- Email: [email protected]
Postoperative nonsteroidal anti-inflammatory drugs and risk of bleeding in pediatric intracapsular tonsillectomy
International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Postoperative nonsteroidal anti-inflammatory drugs and risk of bleeding in pediatric intracapsular tonsillectomy Jill N. D’Souza a, Richard J. Schmidt a,c, Li Xie c, Julie P. Adelman b, Heather C. Nardone a,c,* a
Thomas Jefferson University Hospital, 1020 Walnut St, Philadelphia, PA 19107, United States Temple University, 3401 North Broad St, Philadelphia, PA 19140, United States c Nemours/Alfred I. duPont Hospital for Children, 1600 Rockland Road, Wilmington, DE 19803, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 February 2015 Received in revised form 18 May 2015 Accepted 19 May 2015 Available online 2 July 2015
Objectives: In light of current FDA guidelines on opioid use in children, we sought to determine the risk of post-tonsillectomy hemorrhage (PTH) in children who received ibuprofen with acetaminophen versus those who received narcotic with acetaminophen for postoperative pain control. Methods: This was an IRB-approved retrospective chart review of patients at a tertiary-care pediatric center. The medical records of 449 children who received acetaminophen and ibuprofen following intracapsular tonsillectomy with or without adenoidectomy were reviewed (NSAID group) and compared with medical records of 1731 children who underwent intracapsular tonsillectomy and received acetaminophen with codeine or hydrocodone with acetaminophen postoperatively (narcotic group). Main outcome measure was the incidence of PTH requiring return to the operating room. Secondary outcome measures included incidence of primary PTH, secondary PTH, and postoperative evaluation in the emergency department or readmission for pain and/or dehydration. Results: Incidence of PTH requiring return to the operating room was higher in the NSAID group (1.6%) compared with the narcotic group (0.5%), P = 0.01. Incidence of primary PTH was significantly higher in the NSAID group (2%) versus the narcotic group (0.12%), P < 0.0001. Incidence of secondary PTH was 3.8% in the NSAID group and 1.1% in the narcotic group (P < 0.0001). Conclusion: Use of ibuprofen after intracapsular tonsillectomy in children is associated with statistically significant increase in PTH requiring return to the operating room, as well as an increase in overall rates of both primary and secondary PTH. Ibuprofen provides pain control that is at least equivalent to narcotic and is not associated with respiratory depression. Further study of ibuprofen use in the posttonsillectomy patient is warranted. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tonsillectomy Ibuprofen Post-tonsillectomy hemorrhage Intracapsular
1. Introduction Tonsillectomy is one of the most common surgical procedures performed on pediatric patients in the United States; more than 500,000 surgeries are performed annually [1]. Complications after surgery can include pain, dehydration, and bleeding. Posttonsillectomy hemorrhage (PTH) is one of the most serious complications of tonsillectomy; though rare, PTH can be life
* Corresponding author at: Division of Pediatric Otorhinolaryngology, Nemours/ Alfred I. duPont Hospital for Children, PO Box 269, Wilmington, DE 19899, United States. Tel.: +1 302 651 5895; fax: +302 651 5835. E-mail addresses: [email protected] (J.N. D’Souza), [email protected] (R.J. Schmidt), [email protected] (L. Xie), [email protected] (J.P. Adelman), [email protected] (H.C. Nardone). http://dx.doi.org/10.1016/j.ijporl.2015.05.042 0165-5876/ß 2015 Elsevier Ireland Ltd. All rights reserved.
threatening. Post-tonsillectomy hemorrhage is separated into primary bleeds, which occur within 24 h of surgery, and secondary bleeds, occurring more than 24 h after surgery. The incidence of primary PTH is estimated at 0.2–2.2% and secondary PTH at 0.1–3% [2]. Post-tonsillectomy pain has traditionally been controlled with acetaminophen in combination with a narcotic, such as codeine, hydrocodone, or oxycodone. Deaths from respiratory depression have been reported in children given codeine post tonsillectomy [3]. Many of the deaths have been in children who possess forms of the liver microenzyme CYP2D6, which make them extensive or ultrarapid metabolizers of codeine. For a given dose of codeine, extensive and ultrarapid metabolizers produce relatively high amounts of morphine, the active opioid metabolite, placing them at increased risk for respiratory depression. The FDA has released drug safety warnings regarding pediatric deaths secondary to
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
1473
instructed to use acetaminophen with codeine or hydrocodone as needed for pain (narcotic group). Information on this cohort has been published previously [9]. Technique and equipment for intracapsular tonsillectomy remained consistent among surgeons and over the time periods evaluated. Surgeries in the narcotic group were performed by five attending surgeons, while surgeries in the NSAID group were performed by eight attending surgeons. Of the latter eight surgeons, four were involved in surgeries for both the NSAID and narcotic group. Return to the operating room was dictated by surgeon preference after evaluation of child; generally, medical observation was preferred in the absence of active bleeding and clean tonsillar fossae. If clot or active bleeding was identified, the child was rapidly returned to the operating room. The association between categorical variables was expressed as odds ratios (OR) with 95% confidence intervals. Differences in the relative proportions between categorical variables were analyzed using Fisher exact test and chi-squared analysis, as appropriate. The significance level for all hypothesis tests was set at 0.05.
codeine use after tonsillectomy because of this genetic variation in the CYP2D6 enzyme [4]. The current concerns regarding rapid metabolism of codeine products in children following tonsillectomy [5] have led to widespread reconsideration of post-tonsillectomy pain management regimens for the pediatric population. Nonsteroidal antiinflammatory drugs (NSAIDs), specifically ibuprofen, have gained favor as an alternative to narcotics for postoperative pain control. While NSAIDs provide adequate post-tonsillectomy analgesia [6], their routine use remains controversial as NSAIDs also cause platelet dysfunction that may lead to increased risk of PTH. The clinical impact of this potentially increased bleeding risk with NSAID use remains unclear. Meta-analyses have noted that the postoperative use of NSAIDs may increase the risk of PTH requiring reoperation [7,8]. Given this, one meta-analysis suggests cautious NSAID use [7] while another recommends avoiding NSAID use [8] until further data are available. This study examines whether the use of ibuprofen over a one to two week postoperative period in children following intracapsular tonsillectomy affects the likelihood of clinically significant bleeding requiring return to the operating room. Post-tonsillectomy hemorrhage requiring return to the operating room was selected as primary outcome as we consider it to be the most clinically meaningful complication in children undergoing intracapsular tonsillectomy. Secondarily, the overall incidence of primary and secondary PTH, as well as the incidence of emergency department (ED) evaluation or readmission for pain with or without dehydration is examined.
