Benzodiazepines: sedation and agitation

History

The first benzodiazepine was synthesized by Leo Sternbach, who discovered chlordiazepoxide in 1956 while working for the pharmaceutical company Roche. Chlordiazepoxide was first licensed for use in 1960 and, with its successor diazepam, these rapidly became the most widely prescribed drugs in the world.³ As a group, the benzodiazepines induce sedation, hypnosis, anxiolysis, amnesia and muscle relaxation and they are also anti-convulsant. They are used clinically as pre-medication agents and as sedatives and also in the treatment of disorders such as insomnia, agitation, anxiety, seizures and muscle spasm. By the late 1970s, prescribers began to understand that benzodiazepines could induce dependence. This property is now well-recognized and legal proceedings have been initiated in several countries by groups of patients who allege long-term physical, psychological and cognitive damage consequent on their use of benzodiazepines. Today, benzodiazepines continue to be drugs of choice for premedication and conscious sedation. However, when prescribed for conditions such as insomnia, they are now generally recommended for short-term (2–4 weeks) use only.⁴

Mechanism of action

Benzodiazepines work by altering the effect of gamma-aminobutyric acid (GABA) in the central nervous system. GABA is a major inhibitory neurotransmitter in the adult mammal and it controls the level of excitability of neurons in the brain and spinal cord. GABA is released from vesicles at the pre-synaptic terminal and travels across the synaptic cleft to bind with and activate GABA-specific receptors, of which there are two main types – GABA_A and GABA_B. The most relevant one to this discussion is the GABA_A receptor, which is a complex of five protein sub-units arranged around a central pore.⁵ The receptor complex extends across the cell membrane and when the receptor is activated the central pore becomes permeable to chloride, which then floods into the neuron down a concentration gradient. This influx of chloride causes the neuron to become hyperpolarized, making it less excitable and inhibiting neurotransmission. In the developing brain, GABA produces a different effect. There is a higher intra-cellular chloride concentration in immature neurons and, when the central pore becomes permeable to chloride, the flow of chloride is outward, depolarizing the neuron and making it more excitable, rather than less.⁶ This may explain, at least in part, why children frequently respond differently to benzodiazepines.

Receptors for GABA are widely distributed in the central nervous system, being found in the cortex, thalamus, limbic system, mono-aminergic neurons and motor neurons.⁷ Drugs such as barbiturates can bind at these receptors, activate them directly and cause the chloride channel to remain open for longer. The mechanism of action of benzodiazepines is different. They bind at a location on the receptor known as the benzodiazepine binding site, where their effect is to reduce the quantity of endogenous GABA required to activate the receptor. This difference in how the two groups of drugs affect the GABA receptor means that benzodiazepines have a much wider margin of safety and a reduced potential for fatal overdose when compared to barbiturates.

People have differing responses to benzodiazepines and one way of explaining this may be the difference in how their GABA_A receptors are configured. In humans, each GABA_A receptor contains five sub-units. Many different sub-units have been identified and grouped into families; there are six α-, three β- and three γ- sub-units, as well as δ, ε, θ, ρ and π sub-units. A receptor must contain both α- and β- sub-units in order to be sensitive to GABA. In order to be sensitive to benzodiazepines, the receptor must also contain a γ- sub-unit.⁸ The most common configuration of the five-unit GABA_A receptor in humans is a two α1, two β2, one γ2 receptor.⁹ In this type of receptor, the benzodiazepine binds at the α1–γ2 interface. The different types of α- sub-unit are particularly critical for benzodiazepine binding. If the α- sub-unit is an α4 or α6 type, then that receptor will not be sensitive to benzodiazepines. A separate gene encodes for each sub-unit and the advent of technology to delete or alter individual genes has made it possible to study the specific role of each sub-unit. This ongoing research should improve our understanding of the variation in how individuals respond to benzodiazepines.

Benzodiazepines in sedation

Midazolam

Midazolam is a short-acting, high-potency benzodiazepine and was first synthesized in 1975. Its efficacy, rapid onset, short duration of action and good side-effect profile have made it the current drug of choice for sedation in the dental setting (Figure 2). Midazolam differs from many other benzodiazepines in being water-soluble and this makes its intravenous formulation less irritant. For the purpose of procedural conscious sedation in the UK, the intravenous route is licensed in adults, with additional licences for the rectal and intramuscular routes in children. The buccal route (Buccolam®) is licensed for control of seizures in children. However, off-label use is widespread and dosages for oral, rectal and buccal routes can be found under the heading ‘conscious sedation for procedures’ in the current BNF.¹⁰ There is also a considerable body of experience with intra-nasal administration of midazolam. Where drugs are used for an unlicensed indication or given via an unlicensed route, this information should be included in the consent dialogue with the patient.

Flumazenil

The discovery and subsequent licensing of a benzodiazepine antagonist, flumazenil, in 1987 made it possible to reverse unwanted effects of benzodiazepines quickly (Figure 3). This improved the safety profile of conscious sedation, and thus its acceptability in settings outside the hospital environment.¹¹ Intravenous sedation using midazolam is increasingly available to dental patients in the community setting and has reduced reliance on general anaesthetic for the provision of dental treatment. Flumazenil is currently available for intravenous use only, as it is subject to significant first-pass metabolism when taken orally. Flumazenil is an imidazobenzodiazepine which functions as a competitive antagonist at the benzodiazepine binding site on the GABA_A receptor. Its half-life of 40–80 minutes is shorter than that of other benzodiazepines, including midazolam: this means that the effects of the drug it is used to antagonize may reappear, and further doses may be required.

