Get documents about "The Effect of Preoperative Dexamethasone on the Immediate and
Document Sample
The Effect of Preoperative Dexamethasone on the Immediate
and Delayed Postoperative Morbidity in Children Undergoing
Adenotonsillectomy
Ana Lucia S. Pappas, MD*, Radha Sukhani, MD*, Andrew J. Hotaling, Mm,
Marianna Mikat-Stevens, MD*, Joseph J. Javorski, MD*, Joseph Donzelli, MDt, and
Kalavathi Shenoy, MD*
Departments of *Anesthesiology and tOtolaryngology, Head and Neck Surgery, Loyola University Medical Center,
Maywood, Illinois
In this prospective, randomized, double-blind, incidence of PONV, need for rescue antiemetics, quality
placebo-controlled study, we examined the effect of of oral intake, and analgesic requirements did not differ
preoperative dexamethasone on postoperative nausea between groups. However, during the 24 h after dis-
and vomiting (PONV) and 24-h recovery in children charge, more patients in the control group experienced
undergoing tonsillectomy. One hundred thirty chil- PONV (62% vs 24% in the steroid group) and com-
dren, 2-12 yr of age, ASA physical status I or II, com- plained of poor oral intake. Additionally, more children
pleted the study. All children received oral midazolam in the control group (8% vs 0% in the steroid group)
0.5-0.6 mg/kg preoperatively. Anesthesia was in- returned to the hospital for the management of PONV
duced with halothane and nitrous oxide in 60% oxygen and/or poor oral intake. The preoperative administra-
and maintained with nitrous oxide and isoflurane. In- tion of dexamethasone significantly decreased the inci-
tubation was facilitated by mivacurium 0.2 mg/kg. dence of PONV over the 24 h after home discharge in
Each child received fentanyl 1 pg/kg IV before initia- these children. Implications: In this double blind,
tion of surgery, as well as dexamethasone 1 mg/kg placebo-controlled study, we examined the efficacy of a
(maximal dose 25 mg) (steroid group) or an equal vol- single large dose (1 mg/kg; maximal dose 25 mg) of
ume of saline (control group). Intraoperative fluids preoperative dexamethasone on posttonsillectomy
were standardized to 25-30 mL/kg lactated Ringer’s postoperative nausea and vomiting (PONV) in children
solution. All tonsillectomies were performed under the 2-12 yr of age undergoing tonsillectomy. Compared
supervision of one attending surgeon using an electro- with placebo, dexamethasone significantly decreased
dissection technique. Postoperatively, fentanyl and the incidence of PONV in the 24 h after discharge, im-
acetaminophen with codeine elixir were administered proved oral intake, decreased the frequency of parental
as needed for pain. Rescue antiemetics were adminis- phone calls, and resulted in no hospital returns for the
tered when a child experienced two episodes of retch- management of PONV and/or poor oral intake.
ing and/or vomiting. Before home discharge, the (Anesth Analg 1998;87:57-61)
T and adenoidectomy are the major
onsillectomy pain, poor oral intake, dehydration and fever, how-
ambulatory surgical procedures most frequently ever, continues to be a concern in children undergoing
performed in children (1). The safety of these tonsillectomy in an ambulatory setting (6,7). The re-
procedures in the ambulatory setting has been well ported incidence of PONV after tonsillectomy is 40%-
documented (l-4). The introduction of an electrodis- 73% (8-10). Several anesthetic and antiemetic regi-
section surgical technique has virtually eliminated im- mens have been used to minimize PONV with
mediate postoperative hemorrhage (5). Morbidity re- variable success (8-12).
lated to postoperative nausea and vomiting (PONV), In children undergoing tonsillectomy, dexametha-
sone and other steroid preparations have been used to
minimize tissue injury and edema and related mor-
bidity, such as pain, fever and poor oral intake (13-16).
