Warming up the frozen elephant trunk for aortic arch pathology

Introduction

The frozen elephant trunk (FET) procedure is a complex technique used to treat both chronic and acute aortic arch (AA) disease. However, it necessitates hypothermic circulatory arrest (HCA) to operate in a bloodless environment and ensuring organ protection through metabolic suppression (1-3). While its introduction has reduced circulatory arrest (CA) times compared to the conventional elephant trunk procedure (2), it is still associated with neurological complications and perioperative mortality (4). The advent of antegrade cerebral perfusion (ACP) has led to a gradual increase in CA temperature, shifting from moderate HCA (MoHCA; temperature ≥22–27 ℃) to mild HCA (MiHCA; temperature ≥28–30 ℃), thereby reducing surgical times and the risk of bleeding (5-8). Recent studies support this trend, suggesting that higher temperatures can provide adequate neuroprotection and reduce systemic complications associated with deep hypothermia (9-11). Nevertheless, the debate over the optimal temperature strategy continues; some authors advocate for MiHCA combined with bilateral ACP, while others argue that deep hypothermia remains necessary in certain patients to prevent neurological events, visceral ischemia, and renal injury, especially when ACP is not feasible or when prolonged lower body ischemia is present (12). This study aims to assess the incidence of postoperative neurological events, bleeding, and kidney injury in patients who underwent FET with ACP at a core temperature >30 ℃.

Methods

This retrospective observational study includes all consecutive patients who underwent the FET procedure at the Cardiac Surgery Unit of the University of Pisa from January 2019 to March 2025. The cohort comprises both elective and urgent patients who underwent surgery for acute aortic syndromes or chronic aneurysmal disease. CA and ACP were performed in all the included patients. All procedures were conducted in MiHCA with a core temperature of approximately 30 ℃. Core temperature was continually monitored via a dedicated Foley bladder catheter with a temperature probe. Throughout the procedure, continuous bilateral monitoring of cerebral oxygen saturation was achieved using Near Infrared Spectroscopy (NIRS-Medtronic INVOS).

ACP was either performed through direct cannulation of the supra-aortic vessels, according to the Kazui technique and using a dedicated cannula (TrueFlow RDB), right axillary artery cannulation, left axillary artery, or a combination of axillary artery/arteries and supra-aortic vessels cannulation. Cerebral perfusion flow was maintained between 8 and 10 mL/kg/min until physiological antegrade perfusion was restored. Anesthesia management during CA was standardized with head ice packing and administration of high-dose corticosteroids and phenobarbital to suppress cerebral activity a few minutes before the arrest. Bilateral radial or brachial invasive artery pressure monitoring was always present. At the Cardiac Surgery Unit of the University of Pisa, when necessary, root surgery is typically performed first. After completing the distal arch anastomosis and restoring systemic flow, the arch prosthesis is connected to the aortic root, enabling heart perfusion. Supra-aortic vessel reconstruction is then performed as the final step, with a left subclavian artery extra-anatomic bypass in cases of complex intrathoracic management. All procedures were performed by staff surgeons, with the choice of arterial and venous cannulation, cardioplegia delivery, and ACP configuration made at the surgeon’s discretion. Preoperative, intraoperative, and postoperative data were collected retrospectively through the review of electronic medical records.

