Current Medical Research and Opinion (1995), 13, No. 6, 317-324

Gold Horizontal Line

Double-blind comparison of zopiclone and flunitrazepam
in elderly insomniacs with
special focus on residual effects
O. Dehlin, M.D.,
B. Rubin,* M.D.,

Å. Rundgren,† M.D.

Geriatric Section, Department of Internal Medicine, University Hospital, 22185 Lund,
*Rhône-Poulenc Rorer, Helsingborg, and
Geriatric Clinic, Vasa University Hospital, Gothenburg, Sweden

Received: August 8th 1995

Gold Horizontal Line


In a double-blind study of 102 patients with a mean age of 79 years, zopiclone was compared to flunitrazepam. The patients rated their sleep in a diary. There was no statistically significant difference between the relatively low dose of 5mg zopiclone and 1mg flunitrazepam for eleven out of the twelve variables measuring subjective sleep quality and quantity. There were no differences between the drugs as regards patients' feelings of being rested or alertness.

Key Words

Zopiclone - flunitrazepam - insomnia - elderly - sleep


Subjective sleep quality becomes very much impaired in older age groups1 and this is also reflected in an increased hypnotic drug consumption in old age.1, 7

Benzodiazepines are widely used as hypnotics for older people and the side-effects of such drugs are also well-known.1, 10 One of the most distressing side-effects of benzodiazepines with a medium or long elimination half-life is a residual effect the following day, commonly known as hangover.9 Drugs with a shorter elimination half-life could therefore benefit older people by providing sleep without residual effects the next day.

Zopiclone is a hypnotic compound belonging to the cyclopyrrolone group, non-related to the benzodiazepines, but with a similar activity and equal efficacy in insomnia.4 Zopiclone is rapidly absorbed and has a plasma elimination half-life of 45 hours in young subjects, increased to ~6 hours in patients over 65 years old and to ~8 hours in those over 80 years of age.3 However, this is unlikely to have a profound clinical significance. There are no clinically significant active metabolites.

Flunitrazepam is commonly prescribed to treat insomnia in elderly patients. This benzodiazepine has a fairly long elimination half-life, 1319 hours, which could contribute to hangover effects during the daytime.

This study compared a low dose of zopiclone, 5 mg, and the commonly used dose of flunitrazepam, 1 mg, with its focus on efficacy (quality of sleep) and on residual effects in elderly patients with insomnia. The hypothesis was that the clinical efficacy of the low dose of 5 mg zopiclone should be as efficacious as 1 mg of flunitrazepam but with less influence on alertness.

Methods and Materials

In a multicentre, double-blind, randomized, parallel group study performed at 10 geriatric clinics in Sweden, patients of both sexes over 60 years of age were invited to take part. They had to be in a stable medical condition suffering from insomnia of at least 1 month's duration with a need for hypnotic treatment according to the investigator's judgement. The patients were to be intellectually fit to understand and follow simple instructions. To be included, they had to score 24 points on the MMSE, i.e. the Mini-Mental State Examination.2 Both previously treated and untreated patients were included. At least two of the following criteria had to be fulfilled to be included:

  1. Latency of sleep onset more than 30 minutes.
  2. More than two awakenings each night.
  3. Time taken to go back to sleep after nocturnal awakening of more than 45 minutes.
  4. Less than 6 hours of sleep each night or 80% of desired sleep time.
  5. Morning awakening at least 2 hours before expected time.

Exclusion criteria were patients who were suffering from insomnia due to severe somatic or psychotic diseases; with myasthenia gravis; with known severe liver or renal insufficiency; suffering from the sleep apnoea syndrome; with known allergy to benzodiazepines or zopiclone; undergoing neuroleptic and/or antidepressant treatment; with alcohol or drug abuse.

Each patient received the following treatment regimen:

The beginning and end of the study were single-blinded with the investigators knowing about the placebo treatment.

Efficacy variables

In a patient diary the following variables were recorded daily by the patient:

Clinical assessments were done once a week by the investigators during the entire trial, and the following question was asked: 'How do you judge your sleep for the preceding week?'. Compliance was checked by the patient's response to the diary question: 'Did you take your sleeping pill last night?', and also by counting the pills at the visits to the doctor.5 A nurse checked that the patient complied with the instructions.

Safety variables

At all visits the investigator put the open-ended question: 'Have you noticed anything special?' The answers were noted in the case report form. These were any events, positive or negative, that the subject experienced during the study. Adverse effects or intercurrent illness were considered as study events. The subject was asked the following non-specific question: 'How have you been feeling since your last visit?'.


A total of 107 patients entered the study and five dropped out during the 4 week period because they did not want to continue with the study. According to the protocol they could do so without giving any reason for their decision. The data shown is thus based on the 102 patients who completed the study. Fifty patients, 13 men and 37 women, completed their treatment in the zopiclone group; and 52, 17 men and 35 women, in the flunitrazepam group. They did not differ significantly in age (mean age 79 years, range 6095 years) or body weight. Nor was there any difference in the length of history of insomnia between the groups (mean 10 years, range 160 years).

