UNLABELLED A recent study demonstrated that changes in both heart rate (HR; positive if > or = 10bpm increase) and T-wave amplitude (positive if > or = 25% increase) reliably detect accidental intravascular injection when a full test dose containing epinephrine 0.5 microg/kg is injected intravascularly. We designed this study to prospectively determine whether a smaller dose of epinephrine would produce reliable HR and T-wave changes in sevoflurane-anesthetized children. We studied 80 ASA physical status I infants and children (6-72 mo) undergoing elective surgeries during 1.0 minimum alveolar anesthetic concentration sevoflurane and 67% nitrous oxide in oxygen. After the administration of i.v. atropine 0.01 mg/kg, the patients were randomly assigned to receive either i.v. saline (n = 20), an i.v. test dose (0.1 mL/kg) consisting of 1% lidocaine with 1:200,000 epinephrine (epinephrine 0.5 microg/kg group, n = 20), an i.v. test dose (0.05 mL/kg) (epinephrine 0.25 microg/kg group, n = 20), or an i.v. test dose (0.025 mL/kg) (epinephrine 0.125 microg/kg group, n = 20) via a peripheral vein to simulate the intravascular injection of the test dose. HR and systolic blood pressure were recorded every 20 and 30 s, respectively, and T-wave amplitude of lead II was continuously recorded for subsequent analysis. After the i.v. injection of the test dose, all children in the epinephrine 0.5 and 0.25 microg/kg groups developed positive responses based on the peak T-wave amplitude, whereas all children in the epinephrine 0.5 microg/kg group and 17 children (85%) in the epinephrine 0.25 microg/kg group elicited a positive response according to the peak HR criterion. No false-positive responses were observed with saline injections. Children in the epinephrine 0.125 microg/kg group showed clinically unacceptable efficacy based on either criterion. We conclude that the efficacies of detecting an intravascular injection of the test dose based on the hemodynamic and T-wave criteria are reduced with smaller doses of epinephrine and that HR and T-wave changes are still useful indicators in most patients if epinephrine 0.25 microg/kg is accidentally injected intravascularly. IMPLICATIONS To determine whether an epidurally administered local anesthetic has been unintentionally injected into a blood vessel, a small dose of epinephrine is often added to a local anesthetic. We found that an increase in T-wave amplitude > or = 25% in lead II and a heart rate increase > or = 10 bpm are useful indicators for detecting the accidental intravascular injection of a small dose of epinephrine in sevoflurane-anesthetized children.