Controlled-Release Oxycodone for the Management of Pediatric Postoperative Pain

Vol. 27 No. 4 April 2004 Journal of Pain and Symptom Management 379 Clinical Note Controlled-Release Oxycodone for the Management of Pediatric Postoperative Pain Michelle L. Czarnecki, MSN, RN, CPNP, Matthew D. Jandrisevits, PhD(c), MA, Sarah C. Theiler, PhD(c), MS, Myra Martz Huth, PhD, RN, and Steven J. Weisman, MD Jane B. Pettit Pain and Palliative Care Center Children s Hospital of Wisconsin (M.L.C., M.D.J., S.C.T., M.M.H., S.J.W.); and Medical College of Wisconsin (M.D.J., S.C.T., M.M.H., S.J.W.), Milwaukee, Wisconsin, USA Abstract Studies addressing pain management after pediatric spinal fusion surgery have focused on the use of patient-controlled or epidural analgesia during the immediate postoperative period. Controlled-release (CR) analgesics have been found to be safe and effective in adults. The purpose of this study was to describe the use of oxycodone-cr in pediatric patients after the immediate postoperative period. A retrospective chart review of 62 postoperative spinal fusion patients (10-19 years) was conducted. The mean initial oxycodone-cr dose was 1.24 mg/kg/day. The mean ratio of conversion from parenteral morphine equivalents to oxycodone-cr was 1:1. Mean pain scores decreased from 4.2/10 to 3.7/10 with the transition to oxycodone-cr. Common side effects included dizziness, constipation, and nausea. Oxycodone-CR was used for an average of 13.3 days, which included an average wean time of 6 days. Results of this study demonstrate safe and effective use of oxycodone-cr in the pediatric spinal fusion population. J Pain Symptom Manage 2004;27:379 386. 2004 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Pediatric, spine fusion, pain management, controlled-release analgesics, oxycodone-cr, oxycodone, postoperative pain Introduction Any childhood surgery is a major life event for both the child and family. Children and families verbalize concerns about the surgical procedure, as well as anticipated postoperative Address reprint requests to: Michelle L. Czarnecki, MSN, RN, CPNP, Jane B. Pettit Pain and Palliative Care Center, Children s Hospital of Wisconsin, P.O. Box 1997, MS 792, Milwaukee, WI 53201, USA. Accepted for publication: August 26, 2003. 2004 U.S. Cancer Pain Relief Committee Published by Elsevier Inc. All rights reserved. pain. 1 Spinal fusion surgery is particularly traumatic and the pain experienced afterwards has been described as severe. 2 Following the initial postoperative period, children traditionally have been offered a combination analgesic (opioid/ acetaminophen) for use on an as needed basis. This type of regimen implies that a child must experience pain in order to receive medication. Unfortunately, several studies of children s postoperative pain have revealed that children report severe, unrelenting pain despite analgesics. 3,4 Inadequate analgesia increases patients level of pain, limits their ability to actively participate 0885-3924/04/$ see front matter doi:10.1016/j.jpainsymman.2003.08.007

380 Czarnecki et al. Vol. 27 No. 4 April 2004 in postoperative rehabilitation and activities of daily living, 2 and disrupts their sleep. Uncontrolled pain is also known to impede postoperative functional recovery, 5 and may contribute to a longer hospital stay. 6 A regimen that provides a more consistently delivered analgesic may provide a more consistent level of comfort. In response to these issues, the pain service at Children s Hospital of Wisconsin (CHW) implemented and evaluated a discharge regimen that included the use of controlled-release (CR) analgesics. The goal of this program was to improve ongoing postoperative analgesia as well as decrease the number of times per day a child would need to take medication. Although CR analgesics have been used for chronic pain, they have not commonly been used for acute postoperative pain, nor has their use been reported in children. Oxycodone hydrochloride is a semisynthetic opioid agonist that is available in a fixed combination with acetaminophen or aspirin, as a single-entity preparation, or as a CR preparation. 7 CR and immediate-release (IR) preparations of oxycodone provide equivalent analgesia. 