3. Results There were 449 patients in the NSAID group. Indications included chronic adenotonsillitis in 186 (41.4%), sleep-disordered breathing in 221 (49.2%), or both in 42 (9.4%). Of these, 244 (54.3%) were female, and 205 (45.7%) were male, with a mean age at time of surgery of 9.5 years (SD, 3.2 years). Of 449 patients, 177 (39.4%) received intravenous ketorolac in the immediate postoperative period. There were 1731 patients in the narcotic group. Surgical indications included chronic adenotonsillitis (22.3%), sleep-disordered breathing (69.5%), or both (8.2%). Forty-eight percent were female, and 52% were male, with a mean age at time of surgery of 6 years (SD, 3.4 years). Though the mean patient ages differ between narcotic and NSAID groups, no statistically significant association was found between age at surgery and primary and secondary bleeding (P = 0.18 and P = 0.45, respectively) for the NSAID group. For the narcotic group, no statistically significant association between age and secondary bleeding was found (P = 0.31). Given only two primary bleeding events in the narcotic group, a meaningful statistical analysis regarding primary bleeding and age could not be performed. In the NSAID group, 7 (1.6%) required control of their PTH in the operating room, compared with 8 (0.5%) in the narcotic group. The increased rate of PTH requiring control in the operating room was significantly higher in the NSAID Group (P = 0.01; OR, 3.4; 95% CI, 1.1–10.1). There were nine episodes (2%) of primary PTH in the NSAID group and two episodes (0.12%) of primary PTH in the narcotic group. The incidence of primary PTH was significantly different between groups (P < 0.0001). There were 17 (3.8%) secondary PTHs in the NSAID group and 19 (1.1%) secondary PTHs in the narcotic group, a difference that was statistically significant (P < 0.0001; OR, 3.5; 95% CI, 1.7–7.2) (Table 1).
2. Methods This retrospective study was approved by the Institutional Review Board at Nemours/Alfred I. duPont Hospital for Children (AIDHC). A sample of children who underwent tonsillectomy or adenotonsillectomy for infections or sleep-disordered breathing at AIDHC between 2011 and 2013 was reviewed. Intracapsular tonsillectomy was performed on all children. Intracapsular tonsillectomy technique involves removing approximately 90% of the tonsillar tissue via microdebrider, leaving the capsule intact. Suction electrocautery is then used to ablate the remaining tissue and to obtain hemostasis. Postoperatively, families received verbal and written instruction to use weight-based alternating acetaminophen and ibuprofen as needed for their child’s pain (NSAID group). Data collected included patient age; sex; indication for surgery; use of postoperative ketorolac; occurrence of primary or secondary PTH; evaluation in the ED with or without hospital admission for PTH, pain, or poor oral intake; and need for reoperation because of secondary PTH. Children with known coagulopathies were excluded. The cohort of children who underwent surgery between 2011 and 2013 and used ibuprofen and acetaminophen postoperatively was then compared with a cohort of patients who underwent intracapsular tonsillectomy or adenotonsillectomy at AIDHC between 2002 and 2005. Families in this group were
Table 1 Comparison of outcome measures for narcotic and NSAID groups.
Primary PTH Secondary PTH Secondary PTH requiring postoperative control Evaluation in ED and/or readmission for pain with or without dehydration
Narcotic (n = 1731)
NSAIDs (n = 449)
Odds ratio
95% Confidence interval
P value
3 19 8 57
9 17 7 19
11.78 3.5 3.4 1.3
2.9–55.0 1.7–7.2 1.1–10.1 0. 7–2.3
<0.0001 <0.0001 0.01 0.41
(0.2%) (1.1%) (0.5%) (3.3%)
(2%) (3.8%) (1.6%) (4.2%)
ED, emergency department; NSAID, nonsteroidal anti-inflammatory drug; PTH, post-tonsillectomy hemorrhage.
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
1474
For the NSAID group, chi-squared test performed for surgeons that had at least one patient with primary bleeding shows there is no statistically significant differences between surgeon and primary bleeding (P = 0.85); chi-squared test performed on surgeons that had at least one patient with secondary bleeding shows there is no statistically significant differences between surgeon and secondary bleeding (P = 0.59). Surgeon information was not collected for the narcotic group. In the NSAID group, 161 patients received ketorolac postoperatively for pain control, and 288 did not. When given, ketorolac was administered intravenously, either in the post-anesthesia care unit or, if admitted, on the hospital inpatient floor. There was no ketorolac given preoperatively or intraoperatively. Given its mechanism of action and half-life, administration of ketorolac was considered in relation to the occurrence of primary PTH. Of the patients who received ketorolac, there were four (2.5%) episodes of primary PTH. For those not receiving ketorolac, there were five (1.7%) with primary PTH. Administration of ketorolac was found to have no statistically significant effect on rates of primary (P = 0.73) PTH. In the narcotic group, no patient received ketorolac perioperatively. In the NSAID group, 19 out of 449 (4.2%) patients were evaluated in the ED or readmitted for pain with or without dehydration. In the narcotic group, 57 out of 1731 (3.3%) were evaluated in the ED or readmitted for pain with or without dehydration. The incidence of ED evaluation or readmission for pain with or without dehydration was not statistically significant between groups (P = 0.41) (Table 1). Narcotic and NSAID groups were not similar in terms of indication for surgery. For this reason, a subgroup analysis was
performed according to indication for surgery (sleep-disordered breathing, recurrent tonsillitis, or both). Occurrence of primary and secondary hemorrhage was determined for each subgroup, as was the rate of secondary hemorrhage requiring operative control. For both the narcotic and NSAID groups, no statistically significant difference was found among subgroups in the occurrence of primary hemorrhage, secondary hemorrhage, or in secondary hemorrhage requiring operative control (Tables 2–4). 4. Discussion Tonsillectomy and adenotonsillectomy are frequently performed surgeries in the pediatric population. The postoperative course may be complicated by pain, dehydration, or postoperative hemorrhage. Postoperative pain control in the pediatric population has become an increased topic of discussion with recent FDA warnings regarding increased risk of death in patients taking codeine [3–5,10]. Opioids have been frequently used in this patient population; however, greater understanding of pharmacodynamic and pharmacokinetic differences in individual drug metabolism has revealed a variable response in the pediatric population [5]. Furthermore, a significant proportion of patients undergo tonsillectomy for sleep-disordered breathing such as sleep apnea. It has been suggested that sleep-disordered breathing persists in up to one third of patients after adenotonsillectomy [1]. The use of postoperative opioids has the potential to cause further respiratory depression and distress. Most opioids undergo first-pass metabolism before entering the systemic circulation. This occurs in the liver, where two phases occur. Phase I metabolism promotes oxidation or hydrolysis via the
Table 2 Effect of surgical indication on secondary post-tonsillectomy hemorrhage (PTH). Postoperative pain regimen
NSAID group Narcotic group
Indication Sleep disordered breathing
Chronic tonsillitis
Both
P value
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
221 1373
12 (5.4) 15 (1.1)
186 263
3 (1.6) 4 (1.5)
42 95
2 (4.8) 0
0.13 0.48
Table 3 Effect of surgical indication on secondary post-tonsillectomy hemorrhage (PTH) requiring operative control. Postoperative pain regimen
NSAID group Narcotic group
Indication Sleep disordered breathing
Chronic tonsillitis
Both
P value
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
221 1373
5 (2.3) 5 (0.4)
186 263
0 3 (1.1)
42 95
2 (4.8) 0
0.04 0.19
Table 4 Effect of surgical indication on primary post-tonsillectomy hemorrhage (PTH). Postoperative pain regimen
