| ||||||||||||||
|
|
|||||||||||||

*Departments of Anesthesiology and Pediatrics, Childrens National Medical Center and George Washington University Medical Center, Washington, DC; and
AstraZeneca Pharmaceuticals, Wilmington, Delaware
Address correspondence and reprint requests to Raafat S. Hannallah, MD, Professor of Anesthesiology and Pediatrics, George Washington University Medical Center, Chairman of Anesthesiology, Childrens National Medical Center, 111 Michigan Ave., N.W., Washington, DC 20010.
| Abstract |
|---|
|
|
|---|
Implications: The addition of EDTA does not alter the clinical profile of propofol inpediatric ambulatory surgical patients. With or without EDTA, propofol isassociated with a decrease in ionized calcium with no apparent clinicaleffect.
| Introduction |
|---|
|
|
|---|
As a result of these reports and other studies showing a frequent incidence of handling errors among anesthesia personnel (6,8,9), 0.005% EDTA, an antimicrobial agent, was added to the formulation of propofol (Diprivan®; AstraZeneca, Wilmington, DE) to help prevent accidental extrinsic contamination. Although the addition of EDTA should reduce the risk of complications secondary to accidental extrinsic bacterial contamination, EDTA is a strong chelator of cations and induces hypocalcemia, acute toxicity, and fatalities in dairy cows (10). Its use may have physiologic repercussions (e.g., apoptosis and decreased serum ionized calcium and ionized magnesium levels) (11), particularly in pediatric patients.
The objective of this study was to determine whether the addition of EDTA affects the clinical profile of propofol, calcium and magnesium homeostasis, renal function, or the incidence of adverse events in children undergoing ambulatory surgery.
| Methods |
|---|
|
|
|---|
Anesthesia was induced with a mixture of nitrous oxide (6 L/min), oxygen (3 L/min), and halothane, administered initially at a concentration of 0.5% and titrated upward to a concentration of 3.0%. End-tidal halothane concentrations were recorded at 5, 10, and 15 min after induction.
Two IV catheters, one for drug administration and the other for serial blood sampling, were inserted into each patient. Muscle relaxation was achieved with vecuronium (up to 0.2 mg/kg). Patients received fentanyl 1 to 2 µg/kg, a regional block, or both, as clinically indicated. Other nonsedating medications that were not expected to interfere with the study drug or affect study measurements (e.g., antibiotics) were also given as indicated.
After induction, halothane was discontinued (after a minimum of 5 min), and propofol or propofol EDTA was administered at a minimum rate of 200 µg · kg-1 · min-1 for the first 15 min, with subsequent titration to maintain heart rate and blood pressure within 20% of baseline values. All patients received oxygen (1 L/min) and nitrous oxide (2 L/min) during maintenance. Standard monitors, including three-lead electrocardiogram, Dinamap blood pressure cuff (Critikon, Tampa, Florida), pulse oximeter, and capnograph, were used. Ventilation was controlled to maintain end-tidal CO2 at 35 ± 5 mm Hg.
Hemodynamic measurements (heart rate, systolic and diastolic blood pressure, and mean arterial pressure), respiratory rate, and oxygen saturation were recorded at baseline. These variables, along with end-tidal CO2, were also measured at 1, 2, 3, 4, 5, 10, 15, 30, 45, and 60 min after infusion, every 15 min throughout recovery, and at the time of complete recovery.
Blood samples for assay of serum EDTA concentrations were obtained before study drug infusion, at infusion termination, and at 30 min into recovery. At each specified time point, a 3-mL venous blood sample was collected, allowed to stand for 30 min, and then centrifuged at 3000 rpm for 10 min at 4°C. The serum was harvested, frozen immediately, and maintained at a temperature of 20°C until it was shipped for assay. The concentration of EDTA was measured with a validated assay (PHOENIX method AL-S-1714-01) composed of liquid-liquid extraction, derivatization, and solid-phase extraction, followed by capillary column gas chromatography with mass selective detection.
Additionally, a 0.5-mL venous blood sample for the measurement of ionized calcium and ionized magnesium levels was collected at baseline (before infusion of study drug); at 5, 10, 15, and 30 min of study drug infusion; at the termination of infusion; and at 30 min into recovery. Analyses of ionized calcium and ionized magnesium were performed with an AVL 988-4 Electrolyte Analyzer (PSS Worldmedical, Jacksonville, FL).
