prevention of vomiting after tonsillectomy in children: granisetron versus ramosetron

The LaryngoscopeLippincott Williams & Wilkins, Inc., Philadelphia© 2001 The American Laryngological,Rhinological and Otological Society, Inc.

Prevention of Vomiting After Tonsillectomyin Children: Granisetron VersusRamosetron

Yoshitaka Fujii, MD; Yuhji Saitoh, MD; Noriaki Kobayashi, MD

Objective/Hypothesis: Granisetron, a selective5-hydroxytryptamine type 3 receptor antagonist, iseffective for the prevention of vomiting after tonsil-lectomy in children. Ramosetron (Nasea; Yamanou-chi; Tokyo, Japan), another new antagonist of5-hydroxytryptamione type 3 receptor, has more po-tent and longer-acting properties than granisetron(Kytril; Smith Kline Beecham, London, UK) againstcisplatin-induced emesis. This study was undertakento compare the efficacy and safety of granisetron andramosetron for the prevention of vomiting after pedi-atric tonsillectomy. Study Design: Prospective, ran-domized, double-blinded study. Methods: Ninety pedi-atric patients, aged 4 to 10 years, receivedintravenously granisetron 40 mg/kg or ramosetron 6mg/kg (n 5 45 each) at the end of surgery. The samestandard general anesthetic technique and postoper-ative analgesia were used throughout. Emetic epi-sodes and safety assessment were performed duringthe first 24-hour period and the next 24-hour periodafter anesthesia. Results: The rates of patients beingemesis-free during the period from 0 to 24 hours afteranesthesia were 89% with granisetron and 93% withramosetron, respectively (P 5 .357); the correspond-ing rates during the period from 24 to 48 hours afteranesthesia were 71% and 93%, respectively (P 5 .006).No clinically serious adverse events attributable tothe study drugs were observed in any of the groups.Conclusion: Ramosetron is a better antiemetic thangranisetron for the long-term prevention of postoper-ative vomiting in children undergoing general anes-thesia for tonsillectomy. Key Words: Tonsillectomy,nausea, vomiting, antiemetics, ramosetron,granisetron.

Laryngoscope, 111:255–258, 2001

INTRODUCTIONPostoperative vomiting is a commonly observed ad-

verse event after tonsillectomy with or without adenoid-

ectomy in children.1,2 Most of currently used antiemetics(antihistamines, butyrophenones, dopamine receptor an-tagonists) have been reported to occasionally cause unde-sirable adverse effects, such as excessive sedation, hypo-tension, dry mouth, dysphoria, hallucinations, andextrapyramidal symptoms.3 Granisetron, as well as on-dansetron, is a selective 5-hydroxytryptamine type 3 (5-HT3) receptor antagonist and is effective for the treat-ment of emesis in patients receiving cytotoxic drugs.4

Granisetron reduces the incidence of vomiting after pedi-atric tonsillectomy with or without adenoidectomy.5,6

Ramosetron, (R)-5-[(1-methyl-3-indolyl)carbonyl]-4,5,6,7-tetrohydro-1H-benzimidazol hydrochloride (Nasea, Ya-manouchi, Tokyo, Japan), is another new 5-HT3 receptorantagonist and has more potent and longer-acting activityagainst cisplatin-induced emesis than granisetron.7 Wehave recently demonstrated that prophylactic therapywith ramosetron is more effective than granisetron for theprevention of postoperative nausea and vomiting within a48-hour period after anesthesia in patients undergoingmiddle ear surgery.8 However, there have been no reportscomparing the antiemetic efficacy of ramosetron withgranisetron in children at high risk of vomiting after ton-sillectomy. We designed this prospective, randomized,double-blinded study to compare the efficacy and safety ofgranisetron and ramosetron for the prevention of vomitingafter pediatric tonsillectomy.

MATERIALS AND METHODSAfter approval by our institutional ethics committee and

informed parental consent, we studied 90 American Society ofAnesthesiologists physical status I children, aged 4 to 10 years,undergoing general anesthesia for tonsillectomy with or withoutadenoidectomy. Patients who had experienced postoperative vom-iting, those who had taken an antiemetic medication within 24hours before surgery, and those who had a history of motionsickness were excluded from the study. Patients were not allowedto have solid food after midnight before surgery. Clear liquidswere permitted up to 3 hours before surgery.

