ORIGINAL ARTICLE

Pediatric Tonsillectomy and Ketorolac

Lesley D. Phillips-Reed, DNAP, CRNA, Paul N. Austin, PhD, CRNA,
Ricardo E. Rodriguez, PhD

Background: The use of ketorolac in children undergoing tonsillectomy

remains limited because of the concern about postoperative bleeding.
Methods: A search was performed addressing the question: For patients un-
dergoing a surgical tonsillectomy, does a weight-appropriate single dose of
intravenous ketorolac affect the incidence of postoperative hemorrhage?
Results: Five systematic reviews met the inclusion criteria. A Cochrane
Review included 15 studies with 1,101 pediatric subjects and focused
on perioperative bleeding requiring intervention. Many of the systematic
reviews appraised the same studies. Subgroup analysis often allowed
assessment of the effects of ketorolac administration.
Finding: There was no consensus on the increased risk of bleeding when
nonsteriodal anti-inflammatory drugs such as ketorolac are given to
pediatric patients undergoing tonsillectomy. The conclusions varied
from ketorolac should not be used to it is safe to use with these patients.
Conclusions: The perianesthesia team must carefully weigh the risks and
benefits before deciding to use ketorolac with this subset of patients.
Keywords: ketorolac, pediatric, tonsillectomy.
Ó 2016 The American Society of PeriAnesthesia Nurses.

TONSILLECTOMY IS ONE OF THE MOST complications associated with tonsillectomy

commonly performed pediatric surgical proce-
dures in the United States with more than
500,000 operations performed annually. Tonsillec-
tomy is often regarded as a relatively low-risk oper-
ation yielding excellent results in the management
of obstructive tonsillar hypertrophy, peritonsillar
abscess, and recurrent tonsillitis. Unanticipated
Lesley D. Phillips-Reed, DNAP, CRNA, Staff Certified Regis-
tered Nurse Anesthetist, Arkansas Children’s Hospital, North
Little Rock, AR; Paul N. Austin, PhD, CRNA, is Professor, Texas
Wesleyan University, Graduate Programs of Nurse Anes-
thesia, Boyds, MD; and Ricardo E. Rodriguez, PhD, is Profes-
sor, Texas Wesleyan University, Graduate Programs of Nurse
Anesthesia, Doctorate of Nurse Anesthesia Program, Fort
Worth, TX.
Source of grant or financial support: None to report.
Disclosure: The author has no commercial associations
that might pose a conflict of interest in connection with this
work.
Address correspondence to Paul N. Austin, PhD, CRNA,
Graduate Programs of Nurse Anesthesia, Texas Wesleyan Uni-
versity, 1201 Wesleyan St., Ft. Worth, TX 76105; e-mail
address: paustin@txwes.edu.
Ó 2016 The American Society of PeriAnesthesia Nurses.
1089-9472/$36.00
http://dx.doi.org/10.1016/j.jopan.2015.02.005
include postoperative hemorrhage, airway
obstruction, infection, nausea, vomiting, dehydra-
tion, adverse reaction to anesthesia, and unex-
pected admission to the hospital. Postoperative
pain is an expected complication that frequently
occurs and is challenging to manage.1
Postoperative pain is managed in a variety of ways,
including the intraoperative injection of long-
acting local anesthetics and the perioperative use
of acetaminophen. Parenteral and oral opioids
are also commonly used for the management of
pain after a tonsillectomy. Opioids have the poten-
tial to cause respiratory depression, nausea, and
vomiting. Ketorolac is an example of a nonste-
roidal anti-inflammatory drug (NSAID). It is a
potent analgesic that blocks the complex forma-
tion and production of prostaglandins. It has no
sedative or anxiolytic activity and is available in
oral and parenteral forms.2 The use of ketorolac
in the tonsillectomy patient remains limited
because of concerns of increased hemorrhage
postoperatively from its administration during
the perianesthesia period.3