NSAID group Narcotic group
Indication Sleep disordered breathing
Chronic tonsillitis
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
Both Number of patients
Number (%) with PTH
P value
221 1373
6 (2.7) 1 (0.1)
186 263
3 (1.6) 1 (0.4)
42 95
0 (0) 0 (0)
0.46 0.3
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
CYP450 enzyme, and phase II metabolism conjugates the opioid via glucuronidation to a hydrophilic substance to promote urinary excretion. Codeine undergoes phase I metabolism via the CYP2D6 enzyme. It is a prodrug, and its analgesic properties are linked to the rate it is metabolized into its active form of morphine. Polymorphisms in this gene have given rise to three different rates of metabolism: general metabolism, poor metabolism, and ultrarapid metabolism [5]. General metabolism results in approximately 10% of codeine undergoing activation into morphine. The poor metabolizing form results in almost no morphine being released from the phase I pass of codeine. The ultrarapid metabolizing phenotypes result in production of 50–75% more morphine than the general metabolizers. This high variability in morphine levels has led to fatal outcomes in patients undergoing tonsillectomy and has driven the need to find an agent with adequate pain control that avoids the increased risk of codeine. In addition, opioids are associated with postoperative nausea and vomiting, urinary retention, and constipation. Nonsteroidal anti-inflammatory drugs have been shown to improve postoperative pain in the pediatric population and to decrease postoperative nausea and vomiting [6,7]. Despite these benefits, surgeons have been reluctant to use NSAIDs in the postoperative period because of their potential to increase bleeding rates. Nonsteroidal anti-inflammatory drugs inhibit the cyclooxygenase enzyme, which in turn inhibits prostaglandin production, producing the anti-inflammatory effect associated with pain relief. However, NSAIDS also inhibit platelet aggregation and can cause prolonged bleeding time [11]. Given that PTH can be a life-threatening complication of tonsillectomy, use of NSAIDs postoperatively remains a controversial topic. Several studies in recent years have evaluated the increased risk of bleeding and suggest that there is evidence to support both viewpoints [7,8,12,13]. In a 2003 study by Marret et al., the preoperative use of NSAIDs including ketorolac, ibuprofen, and ketoprofen was found to be associated with increased risk of reoperation for hemostasis [8]. This meta-analysis examined seven prospective studies, with 262 adults and children who received NSAIDs and 243 who received control treatment (placebo or narcotic). For those in the NSAID group, 4.2% required reoperation for primary or secondary bleeding compared with 0.8% in the control group, a difference that was statistically significant (P = 0.02; OR, 3.8; 95% CI, 1.3–11.5). A limitation of this study is that several different NSAID agents were used and that both pediatric and adult populations were studied in combination. A study by Møiniche et al. the same year found that the use of NSAIDs decreases the risk of postoperative nausea and vomiting but found ambiguous evidence that postoperative NSAID use leads to increased rates of postoperative bleeding [7]. The meta-analysis included 13 prospective trials involving both children and adults. Of the 471 who received postoperative NSAIDs, 10.4% had a PTH compared with 8.6% of the 475 who received control therapy, a difference that was not statistically significant. However, the rate of PTH requiring reoperation was higher (3.2%) in the NSAID group compared with the control patients (0.7%), a difference that was statistically significant. Here again, there was a lack of data to evaluate the risk with individual NSAIDs or to evaluate differences between children and adults. A recent systematic review and meta-analysis by Riggin et al. evaluated 18 studies involving children who received diclofenac, ibuprofen, ketoprofen, or ketorolac in various doses and found no significantly increased risk of PTH with NSAID use when various NSAID agents were evaluated separately or in combination [13]. However, when the subgroup of patients receiving NSAIDs in only the postoperative period were analyzed, an increased risk
1475
for PTH was noted. The authors note that additional studies analyzing NSAIDs used only postoperatively are needed. Although retrospective, the current study examines differences solely in a pediatric population who underwent tonsillectomy using a consistent method and for whom parents received standard verbal and written instruction regarding the recommended postoperative pain regimen. In our population, patients receiving postoperative ibuprofen had a statistically significant higher rate of primary and secondary PTH. No statistically significant increase in rate of primary PTH was observed when controlled for ketorolac alone. Of those patients with secondary hemorrhages, 1.6% required return to the operative suite for control. Despite this, our overall rate of PTH remains within the national average, with 1.8% of patients presenting with primary PTH and 3.8% presenting with secondary PTH. The relative increase in PTH must be weighed against the risks of respiratory depression associated with postoperative codeine use, especially in the sleepdisordered breathing population. Furthermore, our study population underwent intracapsular tonsillectomy only. Children that undergo traditional tonsillectomy may be at risk of more severe PTH, as by removing tonsil and capsule in their entirety, larger, extracapsular peritonsillar vessels may be exposed. In the intracapsular technique, a small rind of tonsil remains, and only small caliber, intratonsillar vessels are generally exposed during surgery. A limitation of our study involves variation in the dosage of ibuprofen administered, with some participating physicians prescribing 5 mg/kg/dose and others 10 mg/kg/dose. Though families were instructed to administer weight-based, alternating doses of ibuprofen and tylenol as long as the child experienced postoperative pain, this is a retrospective study, and so the number of NSAID doses actually taken also remains unknown. Both limitations will be addressed in a future prospective study. In addition, the postoperative administration of intravenous ketorolac may confound results as related to the likelihood of a primary PTH to occur. The half-life for ketorolac is approximately 2 h. As ketorolac’s effects on platelet function are reversible and are related to the blood concentration of the medication, the antiplatelet effect should be negligible after 5 to 6 half-lives or 10–12 h. Thus, bleeding attributable to this medication would occur in the immediate postoperative period, and would not impact occurrence of secondary PTH [14]. In this way, ketorolac may confound the results in regards to the likelihood for a primary post-tonsillectomy hemorrhage (PTH) to occur. Though a statistically significant difference in primary PTH was not seen in the current retrospective review, a clinical difference was noted, with 2.5% of those receiving ketorolac experiencing a primary PTH versus 1.7% who did not receive ketorolac. A recent meta-analysis by Chan and Parikh describe a fivefold increased risk of post-tonsillectomy hemorrhage in adults with use of ketorolac, but did not find a corresponding increase in risk in children under the age of 18. They note, however, that it is possible that the various studies included were insufficiently powered, and underestimate risk of PTH with use of ketorolac [15]. Similarly, we believe further investigation with greater numbers of patients is warranted to examine if ketorolac impacts the occurrence of primary PTH. Finally, it is noteworthy that in the narcotic group, children evaluated in the ED for hyperemesis thought to be due to codeine were not considered in the dataset, only those evaluated or readmitted for pain with or without dehydration. As such, it is possible that the number of children in the narcotic group presenting to the ED for nausea and vomiting is underestimated.
