Morphine pharmacokinetics and pain assessment in

Transcription

1 ORIGINAL ARTICLES Morphine pharmacokinetics and pain assessment in premature newborns Christy S. Scott, PharmD, K. Wayne Riggs, PhD, Emily W. Ling, MD, Colleen E. Fitzgerald, RN, Marilyn L. Hill, PhD, Ruth V. E. Grunau, PhD, Alfonso Solimano, MD, and Kenneth D. Craig, PhD Objectives: To determine morphine pharmacokinetics in premature neonates varying in postconceptional age (PCA) and evaluate behavioral pain response in relationship to serum morphine concentrations. Methods: Premature neonates (n = 48), stratified by weeks of PCA (group 1 = weeks, group 2 = weeks, group 3 = weeks, and group 4 = weeks) received morphine infusions. Blood samples were drawn at 48, 60, and 72 hours and at discontinuation of morphine, followed by 3 samples obtained during the next 24 hours. Newborns were videotaped during heel lances and restful states, with morphine at steady-state concentrations and without morphine. Pain was assessed by using the Neonatal Facial Coding System (NFCS). Statistical analysis included regression between NFCS score changes from baseline to painful procedure with and without morphine. Results: Morphine clearance for groups 1, 2, 3, and 4 was calculated as 2.27 ± 1.07, 3.21 ± 1.57, 4.51 ± 1.97, and 7.80 ± 2.67 ml/kg/min, respectively, and correlated with PCA (r = 0.63, P <.001). Pain measured by facial expression was diminished; however, it did not correlate with morphine concentrations. Conclusion: Morphine clearance in premature neonates is less than reported, increasing with PCA. Facial activity discloses morphine analgesia; however, it is unrelated to morphine concentrations. (J Pediatr 1999;135:423-9) From The Faculty of Pharmaceutical Sciences and Departments of Pediatrics and Psychology, The University of British Columbia and the Department of Psychology, British Columbia s Children s Hospital, Vancouver, British Columbia, Canada. Supported by the B.C. Health Research Foundation (grant No ). Presented in part at the 1994 Annual American Pediatric Society and the Society for Pediatric Research Meeting, May 4, 1994, Seattle, Washington; and the 3rd International Symposium on Pediatric Pain, June 8, 1994, Philadelphia, Pennsylvania. Submitted for publication Sept 9, 1998; revisions received Feb 5, 1999, and May 4, 1999; accepted May 12, Reprint requests: Christy Silvius Scott, PharmD, BCPS, Assistant Professor, School of Pharmacy, Division of Pharmacotherapy, The University of North Carolina, CB No. 7360, Beard Hall, Chapel Hill, NC Copyright 1999 by Mosby, Inc /99/$ /21/ Historically, neonates were considered to be relatively insensitive to pain. However, substantial evidence emphasizes that pain comprises physiologic and psychologic forms of stress in premature newborns, and there is increasing evidence of long-term deleterious effects. 3-6 The same analgesic considerations that apply to adults should apply to premature neonates; however, little is known about the optimal treatment. ANOVA C ss k e NFCS PCA V d See editorial, p Analysis of variance Steady-state serum concentrations Apparent terminal disposition rate constant Neonatal Facial Coding System Postconceptional age Volume of distribution Dosage recommendations for opioid medications need to be determined. 2 Only one morphine dose is recommended for term neonates in the literature, with no suggested regimens for premature neonates. 7 Measurement tool development for the evaluation of pain and impact of analgesics is another area that warrants investigation. 8-9 A small number of studies of the pharmacokinetics and pharmacodynamics of morphine have been completed in neonates. However, they are limited by low numbers of patients, and only a few have included premature newborns. Although these studies provide evidence of reduced morphine 423

