Introduction

Paediatric aortic arch repairs, such as the Norwood procedure for hypoplastic left heart syndrome, interrupted aortic arch repair, or coarctation of the aorta, require intraoperative aortic arch clamping. Aortic arch clamping requires careful management to maintain blood flow and organ function in the lower body as well as in brain. Various strategies have been proposed for intraoperative cardiopulmonary bypass (CPB) management, such as circulatory arrest of the lower body under deep hypothermia, or using antegrade descending aortic perfusion via a cannula inserted into the descending aorta to ensure the perfusion of lower body organs including the spinal cord [1].

However, CPB management with hypothermia can lead to platelet and coagulation system dysfunction, resulting in increased intraoperative blood loss and greater blood product requirements [2, 3]. Alternatively, CPB management, combined with antegrade descending aortic perfusion via an additional cannula inserted into the descending aorta, aims to avoid circulatory arrest of the lower body under deep hypothermia. However, this method has limitations. Patients may experience stress when the cannula is removed from the descending aorta, because the heart will be compressed immediately after CPB withdrawal, potentially causing further haemodynamic instability. In addition, ensuring haemostatic control of the cannula insertion site in the descending aorta within the restricted surgical field of small paediatric patients presents challenges.

To address these issues, we proposed a CPB management method involving cannulation into the femoral artery to provide retrograde perfusion to the descending aorta to maintain blood supply to the lower body organs [4]. This method allows for removing the femoral artery cannula postoperatively once blood coagulation stabilises, simplifying the procedure and minimising the risk of haemodynamic instability. Furthermore, the absence of a descending aortic cannula in the small surgical field of paediatric patients enhances visibility, a feature appreciated by cardiovascular surgeons [4]. Subsequently, we explored methods to optimize blood flow volume using smaller cannulas without increasing the CPB circuit pressure by selecting the cannula and extension tube for this method [5].

Although we have reported on innovations and critical aspects to ensure the success of cannulation into the femoral artery as a method of CPB management for paediatric aortic arch repairs [6], the efficacy of this method has yet to be objectively assessed despite more than a decade of successful implementation. Therefore, in this study, we hypothesised that CPB management with blood delivery via femoral artery cannulation reduces lactate elevation compared to CPB management with circulatory arrest of the lower body under deep hypothermia. This study aimed to test the validity of this hypothesis by comparing the degree of lactate elevation during CPB management as the primary outcome. This study also examined the efficacy of this method compared to that of CPB management with antegrade descending aortic perfusion via an additional cannula inserted into the descending aorta.

Methods

Ethics approval and registration

This study was approved by the Ethics Committee of Niigata University, Niigata, Japan (chairperson: Prof. Hirohito Sone), on November 14, 2022 (registration number: 2021 − 0208) and was registered in the UMIN Clinical Trials Registry (registration number: UMIN000052933, registered 28 November 2023, https://rctportal.niph.go.jp/s/detail/um?trial_id=UMIN000052933). Written informed consent for general anaesthesia and surgery and publication was obtained from the parents or persons with parental authority for all paediatric patients undergoing surgery.

Study protocol and patients

This study included patients who underwent paediatric aortic arch repairs between March 2012 and March 2018 at the German Paediatric Heart Centre, Sankt Augustin. Patients who were repumped intraoperatively were excluded. This study retrospectively reviewed the CPB records by investigating the minimum body temperature during CPB (℃), which was measured rectally, the lactate levels in the blood gas analysis at the start of CPB (mmol/L), CPB duration (min), aortic clamp duration (min), total urine output during CPB (mL), and changes in the lactate levels during CPB (mmol/L).

Lower body circulatory arrest (Group A) vs. femoral artery perfusion (Group F)

We first compared a group that underwent lower body circulatory arrest (Group A, circulatory arrest) with a group that received additional retrograde aortic perfusion via femoral artery cannulation (Group F, femoral artery perfusion). Group A included 41 (male/female = 27/14) paediatric patients. Group F included 18 (male/female = 8/10) paediatric patients. Both groups were operated on by the same surgeon. Table 1 shows the daily age (days), height (cm), and weight (kg) of the patients in each group.

Table 1 Patient characteristics for each group
Full size table

Descending aortic perfusion (Group D) vs. femoral artery perfusion (Group F)

We compared a group that received additional antegrade descending aortic perfusion (Group D, descending aortic perfusion) and Group F (n = 18). Fifteen (male/female = 12/3) patients were enrolled in Group D. Groups D and F were operated on by different surgeons. Table 1 shows the daily age (days), height (cm), and weight (kg) of the patients in each group.

Statistical analyses

Numerical data are presented as the mean ± standard deviation, and 95% confidence intervals (CIs). The Welch t test was used to compare two groups, and p < 0.05 was considered to indicate significance.