Patients were randomly assigned to receive intravenously(IV) 40 mg/kg granisetron or 6 mg/kg ramosetron (n 5 45 each) atthe end of surgery. A randomization list was generated, andidentical syringes containing each drug were prepared by person-nel not involved in this study, according to the list. The dose of

From the Departments of Anesthesiology (Y.F., Y.S.) and Otolaryngol-ogy (N.K.), Toride Kyodo General Hospital, Tosride City, Ibaraki, Japan.

Editor’s Note: This Manuscript was accepted for publication Novem-ber 2, 2000.

Send Correspondence to Yoshitaka Fujii, MD, Department of Anes-thesiology, University of Tsukuba Institute of Clinical Medicine, 2-1-1,Amakubo, Tsukuba City, Ibaraki 305, Japan.

Laryngoscope 111: February 2001 Fujii et al.: Antiemetics in Tonsillectomy

255

granisetron used in this study was chosen based on our previousstudies.5,6 No data were available regarding the dose of ramose-tron to be used in children, but the single dose of ramosetronused in this study was extrapolated from an investigation inadults.8

No preanesthetic medications were administered. Anesthe-sia was induced by increasing concentration of sevoflurane in67% nitrous oxide (N2O) and oxygen (O2) via mask. After aninhalation induction of anesthesia, 0.01 mg/kg atropine IV wasadministered, followed by 1 mg/kg fentanyl IV, and tracheal in-tubation was facilitated with 0.1 mg/kg vecuronium IV. Aftertracheal intubation, anesthesia was maintained with N2O/O2

(2:1) and sevoflurane 0.5% to 3.0% (inspired concentration). Ven-tilation was mechanically controlled and adjusted to keep an endtidal concentration of CO2 between 35 and 40 mm Hg throughoutsurgery, as measured by an anesthetic/respiratory gas analyzer(Ultima, Datex, Helsinki, Finland). Neuromuscular blockade wasachieved with vecuronium and was reversed by a combination of0.02 mg/kg atropine IV and 0.04 mg/kg neostigmine IV at thecompletion of surgical procedure, and the trachea was extubatedwhen the patient was awake. Rectal temperature was monitoredand maintained at 37°C 6 1°C using a warming pad. Postopera-tively, all patients were admitted to the hospital and remained fora couple of days. Clear liquids were given only if the patientrequested them, and other oral intake was not allowed for 4 hoursafter recovery from anesthesia. Postoperative analgesia was pro-vided by 15 to 20 mg/kg acetaminophen rectally for mild pain and0.3 mg/kg pentazocine IV for severe pain.

Postoperatively, all episodes of emetic symptoms (retching,vomiting) during the first 24 hours (0–24 h) and the next 24 hours(24–48 h) after anesthesia were recorded by nursing staff who didnot know which treatment each patient had received. Thesenurses observed the patients at various intervals according to thenormal ward routine. Vomiting was defined as the forceful expul-sion of gastric contents from the mouth; retching was defined asthe labored, spasmodic, rhythmic contraction of the respiratorytract muscles, including the diaphragm, chest wall, and abdomi-nal wall muscles, without the expulsion of gastric contents.3

Nausea was not assessed as a separate entity in this studybecause of the young age of the patients. At the end of everyobservation period, the nurses asked the parents about theirchildren’s postanesthetic condition and problems (i.e., adverseevents attributable to the study drugs).

Patient demographic data were analyzed by ANOVA withBonferroni’s correction for multiple comparison and x2 test. Therate of patients who were emesis-free, had retching, or had vom-iting and the incidence of adverse events were compared withFisher’s exact probability test. The number needed to treat(NTT), a useful estimate of clinical relevance of treatment effect,was calculated.9 A P value of less than .05 was considered signif-icant. Values were expressed as mean (SD) or number (%). Poweranalysis was used to determine the number of patients in thestudy based on the assumptions that 1) an emesis-free episode(which was regarded as the primary end point) in a patientreceiving granisetron would be 70%, 2) an improvement from 70%to 95% was considered of clinical importance, and 3) a 5 0.05 witha power (1-b) of 0.8.

RESULTSThe treatment groups were comparable with respect

to patient demographics and types of surgery (Table I).The rates of patients being emesis-free 0 to 24 hours afteranesthesia were 89% with granisetron and 91% withramosetron, respectively; the corresponding rates for theperiod from 24 to 48 hours after anesthesia were 71% and93%. Thus, the rate of patients experiencing an emesis-free episode during 24 to 48 hours after anesthesia wasgreater in patients who had received ramosetron than inthose who had received granisetron (P ,.05) (Table II).The NTT was estimated to be 4.5 during the period from24 to 48 hours after anesthesia to prevent postoperativeemesis by administering ramosetron compared with gran-isetron. Thus, ramosetron showed a more favorable effecton an emesis-free episode than granisetron within thenext 24-hour period after anesthesia. Clinically seriousadverse events attributable to the study drugs were notobserved in any of the groups.