1476
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
5. Conclusion The debate on the ideal pain control regimen for children post-tonsillectomy persists. The current study suggests that NSAIDs increased the likelihood of PTH requiring takeback to the operating room. Conversely, NSAIDs appeared to provide adequate pain relief, resulting in at least an equivalent incidence of ED evaluation or readmission for pain with or without dehydration compared with narcotic use and not conferring an increased risk for suppression of respiratory drive. Use of both postoperative NSAIDs and narcotics is associated with risk. Parental education regarding these risks should be undertaken preoperatively. Future studies aim to evaluate whether the number of doses and the amount of ibuprofen given per dose affects risk for PTH and to continue to examine the effects of ketorolac on primary PTH. Conflicts of interest None. Acknowledgments Heather C. Nardone, MD, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Li Xie, ScM, biostatistician at Nemours/Alfred I. duPont Hospital for Children, is responsible for the data analysis. Author contributions: Heather Nardone, MD – Conception and design; acquisition, analysis, or interpretation of data; drafting of manuscript; critical revision of manuscript; supervision. Richard Schmidt, MD – conception and design; acquisition, analysis, or interpretation of data; critical revision of the manuscript for important intellectual content; supervision. Li Xie, ScM – acquisition, analysis, or interpretation of data; drafting of the manuscript; statistical analysis. Jill D’Souza, MD – acquisition, analysis, or interpretation of data, drafting of manuscript, technical support.
Julie Adelman, MD – acquisition, analysis, or interpretation of data; drafting of manuscript; technical support. References [1] R.F. Baugh, S.M. Archer, R.B. Mitchell, R.M. Rosenfeld, R. Amin, J.J. Burns, et al., Clinical practice guideline: tonsillectomy in children, Otolaryngol. Head Neck Surg. 144 (2011) S1–S30. [2] R. Subramanyam, A. Varughese, J.P. Willging, S. Sadhasivam, Future of pediatric tonsillectomy and perioperative outcomes, Int. J. Pediatr. Otorhinolaryngol. 77 (2013) 194–199. [3] C. Ciszkowski, P. Madadi, M.S. Phillips, A.E. Lauwers, G. Koren, Codeine, ultrarapidmetabolism genotype, and postoperative death, N. Engl. J. Med. 361 (2009) 827– 828. [4] U.S. Food and Drug Administration, Codeine Use in Certain Children After Tonsillectomy and/or Adenoidectomy May Lead to Rare, But Life-threatening Adverse Events or Death, FDA Drug Safety Communication, Silver Springs, MD, 2012. [5] J.A. Racoosin, D.W. Roberson, M.A. Pacanowski, D.R. Nielsen, New evidence about an old drug—risk with codeine after adenotonsillectomy, N. Engl. J. Med. 368 (2013) 2155–2157. [6] D. Michelet, J. Andreu-Gallien, T. Bensalah, J. Hilly, C. Wood, Y. Nivoche, et al., A meta-analysis of the use of nonsteroidal antiinflammatory drugs for pediatric postoperative pain, Anesth. Analg. 114 (2012) 393–406. [7] S. Møiniche, J. Rømsing, J.B. Dahl, M.R. Trame`r, Nonsteroidal antiinflammatory drugs and the risk of operative site bleeding after tonsillectomy: a quantitative systematic review, Anesth. Analg. 96 (2003) 68–77. [8] E. Marret, A. Flahault, C.M. Samama, F. Bonnet, Effects of postoperative, nonsteroidal, antiinflammatory drugs on bleeding risk after tonsillectomy: meta-analysis of randomized controlled trials, Anesthesiology 98 (2003) 1497–1502. [9] R. Schmidt, A. Herzog, S. Cook, R. O’Reilly, E. Deutsch, J. Reilly, Complications of tonsillectomy: a comparison of techniques, Arch. Otolaryngol. Head Neck Surg. 133 (2007) 925–928. [10] L.E. Kelly, M. Rieder, J. van den Anker, B. Malkin, C. Ross, M.N. Neely, et al., More codeine fatalities after tonsillectomy in North American children, Pediatrics 129 (2012) e1343–e1347. [11] A.I. Schafer, Effects of nonsteroidal antiinflammatory drugs on platelet function and systemic hemostasis, J. Clin. Pharmacol. 35 (1995) 209–219. [12] H. Yaman, A. Belada, S. Yilmaz, The effect of ibuprofen on postoperative hemorrhage following tonsillectomy in children, Eur. Arch. Otorhinolaryngol. 268 (2011) 615–617. [13] L. Riggin, J. Ramakrishna, D.D. Sommer, G. Koren, 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. 38 (2013) 115–129. [14] R. Dsida, C.J. Cote´, Nonsteroidal anti-inflammatory drugs and hemorrhage following tonsillectomy: do we have the data? Anesthesiology 100 (2004) 749–751. [15] D.K. Chan, S.R. Parikh, Perioperative ketorolac increases post-tonsillectomy hemorrhage in adults but not children, Laryngoscope 124 (2014) 1789–1793.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Postoperative nonsteroidal anti-inflammatory drugs and risk of bleeding in pediatric intracapsular tonsillectomy Jill N. D’Souza a, Richard J. Schmidt a,c, Li Xie c, Julie P. Adelman b, Heather C. Nardone a,c,* a
Thomas Jefferson University Hospital, 1020 Walnut St, Philadelphia, PA 19107, United States Temple University, 3401 North Broad St, Philadelphia, PA 19140, United States c Nemours/Alfred I. duPont Hospital for Children, 1600 Rockland Road, Wilmington, DE 19803, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 February 2015 Received in revised form 18 May 2015 Accepted 19 May 2015 Available online 2 July 2015
