Introduction

Aortic arch surgery is a major surgical challenge with high mortality rates and an important risk of adverse neurological outcomes as well as any compromise to other target organs or significant bleeding [1]. Strategies such as selective cerebral perfusion (SCP) under different degrees of hypothermia have been attempted in order to allow adequate

cerebral perfusion and to diminish the peripheric tissue damage secondary to ischemia, thus allowing sufficiently long surgical times that allow the surgeon to correct the anatomical defect properly [2]. Although this technique has demonstrated adequate brain protection, it does not completely prevent the involvement of other bodily organs and systems such as the gastrointestinal system, the kidneys and the spinal cord because visceral suffering occurs after 30-40 minutes of SCP [3].

According to this, since year 2013 we implemented in our hospital a new technique, which allows simultaneous cerebral and mesenteric perfusion under moderate hypothermia (28° Celsius) in patients undergoing aortic arch surgery. The use of this perfusion method has been intermittent and its implementation has depended on the decision of the surgical team in each case. In this article, we present a transversal study in patients undergoing aortic arch surgery from January 2013 to March 2017 at the CardioVID hospital in Medellin – Colombia, whose primary objective is to compare the outcomes of the simultaneous cerebral and mesenteric perfusion strategy (intervention group) with the SCP technique (control group), as well as independently compare the results within the intervention group in an exploratory manner, assessing the outcomes according to the duration of the mesenteric perfusion.

Outcomes

Primary outcome of the study is the 30-day postoperative mortality. Exploratory outcomes are incidence of acute kidney injury (AKI) according to KDIGO guidelines 2012 criteria, length of stay (LOS) in the cardiovascular intensive care unit (ICU), total hospital LOS, orotracheal intubation time and metabolic behavior (pH, bicarbonate, lactate and glycemia)

Method and inclusion criteria

All patients undergoing aortic arch surgery due to ascending aorta and aortic arch aneurysm or Stanford type A dissection in our cardiovascular hospital between January 2013 and March 2017 were included in the study independently of which of both perfusion techniques was employed. Along this period 38 patients underwent aortic arch surgery in our center. In 26 patients, the simultaneous cerebral plus mesenteric perfusion was the technique of choice (intervention group) and in 12 patients SCP (control group) was the alternative for the surgical approach. Primary and exploratory outcomes were compared between both study groups. Additionally, intervention group was divided according to the duration of the mesenteric perfusion into 3 groups: <30 minutes, 31 – 59 minutes and >60 minutes, in order to compare intra-group outcomes in an exploratory manner. Perfusion technique was decided by the heart team for each patient according to their personal criteria in a case-based not standardized fashion.

Statistical analysis

A comparison of preoperative and demographic variables was carried out in order to establish the level of homogeneity between the two groups, as shown in Table1. Postoperative variables such as 30-day postoperative mortality, incidence of AKI, ICU and total hospital LOS, orotracheal intubation time and metabolic behavior (pH, bicarbonate, lactate and glycemia) were compared between both groups and also assessed in an intra-group manner inside the intervention group, in order to determine the study results according to the established primary and exploratory outcomes. Comparisons between both groups for dichotomic variables were accomplished using chi-square tests and were reported as P values and odds ratios with their respective 95% confidence intervals. Continuous variables were compared with the Student’s t-test and were reported as P values and 95% confidence intervals. The intra-group exploratory comparisons in the intervention group were performed using chi-square tests for dichotomic variables and analysis of variance (ANOVA) for continuous variables. Results were also reported as P values. Results are considered statistically significant when P values are < 0.05. All statistical analyzes were done with the SPSSv20.0 program (SPSS Inc, Chicago, IL, USA). Long-term survival assessment was performed with a Kaplan-Meier analysis [Table 2].