The following outcomes were evaluated: (I) 30-day mortality; (II) occurrence of any postoperative neurological events (including transient ischemic attacks and disabling or not-disabling strokes); (III) postoperative major bleeding events, defined as either need for mediastinal packing for hemostatic control at the end of index surgery and/or re-sternotomy for bleeding in the first 72 hours after index surgery; and (IV) acute kidney injury (AKI), defined as a ≥150% increase in baseline creatinine within 72 hours and/or need for continuous venovenous hemofiltration (CVVH). Postoperative neurological events were assessed either with neurological examination by local expert neurologist and/or with computed tomography (CT) scan (at assessment and after 48 and 72 hours). Continuous variables were presented as mean ± standard deviation (SD) or as median with I–III quartile, according to data distribution, and were compared using Student’s t-test or the Mann-Whitney U test, as appropriate. Categorical variables were expressed as absolute numbers and percentages and were compared using the chi-square test or Fisher’s exact test, as applicable. To identify variables independently associated with neurological events, bleeding events, and renal events, multivariable regression models were employed. For each of these outcomes, one model was developed based on baseline characteristics and another based on intraprocedural variables, allowing for the separate evaluation of pre-procedural and procedural risk factors. The variables used to develop the model were chosen according to clinical experience and published literature and included, for baseline, age, sex, German Registry of Acute Aortic Dissection Type A (GERAADA) score, history of chronic kidney disease, previous cardiac surgery, acute aortic syndrome as presentation, involvement of the descending aorta and shock status upon arrival, while for intraprocedural, bladder temperature, arterial cannulation strategy (antegrade or retrograde), cardiopulmonary bypass (CPB), cross-clamp, ACP and CA times, cerebral perfusion strategy (unilateral or bilateral), INVOS value and re-sternotomy for bleeding. All statistical tests were two-sided, and P values <0.05 were considered statistically significant. Analysis was performed using SPSS Statistics 28 (IBM; SPSS Inc. Chicago, IL, USA) and MedCalc (MedCalc Soft Ltd., Ostend, Belgium).

Results

Ninety-two patients were included, of whom 74% were male with a mean age of 64±13 years. In 86% of cases, patients presented with acute aortic syndromes, while 14% had chronic thoracic aortic aneurysms. The descending aorta was involved in 68% of cases, and 37% of patients had hemodynamically relevant aortic valve regurgitation. Overall, patients did not present many comorbidities; however, their general condition at presentation was compromised, with a mean GERAADA score of 24%±14%. At arrival, 13% presented hemiparesis or hemiplegia, while 4% had coma, and 9% were intubated. Pericardial effusion was present in 28% of patients; of those, 9% presented with tamponade, and 16% were shocked. Peripheral, visceral, and coronary malperfusion wer observed in 20%, 16%, and 17% of cases, respectively. Complete baseline characteristics are reported in Table 1.

All patients underwent aortic ascending replacement and arch replacement with the FET technique. Root surgery was performed in 26 patients (28%), and aortic valve replacement was done in 19 patients (20%). Canulation with retrograde perfusion was performed in the majority of cases (57%), anterograde perfusion in 29% of cases, and 14% of patients underwent a combination of the two strategies. Total CPB and cross-clamp times were 219±78 and 123±55 min, respectively, while median CA time was 15 (I–III quartile, 13–19) min. Mean bladder temperature was 30±1.6 ℃. Almost all patients (96%) received bilateral ACP, while 4% received unilateral ACP with a median perfusion time of 59 (I–III quartile, 31–113) min. Five patients (5%) experienced intraoperative death due to descending thoracic aorta rupture and massive bleeding. Four patients (4.3%) needed packing at exit from the operating room to control diffuse oozing. Intraoperative data are reported in Table 2. Thirty-day mortality was 14%, and eight patients (9%) died for cardiovascular causes, while five (5%) died for sepsis or respiratory causes. Neurological events were observed in 16% of patients (fifteen patients), of those 9% (eight patients) were disabling and 1% (one patient) led to death. Renal or visceral complications occurred in nine patients (10%), one patient (1%) developed bowel ischemia, and eight patients (9%) required CVVH (2% resulting in chronic dialysis). Re-sternotomy for bleeding was necessary in nineteen patients (21%), and six patients (7%) developed tamponade within the hospital admission. Postoperative events are reported in Table 3.

