- Email: [email protected]
Relationship of Opioid Analgesic Protocols to Assessed Pain and Length of Stay in the Pediatric Postanesthesia Unit Following Tonsillectomy
Relationship of Opioid Analgesic Protocols to Assessed Pain and Length of Stay in the Pediatric Postanesthesia Unit Following Tonsillectomy Jana Smith, ADN, RN, Patricia Newcomb, PhD, RN CPNP, Erin Sundberg, BSN, RN, Paul Shaffer, BSN, RN After tonsillectomy and adenoidectomy in children, postoperative pain management is an essential, yet often challenging, task. In addition to discomfort, lack of pain management can lead to delays in oral intake of patients, resulting in extended stays and increased costs. At one North Texas pediatric facility, postoperative coblation tonsillectomy and adenoidectomy pain management orders include the as-needed use of both intravenous fentanyl and intravenous morphine. Both drugs are effective and both have potential side effects that might prolong the recovery period. Nurses in the postanesthesia care unit retrospectively compared a fentanyl and morphine regimen with a morphine-only regimen to determine whether either protocol made a difference in length-of-stay or pain relief. Analysis of available data revealed no statistically significant differences in length of stay between the groups and trivial differences thought to be clinically irrelevant on other variables. Keywords: children, perianesthesia nursing, analgesia, length-of-stay, pain, opioid, coblation. Ó 2009 by American Society of PeriAnesthesia Nurses.
THE PURPOSE of this retrospective chart review was to evaluate the association between outcomes and administration of two analgesic protocols for postoperative pain management for children after coblation tonsillectomy or tonsillectomy and adenoidectomy (T&A). Specific outcomes of interest included length-of-stay and pain scores in the postanesthesia care unit (PACU).
Jana Smith, ADN, RN, is staff nurse in the postanesthesia unit at Cook Children’s Medical Center, Fort Worth, TX; Patricia Newcomb, PhD, RN CPNP, is an assistant professor in the School of Nursing at University of Texas at Arlington, Arlington, TX; Erin Sundberg, BSN, RN, is a staff nurse in the postanesthesia care unit at Cook Children’s Medical Center, Fort Worth, TX; and Paul Shaffer, BSN, RN, is a staff nurse in the postanesthesia care unit at Cook Children’s Medical Center, Fort Worth, TX. Address correspondence to Patricia Newcomb, School of Nursing, University of Texas at Arlington, 7216 Meadowbrook Dr, Fort Worth, TX 76112; e-mail address: [email protected]. Ó 2009 by American Society of PeriAnesthesia Nurses. 1089-9472/09/2402-0005$36.00/0 doi:10.1016/j.jopan.2009.01.001 86
Background After T&A in children, postoperative (post-op) pain management is an essential, yet often challenging, task. Lack of pain management can lead to delays in children taking oral fluids, a requirement for discharge. Longer stays in the PACU may result in prolonged discomfort for the child, unit crowding, and increased costs; thus, evaluation of interventions to determine those with the greatest potential for effective pain relief during post-op recovery is an ongoing need. Surgical interventions that decrease postoperative tonsillectomy pain include new surgical techniques, such as cold ablation, or coblation, which uses an electrical current passing through a conductive solution (saline) to create a field of sodium ions. As the highly ionized particles are applied to tonsillar tissue, ionic dissociation causes breakdown of molecular bonds within the tissue and results in tissue destruction. Unlike older techniques, coblation generates relatively little heat and little thermal injury. Decreased pain has been demonstrated with Journal of PeriAnesthesia Nursing, Vol 24, No 2 (April), 2009: pp 86-91
ANALGESIC PROTOCOLS IN THE PEDIATRIC PACU
coblation patients,1 but pain still remains one of the major issues in post-op care. The purpose of this study was to determine whether the use of both fentanyl and morphine in post-op coblation patients provided superior pain relief or shorter length-of-stay than the use of morphine alone.
Development of a Clinical Question Nursing observation in the PACU in a pediatric hospital in North Texas during the 2004–2005 period suggested that episodes of airway obstruction sometimes followed the use of fentanyl for pain management. The usual protocol for achieving analgesia included the use of intravenous (IV) fentanyl after extubation, for a total of 4 doses, then IV morphine as needed (PRN); however, nurses also had the option of using morphine only. With the option of using either IV fentanyl or IV morphine, several nurses began using only morphine in preference to fentanyl followed by morphine to avoid airway problems as much as possible. Morphine doses were administered while the patient was still intubated and within 10 minutes of arrival to PACU; then additional doses of morphine were administered as needed after extubation. Postanesthesia care unit staff nurses informally observed that patients who were managed by the morphine-only option seemed to have less postoperative pain and were discharged more quickly. Staff members began to discuss their observations and speculate about the different methods. The institution where the nurses are affiliated strongly endorses evidence-based practice; thus the nursing group sought and received support to investigate their clinical questions regarding the outcomes related to the two analgesic protocols. Through discussion, nurses collectively formed a clinical question addressing the two different methods of pain management: ‘‘Do patients that receive only morphine instead of fentanyl and morphine for post-op T&A pain management have better outcomes (less pain, fewer respiratory adverse effects, and shorter lengths-of-stay) in the PACU?’’ With the help of experienced medical librarians, searches of the MedLine and CINAHL electronic databases were conducted to determine what information was available that might answer the clinical question. Searches revealed many studies of the efficacy and safety of the two drugs, and a single study comparing fentanyl with fentanyl/morphine intraoperatively for pain relief,2 but no studies that examined this specific question. Therefore, it was decided to complete a retrospective chart review. Nurses were able to retrieve data regarding two important outcomes: length of stay and pain scores.
Literature Review Pain is common after a T&A procedure and can prevent a child from consuming adequate fluids postoperatively.
