Aging 3: 313·324, 1991

ORIGINAL ARTICLES

Chronic use of triazolam in patients with periodic leg movements, fragmented sleep and daytime sleepiness M.H. Bonnet* and D.L. Arand** * Dayton VA Hospital, Wright State University, Dayton, Ohio, ** University of California, Los Angeles, California, U.S.A ABSTRACT. Many studies have shown a relationship between fragmented nocturnal sleep and daytime sleepiness. In the current study, 9 patients, aged 55-79, with fragmented nocturnal sleep secondary to periodic leg movements and objective daytime sleepiness, as verified by Multiple Sleep Latency Test (MSLT), had 12 weeks of treatment with 0.125 mg of triazolam following 2 screening nights and 2 placebo baseline nights; 2 final placebo nights were placed 5 nights following the last medication night. The medication increased total sleep time and sleep efficiency throughout the administration period, as compared to average placebo values; total leg movements were not changed. Generally, daytime performance, as measured by a vigilance task, and objective alertness, as measured by MSLT, were improved following the use of triazolam. No adverse reactions or significant side effects were noted. It was concluded that 0.125 mg triazolam, when used for up to 3 months, could improve sleep and daytime function in older patients with periodic leg movements, fragmented sleep, and daytime sleepiness. (Aging 3: 321-332, 1991) INTRODUCTION Carskadon et aL (1, 2) found that 33% of healthy older individuals without sleep complaint

had over 100 events during sleep which caused awakening or brief arousal. Those events included involuntary, periodic movement of the legs (periodic leg movements) and pauses in respiration (sleep apnea) . Periodic leg movements (PLMs) were found to increase as a function of age, with 5% of 30-50 year-old samples, 29% of 50+ year-old samples and 44% of 65+ year-old samples showing a clinically diagnostic number of leg movements (3). The leg movements are relevant because they frequently result in brief arousal or awakening, although this effect and its cause are usually not remembered in the morning. Decreased daytime alertness is a frequent complaint in persons over 55 years of age. Carskadon et aL (1, 2) found that older individuals who had frequent arousal during sleep also had pathological levels of sleepiness during the day. There was a significant negative correlation between daytime nap latencies (the measure of sleepiness) and the number of transient arousals on the prior night. Other studies of experimental sleep disturbance in normal young adults also support the contention that decreased daytime alertness is directly related to the number of brief arousals which occ ur during sleep (4). It was shown that significant daytime compromise can be found in individuals with as few as 40 brief arousals during the night; however, as little as 2·3 hours of

Key words: Aging, benzodiazepine, nocturnal myoclonus, periodic leg movements, sleep disorders. Correspondence: Michael H. Bonnet, Ph.D., VA Hospital, (127),4100 West Third St, Dayton, OH 45428, US.A. Received October 8, 1990; accepted March 13, 1991.

Aging, Vol. 3, N. 4 313

M.H. Bonnet and D.L. Arand

consolidated sleep could minimize daytime sleepiness, at least in young adults (5). As about 33% of patients diagnosed as having PLMs complain of excessive sleepiness (6), it is very likely that a period of consolidated sleep might increase their sleep restoration. Such patients traditionally have not been considered as candidates for benzodiazepine therapy because the long metabolic half-lives of common benzodiazepines could cause additional residual daytime sleepiness. As a result, most treatment studies of PLMs have dealt with insomniacs (711) and/or avoided objective verification of sleepiness (12-14). In a recent inquiry (15), the acute effects of the short -acting benzodiazepine, triazolam 0.125 mg and 0.25 mg, were evaluated in a group of objectively sleepy patients aged 55-75 with verified PLMs and fragmented sleep. In a 3-night administration period, triazolam increased total sleep time and improved daytime alertness and performance. Periodic leg movements are a chronic condition. While current research studies have shown symptomatic improvement on an acute basis, long-term treatment is required or symptoms are certain to return when medication is discontinued. Thus, it is relevant to determine the effectiveness of triazolam in a more chronic design in patients with PLMs and daytime sleepiness. The current study examined a 12-week chronic administration of 0.125 mg triazolam in a group of patients with disturbed sleep and daytime sleepiness secondary to PLMs during sleep to determine whether this treatment could consolidate sleep and improve daytime alertness. The 0.125 mg dose was chosen due to its efficacy in the acute study, consideration of increased plasma triazolam levels in the elderly (16), and the desire to avoid possible medication withdrawal problems (17). METHOD

