OUP user menu

Preventing Opioid-Related Deaths in Children Undergoing Surgery

Senthilkumar Sadhasivam MD, MPH, Charles M. Myer III MD
DOI: http://dx.doi.org/10.1111/j.1526-4637.2012.01419.x 982-983 First published online: 1 July 2012


Tonsillectomy is one of the most commonly done surgical procedures in children. There are many unreported deaths following tonsillectomy in children due to opioids. Genetic variations in liver microenzyme, CYP2D6, had been associated with some of the deaths. Opioids metabolized by CYP2D6 include codeine, tramadol, hydrocodone, and oxycodone. Ultrarapid metabolizers and some extensive metabolizers of CYP2D6 relatively produce more active opioid metabolites resulting in life-threatening adverse effects, especially in young children. Young and obese children with history of sleep apnea are at higher risk of developing serious opioid-related respiratory depression. The adverse outcomes can be avoided and the safety of pain management can be improved by CYP2D6 genetic testing before prescribing these opioids or by using alternative analgesics.

Dear Editor,

Sutters et al. in a small feasibility (N = 47) study of preschool children (3–5 years) conclude that scheduled acetaminophen with hydrocodone every 4 hours around the clock (ATC) for the first 3 days following tonsillectomy is safe and effective in relieving pain at home, though they also observed that time-contingent dosing was associated with moderate to severe side effects and should be addressed in discharge teaching with parents [1].

ATC use of (safe) pain medication following tonsillectomy is an effective way of treating pain following tonsillectomy or other painful surgical procedures. There are many deaths related to opioids in children [2–4], especially with codeine following tonsillectomy [5–8] in young children due to relative overdose and/or P450 microenzyme CYP2D6 ultrarapid metabolizing status. Three recent deaths in children suggest that codeine and potentially other opioids metabolized by the CYP2D6 pathway cannot be considered safe analgesics for young children after tonsillectomy, especially those with sleep apnea [9].

Similar to codeine, hydrocodone is also metabolized by CYP2D6 to hydromorphone [10] and hydromorphone is about five to six times more equipotent than morphine (codeine's active metabolite). Unlike codeine, hydrocodone as a parent drug has some analgesic activity (1/10 of morphine's analgesic efficacy) and would be beneficial in CYP2D6 poor metabolizers; however, in ultrarapid metabolizers, without knowing their CYP2D6 genotype, prescribing ATC hydromorphone in preschool children is not a safe practice. In point of fact, Sutters et al. avoided high-risk patients in the study and they did not analyze three patients that had respiratory complications needing oxygen (5.7%) who got admitted in the hospital following tonsillectomy along with three more patients (5.7%) who had persistent vomiting, facial swelling, and parental concerns. Intent to treat and analyze was not used in more than 10% of their small and convenient sample excluding most of the high-risk patients who possibly had opioid-related adverse effects from the final analysis. The incidence of the reported opioid adverse effects on the day of surgery and postoperative day 1 would have been even worse if they included the patients who had respiratory complications needing oxygen and persistent vomiting.

There are many unreported deaths following tonsillectomy in children and we reviewed the LexisNexis database between 1984 and 2010 for death and injury claims following tonsillectomy [11]. To identify risk factors by category, we specifically looked at anesthesia and opioid-related claims with tonsillectomy (personal communication, Sadhasivam, 2012). To our surprise, 18% of death claims and 5% of injury claims (hypoxic brain injury) were due to opioids. Though, many opioids had been associated with these claims, the opioid most commonly associated with these claims was codeine because of its frequency of use in this population and its unpredictable life-threatening risk in ultrarapid metabolizers even when it is given on as needed basis.

What concerns the authors about the study is not the ATC dosing, but the medication used for ATC dosing, hydrocodone, which goes through the same CYP2D6 pathway, with its active metabolite hydromorphone having 5–6 times more efficacy than morphine. If hydrocodone is used as frequently as codeine, we would see more reports of serious complications, especially in high-risk patients (i.e., young children). It would be safer to do genotype testing (though expensive currently) and avoid ATC hydrocodone in ultrarapid metabolizers to prevent life-threatening complications in an unmonitored home environment. An alternative and less-expensive approach would be to use safer analgesics ATC as we changed our tonsillectomy practice (in children <6 years) more than a year ago with the use of ATC acetaminophen (maximum 75 mg/kg/day), dexamethasone once a day for 3 days and ibuprofen as needed from postoperative day 2 (maximum two doses per day). Despite our high volume, we have not seen any serious or life-threatening complications or an increase in incidence of inadequate pain control or increase in postoperative bleeding.

In our own CYP2D6 genotyping study in children (>6 years) undergoing tonsillectomy, we observed significantly more adverse effects with codeine at home even when it was administered on as needed basis (personal communication, Sadhasivam, 2012). The authors would strongly suggest doing CYP2D6 testing before prescribing ATC codeine, hydrocodone, tramadol, and oxycodone as all are at least partially metabolized by CYP2D6 pathway. Alternatively, using alternative analgesics (i.e., non-opioids and possibly oral morphine with appropriate doses) will lead to safer outcomes following tonsillectomy in young children.

Despite increasing evidence for the clinical value of genetics-guided pharmacotherapy, pharmacogenetic tests are not readily being adopted into clinical practice due to inconsistent evidence, limited access to preoperative genetic testing, and absence of peer-reviewed guidelines to facilitate transitioning from “the bench to the bedside”. In response to these challenges, the Clinical Pharmacogenetics Implementation Consortium (CPIC) was developed to create peer-reviewed, evidence-based, freely accessible guidelines. The CPIC's most recent guideline highlights the importance of genotyping before prescribing codeine and other opioids metabolized by the CYP2D6 [12].

Thank you.



  • Reference: In response to Sutters KA et al.'s article “A Descriptive Feasibility Study to Evaluate Scheduled Oral Analgesics Dosing at Home for the Management of Postoperative Pain in Preschool Children Following Tonsillectomy.” Pain Medicine 2012; 13: 472–483.


View Abstract