Objectives: In light of current FDA guidelines on opioid use in children, we sought to determine the risk of post-tonsillectomy hemorrhage (PTH) in children who received ibuprofen with acetaminophen versus those who received narcotic with acetaminophen for postoperative pain control. Methods: This was an IRB-approved retrospective chart review of patients at a tertiary-care pediatric center. The medical records of 449 children who received acetaminophen and ibuprofen following intracapsular tonsillectomy with or without adenoidectomy were reviewed (NSAID group) and compared with medical records of 1731 children who underwent intracapsular tonsillectomy and received acetaminophen with codeine or hydrocodone with acetaminophen postoperatively (narcotic group). Main outcome measure was the incidence of PTH requiring return to the operating room. Secondary outcome measures included incidence of primary PTH, secondary PTH, and postoperative evaluation in the emergency department or readmission for pain and/or dehydration. Results: Incidence of PTH requiring return to the operating room was higher in the NSAID group (1.6%) compared with the narcotic group (0.5%), P = 0.01. Incidence of primary PTH was significantly higher in the NSAID group (2%) versus the narcotic group (0.12%), P < 0.0001. Incidence of secondary PTH was 3.8% in the NSAID group and 1.1% in the narcotic group (P < 0.0001). Conclusion: Use of ibuprofen after intracapsular tonsillectomy in children is associated with statistically significant increase in PTH requiring return to the operating room, as well as an increase in overall rates of both primary and secondary PTH. Ibuprofen provides pain control that is at least equivalent to narcotic and is not associated with respiratory depression. Further study of ibuprofen use in the posttonsillectomy patient is warranted. ß 2015 Elsevier Ireland Ltd. All rights reserved.
Keywords: Tonsillectomy Ibuprofen Post-tonsillectomy hemorrhage Intracapsular
1. Introduction Tonsillectomy is one of the most common surgical procedures performed on pediatric patients in the United States; more than 500,000 surgeries are performed annually [1]. Complications after surgery can include pain, dehydration, and bleeding. Posttonsillectomy hemorrhage (PTH) is one of the most serious complications of tonsillectomy; though rare, PTH can be life
* Corresponding author at: Division of Pediatric Otorhinolaryngology, Nemours/ Alfred I. duPont Hospital for Children, PO Box 269, Wilmington, DE 19899, United States. Tel.: +1 302 651 5895; fax: +302 651 5835. E-mail addresses: [email protected] (J.N. D’Souza), [email protected] (R.J. Schmidt), [email protected] (L. Xie), [email protected] (J.P. Adelman), [email protected] (H.C. Nardone). http://dx.doi.org/10.1016/j.ijporl.2015.05.042 0165-5876/ß 2015 Elsevier Ireland Ltd. All rights reserved.
threatening. Post-tonsillectomy hemorrhage is separated into primary bleeds, which occur within 24 h of surgery, and secondary bleeds, occurring more than 24 h after surgery. The incidence of primary PTH is estimated at 0.2–2.2% and secondary PTH at 0.1–3% [2]. Post-tonsillectomy pain has traditionally been controlled with acetaminophen in combination with a narcotic, such as codeine, hydrocodone, or oxycodone. Deaths from respiratory depression have been reported in children given codeine post tonsillectomy [3]. Many of the deaths have been in children who possess forms of the liver microenzyme CYP2D6, which make them extensive or ultrarapid metabolizers of codeine. For a given dose of codeine, extensive and ultrarapid metabolizers produce relatively high amounts of morphine, the active opioid metabolite, placing them at increased risk for respiratory depression. The FDA has released drug safety warnings regarding pediatric deaths secondary to
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
1473
instructed to use acetaminophen with codeine or hydrocodone as needed for pain (narcotic group). Information on this cohort has been published previously [9]. Technique and equipment for intracapsular tonsillectomy remained consistent among surgeons and over the time periods evaluated. Surgeries in the narcotic group were performed by five attending surgeons, while surgeries in the NSAID group were performed by eight attending surgeons. Of the latter eight surgeons, four were involved in surgeries for both the NSAID and narcotic group. Return to the operating room was dictated by surgeon preference after evaluation of child; generally, medical observation was preferred in the absence of active bleeding and clean tonsillar fossae. If clot or active bleeding was identified, the child was rapidly returned to the operating room. The association between categorical variables was expressed as odds ratios (OR) with 95% confidence intervals. Differences in the relative proportions between categorical variables were analyzed using Fisher exact test and chi-squared analysis, as appropriate. The significance level for all hypothesis tests was set at 0.05.
codeine use after tonsillectomy because of this genetic variation in the CYP2D6 enzyme [4]. The current concerns regarding rapid metabolism of codeine products in children following tonsillectomy [5] have led to widespread reconsideration of post-tonsillectomy pain management regimens for the pediatric population. Nonsteroidal antiinflammatory drugs (NSAIDs), specifically ibuprofen, have gained favor as an alternative to narcotics for postoperative pain control. While NSAIDs provide adequate post-tonsillectomy analgesia [6], their routine use remains controversial as NSAIDs also cause platelet dysfunction that may lead to increased risk of PTH. The clinical impact of this potentially increased bleeding risk with NSAID use remains unclear. Meta-analyses have noted that the postoperative use of NSAIDs may increase the risk of PTH requiring reoperation [7,8]. Given this, one meta-analysis suggests cautious NSAID use [7] while another recommends avoiding NSAID use [8] until further data are available. This study examines whether the use of ibuprofen over a one to two week postoperative period in children following intracapsular tonsillectomy affects the likelihood of clinically significant bleeding requiring return to the operating room. Post-tonsillectomy hemorrhage requiring return to the operating room was selected as primary outcome as we consider it to be the most clinically meaningful complication in children undergoing intracapsular tonsillectomy. Secondarily, the overall incidence of primary and secondary PTH, as well as the incidence of emergency department (ED) evaluation or readmission for pain with or without dehydration is examined.