Journal of PeriAnesthesia Nursing, Vol -, No - (-), 2016: pp 1-10 1

2 PHILLIPS-REED, AUSTIN, AND RODRIGUEZ
History
Tonsillectomy has been performed for centuries,
both with and without the benefit of anesthesia.
The earliest reference to a tonsillectomy was re-
corded in Hindu medicine approximately 1,000
BC.4 Breathing obstruction and recurrent throat in-
fections are among the most common indications
for a tonsillectomy. Tonsillectomy involves the
removal of the tonsil and its capsule by dissection
between the tonsillar capsule and the muscular
wall. Post-tonsillectomy bleeding is categorized
as primary or secondary. Primary hemorrhage is
defined as bleeding that occurs within the first
24 hours after surgery and is generally attributed
to the surgical technique1 or a reactionary pro-
cess.5 Incidence of primary hemorrhages range
from 0.2% to 2.2%.1 Secondary hemorrhage occurs
beyond the first postoperative day as a result of
sloughing of the primary eschar at the
tonsil bed.1 Secondary hemorrhage typically
occurs 5 to 10 days after tonsillectomy at a rate
of 0.1% to 3%.1
Postsurgical pain after tonsillectomy is a major
concern for anesthesia providers. Ketorolac is indi-
cated for the short-term management of mild to
moderate pain and can provide analgesia compara-
ble to morphine.6 Ketorolac does not cause
respiratory or central nervous system depression.
When used with an opioid, synergistic opioid-
sparing effects can be appreciated.7 The anti-
inflammatory effects of ketorolac help to decrease
the sensitivity of tissue nociceptors that generally
occur with surgery.7
Ketorolac causes a reversible inhibition of cycloox-
ygenase, which interferes with thromboxane
synthesis, resulting in an antiplatelet effect.
Bleeding time may increase, but it typically re-
mains within the normal value range.7 Anesthesia
providers and surgeons have been reluctant to
use ketorolac in patients undergoing tonsillec-
tomy, arguing that ketorolac causes an increased
potential for postoperative bleeding secondary to
an alteration in the normal clotting mechanism
through inhibition of platelet aggregation.
This evidence-based article examines the use of
ketorolac with pediatric patients undergoing
tonsillectomy.
The PICO Question
The patient, intervention, comparison, and
outcome (PICO) format provides an explicit pro-
cess in evidence-based practice to guide in the
search for the best supporting evidence to address
a problem.8 The PICO question guiding the litera-
ture search for the best current evidence was ‘‘For
patients undergoing a surgical tonsillectomy
(patient), does a weight-appropriate single dose
of intravenous ketorolac (intervention), compared
to no ketorolac (comparison), affect the incidence
of postoperative hemorrhage (outcome)?’’ Evi-
dence was sought including subjects of any age
because of the suspicion of the paucity of studies
including only children and the likelihood of
knowledge transfer from studies including adults.
The Search for Evidence
Search Strategy
The search for evidence was conducted using
PubMed, SUMSearch, Cochrane Central Register
of Controlled Trials, and the Cochrane Database
of Systematic Reviews databases. Sources were
included if published from 1989 to 2014. Specific
search terms, used alone or in combination,
included tonsillectomy, adenotonsillectomy, ketor-
olac, non-steroidal anti-inflammatory, bleeding,
and hemorrhage. Wildcards were used with appro-
priate terms. The search was limited to full-text
systematic reviews (SRs) with and without meta-
analysis and randomized controlled clinical trials
(RCTs) available in full-text and published in
peer-reviewed English language journals.
Web sites of professional (American Association of
Nurse Anesthetists and American Society of Anes-
thesiologists) and governmental (Agency for
Healthcare Research and Quality) organizations
were examined, and discussion of the topic with
subject matter experts was conducted to identify
additional evidence sources. PubMed’s ‘‘related
citations’’ section and the reference lists of all