DISCUSSIONThe reported incidence of vomiting after tonsillec-

tomy with or without adenoidectomy in children variesfrom 62% to 73% when no prophylactic antiemetic is giv-en.1,2 This incidence justifies the use of prophylactic anti-emetics after pediatric tonsillectomy. The etiology of post-operative emesis in children undergoing generalanesthesia for tonsillectomy remains unclear but is prob-ably multifactorial.3 A number of factors, including age,

TABLE I.Patient Demographic Data.

Granisetron(n 5 45)

Ramosetron(n 5 45)

Age (y) 6.4 (2.4) 6.6 (2.3)

Sex (male/female) 22/23 21/24

Height (cm) 117.2 (10.0) 120.4 (12.2)

Weight (kg) 23.2 (6.2) 23.9 (5.7)

Duration of operation (min) 40 (16) 44 (17)

Duration of anesthesia (min) 60 (19) 64 (18)

Analgesics administeredpostoperatively

Acetaminophen 35 34

Pentazocine 8 9

Types of operation

Tonsillectomy 19 20

Tonsillectomy with adenoidectomy 26 25

Values are mean (standard deviation) or number.

TABLE II.Number (%) of Patients Experiencing Emesis-Free, Retching, or

Vomiting During the First 24 Hours (0–24 h) and the Next 24Hours (24–48 h) After Anesthesia.

Granisetron(n 5 45)

Ramosetron(n 5 45) P

0 to 24 hours after anesthesia

Emesis-free 40 (89%) 42 (93%) .357

Retching 1 (2%) 1 (2%) .999

Vomiting 5 (11%) 3 (7%) .357

24 to 48 hours after anesthesia

Emesis-free 32 (71%) 42 (93%) .006

Retching 5 (11%) 2 (4%) .217

Vomiting 9 (20%) 2 (4%) .025

Values are number (%).


256

obesity, a history of motion sickness and/or previous post-operative vomiting, operative procedure, anesthetic tech-nique, and postoperative pain are considered to affect theincidence of postoperative emesis. In this study, however,there were no differences between the groups with regardto patient demographics, types of operation, anestheticsadministered, and analgesics used postoperatively. Chil-dren with a history of motion sickness and/or previouspostoperative vomiting were excluded from the study be-cause they had a relatively high incidence of postoperativeemesis.3 Therefore, the difference between the groups inthe rate of patients being emesis-free can be attributed tothe study drug.

Granisetron is effective for the treatment of nauseaand vomiting in patients receiving cytotoxic drugs.4 Re-cently, we have demonstrated that granisetron reducesthe incidence of vomiting after tonsillectomy with or with-out adenoidectomy in children5 and have also demon-strated that 40 mg/kg granisetron is the minimal effectivedose for the prevention of postoperative vomiting.6 There-fore, the same dose of granisetron was administered inthis study. The exact mechanism of granisetron for theprevention of postoperative vomiting is unknown, butgranisetron may act on sites containing 5-HT3 receptorswith demonstrated antiemetic effects.10

Ramosetron, another 5-HT3 receptor antagonist, iseffective for the treatment of nausea and vomiting inducedby anticancer drugs.11 We could not find any report toevaluate the efficacy and safety of ramosetron for theprevention of vomiting after tonsillectomy in children. Inthis clinical trial, we showed that ramosetron, as well asgranisetron, was effective for increasing the rate of pa-tients being emesis-free. The precise mechanism of ramo-setron for the prevention of postoperative vomiting is notknown, but ramosetron may act at the area postrema andthe nucleus tractus, which contains a number of 5-HT3

receptors.12 Thus, the possible mechanism of ramosetronfor the prevention of postoperative vomiting is similar tothat of granisetron. The dose of ramosetron to be used inchildren is not established but was extrapolated from theadult investigation that ramosetron 0.3 mg (approximate-ly 6 mg/kg) reduces the incidence of nausea and vomitingafter middle ear surgery.8 Further studies are needed todetermine the dose-response curves available for the an-tiemetic effect of ramosetron in children.

