Midazolam is widely appreciated for its sedative, anxiolytic, and anticonvulsant properties. It has a rapid onset of action and is commonly used in the management of seizures, procedural sedation, anxiolytic, amnesic, and muscle relaxants. However, despite its effectiveness and general safety profile, midazolam can occasionally produce paradoxical reactions (1). The reactions, though rare, are characterized by effects opposite to those intended, such as agitation, anxiety, aggression, and hyperactivity (2). Recent research indicates that flumazenil is the preferred option for reversing the paradoxical reactions triggered by midazolam (3). Additionally, other studies have demonstrated successful reversal of the reactions using ketamine, physostigmine, and haloperidol (4,5).

Herein, we present a case of an adolescent girl who underwent paradoxical reactions to midazolam, whose condition improved upon administration of ketamine following the failed use of flumazenil. The paradoxical reactions often go unnoticed by physicians due to a lack of awareness and their relatively low incidence. Therefore, early recognition and prompt intervention are crucial to prevent exacerbation and overlook of diagnoses.

An 11-year-old girl was brought to the emergency department after undergoing a seizure shortly after going to bed, approximately 1 hour after she had returned from a camping trip. Upon arrival to the emergency department, vital signs were as follows: blood pressure, 115/75 mmHg; heart rate, 120 beats/minute; respiratory rate, 24 breaths/minute; temperature, 37.4 °C; and oxygen saturation, 92% on room air. She was in postictal confusion and given a single loading dose of intramuscular midazolam 7 mg (0.2 mg/kg) over 5 minutes to prevent the recurrence of seizure. Approximately 10 minutes after the administration of midazolam, she showed increased agitation, confusion, uncontrollable crying, and involuntary repetitive movements. There was no history of recent infections or head injury, arthropod bites, psychiatric disorders, or use of drug or herbal medication. She had benign childhood epilepsy with centrotemporal spikes since her age of 5 years but had not been on an anti-seizure medication.

Upon examining the level of consciousness after the midazolam administration, the girl was found agitated, confused, and disoriented with a Glasgow Coma Scale score of 14 (eye-opening, 4; verbal response, 4; and motor response, 6). A complete neurological examination showed normal findings of the cranial nerves, muscle strength, or deep tendon reflexes. No focal or lateralizing neurological deficits, meningeal irritation signs, and Babinski reflex were observed.

Due to the persistence of her uncontrollable crying and violent behavior, such as fighting, shouting, and biting, the girl received intravenous diazepam 7 mg (0.2 mg/kg), which initially alleviated the behavioral symptoms. However, once the symptoms recurred approximately 1 hour later, a possibility of paradoxical reactions of the initially-administered midazolam was strongly suspected. Other serious conditions, such as central nervous system (CNS) infection, arthropod bites, poisoning, and acute hydrocephalus, were also taken into consideration. To ease the presumed paradoxical reactions, 0.2-mg flumazenil was administered over 30 seconds. Subsequently, same-dose flumazenil was added 3 times at 1-minute intervals, but her symptoms showed no discernible improvement. Following the fourth dose of flumazenil, she developed chest pain and palpitation. Consequently, she was promptly connected to monitoring equipment, and then transferred to the pediatric intensive care unit.

After receiving ketamine for procedural sedation and analgesia prior to conducting a brain computed tomography and lumbar puncture, her condition improved, leading to the complete resolution of symptoms. No acute cerebral pathology was found in the computed tomography and cerebrospinal fluid profile. Initial laboratory tests were unremarkable, including complete blood count, electrolytes, liver and renal function tests, urine toxicology screen, and urinalysis. She was thought to have paradoxical reactions based on the initial use of midazolam, and classical symptoms and signs suggesting the reactions. Twenty-four hours later, she was discharged in good condition and advised to avoid midazolam in case of developing seizures.

Although the paradoxical reactions to midazolam have been extensively documented, making the diagnosis can be challenging due to the rarity of the condition and limited awareness among pediatricians or emergency physicians.