87
Therefore, pain management in pediatric T&A patients is important to relieve the suffering of the child and prevent dehydration, which can result in emergency department visits after discharge. Some practitioners believe that children do not experience or remember pain as adults do. However research shows that the pain experience of children is similar to that of adults.3 Although no evidence was found that specifically compared fentanyl with morphine for pediatric pain management after T&A by coblation, there is evidence that both drugs are safe and effective for children. Morphine is the ‘‘gold standard’’ of opioids.4 Unlike fentanyl, morphine is water soluble with slower onset, peak, and duration. Its onset of action after IV administration is 2–10 minutes and its duration is 2–5 hours, with a peak at about 30 minutes. One of the potential adverse effects of morphine is venodilation sufficient to result in decreased blood pressure or systemic perfusion, especially if intravascular volume is compromised.5 Morphine may also decrease heart rate by direct action on the sinoatrial node.6 However, in a variety of studies from the 1970s to the present, morphine has generally been found safe in respect to changes in hemodynamic parameters in patients without head injury.7-9 The primary adverse effect of morphine is respiratory depression.10 Slow IV administration of morphine decreases the risk of respiratory depression.11 A common side effect of morphine, and a common reason for avoiding it, is nausea and vomiting. Fentanyl is an opioid analgesic commonly used for postoperative pain management in children. Because it is highly lipid soluble, fentanyl has a rapid onset and short duration time.12 When administered intravenously, the onset of action of fentanyl is 3–5 minutes, with duration of 30 minutes to one hour.4 Rapid administration of fentanyl can result in bradycardia but, generally, fentanyl lacks the hemodynamic consequences characteristic of morphine. However, one of the potential adverse effects of fentanyl is chest wall rigidity.10,12 This effect is typically related to the speed of injection and can usually be prevented by administering IV fentanyl slowly.5 Available evidence from a variety of settings and patient populations indicates that fentanyl and morphine are equally effective for postoperative or procedural pain relief.13,14 The decision to use fentanyl in preference to morphine in the pediatric PACU is frequently made on the basis of concerns about adverse effects, such as nausea and vomiting. Morphine is thought to be associated with a greater frequency of postoperative nausea and vomiting than fentanyl. Mukherjee et al2 compared fentanyl with morphine for intraoperative pain management in elective adeno-tonsillectomy in children in the United Kingdom. Intraoperatively, one group was given only fentanyl, and the other group was given morphine and fentanyl. This study showed a lower incidence of vomiting
SMITH ET AL
88
in the first four hours postoperatively in the group given only fentanyl during surgery, but the incidence of vomiting, need for antiemetics, and pain scores 24 hours postoperatively were the same for both groups. An evaluation of postoperative pain, nausea, and vomiting at Bristol Royal Hospital for Children on a small sample demonstrated decreased vomiting with the use of postoperative fentanyl (without antiemetics) and adequate pain relief compared with standard morphine.15 A recent study of pain management in the pediatric PACU after tonsillectomy included morphine with acetaminophen combined with hydrocodone or codeine, and fentanyl with oral acetaminophen combined with hydrocodone or codeine. In this study, the researchers also found no difference in pain relief between the agents.16 In addition to efficacy, timing of opioids for postoperative pain may play a role in postoperative comfort. The Cochrane Database of Systematic Reviews reveals 131 previous studies of preemptive anesthesia in adults and children that suggest that administering analgesia during surgery results in less postoperative pain in the early stages of recovery.17-19 Preemptive analgesia has been described using a variety of drugs and has been administered at various times, including pre-induction, induction, and during surgery. No studies were found describing the preemptive use of opioids before consciousness (specifically during intubation) in the PACU in children. Fentanyl and morphine continue to play important roles in pain management after discharge from the surgical and recovery settings. Romsing et al20 examined the use of acetaminophen in managing pediatric postoperative pain in the outpatient setting and found that acetaminophen does not provide sufficient analgesia after discharge from the surgical setting after a tonsillectomy procedure. Woodhouse et al’s21 comparison of morphine, pethidine (meperidine), and fentanyl used in adult patientcontrolled analgesia demonstrated that, although patients may be intolerant to the particular side effects of one or another of the drugs, all drugs were equally satisfactory for pain relief. In summary, although morphine and fentanyl are associated with concerning adverse effects, they appear to be equally effective for pain relief in a variety of settings and are safe when used appropriately in the pediatric population. Adequately powered and controlled comparisons of fentanyl and morphine in regards to pain relief and length-of-stay, specifically in the pediatric PACU after tonsillectomy or T&A, are lacking, making it difficult to base nursing pain management on evidence in this situation.
Research Questions 1. Is there a difference in nurses’ assessments of pain in children undergoing coblation tonsillectomy or
T&A who receive only morphine for pain management while intubated and thereafter (‘‘morphine only’’) versus those who receive fentanyl after extubation with morphine as needed (‘‘fentanyl first’’)? 2. Is there a difference in length-of-stay in the PACU for children receiving ‘‘morphine only’’ for postoperative pain relief compared with those receiving ‘‘fentanyl first’’? Design and Methods This retrospective chart review compared pain scores and length-of-stay among patients receiving one of two different methods of pain management after tonsillectomy or T&A surgery by coblation. Charts included records of children admitted for day surgery in a tertiary pediatric medical center. A convenience sample of 178 eligible charts was used. The sample included 75 patients who received analgesia protocol 1 (morphine only) and 103 who received protocol 2 (fentanyl first followed by morphine). Charts were chosen consecutively, beginning with the most recent and selecting backwards until 178 charts with sufficient numbers of patients receiving the morphine-only protocol (smaller group) were found. Subjects were limited to children, classified American Society of Anesthesiologits (ASA) Classification I or II, and 7–12 years of age (school age) undergoing tonsillectomy or T&A by coblation. ASA Classification refers to the American Society of Anesthesiologists’ classification of physical status, a measure of physical health, physiologic stability, and coexisting disease.22 The ASA Classification scheme has been shown to predict morbidity23 and is commonly used to identify surgical risk. Children that fall in the classes ASA I and II are basically healthy children, with one or two well-controlled underlying health issues. Children discharged from the PACU back to the operating room (OR) for post-op complications were not included in this study. In addition, children with a body mass index .95th percentile, identified dysmorphic syndromes, developmental delays, craniofacial anomalies, and T&A performed by methods other than coblation were excluded.
Measures Two different scales were used to measure pain. In the immediate post-op period when the patient was not yet capable of verbalizing expressions about his or her pain level, the FLACC Postoperative Pain Scale was used. With this tool, facial expression, leg position, activity, crying, and ability to be consoled were assessed.24,25 When the child was able to verbalize pain, the scale was changed to the Wong-Baker FACES Pain Rating Scale and children were asked if they were hurting or uncomfortable. The Wong-Baker FACES 10-point scale consists of six cartoon faces ranging from smiling face for ‘‘no pain’’ to tearful
ANALGESIC PROTOCOLS IN THE PEDIATRIC PACU
face for ‘‘worst pain.’’26 Postanesthesia care unit nurses in the participating institution routinely use these pain scales to evaluate post-op patients. Correct use of the pain scales is taught in orientation, and nurses must demonstrate competency before use in practice. In the PACU where subjects were recruited, the practice is to transfer the scores from each scale to a single 0–10 scale on the chart so that scores are comparable. The validity of this practice is unknown, but it did provide a means of comparing nurses’ assessment of pain across time in the PACU. Length of PACU stay was defined as the time in hours from admission to the PACU to the point in time that parents state they are comfortable taking the child home after the child met discharge criteria. Discharge criteria included presence of a patent airway, minimum peripheral oxygen saturation $95%, toleration of fluids without nausea or vomiting, and controlled pain documented by a positive response to analgesia. The decision to use parental comfort level following meeting discharge criteria as the discharge time was because the length of time from the point the patient meets discharge criteria to actually obtaining a physician’s order to discharge can vary from 30 minutes to 2 hours depending on unit workload and physician availability. Nausea and vomiting were not included as outcome variables because in almost every case antiemetics such as ondansetron and metaclopromide were administered routinely in the OR and PACU.