3. Results There were 449 patients in the NSAID group. Indications included chronic adenotonsillitis in 186 (41.4%), sleep-disordered breathing in 221 (49.2%), or both in 42 (9.4%). Of these, 244 (54.3%) were female, and 205 (45.7%) were male, with a mean age at time of surgery of 9.5 years (SD, 3.2 years). Of 449 patients, 177 (39.4%) received intravenous ketorolac in the immediate postoperative period. There were 1731 patients in the narcotic group. Surgical indications included chronic adenotonsillitis (22.3%), sleep-disordered breathing (69.5%), or both (8.2%). Forty-eight percent were female, and 52% were male, with a mean age at time of surgery of 6 years (SD, 3.4 years). Though the mean patient ages differ between narcotic and NSAID groups, no statistically significant association was found between age at surgery and primary and secondary bleeding (P = 0.18 and P = 0.45, respectively) for the NSAID group. For the narcotic group, no statistically significant association between age and secondary bleeding was found (P = 0.31). Given only two primary bleeding events in the narcotic group, a meaningful statistical analysis regarding primary bleeding and age could not be performed. In the NSAID group, 7 (1.6%) required control of their PTH in the operating room, compared with 8 (0.5%) in the narcotic group. The increased rate of PTH requiring control in the operating room was significantly higher in the NSAID Group (P = 0.01; OR, 3.4; 95% CI, 1.1–10.1). There were nine episodes (2%) of primary PTH in the NSAID group and two episodes (0.12%) of primary PTH in the narcotic group. The incidence of primary PTH was significantly different between groups (P < 0.0001). There were 17 (3.8%) secondary PTHs in the NSAID group and 19 (1.1%) secondary PTHs in the narcotic group, a difference that was statistically significant (P < 0.0001; OR, 3.5; 95% CI, 1.7–7.2) (Table 1).
2. Methods This retrospective study was approved by the Institutional Review Board at Nemours/Alfred I. duPont Hospital for Children (AIDHC). A sample of children who underwent tonsillectomy or adenotonsillectomy for infections or sleep-disordered breathing at AIDHC between 2011 and 2013 was reviewed. Intracapsular tonsillectomy was performed on all children. Intracapsular tonsillectomy technique involves removing approximately 90% of the tonsillar tissue via microdebrider, leaving the capsule intact. Suction electrocautery is then used to ablate the remaining tissue and to obtain hemostasis. Postoperatively, families received verbal and written instruction to use weight-based alternating acetaminophen and ibuprofen as needed for their child’s pain (NSAID group). Data collected included patient age; sex; indication for surgery; use of postoperative ketorolac; occurrence of primary or secondary PTH; evaluation in the ED with or without hospital admission for PTH, pain, or poor oral intake; and need for reoperation because of secondary PTH. Children with known coagulopathies were excluded. The cohort of children who underwent surgery between 2011 and 2013 and used ibuprofen and acetaminophen postoperatively was then compared with a cohort of patients who underwent intracapsular tonsillectomy or adenotonsillectomy at AIDHC between 2002 and 2005. Families in this group were
Table 1 Comparison of outcome measures for narcotic and NSAID groups.
Primary PTH Secondary PTH Secondary PTH requiring postoperative control Evaluation in ED and/or readmission for pain with or without dehydration
Narcotic (n = 1731)
NSAIDs (n = 449)
Odds ratio
95% Confidence interval
P value
3 19 8 57
9 17 7 19
11.78 3.5 3.4 1.3
2.9–55.0 1.7–7.2 1.1–10.1 0. 7–2.3
<0.0001 <0.0001 0.01 0.41
(0.2%) (1.1%) (0.5%) (3.3%)
(2%) (3.8%) (1.6%) (4.2%)
ED, emergency department; NSAID, nonsteroidal anti-inflammatory drug; PTH, post-tonsillectomy hemorrhage.
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
1474
For the NSAID group, chi-squared test performed for surgeons that had at least one patient with primary bleeding shows there is no statistically significant differences between surgeon and primary bleeding (P = 0.85); chi-squared test performed on surgeons that had at least one patient with secondary bleeding shows there is no statistically significant differences between surgeon and secondary bleeding (P = 0.59). Surgeon information was not collected for the narcotic group. In the NSAID group, 161 patients received ketorolac postoperatively for pain control, and 288 did not. When given, ketorolac was administered intravenously, either in the post-anesthesia care unit or, if admitted, on the hospital inpatient floor. There was no ketorolac given preoperatively or intraoperatively. Given its mechanism of action and half-life, administration of ketorolac was considered in relation to the occurrence of primary PTH. Of the patients who received ketorolac, there were four (2.5%) episodes of primary PTH. For those not receiving ketorolac, there were five (1.7%) with primary PTH. Administration of ketorolac was found to have no statistically significant effect on rates of primary (P = 0.73) PTH. In the narcotic group, no patient received ketorolac perioperatively. In the NSAID group, 19 out of 449 (4.2%) patients were evaluated in the ED or readmitted for pain with or without dehydration. In the narcotic group, 57 out of 1731 (3.3%) were evaluated in the ED or readmitted for pain with or without dehydration. The incidence of ED evaluation or readmission for pain with or without dehydration was not statistically significant between groups (P = 0.41) (Table 1). Narcotic and NSAID groups were not similar in terms of indication for surgery. For this reason, a subgroup analysis was
performed according to indication for surgery (sleep-disordered breathing, recurrent tonsillitis, or both). Occurrence of primary and secondary hemorrhage was determined for each subgroup, as was the rate of secondary hemorrhage requiring operative control. For both the narcotic and NSAID groups, no statistically significant difference was found among subgroups in the occurrence of primary hemorrhage, secondary hemorrhage, or in secondary hemorrhage requiring operative control (Tables 2–4). 4. Discussion Tonsillectomy and adenotonsillectomy are frequently performed surgeries in the pediatric population. The postoperative course may be complicated by pain, dehydration, or postoperative hemorrhage. Postoperative pain control in the pediatric population has become an increased topic of discussion with recent FDA warnings regarding increased risk of death in patients taking codeine [3–5,10]. Opioids have been frequently used in this patient population; however, greater understanding of pharmacodynamic and pharmacokinetic differences in individual drug metabolism has revealed a variable response in the pediatric population [5]. Furthermore, a significant proportion of patients undergo tonsillectomy for sleep-disordered breathing such as sleep apnea. It has been suggested that sleep-disordered breathing persists in up to one third of patients after adenotonsillectomy [1]. The use of postoperative opioids has the potential to cause further respiratory depression and distress. Most opioids undergo first-pass metabolism before entering the systemic circulation. This occurs in the liver, where two phases occur. Phase I metabolism promotes oxidation or hydrolysis via the
Table 2 Effect of surgical indication on secondary post-tonsillectomy hemorrhage (PTH). Postoperative pain regimen
NSAID group Narcotic group
Indication Sleep disordered breathing
Chronic tonsillitis
Both
P value
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
221 1373
12 (5.4) 15 (1.1)
186 263
3 (1.6) 4 (1.5)
42 95
2 (4.8) 0
0.13 0.48
Table 3 Effect of surgical indication on secondary post-tonsillectomy hemorrhage (PTH) requiring operative control. Postoperative pain regimen
NSAID group Narcotic group
Indication Sleep disordered breathing
Chronic tonsillitis
Both
P value
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
221 1373
5 (2.3) 5 (0.4)
186 263
0 3 (1.1)
42 95
2 (4.8) 0
0.04 0.19
Table 4 Effect of surgical indication on primary post-tonsillectomy hemorrhage (PTH). Postoperative pain regimen
