Hemodynamic Responses to Intravascular Injection of Epinephrine-containing Epidural Test Doses in Adults during General Anesthesia

After institutional review board approval and informed consent were obtained, 36 ASA physical status 1 or 2 patients scheduled to undergo general anesthesia for elective surgery were enrolled in this study. Exclusion criteria included age younger than 18 yr or older than 45 yr, history of diabetes mellitus or cardiovascular disease, use of medications affecting the cardiovascular system, daily consumption of alcohol in excess of two beverages per day, or current use of recreational drugs.

All patients were premedicated with 2 mg midazolam and 100 micro gram fentanyl and were randomized to receive 0.5 MAC isoflurane, 1 MAC isoflurane, or 1 MAC total isoflurane and nitrous oxide (0.5 MAC each). The randomization was stratified such that each group consisted of six men and six women. Before induction of anesthesia, heart rate from the electrocardiogram (averaged every eight heart beats) and noninvasive blood pressures (model 9030570, Spacelabs Medical, Redmond, WA) were measured. The blood pressure cuff was placed on the arm without an intravenous catheter. Anesthesia was induced with 4 mg *symbol* kg sup -1 thiopental and muscle relaxation achieved using 0.1 mg *symbol* kg sup -1 vecuronium. The trachea was intubated and ventilation controlled to maintain an end-tidal carbon dioxide tension between 30 and 35 mmHg. Patients' lungs were ventilated at 6–8 breaths/min, with a tidal volume of 10–15 ml *symbol* kg sup -1. Anesthesia was maintained with stable end-tidal concentrations of 0.5 MAC isoflurane (0.6 vol%), [10]1 MAC isoflurane (1.2 vol%), [10]or 0.5 MAC isoflurane with 0.5 MAC nitrous oxide (52%)[11]for 15 min. End-tidal gas concentrations were determined with an infrared end-tidal gas analyzer (Poet IQ model 602–6A, Criticare Systems, Milwaukee, WI). No other medications were administered. Body temperature was measured via an esophageal stethoscope (Mon-a-therm, Mallinckrodt, St. Louis, MO). Four test solutions were administrated to each subject. First saline (3 ml), then in a randomized, double-blind fashion, three test doses of 45 mg lidocaine (3 ml 1.5% lidocaine) containing 7.5 mg, 15 mg, and 30 micro gram epinephrine were injected intravenously. Each injection was performed over 3 s, the most proximal injection port was used, and intravenous fluid was administered at maximal gravitational flow rates. Heart rate was recorded from the electrocardiogram immediately before and for 5 min after injection of each test solution (saline and test doses) at 15-s intervals. Blood pressure cycling was set to every minute, initiated immediately before injection, and continued for 5 min after injection of each test solution. Intravenous fluid administration was standardized such that 750 ml lactated Ringer's solution was administered before injection of study solutions, and an additional 250 ml was infused during the study period. Surgery commenced after the study period, and patients were left undisturbed during the study period.

Demographics were compared with analysis of variance. Baseline heart rate and blood pressure for each test solution were defined as the measurements obtained before injection. Peak measured increases in heart rate and blood pressure were defined as the difference between the baseline values and the greatest measured values after injection of each test solution. Onset of peak measured increases was the time at which these measurements were obtained. Ninety-five-percent confidence intervals applicable to 99% of the general population [12]were calculated for peak increases in heart rate and systolic (SBP), diastolic (DBP), and mean (MBP) blood pressures after injection of saline from all subjects in each anesthetic group. Hemodynamic increases greater than these 95% confidence intervals were defined as positive hemodynamic criteria for detection of intravascular injection of test doses and were determined for each anesthetic group. Dose-effect relationships between epinephrine and peak measured increases in heart rate and SBP, DBP, and MBP were analyzed with linear regression within each anesthetic group. Minimum required doses (MRDs) of epinephrine producing mean peak hemodynamic changes with 95% confidence intervals that exceeded positive hemodynamic criteria were determined from linear regression. Bonferroni correction was used to correct P values for multiple comparisons. A corrected P < 0.05 was considered significant.

Age, height, weight, preinduction hemodynamics, and patient temperature and baseline hemodynamics obtained before injection of test solution did not differ between anesthetic groups (Table 1). Significant dose-effect relationships were identified for epinephrine-induced peak measured increases in heart rate and blood pressures for each anesthetic group (Figure 1, Figure 2, Figure 3, Figure 4). Criteria for positive heart rate and SBP response to intravascular injection of epinephrine-containing test doses and MRDs of epinephrine for production of positive responses were determined (Table 2). Times of onset of peak measured hemodynamic changes were determined (Table 3).