Results

Lower body circulatory arrest (Group A) vs. femoral artery perfusion (Group F)

During CPB, patients in Group A had a minimum body temperature of 16.79 ± 1.48 °C (95% CI 16.34 to 17.24), which was significantly different (p < 0.05) from that in Group F (23.73 ± 4.67 °C, 95% CI 21.57 to 25.89).

The lactate levels at the start of CPB were 2.32 ± 0.70 mmol/L (95% CI 2.10 to 2.53) in Group A and 2.30 ± 1.46 mmol/L (95% CI 1.62 to 2.98) in Group F. The CPB durations were 148.98 ± 22.45 min (95% CI 142.10 to 155.85) in Group A and 158.67 ± 41.64 min (95% CI 139.43 to 177.90) in Group F. The aortic clamp time durations were 75.49 ± 21.97 min (95% CI 68.76 to 82.21) in Group A and 78.11 ± 17.67 min (95% CI 69.95 to 86.27) in Group F. The total urine outputs during CPB were 7.03 ± 5.54 mL (95% CI 5.24 to 8.11) in Group A and 8.94 ± 9.62 mL (95% CI 4.22 to 13.62) in Group F. These parameters did not significantly differ (lactate levels, p = 0.964; CPB duration, p = 0.375; aortic clamp time, p = 0.637; total urine output, p = 0.480).

The mean increase in lactate levels during CPB was 1.66 ± 0.90 mmol/L (95% CI 1.39 to 1.94) in Group A and 1.06 ± 0.94 mmol/L (95% CI 0.62 to 1.50) in Group F; the mean increase in lactate levels was significantly lower in Group F (p = 0.033) (Table 2).

Table 2 Comparison between the lower body circulatory arrest group (Group A) and the cardiopulmonary bypass method with blood delivery via femoral artery cannulation group (Group F)
Full size table

Descending aortic perfusion (Group D) vs. femoral artery perfusion (Group F)

Patients in Group D had minimum body temperatures during CPB of 30.59 ± 2.35 °C (95% CI 29.40 to 31.78), which were significantly different (p < 0.05) from those in Group F (23.73 ± 4.67 °C, 95% CI 21.57 to 25.89) (Table 3). The lactate levels at the start of CPB in Group D did not differ significantly from those in Group F (Group D: 2.35 ± 0.68 mmol/L, 95% CI 2.00 to 2.69 vs. Group F: 2.30 ± 1.46 mmol/L, 95% CI 1.62 to 2.98; p = 0.908).

Table 3 Comparisons between the antegrade descending aortic perfusion group (Group D) and the cardiopulmonary bypass method group with blood delivery via femoral artery cannulation (Group F)
Full size table

Group D differed significantly from Group F in terms of CPB duration (Group D: 90.53 ± 8.84 min, 95% CI 86.06 to 95.01 vs. Group F: 158.67 ± 41.64 min, 95% CI 139.43 to 177.90; p < 0.05), aortic clamp time duration (Group D: 39.93 ± 6.09 min, 95% CI 36.85 to 43.02 vs. Group F: 78.11 ± 17.67 min, 95% CI 69.95 to 86.27; p < 0.05), and total urine output (Group D: 2.67 ± 3.09 mL, 95% CI 1.10 to 4.23 vs. Group F: 8.94 ± 9.62 mL, 95% CI 4.22 to 13.62; p = 0.028).

The mean increase in lactate level during CPB did not differ significantly between Group D and Group F (Group D: 0.92 ± 1.13 mmol/L, 95% CI 0.35 to 1.49 vs. Group F: 1.06 ± 0.94 mmol/L, 95% CI 0.62 to 1.50, p = 0.807) (Table 3).

Discussion

This study aimed to evaluate the efficacy of the CPB method using blood delivery via femoral artery cannulation for paediatric aortic arch repair. We compared the degree of lactate elevation during CPB managed via circulatory arrest of the lower body under deep hypothermia, management via antegrade descending aortic perfusion via an additional cannula inserted into the descending aorta, and management via femoral artery cannulation.

In the CPB method with blood delivery via femoral artery cannulation, blood from the CPB blood reservoir is oxygenated and temperature controlled in an oxygenator and heat exchanger, respectively, before being directed to the systemic flow line. The blood pump maintains a total blood flow of 2.2–2.4 L/min/m2, equivalent to approximately 450–500 ml/min based on a body surface area of approximately 0.2 m2 for a newborn or infant requiring aortic arch repair (approximately 50 cm and 3 kg) in the DuBois formula [7]. The systemic flow line is then split into blood delivery lines to the upper and lower bodies via a Y-connector. Blood flow to the lower body was controlled at approximately 300 mL/min, with the remainder delivered to the upper body [4]. In this study, the minimum body temperature during CPB was 23.73 ± 4.67 °C (95% CI 21.57 to 25.89) in Group F, which was typical of procedures when we started using the CPB method. However, subsequent refinements in technique have enabled the maintenance of body temperature during CPB at approximately 28 °C in all patients with this CPB management. During CPB, blood pressure is monitored via an arterial pressure line inserted into the right radial artery, reflecting cerebral circulation pressure via the right common carotid artery during aortic arch repair, and is maintained at approximately 35 mmHg [4].