8 Additionally, the literature suggests that oxycodone is as efficacious as morphine, comparable to morphine with respect to side effects, and more potent than oral morphine on a per-milligram basis. 9 In a review of the literature, no studies were found that addressed pain management in children using oral CR analgesics in preparation for discharge from the hospital. Surgical correction for scoliosis is associated with high levels of pain 10 that extend beyond the initial inpatient hospital stay. Adolescents reported moderate to severe pain for up to four days following spinal fusion surgery despite receiving opioid analgesics via patient-controlled analgesia (PCA). 2 One study using epidural analgesia in spinal fusion patients reported an average pain score of 3.6 at rest and 5.1 with movement using a 0 10 scale two days after surgery. 11 Studies have shown that the use of around-theclock CR analgesics is effective and safe for chronic, moderate to severe pain in adults. 12,13 CR opioid preparations offer a reliable means of maintaining more constant serum concentrations than immediate-release formulations. 14 Oxycodone-CR has been shown to be as effective as morphine-cr in the treatment of chronic cancer pain, 9 for the control of moderate to severe pain of more than a few days duration, and in postoperative pain. 15 In addition, Sunshine et al. 8 found oxycodone-cr safe and effective for the management of adult postoperative pain. In a study of oxycodone-cr potency, Curtis et al. 15 found that oral oxycodone-cr was 1.8 times more potent than oral morphine- CR for total effect and 2.2 times more potent for peak effect. The purpose of this study was to assess the safety and efficacy of oxycodone-cr in the management of children s pain following spinal fusion surgery. The following questions were addressed: 1. What is the initial dose ratio of morphine PCA to oxycodone-cr and are different pain scores and side effects associated with the dosage ratio? 2. What are the average pain scores in the 24-hours prior to, 24-hours after, and 48- hours after the transition from PCA to oxycodone-cr? 3. What is the frequency of oxycodone-cr side effects compared with morphine PCA? 4. What are the characteristics of oxycodone- CR use, including length of time until transition to oxycodone-cr, length of time on oxycodone-cr, and time required in weaning patients from oxycodone-cr? Methods The data derived from a retrospective review of charts drawn from a convenience sample of children and adolescents, aged 10 to 19 years, admitted to a large midwestern children s hospital. Inclusion criteria were: 1) first posterior and/or anterior spinal fusion for scoliosis between December 2000 and December 2002; 2) administration of morphine or hydromorphone PCA for initial pain management following surgery; and 3) subsequent use of oxycodone-cr. Procedure After obtaining approval from the institutional review board of CHW and the surgeons who had cared for the children, research assistants were trained in the data collection procedures by the principal investigator. Data were parsed from charts by hand onto a spreadsheet through the use of a data collection checklist. Collected data included demographic characteristics, diagnosis, and type of surgery. Information

Vol. 27 No. 4 April 2004 Management of Pediatric Postoperative Pain 381 regarding oxycodone-cr dosage, duration, and wean were also collected, as was information regarding patient-controlled analgesia use for the 24-hour period prior to treatment with oxycodone-cr. Mean pain scores (based on a verbal 0 10 scale) and side effects were recorded for each of the following periods: a) 24 hours prior to transition; b) the first 24 hours after transition, and c) 24 48 hours after transition to oxycodone-cr (for those patients who remained in the hospital during that time period). The chart review also included records of follow-up telephone calls made to patient families after discharge from the hospital, which provided information about medication use at home. In order to protect the privacy of patients in the study, all names and medical record numbers were kept on a separate list from the data collection tool. After all data were hand-entered onto data collection spreadsheets, data were cleaned through double-checking chart data. Eleven percent of the data spreadsheets were randomly selected and checked for accuracy by the principal