British Journal of Anaesthesia 1994; 72: 443-446
Zopiclone as a preoperative night hypnotic: a double-blind
comparison with temazepam and placebo
C. WHITEHEAD, L. SANDERS, I. APPADURAI, I. POWER, M . ROSEN AND J. ROBINSON
SUMMARY
We have examined the hypnotic effects of zopiclone
7.5 mg and temazepam 20 mg compared with
placebo in a double-blind, randomized, clinical
study of 60 patients on the night before operation.
Evaluation was both subjective (visual analogue
scales and a sleep questionnaire), to measure the
quality of sleep, and objective (critical flicker fusion,
object recall and paired associates tasks), to
measure residual impairment. We found that zopi-
clone was an effective single-dose hypnotic with
similar residual effects to the benzodiazepine and it
may therefore provide a suitable alternative to
benzodiazepines. (Br. J. Anaesth. 1994; 72:
443-446)
KEY WORDS
Hypnotics, benzodiazepines: temazepam. Hypnotics, cyclo-
pyrrolones: zopiclone.
Hypnotic agents which facilitate the production and
maintenance of sleep in insomniacs can be used also
for transient insomnia occurring for extraneous
reasons in those who normally sleep well [1]. It has
therefore become common practice to offer patients
a hypnotic agent on the evening before surgery [2].
A hypnotic selected for this purpose should be
without adverse effects and short-acting, leaving the
patient without residual preoperative sequelae. It is
customary to prescribe benzodiazepines for hypnosis
on the night before operation.
Zopiclone is a cyclopyrrolone but has many of the
characteristics associated normally with a benzo-
diazepine: anticonvulsant, myorelaxant, antiaggres-
sive, sedative-hypnotic and "anticonflict" [3]. It is
short-acting with a half-life of 3.5-5 h [4]. Exam-
ination of the dose-response profile for zopiclone in
insomniac patients indicates that maximum benefit
in terms of both sleep induction and maintenance
can be obtained at a dose of 7.5 mg [5]. This dose has
been shown to have some advantages over triazolam
[6], nitrazepam [7] and flurazepam [8] and to be
comparable with temazepam [9]. Residual effects
have been shown to be less with zopiclone than with
nitrazepam [10, 11], a dose of 7.5 mg producing only
marginal hangover effects whilst a dose of 5 mg
results in no detectable effects at 12 h [12]. A dose of
7.5 mg was considered optimal, however, as it
produces significant hypnotic effect with minimum
residual impairment [11, 5]. There is evidence that
memory may be impaired in the morning after
administration with this dose but that this im-
pairment is less with zopiclone than with some
benzodiazepines [5].
Zopiclone has been used as a hypnotic on the
night before operation and been shown to be
comparable with lormetazepam and midazolam using
subjective assessment of quality of sleep and psycho-
motor performance, but no placebo control was used
[13]. However, as temazepam is also used frequently
for this type of hypnotic we have compared zopiclone
7.5 mg with temazepam 20 mg and placebo, assessing
subjective and objective effects, including psycho-
motor performance and memory.
PATIENTS AND METHODS
After obtaining approval from the Hospital Ethics
Committee, we studied 60 patients [14] of ASA
grades I and II, aged 18-65 yr. Exclusion criteria
included evidence of clinically relevant concurrent
disease, a history of drug or alcohol abuse and
concurrent drug therapy with centrally acting drugs
or those known to interact with the trial drugs.
Eligible patients were recruited from the afternoon
surgical lists of those presenting for minor head and
neck surgery (e.g. dental extraction, nasal poly-
pectomy, submucous resection of turbinates, tonsil-
lectomy, cyst removal, etc.). After admission the
evening before surgery, the patients were invited to
participate in the study, a full explanation was given
and written consent obtained. The psychological test
battery was completed for the first time during this
initial interview which lasted approximately 20 min.