Our study involved two cardiac surgeons performing complex surgeries requiring aortic arch repair, such as the Norwood procedure, interrupted aortic arch repair, or repair of coarctation of the aorta. One surgeon performed paediatric aortic arch repair by circulatory arrest of the lower body, and the other used antegrade descending aortic perfusion via a cannula additionally inserted into the descending aorta to provide perfusion to the lower body organs. In approximately 2012, both surgeons transitioned to using femoral artery cannulation. Therefore, patients in Groups A and F were operated on by the same surgeon, while patients in Groups D and F were operated on by two different surgeons. Although the daily ages of patients in Groups D and F, who were operated on by different surgeons, were significantly different, there was no significant difference in the height and weight among the patients in the three groups.

Comparisons between patients in Groups A and F, who underwent identical surgeries, revealed no significant differences in CPB duration or aortic clamp time duration, suggesting that femoral artery cannulation did not impose additional surgical challenges, because the femoral artery is cannulated while anaesthesia induction.

Blood temperature during CPB is surgically controlled because it is correlated with organ metabolism and the impact of reperfusion [8, 9]. In our study, the minimum body temperature during CPB was significantly greater in Group F than in Group A, while the increase in lactate levels was significantly lower in Group F. In contrast, lactate levels at the start of CPB, aortic clamp time duration, and total urine output did not differ significantly between Group F and Group A. Lactate levels in patients in Group F could thus be managed without lowering body temperature as much as those in Group A. This suggests that CPB management using femoral artery perfusion in paediatric aortic arch repair can be performed under moderate hypothermia [10], thus preventing unnecessary deep hypothermia associated with lower body circulatory arrest.

In comparisons between Groups D and F, lactate levels at the start of CPB did not differ significantly, but patients in Group D had significantly greater minimum body temperatures during CPB than patients in Group F. Patients in Group D had larger cannulas (8 Fr in all 15 patients) inserted into the descending aorta than patients in Group F (3 Fr in nine patients < 3 kg; 4 Fr in nine patients > 3 kg into the femoral artery), allowing more flow to the lower body. The CPB and aortic clamp time durations were significantly longer in Group D, which might be due to different surgeons performing the procedures. Despite these differences, Groups D and F did not differ in terms of the degree of increase in lactate levels during CPB. Urine output during CPB was significantly greater in Group F, possibly due to the prolonged duration of CPB. Our results suggest that the use of the CPB method with femoral artery perfusion was comparable to the CPB method with antegrade descending aortic perfusion, given similar lactate elevation levels even with longer CPB durations.

Hypothermia is a risk factor for various adverse events, such as haemoconcentration, leukopenia, thrombocytopenia [11], impaired blood coagulation system [12] related to dysfunction of the coagulation cascade [13], disordered fibrinolysis, disruption of platelet function [14, 15], and acute kidney injury [16]. Hypothermia is also implicated in the development of arrhythmias [17, 18]. In addition, paediatric patients who develop arrhythmias after open-heart surgery require longer ventilatory support and intensive care unit (ICU) stay, as well as longer catecholamine support [19]. Therefore, unnecessary deep hypothermia should be avoided whenever possible. Although the minimum body temperatures during CPB reported here were significantly different between Groups D and F, our subsequent modifications have allowed CPB management at 28 °C with blood delivery via femoral artery cannulation. This method allows CPB to be conducted at temperatures similar to those achieved with antegrade descending aortic perfusion via additional cannulation without complications such as spinal cord ischaemia.

Limitation of this study was that it was a retrospective observational study reviewing the CPB records in a single centre. Therefore, the trend of lactate levels after surgery, the postoperative mechanical ventilation duration, or the length of ICU stay were not investigated. Further investigations into changes in lactate levels during CPB with blood delivery via femoral artery cannulation at our current temperature setting of 28 °C and the questions about the impact on the postoperative courses are needed.

Conclusion

CPB management with blood delivery via femoral artery cannulation offers enhanced visibility and maneuverability in the restricted surgical field of small paediatric patients and neonates undergoing aortic arch repair, even without a descending aorta cannula. Notably, this method alleviates the need for immediate post-CPB removal of the descending aorta cannula under stressful conditions when the patient’s haemodynamic status is unstable, and coagulation status is not yet normal. The femoral artery cannula can be removed once haemodynamic and blood coagulation stability is achieved in the ICU [4]. This study was the first to examine the efficacy of the CPB method with blood delivery via femoral artery cannulation for paediatric aortic arch repairs, showing superiority to CPB management with circulatory arrest of the lower body under deep hypothermia, positioning it as a promising alternative to antegrade descending aorta perfusion for paediatric aortic arch repair.