Procedures After Institutional Review Board approval, data were extracted from relevant paper and electronic hospital records. Dependent variables obtained from charts included pain scores and length-of-stay in the PACU. Pain scores were recorded at least every 30 minutes and recorded as a value from 0–10 regardless of the pain scale used. Length-of-stay was calculated as the difference between admission time and the time at which parents expressed comfort with taking the child home after meeting discharge criteria. All variables were recorded on the PACU nurses’ notes in flow-chart format. Records of medication administration during the PACU admission were extracted to determine which of two analgesic protocols was used with each patient. Protocol 1 consisted of administering IV morphine as preemptive analgesia upon admission to the PACU before extubation. Subsequent doses of morphine were administered ‘‘as needed’’ after extubation. Protocol 2 consisted of administering fentanyl after extubation, up to a maximum of four doses, followed by IV morphine ‘‘as needed.’’ All medication doses were prescribed based on the child’s weight in kilograms according to standard practice.
89
The decision to administer ‘‘as needed’’ morphine was based on the nurse’s assessment of pain intensity. Nursing assessment of pain intensity was the result of a composite of information including scores on pain scales, the child’s statement that he or she was hurting, and nursing observations of the child’s behavior. The decision to administer an opioid was, in summary, made on the basis of a gestalt that included but was not limited to scores on pain scales. When children indicated they were not experiencing pain, analgesia was not administered. Analgesia protocol 1 was the ‘‘new’’ method in this PACU. In this case, IV opioid was administered in the OR at the discretion of the anesthetist as for protocol two described below. This represents standard care. In the PACU, nursing staff administered morphine (0.025 mg/kg to 0.05 mg/kg) while the child was still intubated, and additional doses of morphine were administered as needed after the child regained consciousness. Analgesia protocol 2 reflected traditional practice in this PACU after tonsillectomy or T&A. This protocol included administering an IV opioid, usually fentanyl 0.5 mcg/kg, in the OR by anesthesia staff, before bringing the intubated child into the PACU. After extubation, nursing staff administered 3–4 titrated doses of fentanyl, up to 2–3 mcg/kg, followed by morphine if needed, titrated up to 0.2–0.3 mg/kg. Figure 1 illustrates the two analgesic protocols.
Results An independent samples t-test was used to determine whether there was a significant difference in length-ofstay between the protocol groups. Significance level was set at 0.05. No difference was found (t 5 –1.17,
Patient admitted to PACU
Protocol 1 “morphine only”
Protocol 2 “fentanyl first”
Morphine administered within 10-15 minutes of admission to PACU while child still intubated
No analgesia administered until child is extubated
Morphine is administered in prescribed doses and at prescribed PRN intervals as needed throughout PACU visit.
Fentanyl dose is administered and repeated at prescribed intervals up to a maximum of 4 doses
Morphine is administered if adequate pain control is not achieved.
Figure 1. Opioid analgesic protocols.
SMITH ET AL
90
P 5 .24, and confidence interval (CI) 5 –17.25 to 4.42). Linear regresson was used to evaluate the contributions of protocol, age, gender, history of asthma or reactive airways disease, and history of obstructive sleep apnea or snoring, and length-of-stay in the PACU. Distributions were normal or nearly normal. None of the targeted independent variables were significant predictors of length-ofstay in the PACU for our school-age patients. Age was the variable closest to significance and, when univariate regression was used to examine the contribution of age to length-of-stay, age was mildly predictive (b 5 –3.404, P 5 .049, and CI 5 –6.688 to –0.02). This would represent a decrease of about 3.5 minutes for each additional year of age from 7–12 years.
first group demonstrated a relatively smoother pattern of pain scores. Mean pain scores at all assessment episodes were #4. As noted earlier, all patients received ondansetron or ondansetron and metaclopromide in the OR before admission to the PACU; therefore, nausea and vomiting were not included as targeted outcomes. However, we observed that nausea and vomiting or nausea alone were reported by only 20 (11%) patients. Eight of the children reporting nausea and/or vomiting were exposed to the fentanyl-first protocol and 12 were exposed to the morphine-only protocol.
Limitations Pain was formally assessed and recorded at 30- to 45minute intervals. Pain scores at the first three assessment episodes are presented in Fig 2. The first formal pain assessments were performed at the time an opioid was first administered. In protocol 1 (morphine only) group the first pain assessment was performed while children were still intubated and the first recorded pain assessment in the fentanyl group was performed when the child was extubated. Regressing opioid protocol on minutes from admission to extubation showed that use of the morphine-only protocol resulted in prolonging the time to extubation by about three minutes on average (b 5 2.9, P 5 .02, and CI 5 .442 to 5.27). Although this was statistically significant, it was not considered clinically significant. Repeated measures analysis of variance showed a significant difference between the groups in mean pain scores at the first assessment. By the third half-hourly pain assessment, at 60–90 minutes after admission, there were no differences in assessed pain between the groups. Mean pain scores of children in the morphine-only group spiked from assessment 1 to 2, whereas children in the fentanyl-
10
pain scores
8
6 3.9 4
4.4 4.1
4 3.9
2.5 2
0 first
second
third
pain assessments pain score protocol 1 pain score protocol 2
Figure 2. Mean pain scores at first three assessments (through 60–90 minutes post admission). This figure is available in color online at www.jopan.org.
Retrospective chart review offers a wealth of relevant clinical data at relatively low cost. Limitations of the method generally include unrecoverable, unrecorded or missing information, as well as difficulties verifying information and varying quality of recorded information. The method can indicate associations or, as in this study, no associations but is not suited to demonstrating causeand-effect relationships. In this case, data regarding one of the primary concerns of nurses—respiratory obstruction—were not obtained, thus one of the major arguments advanced by the PACU staff nurses group against using fentanyl was not explored. This could be explored in future prospective studies of children after T&A. In addition, research team members who abstracted data from charts were not blind to the study objectives. The use of two pain scales makes interpretation more difficult, and limiting tonsillectomy techniques to coblation resulted in a sample that probably experienced relatively little pain as a group.