Venous blood samples for the measurement of sodium, potassium, blood urea nitrogen (BUN), creatinine, uric acid, inorganic phosphate, and hydrogen ion concentration (pH) were also collected before and at the termination of study drug infusion. To ensure patient safety, a cumulative total of no more than 5% of a patients intravascular volume was collected during the study.
Recovery characteristics were assessed by time to extubation, spontaneous eye opening, verbal command response, and complete recovery (defined as the achievement of two consecutive scores of 6 on the Steward Recovery Scale). Assessments were performed every 5 min for the first 20 min and then every 15 min until discharge (12). All adverse events were fully documented and assessed.
The study was designed to have >99% power to detect a 0.1 mmol/L difference in ionized calcium between treatment groups. The estimated SD in ionized calcium between patients was 0.13 mmol/L (13). With a planned sample size of 20 patients per treatment group, the power in this study to detect a 0.1 mmol/L difference in ionized calcium with an
level of 0.05 was >0.99. All P values were based on two-tailed t-tests; no adjustment was made for multiple comparisons. A P value of 0.05 was considered statistically significant. Randomized but untreated patients were not included in any analysis. All patients who received one of the study drugs, regardless of the infusion duration or quantity administered, were included in the safety analysis.
Demographic variables were compared between treatment groups by using Wilcoxons ranked sum test for continuous variables and
2 analysis or Fishers exact test for discrete variables.
Hemodynamic variables were analyzed for changes from baseline by use of analysis of covariance, with baseline values as covariates. Anesthetic effectiveness was estimated by comparing time to extubation, time to spontaneous eye opening, time to verbal command response, and time to complete recovery, by means of analysis of covariance. Covariates included age and duration of anesthesia.
Analyses of chemical (ionized calcium and magnesium) and biochemical variables (BUN, creatinine, sodium, potassium, uric acid, inorganic phosphate, and pH) were performed on changes from baseline by using analysis of covariance, with age, duration of anesthesia, and baseline values as covariates. In addition, paired t-tests were used to determine within-group changes from baseline for levels of ionized calcium and ionized magnesium. Time profiles of group means for laboratory and hemodynamic variables were plotted to determine specific trends.
| Results |
|---|
|
|
|---|
|
Hemodynamic measurements generally remained stable during the study period and were similar for both groups, with no age-related differences. Statistically significant differences, however, were found for changes in systolic blood pressure, mean arterial pressure, and heart rate at various time points.
For systolic blood pressure, the mean change from baseline at 5 and 10 min after the start of infusion was significantly smaller (P = 0.02) for the Propofol group compared with the Propofol EDTA group (Fig. 1). The Propofol EDTA group experienced significantly smaller changes from baseline in mean arterial blood pressure at 5 min of infusion (P = 0.049; Fig. 2) and at the time of surgical incision. The decreases in heart rate from baseline were significantly more in Propofol patients compared with Propofol EDTA patients at 45 and 60 min after the start of infusion (P = 0.01 and 0.03, respectively) (Fig. 3). None of these reported differences, however, was considered to be clinically significant.
|
|
|
|
When ionized calcium levels were examined with respect to EDTA concentration and the propofol dose was adjusted for body surface area, no correlation was found.
Mean ionized magnesium levels were similar between treatment groups throughout the study, remaining close to or fluctuating slightly below baseline. Differences between groups were not statistically significant. No statistically or clinically significant differences between treatment groups were found for BUN, creatinine, potassium, sodium, inorganic phosphate, uric acid, or pH.
Recovery times were similar between those who received propofol and those who received propofol EDTA (Table 3). Rapid emergence and recovery were consistent, with extubation occurring at approximately 10 min after termination of infusion and complete recovery occurring at about 25 min. No statistically significant differences were noted between the two treatment groups.
|
| Discussion |
|---|
|
|
|---|
A previous pharmacokinetic study found that the addition of EDTA did not significantly affect serum concentrations of propofol (14), and as a result, EDTA was not expected to alter propofols efficacy. This study extends the results of these earlier pharmacokinetic studies by demonstrating that the addition of EDTA had no discernible effect on the efficacy of propofol as a maintenance anesthetic. Both Propofol and Propofol EDTA patients used comparable amounts of the drug for a similar duration and at a similar rate. Rapid emergence and recovery were observed in each study group and were comparable to those previously reported (15).