Accepted for publication March 31, 1998. The prolonged antiemetic effect of IV dexamethasone
Address correspondence and reprint requests to Ana Lucia
Pappas, MD, Department of Anesthesiology, Loyola University is well documented in chemotherapy-induced nausea
Medical Center, 2160 South First Avenue, Maywood, IL 60153. and vomiting (17). The effect of dexamethasone in
01998 by the International Anesthesia Research Society
0003-2999/98/$5.00 Anestb Analg 1998;87:57-61 57
58 PEDIATRIC ANESTHESIA PAPPAS ET AL. ANESTH ANALG
DEXAMETHASONE, CHILDREN, AND ADENOTONSILLECTOMY 1998;87:57-61
tonsillectomy-associated PONV, however, is contro- All tonsillectomies were performed using an elect-
versial (15,16,18). Because PONV is a multifactorial rodissection technique under the supervision of one
problem, these conflicting results could be related to attending surgeon (AJH) who was also responsible for
the lack of standardization of anesthetic and periop- grading the degree of tonsillar enlargement (1 = ton-
erative factors that contribute to the incidence and sils within tonsillar folds, 2 = tonsils just outside
severity of PONV. tonsillar folds, 3 = tonsils well outside tonsillar folds
The purpose of the present study was to determine but not reaching uvula, 4 = tonsils reaching uvula or
whether one single, large dose of IV dexamethasone past uvula). At the conclusion of surgery, gastric con-
administered before surgery could decrease PONV tents were suctioned via an orogastric tube. Drugs to
and improve oral intake in the first 24 h after home reverse muscle relaxation were not required in any
discharge. To specifically delineate the contribution of case. The children were tracheally extubated when
dexamethasone, all anesthetic and nonanesthetic fac- they awoke and demonstrated satisfactory motor re-
tors that could influence the incidence of PONV were covery. Children were transported to the postanesthe-
standardized. sia care unit (PACU 1); when they were fully awake,
comfortable, and stable, they were transferred to a
step-down recovery unit (PACU 2). All children were
observed in the PACU for a combined period of at
Methods least 180 min. Oral fluids were offered to the children
After institutional review board approval, 130 chil- but they were not required to take them. Children had
dren, 2-12 yr of age, ASA physical status I or II, who to be able to swallow without difficulty before they
were scheduled for ambulatory tonsillectomy and ad- were discharged home. Children were considered
enotonsillectomy were enrolled in the study. Written, ready for discharge when they met institutional crite-
informed parental consent was obtained in all cases. ria: they were awake, alert, comfortable, and able to
The study design was randomized, double-blind, and swallow without difficulty and had stable vital signs,
placebo-controlled. Children who received antiemet- minimal or no nausea, and no retching and vomiting.
its, antihistaminics, steroids, or psychoactive drugs Parents participated in the child’s care for the entire
within 24 h of surgery were excluded, as were chil- PACU stay.
dren who had a history of diabetes and those in whom All vomiting or retching episodes during PACU
IV induction was indicated; e.g., those with gastro- stay were recorded by the PACU nurse caring for the
esophageal reflux. Children were allowed to eat solid child. For the purposes of data collection, only retch-
food until 12 AM on the day before surgery and to have ing and vomiting episodes were documented, because
clear liquids until 3 h before the expected start of nausea is difficult to assess in children. Episodes of
surgery. retching and vomiting occurring <5 min apart were
All children received oral midazolam 0.5-0.6 considered one episode. A rescue antiemetic (metoclo-
mg/kg (maximal dose 20 mg) for preanesthetic med- pramide 0.15 mg/kg) was administered IV when a
ication 2030 min before anticipated induction. After child experienced two episodes of retching and/or
establishing standard monitoring, general anesthesia vomiting. If metoclopramide did not control retch-
was induced using halothane and a gas mixture of ing and/or vomiting after 20 min, a second rescue
60% nitrous oxide and oxygen. IV access was estab- antiemetic (ondansetron 0.15 mg/kg IV) was
lished after the induction of anesthesia. The amount of administered.