Conclusion Effective pain management for children after tonsillectomy or T&A is essential to ensure patient comfort and adequate hydration in the PACU and after discharge. In spite of multiple studies in adults and children, controversies continue regarding which analgesics and surgical techniques offer superior pain relief post tonsillectomy. Research suggests that coblation tonsillectomies result in less pain than other techniques, but there is little consensus on choice of postop analgesic agents. Suggested treatments include single opioids, combinations of opioids, combinations of opioids and acetaminophen or nonsteroidal anti-inflammatory drugs, as well as oral acetaminophen or nonsteroidal antiinflammatory drugs alone. Although limited by the retrospective nature of the available data, this study supports other research that finds no difference in the effectiveness of fentanyl and morphine for alleviating pain in children even when, as in this case, morphine is started before the
ANALGESIC PROTOCOLS IN THE PEDIATRIC PACU
91
child’s complete appreciation of pain. Children in the morphine-only group demonstrated lower pain scores on the first recorded assessment, but this may be attributed to the fact that they were relatively more sedated than the children receiving fentanyl after extubation. If there is little or no difference in the effectiveness of fentanyl and morphine for pain management in children post tonsillectomy, choice of drug agent may be based on other considerations, such as length-of-stay. In this study we found that length-of-stay was not shortened by the use of morphine as a single agent given as needed and started before extubation. We observed serendipitously that this sample demonstrated less nausea
and vomiting than reported in previous studies,16,27 likely as a result of receiving antiemetics routinely before admission to the PACU. For PACU staff committed to the use of opioid analgesics after tonsillectomy, there could be other reasons to prefer morphine or fentanyl. High rates of respiratory adverse events or of other side effects associated more with one drug than another would be adequate reason to use one preferentially. In the absence of safety or adverse effects concerns, cost is a good reason to prefer one over the other. Future studies should be performed to investigate the costs in both nursing time and dollars of the analgesic protocols practiced in pediatric postanesthesia settings.
References 1. Mitic S, Tvinnereim M, Lie E, et al. A pilot randomized controlled trial of coblation tonsillectomy versus dissection tonsillectomy with bipolar diathermy haemostasis. Clin Otolaryngol. 2007;32:261-267. 2. Mukherjee K, Esuvaranathan V, Streets C, et al. Adenotonsillectomy in children: A comparison of morphine and fentanyl for peri-operative analgesia. Anaesthesia. 2001;56:1193-1197. 3. Verghese ST, Hannallah RS. Postoperative pain management in children. Anesth Clin North Am. 2005;23:163-184. 4. Curley MAQ, Moloney-Harmon P. Caring practices: Providing comfort. In: Critical Care Nursing of Infants and Children, 2nd ed. Philadelphia: WB Saunders Company; 2001:547-577. 5. Hazinski MF. Sedation, and neuromuscular blockade in pediatric critical care. In: Manual of Pediatric Critical Care. St Louis: Mosby; 1999:44-72. 6. Hsu HO, Hickey RF, Forbes AR. Morphine decreases peripheral vascular resistance and increases capacitance in man. Anesthesiology. 1979; 50:98-102. 7. Larijani GE, Goldberg ME, Gratz I, et al. Analgesic and hemodynamic effects of a single 7.5-mg intravenous dose of morphine in patients with moderate-to-severe postoperative pain. Pharmacotherapy. 2004; 24:1675-1680. 8. Sabatino G, Quartulli L, DiFabio S, et al. Hemodynamic effects of intravenous morphine infusion in ventilated preterm babies. Early Hum Dev. 1997;47:263-270. 9. El Sayed MF, Taddio A, Fallah S, et al. Safety profile of morphine following surgery in neonates. J Perinatol. 2007;27:444-447. 10. Liu LL, Gropper MA. Postoperative anesthesia and sedation in the adult intensive care unit. Drugs. 2003;63:755-767. 11. Taketomo CK, Hodding JH, Kraus DM. Pediatric Dosage Handbook, 13th ed. Hudson, OH: Lexi-Comp; 2006: 971-976. 12. Fahnenstich H, Steffan J, Kau N, et al. Fentanyl-induced chest wall rigidity and laryngospasm in preterm and term infants. Crit Care Med. 2000;28:836-839. 13. Goodarzi M. Comparison of epidural morphine, hydromorphone, and fentanyl for post-operative pain control in children undergoing orthopedic surgery. Paediatr Anaesth. 1999;9:419-422. 14. Finn J, Wright J, Fong J, et al. A randomised crossover trial of patient controlled intranasal fentanyl and oral morphine for procedural wound care in adult patients with burns. Burns. 2004;30:262-268.
15. White MC, Nolan JA. An evaluation of pain and post-operative nausea and vomiting following the introduction of guidelines for tonsillectomy. Paediatr Anaesth. 2005;15:683-688. 16. Pop RS, Manworren RCB, Guzzetta CE, et al. Perianesthesia nurses’ pain management after tonsillectomy and adenoidectomy: Pediatric patient outcomes. J Perianesth Nurs. 2007;22:91-101. 17. Ozkose Z, Akcabay M, Kemaloglu YK, et al. Relief of posttonsillectomy pain with low-dose tramadol given at induction of anesthesia in children. Int J Pediatr Otorhinolaryngol. 2000;53:207-214. 18. Wnek W, Zajaczkowska R, Wordliczek J, et al. Review Group: Cochrane Pain, Palliation, and Supportive Care Group (2004). Infulence of pre-operative ketoprofen administration (prememptive analgesia) on analgesic requirement and the level of prostaglandins in the early postoperative period. Polish J Pharmacol. 2004;56:547-552. 19. Umuroglu T, Zeynep E, Ciftci H, et al. Analgesia for adenotonsillectomy in children: A comparison of morphine, ketamine, and tramadol. Pediatr Anesth. 2004;14:568-573. 20. Romsing J, Hertel S, Harder A, et al. Examination of acetaminophen for outpatient management of postoperative pain in children. Paediatr Anesth. 1998;8:235-239. 21. Woodhouse A, Ward ME, Mather LE. Intrasubject variability in post-operative patient-controlled analgesia (PCA): Is the patient equally satisfied with morphine, pethidine, and fentanyl? Pain. 1999;80: 545-553. 22. American Society of Anesthesiologists. New classification of physical status. Anesthesiology. 1963;24:111. 23. Menke H, Klein A, John KD, et al. Predictive value of ASA classification for the assessment of perioperative risk. Int Surg. 1993;78:266-270. 24. Merkel SI, Voepel-Lewis T, Sheyevitz JR, et al. The FLACC: A behavioral scale for scoring postoperative pain in young children. Pediatr Nurs. 1997;23:293-297. 25. Willis MHW, Merkel SI, Voepel-Lewis T, et al. FLACC behavioral pain assessment scale: A comparison with the child’s self-report. Pediatr Nurs. 2003;29:195-198. 26. Wong DL, Baker C. Pain in children: Comparison of assessment scales. Pediatr Nurs. 1988;14:9-17. 27. Anderson BJ, Ralph CJ, Stewart AW, et al. The dose-effect relationship for morphine and vomiting after day-stay tonsillectomy in children. Anaesth Intensive Care. 2000;28:155-160.
