HIV/AIDS
BRIEF REPORT
Analyzing Sleep Abnormalities in HIV-Infected Patients Treated with Efavirenz Lucı´a Gallego,1 Pablo Barreiro,1 Rafael del Rı´o,3 Daniel Gonza´lez de Requena,2 Apolinar Rodrı´guez-Albarin˜o,3 Juan Gonza´lez-Lahoz,1 and Vincent Soriano1 1 Service of Infectious Diseases and 2Service of Pharmacy, Hospital Carlos III, Instituto de Salud Carlos III, and 3Service of Clinical Neurophysiology, Hospital La Paz, Madrid, Spain
Ambulatory electroencephalogram monitoring was performed for 18 HIV-infected subjects treated with efavirenz with and without insomnia and for 13 healthy control subjects. All patients receiving efavirenz had longer sleep latencies and shorter duration of deep sleep, although poor sleepers also showed reduced sleep efficiency and shorter duration of rapid eye movement sleep. Efavirenz plasma levels were higher in patients with insomnia and/or reduced sleep efficiency.
Sleep disorders are common in HIV-infected persons [1]. Although HIV infection per se has been implied in causing sleep abnormalities, these abnormalities are more frequent in subjects receiving efavirenz (Efv). Vivid dreams, difficulties in falling asleep, and/or numerous night awakenings are frequently reported by patients after beginning Efv therapy. Although these symptoms tend to resolve within the first weeks of therapy, they persist in a significant proportion of patients, causing measurable quality of life impairment and occasionally leading to Efv withdrawal [2]. It is not well known whether Efv-related insomnia is characterized only by a reduction in time sleeping, or whether moreprofound abnormalities in sleep architecture are produced. To answer this question, we studied electroencephalographic (EEG) activity in HIV-infected patients receiving Efv therapy. The presence of insomnia and the potential relationship beReceived 23 June 2003; accepted 14 September 2003; electronically published 9 January 2004. Financial support: This study was supported in part by grants from the Asociacio´n Investigacio´n y Educacio´n en SIDA (AIES) and Red de Investigacio´n en SIDA project 173 (RIS). Reprints or correspondence: Dr. Vincent Soriano, Calle Nueva Zelanda 54, 4 B, 28035 Madrid, Spain (
[email protected]). Clinical Infectious Diseases 2004; 38:430–2 2004 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2004/3803-0020$15.00
430 • CID 2004:38 (1 February) • HIV/AIDS
tween insomnia and Efv treatment were established through psychiatric interviews. Since preliminary data have suggested that sleep alterations might be more frequent in HIV-infected patients with higher plasma Efv levels [3], we decided, in our study, to examine the clinical diagnosis of insomnia and the EEG findings with respect to plasma Efv levels. Patients and methods. A total of 26 HIV-infected patients receiving a triple antiretroviral combination including Efv (600 mg once a day at bedtime) for 112 weeks were selected in a prospective and nonrandomized fashion. A semistructured psychiatric interview was performed to rule out any history of mental disorders—including sleep alterations—and of consumption of licit or illicit psychoactive substances. A total of 8 patients were excluded after this clinical evaluation; 1 had a history of severe psychiatric morbidity, 3 had chronic sleep abnormalities, 2 were long-term users of hypnotics, and 2 admitted current substance abuse. The presence of insomnia for the 18 patients enrolled was established through sleep diaries and the Pittsburgh Sleep Quality Index (PSQI). Each day for 1 week, each patient recorded, in sleep diaries, variables such as bedtime, total sleep time, time until sleep onset, number of awakenings, use of sleep medications, morning wake time, and a rating of subjective quality of sleep and daytime symptoms of sleep deprivation (fatigue, irritability, and impaired concentration). The PSQI [4] is a composite score generated from 19 items assessing several sleep-associated characteristics (quality, latency, duration, disturbances, use of hypnotics, nighttime restoration, and daytime symptoms of sleep deprivation). Patients having a final score of ⭐5 were considered poor sleepers. Both questionnaires were administered and interpreted by an experienced psychiatrist. On the basis of the results of the psychometric tools, Efvrelated insomnia was diagnosed in 13 of 18 patients, whereas the remaining 5 subjects had no significant sleep alterations. All 18 HIV-positive patients receiving Efv and 13 healthy HIV-negative control subjects underwent ambulatory EEG monitoring. A system with 8 channels of continuous EEG, electromyogram, and electrooculogram monitoring, and a pushbutton marker of bedtime and nighttime arousals were used for 24 h. The data registered by this system were interpreted according to current international guidelines [5] by a single experienced neurophysiologist who was blinded to the results of the sleep psychometric tests. Finally, blood samples were obtained from all HIV-infected patients to determine plasma Efv concentrations using high-performance liquid chromatog-
