324
Randomised controlled trial of nasal nicotine spray
in smoking cessation
Studies with nicotine chewing gum and nicotine
skin patches indicate that nicotine replacement can
help people to give up smoking. The rapidity with
which nicotine is absorbed when given as a nasal
spray suggests that it might be effective for those for
whom the other means of replacement are too slow.
The efficacy and safety of a nasal nicotine spray as
an adjunct to group treatment for stopping smoking
were assessed in a randomised, double-blind,
placebo-controlled trial in which 227 cigarette
smokers attending the Maudsley Hospital Smokers
Clinic received 4 weeks of supportive group
treatment plus active nicotine (0·5 mg per shot) or
placebo nasal spray. The main end-point was
biochemically validated complete abstinence from
smoking from the third week of group treatment until
the 12-month follow-up. Side-effects were assessed
by self-reports and, where necessary, by physical
examination. Of subjects assigned to active
treatment 26% (n=30) were validated abstinent
throughout the year, compared with 10% (n = 11) of
those assigned to placebo (relative abstinence rate
2&middot;6, 95% Cl 1&middot;5-4&middot;5, p<0&middot;001). The advantage of
the active spray was greatest in the heaviest smokers.
Plasma nicotine concentrations from the spray were
typically between one-half and three-quarters of
baseline smoking levels. Tobacco-withdrawal
symptoms, craving for cigarettes, and weight gain in
abstinent subjects were reduced by the active spray.
Minor irritant side-effects were frequent in both
active and placebo sprays, but only 2 subjects had
the spray discontinued as a result. No serious adverse
effects were encountered.
Nasal nicotine spray combined with supportive
group treatment is an effective aid to smoking
cessation.
Introduction
The value of nicotine replacement in the treatment of
dependent smokers is now well established. Numerous
studies have shown that nicotine chewing gum (Nicorette,
Kabi Pharmacia Therapeutics, Helsingborg, Sweden) is
more effective than placebo or various psychological
methods. More recently, encouraging results have been
reported with a nicotine skin patch.1-3 One potentially
important difference between different means of giving
nicotine is the rate of nicotine absorption. It is slowest with
the transdermal route (nicotine patch), which takes 6-8 h to
produce a flat peak. Somewhat faster is buccal absorption
(nicotine gum), with 30 min chewing required for each piece
of nicotine gum. Faster still, and second only to inhalation, is
the nasal route. Testimony to the efficiency of this route
comes from the long history of dry nasal snuff use.4 It has
been suggested that the satisfaction and other positive
reinforcing effects of smoking may depend, at least in part,
on the very rapid rate of nicotine absorption.5 The nasal
route might, therefore, be an effective method of treatment
for those smokers for whom nicotine absorption from the
oral mucosa or skin is too slow.
Preliminary work with a nasal nicotine spray (NNS)
showed that plasma nicotine concentrations peaked 5 min
after a 2 mg dose and that concentrations in the smoking
range were easily obtained after repeated doses.6 Results
from an uncontrolled clinical trial of nasal nicotine have been
encouraging. We report here the results of a 1-year
randomised, double-blind, placebo-controlled trial of NNS
as an adjunct to group treatment for helping smokers give up
the habit.
Subjects and methods
Subjects
Between January, 1989, and March, 1990, 429 consecutive
patients referred to the Maudsley Hospital Smokers Clinic were
screened for eligibility for the trial. To be eligible subjects had to be
current daily cigarette smokers, aged 18-68 years, in good health,
motivated to stop smoking, and willing to adhere to the trial
protocol, which included attendance at follow-ups. Exclusion
criteria were history of cardiovascular disease, hypertension,
diabetes, or severe allergy; current use of psychotropic medication;
use of nicotine gum in the past year; current abuse of alcohol or
other drugs; and pregnancy.
Of 274 who were eligible and willing to participate only those who
turned up for their first group treatment session (n = 227) were
randomised and entered into the trial. They drew a card marked A
or P for allocation to active or placebo group, respectively. 116
subjects were assigned the active nasal spray and 111 placebo. There
were 17 treatment groups ranging in size from 9 to 16 subjects
(mean 13), with about equal numbers assigned to active and placebo
spray in each group. Subjects and therapists were blind to spray
assignment. Subjects gave written informed consent to participate
in the trial, which was conducted in accordance with the Helsinki
Declaration (1983) and approved by an independent ethics
committee. Active and placebo subjects were well matched on
demographic and smoking characteristics (table I).