THE PURPOSE of this retrospective chart review was to evaluate the association between outcomes and administration of two analgesic protocols for postoperative pain management for children after coblation tonsillectomy or tonsillectomy and adenoidectomy (T&A). Specific outcomes of interest included length-of-stay and pain scores in the postanesthesia care unit (PACU).
Jana Smith, ADN, RN, is staff nurse in the postanesthesia unit at Cook Children’s Medical Center, Fort Worth, TX; Patricia Newcomb, PhD, RN CPNP, is an assistant professor in the School of Nursing at University of Texas at Arlington, Arlington, TX; Erin Sundberg, BSN, RN, is a staff nurse in the postanesthesia care unit at Cook Children’s Medical Center, Fort Worth, TX; and Paul Shaffer, BSN, RN, is a staff nurse in the postanesthesia care unit at Cook Children’s Medical Center, Fort Worth, TX. Address correspondence to Patricia Newcomb, School of Nursing, University of Texas at Arlington, 7216 Meadowbrook Dr, Fort Worth, TX 76112; e-mail address: [email protected]. Ó 2009 by American Society of PeriAnesthesia Nurses. 1089-9472/09/2402-0005$36.00/0 doi:10.1016/j.jopan.2009.01.001 86
Background After T&A in children, postoperative (post-op) pain management is an essential, yet often challenging, task. Lack of pain management can lead to delays in children taking oral fluids, a requirement for discharge. Longer stays in the PACU may result in prolonged discomfort for the child, unit crowding, and increased costs; thus, evaluation of interventions to determine those with the greatest potential for effective pain relief during post-op recovery is an ongoing need. Surgical interventions that decrease postoperative tonsillectomy pain include new surgical techniques, such as cold ablation, or coblation, which uses an electrical current passing through a conductive solution (saline) to create a field of sodium ions. As the highly ionized particles are applied to tonsillar tissue, ionic dissociation causes breakdown of molecular bonds within the tissue and results in tissue destruction. Unlike older techniques, coblation generates relatively little heat and little thermal injury. Decreased pain has been demonstrated with Journal of PeriAnesthesia Nursing, Vol 24, No 2 (April), 2009: pp 86-91
ANALGESIC PROTOCOLS IN THE PEDIATRIC PACU
coblation patients,1 but pain still remains one of the major issues in post-op care. The purpose of this study was to determine whether the use of both fentanyl and morphine in post-op coblation patients provided superior pain relief or shorter length-of-stay than the use of morphine alone.
Development of a Clinical Question Nursing observation in the PACU in a pediatric hospital in North Texas during the 2004–2005 period suggested that episodes of airway obstruction sometimes followed the use of fentanyl for pain management. The usual protocol for achieving analgesia included the use of intravenous (IV) fentanyl after extubation, for a total of 4 doses, then IV morphine as needed (PRN); however, nurses also had the option of using morphine only. With the option of using either IV fentanyl or IV morphine, several nurses began using only morphine in preference to fentanyl followed by morphine to avoid airway problems as much as possible. Morphine doses were administered while the patient was still intubated and within 10 minutes of arrival to PACU; then additional doses of morphine were administered as needed after extubation. Postanesthesia care unit staff nurses informally observed that patients who were managed by the morphine-only option seemed to have less postoperative pain and were discharged more quickly. Staff members began to discuss their observations and speculate about the different methods. The institution where the nurses are affiliated strongly endorses evidence-based practice; thus the nursing group sought and received support to investigate their clinical questions regarding the outcomes related to the two analgesic protocols. Through discussion, nurses collectively formed a clinical question addressing the two different methods of pain management: ‘‘Do patients that receive only morphine instead of fentanyl and morphine for post-op T&A pain management have better outcomes (less pain, fewer respiratory adverse effects, and shorter lengths-of-stay) in the PACU?’’ With the help of experienced medical librarians, searches of the MedLine and CINAHL electronic databases were conducted to determine what information was available that might answer the clinical question. Searches revealed many studies of the efficacy and safety of the two drugs, and a single study comparing fentanyl with fentanyl/morphine intraoperatively for pain relief,2 but no studies that examined this specific question. Therefore, it was decided to complete a retrospective chart review. Nurses were able to retrieve data regarding two important outcomes: length of stay and pain scores.
Literature Review Pain is common after a T&A procedure and can prevent a child from consuming adequate fluids postoperatively.