2 SCOTT ET AL THE JOURNAL OF PEDIATRICS OCTOBER clearance, considerable variability in pharmacokinetic data reflects the inclusion of newborns with a wide range of ages. The few studies that addressed the issue of morphine efficacy have used subjective criteria, such as nursing observations. Others did not specify a priori criteria for end points. Although some authors argue that morphine provides analgesia in neonates, a dose-response or concentration-response relationship has not been documented with valid methods of pain assessment. The objectives of this study were to determine morphine pharmacokinetics in neonates grouped by postconceptional age and to objectively evaluate behavioral responses to pain in relationship to steady-state serum morphine concentrations. If morphine concentrations were related to responses, optimal morphine doses would be estimated. METHODS Ethical approval was obtained from The University of British Columbia and The British Columbia s Children s Hospital. Informed consent was obtained from parents or guardians before study entry. Inclusion Criteria Patients in the Special Care Nursery <40 weeks PCA who required morphine by infusion and received the same dose for at least 60 hours were eligible. Neonates were stratified by weeks of PCA into 4 groups: group 1 = 24 to 27 weeks, group 2 = 28 to 31 weeks, group 3 = 32 to 35 weeks, and group 4 = 36 to 39 weeks. Exclusion Criteria Patients were excluded a priori if they: (1) received morphine within 72 hours of study initiation; (2) had Apgar scores at 10 minutes 3; (3) had neurologic or facial pathology; (4) received sedating medications or skeletal muscle relaxants; (5) were infants of substance using mothers (ISUM) or diabetic mothers (IDM); (6) were small for gestational age; (7) were too unstable for extra manipulation; or (8) received morphine >14 days. Serum Morphine Concentrations Morphine infusions were initiated by the patients attending physicians. Usual morphine starting doses were 0.02 to 0.03 mg/kg/h, preceded by a bolus infusion of 0.05 mg/kg. Morphine doses and infusions were prepared by nursing staff; an infusion bag from each baby was sampled once and analyzed for morphine concentration. IMED volumetric infusion pumps (model 965 Micro; IMED Corp, San Diego, Calif) were used, and only known compatible medications were infused through the line through the y- site connector closest to the patient. Usual care was not altered, except for the drawing of 100- to 200-µL blood samples. Control serum samples were drawn before morphine administration or taken from cord blood. Additional blood samples were drawn at 48 ± 4 hours, 60 ± 4 hours, and 72 ± 4 hours and at the time of infusion discontinuation, which was determined by the attending physician. If the duration was <72 hours, the final sample was omitted. If the duration was >72 hours, blood samples from other blood draws were scavenged and collected approximately every 24 hours until discontinuation. Blood samples were also drawn between 3 and 4, 10 and 14, and 20 and 24 hours after the infusion was discontinued. All blood samples were drawn by heel stick or through non-morphine-containing indwelling catheters. Serum samples were stored at 70 C until analysis. Quality control serum samples were spiked with known morphine concentrations, stored at 70 C, and analyzed at 2-week intervals over a 6-month period. Patient samples were analyzed within 6 months after collection. Pain Assessment The Neonatal Facial Coding System was used for pain assessment. 8,21,22 Babies faces were videotaped during 4 conditions: restful state with morphine at steady-state concentrations; painful event with morphine at steady-state concentrations; restful state 4 days after morphine discontinuation; and painful event 4 days after morphine discontinuation. The painful event had 4 distinct components, and each was videotaped for a minimum of 20 seconds: baseline, swab, lance, and recovery. For the restful state, equivalent times were videotaped. Morphine conditions were recorded first because of the usual urgency in instituting therapy. Videotaping 4 days after morphine discontinuation allowed for body drug removal, based on the longest half-life reported in the literature for premature newborns. 12,20 The heel lancing procedure was chosen as the painful event because all infants required it. Two investigators served as videographers (C.S.S. and C.E.F.). The video camera was positioned to record only the infant s face. For masking, each videotaped condition was assigned a computer-generated random number. An additional computer random number system dictated which of 4 videotapes to use for each condition for each baby. Adverse Effects Adverse effects were not a primary focus; however, patients were closely monitored through standard practices. Analysis MORPHINE ASSAY. Morphine serum concentrations were determined by using a high-performance liquid chromatography technique developed for this study (K.W.R.), which is based on significant modifications to techniques previously reported. 24,25 Briefly, morphine was extracted from phosphatebuffered (ph 8.9) serum (0.05 to 0.1 ml) with methylene chloride containing 1% isopropanol. Nalorphine was used as the internal standard. After centrifugation, morphine and nalor-

