Print version ISSN 0034-7094
Rev. Bras. Anestesiol. vol.55 no.4 Campinas July/Aug. 2005
Dexamethasone compared to metoclopramide in the prophylaxis of emesis in children undergoing ambulatory surgical procedures*
Dexametasona comparada a la metoclopramida en la prevención de vómitos pos-operatorios en niños sometidos a procedimientos quirúrgicos ambulatoriales
Antonio Bedin, TSA, M.D.I; Mauro de Souza Leite PinhoII; Cladir Terezinha ZanotelliIII; Ângela Sílvia CaldartIV; Jurandir Coan Turazzi, TSA, M.D.V ; Renato Almeida Couto de Castro, TSA, M.D.VI
pelo CET/SBA de Joinville
IIColoproctologista. Doutorado em Medicina pela Universidade de Birmingham, Inglaterra. Docente do Programa de Mestrado em Saúde e Meio Ambiente da Univille
IIIGraduada em Matemática. Doutorado em Engenharia de Produção, Universidade Federal de Santa Catarina. Docente do Programa de Mestrado em Saúde e Meio Ambiente da Univille
IVGraduada em Farmácia e Bioquímica. Farmacêutica da Secretaria Municipal de Saúde de Joinville
VAnestesiologista. Co-responsável pelo CET/SBA de Joinville. Coordenador do Serviço de Anestesiologia de Joinville
VIAnestesiologista. Responsável pelo CET/SBA de Joinville
BACKGROUND AND OBJECTIVES: Postoperative
vomiting is a common and unpleasant complication. Currently, however, mathematical
models, such as number necessary to treat (NNT) and relative risk reduction
(RRR), have been useful in the decision of which medication to use for prophylaxis.
This study aimed at verifying whether dexamethasone, as compared to metoclopramide,
decreases the incidence of vomiting when intravenously administered to children
anesthetized with sevoflurane for ambulatory pediatric surgeries.
METHODS: Two hundred and thirty seven male children, aged 11 months to 12 years, physical status ASA I and II, undergoing hernia repair were included in this study. They were premedicated with oral midazolam. Anesthesia was induced and maintained with sevoflurane, nitrous oxide, and 1 µg.kg-1 fentanyl. Patients were divided in two groups: group D patients (n = 118) were given 150 µg.kg-1 dexamethasone at induction while group M (n = 119) received 150 µg.kg-1 metoclopramide at induction. The following parameters were evaluated: incidence of vomiting in the first 4 postoperative hours (PO), incidence of vomiting between 4 and 24 PO hours, NNT of both medications and RRR of dexamethasone as compared to metoclopramide.
RESULTS: The incidence of vomiting was 9.32% for group D and 33.61% for group M during the first 4 PO hours, and 1.69% with dexamethasone and 3.36% with metoclopramide between 4 and 24 PO hours. RRR of dexamethasone related to metoclopramide in the first 4 hours was 72%. The number necessary to treat (NNT) for dexamethasone was 3.25 and for metoclopramide it was 15.66.
CONCLUSIONS: Dexamethasone is more effective than metoclopramide in decreasing the incidence of vomiting when used during anesthetic induction with sevoflurane associated to nitrous oxide and fentanyl.