relevant articles were scrutinized for additional
relevant evidence.
In an effort to locate all available evidence, the
search included sources comparing the incidence
of postoperative hemorrhage after surgical tonsil-
lectomy in persons of any age receiving an NSAID
TONSILLECTOMY AND KETOROLAC
and not just ketorolac. Sources included in
appraised SRs were not separately appraised to
take advantage of higher level evidence.
Critical Appraisal of the Literature
Evidence Sources
The search yielded 88 potential sources (Figure 1).
Twenty-one met inclusion criteria based on exam-
ining the title, 14 based on abstract, and 5 based on
the full text.9-13 These were all SRs9-13 examining
NSAIDS, but subgroup analysis allowed for
assessment of ketorolac administration. An
evaluation of evidence is presented in Table 1 for
the five SRs.9-13 There was no evidence located
that was not included in these five SRs. Evidence
was appraised using the method described by
Melnyk and Fineout-Overholt.14 Randomized
controlled trials included in SRs9-13 were not
appraised separately because SRs with meta-
analysis overcome some of the problems of indi-
vidual studies and are considered stronger
evidence sources in the hierarchy of evidence.16
88 sources identified by
systematic search
21 titles and abstracts
reviewed for relevancy
14 abstracts and full text
manuscripts reviewed for
relevancy
11 full text manuscripts
analyzed for inclusion
criteria
5 sources met inclusion for
analysis
Figure 1. Flowchart summary of evidence examining postoperative hemorrhage in patients undergoing tonsillec-
tomy who received perioperative nonsteroidal anti-inflammatory drugs.
3
A number of the investigations included in the 2013
Cochrane Review9 were also included in four other
SRs with meta-analysis.10-13 The SR by Chan and
Parikh10 included six investigations, another11
included 14 investigations, a third12 included nine
investigations, and a fourth13 included five investiga-
tions. All but one SR10 examined multiple NSAIDs,
including ibuprofen,9,11-13 ketoprofen,9,11,12
9,11,12 9,11,12
diclofenac, tenoxicam, rofecoxib,9,11
11,12
indomethacin, celecoxib, flurbiprofen,11 lor-
11
noxicam, naproxen,12 and nimesulide.12
11
Systematic Reviews
The Cochrane Review9 included 15 RCTs
involving 1,101 children, ages 1 to 16 years, using
a rigorous methodology. The search strategy,
appraising, and scoring methods were compre-
hensive and described in detail. The methodolog-
ical quality of each study was determined by
assessing the risk of bias in random sequence
generation, allocation concealment, blinding of
participants and personnel, blinding of outcome
assessors, incomplete outcome data, selective
67 sources excluded based on
lack of title relevancy
7 studies excluded based on
abstract lack of relevancy, not
meeting inclusion criteria, letters
to the editor
3 studies irrelevant or lack of
comparison of modality
6 sources excluded secondary to
being appraised in systematic
reviews, 1 Cochrane Review
replaced with update
 4
Table 1. Summary of Evidence Examining Postoperative Hemorrhage in Patients Undergoing Tonsillectomy Receiving Perioperative
Nonsteroidal Anti-inflammatory Drugs
Evidence Type
N
Evidence Source Level of Evidence* Outcome Comments
Lewis et al (2013)9 SR of 15 RCTs Outcome 1: reoperation for bleeding—14 Cochrane Review
1,101 subjects studies, 1,044 subjects—Peto OR, 1.69 (95% Large CIs suggesting inadequate
Level I CI, 0.71 to 4.01) sample size
Outcome 2: bleeding, no reoperation—10 Large degree heterogeneity (c2 and I2
studies, 745 subjects—Peto OR, 0.99 (95% statistics)
CI, 0.41 to 2.40) Fixed-effects model used
Conclusion: insufficient evidence to
Ketorolac subgroup:
exclude an increased risk of bleeding
Outcome 1: reoperation for bleeding—Peto
when NSAIDs are used in pediatric
OR, 3.82 (95% CI, 1.03 to 14.10)
tonsillectomy
Five trials, 359 children; total events: nine
ketorolac, one control
Heterogeneity: c22 5 0.56, (P 5 .76);
I2 5 0.0%
Overall effect: Z 5 2.01 (P 5 .044)
Outcome 2: bleeding, no
reoperation—Peto OR, 1.19 (95% CI, 0.45 to