Technique

Intervention group (Simultaneous cerebral and mesenteric perfusion)

 The cardiopulmonary bypass (CPB) lines and cannulas were prepared before starting each operation. One “Y” connector was mounted on the arterial line to split the flow into two lines. The first line was inserted percutaneously under direct vision with Seldinger´s technique and connected to an arterial cannula (Edwards OptiSite, 18- French), which is going to be placed into the Innominate artery. In order to achieve bilateral antegrade cerebral perfusion, a self-inflating retrograde coronary sinus perfusion cannula, which is going to be positioned into the left carotid artery, was then attached to the luber on the side of the arterial cannula for the Innominate artery. The second line on the other side of the “Y” connector was attached to an orotracheal (OT) tube, which is going to be inserted into the aortic Dacron graft and the descending aorta for the mesenteric perfusion after opening of the aortic arch as described below. The operations were conducted through midline sternotomies. Cannulation took place after systemic heparinization. The ascending aorta was cross clamped and cold Del Nido cardioplegia was delivered in an antegrade manner to achieve cardiac arrest. The patient was cooled to a rectal temperature of 28° Celsius. During cooling, the aortic arch and neck vessels were prepared for control and procedures on the aortic valve and the ascending aorta were performed. At 28° Celsius, the three neck vessels were clamped individually and ascending aorta cross-clamping was released. Aortic arch pathologic tissues were then transected and the Dacron graft distal anastomosis to the descending aorta was started in an open way. During this period, the brain was continuously perfused through the innominate artery to a flow rate of 10-15 mL/kg/min and a mean arterial pressure of 60 mm Hg [4]. Perioperative cerebral blood flow was monitored continuously with bilateral Near Infrared Sonography (NIRS). Perfusion of the lower body was achieved through the head and neck collateral circulation which had adequate perfusion pressure thanks to the SCP through the innominate artery.

Approximately, when the first quarter of the distal anastomosis between the Dacron graft and the descending aorta was performed, the retrograde arterial cannula was placed into the left Carotid artery and the arterial line attached to the orotracheal tube was inserted through the Dacron graft and within the descending aorta [Figure 1]. The balloon of the orotracheal tube was then inflated to achieve a blood proof occlusion of the aorta and a surgical assistant manually kept the tube in its position while systemic blood flow increased progressively.

After cerebral and mesenteric perfusion was initiated, the rest of the anastomosis of the Dacron graft to the descending aorta was finished. Subsequently the neck vessels were anastomosed onto the neo-arch Dacron graft in a “Peninsula-Style” [5]. Once the anastomoses were done, the OT tube was released and retrograde aortic flow returned to the descending aorta and the neo-arch. The Dacron graft was cross clamped proximal to the Innominate artery and the individual cross- clamping on the neck vessels was released, allowing full systemic flow through the cannula in the Innominate artery. Thereafter slow rewarming and zero balance technique were initiated and anastomosis between ascending aorta Dacron graft and neo-arch Dacron graft was performed.

Control group (selective antegrade cerebral perfusion): 

Surgical procedure was performed similar to the intervention group, only with variations in steps related to mesenteric perfusion. Cerebral selective perfusion was given through an arterial cannula inserted into the Innominate artery and through a retrograde cannula in the left carotid artery in order to achieve bilateral cerebral blood delivery of 10 – 15mL/kg/min. Patients were also cooled until 28° Celsius.

Results

Operative variables are shown in Table 2. One patient from the intervention group died intraoperatively. The 30-day postoperative mortality was 13.2% (5/38) for both groups, 11.5% (3/26) and 16.7% (2/12) for the intervention group and the control group, respectively (OR 0.65; 95% CI 0.09 to 4.53; P = 0.66). Comparative results between intervention and control groups are shown in [Table 3].

Only pH after CPB and on admission in the ICU had a statistically significant difference in favor of the intervention group (7.4 ± 0.1 in the intervention group vs 7.3 ± 0.1 in the control group, 95% CI -0.17 to -0.03 and P= 0.01). Intra-group exploratory comparison within the intervention group is shown in [Table 4]. Less incidence of AKI (P= 0.05) and shorter orotracheal intubation times (P= 0.025) are evidenced in patients within the intervention group and mesenteric perfusion times longer than 30 minutes. Long-term survival is shown in [Figure 2].

Discussion

Open aortic arch surgery requires continuous antegrade selective cerebral perfusion under moderate or deep hypothermia. Although hypothermic SCP has been demonstrated safe and effective in terms of cerebral protection [6,7], safe duration of the hypothermic SCP is limited and hypothermia-related complications constitute an additional disadvantage for this technique [3]. Recent studies have reported operative in-hospital mortality for total aortic arch repair ranging from 2% to 12% and up to 39% in emergency cases [8 – 13]. With the Simultaneous cerebral and mesenteric perfusion technique, the brain and viscera are continuously perfused. This kind of perfusion allows an arch repair surgery to elapse as safe as a standard CPB heart operation even in normothermia, with a guaranteed perfusion to the spinal cord and splanchnic organs [14,15]. Separated mesenteric perfusion reduces ischemia of the spinal cord, kidneys, visceral organs and legs [16].