Key Words: ANTIEMETICS: dexamethasone, metoclopramide; COMPLICATIONS: nausea, vomiting; SURGERY, Ambulatory
JUSTIFICATIVA Y OBJETIVOS: El vómito
pos-operatorio es una complicación común y desagradable. Sin embargo,
actualmente modelos matemáticos tales como número necesario para tratar
(NNT) y reducción del riesgo relativo (RRR) han sido útiles para la
decisión de cual medicación utilizar para la profilaxis. El objetivo
de este estudio fue verificar si la dexametasona, comparada a la metoclopramida,
reduce la incidencia de vómitos cuando administrada por vía venosa
en niños anestesiados con sevoflurano en cirugías pediátricas
MÉTODO: Participaron del estudio 237 niños, del sexo masculino, con edad entre 11meses y 12 años, estado físico ASA I y II, sometidos a herniorrafia inguinal. Como medicación pre-anestésica fue utilizado midazolan por vía oral. Para la inducción y mantenimiento de la anestesia fue utilizado sevoflurano, óxido nitroso y fentanil (1 µg.kg-1). Los pacientes fueron divididos en grupo D (n = 118) dexametasona (150 µg.kg-1) por vía venosa en la inducción y grupo M (n = 119) metoclopramida (150 µg.kg-1) en la inducción. Fueron analizadas las incidencias de vómitos en las primeras 4 horas de pos-operatorio (PO), entre 4 horas y 24 horas de PO, el NNT de ambas medicaciones utilizadas y la RRR de la dexametasona con relación a la metoclopramida.
RESULTADOS: La incidencia de vómitos fue 9,32% en el grupo D y 33,61% en el grupo M durante las primeras 4 horas de PO y 1,69% con la dexametasona y del 3,36% con la metoclopramida entre 4 y 24 horas de PO. El RRR fue 72% de la dexametasona con relación a la metoclopramida en las primeras 4 horas. El NNT de la dexametasona fue 3,25 y de la metoclopramida fue 15,66.
CONCLUSIONES: La dexametasona reduce, de forma más eficiente que la metoclopramida, la incidencia de vómitos cuando utilizada durante la inducción de anestesia con sevoflurano asociado al óxido nitroso y fentanil.
Nausea and vomiting are among the most frequent anesthetic side-effects 1. These complications are common causes of late discharge after anesthetic recovery, anxiety, dehydration and metabolic changes in patients submitted to ambulatory surgeries 2,3.
Although the incidence of postoperative vomiting is higher in children as compared to adults, prophylactic antiemetics are controversial 1,2; some authors recommend them while others state that they should not be routinely used because most popular agents present adverse effects, including sedation, headache, agitation or extrapyramidal symptoms 1,2.
Metoclopramide is commonly used as prophylactic antiemetic drug. However, its use is limited by low efficiency and the possibility of undesirable side effects, such as extrapyramidal symptoms 2. Dexamethasone is a low cost steroid, with low incidence of side effects, prolonged antiemetic effect in oncologic patients, and markedly decreases the incidence of postoperative vomiting in children submitted to tonsillectomy 2. However, there are still few studies with dexamethasone to prevent POV, especially POV related to ambulatory pediatric surgeries such as inguinal hernia correction 2-4.
Recently, some mathematical models have become part of clinical research to aid in the difficult decision of adopting or not a drug. These mathematical models, such as number necessary to treat (NNT) and relative risk reduction (RRR) help clarifying the efficacy of a given medication 5-9.
The purpose of this study was to evaluate whether dexamethasone is more effective than metoclopramide in decreasing postoperative vomiting when used during anesthetic induction with sevoflurane, nitrous oxide and fentanyl, in children submitted to ambulatory surgical procedures.
This clinical prospective study was carried out at Hospital Municipal São José, Joinville, SC, after the Regional Research Ethics Committee approval and the informed consent of children's parents or guardians. Inclusion criteria were male children aged 11 months to 12 years, physical status ASA I or II, with negative feces exam and no history of postoperative vomiting in previous surgeries or motion sickness, submitted to ambulatory inguinal hernia correction under general anesthesia (standard technique).
To prepare allocation tables and distribute patients in this study, Microsoft Excel XP® random number function was used. Patients were premedicated with 0.3 mg.kg-1 oral midazolam one hour before surgery. Anesthesia was induced with sevoflurane, 1 µg.kg-1 fentanyl and 50% oxygen and nitrous oxide, followed by laryngeal mask or tracheal intubation. Incision site was infiltrated with up to 2 mg.kg-1 bupivacaine and 20 mg.kg-1 dipirone was intravenously administered. Anesthesia was maintained with sevoflurane and 50% oxygen and nitrous oxide. Hydration was achieved with 5% glucose solution (10 mL.kg-1) every hour. Oral paracetamol (10 mg.kg-1) every 6 hours was prescribed at hospital discharge.
