Accidental induced seizures in three cynomologus macaques following administration of ceftriaxone dissolved in 1% lidocaine diluent

Abstract

Following initial signs of dural infection, or as prophylactic postoperative therapy, three cynomologus macaques were given an intramuscular or a subcutaneous injection of ceftriaxone dissolved in a 1% lidocaine diluent (CL). Two to 15 min later, all three monkeys experienced a long-lasting generalized apnoeic clonic convulsive seizure. The injected doses of lidocaine (0.7-2 mg/kg) were as low as 7-20-fold less than the experimental intravenous dosage (14.2 ± 3.2 mg/kg) previously reported to induce seizure in healthy rhesus monkeys. Under different clinical conditions, the same three animals were either once (1 animal) or repeatedly given CL without any resultant neurological alterations. The monkeys had a cranial device implanted and two of them were craniotomized. This first report of accidental lidocaine-induced seizure in laboratory non-human primates following CL injection strongly suggests increased susceptibility when lidocaine administration is associated with central nervous system alteration. A novel hypothesis, the possible role of cytokine in lowering the lidocaine seizure threshold, is suggested.

Although intramuscular injections of cephalosporins can cause significant pain at the injected sites, ¹ in routine handling of laboratory non-human primates this is the only practicable administration route. In order to reduce the ensuing pain, Travel ² suggested producing a local conduction block in the sensory nerve endings by using anaesthetic diluents. This practice became widely accepted after Patel et al. ³ showed that coadministration of 1% lidocaine as a diluent did not alter the pharmacokinetics of intramuscular administered cephalosporin. Lidocaine, however, may have marked adverse effects on central nervous system (CNS) functions ⁴ and in humans and animals has largely been associated with acute CNS intoxications^4,5 including grand mal convulsion, unconsciousness and apnoeic seizures. ⁶ The convulsant effect of lidocaine is thought to be mediated through binding on the GABA receptor-ionophore complex.^7,8 The drug exerts its neurotoxicity in a dosedependent manner and the seizure dosage and plasma levels have been well defined in humans and monkeys.^9–12 Three cynomologus macaques were all part of an ongoing brain research that involved electrophysiological recording and blood vessel imaging. The study requested and received all forms of approval required by law and was carried out under the strict observation of the Institutional Animal Care and Use Committee (IACUC). As a part of this project, each animal was implanted with a stainless-steel receptacle and recording cylinder device on its skull. ¹³ Some of these were craniotomized several weeks or months later. The meninges are removed and substituted with artificial dura made up of transparent silicon as described by Arieli et al. ¹⁴ The chamber is opened between one to several times a week for cleaning, which is done under rigorous sterile conditions and the silicone sheets are sampled for bacteriological and mycological examinations. The most frequent infections are those caused by the Gram-positive opportunistic pathogen Staphylococcus aureus and the Gram-negative bacterium Escherichia coli (personal communication). Yeast infections are relatively common. Thus, frequent inspections are of crucial importance since these contaminants are highly hazardous in these exposed anatomical sites. Many different insults to the exposed operated field, including bacterial growth, encourage scar tissue formation that might exclude the animal from meeting the criteria for research.^15–17 In order to successfully cope with this risk, frequent and extensive use of a broad-spectrum bactericidal drug, e.g. ceftriaxone, is a justified preventive and curative strategy.

The first animal (male, 8 years old, 12 kg body weight) had been given ceftriaxone dissolved in 1% lidocaine diluent (CL) on several different occasions including during the pre- and postoperative periods. No adverse effects had ever been observed. The animal was implanted with a recording chamber and head holder (attached to the skull using a head-cap made up of dental acrylic ¹³ ) aimed at keeping the head stationary because sudden rotations cause the brain to move, destabilizing the recording. ¹⁸ It was treated with CL following acute onset of symptoms related to dural infection. One gram of ceftriaxone (Ceftriaxone, Teva^®, Petach-Tikva, Israel) was dissolved in 3.5 mL CL, according to the manufacturer's instructions. A volume of 1.4 mL of the final solution was divided into two syringes (0.7 mL in each syringe) connected to 30 gauge needles. Upon needle insertion and following negative blood aspiration, 0.7 mL of CL was pushed over 5-7 s into each gluteal mass at a one-minute interval. Two minutes later, the animal started to show gradual signs of focal seizure that rapidly evolved into a long-lasting generalized apnoeic clonic convulsant seizure. More specifically, the temporal progression of the symptoms took place in the following sequential order: very rapid monolateral eye blinks spreading to unilateral muscle contraction of the face associated with quasi-vertical fixation of the eye globes oriented at the 11 o'clock position; rapid involuntary jaw opening accompanied by tongue biting to bleeding; abolition of the pupillary reflex associated with semi-miosis without convulsions that were not yet present and no clonic contractions were observed; loss of righting reflex followed by a rapid decrease in responsiveness, which progressed to a total loss of receptiveness to sensory and mechanical stimuli; convulsions commencing at the head and descending to the upper and then the lower limbs; onset and establishment of a generalized aponeic myoclonic convulsant seizure with intermittently arrested laborious and dyspnoic breathing; the oral mucosa was clearly cyanotic. The limbs jerked but the animal could be restrained. The overall time from the first signs of seizure to the last stage was about 25 s. Blood was sampled and a leukogram (10³ cells/μL) reflected absolute neutrophilia with left shift deviation (absolute neutrophil count, 27.7; absolute band count, 1.39). The second monkey (male, 8 years old, 9 kg body weight) had been craniotomized several months earlier (as described elsewhere ¹³ ) and similar to the first animal had been given CL on different occasions without any ensuing adversity. Following overt purulent inflammation of the cap surrounding tissues it was injected with CL. In light of the previous experience, the dosage of lidocaine was reduced to 7 mg/kg. The CL was injected to one site only, in the same manner as described above. Three minutes later the first signs of seizure appeared, and fully resembled the one described above. The third animal (male, 3 years old, 4 kg body weight) was given CL (intramuscularly) before undergoing craniotomy as a preoperative prophylactic therapy. Anaesthesia was induced by ketamine and etmoidate and an endotracheal tube was inserted. Anaesthesia was maintained with a mixture of nitrous oxygen and isoflurane (1-1.5%) throughout surgery (6 h) during which no signs of intraoperative seizure were observed. Upon procedure completion, it was given CL (2 mg/kg lidocaine) injected subcutaneously and disconnected from the anaesthesia machine. Fifteen minutes after recovery (25 min after the CL injection), the animal experienced a seizure resembling the ones reported above.