PHILLIPS-REED, AUSTIN, AND RODRIGUEZ

Chan and Parikh (2014)10 SR and meta-analysis of 10
retrospective and prospective
ketorolac RCTs
1,357 subjects
Level I
3.14)
Five trials, 365 children; total events:
12 ketorolac, 7 control
Heterogeneity: c2 5 6.78 (P 5 .03); I2 5 71%
Overall effect: Z 5 0.35 (P 5 .72)
Summary RR for 10 studies 5 2.04 (95% CI,
1.32 to 3.15; P , .002)
Adults only: RR, 5.64 (95% CI, 2.08 to 15.27;
P , .001)
Children only: RR, 1.39 (95% CI, 0.84 to
2.30; P 5 .20)
Ten studies considered together
significantly increased risk of
hemorrhage with ketorolac (P , .001)
Adults: more than five times increased
risk of bleeding
Children: no significantly increased
risk of bleeding
Random-effects model used

Riggin et al (2013)11 SR and meta-analysis of 36 RCTs Received relative weight and OR
3,193 subjects No increased risk of bleeding with NSAIDs
Level I after tonsillectomy
NSAIDs in general population: OR, 1.30
(95% CI, 0.90 to 1.88); P 5 .16 (n 5 36,
N 5 3,193)
NSAIDs in children only: OR, 1.06 (95% CI,
0.65 to 1.74); P 5 .81 (n 5 18, N 5 1,609)
Ketorolac in general population: OR, 2.01
(95% CI, 0.62 to 6.54); P 5 .24 (n 5 8,
N 5 579)
Ketorolac in children only: OR, 3.40 (95% CI,
0.69 to 16.81); P 5 .13 (n 5 6, N 5 459)
Nine studies reported zero outcomes
Møiniche et al (2003)12 Quantitative SR of 25 RCTs Continuous data calculated using WMD. No
1,853 subjects statistically significant difference in intrao-
Level I perative blood loss, postoperative bleeding,
and postoperative admissions (95% CI).
Reoperation (control 0% to 10%, NSAID 0%
to 12.5%) was statistically significant with
Peto method: Peto OR, 2.33 (95% CI, 1.12 to
4.83) and borderline significant with
method described by Shadish and Haddocky
OR, 1.92 (95% CI, 1.00 to 3.71). NNT 5 60
(95% CI, 34 to 277); 1,853 patients, 970
received NSAID
Marret et al (2003)13 SR with meta-analysis of seven RCTs Severe postoperative bleeding: 5.3% con-
505 subjects trols, 9.2% NSAID; OR, 1.8 (95% CI, 0.9 to
Level I 3.4)
TONSILLECTOMY AND KETOROLAC
Significant risk of bias across many
domains for all reports
Conclusion: ketorolac safe for
children but not for adults
1,747 children, 1,446 adults
Only significant adverse NSAID effect
was when given only postoperatively
and can expect 1:20 type I error
Some studies had no bleeding
reported in either arm
Conclusion: NSAIDs safe for children
undergoing tonsillectomy
Fixed-effects model
Low-power meta-analysis
Results ambiguous
Type II errors possible
Significance of results was dependent
on method used
Conclusion: NSAIDs should be used
cautiously in tonsillectomy
Rigorous methodology
Included studies with score of 3 or
more on 1 to 5 quality scalez
(Continued )
5
 6
Table 1. Continued
Evidence Type
N
Evidence Source Level of Evidence* Outcome Comments

Rate of reoperation: 0.8% controls, 4.2% Conclusion: NSAIDs increase

NSAID reoperation risk. Therefore, should
OR, 3.8 (95% CI, 1.3 to 11.5; P 5 .02) not be used after tonsillectomy
NNH 5 29 (95% CI, 17 to 144) Ketorolac 5 151 patients
c2 (k 2 1 degree of freedom) heterogeneity
statistics not statistically significant: severe
postoperative bleeding: c26 5 3.07 (P 5 .7)
Reoperation: c26 5 1.32 (P 5 .95)
SR, systematic review; RCT, randomized controlled trial; OR, odds ratio; CI, confidence interval; NSAID, nonsteroidal anti-inflammatory drug; RR, relative risk;
WMD, weighted mean difference; NNT, number needed to treat; NNH, number needed to harm.
*From Melynk and Fineout-Overholt14: level I: SRs with or without a meta-analysis; level II: well-designed RCTs; level III: well-designed controlled trials without
randomization; level IV: well-designed case-control and cohort studies; level V: SR of descriptive and qualitative studies; level VI: single descriptive or qualitative
study; and level VII: opinion of authorities and/or reports of expert committees.
y
Method described by Shadish and Haddock15: if both cells of a sample are zero, add 0.5 to all cells of that sample.
z
Score of 2 or greater considered significant bleeding (two 5 physician visit/admission; no operative intervention).

