President's Message

CCAS Database Update

Pediatric
Cardiovascular
Anesthesia Fellowship Update

CCAS Charter Members

Literature Review:

A randomized clinical trial of regional cerebral perfusion versus deep hypothermic circulatory arrest:  Outcomes for infants with functional single ventricle.

Reviewer:  Denise Joffe, MD, Seattle Children’s Hospital

Background:
Early results of the Boston circulatory arrest trial and other studies stimulated cardiac surgeons to come up with improved techniques of cerebral protection during complicated congenital heart surgery.  Goldberg et al’s report from the University of Michigan is one of the early randomized controlled studies evaluating clinical neurological outcome using one technique of regional cerebral perfusion (RCP) compared to deep hypothermic circulatory arrest (DHCA).

Design: 
From October 2001 to February 2005, newborn patients scheduled for surgery with hypoplastic left heart syndrome (HLHS) or a similar variant of single ventricle were randomized to one of two groups.  Patients had the procedures performed either under DHCA or a technique using RCP via cannulation of the innominate artery with perfusion flows of 20 ml/kg/min. The total duration of bypass and circulatory arrest times were compared, as were morbidity (aortic recoarctation, ECMO, cardiac transplantation, vocal cord dysfunction, diaphragmatic paresis) and mortality.  Neurological and psychomotor exams assessing both mental development and psychomotor development (MDI and PDI) were performed prior to second stage surgery and at one year.  The differences in outcome measures between groups, based on an intention-to-treat analysis was made.  In addition, a high risk category consisting of patients less than 36 weeks gestation or less than 2.5 kg, or patients in shock or with clinical seizures pre-op or patients with pulmonary venous obstruction were randomized and also analyzed separately and within the larger cohort.

Results:
A total of 77 patients were enrolled.  38 had DHCA and 39 RCP.  23 (30%) were in the high risk group.  More patients randomized to RCP had a prenatal diagnosis (85% vs 66%).    Mean bypass and aortic cross clamp times were significantly longer in the RCP group.  CPB 129+/- 57 vs 107 +/- 49 min and cross clamp 49 +/- 12 vs 42+/- 11 min.

There were no statistical differences in survival or any other safety factors.  Results for neurodevelopment outcomes (MDI and PDI) demonstrated no statistically significant differences between groups although the results of the RCP group tended to be lower.  As a whole the cohort performed below expected.

Comment:
The major limitation of the Goldberg et al’s study is a small sample size.  This makes it difficult to interpret data with any confidence.  Despite this, Goldberg et al make an important contribution; with the present sample size there is no difference in clinical outcome.  Also, bypass times were all prolonged with the use of RCP.

Another major limitation of the study is the choice of flow rates used in patients perfused with RCP.  Studies by Andropolous et al using cerebral oximetry and cerebral Doppler have demonstrated that flow rates closer to 64 ml/kg/min were necessary to maintain adequate oxygen delivery and flows at 18ºC.  Others have demonstrated that flow rates of 20 ml/kg/min provide inadequate perfusion through the circle of Willis to the left hemisphere.  Even though the cerebral oximeter was used in this study there is no mention of any data or how the monitor was used.  Also, a short period of DHCA was required in the RCP arm of the study, although this was limited to a mean of 6 minutes.  At 18ºC, few people would argue such a short time is detrimental to neurological outcome, however, newer modifications of RCP almost eliminate the need for DHCA. 

The early prenatal diagnosis in the RCP group may have predisposed them to better, quicker resuscitation efforts preoperatively. Even with this advantage no benefit in outcome was demonstrated.  However, neurologic outcomes were limited to two tests and perhaps other clinical assessment scales would detect differences. 

Goldberg et al concluded that these results should be followed up in a multi-centered study.  Although this is necessary, there are several factors that make those recommendations very difficult.  Firstly, many centers are convinced of the benefits of RCP perhaps making it difficult to find participants.  Secondly, the bypass management techniques used around the country vary so widely that it may be difficult to standardize the DHCA arm in a multi-center study.  Thirdly, although speed and alacrity is necessary for both techniques, it is more crucial with DHCA.  Those surgeons who have abandoned the routine use of DHCA may have a disadvantage when operating on patients randomized to that group.
Clearly it is important to come up with an answer.  Controversy surrounding the benefit of the older technique of retrograde cerebral perfusion in adult patients are still ongoing.  However, many technical changes have made it safer and a useful technique in select situations.

As eluded to above, it is crucial that the technique of RCP be perfected before a thorough comparison to DHCA can be made. It seems that the most common marker of optimal perfusion is considered to be the cerebral oximeter.  However, although the present generation INVOS monitors can be useful in select situations, its use as a monitor of cerebral wellbeing may be restricted by technical limitations. (Firstly, it measures a combination of venous and arterial saturations and secondly, because of a left shift in the oxyhemoglobin dissociation curve during profound hypothermia, it does not reflect oxygen availability in the cell.)   The goal is to deliver as much “good” substrate as necessary without furthering ischemic injury or contributing to cerebral edema.