A computer-generated, randomized code was used
to determine to which of the three drug regimens the
patient was allocated. The interviewer and patient
were blind to the allocation of the treatment, the
code for which was not available to the investigator
until after completion of all data collation. Each
group received either zopiclone 7.5 mg, temazepam
20 mg or placebo. To overcome the problem of the
different presentation of each drug, a double-dummy
CAROLINE WHITEHEAD, B.A., R.C.N.T., S.R.N., LALAGE SANDERS,
B.SC.ECON., PH.D., A.F.B.PS.S., C.PSYCHOL., IAN APPADURAI, F.R.C.A.,
IAN POWEH, B.SC., M.D., F.R.C.A., MICHAEL ROSEN, C.B.E., M.B.,
CH.B., F.R.C.A., Department of Anaesthetics, University of Wales
College of Medicine, Heath Park, Cardiff. JAMES ROBINSON, B.SC.,
M.A., PH.D., F.B.PS.S., C.PSYCHOL., School of Psychology, Uni-
versity of Wales College of Cardiff, Park Place, Cardiff. Accepted
for Publication: October 29, 1993.
Correspondence to C.W.
444
technique was used with each patient receiving both
a capsule and a tablet, either one or both of which
was placebo. The trial medications were admin-
istered to the patient by the night nursing staff at
about 22:00, this being the usual time for the
administration of night hypnotics on the ward.
Between 08:00 and 09:00 the following morning,
the patients were interviewed about their night's
sleep using a modification of the Leeds sleep
evaluation questionnaire [15]. The psychological test
battery was then completed for the second time and
the presence or absence of spontaneously reported
side effects was noted.
The test battery was designed to explore changes
in performance and comprised five standard psycho-
metric measures, as detailed below, all with
established use in psychological experimentation
and assessment of drug effects. Of these tests three
were objective measures of performance and two
measured the patients' subjective experiences.
Objective measures
Critical flicker fusion threshold (CFFT) [16, 17]
was used to detect CNS depression as it is thought
widely to be an objective measure of cortical arousal.
The standard hand-held Leeds psychomotor tester
was used to determine the threshold at which a light
flickering at an increasing rate is perceived as being
constant and, conversely, when the frequency is
decreasing the threshold at which a flicker can be
detected. The score used was the mean of three of
each type of trial. Ambient light, illumination and
viewing distance were controlled throughout the
testing sessions.
Object recall test (ORT) [18] was used to assess
short-term spatial recall. The patient was presented
with a card on which were drawn 15 simple everyday
objects and was allowed 1 min in which to study the
card. The card was removed and the patient was
asked to recall as many of these objects as possible in
2 min. The score was the number of objects recalled
correctly, the value used in the analysis was the
difference between evening baseline and morning
test scores.
Paired associates task {PAT) was used to assess
short-term semantic recognition (after the Wechsler
memory scale [19, 20]). The patient listened through
headphones to a presentation of a list of 10 pairs of
words; in six pairs the two words have an obvious
association and in the remaining four there was no
association. Recognition was tested immediately by
presenting the first word of each pair followed by
four words, the task being to identify which of those
four was in the original pair. The score was derived
from the number of words identified correctly, the
non-associated words scoring double (maximum 14
points); the value used in the analysis was the
difference between evening baseline and morning
test scores.
Subjective measures
Visual analogue scales (VAS) [21] were used to
monitor levels of anxiety and mood. For each
BRITISH JOURNAL OF ANAESTHESIA
question the subject was required to draw a per-
pendicular mark through an ungraded 100-mm line,
the ends of which were labelled with bipolar
adjectives. The score for each VAS was derived by
measuring the position of the mark to the nearest
mm; the value used in the analysis was the difference
between evening baseline and morning test scores.
By way of instruction patients were asked to do a
demonstration VAS describing their height [22]
before completing the first VAS battery.
Sleep questionnaire (SQ) [15]. Patients were asked
to complete a modification of the Leeds sleep
evaluation questionnaire (LSEQ) which utilizes VAS
[23] to measure three of the four LSEQ subjective
dimensions of sleep: ease of getting to sleep, quality
of sleep, ease of waking. It also requires the patient
to estimate the time taken to get to sleep and the
length of time slept.
Statistical methods
The raw data analysis was conducted on the SPSS
statistical package running on a Novell Netware
system through a Mertec 486DX PC computer. The
times from the questionnaire and the results from
the CFFT task were analysed using ANOVA (in the
case of CFFT with the baseline as covariate) and a
paired t test where appropriate. The remaining non-
parametric data were analysed using a Kruskal-
Wallis analysis of variance or Mann-Whitney test, as
appropriate.
RESULTS
A total of 63 patients were recruited into the study,
but three were subsequently excluded (two with-
drew, the third took an antihistamine). The three
groups were comparable in age, height, weight and
gender; less than 50% of subjects (25) were smokers
and 25 % of those smoked less than 10 cigarettes per
day (table I).