At multivariable analysis, 30-day mortality was associated with presence of shock at arrival in the OR [odds ratio (OR) =1.336; 95% confidence interval (CI): 0.063–1.485; P=0.012], necessity of resuscitation before surgery (OR =0.321; 95% CI: −1.408 to 0.421; P=0.02), and with GERAADA score (OR =1.557; 95% CI: 0.004–2.018; P=0.002). The GERAADA score (OR =1.038; 95% CI: 1.009–1.068; P=0.009), retrograde perfusion (OR =0.395; 95% CI: 0.145–1.080; P=0.070), and lower right INVOS value (OR =0.934; 95% CI: 0.869–1.005; P=0.066) tended to be associated with the onset of postoperative neurological events (Table 4). Bleeding was associated with the GERAADA score (OR =1.028; 95% CI: 1.000–1.057; P=0.049) and with the cannulation strategy (retrograde OR =2.807; 95% CI: 1.11–7; P=0.028; antegrade OR =0.365; 95% CI: 0.129–1.031; P=0.057) (Table 5). Renal events tended to be associated with preoperative involvement of the descending aorta (OR =7.175; 95% CI: 0.839–61.367; P=0.072), while there was no association with intraoperative variables. For all outcomes, bladder temperature presented no influence (Table 6).

Discussion

The use of MiHCA for acute aortic syndromes has gained traction in recent years, as it appears to provide outcomes comparable or superior to those of MoHCA and deep HCA, particularly in terms of neurological events and renal injury (13-16). The results of our study corroborate this evidence. In our study population, the mean bladder temperature was 30±1.6 ℃, with 115 patients (67%) being treated at a core temperature above 29 ℃. Notably, unlike other studies that analyzed homogeneous groups of elective chronic AA aneurysms (17-19), our experience predominantly involved emergent surgeries for acute aortic dissections. These cases are typically managed with deeper HCA to mitigate the damage associated with the common preoperative hypoperfusion status. Despite the high-risk profile presented by our population, we did not observe an association between postoperative outcomes and core temperature, suggesting that even in the subset of acute pathologies, MiHCA can be safely carried out. Along with MiHCA, selective bilateral cerebral perfusion was performed in more than 90% of patients to guarantee optimal cerebral protection. This strategy resulted in 12% postoperative permanent neurological events, which is comparable to or inferior to other reports (5). Regarding the findings of Zhu et al. (14), who reported a 16% incidence of neurological events, our rate was lower, potentially due to the systematic use of bilateral ACP combined with standard pharmacological protocols, which provided comprehensive brain protection. Other groups did not find a significant difference between unilateral and bilateral ACP in large cohorts in terms of permanent neurological events (20). In contrast, other single-center studies and large meta-analyses did not find a significant benefit of bilateral over unilateral ACP (21,22). Bilateral ACP was associated with lower events in case of longer CA and ACP times (21). Given that our strategy involves reconstructing the supra-aortic vessels as the final step, leading to extended ACP time, we prefer bilateral ACP whenever feasible. In multivariable analysis, core temperature did not influence the occurrence of ischemic stroke. However, ischemic stroke was associated with the preoperative GERAADA score (23) and tended to be affected by the use of retrograde cannulation and lower right INVOS values during surgery. The association with retrograde perfusion aligns with previous literature (24), as retrograde flow may dislodge atherosclerotic or calcific plaques along the descending aorta, resulting in embolic stroke. Additionally, when retrograde systemic perfusion was employed, axillary artery cannulation, later used for ACP, was often not performed. This led to direct cannulation of both the innominate and left common carotid arteries, which in some cases, correlated with higher stroke rates due to the mobilization of intimal plaques during cannula insertion (21). On the other hand, events were strongly associated with the GERAADA score, highlighting the significant impact preoperative critical status has on outcomes.