The MMSE showed no significant differences between the groups: the zopiclone group had a mean of 27.5 ± 2.5 S.D. points and the flunitrazepam group 27.8 ± 2.6 S.D. points.

Almost all patients suffered from conditions for which they took drugs other than hypnotics. Drugs known to interfere with sleep such as bronchodilators, diuretics, beta-blockers, analgesics and antihistamines were evenly distributed between the groups.

The patients' main diagnoses are shown in Table I.

Table I. Main diagnoses of the patients

Diagnosis Zopiclone group Flunitrazepam group
Stroke 11 14
Other neurological disease 2 4
Fracture 7 4
Diabetes mellitus 3 1
Other disease 27 29


All patients completing the study were included in the efficacy analysis. The analyses were made using the mean of the measurements recorded during each week. The treatment effect, calculated as the difference between week 1 and weeks 2 and 3, respectively, was analysed for each variable. The difference between the two treatments was tested using analysis of variance. p-Values below 0.05 were considered statistically significant.

Information and consent

The study was approved by the ethics committee of the University of Gothenburg prior to commencement and all patients gave their informed consent.
The study was performed according to the Declaration of Helsinki (Tokyo Amendment).


All patients tolerated both drugs well. The five drop-outs were not due to side-effects but to a wish from the patients to withdraw, and they gave no reason for their decision. A statistically significant difference between the treatments in difficulty in falling asleep was noted in the data for weeks 2 and 3 (Table II).

Table II.
Means and standard deviations (S.D.) and differences (
p-values) in sleep parameters between zopiclone (Zopiclo) and flunitrazepam (Flunitra)
during the 2 weeks of active treatment

  Week 2 (1. week of active treatment) Week 3 (2. week of active treatment)
  Zopiclo Zopiclo Flunitra Flunitra   Zopiclo Zopiclo Flunitra Flunitra  
  Mean S.D. Mean S.D. p Mean S.D. Mean S.D. p
How did you sleep last night? 30.7 15.8 27.0 20.6 0.08 35.5 21.4 30.8 20.2 0.09
How many times did you wake up? 1.36 0.7 1.38 1.0 0.36 1.58 0.92 1.49 1.17 0.08
Did you wake up too early without being able to go back to sleep? 0.31 0.31 0.25 0.35 0.42 0.30 0.33 0.29 0.34 0.77
How did you feel on awakening? 31.4 18.4 33.2 22.6 0.65 32.2 21.2 29.6 19.0 0.78
How easily did you wake up? 42.3 19.1 45.4 25.0 0.73 44.1 23.5 41.7 20.5 0.40
Do you feel rested? 38.4 19.5 44.3 22.9 0.38 40.2 22.8 40.5 20.4 0.93
Did you have difficulty in falling asleep? 28.9 19.4 23.2 17.2 0.04* 34.6 22.7 24.2 18.5 0.002*
Do you remember having dreamt last night? 75.9 20.0 71.9 24.9 0.22 76.1 22.5 70.9 25.3 0.55
If you remember a dream, was it unpleasant, neutral, pleasant? 41.7 18.7 45.3 23.3 0.06 45.9 21.5 42.7 18.5 0.83
State of calmness 29.8 19.6 31.4 24.8 0.42 33.1 22.8 30.1 23.0 0.97
Alertness during the day 31.7 20.4 33.0 23.1 0.55 33.8 22.6 32.1 23.0 0.69
Need for daytime naps 2.8 0.7 2.9 0.7 0.87 2.9 0.7 2.7 0.7 0.44

*Less difficulty in falling asleep in the flunitrazepam group.

There was no significant difference in compliance ('Did you take your sleeping pill last night?') between the groups for any week, nor for the questions 'Have you noticed anything special?' and 'How have you been feeling since your last visit?'.

No serious or unexpected adverse events occurred and no patient dropped out due to adverse effects. The numbers of adverse effects experienced during the four weeks are shown in Table III. The type of adverse effects reported for three or more patients on either drug during weeks 2 and 3 together, as well as for the initial and final placebo week, are shown in Table IV.

Table III.
Number of adverse effects experienced during the 4 weeks in the two treatment groups

Week Zopiclone group Flunitrazepam group
1. Placebo 23 34
2. Active treatment 24 30
3. Active treatment 23 23
4. Placebo 15 18

Table IV.
Type of adverse effect (number of patients) reported in three or more patients
on either drug during weeks 2 and 3 combined (active treatment)
and during the initial and final placebo weeks

Symptom Zopiclone Flunitrazepam Initial placebo week Final placebo week
Vertigo 10 4 3 5
Depression 7 10 6 3
Arthralgia 7 5 6 0
Diarrhoea 2 6 0 0
Headache 2 6 8 3
Vomiting 0 0 9 3
Dry mouth 0 0 0 3
Myalgia 0 0 0 3
Vision problems 0 0 0 3


The efficacy of zopiclone 5 mg on different subjective sleep parameters in this elderly and relatively heavily disabled population was equal to that of flunitrazepam 1 mg with one exception. The flunitrazepam group showed less difficulty in falling asleep (Table II), which was found also in a study by Wickström et al.14 Most studies comparing zopiclone to other drugs have used 7.5 mg (for an overview see References 4 and 12) but in this elderly patient group 5 mg was sufficient to produce sleep of about equal effective quality as flunitrazepam 1 mg. The dose of 1 mg of flunitrazepam was chosen as it is the most commonly prescribed dose in Sweden for treatment of insomnia.