Subjects To be considered for the study, patients were required to be between 55 and 79 years of age and to report daytime sleepiness. Patients had typically been referred to the sleep disorder center with a history of daytime sleepiness and

314 Aging, Vol. 3, N. 4

usually restless sleep or leg jerks during sleep. The patients had a physical exam and routine laboratory tests. Subjects were required to be in good health; none had chronic pain or any uncontrolled neurologic, hematologic, hepatic, renal, pulmonary, or cardiovascular disease. At the time of study, patients were not using any prescription tranquilizers or hypnotics. Potential subjects with significant psychopathology, uncontrolled diabetes, thyroid dysfunction, or evidence of narcolepsy were excluded. Potential subjects had a sleep history taken and were given a psychological screen. Consenting subjects were scheduled for two consecutive sleep laboratory screening nights followed by daytime testing and Multiple Sleep Latency Test (MSLT). Entry into the study proper required the presence of PLMs as traditionally defined (18) with the additional requirements that the PLMs result in at least 80 brief EEG arousals or awakenings, and that MSLT median latencies of less than 10 minutes were found. Patients with more than 5 mixed or obstructive apneas per hour of sleep, 5 or more central apneas per hour of sleep, or a combination of 6 or more central and obstructive apneas per hour of sleep were excluded from the study.

Study Each patient was recorded for five 2-day periods in the study proper, and thus spent 10 nights and 10 days in the laboratory. On the first 2 nights, either triazolam on both nights (5 Ss) or placebo on both nights (4 Ss) was administered. All subjects received placebO at home for the next 5 nights. On the 2 laboratory nights which followed, patients receiving placebo on the first 2 nights were administered triazolam and vice versa. The assignment of medication and placebo conditions to subjects in the initial 2-week cross over period was random and double-blind. Following this, all subjects received active medication for the next 12 weeks. Subjects spent 2 nights in the laboratory after 6 weeks on medication (Middle Drug) and the 2 final medication nights in the laboratory (Late Drug). After the last medication night, all subjects again received placebo at home for 5 nights before 2 final placebo lab nights. The final 2 laboratory nights were designed to be a baseline

Chronic triazolam in PLMs

recording period, so that changes in sleep from the beginning of the study (laboratory adaptation) and learning on the psychomotor tests could be appropriately assessed (19, 20). Subjects were instructed to take no psychotropic or sedating medication for 7 days prior to the beginning of the study and ·throughout the course of the study. Subjects also refrained from alcohol ingestion for 2 days prior to the study and during the laboratory sessions. Subjects were asked to continue normal caffeine consumption during the study. Patients arrived at the laboratory each evening approximately one hour prior to their normal bedtime (range, 21:00-23:00) and were prepared for standard polysomnographic recordings (21) including: EEG; EOG; EMG· EKG; airflow, as measured by Somniprobe nasal and oral thermistor; chest and abdominal movements (cardiopneumograph or SomniguageTM); left and right anterior tibialis EMG; and %Sa02 (Biox model II or Ohmeda 3700). Patients received placebo or medication 30 minutes prior to lights out. All standard polysomnographic variables (22), as well as apneas, leg movements from either leg, and all brief EEG arousals were scored in each sleep record. Leg movement scoring criteria included an initial increase in leg EMG to at least double the background EMG lasting 0.5 to 5.0 seconds. All leg movements during sleep were scored with the standard requirement that leg EMG increases occurred before any other sign of arousal but without regard to bursts (series of leg movements sometimes required for scoring) (18), because occasional isolated leg jerks associated with awakening or arousal exist in all patients with PLMs and are missed by traditional scoring (15). An EEG arousal was defined as a 3-second change in ongoing EEG including a burst of alpha, EEG speeding or chin EMG increase. The reliability of leg movement, of EEG arousal, and of sleep stage scoring was assessed, and it was found that scoring the first two was more difficult than scoring of normal sleep stages; as a result, an 85% level of agreement was accepted. When reliability was below this level, disagreements were checked and rectified, and the record was rescored by the primary scorer. Following each laboratory night, subjects re-



