You are on page 1of 3

Subhypnotic Doses of Propofol Possess Direct

Antiemetic Properties
Alain Borgeat, MD, Oliver H. G. Wilder-Smith, MD, Michele Saiah, MD, and
Kaplan Rifat, MD
Department of Anaesthesiology, University of Geneva, Geneva, Switzerland

Propofol is associated with a low incidence of post- vs 35% success rate; P < 0.05). Patients successfully
operative nausea and vomiting. In a prospective, treated had a similar incidence of relapse (propofol
randomized, double-blind, placebo-controlled study, 28%; placebo 22%) within the first 30 min after
we investigated the possible direct antiemetic prop- therapy. Thirty-three percent of the propofol-treated
erties of a subhypnotic dose of propofol. Fifty-two patients and 44% of the placebo-treated patients
ASA physical status I or I1 patients, aged 15-60 yr showed a minor increase in sedation. The level of
with nausea and vomiting after minor gynecologic, postoperative pain did not change in either group.
orthopedic, or digestive tract surgery, were included Hemodynamic values remained unchanged in both
in the study and received either propofol (10 mg = groups. Pain on injection (7.6%)or dizziness (3.6%)
1 mL) or placebo (1 mL Intralipid) intravenously in only occurred in the propofol group. We conclude
the postanesthesia care unit. Patients treated with that propofol has significant direct antiemetic prop-
propofol experienced a larger reduction in nausea erties.
and vomiting than patients treated with placebo (81% (Anesth Analg 1992;74:539-41)

N ausea and vomiting are distressing symp-

toms after anesthesia, resulting in significant
morbidity an d longer stays in the recovery
room. General anesthesia conducted with propofol is
associated with less nausea an d vomiting during the
inclusion in the study. Patients were enrolled into the
study postoperatively if they exhibited vomiting of
grade 4 or 5 severity as assessed by one of the
investigators. Institutional and ethical committee ap-
proval was obtained for the study.
early postoperative period than any other anesthetic One hour preoperativeIy, midazolam 0.1 mg/kg
technique (1-3). Some of these studies have sug- IM, was given. Anesthesia was induced with thio-
gested that propofol may have direct antiemetic prop- pental 4 mg/kg IV and maintained with fentanyl,
erties (3). We therefore undertook a prospective, enflurane, and vecuronium. No patient had received
randomized, double-blind, placebo-controlled study neostigmine to reverse neuromuscular blockade.
to investigate the possible antiemetic properties of Medication was blinded and randomized by our
propofol in the early postoperative period. pharmacy, which delivered coded vials of either
propofol or Intralipid. Recovery room patients com-
plaining of nausea and/or vomiting were initially
Methods assessed by one of the investigators a s to the severity
of nausea and vomiting by means of a five-point
We studied 52 ASA physical status I or I1 patients of
rating scale (Table 1). Patients with a score of 4 or 5
both sexes between the ages of 15 an d 60 yr, weigh-
received 1 mL of the treatment drug (propofol =
ing between 50 an d 80 kg, with nausea an d vomiting
10 mg or placebo) intravenously. Sixty seconds later,
after minor elective gynecologic, digestive tract, or
patients were reevaluated with the rating scale. A
orthopedic surgery. Patients with a known allergy to
score of 1 or 2 was considered a success; patients with
propofol or a history of epilepsy were excluded from
a score greater than 2 were given a second milliliter of
the study. All patients gave informed consent before
the treatment drug. Patients with a score greater than
2 60 s after the second dose were documented as
Accepted for publication December 10, 1991. definitive treatment failures. An observation period
Address correspondence to Dr. Borgeat, Department of Anaes-
thesiology, University Hospital of Geneva, 1211 Geneve 4, Swit- of 30 min after treatment ensued, during which time
zerland. relapses of nausea and vomiting together with their

01992 by the International Anesthesia Research Society

0003-2999/921$5.00 Anesth Analg 1992,74.53941 539

Table 1. Nausea and Vomiting Rating Scale Table 4. Treatment Results

1= N o nausea and vomiting Propofol (70) Placebo (70)
2= Residual nausea without vomiting
I1 26 26
3- Minor but persistent vomiting with nausea
Success after first injection 18/26 (69) 9126 (35)
4= Maior nausea with vomiting Additional successes after 3126 (12) -
5 = 5evere nausea with vomiting
- second injection
Relapses within 30 min 6121 (29) 219 (22)