As the elimination half-life is considerably longer for flunitrazepam than for zopiclone one might expect more tiredness (i.e. fewer feelings of being rested the following morning), less alertness and greater need of daytime naps after the ingestion of flunitrazepam, but no such differences were found.

In two patient studies where zopiclone 7.5 mg was compared to flunitrazepam 1 mg, the patients' morning vigilance was better after zopiclone and the morning hangover was less.6, 11 In another study zopiclone 7.5 mg showed the same good sleep quality as flunitrazepam 1 mg, whereas the percentage of patients with improved daytime well-being was higher with zopiclone (37%) as compared to flunitrazepam (30%).8 Also, in another study comparing these two drugs in the same dosages, zopiclone did not affect daytime performance, whereas the findings for flunitrazepam were not so uniform.13

There might be different reasons for our somewhat different findings. The rating instruments used could have been too insensitive to detect small differences. The design of the study with ten participating centres and ten different raters might have increased the scatter and hence made it more difficult to detect statistically significant differences. The patients in the study were old and had many diseases and were therefore perhaps not able to report small changes in sleep variables. On the other hand, all patients were mentally intact as judged from the MMSE test and all patients rated their sleep as worse during the initial and final placebo weeks as compared to the active treatment weeks.


Clinical Data Care AB performed the statistical calculations.
This study was supported by Rhône-Poulenc Rorer, Box 33, 25053 Helsingborg, Sweden.


1. Bliwise, D. L., (1993). Sleep in normal aging and dementia. Sleep, 16, 4081.

2. Folstein, M. F., Folstein, S., and McHugh, P. R., (1975). Mini-Mental State. A practical method for grading the cognitive state of patients for the clinician. J. Psychiatr. Res., 12, 189198.

3. Gaillot, J., Le Roux, Y., Houghton, G. W., and Dreyfus, J. F., (1987). Critical factors for pharmacokinetics of zopiclone in the elderly and in patients with liver and renal insufficiency. Sleep, 10 (Suppl. 1), 721.

4. Goa, K. L., and Heel, R. C., (1991). Zopiclone: A review of its pharmacodynamic and pharmacokinetic properties and therapeutic efficacy as an hypnotic. Drugs, 32, 4865. (Updated version, June 1991.)

5. Nordiska Läkemedelsnämnden, Uppsala, (1989). Good Clinical Trial Practice, Nordic Guidelines. NLN Publication No. 28.

6. Harrison, C., Subhan, Z., and Hindmarch, I., (1985). Residual effects of zopiclone and benzodiazepine hypnotics on psychomotor performance related to car driving. Drugs Exptl. Clin. Res., 12, 823829.

7. Nicholson, A. N., Bradley, C. M., and Pascoe, P. A., (1989). Medications: effect on sleep and wakefulness. In: Kryger, M. H., Roth, T., and Dement, W. C. (Eds), Principles and Practice of Sleep Medicine, W. B. Saunders, Philadelphia, pp. 228-236.

8. Rüther, E., Clarenbach, P., Hajak, G., Fischer, W., and Haase, W., (1992). Zopiclon bei Patienten mit Schlafstörungen. Münchener Med. Wochenschr., 46, 753757.

9. Sheikh, J. I., (1992). Problems associated with long-term benzodiazepine use in the elderly. In Shamoian, C. A. (Ed.), Psychopharmacological Treatment Complications in the Elderly, American Psychiatric Press, Washington, pp. 71-87.

10. Skoog, I., Nilsson, L., Landahl, S., and Steen, B., (1993). Mental disorders and the use of psychotropic drugs in an 85-year-old urban population. In Skoog, I. (Ed.), 'Mental Disorders in the Elderly', Int. Psychogeriatr., 5, 3348.

11. Subhan, Z., and Hindmarch, I., (1984). Effects of zopiclone and benzodiazepine hypnotics on search in short-term memory. Neuropsychobiol., 12, 244248.

12. Wadworth, A. N., and McTavish, D., (1993). Zopiclone: a review of its pharmacological properties and therapeutic efficacy as an hypnotic. Drugs & Aging, 3(5), 441459.

13. Wickström, E., Godtlibsen, O. B., Bredesen, J. E., and Jensen, M. H., (1988). Performance and mood following partial sleep deprivation: a randomized, double-blind crossover study of zopiclone, triazolam, ethanol and placebo. Hum. Psychopharmacol, 3, 311.

14. Wickström, E., Barbo, S. E., Dreyfus, J. F., Jerkö, D., Kleiven, R., Slattbrekk, R., and Stray Tonnesen, J. N., (1983). A comparative study of zopiclone and flunitrazepam in insomniacs seen by general practitioners. Pharmacol., 27 (Suppl. 2), 165172.

Gold Horizontal Line