mained in the laboratory throughout the day for a post-sleep questionnaire, MSLT (at 08:00, 10:00, 12:00, 14:00, 16:00) (23), 30-minute addition tests (07:00, 11:00, 14:30) (24), 30-minute auditory vigilance tests (07:30, 11:30, 15:00) (24, 25), and short-term memory tests (10:30, 15:50) (26). The addition tests were evaluated for number of correctly completed and attempted problems; the auditory vigilance tests were scored for hit rate and false alarm rate, and the short-term memory test was graded for number of words correctly recalled. Subjects completed sleep logs for 2 weeks prior to beginning the study and for 2-week periods early (week 4-5) and late in the study (week 10-11). Subjective comparisons between premedication, early medication, and late medication intervals were therefore possible. During the home medication periods, subjects returned to the laboratory each 2 weeks to pick up a new bottle of medication and to receive or drop off sleep logs. At this time, pills were counted and any problems were assessed. While in the laboratory, subjects received individually packaged medication.

RESULTS A total of 9 (7 male and 2 female) subjects were qualified and enrolled in the study; their average age and weight were 66 years (range 59-76), and 1741b (range 110-220), respectively. Data from the study proper were analyzed by repeated analysis of variance. Nocturnal sleep, leg movement, and respiratory variables were analyzed by first comparing the 2 placebo baseline nights with the 2 late withdrawal nights (nights 6 and 7 off medication) to determine the existence of any late rebound insomnia or laboratory adaptation effects. The initial analysis had terms for time (baseline or withdrawal), night (first or second) and interaction. Summary placebo sleep data are presented in Table 1. The baseline data give no evidence for any medication rebound extending for 6 days after the medication was discontinued. Sleep was significantly improved on the postmedication baseline nights; total sleep time increased from 332 to 350 minutes (p < 0.01), latency to sleep onset decreased, and stage 3 sleep increased,

Aging, Vol. 3, N. 4 315

M.H. Bonnet and D.L. Arand

Table 1 - Pre-medication/Post-medication sleep values. Variable Total sleep time (min) Latency to Stage 1 (min) % Stage 1 % Stage 2 % Stage 3 % Stage 4 % Stage R Latency to REM (min) Wake during sleep (min) Early morning wake (min) Stage changes Sleep efficiency Awakenings No. of leg movements Leg index No. of EEG arousals Arousal index

Pre-medication

Post -medication

332 12 17 47 1 0 10 134 112 4 172 75 43 250 49 184 33

350 4 14 49 2.6 0 10 100 104 3 174 77 33 284 51 264 45

compared to premedication baselin~ nights. These results indicated contin'uing patient habituation to the sleep laboratory setting over the study periods (19, 20), as predicted. Therefore, it was felt that the most reasonable comparison to the medication conditions was to average baseline and recovery values, and compare average values to medication values. The medication analyses had terms for medication condition (df = 2), night of administration (df = 1) and night by condition interaction. No significant night by condition interactions were found, and interaction variance was therefore pooled with error to test the main effects for night and condition. Where significant F-values were found at p < 0.05 with Greenhouse-Geisser criterion, pairwise comparisons were performed with the Neuman-Keuls procedure at the 0.05 level and based upon Greenhouse-Geisser degrees of freedom. Significant differences are noted in the Tables.

