Arch surgery may be mandatory to improve the prognosis of patients with aTAAD with arch involvement. The first case of aortic arch replacement with hypothermic circulatory arrest was performed in Bologna by Pierangeli in 1974[3]. However, some drawbacks of this approach were been reported, including prolonged CPB time, higher incidence of neurologic dysfunction, limited safe time of distal circulatory arrest, and clotting disturbances [4]. To alleviate the side-effects of DHCA, a composite strategy incorporating the hypothermic circulatory arrest and selective cerebral perfusion was been proposed. This led to an apparent consensus that different levels of MHCA assisted by continuous selective cerebral perfusion was the key protocol for open arch repair [5-7].
In fact, the core temperature during arch surgery increased gradually with an acceptable postoperative mortality and morbidity [2, 8-10]. However, the optimal hypothermic level in TAR with SET implantation remained a controversial topic.
Since July 2019, we have attempted a warmer temperature in TAR with SET implantation for aTAAD in indicated patients. We found that there were no significant differences between the control and modified groups in terms of operative mortality, postoperative neurologic deficit, and visceral injury, and no deaths, PND or new renal failure needing CRRT occurred in the modified group.
The in-hospital mortality in the entire cohort was 4.9% (3/61), and all deaths occurred in the control group. This result was comparable with our previous reports [1, 2, 11, 12]. We believed the modified surgical techniques and evolving perfusion and cerebral protection strategies accounted for this. However, it should be noted that the characteristics of patients in this study may have contributed to the acceptable results. First, the mean age in all patients was 50.1±11.3 years, which was almost 10 years younger than the European series of aTAAD treated by frozen elephant trunk [13, 14]. Second, the rates of preoperative comorbidities were low. Third, natural selection was inevitable. Many patients were transferred from hundreds of miles away, and may die on the way to hospital.
Neurologic deficit was one of the main parameters used to evaluate arch surgery. The TND rate was 8.0% (2/25) in the modified group, and 13.9% (5/36) in the control group, although this difference did not reach the level of statistical significance.
The incidence of PND was 4.9% (3/61), and all three cases occurred in the control group (8.3% vs. 0, P = 0.262). This result was comparable with other published series focusing on patients who underwent frozen elephant trunk, in which the rate of PND ranged from 8.8% to 13% [15, 16].
In a study of the application of uSACP and bSACP for varying extents of arch replacement in 307 aTAAD patients performed by Norton et al. [17], in which risk factors for postoperative stroke were analyzed. It was observed that the type of SACP, HCA time, the extent of arch replacement were not significantly associated with new-onset postoperative stroke. Several factors may be related this and the mechanisms of PND require further investigation.
The right axillary artery was routinely used for CPB and SACP, which rarely involved dissection and atherosclerosis, and was associated with lower embolic stroke rates [18]. The blood flow was retrograde from the cannulating site to the arch, which may protect against embolic events in the right hemisphere during proximal aortic manipulations. In addition, the atherosclerotic plaques were always located at the base of the supra-arch vessels, which became the source of the embolus during arch manipulations. We performed isolated anastomosis of supra-aortic vessels with respective graft branches after clamping the vessels, and the arch roof with potential atheromasia was excised totally.
The relative benefits of uSACP compared with bSACP as cerebral perfusion strategies remained undetermined. In a comparative cerebral protection study from the German Registry for aTAAD [19], including 1,081 patients from 44 cardiac centers, it was found that the early neurologic outcomes with uSACP and bSACP were equivalent. A meta-analysis performed by Malvindi et al., consisting of 17 studies with 2,949 bSACPs and 599 uSACPs, concluded that bSACP allowed for longer SACP time, with increasing safety once the SACP time exceeded 40 to 50 minutes [20]. In a study by Tong et al. focusing on the clinical effect of bSACP and uSACP in TAR [21], no significant difference was identified with regard to 30-day mortality and postoperative PND. The four-branched graft anastomosed to the left carotid artery was used for left hemisphere perfusion, and they concluded that this approach avoided the risk of embolic injury by cannulating the left carotid artery.
We concluded that the bSACP represented a more physiologic perfusion, and contributed to a reduced risk of left hemisphere hypoperfusion in cases with an incomplete Willis circle. A flexible arterial cannula, designed for aortic cannulation in children with a diameter of 16-18F, was used to cannulate the left carotid artery. Great attention must be paid to identify the true lumen. The intima was always edematous and fragile, requiring cannulation with extreme caution. The absence of PND in the modified group confirmed that no embolic event related to cannulation in left carotid artery occurred in these patients.