PHILLIPS-REED, AUSTIN, AND RODRIGUEZ

TONSILLECTOMY AND KETOROLAC
reporting bias, and absences from other sources of
bias.
Two authors independently assessed trial quality
and extracted the data for the Cochrane Review.9
There were no studies with a low risk of bias in
all domains. Outcome measures included periop-
erative bleeding requiring surgical intervention,
perioperative bleeding requiring nonsurgical
intervention, and vomiting. Nonsurgical interven-
tions included intravenous fluid therapy, pro-
longed observation and nonsurgical hemostasis.
Subgroup analyses included type of NSAID, timing
of administration, and type of control. Six of the 15
studies examined ketorolac as the intervention.
Subgroup comparisons were achievable for ketor-
olac. Outcome 1, perioperative bleeding that
required surgical intervention, was examined in
five studies and involved 359 children. Outcome
2, perioperative bleeding that required nonsur-
gical intervention, was examined in five studies
and involved 365 children.
The SR by Chan and Parikh10 included retrospec-
tive and prospective studies involving children
and adults. Two authors independently identified
studies involving tonsillectomy and ketorolac
administration during the perioperative period.
Ten studies were included, and individual-level
data were extracted for meta-analysis. Risk of bias
was evaluated based on randomization, treatment
allocation, blinding, treatment and outcome stan-
dardization, selective reporting, and complete
data. Analysis revealed a significant risk of bias
for all reports across many domains, especially
treatment allocation and blinding. A total relative
risk (RR) was calculated from the 10 studies, as
well as independently for pediatric and adult
data. The adult data included three studies, two
retrospective and one prospective. Nine pediatric
studies included three retrospective and six pro-
spective. Overall, postoperative hemorrhage rates
from individual studies and pooled subgroups
were reported.
In the SR with meta-analysis by Riggin et al,11 36
studies met inclusion criteria, which included
1,747 children and 1,446 adults. All were
RCTs involving NSAIDs administered to patients
undergoing tonsillectomy; a control group treated
with paracetamol, placebo, or opioids; and
bleeding measured as an outcome. Two authors
7
independently reviewed and sorted studies based
on NSAID type, children versus adults, type and
dose of NSAID, timing and route of administration,
control, surgical technique, and bleeding-related
outcomes. A meta-analysis was performed using a
statistical meta-analysis program in studies in
which the number of individuals with each
outcome was reported. Subgroup analysis was
conducted for subjects receiving ketorolac. To
combine studies measuring different outcomes,
the most severe outcome was selected from each
study. The studies were grouped by specific out-
comes: bleeding requiring reoperation, bleeding
requiring readmission, secondary bleeds, and
bleeding that could be managed with conservative
or no measures. There was statistical homogeneity
among the 36 studies, nine of which had no severe
outcomes.
In the SR by Møiniche et al,12 data from 25 RCTs
including 33 comparisons were included following
recommendations by the Quality of Reporting of
Meta-analyses statement.17 Adult and pediatric
data included 970 subjects receiving NSAIDs and
883 receiving a non-NSAID treatment or placebo.
All trials were of satisfactory methodological qual-
ity, producing a median Oxford score of 4. After
an ante hoc decision, the incidence of reoperation
because of bleeding was the primary outcome mea-
sure. The presence of 0-event trials was identified
as a methodological problem. Zero-event trials
occur with very rare events or when the trials are
of limited size. To overcome this issue, two statisti-
cal models were used—the Peto method18 and the
method described by Shadish and Haddock15—to
calculate odds ratios (ORs) with 95% confidence
intervals (CIs) for rare events.
The SR with meta-analysis of RCTs by Marret et al13
used a very rigid methodology. Included were
seven randomized double-blind pediatric studies.
Each study was subjected to the Jadad Composite