investigator. Inter-rater reliability between the principal investigator and research assistants report ranged from 89% to 100%, with a mean of 97% agreement. Initially, the outpatient charts of 92 patients who were followed by the pain service at CHW were reviewed. The investigators then eliminated patients in an attempt to design the most homogeneous sample possible. Specifically, 12 children who had received analgesia other than PCA were removed. Next, 9 patients who were off PCA for more than 8 continuous hours prior to the transition to oxycodone-cr were removed. Eight children were removed because they were treated with a CR analgesic other than oxycodone-cr. Finally, the attending physician of 1 patient had not been asked to release the patient s chart information, so that patient was not included. This left a total sample of 62 patients who were included for data analysis. The general procedure used to convert spinal fusion patients at CHW from PCA to oxycodone-cr starts once patients are able to tolerate clear liquids, have been fitted for their brace (if needed), and have at least been sitting up. The preceding 24-hour parenteral morphine use is calculated and converted 1:1 to oral oxycodone- CR, which is divided into every 12-hour dosing. If a patient is receiving hydromorphone PCA, the 24-hour hydromorphone use is multiplied by 5 to give the morphine equivalents. Patients are given the option to keep the PCA button feature for 24 hours (without the continuous infusion), or to immediately transition to oxycodone-ir for breakthrough pain and discontinue the PCA button as well. Statistical Analyses Data were entered into SPSS for Windows 11.0, cleaned for data entry errors, and analyzed. Descriptive statistics, including means, standard deviations, and percentages were used to assess the research questions for this study. Analysis of variance (ANOVA) was used to compare groups of patients who received differing dosages of oxycodone-cr with respect to their level of morphine PCA dosage prior to the transition to oxycodone-cr. For this statistical analysis, an alpha level of 0.05 was assumed. Results Table 1 shows the demographic characteristics of the included sample. The mean age of participants was 14.6 years (SD 2.1). The majority of participants were female, Caucasian, and had a posterior spinal fusion. No statistically significant differences were found between the children who were excluded and the children who were included with respect to age: F(1,90) 3.61 (ns) and weight: F(1,90) 0.08 (ns). In regard to sex, there were fewer girls than boys excluded from the study than were included (χ 2 (1, n 92) 3.986, P 0.05). The first set of research questions focused on the dosage ratio between the PCA (using Table 1 Characteristics of the Sample (n 62) Demographic % n Sex Female 63 39 Male 37 23 Race African American 11 7 Asian American 2 1 Caucasian 74 46 Hispanic/Latino 3 2 Other 10 6 Type of spinal fusion Anterior 11 7 Posterior 73 45 Anterior & posterior 16 10

382 Czarnecki et al. Vol. 27 No. 4 April 2004 morphine equivalents) and the initial oxycodone-cr dose, and whether differing levels of dosage ratio were associated with patient reports of pain and side effects. Prior to transition to oxycodone-cr, patients in this study received a mean of 1.35 mg/kg/day (SD 0.61) of parenteral morphine equivalents. Patients in the sample then received a mean initial dose of oxycodone-cr of 1.24 mg/kg/day (SD 0.54), with a mean morphine PCA:oxycodone-CR ratio of 1:1 mg/kg/day (SD 0.38), as expected by the conversion procedure. Just over 40% (n 25) of patients had their initial dose adjusted while in the hospital. Twenty-seven percent (n 17) of these adjustments were an increase in dose, and 13% (n 8) of these adjustments were a decrease in dose. The average dose change was 0.12 mg/kg/day (SD 0.49), indicating an increase in oxycodone-cr dosage during the postoperative inpatient recovery period. To investigate the effects associated with PCA:oxycodone-CR dosage ratio level, patient data were grouped according to quartile values. The three groups included patients with morphine PCA:oxycodone-CR mg/kg/day ratio values up to the 25th percentile (1:0.74), above the 75th percentile (1:1.29), and those with a value between the two. Analysis of variance (ANOVA) facilitated