Table 2. Sedation Rating Scale

1 = Patient fully awake successfullv ( P < 0.05). Of these, six patients in the
2 = Patient somnolent, response to call
propofol gioup and two patients in the placebo group
3 = Patient somnolent, no response to call, response to tactile
stimulation relapsed within the next 30 min. In the propofol
4 = Patient asleep, response to painful stimulation group, three patients (3126 = 12%) needed a second
administration of the study drug to achieve a success-
ful outcome. (None required a second administration
Table 3. Characteristics of Patients in the placebo group; Table 4.) The sedation scale
increased bv one point in 33% of the patients success-
Propofol Placebo fully treated with propofol and in 44% of patients
group group
successfully treated with placebo (NS). No change in
I1 26 26 sedation was noted for failures in both treatment
Age (L r) 42 ? 4 36 5 3 groups. N o patient showed a change in postoperative
5ex (M b) 917 i21 pain score.
Operating time (min) 98 i 10 89 2 13
5urgen t\ pe (GsO D) 1 52,9 17 I 7
We observed pain on injection in 7.6% of propofoi
_-___ - treatments, none in the placebo group (NS). Dizzi-
G gvne~olc~gic 0 orthopedic, D, digestn e \urger\ ness occurred in one patient (3.6%) in the propofol
Ldlue\ for age and operating time are mean 2
group onlv (NS). No changes in mood and/or hallu-
cinations were reported. Hernodynamic changes
were similar in both groups and nonsignificant.
rdting scale were recorded. Relapse was defined as a
score of more than 2. No drug treatment other than
aspirin was given during the study period; the 0.5 g Discussion
of aspirin was given intravenously for postoperative
This is the first study to have investigated the direct
pain. Other variables documented in the course of
antiemetic properties of propofol. Previous studies
the study were sedation (four-point rating scale,
on this topic (4-6) have noted that propofol anesthe-
Table 2) and postoperative pain (10-point verbal rat-
sia is accompanied by significantly less postoperative
ing scale from 0 = no pain to 10 = worst pain
nausea and vomiting. Our study showed that propo-
possible) after each drug administration, the presence
fol exhibits significant direct antiemetic properties
of pain on injection (verbal complaints), dizziness,
compared with placebo. Recently, Oestman et al. (7)
mood change, and the presence of hallucinations.
found that Intralipid (the soya-oil emulsion in which
Statistical analysis was performed using non-
propofol is formulated) possesses no significant anti-
parametric methods with the Mann-Whitney and
emetic effects. Thus Intralipid is a valid placebo for
Fishers exact tests. We compared demographic data
propofol, particularly in the context of emesis inves-
for the groups using Students t-test. A P < 0.05 was
tigations. The placebo success rate in this study is
considered significant.
also in accordance with the rate quoted in the litera-
ture (8). Relapse rates within the first 30 min of
Iiesults treatment were similar between the groups; we have
no explanation for this phenomenon. In the course of
Both groups were comparable with regard to demo- pilot investigations for the present study, we found
graphic data, type and dose of anesthetics (including that propofol has a very rapid onset of action in
fentdnyl), and type and duration of surgical proce- treating nausea and vomiting, with no or little further
dure (Tdble 3). In the propofol group, 73% had grade improvement in nausea or vomiting 60 s after appli-
5 nausea and vomiting; in the placebo group, 81% cation. We therefore considered this period to be an
showed grade 5 nausea and vomiting (NS). Twenty- adequate interval before injecting a second dose.
one of 26 (81%)patients in the propofol group and 9 Interestingly, there was no significant difference in
of 26 (35%)patients in the placebo group were treated sedation change between groups for successfully