Nasal nicotine spray
The NNS device (Kabi Pharmacia Therapeutics, Helsingborg,
Sweden) consisted of a pocket-sized multidose bottle with a pump
mechanism fitted to a nozzle for insertion into the nostril. The active
spray delivered 0-5 mg nicotine per 50 ul shot. A dose consisted of
two shots (1 mg nicotine), one to each nostril. The placebo spray
contained black pepper oleo resin (piperine) to mimic the sensation
ADDRESSES: Health Behaviour Unit, National Addiction Centre,
Institute of Psychiatry, 4 Windsor Walk, London SE5 8AF
(G. Sutherland, MPhil, J. A. Stapleton, MSc, M. A. H. Russell, FRCP, M J.
Jarvis, MPhil, P. Hajek, phD, Michael Belcher, RMN); and Poisons Unit,
New Cross Hospital, London (C. Feyerabend, PhD) Correspondence
to Ms Gay Sutherland.
 325

TABLE I-SUBJECT CHARACTERISTICS ON ENTRY INTO STUDY TABLE II-PERCENT CONTINUOUSLY ABSTINENT AND
&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;.&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;.&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash;&mdash; BIOCHEMICALLY VALIDATED AT FOLLOW-UP

*Months after randomisation

tNos in parentheses refer to actual number of subjects

subjects also rated-on a 5-point scale ranging from "not at all" to

"very much"-the extent to which NNS relieved craving, and the
Nos m parentheses refer to SD. extent to which it was "helpful" in enabling them to refrain from
*Maximum score 9.
smoking.
Side-effects and positive subjective effects were recorded on a
of nicotine in the nasal passage. Spray use was left to the individual, symptom checklist at all treatment sessions. An open-ended
who was allowed up to a maximum of 5 doses/h (5 mg nicotine) question allowed for reporting of additional side-effects. Overall
and 40 doses per day (40 mg nicotine). Maximum dosage severity of side-effects was rated on a 4-point scale, ranging from
recommendations were based on the average nicotine intake of 1 mg "no problem at all" to "unacceptable". At each follow-up the
per cigarette.8 therapist asked about side-effects since the last visit. A physician,
The first dose of NNS was taken at the first group session, after a blind to spray assignment, examined subjects with signs or
demonstration. Subjects were advised to use the NNS whenever symptoms causing them concern, and all subjects using the spray
they felt an urge to smoke and were told to stop using the spray if for 6 months or longer were referred to an ear, nose, and throat
they resumed smoking. An instruction leaflet was provided. specialist, irrespective of whether they had had adverse reactions.
Subjects were told not to try each other’s sprays for reasons of
hygiene. The recommended duration of NNS use was 3 months but Determination of abstinence
it was not withheld from those expressing a strong wish to continue
The 227 subjects formed the base for all assessments of
beyond this time. No formal dose reduction regimen to wean
abstinence. The principal criterion of abstinence was self-reported
subjects off NNS was imposed.
continuous abstinence from smoking from the start of the last week
of group treatment to the 12 month follow-up, validated by a CO
Group treatment and follow-up concentration of < 10 ppm at the end of group treatment and at all
Supportive group treatment consisted of six sessions over a five follow-ups. A secondary criterion, which allowed some
month, each lasting 60-75 min. The first session, mainly to impart smoking "slips", was based on the recommendations of the
information about spray use and to prepare subjects for stopping National Conference on Smoking Relapse13-CO validated
smoking, was followed by a second 48 h later; thereafter sessions abstinence at the same six sessions as for the main criterion, with not
were held weekly from the date of the first session. Each group was more than 6 consecutive days of smoking and not more than 9 days’
run by two therapists using a group-oriented treatment smoking in total from week 4 (end of group treatment) to the
approach.9,lO Subjects were urged to stop smoking from the first 12-month follow-up. To compare active versus placebo rates of
session but were not excluded from subsequent sessions if they resumption of smoking by use of life-table methods, a "time to first
failed to do so or if they chose not to use the spray. cigarette" was recorded for each subject. For those abstinent at the
Individual follow-ups were scheduled for 2, 3, 6, 9, and 12 end of group treatment this was taken as time to the day on which
months after start of treatment. At these times subjects were they first smoked if they subsequently relapsed. For those who had
contacted by telephone and those reporting abstinence for at least not abstained by the last week of group treatment this interval was
the past week were asked to attend a validation appointment. taken as time to the day on which they first smoked after the start of
Current smokers provided details by telephone only. treatment.

