The diagnosis of sleep disorders due to a general medical condition is used in cases in which a prominent disturbance in sleep is sufficiently severe to warrant independent clinical attention and is judged to be related to a coexistent general medical or neurological condition. As with other sleep disorders, the patient must show impairment in social or occupational functioning. This diagnosis is not used in cases of delirium. Four subtypes of general medical condition-related sleep disorders are included in DSM-IV: 1) insomnia, 2) hypersomnia, 3) parasomnia, and 4) mixed type.

A wide variety of medical conditions may cause sleep-related complaints (

Table 89-4). Epidemiological studies show that certain medical disorders, including chronic obstructive pulmonary disease, coronary artery disease, congestive heart failure, peptic ulcer disease, prostate problems, and arthritis, are particularly likely to be associated with insomnia (Katz and McHorney 1998; Klink et al. 1992; Raiha et al. 1994). The total number of physical problems or symptoms is also related to insomnia (Henderson et al. 1995).

The general tenet of treatment for sleep disorders due to a general medical condition is to optimally treat the underlying condition. If stabilization of the medical condition does not alleviate sleep complaints, adjunctive treatment with behavioral or medication approaches can be considered (Mitler et al. 1991).

Behavioral/Psychotherapeutic Treatment

In chronic medical illnesses, sleep disturbances may be the direct sequelae of the underlying illness, or they may be secondary to the drugs or procedures used to treat the medical condition. Conditioning factors may also be involved in perpetuating disturbed sleep independently of these primary causes. For example, patients with chronic pain often engage in maladaptive behavior patterns (e.g., excessive amounts of time spent in the bedroom and/or lying down) in a misguided effort to control pain and sleep. Stimulus control and sleep restriction therapies, two interventions aimed at curtailing sleep-interfering activities and time spent in bed, combined with cognitive therapy, have yielded significant sleep improvements in patients with chronic pain (Currie et al. 2000). Relaxation training has proved useful to alleviate sleep disturbances secondary to cancer (Cannici et al. 1983). Environmental factors, such as noise and excessive light, as well as the need to perform nursing procedures, can greatly contribute to sleep disturbances among hospitalized patients (Berlin 1984). Attention to these factors is essential before psychopharmacological interventions are attempted.

Older adults are particularly at risk for insomnia secondary to medical problems. Although they are often excluded from treatment studies based on the assumption that they would not respond to treatment, one recent study (Lichstein et al. 2000) focused specifically on older adults with insomnia secondary to medical (n = 12; pain, respiratory, prostate, neurological) or psychiatric conditions (n = 11; anxiety or depression). The study found that treated patients improved their sleep efficiency and sleep quality more than untreated control subjects after a brief intervention (four sessions) with relaxation and stimulus control procedures. Treatment response was comparable for patients with insomnia secondary to medical or psychiatric conditions.

Pharmacological Treatments

Pharmacological treatments must be tempered by an understanding of the coexistent medical conditions. Without such knowledge, the clinician runs the risk of actually worsening the medical condition and the sleep complaint with the administration of inappropriate medications. For instance, prescription of benzodiazepines or other sedative-hypnotic medications to patients with chronic obstructive pulmonary disease could potentially worsen hypoxemia during the night. The use of antihistamines or anticholinergic antidepressants could worsen the sleep disturbance of patients with dementia.

The clinician also must understand the changes in drug metabolism that may be caused by a medical condition. For instance, impaired hepatic metabolism can lead to prolonged half-lives of benzodiazepine drugs metabolized by oxidation. In such cases, a benzodiazepine metabolized by conjugation may be used effectively. Dosage adjustments may also be necessary for patients receiving hemodialysis. In general, the elderly have a prolonged elimination half-life for most medications used as hypnotics, largely as a result of increased volume of distribution. Reduced dosages must often be used in the elderly for this reason.

Clearly, each patient and each medical condition must be considered individually. As a general approach, however, pharmacological treatment of sleep disorders due to a general medical condition should follow the guideline of using the lowest effective dose for the shortest amount of time. Again, as a general rule, drugs with the fewest active metabolites and the shortest duration of action should be selected. Short-acting benzodiazepine receptor agonists with no active metabolites are often helpful for hospitalized patients with insomnia and medical illness. Patients with no history of ventricular arrhythmia and low risk for hypotension may also benefit from low-dose trazodone.