IV fluid administered was standardized to 25-30 The need for postoperative pain medication was
mL/kg of lactated Ringer’s solution during the intra- assessed by the PACU nurses. IV fentanyl in incre-
operative period, followed by a maintenance-rate in- ments of 0.5 pg/kg was given for pain during early
fusion until discharge. Endotracheal intubation was recovery until the child was comfortable. Once chil-
facilitated by mivacurium 0.2 “g/kg. Anesthesia was dren demonstrated the ability to swallow, they were
maintained with nitrous oxide and isoflurane adjusted given acetaminophen with codeine elixir (0.75-
to maintain heart rate and blood pressure values 1 mg/kg codeine) to control pain. Children were of-
within 20% of the baseline induction value. Each child fered liquids 45 min after arrival to PACU. The quality
received fentanyl 1 pg/kg before surgery. Dexameth- of oral intake was judged as follows: excellent = child
asone 1 mg/kg (maximal dose 25 mg, steroid group) requests it, good = child accepts it when offered,
or an equal volume of saline (control group) was fair = child accepts it when coaxed, and poor = child
administered IV in a randomized double-blind fash- refuses.
ion after the induction of anesthesia before surgery. All patients were discharged home with prescrip-
Randomization was guided by a computer-generated tions for acetaminophen with codeine and plain acet-
number table. The study drugs were prepared by the aminophen to be taken concurrently every 4 h while
pharmacy, and an appropriate code number was awake. Parents were instructed to follow the pain
assigned. medication schedule diligently. An evaluation diary/
ANESTH ANALG PEDIATRIC ANESTHESIA PAPPAS ET AL. 59
1998;87:5741 DEXAMETHASONE, CHILDREN, AND ADENOTONSILLECTOMY
log book was provided to parents, and they were Table 1. Demographic Characteristics, Duration of
instructed to record oral intake (graded in the manner Anesthesia and Surgery, Oral Intake, and PACU Stay
similar to that in the PACU), number of episodes of Steroid Control
retching and vomiting, compliance with oral pain group group
medications, and state of hydration (frequency of (n = 63) (n = 65)
voiding). Parents were asked to call the otolaryngol- Age W 6.0 t 2.5 5.8 2 2.7
ogy service if the child experienced more than two Weight (kg) 27.1 ? 13.5 26.0 2 13.3
episodes of vomiting or had poor oral intake and Anesthesia duration (mm) 62.3 2 18.5 63.0 5 15.3
failed to void for >12 h after surgery. A rescue anti- Surgery duration (min) 39.4 -c 13.3 40.2 k 14.7
emetic (trimethobenzamide hydrochloride supposi- Time to first oral intake in 84.3 t 28.3 83.6 + 42.3
PACU (min)
tory 200 mg for children weighing >15 kg and 100 mg
PACU stay duration (min) 183.1 + 19.1 191.7 + 2&o*
for children weighing ~15 kg to be administered ev-
ery 6 h as needed for a maximum of three doses) was PACU = postanesthesia care unit.
* P i 0.05 versus the steroid group.
prescribed for the children whose parents had called
and reported more than two episodes of vomiting.
Twenty-four hours after discharge, one of the investi-
with atypical pseudocholinesterase). Of the remaining
gators (ALP or RS) called the parents to document the
128 children, 63 received IV dexamethasone (steroid
first 24-h recovery data and compliance with analgesic
group) and 65 received saline (control group).
instructions.
There were no significant differences between the
Age, weight, surgery time, anesthesia time, dura-
two groups with respect to age, weight, gender distri-
tion of PACU stay, fentanyl requirement in PACU,
bution, blood loss during surgery, duration of anes-
incidence of PONV, need for rescue antiemetics, qual-
ity of oral intake in PACU, quality of oral intake after thesia and surgery, and time to oral intake (Table 1).
home discharge, compliance with oral pain medica- The duration of PACU stay, however, was signifi-
tions, calls to physicians, and hospital returns were cantly longer in the control group. Distributions of
documented for each patient. patients in the two study groups was comparable with
A postoperative emesis incidence of 40%-70% has respect to tonsillar size, surgical indication, and sur-
been reported after adenotonsillectomy in children gical procedure.