Objective measures
Of the three objective measures only the results of
the CFFT showed a statistically significant overall
treatment effect (P < 0.05). The placebo group
showed a minor improvement in the morning
assessment, but this diurnal variation was not
apparent in either of the two active drug groups
(table II). Further analysis showed a significant
difference between zopiclone and placebo (P < 0.01),
but not between zopiclone and temazepam
(P > 0.05).
There was no significant difference between the
groups in the scores for the memory tasks (table III).
TABLE I. Patient characteristics (mean (range or SD) or number)
Zopiclone Temazepam Placebo
(n = 20) (n = 19) (n = 21)
Age (yr) 31.5(20-59) 30.6(18-60) 33.2 (18-65)
Height (cm) 167.8(9) 168.4(11) 166 2(9)
Weight (kg) 73.0(14) 72.2(15) 74.3(11)
Smokers 9 7 9
Gender (F/M) 10/10 12/7 15/6
PREOPERATIVE NIGHT HYPNOTIC 445
TABLE II. Mean (SD) results of the critical flicker fusion threshold patients who had received placebo, whilst those
(0-50 Hz) receiving zopiclone or temazepam recorded similar
Zopiclone Temazepam Placebo P scores (P < 0.001). Further analysis showed a signifi-
(n = 20) (n=19) (n = 21) (between groups) cant difference between the zopiclone and placebo
groups (P < 0.001) but not between those given
Baseline 29.2(3.7) 29.2(4.3) 28.4(2.9) zopiclone or temazepam (P > 0.05). Quality of
0.05
Morning 28.6(3.4) 29.3(3.9) 30.0(2.7) sleep was also comparable between those given the
active drug and poorer in those given placebo
(P < 0.05). Further analysis showed a significant
TABLE III. Memory tests. Mean (SD) change from baseline: a minus difference between the zopiclone and placebo groups
number would indicate impairment. There were no significant
differences between groups (P < 0.05), but not between those given zopiclone or
temazepam (P > 0.05). The placebo group reported
Zopiclone Temazepam Placebo the least difficulty in waking but there was no
(n = 20) (n=19) (n = 21) significant difference between groups (P > 0.05).
Object recall 0.0(1.8) 0.6(1.6) 0.8(1.6) There was a significant difference between groups
Paired associates 1.0(3.4) 1.7(2.5) 0.5 (3.6) in the amount of time they reported having taken to
fall asleep (P < 0.05) and the length of time slept
(P < 0.01). In each of these cases further analysis
TABLE IV. Subjective tests. Mean (SD) change from baseline: a confirmed that the scores from the zopiclone group
minus number indicates reduction. There were no significant were significantly better than those from the placebo
differences between groups group (P < 0.01 and P < 0.005, respectively) but not
Zopiclone Temazepam Placebo different from the temazepam group (P > 0.05).
(n = 20) (" = 19) (" = 21)
Anxiety -6.5(22) -3.3(32) -7.9(26) DISCUSSION
Confusion 18.6 (26) 12.5 (29) 13.4(30)
Depression -0.4(14) -0.6(9) 4.2 (20) It would seem that zopiclone is an effective hypnotic
Tired on waking 14.2 (35) 16.2 (30) 18.3 (35) on the night before operation in reducing the length
Tired at interview 8.2 (35) 3.5 (30) 10.8(31) of time taken to fall asleep and increasing the
duration of sleep. Furthermore, patients given
zopiclone appear to experience greater ease of getting
However, in the ORT, less than 33 % of patients to sleep and superior quality of sleep compared with
given zopiclone and just less than 50% of those those given placebo. Both groups given an active
given temazepam showed an improvement in the drug had mean sleep scores of more than 50. As these
morning test compared with nearly 66 % of those scales require the patient to compare the night's
given placebo. sleep before operation with their normal night's
sleep, scores greater than 50 may suggest that it was
Subjective measures easier and quicker to get to sleep than usual. A linear
Mean scores indicated a reduction in anxiety and interpretation of individual responses to VAS may
an increase in confusion for all three groups in the be inappropriate.
