BIOPHARMACEUTICS & DRUG DISPOSITION Biopharm. Drug Dispos. 25: 297–301 (2004) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/bdd.417 Characterization of Early Plasma Concentrations of Midazolam in Pigs after Administration by an Autoinjector Aharon Levy*, Moshe Kushnir, Shira Chapman, Rachel Brandeis, Zvi Teitelbaum and Eran Gilat Department of Pharmacology, Israel Institute for Biological Research, Ness Ziona, POB 19, 74100, Israel ABSTRACT: The treatment of organophosphate-induced poisoning is based mainly on atropine and an oxime. Prompt anticonvulsive intervention is usually also required to terminate the ensuing seizure activity and to prevent delayed permanent brain damage. Midazolam, a water-soluble benzodiazepine agonist, has the advantage of rapid absorption following intramuscular administration. In mass casualty situations, the availability of an autoinjector, filled with midazolam, might be a further advantage. In the present study, the plasma pharmacokinetics of midazolam after administration by an autoinjector was compared with conventional intramuscular (i.m.) administration in two groups of four pigs each. During the first 15 min after injection, significantly higher plasma concentrations of midazolam were detected following autoinjector administration, compared with the i.m. injection. The physiological reflection of the accelerated midazolam absorption was a marked reduction in the time interval required for muscle relaxation, induced by midazolam. It is concluded that a midazolam autoinjector might be helpful in the mass casualty scenario following organophosphate poisoning. Copyright # 2004 John Wiley & Sons, Ltd. Key words: anticonvulsants; benzodiazepines; organophosphates; sedation; seizures Introduction soluble benzodiazepine agonist, has the advan- tage of a higher bioavailability and a more rapid One of the most important elements for the absorption and onset of activity, following successful treatment of organophosphate (OP) intramuscular (i.m.) administration [6]. There- cholinesterase-inhibitor poisoning is its prompt fore, the drug has been suggested as a possible application. While traditional routine therapy is replacement for diazepam in the US army [7]. based on the administration of the muscarinic Midazolam has a further advantage when ad- blocker atropine and an oxime that reactivates ministered through the nasal route [8,9], the inhibited enzyme, rapid intervention with an although at this point, at least for battlefield anticonvulsant for the quick termination of application, this route does not seem practical seizure activity is also often required [1,2]. yet. In a comparison between the various Inefficient or delayed anticonvulsive treatment benzodiazepines for the control of soman- may lead to permanent brain damage [3–5]. induced seizures in guinea-pigs, midazolam was Diazepam (or its pro-drug avizafone) is used in found to be the most potent and rapidly acting the treatment strategy of various countries. [10]. Midazolam also had a better performance Compared with diazepam, midazolam, a water- when compared with diazepam in monkeys [11]. A few papers in the literature reported the different pharmacokinetics of drugs following *Correspondence to: Department of Pharmacology, Israel Institute for Biological Research, Ness Ziona, P.O. Box 19, 74100, Israel. intramuscular administration using an autoinjec- E-mail: [email protected] tor. In allergic children, epinephrine absorption Received 12 August 2003 Revised 27 November 2003 Copyright # 2004 John Wiley & Sons, Ltd. Accepted 10 June 2004 298 A. LEVY ET AL. was faster following injection with the automatic arrival to the Institute’s animal facility. Four EpiPen injector [12]. The time to peak concentra- comparable experiments were performed. For tion was significantly reduced compared with each experiment, one pig was injected with an subcutaneous administration, which in the case autoinjector and the other by conventional needle of anaphylaxis could be extremely important. and syringe, the pigs being matched by weight to The spring-loaded self-injector Atropen, contain- a difference not greater than 100 g. Because pigs ing the equivalent of 2 mg atropine sulfate, was grow very fast, a cross-over design was not studied in 15 healthy volunteers. It was found to possible. produce a significantly earlier increase in heart Autoinjectors, normally used for atropine, rate compared with atropine in a conventional were supplied by Shalon, Israel, each containing syringe injected i.m. [13]. In another study, the 5 mg of midazolam hydrochloride in 0.7 ml performance of an atropine autoinjector (Mark I, of saline plus preservatives. For conventional used in the US Army) was tested in 20 human i.m. injection, the content of some of these volunteers against a different injector referred to autoinjectors was released and measured. They as an MCP [14]. The Mark I injector exhibited a were found to release consistent volumes significantly higher atropine concentration in (0.734 Æ 0.002 