Pitfalls in neonatal cardiac surgery using antegrade cerebral perfusion

Brain Monitoring and Protection During Pediatric Cardiac Surgery

With advances in medical care, survival after cardiac surgery for congenital heart disease has dramatically improved, and attention is increasingly focused on longterm functional morbidities, especially neurodevelopmental outcomes, with their profound consequences to patients and society. There are multiple reasons for concern about brain injury. Some cardiac defects are associated with brain anomalies and altered cerebral blood flow regulation. Brain imaging studies have demonstrated that injury to gray and white matter is quite frequent before heart surgery in neonates. Cardiopulmonary bypass and deep hypothermic circulatory arrest are associated with shortand longer-term adverse neurologic outcome. Additional brain injury can occur during the patient's recovery from surgery. Strategies to optimize neurologic outcome continue to evolve. With new technological developments, perioperative neurologic monitoring of small children has become easier, and data suggest these modalities usefully identify adverse neurologic events and might predict outcome. Monitoring methods to be discussed include processed electroencephalography, near infrared spectroscopy, and transcranial Doppler ultrasound. Alternative perfusion techniques to deep hypothermic circulatory arrest have been developed, such as regional antegrade cerebral perfusion during cardiopulmonary bypass. Other neuroprotective strategies employed during open-heart surgery include temperature regulation, acid-base management, degree of hemodilution, blood glucose control and anti-inflammatory therapies. Evidence of the impact of these measures on neurologic outcome is examined, and deficiencies in our current understanding of neurologic function in children with congenital heart disease are identified.

Monitoring of selective antegrade cerebral perfusion using near infrared spectroscopy in neonatal aortic arch surgery

BACKGROUND AND OBJECTIVE

To prevent neurological complications, low-flow antegrade cerebral perfusion (ACP) is used during repair of complex congenital heart defects. To overcome technical problems, continuous monitoring of cerebral blood flow and oxygenation is mandatory. The aim of the study was to evaluate the effect of different ACP flow rates on cerebral oxygen saturation obtained by near infrared spectroscopy.

METHODS

Ten consecutive neonates undergoing Norwood stage I were included. In addition to near infrared spectroscopy (Invos 5100; Somanetics Corp., USA) on both hemispheres, mean arterial pressure and transcranial Doppler flow velocity were measured continuously and arterial and jugular venous oxygen saturation intermittently. Cerebral oxygen extraction ratio was calculated. Measurement points were obtained after starting bypass, during ACP with flow rates of 30, 20 and 10 mL kg(-1) min(-1) and immediately after ACP. ANOVA and Tukey-Kramer multiple comparison test were used for statistics.

RESULTS

The near infrared spectroscopy signal could be obtained in all children at all measurement points, whereas transcranial Doppler failed in 1 neonate at a flow rate of 30 mL kg(-1)min(-1), in 3 neonates at 20 mL kg(-1) min(-1) and in 4 neonates at 10 mL kg(-1)min(-1). With the reduction of flow there was a significant decrease of cerebral oxygen saturation on both hemispheres (right: 78+/-8 to 72+/-9 and 66+/-8, P < 0.001; left: 71+/-7 to 65+/-7 and 60+/-7, P < 0.001), of jugular venous oxygen saturation (94+/-6 to 89+/-13 and 83+/-15, P < 0.001) and a significant increase in oxygen extraction ratio (9.1+/-8 to 14.8+/-14 and 21+/-16, P < 0.001) respectively, for 30, 20, 10 mL kg(-1)min(-1).

CONCLUSION

Near infrared spectroscopy reliably detects flow alterations during ACP with profound hypothermia.

Does antegrade cerebral perfusion protect the brain during deep hypothermic circulatory arrest?

PURPOSE

This study compares cerebral protection using no cerebroplegia and using antegrade cerebroplegia with variable flow rates during deep hypothermic circulatory arrest (DHCA).

METHODS

Twenty healthy neonatal piglets (2.5-3.8 kg) underwent 60 minutes of DHCA. No cerebroplegia was used in group 1 (n = 5). Cold (16 degrees C) antegrade cerebral perfusate was administered through the innominate artery at 10 mL/kg per minute in group 2 (n = 5), at 25 mL/kg per minute in group 3 (n = 5), and at 50 mL/kg per minute in group 4 (n = 5). Venous samples for lactate, pyruvate, S-100B protein, and creatine kinase BB (CKBB) were drawn from the jugular vein before and after discontinuation of cardiopulmonary bypass--lactate at 5 minutes postbypass, pyruvate at 5 minutes postbypass, S-100B protein at 30 minutes postbypass, and CKBB at 6 hours postbypass. Piglets were killed 6 hours postbypass and their brains were harvested for histological/immunologic studies. Extent of damage was assessed using a semiquantitative score of 0 to 4 based on a validated method.