87
Therefore, pain management in pediatric T&A patients is important to relieve the suffering of the child and prevent dehydration, which can result in emergency department visits after discharge. Some practitioners believe that children do not experience or remember pain as adults do. However research shows that the pain experience of children is similar to that of adults.3 Although no evidence was found that specifically compared fentanyl with morphine for pediatric pain management after T&A by coblation, there is evidence that both drugs are safe and effective for children. Morphine is the ‘‘gold standard’’ of opioids.4 Unlike fentanyl, morphine is water soluble with slower onset, peak, and duration. Its onset of action after IV administration is 2–10 minutes and its duration is 2–5 hours, with a peak at about 30 minutes. One of the potential adverse effects of morphine is venodilation sufficient to result in decreased blood pressure or systemic perfusion, especially if intravascular volume is compromised.5 Morphine may also decrease heart rate by direct action on the sinoatrial node.6 However, in a variety of studies from the 1970s to the present, morphine has generally been found safe in respect to changes in hemodynamic parameters in patients without head injury.7-9 The primary adverse effect of morphine is respiratory depression.10 Slow IV administration of morphine decreases the risk of respiratory depression.11 A common side effect of morphine, and a common reason for avoiding it, is nausea and vomiting. Fentanyl is an opioid analgesic commonly used for postoperative pain management in children. Because it is highly lipid soluble, fentanyl has a rapid onset and short duration time.12 When administered intravenously, the onset of action of fentanyl is 3–5 minutes, with duration of 30 minutes to one hour.4 Rapid administration of fentanyl can result in bradycardia but, generally, fentanyl lacks the hemodynamic consequences characteristic of morphine. However, one of the potential adverse effects of fentanyl is chest wall rigidity.10,12 This effect is typically related to the speed of injection and can usually be prevented by administering IV fentanyl slowly.5 Available evidence from a variety of settings and patient populations indicates that fentanyl and morphine are equally effective for postoperative or procedural pain relief.13,14 The decision to use fentanyl in preference to morphine in the pediatric PACU is frequently made on the basis of concerns about adverse effects, such as nausea and vomiting. Morphine is thought to be associated with a greater frequency of postoperative nausea and vomiting than fentanyl. Mukherjee et al2 compared fentanyl with morphine for intraoperative pain management in elective adeno-tonsillectomy in children in the United Kingdom. Intraoperatively, one group was given only fentanyl, and the other group was given morphine and fentanyl. This study showed a lower incidence of vomiting
SMITH ET AL
88
in the first four hours postoperatively in the group given only fentanyl during surgery, but the incidence of vomiting, need for antiemetics, and pain scores 24 hours postoperatively were the same for both groups. An evaluation of postoperative pain, nausea, and vomiting at Bristol Royal Hospital for Children on a small sample demonstrated decreased vomiting with the use of postoperative fentanyl (without antiemetics) and adequate pain relief compared with standard morphine.15 A recent study of pain management in the pediatric PACU after tonsillectomy included morphine with acetaminophen combined with hydrocodone or codeine, and fentanyl with oral acetaminophen combined with hydrocodone or codeine. In this study, the researchers also found no difference in pain relief between the agents.16 In addition to efficacy, timing of opioids for postoperative pain may play a role in postoperative comfort. The Cochrane Database of Systematic Reviews reveals 131 previous studies of preemptive anesthesia in adults and children that suggest that administering analgesia during surgery results in less postoperative pain in the early stages of recovery.17-19 Preemptive analgesia has been described using a variety of drugs and has been administered at various times, including pre-induction, induction, and during surgery. No studies were found describing the preemptive use of opioids before consciousness (specifically during intubation) in the PACU in children. Fentanyl and morphine continue to play important roles in pain management after discharge from the surgical and recovery settings. Romsing et al20 examined the use of acetaminophen in managing pediatric postoperative pain in the outpatient setting and found that acetaminophen does not provide sufficient analgesia after discharge from the surgical setting after a tonsillectomy procedure. Woodhouse et al’s21 comparison of morphine, pethidine (meperidine), and fentanyl used in adult patientcontrolled analgesia demonstrated that, although patients may be intolerant to the particular side effects of one or another of the drugs, all drugs were equally satisfactory for pain relief. In summary, although morphine and fentanyl are associated with concerning adverse effects, they appear to be equally effective for pain relief in a variety of settings and are safe when used appropriately in the pediatric population. Adequately powered and controlled comparisons of fentanyl and morphine in regards to pain relief and length-of-stay, specifically in the pediatric PACU after tonsillectomy or T&A, are lacking, making it difficult to base nursing pain management on evidence in this situation.
Research Questions 1. Is there a difference in nurses’ assessments of pain in children undergoing coblation tonsillectomy or
T&A who receive only morphine for pain management while intubated and thereafter (‘‘morphine only’’) versus those who receive fentanyl after extubation with morphine as needed (‘‘fentanyl first’’)? 2. Is there a difference in length-of-stay in the PACU for children receiving ‘‘morphine only’’ for postoperative pain relief compared with those receiving ‘‘fentanyl first’’? Design and Methods This retrospective chart review compared pain scores and length-of-stay among patients receiving one of two different methods of pain management after tonsillectomy or T&A surgery by coblation. Charts included records of children admitted for day surgery in a tertiary pediatric medical center. A convenience sample of 178 eligible charts was used. The sample included 75 patients who received analgesia protocol 1 (morphine only) and 103 who received protocol 2 (fentanyl first followed by morphine). Charts were chosen consecutively, beginning with the most recent and selecting backwards until 178 charts with sufficient numbers of patients receiving the morphine-only protocol (smaller group) were found. Subjects were limited to children, classified American Society of Anesthesiologits (ASA) Classification I or II, and 7–12 years of age (school age) undergoing tonsillectomy or T&A by coblation. ASA Classification refers to the American Society of Anesthesiologists’ classification of physical status, a measure of physical health, physiologic stability, and coexisting disease.22 The ASA Classification scheme has been shown to predict morbidity23 and is commonly used to identify surgical risk. Children that fall in the classes ASA I and II are basically healthy children, with one or two well-controlled underlying health issues. Children discharged from the PACU back to the operating room (OR) for post-op complications were not included in this study. In addition, children with a body mass index .95th percentile, identified dysmorphic syndromes, developmental delays, craniofacial anomalies, and T&A performed by methods other than coblation were excluded.
Measures Two different scales were used to measure pain. In the immediate post-op period when the patient was not yet capable of verbalizing expressions about his or her pain level, the FLACC Postoperative Pain Scale was used. With this tool, facial expression, leg position, activity, crying, and ability to be consoled were assessed.24,25 When the child was able to verbalize pain, the scale was changed to the Wong-Baker FACES Pain Rating Scale and children were asked if they were hurting or uncomfortable. The Wong-Baker FACES 10-point scale consists of six cartoon faces ranging from smiling face for ‘‘no pain’’ to tearful
ANALGESIC PROTOCOLS IN THE PEDIATRIC PACU
face for ‘‘worst pain.’’26 Postanesthesia care unit nurses in the participating institution routinely use these pain scales to evaluate post-op patients. Correct use of the pain scales is taught in orientation, and nurses must demonstrate competency before use in practice. In the PACU where subjects were recruited, the practice is to transfer the scores from each scale to a single 0–10 scale on the chart so that scores are comparable. The validity of this practice is unknown, but it did provide a means of comparing nurses’ assessment of pain across time in the PACU. Length of PACU stay was defined as the time in hours from admission to the PACU to the point in time that parents state they are comfortable taking the child home after the child met discharge criteria. Discharge criteria included presence of a patent airway, minimum peripheral oxygen saturation $95%, toleration of fluids without nausea or vomiting, and controlled pain documented by a positive response to analgesia. The decision to use parental comfort level following meeting discharge criteria as the discharge time was because the length of time from the point the patient meets discharge criteria to actually obtaining a physician’s order to discharge can vary from 30 minutes to 2 hours depending on unit workload and physician availability. Nausea and vomiting were not included as outcome variables because in almost every case antiemetics such as ondansetron and metaclopromide were administered routinely in the OR and PACU.
