In Reply:-We appreciate the comments by Drs. Hammer and Krane. Indeed, we believe that the pressurized infusion was the main cause of the localized arterial hypoxemia, because the high pressure reversed the normal arterial/venous pressure gradient, causing the transfusion to flow directly across the shunt. The shunting was so immediate and dramatic that we immediately suspected an anatomic shunt rather than retrograde flow through capillary beds. This was mentioned as a possible theoretical explanation for venoarterial admixture in general. The postoperative arteriography demonstrated the suspected abnormal anatomic venoarterial connection.

It is likely that all pediatric anesthesiologists believe that small distal veins should not be used for rapid transfusion. However, in this case, both peripheral lines were 18-gauge catheters. These are by no means small catheters, particularly in an 8.7-kg patient! Why were they placed so distally? Because these distal veins were much larger than the antecubital veins in this patient, possibly because of arterialization of the veins secondary to the anatomic shunts.

According to Poiseuille's law, resistance to flow is affected by both radius and length of the catheter used. An 18-gauge, 3.2-cm catheter will allow much more rapid transfusion than an 18-gauge 8-cm catheter; in situations involving massive hemorrhage, this difference in flow can be critical.

We appreciate reiteration of the point that calcium chloride can be very irritating to small veins and tissues; when distal administration of calcium is considered, calcium gluconate is less damaging.

Kristine Henderson, M.D.

Bayou Anesthesia Associates; Fort Walton Beach, Florida 32547;

(Accepted for publication March 10, 1999.)