3 THE JOURNAL OF PEDIATRICS VOLUME 135, NUMBER 4 SCOTT ET AL Table I. Clinical profile of infants completing protocol Clinical indications Mechanical PCA at GA at PNA at Birth Analgesia for Analgesia ventilation PCA entry entry entry weight Sex Sedation surgery (other) during group (wk) (wk) (d) (g) (M:F) (n) (n) (n) morphine (n) 1 (n = 10) 26.8 ± ± ± ± 174 8: (n = 16) 29.8 ± ± ± ± : (n = 15) 33.1 ± ± ± ± 412 8: (n = 7) 37.8 ± ± ± ± 464 4: Results are presented as mean ± SD. GA, Gestational age; PNA, postnatal age. phine were back-extracted from the retained organic layer by using 5 mmol/l phosphoric acid. The aqueous layer was subsequently removed for injection (30 µl) into the high-performance liquid chromatography column (Hewlett Packard 1050; Zorbax RX-C mm I.D. 15 cm). The mobile phase consisted of acetonitrile 24%, ph 3.6 phosphate buffer 76%, 1 mmol/l sodium dodecylsulfate, 1 mmol/l EDTA, and 10 mmol/l potassium chloride (flow rate, 0.5 ml/min). A Hewlett Packard 1049A amperometric detector, equipped with a glassy carbon electrode and operated at 0.6 V, was used for all analyses. Standard curves were linear over a working range of 0.5 to 60 ng/ml (r 0.99; limit of quantitation 0.5 ng/ml). The coefficients of variation for intraand inter-day precision were 14.2%, 6.5%, and 2.8% and 14.5%, 8.3%, and 6.2% at concentrations of 1, 10, and 30 ng/ml, respectively. Accuracy of the assay averaged +3.75%, 0.83%, and +1.01% at the 1-, 10-, and 30-ng concentration points, respectively. No chromatographic interference from concurrently administered drugs was observed in patient control samples. PHARMACOKINETIC CALCULATIONS. Steady state was assumed to be reached if 48- and 60-hour (for infusions discontinued before 72 hours) or 60- and 72-hour serum morphine concentrations were within 15% of each other. Elimination rate constant was determined from linear regression lines obtained for the post-infusion samples by using k e = (slope)(2.303), where k e is the apparent terminal disposition rate constant. Total body clearance was calculated as: Cl = k o /C ss, where k o is infusion rate and C ss is serum morphine steady-state concentration. For patients not achieving steady-state concentrations, k e was the only parameter determined. Volume of distribution was estimated from V d = Cl/k e and elimination half-life from t 1 2 = 0.693/k e. FACIAL ACTIVITY. Videotapes were analyzed to measure facial activity by one coder, who had passed the NFCS proficiency test and was masked to each infant s age, event, and condition. To establish reliability, a second coder coded 25% of randomly selected videotape samples. Satisfactory agreement was set at NFCS reliability coefficient of The 4 procedures were analyzed for the first 10 seconds. Each NFCS facial action was identified as present or absent, during five 2-second intervals. Baseline began 30 seconds before the laboratory technician initiated physical contact. The swabbing procedure began with the technician lifting the heel, and lancing followed penetration with the lancet. Recovery commenced 60 seconds after conclusion of the lance. Statistical Analysis Pharmacokinetic data were analyzed across the groups by using 1-way analysis of variance with post-hoc analyses (Tukey HSD), with a significance level of P <.05. Regression correlation analyses were used to examine whether pharmacokinetic parameters varied with PCA. To examine behavioral reactions, facial activity was reduced to a summary score for each of the events. A principal components analysis of NFCS facial actions indicated a single factor accounted for reactions during heel lancing during the no-morphine condition. Scores for all events were based on the sum of weighted contributions of facial actions to this factor, with factor score coefficients as the weights. The effects of PCA (a between-subjects variable), condition, and procedure (within-subjects variable) were examined by using a ANOVA (followed by Tukey tests). Because morphine conditions preceded no morphine conditions, analysis of covariance, co-varying for age, was conducted with factors paralleling the ANOVA. Regression analyses also examined the relationship between facial activity and PCA, magnitude of response to the painful event, and serum morphine concentrations. RESULTS A total of 496 neonates received morphine infusions during the study period, October 9, 1991, through December 4, 1993; 48 individual newborns (33 boys) completed the protocol (Table I). Excluded infants included 47 consent re- 425