raphy. The blood samples were obtained in the morning before the patients had eaten breakfast, ∼12 h after the last Efv dose was taken. The EEG recording was performed thereafter. Drug adherence for all HIV-positive patients was assessed using the validated Simplified Medication Adherence Questionnaire [6]. Results. Sex distribution (84% male) and mean age SD (42 7 years) were comparable between HIV-infected patients and healthy volunteers. One-half of HIV-positive individuals were homosexual men, 5 (28%) were heterosexuals, and 3 (17%) were former injection drug users. At the time of sleep evaluation, the mean CD4 count (SD) was 420.3 172.4 cells/mm3, the median virus load was 90 HIV-RNA copies/mL (range, !50–266,255 copies/mL), and the mean time of exposure to Efv (SD) was 6 3 months in the HIV-positive population. The sleep patterns observed in the 18 HIV-positive patients
receiving Efv and in the 13 healthy control subjects are summarized in table 1. Patients with Efv-related insomnia showed low values of sleep efficiency (mean, 81%), whereas this parameter—which reflects the proportion of sleep during the night—remained within normal limits (i.e., 190%) both in HIV-positive patients without insomnia and in healthy control subjects (P ! .05). The total time awake was significantly longer in poor sleepers receiving Efv, mainly due to the increase in the number of night arousals (mean number per night, 4.1 vs. 2.6). A significant increase in sleep latency, which implies difficulties in falling asleep, was noticed among HIV-positive patients receiving Efv, compared with sleep latency in healthy volunteers. Compensatory napping, manifest by EEG sleep activity registered during the day, was more frequent in subjects with Efv-related insomnia, probably in response to the loss of nocturnal sleep restoration.
Table 1. Main sleep parameters for HIV-infected patients receiving efavirenz (Efv) who underwent ambulatory electroencephalographic (EEG) monitoring. Mean value SD HIV-positive patients receiving Efv Variable
With insomnia (n p 13)
Without insomnia (n p 5)
Healthy control subjects (n p 13)
429.7 95.8
Sleep time, min 438.6 51.7
436.4 23.4
Napping
32.5 38.1*
12.1 9†
10.3 6†
Night sleep
397 82.6
426.5 48.5
426.1 18.3
Total in 24 h
No. of complete night sleep cycles
3.62 1.4
3.6 0.5
Sleep efficiency, %a
80.7 14.6*
96 1.6†
94.5 2.3†
4.2 0.8
Sleep latency, minb
30.9 18.5*
34 49.3*
10.9 3.0†
Total time awake, % of sleep time
19.6 14.6*
3.7 0.2†
5.0 0.9†
Stage 1
13.1 8.8*
6.8 3
4.0 1.4†
Stage 2
43.6 6.5*
56.2 7.3
Stage 3
5.6 5.1*
Stage 4
3.5 4.4*
4.3 5*
14.2 6.3*
22.6 2.8
Stage 1
3.3 4.6
2.2 1.4
NA
Stage 2
3.9 1.9
2.9 2.5
NA
Stage 3
0.9 1.2
2.7 2.9
NA
Stage 4
0.7 1.2
3 6.1
NA
NREM sleep stage, % of sleep time
REM sleep, % of sleep time
50.1 4.1
†
6 2.4*
8.9 1.9† 12.7 1.8†
†
22.6 2.1†
No. of arousals By sleep stage
REM sleep Total per h
9.6 7.8
5.2 4.3
NA
4.1 2.4
2.6 1.5
2.1 1.4
NOTE. In each row, statistically significant differences were found between values labeled with an asterisk (*) and those labeled with a dagger (†) (Wilcoxon or x2 tests; P ! .05 ). NREM, nonrapid eye movement; NA, not available; REM, rapid eye movement. a
Sleep efficiency was defined as the proportion of EEG sleep activity during night sleep in bed. Sleep latency was defined as the time between going to bed and the onset of sleep activity detected by EEG. b
HIV/AIDS • CID 2004:38 (1 February) • 431
With respect to sleep patterns, patients with insomnia had less rapid eye movement sleep (REM) than did HIV-positive patients without insomnia and healthy control subjects (mean percentage of sleep time SD, 14.2 6.3 vs. 22.6 2.8 and 22.6 2.1%, respectively; P ! .05). Moreover, the proportion of nonrapid eye movement (NREM) stages 3 and 4 (deep sleep) was significantly reduced in HIV-infected patients receiving EFV, irrespective of the diagnosis of insomnia, in comparison with that in healthy control subjects (mean percentage of sleep time SD, 4 4.2 vs. 10.8 1.8, P ! .05). As sleep fragmentation was one of the most striking findings in our study, the distribution of arousals along the different