an examination of whether patients in the three ratio groups differed with respect to two variables of patient experience during the 24-hour period following transition to oxycodone-cr. These variables included mean pain scores and mean number of different types of side effects (Table 2). ANOVA results indicated that patients in the three ratio groups did not differ with respect to mean pain scores, F(2, 57) 1.63 (ns), or number of different types of side effects reported, F(2, 58) 0.21 (ns). The second research question focused on the average pain scores and side effects associated with the transition to oxycodone-cr while in the hospital. Table 3 shows the documented average pain scores while on PCA for the 24 hours prior to the transition to oxycodone-cr, for the 24-hour period after transition, and the period from 24 48 hours after transition (for those patients still hospitalized). There was a mean initial improvement in pain scores of one-half a point following transition to oxycodone-cr that persisted through the following 24-hour period. Although the children in the current study all experienced spinal fusion surgery, 11% had only anterior, 73% had only posterior, and 16% had anterior and posterior spinal fusions. Because the pain produced by each type might be different, a Kruskal-Wallis nonparametric analysis of variance was used to compare the differencesinpainscoresatthreetimepoints(24hours before initiation of oxycodone-cr, 24 hours after initiation of oxycodone-cr, and 48 hours after initiation of oxycodone-cr). The differences between pain scores at 24 hours after and 24 hours before oxycodone-cr, 48 hours after and 24 hours before, and 48 hours after and 24 hours after yielded P 0.793, P 0.831, P 0.533, respectively, indicating no significant difference in pain scores for the three types of spinal fusion surgery. This study focused on typical opioid side effects including sedation, dizziness, constipation, Table 2 Documented Mean Pain Scores and Mean Number of Side Effects of Patients Grouped by Ratio of PCA to Oxycodone-CR Dose Number of Dose Ratio Pain Scores Side Effects PCA:Oxycodone-CR (mg/kg/day) M SD M SD Less than 1:0.74 a 3.76 1.89 0.60 0.91 1:0.74 to 1:1.29 b 4.01 1.52 0.74 0.93 Greater than 1:1.29 c 3.05 1.66 0.60 0.63 Note: Data refer to the 24-hour period following transition to oxycodone-cr. Total n 61 due to incomplete data for one patient. a n 15. b n 31. c n 15. Table 3 Mean Pain Scores 24 Hours Prior, 24 Hours After, and 48 Hours After Transition Pain Score Time Relative to Oxycodone-CR Transition M SD 24 hours prior a 4.21 1.46 24 hours after b 3.70 1.67 48 hours after c 3.77 1.60 Note: Data were collected from the same patients over three separate time intervals. Differing n results from availability of pain score chart documentation. a n 61. b n 61. c n 55.

Vol. 27 No. 4 April 2004 Management of Pediatric Postoperative Pain 383 pruritus, nausea, vomiting, nausea reported together with vomiting, and respiratory depression (Figure 1). In general, reports of side effects did not change following transition to oxycodone-cr, with the exception of a moderate reduction in pruritus. Overall, however, the number of patients with side effects of any kind decreased from 56.5% with PCA to 43.9% (a 12.6% decrease) 48 hours after transitioning to oxycodone-cr. Figure 1 displays the documented side effects of patients over three time intervals: 1) while on PCA for the 24-hours prior to the transition to oxycodone-cr; 2) for the 24- hour period after transition, and 3) for the period 24 to 48 hours after transition. The last research question investigated the characteristics of oxycodone-cr use. Patients were transitioned from PCA to oxycodone-cr in an average of 3.9 (SD 1.7) days. The average length of stay in the hospital was 6.5 days (SD 1.8). The mean duration of oxycodone- CR use was 13.3 days (SD 4.4), including a mean wean time of 6.4 days (SD 2.4). Discussion The initial dose of oxycodone-cr was individually calculated based on each patient s PCA use (in morphine equivalents) during the previous 24-hour period. Because the analgesia was patient-controlled, there was significant variation in PCA use, and, therefore, in oxycodone-cr doses. The conversion ratio in the current study was 1:1 (parenteral morphine:oral oxycodone- CR), which