morning compared with the previous evening, whilst The scores for patients given zopiclone were
the placebo group alone recorded an increase in comparable with those of the patients given temaze-
depression (table IV)- The groups were similar in the pam throughout the study. This accords with the
level of tiredness reported, both retrospectively at work of Rettig and colleagues [13] who found
waking, and at the time of interview. There was no zopiclone was comparable with lormetazepam and
significant difference between groups in any of the midazolam when used in this context; however the
subjective measures. Although there was no signifi- lack of a placebo control restricts interpretation of
cant difference in the measured anxiety, only 25 % of their results. These findings accord with the review
the zopiclone group (five) recorded an increase in by Jonas and colleagues [24] which showed that
anxiety in the morning compared with more than zopiclone and zolpidem were in many ways
50% of each of the other groups (11 in the indistinguishable from benzodiazepine hypnotics.
temazepam and 12 in the placebo group). There was a trend for both these groups of patients
On two of three dimensions of the SQ there was a to report that waking was more difficult than usual.
significant difference between groups (table V). There were no clinically significant hangover effects
Getting to sleep was reported to be most difficult by with either of the active drugs, although it may be

TABLE V. Mean (SD) results of the sleep questionnaire

Zopiclone Temazepam Placebo P

(n = 20) (n = 19) (n = 21) (between groups)
Time from drug to sleep (h) 0.7 (0.4) 0.9 (1.0) 1.4(1.0) 0.043
Length of sleep (h) 6.8(1.1) 6.2(1.4) 5.5(1.4) 0.009
VAS
Ease of getting to sleep 57.9 (14) 55.4(13) 37.0(15) 0.0001
Quality of sleep 54.5 (31) 52.8 (22) 31.9(22) 0.0164
Ease of waking 60.5 (25) 53.5 (18) 43.2 (18) —
446
that they masked diurnal improvement in perform-
ance more apparent in the placebo group. There was
no evidence that either of the active drugs affected
the mood of the patients, as measured by VAS. But
two trends were apparent: the placebo group tended
to score higher on depression, possibly because of
the poor night's sleep, and fewer patients given
zopiclone recorded an increase in anxiety in the
morning compared with the other two groups.
It could be argued that these measures are
dependent on the patient's ability to estimate the
passage of time when in a drugged state and loss of
critical judgement may be a side effect of centrally
acting drugs. However, insomnia is a subjective
experience and there is wide individual variation on
acceptable sleep variables. The subjective estimates
of "time to go to sleep" and "sleep duration" by
normal subjects have shown good correlation with
objective recordings [25] and subjective measure-
ment is a useful means of assessing the sleep
facilitation properties of different hypnotics.
When measuring dimensions of effect, subject
preselection can minimize the effects of the in-
teraction between personality and performance [26].
This sample may not have included those patients
who would have been inclined to extreme pre-
operative insomnia. The study criteria prevented the
inclusion of those who demonstrated excessive
anxiety at the initial interview and those whose
possible diagnoses included cancer.
It may be noted that many subjects claimed at the
preliminary interview to be "good sleepers" and
doubted their need for chemical assistance. However,
laboratory reports have demonstrated the "first
night" effect, where environmental changes have
disrupted sleep patterns [27]. An increase in the
perceived time taken to get to sleep and more
nocturnal awakenings are typical symptoms. Wein-
man [27] has attributed these alterations to many
factors, including unusual noises, changes in routine,
anxiety and pain or discomfort; each of these may be
associated with the preoperative night.
We conclude that zopiclone is an effective single-
dose hypnotic, comparable with temazepam, and
may provide a suitable alternative to benzo-
diazepines.
REFERENCES
1. British National Formulary. London: British Medical Associ-
ation and Royal Pharmaceutical Society of Great Britain,
1993; No. 25.
2. Sanders LD, Ycomans WA, Rees J, Rosen M, Robinson JO.
A double blind comparison between nitrazepam, lorazepam,
lormetazepam and placebo as preoperative night sedatives.
European Journal of Anaeithesiology 1988; 5: 377-383.
3. Julou L, Bardonne MC, Blanchard JC, Garret C, Stutzmann
JM. Pharmacological studies on zopiclone. Pharmacology
1983; 27 (Suppl. 2): 46-58.
4. Ueki S. Behavioral pharmacology of zopiclone. Sleep 1987;
10 (Suppl. 1): 1-6.
BRITISH JOURNAL OF ANAESTHESIA
5. Nair NPV, Schwartz G, Dimitri R, Le Morvan P,
Thavundayil JX. A dose range finding study of zopiclone in
insomniac patients. International Clinical Pharmacology 1990;
5 (Suppl. 2): 1-10.