The temporal and circumstantial relationships between CL administration, seizures and CNS alteration are evident. In humans, accidental lidocaine-induced seizure (ALIS) is a well-documented iatrogenic pathology largely attributed to inadvertent intravenous administration. Munson et al. ¹⁰ reported that the experimental intravenous dosages of lidocaine and drug plasma level that caused seizures in healthy rhesus monkeys were 14.2 ± 3.2 mg/kg and 24.5 ± 4.5 μg/mL, respectively. Similar values were found by others.^11,19 Although lidocaine serological levels were not assayed here, several features point to the possible involvement of an additional pathogenic factor; e.g. inflammation mediators acting either in a cumulative or in a synergetic fashion with the lidocaine through a common pathway, the GABA inhibitory system. This is a reasonable hypothesis since the injected dosages of lidocaine fell well below the values found by Munson and others (as low as 7-20-fold less), the first was injected in two different sites that strongly reduced the likelihood of inadvertent intravenous administration of the entire dose, and lastly all the animals had altered CNS conditions, which strongly implies a link to ALIS. Furthermore under different clinical conditions (not related to overt infections), two of the animals had repeatedly been given CL, using the same dosage and in the same way as described previously, without any ensuing adversity. In CNS pathological states such as infections, ischaemic, traumatic and/or excitotoxic damage, the intracerebral immune response is mediated by the cytokines. Tumour necrosis factor-α, interleukin (IL)-1, IL-6 and IL-8 have been reported to play a role in this response^20–23 and enhance seizure activity. Intracerebral application of the IL-1 receptor antagonist, IL-1ra, that antagonizes the effects of IL-1β was found to have potent anticonvulsant properties. ²⁴ Transgenic mice overexpressing IL-1ra showed markedly reduced susceptibility to seizure. ²⁵ Of relevance is that IL-1β was found to affect seizure activity by increasing glutamatergeic neurotransmission. ²⁶ Seizure susceptibility was found to be significantly increased in transgenic mice overexpressing IL-6 in glia. Furthermore, these transgenic mice had increased susceptibility to glutamergic agonists. ²⁷ IL-1β was found to enhance focal electrographic seizures induced via focal kainate application receptors through an increase in glutamatergeic neurotransmission. ²⁶ A number of studies employing different experimental conditions have shown that the lidocaine seizure threshold is not invariant.^7,8,28–31 In particular, Post et al. ²⁸ noted progressive susceptibility to lidocaine-induced convulsion as a result of chronic administration. There are very few other circumstances in which animals might be subjected to chronic or frequent administration of CL, which is the case for craniotomized monkeys in prolonged brain research projects. Thus, the lidocaine seizure threshold might be lowered over time in these animals as a function of frequency of CL administration. Whether the lidocaine seizure threshold can be modified or modulated by states of dural infections or any other CNS insult has never been investigated and no case of ALIS in monkeys – whether craniotomized or intact – has so far been reported.

In summary, the cases reported here strongly suggest that cynomologus macaques might undergo ALIS following CL injection. Susceptibility to lidocaine-induced seizure in cynomologus macaques is not invariable and might be significantly enhanced under certain circumstances such as repetitive administration of lidocaine and/or CNS alterations including infections. Seizure dosage, defined as the dose of the drug calculated in mg/kg of body weight, required to evoke electrical seizure activity ¹⁰ of lidocaine in highly susceptible monkeys might be as low as 20 times below the figure reported to induce experimental seizure in healthy rhesus monkeys.