treated patients; there was no change for the failures. Both these factors make hypnosis an unlikely expla-
This strongly suggests that propofol was truly sub- nation for propofols antiemetic properties. Propofol
hypnotic in the doses administered (10 mg) in this has a profile of central nervous system depression
study. Vigilance is obviously strongly influenced by that differs significantly from other anesthetic drugs
the act of vomiting, hence patients vomiting are more (13). In contrast to thiopental, for example, propofol
awake than those not vomiting. However, the design uniformly depresses central nervous system struc-
of our study cannot categorically exclude more sub- tures, including subcortical centers (13). Most drugs
tle, subclinical differences in sedation between the of known antiemetic efficacy exert their therapeutic
groups. effects via subcortical structures (14,15). We therefore
Propofol causes pain on injection (9), particularly suggest that propofol exerts its antiemetic action by
for small catheters or veins (10). Pain on injection was the modulation of some subcortical pathways. Fur-
slight and only occurred in the propofol group. The ther studies are needed to elucidate the precise mech-
occurrence of pain did not unblind the investigator as anisms involved in the antiemetic properties of
it is not clear or proven that when comparing propo- propofol as well as to optimize the dosage regimen.
fol and Intralipid, only the injection of propofol In conclusion, this study demonstrated that propo-
causes pain. Moreover, this side effect was only fol in subhypnotic doses possesses direct antiemetic
expressed after the end of the emetic episode. Dizzi- properties in the context of minor elective surgery. At
ness occurred only once in the propofol group (no the dose administered (10 mg), side effects were rare
significant difference between groups). and minor.
Both groups were well-matched for factors (e.g.,
sex, duration and type of surgery, anesthetic tech- References
nique) known to affect the incidence and severity of
1. Doze V, Shafer A, White P. Propofol-nitrous oxide versus
postoperative nausea and vomiting. Both treatments thiopental-isoflurane-nitrous oxide for general anesthesia. An-
were identical in appearance to minimize psycholog- esthesiology 1988;69:63-71.
ical effects, which are known to play an important 2. Korttila K, Oestman PL, Faure E, et al. Randomized compari-
son of recovery after propofol-nitrous oxide versus thiopental-
role in the etiology of nausea and vomiting. Pain on isoflurane-nitrous oxide anaesthesia in patients undergoing
injection, when present, was slight, and is therefore ambulatory surgery. Acta Anaesthesiol Scand 1990;34:400-3.
unlikely to have affected treatment outcome. Thus, it 3. McCollum JSC, Milligan KR, Dundee J W , The antiemetic
is likely that the differences observed were due to the action of propofol. Anaesthesia 1988;43:23940.
4. Gunawardene RD, White DC. Propofol and emesis. Anaesthe-
treatment. It is of interest that 80% of the failures in sia 1988;43(Suppl):6!%7.
the propofol group were related to a type of surgery 5. McCollum JSC, Milligan KR, Dundee J W . Has propofol an
associated with vagal stimulation (digestive sur- antiemetic action? Br J Anaesth 1987;59:654P-W.
gery)-a relationship not present in the placebo 6. Borgeat A, Popovic V, Meier D, Schwander D. Comparison of
propofol and thiopentalihalothane for short-duration ENT
group. surgical procedures in children. Anesth Analg 1990;71:511-5.
In our pilot study, a dose of 10 mg of propofol 7. Oestman PL, Faure E, Glosten B, Kemen M, Robert MK,
proved effective without side effects. Our initial study Bedwell S. Is the antiemetic effect of the emulsion formulation
of propofol due to the lipid emulsion? Anesth Analg 1990;71:
was not designed to investigate the effects of varying 536-40.
dosages. PropofoI 10 mg was effective for our patients 8. Beecher HK. The powerful placebo. JAMA 1955;159:16024.
who weighed between 50 and 80 kg. We cannot 9. Nightingdale P, Healy TEJ, Hargreaves J , McGuiness K, Kay B.
recommend dosages for patients outside this weight Propofol in emulsion form: induction characteristics and
venous sequelae. Eur J Anaesthesiol 1985;2:361-8.
range; however, a larger dosage of propofol may be 10. Kawar P, Dundee JW. Frequency of pain on injection and
associated with undesirable effects. venous sequelae following the IV administration of certain
Perhaps hypnotic agents exert their antiemetic anaesthetics and sedatives. Br J Anaesth 1982;54:935-9.
11. Moyer JH. Effective antiemetic agents. Med Clin N Am 1957;
action primarily via sedation (11). In our study we 41:40532.
observed no significant sedative effects owing to 12. Kay NH, Sear JW, Uppington J, Cockshott ID, Douglas EJ.
propofol. Additionally, the duration of antiemetic Disposition of propofol in patients undergoing surgery. A
action in our study was much longer than the normal comparison in men and women. Br J Anaesth 1986;58:10759.
13. Cavazzuti M, Porro CA, Barbieri A, Galetti A. Brain and spinal
duration of hypnotic action (5-7 min 1121) achieved cord metabolic activity during propofol anaesthesia. Br J
with the 15-fold higher doses of propofol used for Anaesth 1991;66:490-5.
anesthetic induction. This is also seen in the pro- 14. Lerman J, Eustis S, Smith DR. Effect of droperidol pretreat-
ment on post anesthetic vomiting in children undergoing
longed decrease in nausea and vomiting after propo- strabismus surgerv. Anesthesiology 1986;65:322-5.
fol anesthesia, still evident for a long time after any 15. Cohen SE, Woods WA, Wyner 1. Antiemetic efficacy of dro-
anesthetic or hypnotic effect of propofol has worn off. peridol and metoclopramide. Anesthesiology 1984;60:67.