Paradoxical reactions

In the vast majority of patients benzodiazepines produce anxiolysis and sedation and success rates for dental sedation are high. A case series of intravenous midazolam sedation in 562 patients under 16 years of age, a difficult cohort for benzodiazepine sedation, reported a success rate of almost 99% as defined by completion of the planned dental treatment in sedated patients.¹² Ten of their patients could not be sedated in accordance with the study protocol, owing to various difficulties with cannulation; but, in total, they successfully sedated and treated 97% of the patients referred to them.

However, there are some patients who experience a paradoxical reaction to benzodiazepines and who, instead of anxiolysis and sedation, exhibit a variety of responses including agitation, aggression, hyperactivity, sexual disinhibition, hallucinations, hostility and rage. The reported incidence of these paradoxical reactions varies considerably in the literature. One early retrospective study, of 16,000 patients treated over a 13-year period, reported a 29% incidence of adverse psychological reactions.¹³ These patients were sedated with an average of 20 mg of intravenous diazepam, sometimes in combination with methohexitone; an unlikely regimen in current practice. In more recent studies, the prevalence of paradoxical reactions is felt to be less than 1%.^14,15,16 What has become clear in the decades of experience with benzodiazepines is that there are certain sub-groups at greater than average risk. Children, the elderly, psychiatric patients, those with a history of alcohol abuse or alcohol-related aggression, those with a history of poor impulse control and those with intellectual impairment all have an increased incidence of paradoxical reaction,^15,17,18 as indeed do those who have previously experienced a paradoxical reaction. Careful patient selection, with particular attention being paid to these patient sub-groups, should reduce the incidence of paradoxical reaction.

Mechanism

We do not yet fully understand paradoxical reactions. There are many theories, with one of the most widely accepted being that the anxiolytic and amnesic effects of benzodiazepines lead to a loss of normal social restraint. This may possibly explain why groups of patients, for example children and those with intellectual disability, who have not developed the skills to control their behaviour in adverse circumstances, are at increased risk.¹⁸ Another theory is based on the fact that reduction in serotonin neurotransmission is responsible for some types of aggressive behaviour.¹⁹ As benzodiazepines have a recognized ability to effect a reduction in serotonin neurotransmission, this mechanism may account for some instances of paradoxical reaction. A case report of paradoxical reactions to midazolam in identical twins²⁰ led to the suggestion of a genetic basis for the phenomenon, and there may be genetically determined variability in receptor density or receptor structure which accounts for abnormal responses. The differences in intra-cellular chloride concentration and response to GABA found in immature neurons, which were discussed earlier, may be evidence for a difference in physiology accounting for the difference in response. With further research we may well find that there are different mechanisms underlying subtly different forms of paradoxical reaction.

Management

Whatever the mechanism of the paradoxical reaction, the key to managing it successfully lies in recognizing it. Diagnosis of any emergency is more difficult in the sedated patient and careful pre-operative attention to review of the patient's medical history will help in the prediction, prevention and diagnosis of untoward events. There are many other reasons for agitation in the sedated patient; the most common being pain or the derangement of blood glucose or blood gases. If these other possible causes of agitation are eliminated, then a paradoxical reaction is the most likely cause for agitation in a patient sedated with benzodiazepines.

In the first instance, supportive measures to manage the airway and cardio-pulmonary function must be provided as necessary. As a paradoxical reaction is caused by the administration of a benzodiazepine, it will disappear as the effect of the drug wears off. If the reaction is not severe, it may be possible to allow it to resolve spontaneously. However, in many cases treatment is indicated and various pharmacological measures to control the reaction have been proposed. One of the earliest suggestions was physostigmine,¹⁵ but other suggestions have included haloperidol,²¹ caffeine²² and ketamine.²³ Presently, the weight of opinion appears to favour the use of flumazenil.^24,25,26 Its specificity, safety and relative absence of contra-indications make it an attractive choice. It has even been suggested that small doses (0.1–0.4 mg) of flumazenil can reverse the paradoxical reaction without reversing the amnesic and sedative effects of the original benzodiazepine.²⁷ In such a scenario it might be possible to continue with treatment. When flumazenil is used, care must always be taken to observe the patient for re-emergence of symptoms: the half-life of flumazenil is shorter than that of the benzodiazepines it is used to reverse.

Conclusion

Provision of treatment for the anxious patient is a familiar challenge in dental practice. Avoidance of necessary treatment is all too common in anxious patients and strategies to help such patients access dental care are part of treatment planning at all levels; from the individual dentist in primary care to the creators of public health policy. A significant tool in our armamentarium has been the development of safe and predictable sedation for use in primary care. The aim of such sedation is to ensure a pleasant treatment experience for the anxious patient, but as sedationists it is incumbent upon us to be aware that agitation may occasionally be an unintended consequence. We must also be aware that agitation may have many causes and be familiar with the differential diagnosis in order to manage it appropriately. Furthermore, it is essential that we recognize the increased likelihood of the occurrence of agitation in certain groups of patients and appreciate the implications of this for patient selection.

In the particular case of paradoxical reactions, further research, leading to greater understanding of the underlying mechanisms, should help us to reduce the incidence of such adverse events in future and better inform our management of them when they occur. As with all adverse events, our concern must be analysis of the problem with this aim in mind. If we are successful, we will improve outcomes for patients and dentists alike.