(B-10). This incidence was taken into consideration Incidence of PONV, need for rescue antiemetics,
when selecting the patient sample size for the current quality of oral intake, and analgesic requirements
study. A decrease in the incidence rate to 50% was were not significantly different between the two
considered to be clinically relevant. It was desired that groups before PACU discharge (Table 2). However,
this be detected with a P valve ~0.05 at a power of during 24-h period after discharge from the PACU, a
90%. The current study therefore was targeted for a significantly greater percentage of patients in the con-
minimal sample size of 110 patients (55 in each group). trol group experienced PONV. Parental calls to the
A larger number of children (65 in each group) was physician for complaints of pain, inability to maintain
recruited to overcome the factor of parental noncom- oral pain medication schedule, poor oral intake, and
pliance and to ensure that the satisfactory recovery PONV were also higher in the control group. Al-
data could be obtained for the projected number of though none of the patients in the steroid group re-
patients determined by power analysis. turned to the hospital, five children in the control
Student’s t-test was used to analyze group differ- group returned to the hospital for the management of
ences in patient demographics, duration of surgery, PONV or for IV hydration in the first 24 h after sur-
duration of anesthesia, and recovery times. ,$ analysis gery. Compared with those in the steroid group, sig-
and Fisher’s exact test were used as appropriate to nificantly fewer children in the control group had
compare symptom frequencies. A P valve CO.05 was good to excellent oral intake after home discharge
considered statistically significant. All data are pre- (Table 2).
sented as mean -+ SD. Five children in the steroid group experienced par-
tial laryngospasm at induction, and one experienced
bronchospasm after intubation. Among the children in
the control group, four experienced partial laryngo-
Results spasm at induction, and one experienced complete
Of the 130 children enrolled in the study, 2 were laryngospasm at emergence. These complications re-
excluded (1 child had generalized edema 1 day after solved with appropriate interventions and did not
surgery and was diagnosed with acute glomerulone- influence patient recovery. All the patients met dis-
phritis unrelated to the procedure; 1 child had pro- charge criteria and were discharged home from the
longed paralysis after mivacurium and was diagnosed PACU.
60 PEDIATRIC ANESTHESIA PAPPAS ET AL. ANESTH ANALG
DEXAMETHASONE, CHILDREN, AND ADENOTONSILLECTOMY 1998;87:57-61
Table 2. Frequency of Recovery Characteristics in the indicate that, in children undergoing ambulatory tonsil-
Early and Late Recovery Periods lectomy, a large single dose (1 mg/kg, maximal dose
Steroid Control 25 mg) of IV dexamethasone administered at the induc-
tion of anesthesia decreased PONV and improved oral
group group
(n = 63) (n = 65) intake during the first 24-h period after discharge. Ad-
Early (PACU) ditionally, significantly fewer of these children sought
Fentanyl administered 62 66 medical attention with respect to postoperative phone
Retching/vomiting 38 29 calls to a physician, and none returned to the hospital for
Rescue antiemetic 22 18 the management of PONV and/or poor oral intake. Of
Two rescue antiemetics 6 10 note, this beneficial effect of dexamethasone was not
Good to excellent oral intake 84 77 evident during early recovery (PACU stay).
Late (Discharge to 24 h)
Retching/vomiting 24 62* In a prospective, randomized, double-blind study,
Rescue antiemetic 3 10 Volk et al. (16) observed no differences in postopera-
Good to excellent oral intake 83 58” tive recovery variables, such as oral intake, level of
Return to hospital for PONV 0 8* activity, and analgesic use, in children who received a
and hydration single dose of dexamethasone 10 mg IV versus pla-
Frequency of phone calls 8 25” cebo, although they did not specifically examine
Total PONV. Catlin and Grimes (15) examined the incidence
Overall PONV 48 88*
Overall rescue antiemetics 25 28 of vomiting in addition to pain, fever, and oral intake
in 25 children (4-12 yr of a e) who received placebo or
Values are expressed as %. 8
PACU = postanesthesia care unit, PONV = postoperative nausea and
dexamethasone (8 mg/m ) IV before tonsillectomy.
vomiting. The only difference they observed was in return to a
* P < 0.05 versus the steroid group. full diet, which was faster after dexamethasone. These
studies were published in the otolaryngology litera-
ture and were not standardized for anesthetic and
Discussion perioperative factors.