NSAID group Narcotic group
Indication Sleep disordered breathing
Chronic tonsillitis
Number of patients
Number (%) with PTH
Number of patients
Number (%) with PTH
Both Number of patients
Number (%) with PTH
P value
221 1373
6 (2.7) 1 (0.1)
186 263
3 (1.6) 1 (0.4)
42 95
0 (0) 0 (0)
0.46 0.3
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
CYP450 enzyme, and phase II metabolism conjugates the opioid via glucuronidation to a hydrophilic substance to promote urinary excretion. Codeine undergoes phase I metabolism via the CYP2D6 enzyme. It is a prodrug, and its analgesic properties are linked to the rate it is metabolized into its active form of morphine. Polymorphisms in this gene have given rise to three different rates of metabolism: general metabolism, poor metabolism, and ultrarapid metabolism [5]. General metabolism results in approximately 10% of codeine undergoing activation into morphine. The poor metabolizing form results in almost no morphine being released from the phase I pass of codeine. The ultrarapid metabolizing phenotypes result in production of 50–75% more morphine than the general metabolizers. This high variability in morphine levels has led to fatal outcomes in patients undergoing tonsillectomy and has driven the need to find an agent with adequate pain control that avoids the increased risk of codeine. In addition, opioids are associated with postoperative nausea and vomiting, urinary retention, and constipation. Nonsteroidal anti-inflammatory drugs have been shown to improve postoperative pain in the pediatric population and to decrease postoperative nausea and vomiting [6,7]. Despite these benefits, surgeons have been reluctant to use NSAIDs in the postoperative period because of their potential to increase bleeding rates. Nonsteroidal anti-inflammatory drugs inhibit the cyclooxygenase enzyme, which in turn inhibits prostaglandin production, producing the anti-inflammatory effect associated with pain relief. However, NSAIDS also inhibit platelet aggregation and can cause prolonged bleeding time [11]. Given that PTH can be a life-threatening complication of tonsillectomy, use of NSAIDs postoperatively remains a controversial topic. Several studies in recent years have evaluated the increased risk of bleeding and suggest that there is evidence to support both viewpoints [7,8,12,13]. In a 2003 study by Marret et al., the preoperative use of NSAIDs including ketorolac, ibuprofen, and ketoprofen was found to be associated with increased risk of reoperation for hemostasis [8]. This meta-analysis examined seven prospective studies, with 262 adults and children who received NSAIDs and 243 who received control treatment (placebo or narcotic). For those in the NSAID group, 4.2% required reoperation for primary or secondary bleeding compared with 0.8% in the control group, a difference that was statistically significant (P = 0.02; OR, 3.8; 95% CI, 1.3–11.5). A limitation of this study is that several different NSAID agents were used and that both pediatric and adult populations were studied in combination. A study by Møiniche et al. the same year found that the use of NSAIDs decreases the risk of postoperative nausea and vomiting but found ambiguous evidence that postoperative NSAID use leads to increased rates of postoperative bleeding [7]. The meta-analysis included 13 prospective trials involving both children and adults. Of the 471 who received postoperative NSAIDs, 10.4% had a PTH compared with 8.6% of the 475 who received control therapy, a difference that was not statistically significant. However, the rate of PTH requiring reoperation was higher (3.2%) in the NSAID group compared with the control patients (0.7%), a difference that was statistically significant. Here again, there was a lack of data to evaluate the risk with individual NSAIDs or to evaluate differences between children and adults. A recent systematic review and meta-analysis by Riggin et al. evaluated 18 studies involving children who received diclofenac, ibuprofen, ketoprofen, or ketorolac in various doses and found no significantly increased risk of PTH with NSAID use when various NSAID agents were evaluated separately or in combination [13]. However, when the subgroup of patients receiving NSAIDs in only the postoperative period were analyzed, an increased risk
1475
for PTH was noted. The authors note that additional studies analyzing NSAIDs used only postoperatively are needed. Although retrospective, the current study examines differences solely in a pediatric population who underwent tonsillectomy using a consistent method and for whom parents received standard verbal and written instruction regarding the recommended postoperative pain regimen. In our population, patients receiving postoperative ibuprofen had a statistically significant higher rate of primary and secondary PTH. No statistically significant increase in rate of primary PTH was observed when controlled for ketorolac alone. Of those patients with secondary hemorrhages, 1.6% required return to the operative suite for control. Despite this, our overall rate of PTH remains within the national average, with 1.8% of patients presenting with primary PTH and 3.8% presenting with secondary PTH. The relative increase in PTH must be weighed against the risks of respiratory depression associated with postoperative codeine use, especially in the sleepdisordered breathing population. Furthermore, our study population underwent intracapsular tonsillectomy only. Children that undergo traditional tonsillectomy may be at risk of more severe PTH, as by removing tonsil and capsule in their entirety, larger, extracapsular peritonsillar vessels may be exposed. In the intracapsular technique, a small rind of tonsil remains, and only small caliber, intratonsillar vessels are generally exposed during surgery. A limitation of our study involves variation in the dosage of ibuprofen administered, with some participating physicians prescribing 5 mg/kg/dose and others 10 mg/kg/dose. Though families were instructed to administer weight-based, alternating doses of ibuprofen and tylenol as long as the child experienced postoperative pain, this is a retrospective study, and so the number of NSAID doses actually taken also remains unknown. Both limitations will be addressed in a future prospective study. In addition, the postoperative administration of intravenous ketorolac may confound results as related to the likelihood of a primary PTH to occur. The half-life for ketorolac is approximately 2 h. As ketorolac’s effects on platelet function are reversible and are related to the blood concentration of the medication, the antiplatelet effect should be negligible after 5 to 6 half-lives or 10–12 h. Thus, bleeding attributable to this medication would occur in the immediate postoperative period, and would not impact occurrence of secondary PTH [14]. In this way, ketorolac may confound the results in regards to the likelihood for a primary post-tonsillectomy hemorrhage (PTH) to occur. Though a statistically significant difference in primary PTH was not seen in the current retrospective review, a clinical difference was noted, with 2.5% of those receiving ketorolac experiencing a primary PTH versus 1.7% who did not receive ketorolac. A recent meta-analysis by Chan and Parikh describe a fivefold increased risk of post-tonsillectomy hemorrhage in adults with use of ketorolac, but did not find a corresponding increase in risk in children under the age of 18. They note, however, that it is possible that the various studies included were insufficiently powered, and underestimate risk of PTH with use of ketorolac [15]. Similarly, we believe further investigation with greater numbers of patients is warranted to examine if ketorolac impacts the occurrence of primary PTH. Finally, it is noteworthy that in the narcotic group, children evaluated in the ED for hyperemesis thought to be due to codeine were not considered in the dataset, only those evaluated or readmitted for pain with or without dehydration. As such, it is possible that the number of children in the narcotic group presenting to the ED for nausea and vomiting is underestimated.