The first goal of our study was to determine hemodynamic criteria for accurate identification of intravascular injection of an epinephrine-containing test dose during general anesthesia in the absence of epidural anesthesia. Previous hemodynamic criteria have been derived exclusively from awake subjects. [3]Positive criteria were determined for each anesthetic group and calculated to apply to 99% of the general population of healthy, young adults undergoing general anesthesia with isoflurane and nitrous oxide. In such a population, increases in heart rate greater or equal to 8 beats/min, SBP greater or equal to 13 mmHg, DBP greater or equal to 7 mmHg, or MBP greater or equal to 9 mmHg will identify intravascular injection of epinephrine. Other populations, such as the elderly, [13]and other general anesthetics [14]may have different cardiovascular profiles and thus require separate determination of hemodynamic criteria. Our positive criteria derived from healthy adults during general anesthesia with isoflurane suggest that smaller magnitude of hemodynamic changes can be used than are typically used for awake, healthy adults (i.e., increase in heart rate greater or equal to 20 beats/min and SBP greater or equal to 15 mmHg). [3]This is likely due to lesser magnitude of hemodynamic variability observed during general anesthesia with isoflurane. We speculate that decreased hemodynamic variability was due to the direct and indirect cardiovascular depression seen after administration of isoflurane and nitrous oxide. Administration of isoflurane results in dose-dependent reduction in heart rate, [15]systemic vascular resistance, [16]and myocardial contractility. [15]Isoflurane also has indirect cardiovascular effects via dose-dependent attenuation of the sympathetic nervous system. [17–19]Similarly, administration of nitrous oxide also produces dose-dependent cardiac depression [20]and vasodilation. [21]Other possible explanations for decreased hemodynamic variability include immobility, negative chronotropic effects of fentanyl and midazolam, and lack of environmental stimulation. In summary, our findings indicate that increases in heart rate greater or equal to 8 beats/min, SBP greater or equal to 13 mmHg, DBP greater or equal to 7 mmHg, or MBP greater or equal to 9 mmHg will identify intravascular injection of epinephrine in the general population of healthy young adults during general anesthesia with isoflurane and nitrous oxide.

The second goal of our study was to determine hemodynamic responses to epinephrine-containing test doses during general anesthesia in the absence of epidural anesthesia. Our data demonstrate that peak, measured hemodynamic responses vary with depth and type of general anesthesia. During anesthesia with 0.5 MAC isoflurane, MRDs of epinephrine required to produce mean peak hemodynamic changes exceeding positive criteria ranged from 7 to 15 micro gram. Thus, intravascular injection of the standard epidural test dose containing 15 micro gram epinephrine should, on average, produce positive changes in heart rate and blood pressures. However, during anesthesia with 1 MAC isoflurane, 17–19 micro gram epinephrine is the predicted MRD for production of positive mean increases in heart rate, DBP, and MBP.

Thus, the standard 15-micro gram epinephrine test dose probably will not produce sufficiently large increases in these hemodynamics, on average, for detection of intravascular injection. Use of an epinephrine dose greater than 15 micro gram during 1 MAC isoflurane should enhance identification of intravascular injection but may increase the likelihood of cardiac dysrhythmias. [22]Alternatively, the MRD of epinephrine predicted to produce positive increases in SBP during 1 MAC isoflurane is 13 micro gram; thus, increases in SBP may be monitored for identification of intravascular injection.

In contrast, addition of 0.5 MAC nitrous oxide to 0.5 MAC isoflurane also resulted in 1 MAC total anesthesia, yet did not reduce hemodynamic responses to epinephrine (MRDs ranged from 6 to 12 micro gram). Because isoflurane [15]and nitrous oxide [20]directly depress cardiovascular function, different effects of nitrous oxide and isoflurane on the sympathetic nervous system may explain this finding. Isoflurane is a potent inhibitor of the sympathetic nervous system, [17–19,23]whereas nitrous oxide preserves and may enhance sympathetic function. [17,24,25]Thus, the relative preservation of sympathetic function with administration of nitrous oxide may preserve hemodynamic responses to epinephrine.

There are several limitations to our study. First, our data cannot identify a specific epinephrine dose that is 100% sensitive and specific. Rather, MRDs should produce, on average, mean hemodynamic changes that exceed positive criteria. Individual patient responses will vary, and not all patients will experience positive hemodynamic increases after intravascular injection of an MRD. We encourage readers to measure multiple hemodynamic variables to improve sensitivity of the epinephrine test dose for detection of unintentional intravascular injection. Another limitation is the relatively small number of subjects enrolled. However, the significance of our findings should be enhanced by the study design. Each subject was used as their own control, and generalized applicability of findings was improved by the use of 95% confidence intervals and a dose-effect study design. Our use of intermittent and noninvasive measurement of blood pressure may have decreased the sensitivity of our results, yet may have enhanced the clinical value of our findings because of the common use of oscillometric blood pressure monitoring. We note that previous studies examining epidural test doses have used similar methodology. [4,26–28]The standardized administration of fentanyl and midazolam as premedication may have affected patients' responses to epinephrine, but administration of pre-medicants is common clinical practice. Finally, our study was performed without the influence of surgery or epidural anesthesia. Surgical stimulation would be expected to provoke hemodynamic stimulation, and epidural anesthesia may decrease sympathetic tone [29,30]depending on the extent of neural block [31]and type of local anesthetic used. [32]Either situation could alter our results.

In conclusion, MRDs of epinephrine to produce mean positive hemodynamic increases vary with depth and composition of general anesthetic. The standard test dose with 15 micro gram epinephrine should produce, on average, multiple positive hemodynamic responses identifying intravascular injection during anesthesia with 0.5 MAC isoflurane with or without 0.5 MAC nitrous oxide. However, during general anesthesia with 1 MAC isoflurane, 15 micro gram epinephrine is probably insufficient to produce positive increases in heart rate, DBP, or MBP, and either a larger dose of epinephrine should be used or SBP should be monitored.