Although the safety of tonsillectomy and adenotonsil- In the anesthesia literature, Splinter and Roberts
lectomy as outpatient surgical procedures is well doc- (18) described the effect of dexamethasone on postton-
umented, the outpatient setting has not been univer- sillectomy nausea and vomiting. All confounding
sally adopted (2-4). There is also no consensus about perioperative factors were controlled except anesthetic
the time for which these children must be observed induction: more patients in the dexamethasone group
after surgery to ensure a safe discharge with minimal received propofol induction (50% vs 40%). Unlike the
risk of postdischarge complications (4,6). The three present study, a decreased incidence of nausea and
most common postdischarge complications that re- vomiting in dexamethasone-treated children was ob-
quire the children to return to the hospital after am- served both during early (PACU) recovery and de-
bulatory tonsillectomy are hemorrhage, persistent layed (24-h) recovery. Because an IV induction with
PONV, and poor oral intake (5,6,8). propofol reduces PONV during early recovery, we
speculate that the lower incidence of PONV reported
To minimize PONV and improve oral intake, anes-
thesiologists have focused primarily on anesthetic during early recovery may be related to the greater
frequency of propofol use in dexamethasone-treated
techniques with minimal emetic potential and on the
children.
administration of different antiemetic drugs (8-12).
The efficacy of dexamethasone in minimizing late
Surgeons, however, have promoted the use of cortico-
PONV in this study and several other published stud-
steroids (methylprednisolone and dexamethasone) to
ies is consistent with its biological half-life of 36-48 h.
minimize morbidity related to postsurgical edema and A prolonged antiemetic effect has also been demon-
inflammation, such as poor oral intake, pain, and fever strated when dexamethasone was used to control
(13,15,16). The efficacy of dexamethasone as an anti- chemotherapy-induced nausea and vomiting (17). The
emetic has been well established in chemotherapy- dose of IV dexamethasone used in the present study
induced nausea and vomiting (17), but studies of its (1 mg/kg, maximal dose 25 mg) was larger than that
antiemetic potential in children undergoing tonsillec- reported in previous studies of patients undergoing
tomy have produced conflicting results (11,15,16,18). tonsillectomy (15,16,18). This dose range has been rec-
PONV is a multifactorial problem, and several anes- ommended for airway obstruction (19). Complications
thetic and nonanesthetic factors must be standardized to from corticosteroid therapy are typically related to its
examine the antiemetic potential of any specific drug. In long-term use, and risks of steroid therapy of <24-h
the present study, the anesthetic technique, amount of IV duration are negligible (20).
hydration, narcotic analgesic dose, and antiemetic ther- The preoperative administration of a single large
apy were standardized. Data from the present study dose of IV dexamethasone significantly decreased the
ANESTH ANALG PEDIATRIC ANESTHESIA PAPPAS ET AL. 61
1998B7t57-61 DEXAMETHASONE, CHILDREN, AND ADENOTONSILLECTOMY
overall incidence of PONV, specifically during the 7. Crysdale WS, Russel D. Complications of tonsillectomy and
adenoidectomy in 9409 children observed overnight. Can Med
24 h after discharge in children undergoing ambula-
Assoc J 1986;135:1139-42.
tory adenotonsillectomy and tonsillectomy. Preopera- 8. Litman RS, Wu CL, Catanzaro FA. Ondansetron decreases eme-
tive dexamethasone administration also improved sis after tonsillectomy in children. Anesth Analg 1994;78:
postoperative oral intake, reduced the number of 478-81.