Procedures After Institutional Review Board approval, data were extracted from relevant paper and electronic hospital records. Dependent variables obtained from charts included pain scores and length-of-stay in the PACU. Pain scores were recorded at least every 30 minutes and recorded as a value from 0–10 regardless of the pain scale used. Length-of-stay was calculated as the difference between admission time and the time at which parents expressed comfort with taking the child home after meeting discharge criteria. All variables were recorded on the PACU nurses’ notes in flow-chart format. Records of medication administration during the PACU admission were extracted to determine which of two analgesic protocols was used with each patient. Protocol 1 consisted of administering IV morphine as preemptive analgesia upon admission to the PACU before extubation. Subsequent doses of morphine were administered ‘‘as needed’’ after extubation. Protocol 2 consisted of administering fentanyl after extubation, up to a maximum of four doses, followed by IV morphine ‘‘as needed.’’ All medication doses were prescribed based on the child’s weight in kilograms according to standard practice.
89
The decision to administer ‘‘as needed’’ morphine was based on the nurse’s assessment of pain intensity. Nursing assessment of pain intensity was the result of a composite of information including scores on pain scales, the child’s statement that he or she was hurting, and nursing observations of the child’s behavior. The decision to administer an opioid was, in summary, made on the basis of a gestalt that included but was not limited to scores on pain scales. When children indicated they were not experiencing pain, analgesia was not administered. Analgesia protocol 1 was the ‘‘new’’ method in this PACU. In this case, IV opioid was administered in the OR at the discretion of the anesthetist as for protocol two described below. This represents standard care. In the PACU, nursing staff administered morphine (0.025 mg/kg to 0.05 mg/kg) while the child was still intubated, and additional doses of morphine were administered as needed after the child regained consciousness. Analgesia protocol 2 reflected traditional practice in this PACU after tonsillectomy or T&A. This protocol included administering an IV opioid, usually fentanyl 0.5 mcg/kg, in the OR by anesthesia staff, before bringing the intubated child into the PACU. After extubation, nursing staff administered 3–4 titrated doses of fentanyl, up to 2–3 mcg/kg, followed by morphine if needed, titrated up to 0.2–0.3 mg/kg. Figure 1 illustrates the two analgesic protocols.
Results An independent samples t-test was used to determine whether there was a significant difference in length-ofstay between the protocol groups. Significance level was set at 0.05. No difference was found (t 5 –1.17,
Patient admitted to PACU
Protocol 1 “morphine only”
Protocol 2 “fentanyl first”
Morphine administered within 10-15 minutes of admission to PACU while child still intubated
No analgesia administered until child is extubated
Morphine is administered in prescribed doses and at prescribed PRN intervals as needed throughout PACU visit.
Fentanyl dose is administered and repeated at prescribed intervals up to a maximum of 4 doses
Morphine is administered if adequate pain control is not achieved.
Figure 1. Opioid analgesic protocols.
SMITH ET AL
90
P 5 .24, and confidence interval (CI) 5 –17.25 to 4.42). Linear regresson was used to evaluate the contributions of protocol, age, gender, history of asthma or reactive airways disease, and history of obstructive sleep apnea or snoring, and length-of-stay in the PACU. Distributions were normal or nearly normal. None of the targeted independent variables were significant predictors of length-ofstay in the PACU for our school-age patients. Age was the variable closest to significance and, when univariate regression was used to examine the contribution of age to length-of-stay, age was mildly predictive (b 5 –3.404, P 5 .049, and CI 5 –6.688 to –0.02). This would represent a decrease of about 3.5 minutes for each additional year of age from 7–12 years.
first group demonstrated a relatively smoother pattern of pain scores. Mean pain scores at all assessment episodes were #4. As noted earlier, all patients received ondansetron or ondansetron and metaclopromide in the OR before admission to the PACU; therefore, nausea and vomiting were not included as targeted outcomes. However, we observed that nausea and vomiting or nausea alone were reported by only 20 (11%) patients. Eight of the children reporting nausea and/or vomiting were exposed to the fentanyl-first protocol and 12 were exposed to the morphine-only protocol.
Limitations Pain was formally assessed and recorded at 30- to 45minute intervals. Pain scores at the first three assessment episodes are presented in Fig 2. The first formal pain assessments were performed at the time an opioid was first administered. In protocol 1 (morphine only) group the first pain assessment was performed while children were still intubated and the first recorded pain assessment in the fentanyl group was performed when the child was extubated. Regressing opioid protocol on minutes from admission to extubation showed that use of the morphine-only protocol resulted in prolonging the time to extubation by about three minutes on average (b 5 2.9, P 5 .02, and CI 5 .442 to 5.27). Although this was statistically significant, it was not considered clinically significant. Repeated measures analysis of variance showed a significant difference between the groups in mean pain scores at the first assessment. By the third half-hourly pain assessment, at 60–90 minutes after admission, there were no differences in assessed pain between the groups. Mean pain scores of children in the morphine-only group spiked from assessment 1 to 2, whereas children in the fentanyl-
10
pain scores
8
6 3.9 4
4.4 4.1
4 3.9
2.5 2
0 first
second
third
pain assessments pain score protocol 1 pain score protocol 2
Figure 2. Mean pain scores at first three assessments (through 60–90 minutes post admission). This figure is available in color online at www.jopan.org.
Retrospective chart review offers a wealth of relevant clinical data at relatively low cost. Limitations of the method generally include unrecoverable, unrecorded or missing information, as well as difficulties verifying information and varying quality of recorded information. The method can indicate associations or, as in this study, no associations but is not suited to demonstrating causeand-effect relationships. In this case, data regarding one of the primary concerns of nurses—respiratory obstruction—were not obtained, thus one of the major arguments advanced by the PACU staff nurses group against using fentanyl was not explored. This could be explored in future prospective studies of children after T&A. In addition, research team members who abstracted data from charts were not blind to the study objectives. The use of two pain scales makes interpretation more difficult, and limiting tonsillectomy techniques to coblation resulted in a sample that probably experienced relatively little pain as a group.