To evaluate a drug expected to decrease vomiting in 50% of a population submitted to surgical procedures in which the expected POV incidence is 40% (according to our previous surveys) and considering acceptable alpha = 0.05 and beta = 0.2, at least 91 patients are needed in each group, according to chart published by Apfel 9. For this reason, 240 patients were randomly divided in two groups: group D (n = 120) received 150 µg.kg-1 intravenous dexamethasone and group M (n = 120) received 150 µg.kg-1 intravenous metoclopramide, both at induction. Bolus intravenous ondansetron (100 µg.kg-1) was used to treat postoperative vomiting.
Evaluated variables were the incidence of vomiting in the first 4 postoperative hours (vomiting defined as gastric content expelling), the incidence of multiple vomiting episodes in the first 4 postoperative hours (defined as different vomiting episodes observed in intervals higher than 5 minutes) and vomiting in the first 24 postoperative hours (confirmed by telephone contact).
The following parameters were also evaluated: metoclopramide side effects (sedation, agitation, diarrhea, central nervous system depression), dexamethasone side effects (epigastric pain, anxiety, insomnia) and change to inpatient regimen due to postoperative vomiting. The relative risk reduction (RRR) of POV was calculated with the formula RRR = (1-(PRD/PRM))x100, where PRD is the probability of POV decrease with dexamethasone and PRM the probability of POV decrease with metoclopramide during 24 hours for recurrent vomiting and between 4 and 24 hours. The number necessary to treat (NNT) for dexamethasone and metoclopramide was calculated by the formula NNT=1/(PEA-POT), where PEA is the expected probability of the adverse event and POT the observed probability of the event in the group treated during the first 4 postoperative hours.
Descriptive statistics (mean and standard deviation) was used to characterize studied population. Student's t test was used to compare differences between mean values of age, body mass index and height, and Chi-square test (X2) was used to check whether there has been significant difference in the incidence of vomiting between groups, with a confidence limit of 95% (p <0.05). Excel XP® was our chosen computer program.
Two hundred and thirty seven patients, 118 in group D (dexamethasone) and 119 in group M (metoclopramide) took part in this study (Table I). Two group D patients and one group M patient were excluded because telephone contact was impossible after hospital discharge. Three group D patients and 2 group M patients needed tracheal intubation, but none of them presented POV within 24 hours. There were no significant differences in body mass index, height, age, surgery duration and time between surgery completion and hospital discharge (postoperative recovery period) according to Student's t test for two samples (Table I and Table II).
The incidence of vomiting of 9.32%(11) in Group D and of 33.61% (40) in group M during the first 4 postoperative hours was statistically different between groups, X2 = 5.6, for a significance level of 0.05, 1 level of freedom and a critical Chi-square of 3.84 (Figure 1).
The incidence of recurrent vomiting (more than 1 episode) of 5.08% (6) in group D and of 15.97% (19) in group M during the first 4 postoperative hours was not statistically different between groups, X2 = 0.002, for a level of significance of 0.05, 1 level of freedom and critical Chi-square of 3.84 (Figure 2).
The incidence of vomiting of 1.69% (2) in group D and of 3.36% (4) in group M after hospital discharge and before completing 24 postoperative hours was not statistically different, X2 = 0.41, for a significance level of 0.05, 1 level of freedom and critical Chi-square of 3.84 (Figure 3). All patients presenting vomiting after discharge were among those who had already presented vomiting in the first 4 postoperative hours.
There were no side effects both with dexamethasone and metoclopramide. There were also no hospital admissions due to POV.
