Literature Review
Optimal blood gas management during deep hypothermic paediatric cardiac surgery: alpha-stat is easy, but pH-stat may be preferable.
Laussen PC. Paediatric Anaesthesia; 2002; 12:199-204
We have here an excellent, concise, and easy to read review article on blood gas management during deep hypothermic pediatric cardiac surgery. I will attempt to summarize it, but I do recommend that you read the original.
Hypothermia remains the lodestone of brain protection during cardiopulmonary bypass - but there is a price to pay. Hypothermia increases blood viscosity, which impedes microcirculation. Hemodilution helps, but there is no consensus for the degree of hemodilution to be used. Hypothermia reduces cerebral metabolic rate, but also causes cerebral vasoconstriction. Hypothermia shifts the oxygen dissociation curve to the left, and there is increased CO2 solubility and alkalosis, resulting in a further shift of the oxygen dissociation curve to the left. Thus it is possible that there will be impaired unloading of O2 to the cerebral cells. Twenty years ago, neurological complications, namely seizure disorders, occurred in between 2 and 25% in children undergoing cardiopulmonary bypass. There has been a steady improvement in morbidity, even with an ongoing debate about how best to achieve neurological protection. In a recent report from Children’s Hospital, Boston, the incidence of seizures was reported as < 1% in neonates and infants following deep hypothermia and either low flow bypass or circulatory arrest. The low incidence of complications makes it very difficult to select variables that can define outcome for a particular intervention. Rather, late outcome variables such as neurophsyiological development may be more significant, although in some cases the differences may be only subtle. This problem is apparent in studies evaluating blood gas management.
At first, though there were no critical studies, temperature corrected values for arterial pH and CO2, i.e. pH-stat, were used during hypothermia. As CO2 solubility increases with hypothermia, the partial pressure decreases and CO2 is added to maintain normal pH and CO2. This results in relative hypercarbia and acidemia. Arguments against pH-stat centered on the potential for impaired cellular function following reperfusion and associated increased cerebral blood flow.
Following analysis of comparative physiology in cold-blooded vertebrates and invertebrates, the alpha-stat method became widely used. Here, pH and CO2 were maintained at normal levels as measured at 37 C, regardless of core body temperature. This method results in relative alkalosis, but it maintains electrochemical neutrality. The intracellular pH remains constant, and this may have a beneficial effect on protein structure and function, enzyme activity, and substrate ionization.
Prospective studies of neuropsychometric outcome after cardiac surgery in adults have supported the more alkaline alpha-stat approach. However, the mechanisms of cerebral injury in adults and young children are different. In adults, the injury relates to atheromatous emboli or vascular stenosis. In contrast, in neonates and infants, brain injury results from global hypoperfusion, more so during periods of decreased or arrested systemic blood flow.
Cerebral blood flow is increased with pH-stat strategy. This may increase the risk of microembolism, but at the same time improved global and regional cooling from vasodilatation is an advantage. With the pH -stat method there is less “steal” of cerebral blood flow. Pulmonary atresia with aortopulmonary collateral blood vessels is known to be a risk factor for developing choreoathetosis. Therefore, it was suggested that the alpha-stat method was responsible for the steal of blood from the cerebral tothe pulmonary circulation because of the opposing effects of alkalosis on cerebral and pulmonary resistance.
Hypothermia reduces cerebral oxygen consumption in an exponential fashion and blood flow in linear fashion, therefore, in children the rate of flow to metabolic rate increases with hypothermia. It has been suggested that this effect, rather then reflecting luxuriant flow (high venous oxygen saturation), may reflect impaired oxygen transfer because of leftward shift of the oxyhemoglobin dissociation curve and that the pH-stat strategy may improve oxygen delivery. Recently, in piglet studies, it was demonstrated that the pH-stat strategy was associated with improved functional neurological outcome. This finding was based on brain histology following sacrifice revealing significant injury in the alpha-stat group.
The intracellular acidosis that develops during deep hypothermic cardiac arrest recovers more rapidly with pH-stat than with alpha-stat, and there is no evidence that such acidosis delays or prevents full brain recovery. Studies with newborn animal models have demonstrated hypercapnia to be neuroprotective against ischemic injury.
In a prospective, randomized study was carried out on 182 newborn infants at the Children’s Hospital, Boston, comparing alpha-stat versus pH-stat during deep hypothermic bypass. Early subtle benefits were noted in the pH-stat group, but the use of either strategy for acid-base management was not consistently related to either improved or impaired neurodevelopmental outcome at one year of age.
Based upon animal data, the pH-stat method for blood gas management during deep hypothermic cardiopulmonary bypass appears to be a strategy that is more appropriate. However, as demonstrated by recent clinical data, it is difficult to reconcile laboratory findings with clinical outcomes.
Optimal neurological protection depends on a number of variables. The blood gas strategy used may have a significant additive effect, depending on the combination of the variables. The degree of hypothermia is important. At mild or moderate hypothermic temperatures, there is little change in pH, and an alpha-stat strategy is appropriate; while at deep hypothermic levels with low flow or circulatory arrest, an alpha-stat strategy is appropriate. In order to gain more definitive insights, and to control for the many variables, a large study to achieve sufficient power is required, as was suggested by the author of the original paper.
Reviewed by: Hoshang J. Khambatta, MD #include ./footer_include.iphtml