Conclusion Effective pain management for children after tonsillectomy or T&A is essential to ensure patient comfort and adequate hydration in the PACU and after discharge. In spite of multiple studies in adults and children, controversies continue regarding which analgesics and surgical techniques offer superior pain relief post tonsillectomy. Research suggests that coblation tonsillectomies result in less pain than other techniques, but there is little consensus on choice of postop analgesic agents. Suggested treatments include single opioids, combinations of opioids, combinations of opioids and acetaminophen or nonsteroidal anti-inflammatory drugs, as well as oral acetaminophen or nonsteroidal antiinflammatory drugs alone. Although limited by the retrospective nature of the available data, this study supports other research that finds no difference in the effectiveness of fentanyl and morphine for alleviating pain in children even when, as in this case, morphine is started before the
ANALGESIC PROTOCOLS IN THE PEDIATRIC PACU
91
child’s complete appreciation of pain. Children in the morphine-only group demonstrated lower pain scores on the first recorded assessment, but this may be attributed to the fact that they were relatively more sedated than the children receiving fentanyl after extubation. If there is little or no difference in the effectiveness of fentanyl and morphine for pain management in children post tonsillectomy, choice of drug agent may be based on other considerations, such as length-of-stay. In this study we found that length-of-stay was not shortened by the use of morphine as a single agent given as needed and started before extubation. We observed serendipitously that this sample demonstrated less nausea
and vomiting than reported in previous studies,16,27 likely as a result of receiving antiemetics routinely before admission to the PACU. For PACU staff committed to the use of opioid analgesics after tonsillectomy, there could be other reasons to prefer morphine or fentanyl. High rates of respiratory adverse events or of other side effects associated more with one drug than another would be adequate reason to use one preferentially. In the absence of safety or adverse effects concerns, cost is a good reason to prefer one over the other. Future studies should be performed to investigate the costs in both nursing time and dollars of the analgesic protocols practiced in pediatric postanesthesia settings.
References 1. Mitic S, Tvinnereim M, Lie E, et al. A pilot randomized controlled trial of coblation tonsillectomy versus dissection tonsillectomy with bipolar diathermy haemostasis. Clin Otolaryngol. 2007;32:261-267. 2. Mukherjee K, Esuvaranathan V, Streets C, et al. Adenotonsillectomy in children: A comparison of morphine and fentanyl for peri-operative analgesia. Anaesthesia. 2001;56:1193-1197. 3. Verghese ST, Hannallah RS. Postoperative pain management in children. Anesth Clin North Am. 2005;23:163-184. 4. Curley MAQ, Moloney-Harmon P. Caring practices: Providing comfort. In: Critical Care Nursing of Infants and Children, 2nd ed. Philadelphia: WB Saunders Company; 2001:547-577. 5. Hazinski MF. Sedation, and neuromuscular blockade in pediatric critical care. In: Manual of Pediatric Critical Care. St Louis: Mosby; 1999:44-72. 6. Hsu HO, Hickey RF, Forbes AR. Morphine decreases peripheral vascular resistance and increases capacitance in man. Anesthesiology. 1979; 50:98-102. 7. Larijani GE, Goldberg ME, Gratz I, et al. Analgesic and hemodynamic effects of a single 7.5-mg intravenous dose of morphine in patients with moderate-to-severe postoperative pain. Pharmacotherapy. 2004; 24:1675-1680. 8. Sabatino G, Quartulli L, DiFabio S, et al. Hemodynamic effects of intravenous morphine infusion in ventilated preterm babies. Early Hum Dev. 1997;47:263-270. 9. El Sayed MF, Taddio A, Fallah S, et al. Safety profile of morphine following surgery in neonates. J Perinatol. 2007;27:444-447. 10. Liu LL, Gropper MA. Postoperative anesthesia and sedation in the adult intensive care unit. Drugs. 2003;63:755-767. 11. Taketomo CK, Hodding JH, Kraus DM. Pediatric Dosage Handbook, 13th ed. Hudson, OH: Lexi-Comp; 2006: 971-976. 12. Fahnenstich H, Steffan J, Kau N, et al. Fentanyl-induced chest wall rigidity and laryngospasm in preterm and term infants. Crit Care Med. 2000;28:836-839. 13. Goodarzi M. Comparison of epidural morphine, hydromorphone, and fentanyl for post-operative pain control in children undergoing orthopedic surgery. Paediatr Anaesth. 1999;9:419-422. 14. Finn J, Wright J, Fong J, et al. A randomised crossover trial of patient controlled intranasal fentanyl and oral morphine for procedural wound care in adult patients with burns. Burns. 2004;30:262-268.
15. White MC, Nolan JA. An evaluation of pain and post-operative nausea and vomiting following the introduction of guidelines for tonsillectomy. Paediatr Anaesth. 2005;15:683-688. 16. Pop RS, Manworren RCB, Guzzetta CE, et al. Perianesthesia nurses’ pain management after tonsillectomy and adenoidectomy: Pediatric patient outcomes. J Perianesth Nurs. 2007;22:91-101. 17. Ozkose Z, Akcabay M, Kemaloglu YK, et al. Relief of posttonsillectomy pain with low-dose tramadol given at induction of anesthesia in children. Int J Pediatr Otorhinolaryngol. 2000;53:207-214. 18. Wnek W, Zajaczkowska R, Wordliczek J, et al. Review Group: Cochrane Pain, Palliation, and Supportive Care Group (2004). Infulence of pre-operative ketoprofen administration (prememptive analgesia) on analgesic requirement and the level of prostaglandins in the early postoperative period. Polish J Pharmacol. 2004;56:547-552. 19. Umuroglu T, Zeynep E, Ciftci H, et al. Analgesia for adenotonsillectomy in children: A comparison of morphine, ketamine, and tramadol. Pediatr Anesth. 2004;14:568-573. 20. Romsing J, Hertel S, Harder A, et al. Examination of acetaminophen for outpatient management of postoperative pain in children. Paediatr Anesth. 1998;8:235-239. 21. Woodhouse A, Ward ME, Mather LE. Intrasubject variability in post-operative patient-controlled analgesia (PCA): Is the patient equally satisfied with morphine, pethidine, and fentanyl? Pain. 1999;80: 545-553. 22. American Society of Anesthesiologists. New classification of physical status. Anesthesiology. 1963;24:111. 23. Menke H, Klein A, John KD, et al. Predictive value of ASA classification for the assessment of perioperative risk. Int Surg. 1993;78:266-270. 24. Merkel SI, Voepel-Lewis T, Sheyevitz JR, et al. The FLACC: A behavioral scale for scoring postoperative pain in young children. Pediatr Nurs. 1997;23:293-297. 25. Willis MHW, Merkel SI, Voepel-Lewis T, et al. FLACC behavioral pain assessment scale: A comparison with the child’s self-report. Pediatr Nurs. 2003;29:195-198. 26. Wong DL, Baker C. Pain in children: Comparison of assessment scales. Pediatr Nurs. 1988;14:9-17. 27. Anderson BJ, Ralph CJ, Stewart AW, et al. The dose-effect relationship for morphine and vomiting after day-stay tonsillectomy in children. Anaesth Intensive Care. 2000;28:155-160.