Paradoxical reactions to midazolam are idiosyncratic occurrences that may affect 3.4% of children, and occur in any patient, with reported incidences even in newborns (5,6). Factors contributing to the reactions include younger age, increased midazolam dosage, history of psychiatric disorders, and genetic predisposition (2,7). These manifestations may differ across various reports but typically occur within 5 minutes of intravenous administration of midazolam. Most reactions are preceded by a brief period of apparent and transient sedation, after which the child abruptly becomes alert and exhibits intense agitation, often accompanied by crying (4). Following the administration of midazolam to prevent the recurrence of seizures, the case patient displayed an acute alteration in mental status, characterized by agitation, heightened inattention and disorientation, uncontrollable crying, and excessive random movements.

While the exact mechanism behind the paradoxical reactions remains unclear, various theories have been proposed. First, density or genetic diversity of affinity to the gamma-aminobutyric acid receptors can impact the pharmacodynamics of benzodiazepines. Research has revealed that even a single mutation in the gamma-aminobutyric acid receptor subunits can lead to varied responses of the receptors to diazepam. Individuals with such a mutation might exhibit heightened sensitivity to adverse effects of benzodiazepines, necessitating a lower dosage than typical, or they may experience divergent reactions to the drugs (8,9). Second, the inhibitory effects of benzodiazepines might result in diminished cortical control in certain individuals, potentially resulting in heightened excitement (5). Third, midazolam undergoes metabolism facilitated by cytochrome P450 3A4. The activity of these enzymes can be influenced by different medical conditions and drug interactions, potentially leading to changes in the metabolism of the drugs and consequent prolonged effects (10). Finally, there have been suggestions that benzodiazepines may be associated with central cholinergic activity, potentially explaining symptoms like dystonia and tremors observed in paradoxical reactions. This hypothesis gains credibility from the historical use of physostigmine as an antidote for the paradoxical reactions (11).

Presence of paradoxical reactions is determined clinically, necessitating the administration of midazolam thorough medical history and physical examination, as well as scrutiny over the other culprit medications. No specific biochemical metabolite is associated with the reactions. Differential diagnosis encompasses increased intracranial pressure, insect bites, CNS infections, and poisoning (12). The case patient’s condition appeared to align with paradoxical reactions, likely due to the use of midazolam, the rapid onset of the unexpected response, and the absence of typical seizure. Nonetheless, considering her history of camping, we also considered the possibility of a scorpion bite or CNS infection.

Although there is scarce literature specifically discussing the spontaneous resolution of paradoxical reactions to the authors’ knowledge, the prevailing view is that the reactions typically necessitate intervention rather than waiting for spontaneous resolution. Adding midazolam is proven ineffective and to exacerbate the situation or cause delayed awakening (2). Recent research indicates that flumazenil is a preferred option for reversing the paradoxical reactions (3). Other studies have demonstrated successful reversal of the reactions using physostigmine, ketamine, and haloperidol (4,5). Flumazenil is a competitive benzodiazepine antagonist with an onset of within 1-3 minutes and a peak effect in 6 minutes, which is widely used for reversing sedation, respiratory depression, hypnosis, and paradoxical reactions (3). In pediatric patients, it is recommended to administer flumazenil as a series of small injections with an initial intravenous dose of 0.01 mg/kg (over 15 seconds; maximum, 0.2 mg). If no improvement is observed after 45 seconds, additional doses of 0.01 mg/kg (maximum, 0.2 mg) can be given, with further doses repeated every 60 seconds as needed. This can be done up to a maximum of 4 additional doses, not exceeding a total dose of 0.05 mg/kg or 1 mg, whichever is lower (3,5). Flumazenil should be used with caution in patients with a history of dysrhythmia or seizures controlled with benzodiazepines, as it can trigger convulsions in those dependent on benzodiazepines, as well as cause dysrhythmia and chest pain (3).

Ketamine acts as a noncompetitive antagonist at the N-methyl-D-aspartate receptors, and also interacts with numerous other receptors. It has a short duration of action and is commonly employed as both anesthetic and analgesic agents (13). Furthermore, the administration of low-dose ketamine (0.5 mg/kg) has been identified as an effective treatment for reversing paradoxical reactions to midazolam (4,5). In addition to flumazenil’s failure to reverse the reactions in the case patient, she experienced chest pain, which spontaneously resolved. Additionally, the reactions ceased following the use of ketamine.

To conclude, paradoxical reactions to midazolam, though rare, present clinical challenges due to their unpredictability and potential severity. This report highlights the need for prompt recognition and management of the adverse effects. It suggests that standard reversal agents like flumazenil may not always be effective, while alternatives like ketamine may work.