4 SCOTT ET AL THE JOURNAL OF PEDIATRICS OCTOBER 1999 Table II. Pharmacokinetic data k o (all) k o (C ss ) C ss Cl k e t 1 2 Group (µg/kg/h) (µg/kg/h) (ng/ml) (ml/kg/min) (h 1 ) (h) (L/kg) ± ± ± 99.4* 2.27 ± ± ± ± 0.67 (n = 10) (n = 9) (n = 9) (n = 9) (n = 10) (n = 10) (n = 9) ± ± ± ± ± ± ± 1.01 (n = 16) (n = 13) (n = 13) (n = 13) (n = 16) (n = 13) (n = 16) ± ± ± ± ± ± ± 0.69 (n = 14) (n = 13) (n = 13) (n = 13) (n = 13) (n = 13) (n = 12) ± ± ± ± ± ± ± 3.32 (n = 4) (n = 3) (n = 3) (n = 3) (n = 3) (n = 3) (n = 2) Results are presented as mean ± SD. k o (all), Morphine infusion rate for all patients completing study; k o (C ss ), morphine infusion rate for patients reaching C ss ; Cl, total morphine clearance rate; t 1 2, half-life. *P =.002: group 1 versus groups 2, 3, and 4. P =.023: group 1 versus groups 3 and 4. P <.001: group 4 versus groups 1, 2, and 3. V d Table III. Principal components analysis of NFCS facial activity Variable Factor loading Factor score coefficient Eye squeeze Brow bulge Nasolabial fold Lips open Taut tongue fusals and 17 deaths before completion of the entire protocol. One hundred six and 14 patients received morphine <60 hours and >14 days, respectively; 14 neonates received morphine within 72 hours of study entry. Neurologic (58) and facial (3) pathology accounted for 61 exclusions. Fourteen and eight neonates received sedating medications and skeletal muscle relaxants, respectively, and were excluded. Infants of substance using mothers, infants of diabetic mothers, and infants small for gestational age accounted for 10, 25, and 38 excluded newborns, respectively. Babies too unstable for extra manipulation (n = 25) were excluded. In groups 2 and 3, the number goal was reached before study conclusion; 46 and 23 babies, respectively, were excluded. Pharmacokinetics Pharmacokinetic data for 44 of the 48 newborns are presented in Table II. Three serum samples in group 1 and one sample in group 2 were lost because of improper storage. Two infants, one each in groups 1 and 2, did not have all 3 post-infusion blood samples drawn and were excluded from analyses. Six patients (one in group 1, three in group 2, one in group 3, and one in group 4) did not reach morphine steady-state concentrations; therefore, Cl and V d were not calculated. In groups 3 and 4, k e estimation in one infant was not possible because of rapid drug elimination; 2 of 3 plasma concentrations were below assay detection limits. Analysis of each morphine infusion revealed that all contained the specified concentration. Analysis of quality control serum samples indicated no change in morphine concentration over 6 months. Morphine infusion doses ranged from 18 to 60 µg/kg/h (mean, 24.0 ± 8.5 µg/kg/h), for a mean of 80.4 ± 34.8 hours (range, 56.5 to hours). No significant differences in infusion rates or lengths were observed among groups. C ss produced by these infusions varied from 32.3 to ng/ml. Group 1 had significantly higher morphine concentrations than groups 2 to 4 (ANOVA, P =.002). Group 4 had a significantly higher Cl than the other 3 groups (ANOVA, P <.001), whereas group 1 had a significantly lower Cl than groups 3 (P =.023) and 4 (P <.001). There were no significant differences in V d among the groups. Regression analysis indicated a positive correlation between increasing PCA and Cl (r = 0.63, P <.001). A negative correlation is noted between increasing PCA and C ss (r = 0.55, P <.001). Pain Assessment NFCS inter-coder reliability coefficient was Five facial actions occurred more frequently than the cutoff score for infrequent actions. The principal components analysis yielded a single factor accounting for 71.2% of variance (Table III). Forty-eight newborns received consistent morphine infusions for at least 60 hours, meeting criteria for facial activity analysis. Facial analysis revealed significant effects for age (P <.009), procedure (P <.001), and condition (P 426