MiHCA has been shown to be protective against postoperative renal impairment according to previous publications (15,19). In our cohort, higher core temperatures did not appear to be associated with the development of renal impairment, which was observed in only a small proportion of patients. Specifically, no patients increased postoperative creatinine above 150% respect to baseline, and only 6% required temporary renal replacement therapy. This lower incidence of AKI might be explained by the short CA time observed in our patients. Previous studies have demonstrated that the onset of renal dysfunction is associated with preoperative patient conditions and prolonged CA times, typically exceeding 60 min (18). We achieved short CA times through a surgical strategy that involved performing the distal anastomosis first, followed by the restoration of antegrade systemic perfusion and proximal anastomosis with the interruption of cross-clamping. Present data also showed that the involvement of the descending aorta can affect the development of renal impairment. This finding is consistent with results published by Li et al., who identified the involvement of the renal arteries as a factor leading to the necessity of CRRT in a machine-learning model (25).

The evaluation of intraoperative variables did not show correlation with the onset of renal dysfunction, confirming that our surgical protocol, paired with MiHCA, is effective in preventing visceral ischemia. In particular, the early restoration of antegrade perfusion through the side branch of the ascending aorta prosthesis and short CA time likely ensured optimal visceral protection. The rate of bleeding events that required surgical revision within the first 48 hours from surgery (11%) is consistent with previous analysis (5) and correlated with a higher GERAADA score at admission. This correlation can be attributed to the parameters collected at admission to calculate the score, which highlight worse preoperative conditions that can lead to postoperative complications. MiHCA also facilitated a reduction in CPB time and total surgical time. Compared to other studies, we observed slightly longer CPB times (14,15), which may be due to the high number of emergent acute aortic syndromes in our cohort, resulting in more complex and time-consuming repair procedures.

Thirty-day mortality was higher in our cohort compared to other reports (5,15,26). This could be attributed to the high incidence of acute aortic syndromes in our patient population, along with the high-risk profiles as indicated by the GERAADA score. Detter et al. (27) showed that early mortality after FET was mainly influenced by preoperative critical state, including signs of cerebral or systemic malperfusion, preoperative cardiogenic shock with resuscitation, and orotracheal intubation upon arrival in the operating room. Conversely, other studies assessing the impact of MiHCA in elective FET procedures for AA aneurysms reported lower mortality and postoperative complication rates, reinforcing the rationale for reducing the degree of hypothermia during CA (14). Our results are encouraging, supporting the use of MiHCA and paving the way for the adoption of even higher temperatures provided that bilateral ACP is maintained, and CA time is relatively short. In-vivo studies have demonstrated the safety of MiHCA within certain time limits, indicating that CA times exceeding 60 min at 32 ℃ may eventually lead to neurological and visceral injury (28). To address this issue, authors have proposed new strategies to maintain systemic flow throughout the surgery, such as inflating a Foley catheter in the descending aorta to deliver systemic flow to the lower body (24,29,30). Following the introduction of ACP, we transitioned from deep HCA to MoHCA. With the improvement of devices and the introduction of FET procedure, which reduced surgical time for distal anastomosis, we were further prompted to shift from MoHCA to MiHCA. Once we standardize methods to avoid interrupting lower body systemic flow, aided by advancements in FET devices, we anticipate being able to transition from MiHCA to normothermic CA.

The main limitation of this study is represented by its retrospective nature; nevertheless, our surgical perfusion and anesthesia strategies are well-standardized and homogeneous, limiting the selection bias. Another limitation is the lack of a control group represented by MoHCA and deep HCA. We decided not to include patients treated before 2020, when MoHCA was still used, because the operative management during that period was less standardized. This lack of standardization could have introduced additional variables that might have impacted the outcomes.

Conclusions

MiHCA is safe for patients requiring complex AA surgery and in those with acute aortic syndromes, with low rates of neurological events, renal dysfunction, and no increase in postoperative bleeding. However, it must be paired with bilateral ACP and a well-standardized technical strategy and anesthesia management. Outcomes are, however, impacted by patients’ preoperative conditions. Further studies are necessary to support the introduction of normothermic CA.

Acknowledgments

None.

Footnote

Funding: None.

