to the corresponding author with permission from the ethics committee. Author contributions Yumeng Ji: Conceptualization; Data curation; Methodology; Writing—original draft; Writing—review & editing. Juntao Qiu: Conceptualization; Data curation; Methodology; Writing—original draft; Writing—review & editing. Kai Zhang: Conceptualization; Data curation; Methodology; Writing—original draft; Writing—review & editing. Huiyuan Qin: Writing—review & editing. Jian Song: Data curation. Chenyu Zhou: Data curation. Shiqi Gao: Data curation. Bin Hou: Data curation. Fangfang Cao: Data curation. Ji Wang: Data curation. Cuntao Yu, MD: Conceptualization; Funding acquisition; Supervision; Writing—review & editing. Wei Gao: Conceptualization; Supervision; Writing—review & editing Reviewer information Interdisciplinary CardioVascular and Thoracic Surgery thanks Eugenio Neri, Anthony L. Estrera, Erik Beckmann and the other anonymous reviewers for their contribution to the peer review process of this article. REFERENCES Associated Data Supplementary Materials Data Availability Statement The data underlying this paper cannot be shared publicly because of relevant data protection regulations. However, the data will be shared on reasonable request to the corresponding author with permission from the ethics committee.