5 THE JOURNAL OF PEDIATRICS VOLUME 135, NUMBER 4 SCOTT ET AL <.001). A significant interaction between procedure and condition was also observed (P <.001). No interaction between age and procedure or age and condition was observed. Mean facial activity scores for the groups for each procedure during the 4 conditions are shown in Fig 1. Facial expression during the painful event varied significantly across morphine and no-morphine conditions during swab and lance procedures (P <.01 for both procedures). There were no differences between the equivalent segments during the restful state. Fig 2 shows total facial activity scores for each PCA group, summing across both procedures and condition. There were no significant differences between groups 1 and 2 or between groups 3 and 4. However, groups 3 and 4 displayed significantly more facial activity than group 1 or group 2 (P <.01). The mean time interval ± SD between morphine and no-morphine conditions was 5.7 ± 2.5 days. After age was removed as an influence on NFCS scores, event, (P <.001), condition, (P <.001), and an interaction between event and condition (P <.001) continued to have significant effects. The Pearson product-moment correlation between PCA and facial activity was r = 0.44 (P <.01). The correlation between morphine serum concentrations and change in NFCS was not significant (r = 0.16, P >.05). DISCUSSION The study findings provide evidence that morphine provides analgesia in premature neonates and that morphine clearance increases with PCA. Neonates aged 24 to 27 weeks had the lowest clearance, as expected. Authors have suggested that gestational age weakly correlates with morphine clearance. 16,18 Our data suggest that PCA more accurately reflects morphine clearance. In the absence of urinary excretion data for morphine and its metabolites, Fig 1. Mean NFCS scores of the 4 PCA groups combined for each procedure during each condition. No morphine and painful event, ; morphine and painful event, ; no morphine and restful state, ; morphine and restful state, Facial expression was diminished by morphine during swab and lance procedures (*P <.01, ANOVA with post-hoc Tukey test). Fig 2. Mean ± SD NFCS sum of the 4 procedures (baseline, swab, lance, and recovery) over all conditions for each group. Group 1 = 24 to 27, group 2 = 28 to 31, group 3 = 32 to 35, and group 4 = 36 to 39 weeks PCA, respectively. Groups 3 and 4 displayed significantly more facial activity than either group 1 or group 2 (*P <.01, ANOVA with post-hoc Tukey test). we can only speculate that the lower Cl values, and hence higher morphine concentrations observed in babies in group 1, are probably reflective of less mature drug elimination pathways (renal, hepatic) than in babies in groups 3 and 4. Alterations in binding do not appear to be a contributing factor because there were no significant differences in volumes of distribution. Mean morphine half-lives varied from 7.0 to 13.5 hours. Administering morphine by infusion may therefore be inappropriate. Morphine infusion rates were often increased before reaching C ss. These infants are at potential risk of receiving too much medication as morphine concentrations increase while approaching steady-state concentrations. Concentrations may also be higher than necessary for the expected and desired therapeutic effect, especially for infants <28 weeks PCA. Side effects could not be attributed to morphine in this study; however, adverse effect determination was not an 427

6 SCOTT ET AL THE JOURNAL OF PEDIATRICS OCTOBER 1999 outcome measure. Of note, almost all patients in this study were receiving mechanical ventilation, so conclusions cannot be made about morphine-induced respiratory depression. Morphine steady-state concentrations were chosen as the target for facial activity measurement. Six young babies did not reach steady-state because of unexpectedly long drug halflives. Determined pharmacokinetic parameters may overestimate true morphine elimination and clearance rates; however, the number of babies who did not achieve steady-state concentrations was small. Although pain is a subjective response, requiring clinicians to infer its presence and severity, objective evidence can be provided with explicit behavioral criteria. 8 Facial expression has been well studied and is the most effective measure of pain in neonates. 