The relative risk reduction (RRR) of dexamethasone related to metoclopramide was 72% in the first 4 hours, 72.77% for recurrent vomiting in the first 4 hours and 50% between 4 and 24 hours (Figure 4).
The number necessary to treat (NNT) was 3.25 for dexamethasone and 15.66 for metoclopramide (Figure 5).
Nausea is a subjective and unpleasant sensation at the epigastrium and oropharynx, associated to an urgent need to vomit 3. Vomiting is characterized by diaphragm, abdominal muscles, respiratory muscles and chest wall spasmodic contraction followed by gastric content expelling through the mouth as result of this abdominal wall and diaphragm contraction and cardia opening 10.
Nausea is a difficult sensation for children to describe, so most pediatric studies have considered vomiting alone. This explains some differences in results of studies with children or adults 4,11-13.
Vomiting center, located in brainstem reticular formation, controls and coordinates the complex vomiting process. This area receives impulses from other central nervous system regions, including trigger chemoreceptor area, cerebellum, vestibular system, cortical centers and solitaire tract nucleus 3. These areas are rich in serotoninergic, muscarinic, histaminic, opioid and dopaminergic receptors. Blocking of these receptors has been postulated as the action mechanism of antiemetic drugs. The efferent impulse of the vomiting center follows via phrenic nerve and spinal nerves of abdominal wall muscles during vomiting 3,13,14.
It has been reported a general POV incidence of 33.2% in 983 pediatric patients submitted to different surgeries in a community hospital 18. A different study has reported that POV may vary from 15% to 54% in lower abdomen surgeries, such as hernia correction, hydrocelectomies and orchiopexis 14. A great number of factors contribute for the incidence of POV and not all can be controlled by the anesthesiologist. Pediatric patients present a higher incidence of POV as compared to adults 2,12.
During childhood, the incidence of POV is similar for males and females, but at adolescence, the incidence is higher among females. Both children and adults with history of motion sickness or previous POV are at higher risk of POV. Premedicated patients, who are more relaxed at the beginning of surgery, have a lower incidence of POV 2.
Previous studies have suggested that obese patients were at higher risk of POV 2. However, current data do not relate obesity to increased incidence of POV 3. Adult smokers have a lower incidence of POV, but there are no data on the incidence of POV in children living with smoking adults 2. Some surgical procedures are associated to high incidence of POV, such as strabismus, ear, nose and throat procedures, middle ear surgeries, tonsillectomies, orchiopexis and craniofacial surgeries.
The incidence of POV increases with surgery length in approximately 6% for every 30 minutes. Opioids, nitrous oxide, volatile anesthetics and anti-cholinesterase drugs increase the incidence of POV, while propofol seems to decrease it. Studies have shown correlation between pain intensity and incidence of POV in children 2. Sudden movements in the recovery unit may trigger POV. Time for first postoperative meal may also influence the incidence of POV during recovery 2.
The probability of POV in adults may be estimated by the number of associated risk factors, such as female gender, previous history of POV or motion sickness, non-smoker and administration of opioids for postoperative analgesia. If none, one, two, three or four factors are present, the risk of POV is approximately 10%, 20%, 40%, 60% and 80%, respectively 9.
In children, however, predictive factors have only a weak correlation with the actual incidences 12. There are enough data to suggest that vomiting prophylaxis in patients with moderate (30% to 60%) or high risk (above 60%) is justified because it decreases costs and increases patients' satisfaction. However, this would not be justified in low risk patients (below 30%) 1,9,14.
Several drugs have been used to prevent POV after their efficacy has been established by clinical trials 9. The choice of the drug should be based on the evidence of efficacy, few side effects and low cost 1,9.
Basic action of metoclopramide is facilitating acetylcholine release from enteric neurons. This action may be mediated by several mechanisms. There might be suppression of inhibitory interneurons by antagonism of 5-HT3 receptors and stimulation of excitatory neurons by 5-HT4 receptors. Metoclopramide is dopamine (D2) antagonist. It is very likely that central dopaminergic effects contribute to antiemetic and nausea inhibitory effects of metoclopramide 2.