RESULTS

Evidence for significant apoptosis and necrosis was apparent in all groups. The mean H&E score was 2.2 for group 1, 2.3 for group 2, 2.5 for group 3, and 2.3 for group 4. The mean terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling score was 1.0 for group 1, 1.2 for group 2, 1.7 for group 3, and 0.8 for group 4. Pathological changes were not greater in the piglets that did not have antegrade cerebral perfusion. Serum lactate, pyruvate, S-100B protein, and CKBB did not distinguish between perfusion strategies.

CONCLUSIONS

In neonates, unmodified antegrade cerebral perfusion at flow rates of 10, 25, and 50 mL/kg per minute during DHCA does not provide additional protection of the brain as determined by histology, immunology, serum lactate, pyruvate, S-100B protein, and CKBB.

Aortic arch reconstruction without circulatory arrest: review of techniques, applications, and indications

The majority of neonatal intracardiac repairs can now be performed with the use of low-flow cardiopulmonary bypass. However, aortic arch reconstruction still requires a period of circulatory arrest. Recently, a number of surgeons have reported techniques of limiting or completely avoiding circulatory arrest during arch reconstruction in an attempt to reduce the risk of neurologic injury. Several techniques are currently in use that have been successfully applied during biventricular repair and the Norwood operation for a wide range of aortic arch pathology including hypoplastic or interrupted aortic arch, the hypoplastic left heart syndrome, and its equivalents. These techniques maintain continuous low-flow cerebral perfusion via the innominate artery, directly or indirectly. In the last 2 years at the Montreal Children's Hospital (Montreal, Canada) we have consistently used such techniques for all arch reconstructions, including the Norwood operation, completely avoiding the use of circulatory arrest. These techniques are still in evolution with regard to flow rates, temperature, and safe period of low flow. Although the early results are encouraging, long-term follow-up with respect to neurodevelopmental outcome is essential to help us decide which techniques are optimal. This chapter provides an overview of the currently used techniques allowing cerebral perfusion during aortic arch reconstruction and summarizes our early experience at the Montreal Children's Hospital.

Neonatal aortic arch reconstruction avoiding circulatory arrest and direct arch vessel cannulation

BACKGROUND

Aortic arch reconstruction in neonates routinely requires deep hypothermic circulatory arrest. We reviewed our experience with techniques of continuous low-flow cerebral perfusion (LFCP) avoiding direct arch vessel cannulation.

METHODS

Eighteen patients, with a median age of 11 days (range 1 to 85 days) and a mean weight of 3.2 +/- 0.8 kg, underwent aortic arch reconstruction with LFCP. Seven had biventricular repairs with arch reconstruction, 9 underwent the Norwood operation and 2 had isolated arch repairs. In 1 Norwood and 7 biventricular repair patients, LFCP was maintained by advancing the cannula from the distal ascending aorta into the innominate artery. In 8 of 9 Norwood patients, LFCP was maintained by directing the arterial cannula into the pulmonary artery confluence and perfusing the innominate artery through the right modified Blalock-Taussig shunt fully constructed before cannulation for cardiopulmonary bypass. In 2 patients requiring isolated arch reconstruction, the ascending aorta was cannulated and the cross-clamp was applied just distal to the innominate artery.

RESULTS

LFCP was maintained at 0.6 +/- 0.2 L x min(-1) x m(-2) for 41.0 +/- 13.9 minutes at 18.5 degrees C +/- 1.1 degrees C. In 10 of the 18 patients, blood pressure during LFCP was 15 +/- 8 mm Hg remote from the innominate artery (left radial, umbilical or femoral arteries). In 8 of the 18 patients, right radial pressure during LFCP was 24 +/- 10 mm Hg. The mean mixed-venous saturation was 79.8% +/- 10% during LFCP. Two patients had preoperative seizures, whereas none had seizures postoperatively. One patient died.

CONCLUSIONS

Neonatal aortic arch reconstruction is possible without circulatory arrest or direct arch vessel cannulation. These techniques maintained adequate mixed-venous oxygen saturations with no associated adverse neurologic outcomes.