Measures Statistical methods

At assessment subjects provided a smoking history, completed a A sample size in excess of 200 was required to detect as significant
Smoking Motivation Questionnaire (SMQ)," and gave a blood the difference between an abstinence rate of 30% for the active
sample 2 min after smoking a cigarette for measurement of peak group and one of 15% for the placebo group, with 80% power and
plasma nicotine concentrations (referred to in this paper as smoking a =
0-05. These estimates of abstinence were based on the results of
concentrations). Peak post-spray blood samples were taken after 1 a nicotine gum trial conducted in the same clinic.14 To allow for the
and 4 weeks of group treatment and at follow-ups in abstinent effect of having to assign couples or friends to the same spray (to
subjects using the spray. These samples were taken 5 min after a preserve blinding), the minimum sample size was set at 220 since
dose (1 mg nicotine). Nicotine was assayed by use of gas 20% of the sample was expected to fall into this category. If these
chromatography." End expired-air carbon monoxide (CO) was subjects did not behave independently in their response to
measured (EC50 monitor, Bedfont, Technical Instruments, UK) at treatment, there could have been an effective 10% loss in sample
all sessions. Subjects were weighed (indoor clothes, minus shoes) at size.
assessment, after 4 weeks, and at all follow-ups. Amount of spray The difference between active and placebo groups in abstinence
used and any smoking was recorded in a daily diary. rates at the 1-year follow-up was assessed by the X2 test and relapse
Mood disturbance associated with tobacco withdrawal was throughout the year by the log rank test. The influence of each of the
assessed at all sessions by use of self-completion questionnaire, as demographic and pretreatment smoking characteristics on
was craving for cigarettes. The scores for the questionnaire likelihood of abstinence and the interaction of these variables with
items&mdash;depression, irritability, restlessness, and poor spray assignment were examined by use of logistic regression
concentration-were combined to form an index of withdrawal for models. Analysis of variance was used to assess differences between
each subject at each time point. The scores for craving for cigarettes, the two groups in withdrawal symptoms, craving, and weight gain.
difficulty in stopping smoking, time with urges to smoke, and In the case of withdrawal symptoms a baseline (pretreatment)
strength of urges provided an index of craving. At each session measure of mood state was used as a covariate to improve precision.
326
1 111 It:: since 1[1IIUUIIII[1l1UII I ki I IV)
Fig 1-Time to first smoking.
T=group treatment period T=follow-up with
validation.
Software used included GLIM, SPSSPC, and CIA. In view of the
evidence suggesting efficacy of nicotine replacement and of the
results of pilot work with NNS, one-sided probability levels (p)
were used when comparing abstinence rates. In all other
comparisons 2-sided probability levels (2p) were used.
Results
Attendance
Subjects assigned to the active spray attended more group
meetings (mean 4-6, SD 16) than did those assigned to
placebo (mean 4-0, SD 1-8) (t=2.5, 2p=0014). Most
subjects failing to achieve abstinence did not complete the
On average, 96% in the active group and
treatment course.
95% in the placebo group were followed up at each of the
five occasions.
Abstinence
26% of the active treatment group were validated as being
completely abstinent throughout the whole year, compared
with 10% of placebo subjects (X2 9-75, p < 0-002) (table 11).