Scale19 in an effort to decrease bias because of
the inclusion of low-quality trials defined as scores
less than 2. All studies were of at least moderate
quality with a Jadad Assessment Scale score of 3
or more. The Mantel-Haenszel20 procedure was
used, which pooled ORs that were assigned
weights proportional to the inverse of the
within-study OR variance. Heterogeneity statistics
such as a c2 (k 2 1 degree of freedom) were
described. For primary evaluation criterion, the
8 PHILLIPS-REED, AUSTIN, AND RODRIGUEZ
number needed to harm—the number of NSAID
subjects required before one more NSAID subject
would experience a harmful outcome—was
computed as the inverse of the difference of the
proportion of patients who required reoperation
in the NSAID and control groups. All tests were
two sided, and P , .05 was considered statistically
significant. The primary outcome measure was the
need for surgical electrocautery to stop bleeding;
the secondary evaluation criterion was postopera-
tive bleeding requiring a change in postoperative
management.
Findings From the Evidence
Major characteristics of the evidence sources9-13
are presented in the table. Collectively, these
accounted for a total of 46 studies, which
included 12 studies examining ketorolac. All
five SRs9-13 included a number of investigations
examining subjects who underwent
tonsillectomy or adenotonsillectomy and who
received ketorolac. Two SRs9,13 focused only
on children and excluded all patients older
than 16 years. Two SRs9,12 used the Peto OR
method for rare and nonoccurring events. The
disadvantage to the Peto method is that
relevant data might be censored, thus allowing
bias to be introduced.
The Cochrane Review9 reported a subgroup anal-
ysis for ketorolac. Five trials involving 359 subjects
compared ketorolac with another analgesic or pla-
cebo and assessed cases of perioperative bleeding
that required further surgical intervention. The
intervention group had an increased risk (Peto
OR, 3.82; 95% CI, 1.03 to 14.10), and the statistical
test for differences between subgroups was not
significant (P 5 .1). Five studies with 365 subjects
assessed ketorolac and perioperative bleeding that
required nonsurgical intervention. The difference
in the effect estimates was smaller (Peto OR,
1.19; 95% CI, 0.45 to 3.14) and again showed no
statistical evidence of a difference between sub-
groups (P 5 .36). Based on results consistent
with an increased and decreased risk of bleeding,
it was concluded that insufficient evidence exists
to exclude an increased risk of bleeding when
NSAIDs are used in pediatric tonsillectomy.
Ten studies examining ketorolac were reviewed by
Chan and Parikh,10 and a summary RR of 2.04
(95% CI, 1.32 to 3.15; P , .002) for post-
tonsillectomy hemorrhage with perioperative ke-
torolac administration was calculated. This finding
implied a significantly increased risk of post-
tonsillectomy hemorrhage associated with ketoro-
lac use (P , .001). When pediatric and adult data
were assessed independently, there was a greatly
increased risk in adults (RR, 5.64; 95% CI, 2.08 to
15.27; P , .001) compared with children (RR,
1.39; 95% CI, 0.84 to 2.30; P 5 .20). From these re-
sults, adults receiving ketorolac had over five times
increased risk of bleeding compared with the
controls. The three adult studies included two
retrospective RCTs and one prospective RCT. All
three studies were consistent in reporting an
increased risk of bleeding with ketorolac.
In the nine pediatric trials included in the SR by
Chan and Parikh,10 the retrospective and prospec-
tive results were consistent. The three retrospec-
tive studies produced an RR of 1.31 (95% CI,
0.66 to 2.59; P 5 .45), and the six prospective
studies yielded an RR of 1.49 (95% CI, 0.71 to
3.13; P 5 .30). The timing of ketorolac administra-
tion did not affect the RR and showed an equiva-
lent risk of postoperative bleeding: preoperative
(RR, 1.43; 95% CI, 0.6 to 3.42; P 5 .43) and intra-
operative or postoperative (RR, 1.37; 95% CI,
0.74 to 2.54; P 5 .32). These results suggest that
there was no significantly increased risk of
bleeding in children.
The SR by Riggin et al11 found no significant
increased risk of post-tonsillectomy bleeding in