Another advantage of this technique is the avoidance of distal aortic graft clamping because on one side, there is no room for clamping in some cases and on the other side balloon occlusion does not deform the graft, which enables an easy anastomosis in the deep chest [16]. We perfuse both carotid arteries separately providing accurate cerebral perfusion regardless of the integrity of the Willis circle. In some cases, where prolonged surgeries are expected, we also cannulate the left subclavian artery independently to improve perfusion through the vertebral arteries.

It is relevant to notice that both perfusion techniques compared in this study were used under the same degree of hypothermia. According to similar series described in the literature, separated mesenteric perfusion technique is associated with an earlier patient rewarming allowing shorter CPB times [3,16]. There are some cases of special surgical difficulty where thanks to this perfusion technique, prolonged surgery may be performed with certain safety regarding systemic perfusion, allowing accurate anastomosis and thus reducing bleeding.

When comparing the intervention group with the control group, 30-day postoperative mortality was similar for both groups and only pH after CPB and on admission in the ICU had a statistically significant difference in favor of the intervention group. No other statistically significant differences were found for the primary nor the secondary outcomes. However, this isolated finding reflects a better metabolic performance and better tissue perfusion. It is possible that by increasing the sample’s size over time, statistically significant findings for other perfusion/metabolic variables can be found in favor of mesenteric perfusion.

It has been widely documented that significant visceral suffering occurs after 30-40 minutes of SCP. This is another reason why we did not find statistically significant differences when comparing the intervention group with the control group, since in the vast majority of surgeries in both groups the time of SCP was shorter than 30 minutes (29 of 38 patients). In obedience to this, we compared the results within the intervention group according to the duration of the mesenteric perfusion in order to stablish the degree of visceral protection over time provided by this perfusion method. The intragroup exploratory comparison shows promising results such as the statistically significant reduction of the incidence of AKI and shorter orotracheal intubation times in patients under mesenteric perfusion times longer than 30 minutes. In addition, there are other findings with clinical but not statistical significance in our study such as mean postoperative creatinine, ICU LOS mean time, total hospital LOS mean time and mean lactate levels, which could also reach statistical significance by increasing the sample size. This indicates that after 30 minutes, mesenteric perfusion improves the visceral protection, thus reducing the associated morbidity. It must be kept in mind that in those patients whose perfusion times are longer than 30 minutes, a higher morbidity is inherently associated with the procedure, independently of the perfusion technique employed, since they are longer surgeries and more extensive aortic diseases. This gives more value to our findings because the results are even better in those patients undergoing mesenteric perfusion for more than 60 minutes.

Given our results it is reasonable to take two possible behaviors. The first would be to assess the technical difficulty of surgery and the possible length of SCP. Those patients in which there is the certainty of a SCP shorter than 30 minutes, surgery may be performed with standard SCP. Conversely, in complex patients with SCP longer than 30 minutes, the decision should be to use the simultaneous cerebral and mesenteric perfusion. The second behavior would be to initiate simultaneous cerebral and mesenteric perfusion from the beginning of the procedure independent of the duration and technical complexity of the surgery. To do this, mesenteric perfusion should be administered through the descending aorta with a less rigid element than an orotracheal tube in order to allow a comfortable anastomosis at the beginning, remembering that in our original technique the orotracheal tube is inserted into the descending aorta after the first quarter of the distal anastomosis has been performed. An alternative for this would be a large diameter Foley catheter (24 French) that is malleable and mobile enough for this end.

Study Limitations

Perfusion technique choice was not based on a standardized protocol, but on the personalized heart team decision, which could induce selection bias in our study. Doors are left open for further investigation in this area. Randomized prospective studies would be more appropriate to assess the effectiveness of our perfusion strategy. Despite being an experience center for aortic surgery (> 15 cases/year), given the normal scarcity of aortic arch procedures, the study is based on a small sample. Larger volumes are required to probably find statistically significant differences.

Doors are left open to prolong this study in the time and it is also an invitation for other cardiovascular surgery groups in the world to assess the effectiveness of our technique.

Conclusion

Simultaneous cerebral and mesenteric perfusion is a simple, feasible and inexpensive perfusion technique, which could be used in a safe manner and even replace the habitual selective cerebral perfusion in order to improve outcomes in patients undergoing very complex aortic arch surgery with long selective cerebral perfusion and surgery times. It allows easier anastomosis without the constraint of time. Doors are left open for further research with larger patient’s cohorts.