8 In this study facial activity reflected analgesic impact, because morphine reduced heel stick associated facial activity. The findings indicate that facial activity provides a criterion measure for analgesia in premature neonates. The sensitivity of the NFCS measure is reflected in continuing reactivity to tissue damage produced by lancing the heel. More activity with increased PCA attests to the limited expressive capabilities of the early preterm newborn. The findings are consistent with earlier reports of a positive correlation between the magnitude of facial activity and gestational age. 8 The diminished activity stands in contrast to findings indicating central sensitization or hyperalgesia during tissue damage in the very premature newborn. 26 In contrast to this heightened sensitivity, communication of distress appears diminished. Although the number of excluded infants in this study was high, every infant who received morphine by infusion was considered for the study. A large number of the exclusions occurred because infusion rates were not continued for sufficient time to reach steady-state concentrations and because babies were weaned instead of discontinued from morphine at the rate that achieved steady-state concentrations. Many infants were excluded because the goal number of patients had been reached in 2 groups. Infants of substance using mothers (ISUM) or those small for gestational age were excluded because of the risk of hyperirritability, and infants of diabetic mothers (IDM) were excluded because of the potential for decreased activity. Movement of the babies was often necessary for videotaping; unstable infants were excluded. Group 4 infants were difficult to recruit, because many did not require morphine for sedation. Patients were excluded if they had received morphine recently to detect medications interfering with the morphine assay, as determined by analyzing baseline blood samples. This study attempted to correlate morphine serum concentrations with analgesia in premature neonates by using a recognized objective validated scale. Because steady-state morphine concentrations varied greatly, it is not surprising that a correlation was not found. Studies completed in older children and adults (with subjective and non-validated methods of pain relief) also determined that there is not a simple relationship between serum concentration and analgesic efficacy This study replicates this finding in premature infants, and therefore this phenomenon is generalizable across a wide age range. One study including term neonates and a small number of premature neonates reported that analgesia was attained with morphine serum concentrations of 125 ng/ml. 18 However, the investigators used a nonvalidated, subjective comfort score. Although clinicians exercise clinical judgment to assess pain on a daily basis, in newborn patients receiving skeletal muscle relaxants or sedating medications that do not provide analgesia (eg, benzodiazepines or barbiturates), the pain response will not be evident. 32 Systematic research investigations to determine the appropriate doses of morphine and other analgesics for premature neonates are needed. It is imperative that these studies use objective, validated measures of pain. REFERENCES 1. Anand KJS, Hickey PR. Pain and its effects in the human neonate and fetus. N Engl J Med 1987;317: Yaster M. Analgesia and anesthesia [editorial]. J Pediatr 1987;111: Anand KJS, Hickey PR. Halothanemorphine compared with high-dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992;326: Grunau RVE, Whitfield MF, Petrie JH. Pain sensitivity and temperament in extremely low-birth-weight premature toddlers and pre-term and fullterm controls. Pain 1994;58: Grunau RVE, Whitfield MF, Petrie JH. Children s judgments about pain at age 8-10 years: Do extremely low birthweight ( 1000 g) children differ from full birthweight peers? J Child Psychol Psychiatry 1998;39: Johnston CC, Stevens B, Yang F, Horton L. Developmental changes response to heelstick in preterm infants: a prospective cohort study. Dev Med Child Neurol 1996;38: Koren G, Butt W, Chinyanga H, Soldin S, Tan YK, Pape K. Postoperative morphine infusion in newborn infants: assessment of disposition characteristics and safety. J Pediatr 1985;107: Craig KD, Whitfield MF, Grunau RVE, Linton J, Hadjistavropoulos HD. Pain in the preterm neonate: behavioral and physiological indices. Pain 1993;52: Anand KJS, Craig KD. New perspectives on the definition of pain. Pain 1996;67: Lynn AM, Slattery JT. Morphine pharmacokinetics in early infancy. Anesthesiology 1987;66: Farrington EA, McGuinness GA, Johnson GF, Erenberg A, Leff RD. Continuous intravenous morphine infusion in postoperative newborn infants. Am J Perinatol 1993;10: Mecurio M, Nelli C, Gettner P, Sherwonit E, Williams J, Enrenkranz R. Morphine pharmacokinetics in preterm newborns [abstract]. Pediatr Res 1989;25:71A. 13. McRorie TI, Lynn AM, Nespeca MK, 428