6. Autret E, Maillaird F, Autret A. Comparison of the clinical
hypnotic effects of zopiclone and triazolam. British Journal of
Clinical Pharmacology 1987; 31: 621-623.
7. Jovanovic UJ, Dreyfus JF. Polygraphical sleep recording in
insomniac patients under zopiclone or nitrazepam. Phar-
macology 1983; 27 (Suppl. 2): 136-145.
8. Mamelack M, Buck L, Csima A, Price V, Smiley A. Effects
of flurazepam and zopiclone on the performance of chronic
insomniac patients: a study of ethanol-drug interaction. Sleep
1987; 10 (Suppl. 1): 79-87.
9. Wheatley D. Zopiclone: a non-benzodiazepine hypnotic—a
controlled comparison with temazepam in insomnia. Brttish
Journal of Psychiatry 1985; 146: 312-314.
10. Anderson AA. Zopiclone and nitrazepam: a multicentre
placebo controlled comparative study of efficacy and tolerance
in insomniac patients in general practice. Sleep 1987; 10
(Suppl. 1): 54-62.
11. Klimm HD, Dreyfus JF, Delmotte M. Zopiclone versus
nitrazepam: a double-blind comparative study of efficacy and
tolerance in elderly patients with chronic insomnia. Sleep
1987; 10 (Suppl. 1): 79-87.
12. Broadhurst A, Cushnaghan RC. Residual effects of zopiclone
(Imovane). Sleep 1987; 10 (Suppl. 1): 48-54.
13. Rettig HC, de Haan P, Zuurmond WWA, v Leeuwen L.
Effects of hypnotic on sleep and psychomotor performance. A
double blind randomized study of lormetazepam, midazolam
and zopiclone. Anaesthesia 1990; 45: 1079-1082.
14. Altman DG. Statistics and ethics in medical research III:
How large a sample? Brttish Medical Journal 1980; 281:
1336-1338.
15. Hindmarch I. The Leeds Sleep Evaluation Questionnaire
(LSEQ) as a measure of the subjective response to nocturnal
treatment with benzodiazepines. In: Burrows GD, ed.
Biological Psychiatry: Recent Studies. London: J Libbey,
1984; 228-239.
16. Wittenborn JR. Effects of benzodiazepines on psychomotor
performance. British Journal of Clinical Pharmacology 1979;
7: 61S-67S.
17. Smith JM, Misiak H. Critical Flicker Fusion (CFF) and
psychotropic drugs in normal human subjects—a review.
Psychopharmacology 1976; 47: 175-182.
18. Bethune DW. Test of delayed memory recall suitable for
assessing postoperative amnesia. Anaesthesia 1981; 36:
942-948.
19. Weschler D. Standardized memory scale for clinical use.
Journal of Psychology 1945; 19: 87-95.
20. Sanders LD, Piggott SE, Isaac PA, Okell R, Roberts B,
Rosen M, Robinson JO. Reversal of benzodiazepine sedation
with the antagonist flumazenil. British Journal of Anaesthesia
1991; 66: 445-453.
21. Revill SI, Robinson JO, Rosen M, Hogg MJ. The reliability
of a linear analogue for evaluating pain. Anaesthesia 1976; 31:
1191-1198.
22. Robinson JO, Sanders LD. Subjective assessment. In:
Klepper ID, Sanders LD, Rosen M, eds. Ambulatory
Anaesthesia: Impairment and Recovery. Oxford: Blackwells
Scientific Publication, 1991; 158-167.
23. Bond A, Lader M. The use of analogue scales in rating
subjective feelings. British Journal of Medical Psychology
1974; 47: 211-218.
24. Jonas JM, Coleman BS, Sheridan AQ, Kalinske RW.
Comparative clinical profiles of triazolam versus other
shorter-acting hypnotics. Journal of Clinical Psychiatry 1990;
S3 (Suppl.): 19-33.
25. Johns MW. Validity of subjective reports of sleep latency in
normal subjects. Ergonomics 1977; 20: 683-690.
26. Hindmarch I. Psychomotor function and psychoactive drugs.
British Journal of Pharmacology 1980; 10: 189-209.
27. Weinman J. An Outline of Psychology as Applied to Medicine,
2nd Edn. Bristol: Wright, 1987.