Conflicts of Interest: The authors have no conflicts of interest to declare.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Karck M, Kamiya H. Progress of the treatment for extended aortic aneurysms; is the frozen elephant trunk technique the next standard in the treatment of complex aortic disease including the arch? Eur J Cardiothorac Surg 2008;33:1007-13. [Crossref] [PubMed]
2. Moula AI, Romeo JLR, Parise G, et al. The evolution of arch surgery: Frozen elephant trunk or conventional elephant trunk? Front Cardiovasc Med 2022;9:999314. [Crossref] [PubMed]
3. Aalaei-Andabili SH, Scali S, Klodell C, et al. Outcomes of Antegrade Stent Graft Deployment During Hybrid Aortic Arch Repair. Ann Thorac Surg 2017;104:538-44. [Crossref] [PubMed]
4. Hanif H, Dubois L, Ouzounian M, et al. Aortic Arch Reconstructive Surgery With Conventional Techniques vs Frozen Elephant Trunk: A Systematic Review and Meta-Analysis. Can J Cardiol 2018;34:262-73. [Crossref] [PubMed]
5. Wang L, Zhong G, Lv X, et al. Clinical outcomes of mild versus moderate hypothermic circulatory arrest with antegrade cerebral perfusion in adult aortic arch surgery: A systematic review and meta-analysis. Perfusion 2024;39:266-80. [Crossref] [PubMed]
6. Russo CF, Mariscalco G, Colli A, et al. Italian multicentre study on type A acute aortic dissection: a 33-year follow-up†. Eur J Cardiothorac Surg 2016;49:125-31. [Crossref] [PubMed]
7. Colli A, Carrozzini M, Francescato A, et al. Acute DeBakey Type I aortic dissection without intimal tear in the arch: is total arch replacement the right choice? Interact Cardiovasc Thorac Surg 2018;26:84-90. [Crossref] [PubMed]
8. Yan TD, Bannon PG, Bavaria J, et al. Consensus on hypothermia in aortic arch surgery. Ann Cardiothorac Surg 2013;2:163-8. [Crossref] [PubMed]
9. Colli A, Carrozzini M, Galuppo M, et al. Analysis of early and long-term outcomes of acute type A aortic dissection according to the new international aortic arch surgery study group recommendations. Heart Vessels 2016;31:1616-24. [Crossref] [PubMed]
10. Schachner B, Ratschiller T, Benedikt P, et al. Aortic Arch Surgery Under Warm Conditions (Moderate to Mild Hypothermia). Arch Clin Biomed Res 2020;4:551-60.
11. Shen K, Tan L, Tang H, et al. Total Arch Replacement With Frozen Elephant Trunk Using a NEW "Brain-Heart-First" Strategy for Acute DeBakey Type I Aortic Dissection Can Be Performed Under Mild Hypothermia (≥30°C) With Satisfactory Outcomes. Front Cardiovasc Med 2022;9:806822. [Crossref] [PubMed]
12. McClure RS. A Standardised Classification for the Various Total Arch Replacement Techniques: See the Benchmarks, Then Raise the Bar. Can J Cardiol 2024;40:1334-7. [Crossref] [PubMed]
13. Zierer A, El-Sayed Ahmad A, Papadopoulos N, et al. Selective antegrade cerebral perfusion and mild (28°C-30°C) systemic hypothermic circulatory arrest for aortic arch replacement: results from 1002 patients. J Thorac Cardiovasc Surg 2012;144:1042-49. [Crossref] [PubMed]
14. Zhu K, Qian S, Pan X, et al. Short-term outcomes of mild (≥30 °C) vs. moderate hypothermic circulatory arrest in aortic arch surgery. J Thorac Dis 2024;16:5815-25. [Crossref] [PubMed]
15. Li Z, Tang H, Deng C, et al. Impact of mild hypothermic circulatory arrest on surgical outcomes in acute type a aortic dissection patients: a single-centre study. BMC Cardiovasc Disord 2025;25:32. [Crossref] [PubMed]