High metoclopramide doses (1 to 2 mg.kg-1) have been successfully used to prevent or decrease chemotherapy-induced vomiting (CIV). In much lower doses (0.1 to 0.2 mg.kg-1) it has been used to treat POV 2. These lower doses were used in an attempt to decrease sedative effects and the high incidence of dystonic reactions associated to CIV doses. Even with them, children seem to be more prone to these side effects as compared to adults. Reported metoclopramide efficacy, when administered in lower doses for POV, is very variable 2. A great number of factors may contribute to inconsistent results, including administration route, type of surgery and differences in anesthetic techniques. Due to metoclopramide short half-life, it is unlikely that its administration during anesthetic induction may produce any antiemetic effect after the emergence period 2,3. Systematic reviews have questioned the efficacy of 10 mg metoclopramide in adults 4.
One review has stated that for 250 µg.kg-1, NNT for metoclopramide in children is 7.9 15.
Steroids, especially dexamethasone, have been used for CIV for many years. However, their exact action mechanism in preventing nausea or vomiting is still unknown. Prostaglandins antagonism, endorphins release and tryptophan depletion are suggested. However it is still unknown whether these are peripheral or centrally mediated effects 2.
It is also possible that anti-inflammatory or membrane stabilizing effects contribute for the antiemetic effects of these drugs 2. Prophylactic dexamethasone has decreased the incidence of POV. Most frequently studied doses in adults were intravenous 8 mg to 10 mg. In children, typical doses vary from 0.1 mg.kg-1 to 1.5 mg.kg-1. In such doses, NNT to prevent POV was 7.1 for adults and 3.8 for children 16. However, it is possible that doses below 5 mg in adults are also effective 2. Intravenous 10 mg dexamethasone before anesthetic induction was more effective than 10 mg at surgery completion in preventing POV in the first 2 postoperative hours 1,2.
From a strictly statistic point of view, one should compare the antiemetic drug to a control group using placebo to prevent postoperative vomiting. But it is ethically not justifiable to deny prophylaxis and treatment when a previous meta analysis study has already quantified the effect of an antiemetic drug. So, it is recommended to compare the new drug to a well-established antiemetic agent 9. Metoclopramide is an antiemetic agent used for more than 40 years 4 although its effectiveness is being currently questioned 1,4,15. Dexamethasone is used as antiemetic agent in oncologic patients for more than 10 years, however there are still few studies on postoperative vomiting prophylaxis in children 1,2,9,16.
Mathematical models, such as relative risk reduction and number necessary to treat, are examples of how mathematics may be useful for decision making in Anesthesiology. The relative risk reduction (RRR) is calculated by the formula RRR=(1-(PRB/PRA))x100, where PRB is the probability of decreasing POV with drug B and PRA the probability of decreasing POV with drug A. Relative risk reduction is expressed in percentage 6,7. In our study, RRR=72% when dexamethasone was compared to metoclopramide during the first 4 postoperative hours, expresses that dexamethasone is 72% more effective as compared to metoclopramide.
The number necessary to treat (NNT) is obtained by the formula NNT=1/(PEA-POT), where PEA is the expected probability of the adverse event and POT is the observed probability in the treated group 6,7. In our study, dexamethasone in the first 4 postoperative hours had NNT = 3.25, better than metoclopramide (NNT = 15.66), meaning that with dexamethasone, 3.25 patients have to be treated to prevent one POV episode, while with metoclopramide, 15.66 patients have to be treated to prevent one POV episode.
The understanding of POV studies has improved after the concepts of relative risk reduction and number necessary to treat were expanded to this field of study. This has determined the number of patients to be treated to obtain a desired effect. The advantage of this approximation is to allow indirect comparisons of several therapeutic regimens. These mathematical models are not opposed to traditional statistical tests, but rather add to them 6-8.