=
The proportion remaining abstinent over time (fig 1) was
also greater for the active than for the placebo group (log
rank X2 = 18 -4, p < 0&deg;001). Placebo subjects were less likely
to achieve abstinence during group treatment and more
likely to relapse in the first 2 months after the end of group
treatment. The relapse rates during this interval were
28% for the active and 51 % for the placebo group. After 2
months the relapse rates were similar in the two spray
groups.
On the "slips allowed" criterion, 28% of the active group
and 13% of the placebo group were abstinent throughout
the year (relative abstinence rate 2-3, 95% CI 1 -3-36). Only
6 additional subjects (3 active, 3 placebo) were classified as
successful under this criterion-5 had smoked on 1 day (1
cigarette or less) and 1 had smoked on 2 days-which
indicates that few subjects avoided full relapse after an initial
slip. The active spray was equally effective across all 17 sets
of treatment groups.
45 subjects (23 active, 22 placebo) entered the trial with
people they knew and were assigned the same spray, but did
not necessarily attend the same treatment groups, as their
partner or friends. There was no statistical evidence of
similarity between partners or friends, in terms of success at
Fig 2-Relation between treatment (active or placebo spray),
pre-treatment smoking plasma nicotine concentration, and
abstinence at 3 months.
biochemical
Predicted probability of abstinence and 95% confidence intervals based
on a linear logistic model.
stopping smoking. Consequently there was no need to
reduce the effective
sample size or exclude these subjects
from the analyses.
Pretreatment predictors of abstinence
There wasevidence of an interaction between spray
assignment and both pretreatment smoking plasma nicotine
concentration (2p 0&deg;069) and the questionnaire measure of
=
pharmacological dependence (SMQ "addictive" score,
2p=0-08) in predicting which subjects would be
continuously abstinent for the first 3 months. For the
placebo group, the likelihood of abstinence declined sharply
with higher pretreatment smoking plasma nicotine
concentrations, whereas for those on the active spray the
likelihood of abstinence was not affected by their nicotine
intake from cigarettes (fig 2). For pretreatment plasma
nicotine concentrations of 20, 40, and 60 ng/ml, those
treated with the active spray were estimated to be,
respectively, 1-3, 3-0, and 7-9 times more likely than those
treated with the placebo to have been abstinent for the first 3
months. The pattern of the spray by SMQ interaction was
very similar to that for spray by plasma nicotine. None of the
other characteristics in table I interacted with spray
assignment in predicting abstinence; nor was there evidence
that any of these variables predicted abstinence overall,
irrespective of spray assignment.
Use of nasal spray
During the first 2 days of treatment 99% of those assigned
active and 97% of those assigned placebo used the spray,
their average number of daily doses being 9 and 12 doses,
respectively. By the end of group treatment, 87% of active
and 47% of placebo abstainers were still using the spray,
average daily doses being 13 and 7, respectively. At this time
the mean plasma nicotine concentration in subjects still
using the active spray was 16-0 ng/ml, 40% of their
pretreatment concentration of 39-7 ng/ml. None of the
abstinent subjects assigned to placebo continued to use the
spray beyond 6 months but 13 abstainers (43 %) in the active
group were still using the spray at 12 months. Their mean
plasma nicotine concentration at 12 months was 34-9 ng/ml,
79% of their pretreatment smoking level. At 1 month the

Fig 3-Percentage of subjects reporting that the spray relieved
craving for cigarettes "very much" or "quite a bit".
n refers to number of subjects using the spray and completing the
question at each time point.
mean plasma nicotine concentration in these subjects had
been 20-6 ng/ml, 49% of their smoking level.
Tobacco
withdrawal symptoms and craving
Analyses of withdrawal symptoms and craving for
cigarettes were confined to abstinent subjects. Those on the
active spray experienced significantly less withdrawal
discomfort during the first 48 h of treatment than did those
receiving the placebo (F 11 -4,2p 0-001). After a week of
= =
spray use the severity of withdrawal lessened and the
difference between the two spray groups was no longer
significant. Subjects assigned the active spray also
experienced significantly less craving for cigarettes during
the first 48 h (F = 6-7, 2p 0’011) than did the placebo
=
group. The active spray was found to relieve this craving to a
greater extent than did the placebo (fig 3), and it was also
rated as being more helpful than the placebo.
