The efficacy of simulated intravascular test dose in sedated patients.

Abstract

Sedation usually decreases the reliability of subjectively detecting an intravascular test dose, but the efficacies of objective hemodynamic and T-wave criteria remain undetermined. Sixty healthy patients were randomly assigned to receive IV midazolam in 1-mg increments until they were lightly sedated, fentanyl 2 microg/kg followed by incremental midazolam until they were similarly sedated, or no sedative (n = 20 each). Then, normal saline 3 mL was administered IV, followed 4 min later by 1.5% lidocaine 3 mL plus epinephrine 15 microg (1:200,000) in all subjects. Heart rate (HR), systolic blood pressure (SBP) measured by a radial arterial catheter, and lead II of the electrocardiogram were continuously recorded for 4 min after the saline and test dose injections. An IV test dose produced significant increases in HR and SBP and decreases in T-wave amplitude in all subjects. However, the mean maximum increase in HR in patients sedated with midazolam plus fentanyl (31 +/- 14 bpm [mean +/- SD]) was significantly less than in those administered midazolam alone or no sedative (42 +/- 12 and 44 +/- 10 bpm, respectively; P < 0.05). A sensitivity of 100% was obtained on the basis of the traditional HR criterion (positive if > or =20 bpm increase) in patients sedated with midazolam or no sedative, but it was 70% in those with midazolam plus fentanyl (P < 0.05 versus the other two groups). Irrespective of the treatment, sensitivities and specificities of 100% were obtained according to the SBP (positive if > or =15 mm Hg increase) and T-wave (positive if > or =25% decrease in amplitude) criteria. An increase in SBP and a decrease in T-wave amplitude are more reliable than an HR response for detecting accidental intravascular injection of the epinephrine-containing test dose in subjects sedated with midazolam and fentanyl. To determine whether an epidural catheter is in a blood vessel, an epidural test dose containing 15 microg epinephrine is used. We found that an increase in systolic blood pressure and a decrease in T-wave amplitude seem to be more reliable than a heart rate change for detecting inadvertent intravascular injection of an epinephrine-containing test dose in patients sedated with midazolam and fentanyl.