16. Leshnower BG, Myung RJ, Thourani VH, et al. Hemiarch replacement at 28°C: an analysis of mild and moderate hypothermia in 500 patients. Ann Thorac Surg 2012;93:1910-5; discussion 1915-6. [Crossref] [PubMed]
17. Urbanski PP, Lenos A, Bougioukakis P, et al. Mild-to-moderate hypothermia in aortic arch surgery using circulatory arrest: a change of paradigm? Eur J Cardiothorac Surg 2012;41:185-91. [Crossref] [PubMed]
18. Nota H, Asai T, Suzuki T, et al. Risk factors for acute kidney injury in aortic arch surgery with selective cerebral perfusion and mild hypothermic lower body circulatory arrest. Interact Cardiovasc Thorac Surg 2014;19:955-61. [Crossref] [PubMed]
19. Zheng HJ, Liu X, He P, et al. Early outcome of simplified total arch reconstruction under mild hypothermia (30-32 °C) with distal aortic perfusion. J Cardiothorac Surg 2023;18:323. [Crossref] [PubMed]
20. Ji Y, Qiu J, Zhang K, et al. Comparing unilateral and bilateral cerebral perfusion during total arch replacement for acute type A aortic dissection. Interdiscip Cardiovasc Thorac Surg 2024;40:ivae205. [Crossref] [PubMed]
21. Preventza O, Simpson KH, Cooley DA, et al. Unilateral versus bilateral cerebral perfusion for acute type A aortic dissection. Ann Thorac Surg 2015;99:80-7. [Crossref] [PubMed]
22. Santos K, Velasco EM, Mawasi M, et al. Unilateral Versus Bilateral Antegrade Cerebral Perfusion in Aortic Arch Surgery: Systematic Review and Meta-Analysis of Randomised Controlled Trials and Propensity-Matched Studies. Heart Lung Circ 2025; Epub ahead of print. [Crossref]
23. Czerny M, Siepe M, Beyersdorf F, et al. Prediction of mortality rate in acute type A dissection: the German Registry for Acute Type A Aortic Dissection score. Eur J Cardiothorac Surg 2020;58:700-6. [Crossref] [PubMed]
24. Pacini D, Leone A, Di Marco L, et al. Antegrade selective cerebral perfusion in thoracic aorta surgery: safety of moderate hypothermia. Eur J Cardiothorac Surg 2007;31:618-22. [Crossref] [PubMed]
25. Li K, Li Y, Gao Q, et al. Machine Learning in Risk Prediction of Continuous Renal Replacement Therapy After Surgical Repair of Acute Type A Aortic Dissection. J Cardiothorac Vasc Anesth 2025;S1053-0770(25)00494-X.
26. El-Sayed Ahmad A, Papadopoulos N, Risteski P, et al. Is More than One Hour of Selective Antegrade Cerebral Perfusion in Moderate-to-Mild Systemic Hypothermic Circulatory Arrest for Surgery of Acute Type A Aortic Dissection Safe? Thorac Cardiovasc Surg 2018;66:215-21. [Crossref] [PubMed]
27. Detter C, Bax L, Panuccio G, et al. Complicated acute type A aortic dissection and severe aortic atherosclerosis predict early mortality after frozen elephant trunk procedure. Eur J Cardiothorac Surg 2025;67:ezaf213. [Crossref] [PubMed]
28. Fu D, Chen C, He L, et al. Protective Effect of Mild Hypothermia on Spinal Cord Ischemia-Induced Delayed Paralysis and Spinal Cord Injury. Neurochem Res 2022;47:1212-25. [Crossref] [PubMed]
29. Goto Y, Hosoba S, Fukumoto Y, et al. Mild Systemic Hypothermic Circulatory Arrest Using a Frozen Elephant Trunk Graft with Endo-Balloon Occlusion for Total Arch Replacement. Heart Surg Forum 2020;23:E673-6. [Crossref] [PubMed]
30. Li Q, Qu H, Liu T, et al. Total aortic arch replacement surgery with a Core temperature of 34 °C. J Cardiothorac Surg 2019;14:184. [Crossref] [PubMed]