TABLE III-NUMBER OF SUBJECTS WITH EFFECTS ON SYMPTOM
CHECKLISTAT LEAST ONCE
*2p < 0 001
t2p<005
ttn refers to subjects ever having completed checklist
327
Weight gain
Analysis of weight gain was also confined to subjects
abstaining from smoking. Use of the active spray effectively
reduced weight gain associated with stopping smoking.
After 12 months’ abstinence from cigarettes the mean
weight gain was 3-0 kg (SD =4-0) in those still using the
active spray and 5-8 kg (SD=2’9) in those assigned to
placebo (difference=2-8 kg, 95% CI=0-1-5-6 kg). The
mean weight gain in those subjects who had stopped using
the active spray was 5-5 kg (SD 3-3). Men and women did
=
not differ in weight gained and the moderating effect of the
active spray was equally evident in both sexes.
Side-effects
NNS was generally well tolerated and only 2 subjects (1
active after 10 months, 1 placebo after 10 weeks) were
advised to discontinue NNS use because of adverse
reactions. The first had had occasional soreness of the nasal
mucosa and blocked nose and was found on examination to
have some pre-existing obstruction to airflow through both
nostrils. The second reported sore eyes and blood spotting
and tenderness in the nostrils, although only tenderness was
evident on examination. Nearly all subjects reported having
one or more side-effects at least once during the first month
(table III). The commonest symptom was local transient
irritation and was more frequent with the active spray. Some
of the systemic symptoms may have been withdrawal effects
(eg, headache, light-headedness, dizziness). The frequency
of reporting most of these symptoms decreased over time for
both groups. Of the three positive subjective effects-
alertness, calmness, and feeling good or "high"-only the
third showed a statistical difference between the active and
placebo spray groups.
Of the other effects, reported by subjects or diagnosed on
examination, the most troublesome were a sore area in the
nostril (9% active, 7% placebo), blocked nose (7% active,
7% placebo), nasal blood spotting (6% active, 6% placebo),
minor epistaxis (6% active, 5% placebo), nasal ulceration
(3% active, 3% placebo), and vomiting (2% active). Most of
these occurred early during group treatment although half
the occurrences of sore area in nostril, minor nose bleed, and
ulceration were first seen in long-term NNS users after 3
months. Subjects with these symptoms were advised to use
the spray sparingly and to "rest" the affected nostril until
symptoms disappeared. Abrasion due to careless insertion of
the spray nozzle was thought to have accounted for the nasal
blood spotting in most cases. There was no evidence from
ear-nose-throat examinations of any long-term adverse
effects attributable to the spray.
Despite the frequency of irritant effects, only 6 (5%)
subjects on active spray and 8 (8%) on placebo rated
side-effects as unacceptable. 4 of these subjects (2 active, 2
placebo) chose to continue using the spray, and
subsequently gave the effects better ratings.
Discussion
Treatment with the active spray produced substantially
better results than did the placebo, with more than double
the success rate at 1 year. The extent of withdrawal and
craving was less with nicotine than with placebo in the first
few days, and a higher proportion of subjects assigned the
active spray reported that it relieved craving and that it had
been helpful in enabling them to stop smoking. The active
spray had a greater effect on relief of craving and in helping
subjects to stop smoking than on general measures of
328
withdrawal and craving. This difference suggests that it may
not only be the degree of withdrawal experienced but also
the ability to relieve these symptoms, that contribute to
success (or failure). There was some evidence that the
sensory impact of the droplets and the action of spraying in
itself were perceived as helpful since even among placebo
subjects 25 % reported that craving was relieved by the spray
in the first 2 weeks.
Despite its initial irritant effect, a high proportion of
subjects were able to use the spray adequately and obtain a
substantial degree of systemic nicotine replacement. Of
those assigned the active spray, 90% of those abstinent at 4
weeks were still using it at this time. After 2 months nicotine
concentrations were about half previous smoking levels and
after 6 months about three-quarters of smoking levels. The
ease with which nicotine concentrations approaching
smoking concentrations were achieved, the rapid rate of
nicotine absorption, and the ability to relieve craving are
likely to have accounted for the efficacy of NNS.