is somewhat lower than that of Glare et al., 16 who report a relative milligram potency ratio of 3:1 (oral oxycodone:parenteral morphine) in adult cancer patients, although Glare et al. 16 used oxycodone-ir as opposed to the oxycodone-cr used in the current study. Although our initial calculations were successful in more than half the patients, 40% of patients did require a dose adjustment (27% required an increase in dose, 13% required a decrease in dose, total: 40%). Our data are somewhat lower than Citron et al., 12 who found among a sample of adult cancer patients that 57% of patients increased their total daily dose and 14% decreased their total daily oxycodone-cr dose. The adjustments required in our study may suggest the need for a slightly higher conversion to achieve optimal pain control with the initial transition, while also supporting the need for individualized dosing. The goal of this program was to optimize pain control with oral analgesics. The children in this study were transitioned from PCA to oxycodone-cr in an average of 4 days. The mean pain score following the transition to oxycodone-cr was considerably lower than the pain scores reported by Kotzer et al. 17 following spinal fusion surgery. They assessed pain twice a day using a 0 10 scale and reported a mean of 4.99 (SD 2.17) on postoperative day 4, while 58% of the children were still receiving opioids via PCA. Differences in pain scores may support the efficacy of oral CR analgesics for postoperative spinal fusion pain. However, while Kotzer Fig. 1. Percentage of patients who reported selected side effects 24 hours prior (n 62), 24 hours after (n 62), and 48 hours after (n 57) transition to oxycodone-cr. Data were collected from the sample over three separate time intervals, with differing n resulting from patient discharge prior to the third period.

384 Czarnecki et al. Vol. 27 No. 4 April 2004 et al. 17 took the mean of twice-daily pain assessments, and the current study took the mean of all pain assessments for a day; neither study considered the activity experienced during or just before the pain assessment. Therefore, it is difficult to conclude whether or not pain scores are comparable. Additionally, although the verbal 0 10 scale is commonly used in pediatric practice, no studies were found describing its validity and reliability among pediatric patients. After an initial decrease in pain scores with the transition to oxycodone-cr, pain scores in the current study showed an insignificant increase. Although this negligible increase is not considered to be clinically significant, there are some factors that may have contributed to an increase in pain at this point in the postoperative period. For example, an increase in pain may reflect an increase in patient activity. A second possibility may be that the PCA opioid was having an effect on pain scores during the first 24 hours after transition, and that with blood levels decreased, pain scores increased during the 24 48 hour time frame. Lastly, the children who opted to keep their PCA button for the first 24 hours after transition may have had a combined effect resulting in lower pain scores. The current study found side effects commonly reported with opioid use such as dizziness, constipation, sedation, nausea, vomiting, and pruritus before and after transitioning to oxycodone-cr. These side effects have been previously reported with oxycodone-cr used in a variety of patient populations. 8,9,12,13,15,18 Sedation and pruritus were the most commonly documented side effects while using PCA. This differs from others, 17,19 who report nausea and vomiting as the most frequently reported side effect, but is similar to Beaulieu et al., 19 who report a similar frequency of pruritus with morphine PCA use in adolescent spinal fusion patients. Patients in the current study were oftentimes not taking anything by mouth, which could have influenced the lower reports of nausea and vomiting. One of the most concerning side effects of any opioid is respiratory depression. In the current study, respiratory depression was one of the least frequently reported side effects, with both PCA and oxycodone-cr use, and no patient required aggressive intervention. Beaulieu et al. 19 report a 7% occurrence of respiratory depression in adolescent spinal fusion patients using similar amounts of morphine via PCA. The low incident of respiratory depression in patients using oxycodone-cr in the current study is consistent with the results of other studies with other patient populations. 