We could find no report to compare the antiemeticefficacy of ramosetron with granisetron for the preventionof postoperative vomiting in children undergoing generalanesthesia for tonsillectomy, with a relatively high inci-dence of postoperative emesis.1,2 In this study, we demon-strated that the rate of patients being emesis-free 24 to 48hours after anesthesia was greater in patients who hadreceived ramosetron than in those who had received gran-isetron (P ,.05) and also demonstrated no differences inan emetic-free episode in the period from 0 to 24 hoursafter anesthesia between the groups. These findings sug-gest that ramosetron has a more potent antiemetic effectthan granisetron, lasting up to 48 hours. The exact reasonfor the difference in these two 5-HT3 receptor antagonists,granisetron and ramosetron, for the prevention of postop-erative vomiting is unknown but may be related to the

elimination half-life (3.1 h for granisetron vs. 5.8 hours forramosetron in healthy adult volunteers)13,14 and/or theaffinities of 5-HT3 receptor antagonists (1 for granisetronvs. 41 for ramosetron).7

Droperidol and metoclopramide have both been re-ported to be effective in the control of postoperative emesisin children.3 We compared the efficacy of droperidol, met-oclopramide, and granisetron for the prevention of vomit-ing after pediatric tonsillectomy.15,16 Consequently, gran-isetron is a better antiemetic than droperidol ormetoclopramide for the reduction of the incidence of post-operative vomiting. On the basis of our results, it is pre-dicted that prophylactic therapy with ramosetron wouldbe more effective than these two antiemetics for the pre-vention of postoperative vomiting in children undergoinggeneral anesthesia for tonsillectomy.

Dimenhydrinate is one of antihistamines that act onthe vomiting center and the vestibular pathways and is aneffective, safe, and inexpensive antiemetic in pediatricpatients undergoing strabismus surgery.17 However,Hamid et al.18 have compared the efficacy of ondansetronand dimenhydrinate for the prevention of vomiting afteradenotonsillectomy in children and have demonstratedthat ondansetron is more effective than dimenhydrinatefor prophylaxis against postoperative vomiting in thispopulation. These findings suggest that the antiemeticeffectiveness of 5-HT3 receptor antagonists is superior toantihistamines for the prevention of vomiting after pedi-atric tonsillectomy.

Granisetron lacks the sedative, dysphoric, and extra-pyramidal symptoms associated with other non–5-HT3

receptor antagonists, such as droperidol and metoclopra-mide.19 We have recently demonstrated that granisetronis relatively free of adverse effects and is safe for thecontrol of vomiting after tonsillectomy in children.5,6 Ad-verse events attributable to these two drugs observed inthis study were not clinically serious in either group.Thus, ramosetron, as well as granisetron, is considered tobe relatively free of adverse effects.

The failure to include a control group receiving pla-cebo may be considered a deficiency in our study design.We have already shown that the antiemetic efficacy ofgranisetron is superior to placebo for the prevention ofvomiting after tonsillectomy in pediatric patients under-going general anesthesia, which consists of sevoflurane in67% N2O and 33% O2.5,6 Consequently, we excluded pa-tients receiving placebo in the current study because theanesthetic technique was not different. Moreover, Aspinaland Goodman20 have shown that there is a lack of reliableclinical information concerning placebo-controlled trials ofanother 5-HT3 receptor antagonist, ondansetron, for theprevention of postoperative nausea and vomiting.

In Japan, a 5-HT3 receptor antagonist, granisetron($102.00 for 3 mg) or ramosetron ($100.00 for 0.3 mg), ismuch more expensive than other commonly used and well-established antiemetics, droperidol ($1.80 for 1.25 mg)and metoclopramide ($0.60 for 10 mg). However, the use ofthese non–5-HT3 receptors as antiemetics has been lim-ited because these drugs occasionally cause excessive se-dation and/or extrapyramidal symptoms3. In several stud-ies by Fujii et al.,15,16 the efficacy of graniserton was


257

superior to droperidol or metoclopramide for prophylaxisagainst postoperative vomiting. Therefore, the choice ofantiemetics should not be based only on the calculation ofcosts, but also should take into consideration the prefer-ence of patients, as well as the overall cost of care if emesiswere to occur. Our results showed that ramosetron wasmore effective than granisetron in increasing the rate ofpatients being emesis-free during 24 to 48 hours afteranesthesia. The prophylactic use of ramosetron for reduc-ing the incidence of vomiting after pediatric tonsillectomymay be justified by its incremental effectiveness (i.e., an-tiemetic efficacy).