The finding that abstinent active spray users gained less
weight than placebo successes or active spray successes who
had discontinued spray use is consistent with results of some
nicotine gum studies15 and has important implications, since
weight gain may be related to relapse.16 The weight gain in
abstinent non-users of the spray was considerably higher
than the average of 2-3 kg reported in a recent review.17 It is
not clear whether this discrepancy is due to the fact that valid
measures of both abstinence and weight have not always
been used in other studies or to the particular characteristics
of our subject population.
Both the active and placebo sprays were initially
associated with several irritant side-effects which, although
common, tended to be transient and of a similar order to
those encountered by subjects getting used to nicotine gum.
Most subjects developed tolerance to these effects and very
few found them unacceptable. There was evidence from the
reporting of adverse symptoms that the active spray was
more liable than the placebo to produce these effects. This
difference was unexpected since preliminary work had
revealed no differences in irritancy. It is unlikely to have
resulted in unblinding since side-effects were recorded by
questionnaire and not discussed openly at group meetings.
Only a rough guide to the relative effectiveness of NNS
compared with other nicotine replacement therapies can be
obtained from the findings reported here. Although the
two-and-a-half-fold advantage of active substance over
placebo has been found in some nicotine gum and patch
trials, it is above the average figures reported.18 We applied
stringent criteria for success, which has not always been the
case in other studies. The fact that active and placebo
subjects were treated in the same clinic groups may have
served to attenuate the difference in outcome between the
two sprays. The therapeutic effect of group treatment is
likely to have been greater for placebo subjects than for those
on active spray, since the former were likely to have been
helped by the presence of subjects succeeding (with the help
of the active spray), whereas subjects on the active spray
were likely to have been discouraged by the proximity of
placebo subjects struggling but failing to abstain. Groups in
which all subjects receive an active spray might be expected
to achieve higher absolute success rates.
That the active spray was of greater benefit to the more
dependent smokers with higher smoking blood nicotine
levels is particularly interesting. No nicotine patch studies
have reported an interaction between treatment (active/
placebo) and dependence. Only a few of the nicotine gum
studies have reported such a finding.19,20 Our finding does
not imply that NNS would benefit only the most highly
dependent but it suggests that NNS could be an additonal
means of helping highly dependent smokers, who would
otherwise be less likely than the less dependent to succeed.
NNS is likely to have greater dependence potential than
nicotine gum or patch, since its pharmacokinetic profile
more closely approximates that of cigarettes. In addition to
its rapid rate of nicotine absorption, the spray has three other
characterisics thought to influence dependence potential-
namely, ease of producing high nicotine levels, high rates of
use, and few side-effects (after acquisition of tolerance).21
43% of active spray successes used the spray for the whole
year; they represent 11 % of those assigned active spray. The
corresponding figures from our clinic for long-term gum use
are 25% and 6-3%, respectively.15
Our study was designed to examine the efficacy of NNS
as an aid to stopping smoking when spray use was
unrestricted. The extent to which the therapeutic advantage
of the spray might be reduced by limiting duration of use
needs to be addressed. It is probable that many highly
dependent smokers will do better with long periods of
nicotine replacement. The small degree of risk of long-term
nicotine replacement therapy has to be weighed against the
substantial health benefits of stopping smoking. Whether
our findings can be generalised to treatment in other settings
is unknown. The clinical position of NNS, with respect to
the other nicotine replacement therapies, needs to be
clarified, but our results suggest that NNS may be the
preferred treatment for the more heavily dependent
smokers.
We thank the Medical Research Council and the Imperial Cancer Research
fund for financial support; Mr J. N. Thomas, for the clinical examinations;
Mrs Mary Hayward and Mr Jonathan Foulds for assisting with follow-ups;
Kabi Pharmacia Therapeutics AB for supplying the nasal sprays, and their
representative, Dr Mikael Franzon.
REFERENCES
1. Abelin T, Buehler A, Muller B, Vesanen K, Imhof PR. Controlled trial of
transdermal nicotine patch in tobacco withdrawal. Lancet 1989; i: 7-10.