Although the optimal hypothermic level for organ protection in arch surgery remains to be established, the application of aggressive temperature has been reported in recent years[8, 22]. The use of moderate-to-mild levels of hypothermia ameliorates the adverse effects of deep hypothermia, including longer CPB times, coagulopathy, and other complications that result from temperature-related systemic vasoconstriction[23].
In a study of the clinical effect of different levels of systemic hypothermia during arch repair in aTAAD patients, Zierer et al.[8] found that SACP with MHCA (30℃) was a safe tool for brain protection and helped to reduce hypothermia-related side-effects in comparison with the DHCA group. In another study [24], 1,002 patients undergoing hemiarch (684, 68%) or total arch (318, 32%) replacement via SACP with a mean core temperature of 30℃ were analyzed. The rates of early mortality, paraplegia and acute renal failure requiring hemofiltration were 5%, 0.3%, and 4% respectively. It was concluded that a slightly higher cerebral perfusion pressure in combination with warm antegrade cerebral perfusion might allow for an important collateral flow from the brain to the spinal cord that is evidenced as a substantial backflow of blood from the descending aorta requiring continuous suction during open distal anastomosis.
In a comparative study performed by Shen et al.[25] , the arch-first technique was applied for open total arch replacement in aTAAD patients under moderate systemic hypothermia (27.4℃) and antegrade cerebral perfusion. Compared with the control group treated with DHCA (24.2℃), the arch-first group had a significantly lower rate of distal organ complications (renal failure, hepatic failure and paraplegia).
The damaging effects of hypothermia on cerebrovascular autoregulation should not be ignored. Several animal studies have shown that cold cerebral perfusion can lead to neurologic injury due to its negative effects on cerebral autoregulation [26, 27], whereas mild systemic hypothermia preserved cerebral autoregulation and favored equal distribution of flow in the brain [28].
In our study, the occurrence of visceral injury in the setting of a higher core temperature was not noted. In the modified group, neither new renal failure needing CRRT nor hepatic and gastrointestinal injury were reported. One patient developed delayed paraplegia 3 days after surgery, and he recovered well 7 days later after cerebrospinal fluid drainage. The critical intercostal artery originated from the false lumen, and complete thrombus formation was observed in postoperative CTA performed 4 days after surgery. We considered the cause of delayed paraplegia may be related to rapid thrombus formation in the false lumen and inadequate collateral circulation in the acute phase, rather than the thermal strategy or distal open anastomosis.
Furthermore, the satisfactory visceral protection may be partially explained by the shorter lower body circulatory arrest time. In a study of open arch repair with moderate-to-mild (≥28℃) systemic hypothermia performed by Ahmad et al. [6], an acceptable organ protection effect was achieved with a mean lower body circulatory arrest time of 46±8 min in a cohort of patients with a mean age of 68±16 years. In our study, the mean time of open distal arch anastomosis was 16±4 min in the modified group. The distal dissected arch was sandwiched by the four-branched graft and the unstented part of the SET, and additional reinforced sutures were not usually needed. Therefore, distal reperfusion with approximate flow and pressure can be initiated once the distal anastomosis is completed.
In our previous study of 456 aTAAD patients treated by the same procedure with MHCA (25℃) and uSACP, the mortality, incidence of PND, renal failure, and spinal cord injury were 8.1%, 4.8%, 4.4%, and 2.4%, respectively [11]. The early postoperative outcomes were comparable to the results obtained in this study. We speculated that the effect of visceral protection using a higher core temperature was not inferior to that achieved by traditional thermal management strategy in the setting of TAR and SET implantation for aTAAD.
In this series, the duration of the operation, CPB, ventilation, and ICU stay were significantly lower in the modified group compared with that of the control group. We regarded cannulation of the left carotid artery as a convenient manipulation that costed only a small amount of time. The differences observed with respect to the duration of the operation, CPB, ventilation, and ICU stay may be explained to a large extent by the difference in hypothermia level. The clotting disorders related to prolonged CPB and low level MHCA were alleviated, which was confirmed by the significantly reduced volume of chest tube drainage in the first 48 h and RBC transfusion in the modified group.
Study Limitations
Certain limitations of this study should be noted. A retrospective comparative study cannot provide the same validity of evidence as a prospective study design. Although the comparable baseline characteristics in both groups, the existence of selection bias cannot be excluded.
In this study, 25 aTAAD patients were treated by two surgeons in modified group, and 36 patients were treated by another two surgeons in the control group. Although all four surgeons were experienced, the differences in operative data relating to the different surgeons treating the two groups may be magnified in the context of a limited sample.
Due to its small sample and retrospective nature, the statistical significance of our results require further confirmation in well-designed trials with larger sample sizes.