The management of insomnia: an update
- John W.G. Tiller
- Aust Prescr 2003;26:78-81
- 1 August 2003
- DOI: 10.18773/austprescr.2003.061
Insomnia is a common symptom but hypnotics should be avoided if possible. Management aims to identify and treat underlying causes, such as psychiatric disorders and medical problems. If symptomatic relief is still required in addition to medical, psychological and social interventions, hypnotics can be considered. Hypnotics should preferably be used intermittently, for less than two to four weeks. The newer non-benzodiazepine hypnotics - zopiclone, zolpidem and zaleplon - are not free of the problems surrounding the use of benzodiazepines.
Insomnia is a common symptom with up to 25% of Australians reporting trouble getting enough sleep.1There are many causes for this perceived inadequacy of sleep, but subjective perceptions do not necessarily mean the patient is not sleeping.
Symptoms associated with insomnia may suggest an underlying medical, surgical, psychological or environmental problem. Treating any underlying problem can help to alleviate the insomnia.
Psychiatric disorders are typically associated with insomnia. Anxiety disorders can cause early insomnia (difficulty in getting to sleep) associated with rumination over particular worries or concerns. With depression, it is typical to have middle insomnia (waking in the early hours of the morning) and late insomnia (waking earlier in the morning than is usual and being unable to get back to sleep). The depressive pattern may also have an associated anxiety disorder so the patient's sleep is disturbed throughout the night. Patients may present with insomnia and only acknowledge their low mood or loss of interest after enquiry.
Middle insomnia is typical with alcohol abuse. The patient goes to sleep in the evening when intoxicated only to wake a few hours later when their blood alcohol concentration drops.
If a patient's bedroom is too hot, too cold, too noisy, or their bed cramped or uncomfortable, addressing those factors may resolve the problem. A crying baby, or a sick or restless child or other family member may disturb sleep. The assistance of a partner, other relative, or brief period of respite may address the sleeplessness.
Many illnesses including cardiac and respiratory failure and pain syndromes may contribute to insomnia.
Flying across several time zones may also result in insomnia. The therapeutic key is to settle into the new time zone as quickly as possible. This is aided by a regular local sleep-wake cycle and particularly by re-setting sleep rhythms with early morning light and exercise. It is possible to adjust approximately one hour per day, a task which is easier when the sleep cycle is extended rather than shortened. This is quicker following east to west travel than west to east. A similar disturbance may occur when shift workers start and end work cycles. Occasionally, the brief use of a hypnotic may help adaptation to a new sleep pattern. Taking a hypnotic during flight should generally be avoided as immobility may predispose to deep vein thrombosis.
In addition to routine clinical evaluation, it is worth asking in detail about the patient's sleep pattern (seeBox 1). Asking the patient to complete a sleep log over a few days is also useful. The log should be completed as each day progresses, as retrospective entries tend to minimise sleep and maximise disturbances. Patients may enter factors you had not considered, but which may be relevant to the sleep disturbance (Fig. 1).
The prescription of hypnotics should only follow a careful evaluation and consideration of other approaches including psychological interventions such as cognitive behaviour therapy. In general hypnotics should only be prescribed if the duration of use is likely to be less than four weeks, and preferably less than one or two weeks.
Key elements when prescribing are to manage patient expectations of the duration of treatment and likely outcomes, and to have an 'exit' strategy. Explain the likely duration of therapy, when medicines should be used and when they should not be used, common adverse events, and the risks of tolerance, dependency, withdrawal and discontinuation syndromes if use is prolonged.
The exit strategy is a clear plan of change for the patient so that they should not need continued drug treatment. For example, you might expect an antidepressant to have started to work in two to four weeks so that depression-related insomnia should have resolved by that time. Most patients do not need a hypnotic for depression-related insomnia. A few value initial help with sleep, but hypnotics should not be continued once the depression is relieved. Some personal and social crises can result in the patient becoming so distressed and dysfunctional with insomnia that a few nights assisted sleep helps them reintegrate. They could then be expected to cope with the stresses in their life without the need for ongoing drug treatment. Bereavement would not normally necessitate hypnotics, although they can sometimes be briefly helpful when the bereaved patient is not coping with insomnia.
The hypnotics predominantly used in Australia are benzodiazepines, or non-benzodiazepines acting through benzodiazepine receptors. Other classes of drugs are also used, but are potentially more toxic and would rarely seem to offer any advantage over a benzodiazepine or related drug.
These drugs all have similar actions including sedative-hypnotic, anxiolytic, anticonvulsant, muscle relaxing, and amnesic effects. Although some of the drugs are marketed for different indications, their major differences in practice are brought about by differences in pharmacokinetics.
Half-life (Table 1)
Drugs with longer half-lives may cause appreciable impairment in the morning (on waking). A single dose of temazepam or oxazepam can have actions well into the next day, and nitrazepam and flunitrazepam even more so. There has been a recent campaign to use temazepam tablets rather than capsules (because of the risk of people injecting the contents), however the onset of action and time to maximum effect of temazepam tablets can be slower than one would wish in a hypnotic.
Adverse effects of benzodiazepines
Adverse effects can be anticipated from the normal actions of hypnotics. Excessive or daytime sedation may occur, particularly with drugs that have a longer half-life. The sedative and muscle relaxing activity may combine to increase the risk of ataxia or falls, particularly in the elderly.