adults and children receiving ketorolac. Outcome
data from eight statistically homogenous ketorolac
studies (heterogeneity P 5 .717) were subana-
lyzed, and five reported no adverse outcomes in
either treatment arm. The one significant result
was in the secondary subanalysis of studies in
which subjects were given NSAIDs postopera-
tively only, which revealed an increased risk of
bleeding (OR, 2.02; 95% CI, 1.25 to 3.27;
P 5 .0042). There was no significant bleeding
risk in the preoperative or perioperative analysis.
The SR by Møiniche et al12 examined intraopera-
tive blood loss, postoperative bleeding, and admis-
sion as measures of bleeding along with
reoperation because of bleeding as endpoints
and found that the necessity for reoperation was
0% to 12.5% in the NSAID group compared with
TONSILLECTOMY AND KETOROLAC
0% to 10% in the control group. The OR was statis-
tically significant (Peto OR, 2.33; 95% CI, 1.12 to
4.83) and borderline significant with the method
described by Shadish and Haddock15 (OR, 1.92;
95% CI, 1.00 to 3.71). The number needed to treat
was 60 (95% CI, 34 to 277).
When the NSAID was given postoperatively,
Møiniche et al12 found that 3.2% of the subjects
required a reoperation compared with 0.7% of
the control subjects. This difference was statisti-
cally significant (Peto OR, 4.36; 95% CI, 1.69 to
11.26; OR, 3.27; 95% CI, 1.39 to 7.68), and the
number needed to treat was 40 (95% CI, 23 to
123), meaning that for every 40 subjects who
receive an NSAID dose postoperatively, one sub-
ject will require a reoperation. Of the four bleeding
endpoints used, only reoperation showed a statis-
tically significant result in favor of the non-NSAID
treatment and was dependent on the selection of
the statistical model. Event rates of admission
because of bleeding were very small, 2.5% without
and 5.1% with NSAIDs, and calculations were
based on 506 subjects. A post hoc power analysis
using these numbers and a significance level of
5% revealed a statistical power of approximately
50% only. Both Riggin et al11 and Møiniche et al12
found the timing of NSAID administration to be sig-
nificant.
Marret et al13 reported that of the 243 subjects
who did not receive NSAIDs, 13 subjects had pri-
mary or secondary bleeding (5.3%; range, 0% to
25%), and of the 262 subjects who received
NSAIDs postoperatively, 24 subjects (9.2%; range,
0% to 25%) had postoperative bleeding (OR, 1.8;
95% CI, 0.9 to 3.4). The overall incidences of post-
operative bleeding treated medically or surgically
and of postoperative bleeding requiring reopera-
tion were 7.3% and 2.6%, respectively. A significant
difference in the rate of reoperation was reported
with 0.8% for the controls and 4.2% for the NSAID-
treated subjects (OR, 3.8; 95% CI, 1.3 to 11.5;
P 5 .02) representing a very large increase in the
OR. The number needed to harm was 29
References
1. Baugh RF, Archer SM, Mitchell RB, et al. Clinical practice
guideline: Tonsillectomy in children. Otolaryngol Head Neck
Surg. 2011;144(1 suppl):S1-S30.
9
(95% CI, 17 to 144), which suggests that using
NSAIDs in 29 subjects after tonsillectomy would
be accompanied by a hemorrhage severe enough
to require reoperation of one subject. Six of the
seven trials yielded an OR greater than one for re-
operation for hemostasis, although the differences
were not statistically significant. One of the
included studies suggested an increase in immedi-
ate postoperative bleeding.
Summary
The results differed for bleeding associated with
the administration of an NSAID after tonsillectomy
even between the SRs.9-13 This may be because of
the differences in outcomes measured or to the
heterogeneity of the control groups. Bleeding
requiring further surgical intervention is an
uncommon event after tonsillectomy. The
interpretations from SRs9-13 available to date
yielded incongruent conclusions on the
increased risk of bleeding after tonsillectomy
when NSAIDs were given. The 2013 Cochrane
Review9 concluded that there is insufficient evi-
dence to exclude an increased risk of bleeding
when NSAIDs are used in pediatric tonsillectomy.
The subgroup of children receiving ketorolac had
an increased risk of bleeding. One SR10 concluded