ml n ¼ 5). Thus, using a 1 ml plasma 10 min after injection. The decrease in syringe and 23G-100 needle (identical in size to salivation was significantly more pronounced the autoinjector needle), 0.73 ml of this midazo- during the first 15 min and the increase in heart lam solution was injected i.m. into the gluteus rate was statistically significant up to 30 min muscle as a single bolus. The animals were following administration. The more rapid monitored visually, as well as physically by changes were attributed to a broader dispersion lifting their head, to determine muscle relaxation of the drug in the injected tissue. A Swedish and sedation. An ANOVA test was performed on group compared the pharmacokinetics of HI-6 the results for the ‘time to muscle relaxation’, to and atropine following injection with an Astra determine whether the differences between the Tech autoinjector to anaesthetized pigs with that two groups were statistically significant. of conventional intramuscular delivery [15]. The One ml blood samples were withdrawn from time interval for obtaining the therapeutic con- the pig subclavian vein at 1, 2, 4, 7, 10, 15, 20, 30 centration of HI-6 (defined as 4 mg/ml) was and 60 min following injection, as well as after 2, reduced from around 5 min to 1 min. 3, 4 and 5 h. Plasma was separated and kept No data are available for the pharmacokinetics frozen at À708C until assayed. of midazolam administered by autoinjectors. The Midazolam was determined by radio receptor importance of prompt anticonvulsant adminis- assay (RRA), using an adaptation of Citron’s tration after OP poisoning is the subject of the method for scopolamine [16]. Briefly, 200 mlof present study. standard midazolam samples in plasma (in the range 3.9–1000 ng/ml) were prepared and added to 5 ml of 0.05 m phosphate buffer (PB) pH 7.4. The diluted standard solutions, as well as the Materials and Methods diluted study samples, were loaded on activated Sep-Pak C18 columns (Waters, Milford MA), The study was approved by the IIBR Committee washed with 5 ml PB and then with 5 ml of for Animal Experimentation, with emphasis on double distilled water containing 20% v/v the prevention of any unnecessary animal suffer- methanol. The drug was extracted from the ing, the use of the smallest number necessary and columns with 5 ml of 100% methanol, and dried was according to the recommendations of the overnight under vacuum and centrifugation. The Guide for the Care and Use of Laboratory Animals, samples and standards were resuspended in National Academy Press, Washington DC, 1996. 200 ml of PB and 25 ml in duplicates were added White domestic young piglets (supplied by to the competitive binding mixture. [N-Methyl- Lahav, Israel), weighing around 12 kg, were used 3H]-Flunitrazepam (TRK-590, Amersham UK), throughout the study, 2–3 days following their having a specific activity of 96 Ci/mmole, served Copyright # 2004 John Wiley & Sons, Ltd. Biopharm. Drug Dispos. 25: 297–301 (2004) PLASMA CONCENTRATIONS OF MIDAZOLAM 299 as the radioactive ligand in the assay, in which 450 midazolam competed for its binding to rat brain 400 300 membranes, as described previously [16]. The 350 AUC AI=18,530±2215 ng.min/ml 300 ± minimal concentration of midazolam in the 300 AUC IM=18,550 1653 ng.min/ml SEM) standard curve, i.e. 3.9 ng/ml, determined ± 250 Autoinjector, n=4 the limit of sensitivity of the assay as seen in 200 Conventional Injection,n=4 Figure 1. Since standard solution samples were 150 (Averages always processed concomitantly with the study Midazolam (ng/ml) 100 samples in the same assay, and spiking of 50 standards in plasma gave similar results, the 0 recovery of midazolam from standards and 0 30 60 90 120 150 180 210 240 270 300 Time (min) samples was considered identical. To determine the statistical significance of the differences Figure 2. Pharmacokinetic follow-up on midazolam concen- between the two modes of injection, an ANOVA trations in plasma after autoinjector (AI) administration to pigs in comparison with conventional syringe and needle AUC test was employed and the individual injection (i.m.). Four pigs were used in each group (aver- values for each animal were compared between age Æ SEM) the two groups. two curves, using GraphPad Prism software, Results were virtually identical (18 530 Æ 2215 and 18 550 Æ 1653 ng.min/ml). Using the pharmaco- Midazolam concentrations in plasma were de- kinetic relationship Dose.F ¼ AUC.CLp, where termined at 13 time points following injection, in CLp/F represents the total plasma clearance, a four pigs for each experimental group. The clearance value of 16 l/h was obtained for these complete data are presented in Figure 2. Accord- young pigs (or approximately 1.3 l/kg.h). ing to the principles of pharmacokinetics, if Despite having similar AUCs, it was apparent the dose was indeed identical in the two that the two pharmacokinetic curves were not groups, similar areas under the curve (AUC0À300) identical. To emphasize the differences, the should be obtained.