The anxiolytic action can be helpful in relieving distress when settling to sleep. However, this can be disadvantageous if it inhibits alertness and responsiveness the following morning.
The anticonvulsant action can result in withdrawal fits if the benzodiazepine is withdrawn abruptly. This risk may increase if a benzodiazepine or related drug is substituted by a sedative antipsychotic or tricyclic antidepressant which is pro-convulsant.
Amnesic effects can result in patients forgetting events soon after taking a dose. They may take extra doses if they forget they have already taken their medication. Some may 'forget' previous cautions about concurrent use of alcohol and anterograde amnesia has been associated with such combined use. Disinhibited behaviour may follow ingestion and hallucinations have been reported, especially at higher doses. Hypnotics, particularly those with a long half-life, can cause cognitive problems the following day.
Zaleplon, zopiclone and zolpidem, although structurally not benzodiazepines, act on the same receptor. While there is a suggestion from animal studies that the new drugs have a more specific hypnotic action, this has not yet been shown in humans. The new drugs are not free of the adverse effects of benzodiazepines and are not necessarily safer medicines. On the contrary, one study suggested that elderly patients taking zolpidem had almost double the risk of hip fracture compared with no medication (adjusted odds ratio 1.95, CI*1.09-3.51). This risk is greater than that seen with benzodiazepines (1.46, CI 1.21-1.76), antipsychotic medications (1.61, CI 1.29-2.01) and antidepressants (1.46, CI 1.22-1.75).5
Zopiclone has a bitter taste as its commonest adverse effect. Zolpidem causes hallucinations in a small proportion of patients, and it should be stopped if these occur. Zaleplon has a short half-life making it useful for sleep-onset insomnia. It may be used during the night if a patient cannot fall asleep as it has less risk of morning sedation. Interestingly, within-night discontinuation effects have not been reported, though one might otherwise expect them given its rapid onset and offset.
Several other types of drugs are sometimes used as hypnotics, but in general their use is limited by toxicity. These drugs are primarily used for other indications, but as drowsiness is one of their adverse effects they are sometimes prescribed for insomnia.
Sedative antihistamines have prominent anticholinergic effects which can result in confusion, especially in children or the elderly, and should have little or no place in the management of insomnia. They may be effective hypnotics in the short term, but many patients rapidly develop tolerance.
Antidepressants should not be used for insomnia unless the patient also has depression. Tricyclic antidepressants have sedative antihistaminic effects, even at low doses which are sub-therapeutic for depression. They have significant anticholinergic effects leading to confusion, alpha adrenergic blocking effects that can result in marked postural hypotension, and quinidine-like effects with the potential for atrioventricular block and prolongation of the QTcinterval. Tolerance develops to the sedative effects of tricyclic antidepressants and their potential toxicity generally outweighs their benefit as hypnotics, especially in overdose.
Antipsychotic drugs have been little studied as hypnotics. The toxicity of typical antipsychotics limits their use as hypnotics, while atypical antipsychotics are not suitable and they are only approved for use in schizophrenia.
By establishing a clear expectation of short-term use when starting treatment with a hypnotic you are more likely to avoid your patient falling into the trap of long-term dependency. If the insomnia persists, it is particularly difficult to stop treatment as the patient fears that stopping the hypnotic will make their insomnia worse.
If it is difficult to stop treatment it is worth reviewing the patient's history, and the possibility of underlying disorders, or dependency on the hypnotic. Reinforcement of sleep hygiene techniques, a gradual process of dose reduction, and intermittent use, with the availability of the hypnotic as 'rescue medication' when needed, may assist the patient in reducing their hypnotic use. Long-term use is particularly an issue for patients who were prescribed hypnotics before problems with this use were recognised. Even if one attempt at discontinuing is unsuccessful, review this regularly as an approach at a different time may have more success. Furthermore, if the patient is actively involved in the process, they can choose a night when they would feel more comfortable about reducing the dose and avoid challenging themselves on a night when they fear sleep will elude them.
There is the occasional patient who does not increase their hypnotic dose, or frequency of use, and remains well on a stable dose, but becomes profoundly dysfunctional if it is ceased. When reviewing such a patient consider dose reduction, or stopping the medicine, as well as the possibility of other illnesses or problems perpetuating the patient's seeming need for a hypnotic.
Patients should not be given hypnotics when other interventions would be more appropriate. Always address underlying disorders and attend to the patient's sleep hygiene before considering prescribing. Before any hypnotic is prescribed, it is important for the patient to have a clear understanding of the expected outcome and that continued use will be unnecessary. With patients on long-term treatment the aim is to cease hypnotics, not change to a newer drug.
National Prescribing Service patient education material 'Getting a good night's sleep'. (Go to Topics, Benzodiazepines, Patient education material'A reduction plan for sleeping tablets and sedatives'page 2)
Professor Tiller has conducted sponsored studies or been a consultant to manufacturers of antidepressants, benzodiazepines, antipsychotics and non-benzodiazepine hypnotics.
The following statements are either true or false.
1. Nitrazepam has a fast onset of action because of its short half-life.
2. Zolpidem may cause more falls in the elderly than benzodiazepines do.
Answers to self-help questions1. False
Professor of Psychiatry, The University of Melbourne, Albert Road Clinic, Melbourne