9. Furst SR, Rodarte A. Prophylactic antiemetic treatment with
phone calls from parents, and prevented hospital re-
ondansetron in children undergoing tonsillectomy. Anesthesi-
turns for the management of PONV and poor oral ology 1994;81:799-803.
intake during the 24 h after discharge. These beneficial 10. Ferrari LR, Donlon JV. Metoclopramide reduces the incidence of
effects of dexamethasone, however, were not evident vomiting after tonsillectomy in children. Anesth Analg 1992;75:
351-4.
during early recovery. We speculate that an antiemetic
11. Ved SA, Walden TL, Montana J, et al. Vomiting and recovery
drug in addition to dexamethasone may minimize after outpatient tonsillectomy and adenoidectomy in children:
early PONV in these children. Further controlled stud- comparison of four anesthetic techniques using nitrous oxide
ies, however, are necessary to verify this practice. with halothane and propofol. Anesthesiology 1996;85:4-10.
12. Pandit UA, Malviya S, Lewis IH. Vomiting after outpatient
tonsillectomy and adenoidectomy in children: the role of nitrous
The authors are grateful to Dr. Stephen Slogoff (Chairman, Depart- oxide. Anesth Analg 1995;80:230-3.
ment of Anesthesiology, Loyola University Medical Center) for his 13. Tom LWC, Templeton JJ, Thomson ME, et al. Dexamethasone in
invaluable help and advice in the preparation of this manuscript. adenotonsillectomy. Int J Pediatr Otorhinolaryngol 1996;37:
115-20.
14. Anderson HA, Rice BJ, Cantrell RW. Effects of injected depos-
teroid on post tonsillectomy morbidity: a double-blind study.
Arch Otolarygnol 1975;101:86-8.
References 15. Catlin FI, Grimes WJ. The effect of steroid therapy on recovery
from tonsillectomy in children. Arch Otolaryngol Head Neck
1. Pratt LW, Gallagher RA. Tonsillectomy and adenoidectomy,
Surg 1991;117:649-52.
incidence and mortality 1968-1972. Otolaryngol Head Neck
16. Volk MS, Martin I’, Brodsky L, et al. The effect of preoperative
Surg 1979;87:159-66.
2. Chiang TM, Sukis AE, Ross DE. Tonsillectomy performed on steroids on tonsillectomy patients. Otolaryngol Head and Neck
outpatient basis. Arch Otolaryngol 1968;88:105-8. Surg 1993;109:726-30.
3. Maniglia AJ, Kushner H, Cozzi L. Adenotonsillectomy: a safe 17. Jones AL, Hill AS, Soukop M, et al. Comparison of dexameth-
outpatient procedure. Arch Otolaryngol Head Neck Surg 1989; asone and ondansetron in the prophylaxis of emesis induced by
115:92-4. moderately emetogenic chemotherapy. Lancet 1991;338:483-7.
4. Colclasure JB, Graham SS. Complications of outpatient tonsil- 18. Splinter WM, Roberts DJ. Dexamethasone decreases vomiting
lectomy and adenoidectomy: a review of 3,340 cases. Ear Nose by children after tonsillectomy. Anesth Analg 1996;83:913-6.
Throat J 1990;69:155-9. 19. Hawkins DB, Crockett DM, Shum TK. Corticosteroids in airway
5. Weimert TA, Babyak JW, Richter HJ. Electrodissection tonsillec- management. Otolaryngol Head Neck Surg 1983;91:593-6.
tomy. Arch Otolaryngol Head Neck Surg 1990;116:186-8. 20. Melby JC. Drug spotlight program: systemic corticosteroid
6. Carithers JS, Gebhart DE, Williams JA. Postoperative risks of therapy-pharmacology and endocrinologic considerations.
pediatric tonsilloadenoidectomy. Laryngoscope 1987;97:422-9. Ann Intern Med 1974;81:505-12.