7 THE JOURNAL OF PEDIATRICS VOLUME 135, NUMBER 4 SCOTT ET AL Opheim KE, Slattery JT. The maturation of morphine clearance and metabolism. Am J Dis Child 1992;146: Choonara I, Lawrence A, Michalkiewicz A, Bowhay A, Racliffe J. Morphine metabolism in neonates and infants. Br J Clin Pharmacol 1992;34: Choonara IA, McKay P, Hain R, Rane A. Morphine metabolism in children. Br J Clin Pharmacol 1989;28: Bhat R, Chari G, Gulati A, Aldana O, Velamati R, Bhargava H. Pharmacokinetics of a single dose of morphine in preterm infants during the first week of life. J Pediatr 1990;117: Barrett DA, Elias-Jones AC, Rutter N, Shaw PN, Davis SS. Morphine kinetics after diamorphine infusion in premature neonates. Br J Clin Pharmacol 1991;32: Chay PCW, Duffy BJ, Walker JS. Pharmacokinetic-pharmacodynamic relationships of morphine in neonates. Clin Pharmacol Ther 1992;51: Hartley R, Green M, Quinn M, Levene MI. Pharmacokinetics of morphine infusion in premature neonates. Arch Dis Child 1993;69: Mikkelsen S, Feilberg VL, Christensen CB, Lundstrom KE. Morphine pharmacokinetics in premature and mature newborn infants. Acta Paediatr 1994;83: Grunau RVE, Craig KD. Pain expression in neonates: facial action and cry. Pain 1987;28: Taddio A, Stevens B, Craig KD, Rastogi P, Ben-David S, Shennan A, et al. Efficacy and safety of lidocaine-prilocaine cream for pain during circumcision. N Engl J Med 1997;336: Guinsburg R, Kopelman BI, Anand KJS, De Almeida MGB, Peres CA, Miyoshi MH. Physiological, hormonal and behavioral responses to a single fentanyl dose in intubated and ventilated preterm neonates. J Pediatr 1998;132: Todd RD. Muldoon SM, Watson RL. Determination of morphine in cerebrospinal fluid and plasma by highperformance liquid chromatography with electrochemical detection. J Chromatogr 1982;232: Glare PA. Walsh TD, Pippenger CE. A simple, rapid method for the simultaneous determination of morphine and its principal metabolites in plasma using high-performance liquid chromatography and fluorometric detection. The Drug Monitor 1991;13: Andrews KA, Fitzgerald M. The cutaneous withdrawal reflex in human neonates: sensitization, receptive fields and the effects of contralateral stimulation. Pain 1994;56: Okkola KT, Maunuksela E, Korpela R, Rosenberg PH. Kinetics and dynamics of postoperative intravenous morphine in children. Clin Pharmacol Ther 1988;44: Dahlstrom B, Tamsen A, Paalzow L, Hartvig P. Patient controlled analgesic therapy. Part IV. Pharmacokinetics and analgesic concentrations of morphine. Clin Pharmacokinet 1982;7: Nayman J. Measurement and control of postoperative pain. Ann R Coll Surg Engl 1979;61: Dahlstrom B, Bolme P, Feychting H, Noack G, Paalzow L. Morphine kinetics in children. Clin Pharmacol Ther 1979;26: Lynn AM, Opheim KE, Tyler DC. Morphine infusion after pediatric cardiac surgery. Crit Care Med 1984;12: Quinn MW, Otoo F, Rushforth JA, Dean HG, Puntis JWL, Wild J, et al. Effect of morphine and pancuronium on the stress response in ventilated preterm infants. Early Hum Dev 1992; 30: O N THE MOVE? Send us your new address at least six weeks ahead Don t miss a single issue of the journal! To ensure prompt service when you change your address, please photocopy and complete the form below. Please send your change of address notification at least six weeks before your move to ensure continued service. We regret we cannot guarantee replacement of issues missed due to late notification. JOURNAL TITLE: Fill in the title of the journal here. OLD ADDRESS: Affix the address label from a recent issue of the journal here. NEW ADDRESS: Clearly print your new address here. Name Address City/State/ZIP COPY AND MAIL THIS FORM TO: OR FAX TO: OR PHONE: Journal Subscription Services Mosby, Inc. Outside the U.S., call Westline Industrial Dr St. Louis, MO