CONCLUSIONRamosetron is a better antiemetic than granisetron

for the long-term prevention of postoperative vomiting inchildren undergoing general anesthesia for tonsillectomywith or without adenoidectomy.

BIBLIOGRAPHY1. Furst SR, Rodarte A. Prophylactic antiemetic treatment with

ondansetron in children undergoing tonsillectomy. Anes-thesiology 1994;81:799–803.

2. Litman RS, Wu CL, Catanzaro FA. Ondansetron decreasesemesis after tonsillectomy in children. Anesth Analg 1994;78:478–481.

3. Watcha MF, White PF. Postoperative nausea and vomiting:its etiology, treatment, and prevention. Anesthesiology1992;77:162–184.

4. Bermudez J, Boyle EA, Minter WD, Sanger GJ. The anti-emetic potential of the 5-hydroxytryptamine3 receptor an-tagonist BRL 43694. Br J Cancer 1988;58:644–650.

5. Fujii Y, Tanaka H, Toyooka H. Granisetron reduces vomitingafter strabismus surgery and tonsillectomy in children.Can J Anaesth 1996;43:35–38.

6. Fujii Y, Toyooka H, Tanaka H. Effective dose of granisetronfor preventing postoperative emesis in children. Can JAnaesth 1996;43:660–664.

7. Fujihara A, Akuzawa S, Miyata K, Miyake A. Ramosetronhydrochloride: affinity for cloned human 5-HT3 receptorand 5-HT3 receptor antagonistic and antiemetic effect inthe ferret. Kiso to Rinsho [Japanese] 1996;30:1965–1972.

8. Fujii Y, Tanaka H, Kobayashi N. Prevention of postoperativenausea and vomiting after middle ear surgery. Laryngo-scope 1999;109:1988–1990.

9. Tramer M, Moore A, McQuay H. Prevention of vomiting afterpaediatric strabismus surgery: a systemic review using thenumbers-needed-to-treat methods. Br J Anaesth 1995;75:556–561.

10. Carmichael J, Cantwell BMJ, Edwards CM, et al. A pharma-cokinetic study of granisetron (BRL 43694A), a selective5-HT3 receptor antagonist: correlation of anti-emetic re-sponse. Cancer Chemother Pharmacol 1989;24:45–49.

11. Noda K, Ikeda M, Yoshida O, et al. Clinical evaluation oframosetron against the nausea and vomiting induced byanticancer drugs. Jpn J Clin Exp Med 1994;71:2765–2776.

12. Kamato T, Ito H, Nagakura Y, et al. Mechanism of cisplatin-and m-chlorophenylbiguanide-induced emesis in ferret.Eur J Pharmacol 1993;238:369–376.

13. Kawabata Y, Sakiyama H, Muto S, et al. Clinical evaluationand pharmacokinetics of ramosetron against the nauseaand vomiting induced by anticancer drugs. NishinihonJ Urol 1994;56:1445–1456.

14. Kumakura H, Koyanagi J, Nishioka Y, et al. Phase I study ofgranisetron: pharmacokinetics of granisetron followingsingle and repeat intravenous drips infusion in Japanesehealthy volunteers. J Clin Ther Med 1990;5:25–34.

15. Fujii Y, Saitoh Y, Tanaka H, Toyooka H. Comparison ofgranisetron and droperidol in the prevention of vomitingafter strabismus surgery or tonsillectomy in children. Pae-diatr Anaesth 1998;8:241–244.

16. Fujii Y, Toyooka H, Tanaka H. Antiemetic efficacy of granis-etron and metoclopramide in children undergoing ophthal-mic or ENT surgery. Can J Anaesth 1996;43:1095–1099.

17. Vener DF, Alison SC, Sikich N, et al. Dimenhydrinate de-creases vomiting after strabismus surgery in children.Anesth Analg 1996;82:728–731.

18. Hamid SK, Selby IR, Sikich N, Lerman J. Vomiting afteradenotonsillectomy in children: a comparison of ondanse-tron, dimenhydrinate, and placebo. Anesth Analg 1998;86:496–500.

19. Yarker YE, McTavish D. Granisetron: an update of its ther-apeutic use in nausea and vomiting induced by antineo-plastic therapy. Drugs 1994;48:761–793.

20. Aspinal RL, Goodman NW. Denial of effective treatment andpoor quality of clinical information in placebo controlledtrials of ondansetron for postoperative nausea and vomit-ing: a review of published trials. BMJ 1995;311:844–846.


258

prevention of vomiting after tonsillectomy in children: granisetron versus ramosetron

Documents