1476
J.N. D’Souza et al. / International Journal of Pediatric Otorhinolaryngology 79 (2015) 1472–1476
5. Conclusion The debate on the ideal pain control regimen for children post-tonsillectomy persists. The current study suggests that NSAIDs increased the likelihood of PTH requiring takeback to the operating room. Conversely, NSAIDs appeared to provide adequate pain relief, resulting in at least an equivalent incidence of ED evaluation or readmission for pain with or without dehydration compared with narcotic use and not conferring an increased risk for suppression of respiratory drive. Use of both postoperative NSAIDs and narcotics is associated with risk. Parental education regarding these risks should be undertaken preoperatively. Future studies aim to evaluate whether the number of doses and the amount of ibuprofen given per dose affects risk for PTH and to continue to examine the effects of ketorolac on primary PTH. Conflicts of interest None. Acknowledgments Heather C. Nardone, MD, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Li Xie, ScM, biostatistician at Nemours/Alfred I. duPont Hospital for Children, is responsible for the data analysis. Author contributions: Heather Nardone, MD – Conception and design; acquisition, analysis, or interpretation of data; drafting of manuscript; critical revision of manuscript; supervision. Richard Schmidt, MD – conception and design; acquisition, analysis, or interpretation of data; critical revision of the manuscript for important intellectual content; supervision. Li Xie, ScM – acquisition, analysis, or interpretation of data; drafting of the manuscript; statistical analysis. Jill D’Souza, MD – acquisition, analysis, or interpretation of data, drafting of manuscript, technical support.
Julie Adelman, MD – acquisition, analysis, or interpretation of data; drafting of manuscript; technical support. References [1] R.F. Baugh, S.M. Archer, R.B. Mitchell, R.M. Rosenfeld, R. Amin, J.J. Burns, et al., Clinical practice guideline: tonsillectomy in children, Otolaryngol. Head Neck Surg. 144 (2011) S1–S30. [2] R. Subramanyam, A. Varughese, J.P. Willging, S. Sadhasivam, Future of pediatric tonsillectomy and perioperative outcomes, Int. J. Pediatr. Otorhinolaryngol. 77 (2013) 194–199. [3] C. Ciszkowski, P. Madadi, M.S. Phillips, A.E. Lauwers, G. Koren, Codeine, ultrarapidmetabolism genotype, and postoperative death, N. Engl. J. Med. 361 (2009) 827– 828. [4] U.S. Food and Drug Administration, Codeine Use in Certain Children After Tonsillectomy and/or Adenoidectomy May Lead to Rare, But Life-threatening Adverse Events or Death, FDA Drug Safety Communication, Silver Springs, MD, 2012. [5] J.A. Racoosin, D.W. Roberson, M.A. Pacanowski, D.R. Nielsen, New evidence about an old drug—risk with codeine after adenotonsillectomy, N. Engl. J. Med. 368 (2013) 2155–2157. [6] D. Michelet, J. Andreu-Gallien, T. Bensalah, J. Hilly, C. Wood, Y. Nivoche, et al., A meta-analysis of the use of nonsteroidal antiinflammatory drugs for pediatric postoperative pain, Anesth. Analg. 114 (2012) 393–406. [7] S. Møiniche, J. Rømsing, J.B. Dahl, M.R. Trame`r, Nonsteroidal antiinflammatory drugs and the risk of operative site bleeding after tonsillectomy: a quantitative systematic review, Anesth. Analg. 96 (2003) 68–77. [8] E. Marret, A. Flahault, C.M. Samama, F. Bonnet, Effects of postoperative, nonsteroidal, antiinflammatory drugs on bleeding risk after tonsillectomy: meta-analysis of randomized controlled trials, Anesthesiology 98 (2003) 1497–1502. [9] R. Schmidt, A. Herzog, S. Cook, R. O’Reilly, E. Deutsch, J. Reilly, Complications of tonsillectomy: a comparison of techniques, Arch. Otolaryngol. Head Neck Surg. 133 (2007) 925–928. [10] L.E. Kelly, M. Rieder, J. van den Anker, B. Malkin, C. Ross, M.N. Neely, et al., More codeine fatalities after tonsillectomy in North American children, Pediatrics 129 (2012) e1343–e1347. [11] A.I. Schafer, Effects of nonsteroidal antiinflammatory drugs on platelet function and systemic hemostasis, J. Clin. Pharmacol. 35 (1995) 209–219. [12] H. Yaman, A. Belada, S. Yilmaz, The effect of ibuprofen on postoperative hemorrhage following tonsillectomy in children, Eur. Arch. Otorhinolaryngol. 268 (2011) 615–617. [13] L. Riggin, J. Ramakrishna, D.D. Sommer, G. Koren, 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. 38 (2013) 115–129. [14] R. Dsida, C.J. Cote´, Nonsteroidal anti-inflammatory drugs and hemorrhage following tonsillectomy: do we have the data? Anesthesiology 100 (2004) 749–751. [15] D.K. Chan, S.R. Parikh, Perioperative ketorolac increases post-tonsillectomy hemorrhage in adults but not children, Laryngoscope 124 (2014) 1789–1793.