that ketorolac can be used safely in children but is
associated with a fivefold increased bleeding risk
in adults. Another SR11 concluded that NSAIDs
can be considered safe in pediatric tonsillectomy.
The fourth SR12 concluded that NSAIDs should
be used cautiously in tonsillectomy. The final
SR13 concluded that NSAIDs should not be used af-
ter tonsillectomy.
The evidence9-13 offered conflicting conclusions.
Given the danger of bleeding after tonsillectomy,
the perianesthesia team should weigh carefully
the risks and complications of NSAIDs and
consider using alternative analgesics. Future
studies need to be carefully designed and
controlled to avoid the risks of bias identified in
prior efforts.
2. Gillis JC, Brogden RN. Ketorolac. A reappraisal of its phar-
macodynamic and pharmacokinetic properties and therapeutic
use in pain management. Drugs. 1997;53:139-188.
10
3. Judkins JH, Dray TG, Hubbell RN. Intraoperative ketorolac
and posttonsillectomy bleeding. Arch Otolaryngol Head Neck
Surg. 1996;122:937-940.
4. McNeill RA. A history of tonsillectomy: Two millenia of
trauma, haemorrhage and controversy. Ulster Med J. 1960;29:
59-63.
5. Isaacson G. Tonsillectomy healing. Ann Otol Rhinol
Laryngol. 2012;121:645-649.
6. Sinha VR, Kumar RV, Singh G. Ketorolac tromethamine for-
mulations: An overview. Expert Opin Drug Deliv. 2009;6:
961-975.
7. Forrest JB, Heitlinger EL, Revell S. Ketorolac for postop-
erative pain management in children. Drug Saf. 1997;16:
309-329.
8. Richardson WS, Wilson MC, Nishikawa J, Hayward RS. The
well-built clinical question: A key to evidence-based decisions.
ACP J Club. 1995;123:A12-A13.
9. Lewis SR, Nicholson A, Cardwell ME, Siviter G, Smith AF.
Nonsteroidal anti-inflammatory drugs and perioperative
bleeding in paediatric tonsillectomy. Cochrane Database Syst
Rev 2013;CD003591.
10. Chan DK, Parikh SR. Perioperative ketorolac increases
post-tonsillectomy hemorrhage in adults, but not children.
Laryngoscope. 2014;124:1789-1793.
11. Riggin L, Ramakrishna J, Sommer DD, Koren G. A 2013
updated systematic review & meta-analysis of 36 randomized
controlled trials; no apparent effects of non steroidal anti-
inflammatory agents on the risk of bleeding after tonsillectomy.
Clin Otolaryngol. 2013;38:115-129.
12. Møiniche S, Rømsing J, Dahl JB, Tramer MR. Nonsteroidal
antiinflammatory drugs and the risk of operative site bleeding
after tonsillectomy: A quantitative systematic review. Anesth
Analg. 2003;96:68-77.
PHILLIPS-REED, AUSTIN, AND RODRIGUEZ
13. Marret E, Flahault A, Samama CM, Bonnet F. Effects of
postoperative, nonsteroidal, antiinflammatory drugs on
bleeding risk after tonsillectomy: Meta-analysis of randomized,
controlled trials. Anesthesiology. 2003;98:1497-1502.
14. Melnyk BM, Fineout-Overholt E. Making the case for
evidenced-based practice and cultivating a spirit of inquiry. In:
Melnyk BM, Fineout-Overholt E, eds. Evidence-Based Practice
in Nursing and Health Care: A Guide to Best Practice, 2nd
ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011:3-24.
15. Shadish WR, Haddock CK. Combining estimates of effect
size. In: Cooper H, Hedges LV, eds. The Handbook of Research
Synthesis, 1st ed. New York, NY: The Russell Sage Foundation;
1994:261-280.
16. Guyatt G, Furukawa TA. Advanced topics in the validity
of therapy trials: An illustration of bias and random error. In:
Guyatt G, Rennie D, Meade M, Cook D, eds. Users’ Guide to
the Medical Literature, 2nd ed. New York, NY: McGraw-Hill;
2008:109-112.
17. Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D,
Stroup DF. Improving the quality of reports of meta-analyses
of randomised controlled trials: The QUOROM statement. Lan-
cet. 1999;354:1896-1900.
18. Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta
blockade during and after myocardial infarction: An overview
of the randomized trials. Prog Cardiovasc Dis. 1985;27:
335-371.
19. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality
of reports of randomized clinical trials: Is blinding necessary?
Control Clin Trials. 1996;17:1-12.
20. Lachin JM. Power and sample size evaluation for the
Cochran-Mantel-Haenszel mean score (Wilcoxon rank sum)
test and the Cochran-Armitage test for trend. Stat Med. 2011;
30:3057-3066.