2. Tonnesen P, Norregaard J, Simonsen K, Sawe U. A double-blind trial of
a 16-h transdermal nicotine patch in smoking cessation. N Engl J Med
1991; 325: 311-15.
3. Transdermal Nicotine Study Group. Transdermal nicotine for smoking
cessation: six-month results from two multicenter controlled clinical
trials. JAMA 1991; 266: 3133-38.
4. Russell MAH, Jarvis MJ, Feyerabend C. A new age for snuff. Lancet
1980; i: 474-75.
5. Russell MAH. Nicotine replacement: the role of blood nicotine levels,
their rate of change, and nicotine tolerance. In: Pomerleau OF,
Pomerleau CS, eds. Nicotine replacement: a critical evaluation. New
York: Alan R Liss, 1988: 187-217.
6. Sutherland G, Russell MAH, Stapleton J, Feyerabend C, Ferno O. Nasal
nicotine spray: a rapid nicotine delivery system. Psychopharmacol (in
press).
7. Jarvis MJ, Hajek P, Russell MAH, West RJ, Feyerabend C. Nasal
nicotine solution as an aid to cigarette withdrawal: a pilot clinical trial.
Br J Addict 1987; 82: 983-88.
8. Benowitz NL, Jacob P. Daily intake of nicotine during cigarette smoking.
Clin Pharmacol Ther 1984; 35: 499-504.
9. Hajek P,Belcher M, Stapleton J. Enhancing the impact of groups: an
evaluation of two group formats for smokers. Br J Clin Psychol 1985;
24: 289-94.
10. Hajek P. Withdrawal-oriented therapy for smokers. Br J Addict 1989; 84:
591-98.
11. West R, Russell MAH. Pre-abstinence smoke intake and smoking
motivation as predictors of severity of cigarette withdrawal symptoms.
Psychopharmacol 1985; 87: 334-36.
12. Feyerabend C, Russell MAH. A rapid gas-liquid chromatographic
method for the determination of cotinine and nicotine in biological
fluids. J Pharm Pharmacol 1990; 42: 450-52.

13. Bigelow G, Ossip-Klein DJ, eds. Classification and assessment of
smoking behaviour. Health Psychol 1986; 5 (suppl): 3-11.
14. Jarvis MJ, Raw M, Russell MAH, Feyerabend C. Randomised
controlled trial of nicotine chewing-gum. Br Med J 1982; 285: 537-40.
15. Hajek P, Jackson P, Belcher M. Long-term use of nicotine chewing gum
occurrence, determinants, and effect on weight gain. JAMA 1988; 260:
1593-96.
16. Klesges RC, Klesges LM. Cigarette smoking as a dietary strategy in a
university population. Int J Eating Disorders 1988; 7: 413-19.
17. US Department of Health and Human Services. The health benefits of
smoking cessation: a report of the Surgeon General. Washington, DC:
US Department of Health and Human Services, 1990.
Randomised study of two doses of cisplatin with
cyclophosphamide in epithelial ovarian cancer
Cisplatin is generally accepted to be the most
active cytotoxic agent for the treatment of ovarian
cancer but the optimum dose remains unclear. We
have performed a randomised trial to assess the
importance of cisplatin dose in the treatment of
advanced epithelial ovarian cancer. Patients were
randomly assigned treatment with 50 mg/m2 (low
dose) or 100 mg/m2 (high dose) cisplatin plus 750
mg/m2 cyclophosphamide, for a maximum of six
cycles with intervals of 3 weeks.
We planned to recruit 300 patients, but an interim
analysis on the first 165 indicated a highly significant
survival difference (p=0&middot;0008). Recruitment was
therefore stopped and the trial patients were
followed-up for 12 months longer. The relative
progression rate (high-dose/low-dose) after 12
months’ extra follow-up was 0&middot;55 (95% confidence
interval 0&middot;37-0&middot;81, p=0&middot;003) and the relative death
rate 0&middot;53 (0&middot;34-0&middot;81, p=0&middot;003). Overall median
survival was 69 weeks in the low-dose group and
114 weeks in the high-dose group. Residual disease
extent before chemotherapy had an important
influence&mdash;patients with lesions of less than 2 cm did
best; if given high-dose cisplatin their median
survival was 3 years. 56 low-dose and 45 high-dose
patients completed six cycles of chemotherapy; 15
and 9 patients, respectively, were withdrawn early
because of progressive disease and treatment was
stopped in 6 and 25, respectively, because of
unacceptable side-effects or patient refusal. Toxic
effects were significantly greater in the high-dose
group, especially those on the nervous system and
ears, alopecia, vomiting, and anaemia.