Sleep data A summary of sleep stage values in the three conditions is seen in Table 2. While time in bed

316 Aging, Vol. 3, N. 4

F

p

9.7 11.6 1.4 1.0 9.2 4.0 0.19 1.13 0.94 0.13 0.03 1.06 1.75 1.33 0.17 4.09 2.83

0.01 0.009 NS NS 0.02 0.08 NS NS NS NS NS NS NS NS NS 0.08 NS

CONDITION

was held constant, total sleep time increased significantly (about 45 minutes) in the early medication condition, and remained significantly increased through the late medication condition (31-minute increase), compared to placebo. This increase in sleep was primarily in percentage Stage 2, which was also increased (p < 0.1) in medication conditions, as compared to placebo. Increased total sleep was also reflected in significantly increased sleep efficiency, and significantly decreased wakefulness throughout medication use. Wake during the night was examined in total and by thirds of the night (Wake 1/3, Wake 2/3, Wake 3/3). Wake time was primarily reduced during the first 2/3 of the night (significant only for the second period), but tended to remain lower than the placebo level throughout the night. In terms of awakening!disturbance variables, triazolam had no significant effect on total awakenings (i.e., number of 30-second or greater periods scored wake in the sleep record), early final awakening (wake time in the morning immediately prior to "lights on" time), or sleep

Chronic triazolam in PLMs

Table 2 - Sleep variables. Variable

Placebo*

Early Drug

Middle Drug

Late Drug

F CONDITION

457 340 8 16 48 2 0 10 108 76 4 173 38 43 3l.5 33.5

465 385

470 377 9 15 55 1 0 10 84 82 1 173 30 27 19.5 37.5

463 371 8 16 53 2 0 9 84 82 2 198 36 32 21 31

0.62 7.37 0.26 l.40 2.82 2.30 l.00 1.79 7.68 10.6 2.05 2.04 1.70 1.72 3.46 l.06

Time in bed (min) Total Sleep Time (min) Latency to Stage 1 (min) % Stage 1 % Stage 2 % Stage 3 % Stage 4 % Stage REM Wake during sleep (min) Sleep efficiency Early morning wake (min) Sta$e changes Awakenings Wake 1/3 (min) Wake 2/3 (min) Wake 3/3 (min)

10

20 50 1 0 12 73 84 0 180 38 30 21 22

p

NS 0.01 NS NS 0.1 NS NS NS 0.01 0.01 NS NS NS NS 0.04 NS

Differences

PLall PLall

* Average of 4 Placebo nights. PL: Placebo.

Leg movement and arousals

stage changes.

Respiratory events T riazolam had no significant impact on central apnea index, obstructive apnea index, longest apnea, baseline oxygen saturation level, minimal oxygen saturation level, or number ofdesaturations below 85% or greater than 6% Sa0 2 -

Leg movement and arousal data are summarized in Table 3. Triazolam did not decrease leg movements during sleep, and was also ineffective in reducing the total number of brief EEG arousals; in fact, a significant increase in EEG arousals was found in the late drug condition compared to placebo. When total leg

Table 3 - Leg movements and arousals. Variables No. of leg movements Leg index No. of EEG arousals Arousal index EEG arousals corrected for TST and leg jerks

F

Placebo

Early Drug

Middle Drug

Late Drug

CONDITION

270 50 225 39

312 52 213 34

312 52 246 39

277 48 288 46

1.17 0.20 3.77 5.04

NS NS 0.04 0.01

LD>all LD>PL=ED

0.16

0.12

0.15

0.17

4.77

0.01

ED

Chronic use of triazolam in patients with periodic leg movements, fragmented sleep and daytime sleepiness.

Many studies have shown a relationship between fragmented nocturnal sleep and daytime sleepiness. In the current study, 9 patients, aged 55-79, with f...
1MB Sizes 0 Downloads 0 Views