EDTA infusion totaling 40 mg/kg can be used for emergency treatment of hypercalcemia (16). Calcium-EDTA infusion totaling 50 to 75 mg/kg has long been the standard treatment for patients with lead poisoning (17). However, chelation with EDTA can cause hypocalcemia, which in turn can lead to several undesirable sequelae (e.g., tetany, seizures, cardiac arrhythmias, myocardial depression, and respiratory arrest) (18). Because hypocalcemia occurs during stress (19), the administration of EDTA during surgery is a concern. However, the results of this study show that 0.005% EDTA did not affect the safety profile of propofol (i.e., ionized calcium and magnesium levels, laboratory indicators of renal function, hemodynamic values, and incidence of adverse events). Exposure to EDTA in this study is negligible when compared with therapeutic use.
Ionized calcium and ionized magnesium levels were decreased by both formulations of propofol used in this study, which would seem to indicate that EDTA had no measurable effect of these levels. The significant decrease in ionized calcium seen in both groups is difficult to explain but is probably caused by the dilutional effect of fluid administration during surgery. Ionized calcium levels remained depressed throughout infusion but returned to normal without treatment within 30 minutes of discontinuing propofol. No patient, even those who were hypocalcemic by laboratory definition, demonstrated any adverse clinical effects secondary to this phenomenon. The clinical importance of these slightly decreased ionized calcium levels in healthy patients seems to be of minimal significance. However, patients with compromised cardiac function, such as those in intensive care units or undergoing cardiovascular surgery, may be at increased risk. Infants, who are more dependent on extracellular calcium, may also require special consideration. Additional studies are needed in these patient populations.
EDTA serum levels were nontoxic. Renal function remained unchanged in both patient groups, with sodium, potassium, uric acid, inorganic phosphate, and pH levels showing no significant differences. Any long-term renal effects cannot be described in this study.
Oxygen saturation was similar in both groups throughout the study, and overall, hemodynamic values remained stable during maintenance of anesthesia. There were statistically significant hemodynamic differences between the two groups at various times during the infusion, but these were not clinically significant and might reflect only the normal responses to different levels of surgical stimulation. These hemodynamic variations were not related to changes in serum calcium levels. Overall, the decreases in heart rate seen in this study were similar to those previously reported in adults and children (15). The significantly lower heart rates occurred later in the infusion time course and may reflect a decrease in surgical stimulation.
Anesthesia is often induced in children with an inhaled anesthetic before the initiation of maintenance anesthesia (20). The method of induction used in this study reflects standard clinical practice in children at the time of study design. Sevoflurane has since become a more popular induction drug. Although it was not an objective of the study, the effects of halothane on propofol maintenance were examined, particularly with respect to the marked hypotensive episodes seen early in propofol infusion. The timing of these occurrences did not correlate with decreased ionized calcium or increased EDTA levels. Instead, these periods of hypotension seem to be related to the sequential exposure of patients to halothane for induction, followed by propofol for maintenance. Hypotension occurring during propofol infusion has been previously reported (15), but the patients in those reports differed in that anesthesia was induced with propofol, patients had higher infusion levels, and hypotension was seen later. In this study, halothane exposure, combined with avoidance of anticholinergic drugs, may have resulted in an earlier decrease in blood pressure. Other adverse events seen in this study were generally mild or moderate and were typical of those previously noted with the use of propofol (7).
Preparations of propofol that contain other antimicrobial agents, such as sulfide, are now available (Gensia Sicor, Irvine, CA). As noted previously, this study, designed and executed between 1995 and 1996, predates the introduction of these products. No comparisons can therefore be made.
In conclusion, the addition of EDTA to propofol does not affect the clinical profile of propofol. Although both formulations of propofolwith and without EDTAproduced a statistically significant decrease in concentrations of ionized calcium during infusion, there was no apparent clinical effect on this population of ambulatory surgery patients. Further studies are needed to examine the effects of EDTA in critically ill patients and infants.
| Acknowledgments |
|---|
The authors would like to acknowledge the efforts of Janet M. Norden, RN, MSN, who was the research nurse coordinator for this study.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
D. N. Cornfield, K. Tegtmeyer, M. D. Nelson, C. E. Milla, and M. Sweeney Continuous Propofol Infusion in 142 Critically Ill Children Pediatrics, December 1, 2002; 110(6): 1177 - 1181. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|