8,15 Therefore, reported results support the safety of oxycodone-cr in adult and pediatric patients. Once patients were converted to oxycodone- CR, constipation, nausea, and dizziness were the most frequently reported side effects. These results are similar to Curtis et al., 15 who reported somnolence, nausea, and dizziness to be most common with adults using oxycodone-cr. Interestingly, Sunshine et al. 8 had no reports of nausea with oxycodone-cr use in adults. Overall, the current study found less nausea and vomiting with oxycodone-cr than with PCA. Pruritis significantly decreased with the transition of oxycodone-cr. The low frequency reported in the current study is consistent with low frequencies reported by Curtis et al. 15 in studying both oral oxycodone-cr and morphine-cr. The infrequent reports of sleepiness or sedation found in this study are inconsistent with other reports 8,15 that somnolence is the most commonly reported adverse event with oxycodone-cr use. Although one might expect a difference based on dose, no significant differences were found in the frequency of side effects based on mg/kg/day dosing in the current study. This is consistent with Cheville et al. 14 who found no difference with regard to side effects between treatment groups receiving oxycodone-cr as well as oxycodone-ir, and a placebo group receiving a placebo and oxycodone-ir. Perhaps the most important conclusion to be drawn with regard to side effects is that the overall incidence of side effects in the current study was lower with oxycodone-cr than with parenteral morphine. Specifically, more patients reported no side effects of any kind. As well, sedation and pruritus were reported less often with oxycodone-cr, and no patients experienced life-threatening respiratory depression. The pattern of decreasing side effects over time suggests that oxycodone-cr was tolerated fairly well by the second day, and is consistent with Citron et al., 12 who also report a decrease in side effects over time with oxycodone-cr.

Vol. 27 No. 4 April 2004 Management of Pediatric Postoperative Pain 385 The large range of time to transition to oral analgesics may be explained in part by a group of children who had sequential spinal fusions (i.e., anterior fusion followed by posterior fusion a few days later). Beaulieu et al. 19 reported a mean duration of PCA use of 3.1 days following pediatric spinal fusion surgery. The longer time until transition in the current study may be due to prolonged nothing by mouth status. The average parenteral morphine use prior to transitioning to oxycodone-cr in the current study was 56 µg/kg/hour, which is higher than reported in other studies. 17,19 Kotzer et al. 17 reported a mean morphine use of 44 µg/kg/ hr on postoperative day 3, and 40 µg/kg/hr on post operative day 4. However, patients in the Kotzer et al. 17 study also received RN-administered opioids in addition to the reported PCA use, which may explain the lower PCA use reported in that study. Once patients were transitioned, they were prescribed oxycodone-cr to use at home. Total oxycodone-cr use was 13.3 days on average, including both inpatient and outpatient use. Of those 13.3 days, an average of 6 days were used to wean from the original dose after discharge. Patients began weaning when pain was well managed, and the need for intermittent oxycodone-ir had decreased. Because no studies were found addressing the time of use of oxycodone-cr, it is difficult to comment on the length of use reported in the current study. Limitations The greatest limitation of this study is the retrospective design. The lack of control over the availability, consistency, and accuracy of children s recorded pain scores and side effects limit the conclusions that can be drawn from these findings. Another limitation is the homogenous sample with respect to sex and race. However, this homogeneity reflects the demographics of idiopathic scoliosis. 