Although the higher dose of cisplatin clearly leads
to better results in terms of survival, its overall clinical
benefit in the management of ovarian cancer will
depend on further improvements in measures to
alleviate toxic effects.
329
18. Lam L, Sze PC, Sacks HS, Chalmers TC. Meta-analysis of randomized
controlled trials of nicotine chewing gum. Lancet 1987, ii: 27-29.
19. Jackson PH, Stapleton JA, Russell MAH, Merriman RJ. Predictors of
outcome in a General Practitioner intervention against smoking.
Prevent Med 1986; 15: 244-53.
20. Fagerstrom KO. Efficacy of nicotine chewing gum: a review. In:
Pomerleau OF, Pomerleau CS, eds. Nicotine replacement: a critical
evaluation. New York: Alan R Liss, 1988: 109-28.
21. Hughes JR. Dependence potential and abuse liability of nicotine
replacement therapies. In: Pomerleau OF, Pomerleau CS, eds.
Nicotine replacement: a critical evaluation. New York: Alan R Liss,
1988: 261-77.
Introduction
There is a general consensus that the most effective
cytotoxic agent for the management of ovarian cancer is
cisplatin. In direct single-agent comparisons with alkylating
agents cisplatin is superior,l and the lack of obvious survival
advantage in studies that have compared regimens with and
without cisplatin may well reflect benefits of cisplatin given
as second-line therapy.2 The cisplatin analogue carboplatin
has lower gastrointestinal, renal, and nervous-system
toxicity. Randomised trials have suggested equivalence with
cisplatin,3and the drug is increasingly used in the
management of ovarian cancer, since it is better tolerated
than cisplatin when given in full dose. However, long-term
survival data for carboplatin are not yet available. Thus, in
many centres, standard therapy for ovarian cancer continues
to be a combination of cyclophosphamide and cisplatin.
This two-drug combination is as effective as a more complex
four-drug combination.4 Retrospective studies have
suggested the importance of dosage of cisplatin for
treatment outcome.5 However, the best dose of cisplatin has
not been established by randomised study. High doses of
cisplatin (100 mg/m2 or more) are associated with a higher
rate of side-effects, but lower doses (50 mg/m2) are generally
well tolerated. It is still not clear whether this apparent
difference in toxicity is accompanied by significant
differences in efficacy.
We have conducted a randomised prospective study to
compare therapeutic efficacy, in terms of progression-free
and overall survival, of 50 and 100 mg/m2 cisplatin given
with cyclophosphamide to patients with ovarian cancer.
Differences in toxicity were also sought. The trial was closed
ADDRESSES Western Infirmary, Royal Infirmary, Stobhill
General Hospital, Glasgow (Prof S. B. Kaye, FRCP, C. R. Lewis,
FRACP, J Paul, BSc, M. Soukop, FRCP, E. M. Rankin, MRCP, J. Cassidy,
MRCP, J. A. Davis, MB, N. S. Reed, FRCP, A. MacLean, MRCOG, J. H.
Kennedy, MRCOG, R. P. Symonds, FRCR); Ninewells Hospital,
Dundee (I D. Duncan, FRCOG, H. K. Gordon, MRCOG); Aberdeen
Royal Infirmary, Aberdeen (H. C. Kitchener, MRCOG, D. J.
Cruickshank, MRCOG, G. A. Swapp, FRCOG, T. K. Sarkar, FRCR); Belfast
City Hospital, Belfast (R. J. Atkinson, FRCOG); and Airedale
Hospital, West Yorkshire, UK (S. M. Crawford, FRCP).
Correspondence to Prof S. B. Kaye, FRCP, Beatson Oncology Centre,
Western Infirmary, Glasgow G11 6NT, UK.