POV is classically described in the literature as an obstacle to early discharge, which is exactly the aim of ambulatory anesthesia 1,10. In our study, however, there was no statistically significant difference between groups in time between surgery completion and hospital discharge, and there was no hospital admission due to incoercible vomiting.
This study has concluded that dexamethasone was more effective than metoclopramide in decreasing the incidence of postoperative vomiting, when used during anesthetic induction with sevoflurane associated to nitrous oxide and fentanyl, in male children submitted to ambulatory inguinal hernia correction
01. Watcha MF - Management of postoperative vomiting in pediatric patients. Cur Op Anaesth, 2003;16:575-583. [ Links ]
02. Olutoye O, Watcha MF - Management of postoperative vomiting in pediatric patients. Int Anesthesiol Clin, 2003;41:99-117. [ Links ]
03. Scuderi PE - Pharmacology of antiemetics. Int Anesthesiol Clin, 2003;41:41-66. [ Links ]
04. Gan TJ, Meyer T, Apfel CC et al - Consensus guidelines for managing postoperative nausea and vomiting. Anesth Analg, 2003;97:62-71. [ Links ]
05. Carvalho WA, Vianna PTG, Braz JRC - Náuseas e vômitos em anestesia: fisiopatologia e tratamento. Rev Bras Anestesiol, 1999;49:65-79. [ Links ]
06. Schulman SR, Schardt C, Erb T - Evidence-based medicine in anesthesiology. Cur Op Anaesthesiol, 2002;15:661-668. [ Links ]
07. Angelis RMC, Avezum Jr A, Cavalcanti AB et al - Anestesiologia baseada em evidências. O que é e como praticar. Rev Bras Anestesiol, 2004;54:582-594. [ Links ]
08. Duval Neto GF - Anestesiologia na era da medicina baseada em evidências. Rev Bras Anestesiol, 2004;54:141-144. [ Links ]
09. Apfel CC, Roewer N, Korttila K - How to study postoperative nausea and vomiting. Acta Anaesthesiol Scand, 2002;46:921-928. [ Links ]
10. Scuderi PE, Conlay LA - Postoperative nausea and vomiting and outcome. Int Anesthesiol Clin, 2003;41:165-174. [ Links ]
11. Apfel CC, Roewer N - Risk assessment of postoperative nausea and vomiting. Int Anesthesiol Clin, 2003;41:13-32. [ Links ]
12. Eberhart LHJ, Morin AM, Guber D et al - Applicability of risk scores for postoperative nausea and vomiting in adults to paediatric patients. Br J Anaesth, 2004;93:386-392. [ Links ]
13. Everett LL - Can the risk of postoperative nausea and vomiting be identified and lowered during the preoperative assessment? Int Anesthesiol Clin, 2002;40:47-62. [ Links ]
14. Villeret I, Laffon M, Duchalais A et al - Incidence of postoperative nausea and vomiting in paediatric ambulatory surgery. Paediatr Anaesth, 2002;12:712-717. [ Links ]
15. Henzi I, Walder B, Tramer MR - Metoclopramide in the prevention of postoperative nausea and vomiting: a quantitative systematic review of randomized, placebo-controlled studies. Br J Anaesth, 1999;83:761-771. [ Links ]
16. Henzi I, Walder B, Tramer MR - Dexamethasone for the prevention of postoperative nausea and vomiting: a quantitative systematic review. Anesth Analg, 2000;20:186-194. [ Links ]
Dr. Antonio Bedin
Address: R. Doutor Roberto Koch, 72 Bairro Atiradores
ZIP: 89203-088 City: Joinville, Brazil
Submitted for publication September 13, 2004
Accepted for publication March 14, 2005
* Received from Hospital Municipal São José, Joinville, SC