2,19 Subjectivity in interpreting the clinical findings, for example, categorization of side effects, is an additional limitation. Further oxycodone-cr research that is randomized, blinded, and includes a control group would extend the body of knowledge on pediatric pharmacological pain management. Conclusions This is the first study to demonstrate safe and effective use of CR analgesics in pediatric postoperative spinal fusion patients. The results of this study suggest that oxycodone-cr is safe and effective for pediatric patients transitioning to oral analgesics following PCA. Additionally, a mean conversion ratio from parenteral morphine to oxycodone-cr in children having spinal fusion surgery was explored. The ability to convert patients directly from PCA opioids to oxycodone-cr may support the use of CR analgesics without first introducing an immediate release formulation. These conclusions would be strengthened by a controlled randomized trial comparing oxycodone-ir alone and oxycodone-ir with oxycodone-cr in this patient population. Optimal pain control in children beyond the initial postoperative period with oral analgesics could lead not only to improved comfort, but could also improve sleep, possibly expedite the return to school, and improve the quality of life children experience following surgery. Acknowledgments This study was funded by Children s Hospital of Wisconsin Foundation and supported by the Jane B. Pettit Pain and Palliative Care Center. References 1. Lamontagne L, Hepworth J, Byington K, Chang CY. Child and parent emotional responses during hospitalization for orthopaedic surgery. MCN Am J Matern Child Nurs 1997;22:299 303. 2. Kotzer A. Factors predicting postoperative pain in children and adolescents following spinal fusion. Issues Compr Pediatr Nurs 2000;28:83 102. 3. Palermo T, Lambert S. A descriptive study of children s beliefs concerning the use of analgesics in treating postoperative pain. Child Health Care 1997;26:47 59. 4. Palermo T, Drotar D. Prediction of children s postoperative pain: the role of presurgical expectations and anticipatory emotions. J Pediatr Psychol 1996;21:683 698. 5. Melzack R. The tragedy of needless pain. Sci Am 1990;262;27 33. 6. Lambert SA. The effects of hypnosis/guided imagery on the postoperative course of children. J Dev Behav Pediatr 1996;17;307 310.

386 Czarnecki et al. Vol. 27 No. 4 April 2004 7. Anderson R, Saiers J, Abram S, Schlicht C. Accuracy in equianalgesic dosing: conversion dilemmas. J Pain Symptom Manage 2001;21:397 406. 8. Sunshine A, Olson N, Colon A, et al. Analgesic efficacy of controlled-release oxycodone in postoperative pain. J Clin Pharmacol 1996;36:595 603. 9. Heiskanen T, Kalso E. Controlled-release oxycodone and morphine in cancer-related pain. Pain 1997; 73:37 45. 10. Amaranth L, Andrish J, Gurd A. Efficacy of intermittent epidural morphine following posterior spinal fusion in children and adolescents. Clin Orthop 1989; 249:223 226. 11. Arms D, Smith J, Osteyee J, Gartrell A. Postoperative epidural analgesia for pediatric spine surgery. Orthopedics 1998;21:539 544. 12. Citron M, Kaplan R, Parris WC, et al. Long-term administration of controlled-release oxycodone tablets for the treatment of cancer pain. Cancer Invest 1998;16:562 571. 13. Roth S, Fleischmann R, Burch F, et al. Aroundthe-clock, controlled-release oxycodone therapy for osteoarthritis-related pain: placebo-controlled trial and long-term evaluation. Arch Intern Med 2000; 160:853 860. 14. Cheville A, Chen A, Oster G, et al. A randomized trial of controlled-release oxycodone during inpatient rehabilitation following unilateral total knee arthroplasty. J Bone Joint Surg 2001;83-A:572 576. 15. Curtis G, Johnson G, Clark P, et al. Relative potency of controlled-release oxycodone and controlled-release morphine in a postoperative pain model. Eur J Clin Pharmacol 1999;55:425 429. 16. Glare PA, Walsh TD. Dose-ranging study of oxycodone for chronic pain in advanced cancer. J Clin Oncol 1993;11(5):973 978. 17. Kotzer A, Foster R. Children s use of PCA following spinal fusion. Orthop Nurs 2000;19(5):19 30. 18. Salzman R, Roberts M, Wild J, et al. Can a controlled-release oral dose form of oxycodone be used as readily as an immediate-release form for the purpose of titrating to stable pain control. J Pain Symptom Manage 1999;18:271 279. 19. Beaulieu P, Cyrenne L, Mathews S, et al. Patientcontrolled analgesia after spinal fusion for idiopathic scoliosis. Int Orthop 1996;20:295 299.