Acknowledgments

We sincerely thank Dr. Joseph E. Bavaria for his kind cooperation and his complete willingness to advise us during the preparation and revision of this article. We also thank Manuel Callejas for his always kind disposition to make such a great illustration of the surgical technique and cannulation.

1. Ergin MA, Galla JD, Lansman SL, Quintana C, Bodian C, et al. (1994) Hypothermic circulatory arrest in operations on the thoracic aorta Determinants of operative mortality and neurologic outcome. J Thorac Cardiovasc Surg 107: 788-797.
2. Suzuki T, Asai T, Nota H, Kuroyanagi S, Kinoshita T, et al. (2013) Selective cerebral perfusion with mild hypothermic lower body circulatory arrest is safe for aortic arch surgery. Eur J Cardiothorac Surg. 43: e94–98.
3. Panos A, Myers PO, Kalangos A(2008) Novel Technique for Aortic Arch Surgery Under Mild Hypothermia. Ann Thorac Surg 85: 347–34 8.
4. Jacobs M, de Mol B, Veldman D (2001) Aortic arch and proximal supraaortic arterial repair under continuous antegrade cerebral perfusion and moderate hypothermia. Cardiovasc Surg;9: 396–402.
5. Itoh A, Fischbein M, Arata K, Craig Miller D (2010) “Peninsula-Style” Transverse Aortic Arch Replacement in Patients With Bicuspid Aortic Valve. Ann Thorac Surg 90: 1369 –1371.
6. Khaladj N, Shrestha M, Meck S, Peterss S, Kamiya H, et al. (2008) Hypothermic circulatory arrest with selective antegrade cerebral perfusion in ascending aortic and aortic arch surgery: a risk factor analysis for adverse outcome in 501 patients. J Thorac Cardiovasc Surg 135: 908-914.
7. Leshnower BG, Myung RJ, Kilgo PD, Vassiliades TA, Vega JD, et al. (2010) Moderate hypothermia and unilateral selective antegrade cerebral perfusion: a contemporary cerebral protection strategy for aortic arch surgery. Ann Thorac Surg 90: 547-554.
8. Ueda T, Shimizu H, Hashizume K, Koizumi K, Mori M, et al. (2003) Mortality and morbidity after total arch replacement using branched arch graft with selective antegrade cerebral perfusion. Ann Thorac Surg. 76: 1951-1956.
9. Spielvogel D, Halstead JC, Meier M, Kadir I, Lansman SL, et al. (2005) Aortic arch replacement using a trifurcated graft: simple, versatile, and safe. Ann Thorac Surg 80:90-95.
10. Sasaki H, Ogino H, Matsuda H, Minatoya K, Ando M (2007) Integrated total arch replacement using selective cerebral perfusion: a 6-year experience. Ann Thorac Surg. 83: S805-10.
11. Kazui T, Yamashita K, Washiyama N, Terada H, Bashar AH, et al. (2007) Aortic arch replacement using selective cerebral perfusion. Ann Thorac Surg 83: S796-98.
12. Minakawa M, Fukuda I, Yamauchi S, Watanabe K, Kawamura T, et al. (2010) Early and long-term outcome of total arch replacement using selective cerebral perfusion. Ann Thorac Surg 90:72-77.
13. Ergin MA, Galla JD, Lansman SL, Quintana C, Bodian C, et al. (1994) Hypothermic circulatory arrest in operations on the thoracic aorta. Determination of operative mortality and neurologic outcome. J Thorac Cardiovasc Surg 107: 788-797
14. Kirklin J, Barrat-Boyes B (1993) Hypothermia and total circulatory arrest. In: Cardiac Surgery, 2nd ed. New York, NY; Churchill- Livingstone, 1993:1712–1713.
15. Laven B, Orvietto M, Chuang (2004) Renal tolerance to prolonged warm ischemia time in a laparoscopic versus open surgery porcine model. J Urol 172: 2471– 2474.
16. Matsuyama S, Tabata M, Shimokawa T, Matsushita A, Fukui T, et al. (2012) Outcomes of total arch replacement with stepwise distal anastomosis technique and modified perfusion strategy. J Thorac Cardiovasc Surg 143: 1377–1381.