phases of sleep was analyzed in more detail among HIV-infected patients. Poor sleepers receiving Efv showed a higher number of awakenings during REM sleep than did HIV-positive subjects without insomnia. Conversely, the frequency of arousals was lower during deep-sleep phases in subjects with insomnia, compared with the frequency in normal sleepers. The differences did not reach statistical significance, most likely due to the small size of the study population (table 1). Among HIV-infected individuals, those with insomnia showed higher drug concentrations than did patients with normal sleep (mean SD, 4.3 2.9 vs. 2.7 0.7 mg/L, respectively; P value not significant). Moreover, sleep efficiency of !90% was nearly 2 times more common among patients with plasma Efv concentrations of 14 mg/L than among patients with Efv levels below this value (found in 62.5% vs. 37.5% of patients, respectively; P p .04). Discussion. This study explored the underlying alterations in EEG patterns among HIV-positive patients with Efv-related insomnia. The selected cohort had no history of insomnia or other mental disorders, and was not receiving substances with potential for sleep interference, such as psychoactive drugs, steroids, decongestants, or beta blockers [7]. Heavy alcohol or caffeine consumption was also ruled out in a psychiatric evaluation. Thus, the introduction of Efv was the only recognizable cause of the observed sleep alterations. EEG records showed that patients with Efv-related insomnia presented a remarkable reduction in sleep efficiency, which seemed to be due to difficulties in falling asleep and to a higher number of sleep interruptions. Regarding sleep architecture, the reduction in the duration of REM sleep was frequent among poor sleepers. The suppression of this stage of sleep can result in an increased intensity of the remaining REM episodes, a phenomenon that has been related to the development of vivid
432 • CID 2004:38 (1 February) • HIV/AIDS
dreams [8] and that is often reported by patients receiving Efv. The enhancement of dreaming activity could precipitate the occurrence of premature arousals, as shown by the fact that sleep interruptions took place during REM sleep stages in 52% of poor sleepers receiving Efv, but in only 32% of HIV-positive subjects without insomnia. A significant reduction in the duration of deep sleep was noticed in HIV-infected individuals, both in normal and poor sleepers, in comparison with healthy volunteers. These phases of sleep appear to be involved in physical restoration, and reduction in the amount of deep sleep may result in fatigue, somnolence, and distractibility during the day [9]. In conclusion, Efv-related neurologic adverse effects (especially insomnia, but also other neuropsychological symptoms) may be explained by alterations in sleep architecture. EEG monitoring may be a helpful tool to detect objective sleep abnormalities in patients complaining of insomnia while receiving Efv, especially for those with few therapeutic alternatives to Efv. The observance that plasma Efv levels correlate with reductions in sleep efficiency opens a door to investigate whether adjusting Efv dosages could ameliorate sleep disturbances without compromising the drug’s virologic efficacy.
References 1. Rubinstein M, Selwyn P. High prevalence of insomnia in an outpatient population with HIV infection. J Acquir Immune Defic Syndr Hum Retrovirol 1998; 19:260–5. 2. Blanch J, Martı´nez E, Rousaud A, et al. Preliminary data of a prospective study on neuropsychiatric side effects after initiation of efavirenz. J Acquir Immune Defic Syndr 2001; 27:336–43. 3. Nun˜ez M, Gonza´lez de Requena D, Gallego L, Jime´nez-Na´cher I, Gonza´lez-Lahoz J, Soriano V. Higher efavirenz plasma levels correlate with the development of insomnia. J Acquir Immune Defic Syndr 2001; 28: 399–400. 4. Buysse D, Reynolds C, Monk T, Berman S, Kupfer D. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989; 28:193–213. 5. Gilliam F, Kuzniecky R, Faught E. Ambulatory EEG monitoring. J Clin Neurophysiol 1999; 16:111–5. 6. Knobel H, Alonso J, Casado J, et al. Validation of a simplified medication adherence questionnaire in a large cohort of HIV-infected patients: the GEEMA Study. AIDS 2002; 16:605–13. 7. Obermeyer W, Benca R. Effects of drugs on sleep. Neurologic Clinics 1996; 14:827–40. 8. Thompson D, Pierce D. Drug-induced nightmares. Ann Pharmacother 1999; 33:93–8. 9. Horne J. Human slow wave sleep: a review and appraisal of recent findings, with implications for sleep functions, and psychiatric illness. Experientia 1992; 48:941–54.