Conclusion:
The widespread adoption of RCP without adequate supporting clinical studies is similar to the widespread use of the Sano shunt.  It is based on similar logic:  from a theoretical point of view, both modifications make sense.  In both cases, studies are being done after the fact. 

In the case of RCP, few studies have randomized patients to either RCP (without periods of DHCA ) or DHCA and examined clinical neurologic outcomes.  Most studies have been retrospective, or case series and have examined surrogate markers of neurological outcome or have looked at surrogate markers of perfusion such as renal function, GI function, liver function and respiratory function.  Other studies involved neurologic outcome in animal models.  In other studies, the RCP arm included patients with extended periods of DHCA.

Although this study only addressed the effect of RCP on neurologic outcome, if further studies demonstrate that other organ systems benefit from maintaining some perfusion during repair, then that should be considered a strong advantage to RCP.

It is important to keep in mind that although early results of the Boston circulatory arrest study were quite convincing for a negative outcome of DHCA, results at 8 years no longer demonstrated significant differences in neurological outcomes between groups.  Importantly, the low flow arm of the study used flows rates of 50 ml/kg/min which is much closer to what Andropolous et al consider necessary (RCP was not necessary since the study patients were having arterial switch procedures).

Few people would argue that the “routine” use of DHCA in order to facilitate repair is justified (as for an arterial switch).  Most would agree that a duration of DHCA greater than 40 minutes is associated with a poorer neurologic outcome.  In addition, most surgeons agree that the technique of bypass management such as duration of cooling, acid base management, inspired oxygen concentration on pump, hematocrit, short periods of reperfusion, etc are crucial variables that can improve the safe limits of DHCA.  However, there is often disagreement about which factors are the most important and how they should be managed.

Purported advantages of DHCA include a greater technical ease and more precision with anastomoses, less clutter in the field and less risk of emboli.  In the early days especially, the immunologic, and hematologic advantage of not being on bypass for the extra 30-60 (or more) minutes was thought to be crucial.  However, with improved bypass and filtration techniques the latter advantage appears to be less clinically relevant.  In addition, the techniques of RCP seem to provide the surgeon with good exposure and there are no studies demonstrating the superiority of the technical repair of one technique over the other.

As with many other issues in pediatric congenital heart surgery, the subject generates intense debate and controversy.  Despite the obstacles, well designed large studies will be necessary to sort out the issues.

References:
Bellinger DC, Jonas RA, Rappaport LA et al.  Developmental and neurologic status of children after heart surgery with hypothermic circulatory arrest or low flow cardiopulmonary bypass.  N Engl J Med 1995;332:549-55

Pigula FA, Nemoto EM, Griffith BP, Siewers RD.   Regional low-flow perfusion provides cerebral circulatory support during neonatal aortic arch reconstruction,  J Thorac Cardiovasc Surg 2000;119(2):331-9

Pigula FA , Gandhi SK, Siewers RS, Davis PJ, Webber SA ,Nemoto EM.  Regional low-flow perfusion provides somatic circulatory support during neonatal aortic arch surgery. Ann Thorac Surg 2001;72(2):401-6

Jonas RA.  Comprehensive surgical management of congenital heart disease.  Oxford University Press, NY, NY 2004

Andropoulos DB, Stayer SA, McKenzie ED, Fraser CD.  Novel cerebral physiologic monitoring to guide low-flow cerebral perfusion during neonatal aortic arch reconstruction.  Journal of Thoracic & Cardiovascular Surgery, 2003, 125:491-9.

Andropoulos DB, Stayer SA, McKenzie ED, Fraser CD.  Regional low-flow cerebral perfusion provides comparable blood flow and oxygenation to both cerebral hemispheres during neonatal aortic arch reconstruction.  J Thorac Cardiovasc Surg 2003;126:1712-17.

Hofer A, Haizinger B, Geiselseder G, et al. Monitoring of selective antegrade cerebral perfusion using near infrared spectroscopy in neonatal aortic arch surgery.  Eur J Anaesthesiol. 2005 Apr;22(4):293-8.

Pochettino A,  Cheung AT .  Pro: Retrograde cerebral perfusion is useful for deep hypothermic circulatory arrest.  J Cardiothoracic Vasc Anesth 2003;17(6):764-7

Bellinger DC, Wypij D, du Plessis AS et al.  Neurodevelopmental status at eight years in children with D-transposition of the great arteries:  The Boston Circulatory arrest trial.  J Thorac Cardiovasc Surg 2003;126:1385-96

© 2007 CONGENITAL CARDIAC ANESTHESIA SOCIETY