Methadone dosage for prevention of opioid withdrawal in children

Transcription

1 Paediatric Anaesthesia : Methadone dosage for prevention of opioid withdrawal in children RAJASHEKHAR SIDDAPPA MD*, JAMES E. FLETCHER MBBS, ANDREW M.B. HEARD MBChB, DONNA KIELMA RN, MICHAEL CIMINO MS, RPH AND CHRISTOPHER M.B. HEARD MBChB* MS, RPH *Division Pediatric Critical Care, Children s Hospital of Buffalo, Buffalo, NY, Department of Anesthesiology, UNC, Chapel Hill, NC, USA, Department of Anaesthesia, Royal, Perth Hospital, Perth, Australia, Clinical Pharmacist, Children s Hospital of Buffalo, Buffalo, NY, USA and QA Coordinator, Children s Hospital of Buffalo, Buffalo, NY, USA Summary Background: Opioids are frequently used for sedation in the Paediatric Intensive Care Unit (PICU). With time the dosing often increases because of tolerance. On cessation of the sedation there is a risk of the opioid withdrawal syndrome. The aim of our study was to evaluate methadone dosing as a risk factor for opioid withdrawal and to determine optimal dose and efficacy of methadone to prevent withdrawal. Method: We undertook a clinical, retrospective, chart review study. Data were analysed from the quality improvement initiative database of a tertiary-care 18 bed PICU. Results: Data from 30 children who received an opioid infusion for 7 days and subsequently received methadone for opioid withdrawal (between January 2000 and July 2001) were analysed. Nurses documented the presence or absence of withdrawal signs daily. Our unit protocol has recommended converting the patient s opioid dose into fentanyl equivalents and a dose of methadone equal to the total daily dose of fentanyl to be given three times a day. Twenty patients had no or minimal withdrawal symptoms and 10 experienced significant withdrawal. Age, weight, PRISM score, lorazepam dose, muscle relaxant use and fentanyl dose were not statistically significantly between these groups. Receiver Operator Characteristics analysis showed that 80% of the suggested methadone dose was effective in minimizing withdrawal symptoms. The odds ratio for withdrawal with <80% of the predicted methadone dose was 21. Conclusions: Inadequate methadone is a risk factor for opioid withdrawal. A daily starting methadone dose equivalent to 2.5 times the daily fentanyl dose is effective in minimizing withdrawal symptoms. Keywords: critical care; drug tolerance; methadone; narcotics; pediatrics; substance; withdrawal syndrome Correspondence to: Dr Christopher MB Heard, Division Pediatric Critical Care, Children s Hospital of Buffalo, 219 Bryant Street, Buffalo, NY 14222, USA ( heardop1@adelphia.net). Ó 2003 Blackwell Publishing Ltd 805

2 806 R. SIDDAPPA ET AL. Introduction Sedation and analgesia are an indispensable component of Paediatric Intensive Care Unit (PICU) patient management. In the United States, fentanyl is the most commonly used opioid for sedation and analgesia in PICUs (1) and opioids are known to decrease the concentration of catecholamines in mechanically ventilated preterm babies (2) and improve clinical outcome in neonates undergoing cardiac surgery (3). However, tolerance, dependence and withdrawal symptoms are potential deleterious effects which are related to the total dose and duration of fentanyl infusion (4,5). A fentanyl infusion of 5 days or a total cumulative dose of 1.5 mgækg )1 during the hospital stay was associated with 50% chance of developing narcotic withdrawal, whereas a fentanyl infusion of 9 days or a total cumulative fentanyl dose of 2.5 mgækg )1 during the hospital stay had an incidence of 100% withdrawal (6). With the reduced use of neuromuscular blocking agents and heightened awareness for adequate sedation, use of higher doses of opioids for patient comfort may become more commonplace and will increase the risk for the opioid withdrawal syndrome. For those patients considered to be at risk of withdrawal, several options are available. These include oral morphine (5), methadone (7,8), clonidine (9), transdermal clonidine patch (10) and subcutaneous fentanyl (11). Among these choices methadone seems to be most suitable. The oral bioavailability of methadone is 80 90%, with elimination half-life of h and it is equipotent with morphine (12). Convenience of oral route, less frequent dosing due to its longer half-life and easy calculation because of its equal potency with morphine makes methadone attractive to use for prevention of opioid withdrawal in children. There is a huge variability in recommendations regarding the methadone dose to prevent opioid withdrawal in children (5,7,8,13 17). Our study was undertaken to evaluate the optimal dose of methadone, which prevented withdrawal syndrome in children. We hypothesized that the withdrawal syndrome is because of the use of an inadequate dose of methadone, and that a dose of methadone correctly calculated from the fentanyl infusion rate, prior to its discontinuation, would significantly decrease the incidence of the opioid withdrawal syndrome. Materials and methods The study was reviewed and approved by the institutional review board. This is a retrospective, chart review and quality improvement initiative data bank review study. Data of patients admitted to PICU during January 2000 to July 2001 were analysed. The PICU staff of our hospital had introduced a clinical pathway to assist with opioid withdrawal management. This pathway provided guidelines for determining those at risk for withdrawal, and a dosing and duration recommendation for methadone therapy. This pathway was available for use by PICU staff if they felt it appropriate. At risk patients were identified as those who had received opioid infusion for 7 days. For these at risk patients, the nurse caring for the patient assessed the patient for signs and symptoms of opioid withdrawal and completed a daily yes/no questionnaire that was added to the patient s daily flow sheet. The symptoms recorded by the nurse are shown in Table 1. The PICU physician evaluated the symptoms recorded to confirm opioid withdrawal. The methadone dose was based upon the dose of opioid the patient was receiving for sedation. Fentanyl is 100 times more potent than methadone and its duration of action is min (12). Duration of action of methadone is h (12). Considering duration of action of methadone as 24 h and duration of fentanyl 45 min (0.75 h) equipotent methadone dose is equal to 3 total daily fentanyl dose (13). This is calculated by the formula. Methadone dose (mg) ¼ [fentanyl daily dose (mg) fentanyl potency duration of action of fentanyl] =duration of action of methadone ¼ fentanyl dose 100 0:75=24 ¼ fentanyl daily dose 3 The recommended dose of methadone per day is three times the daily fentanyl dose per day. This methadone dose was initially given intravenously in three divided doses per day for 48 h. The fentanyl dose was tapered at 50% decrements starting from

3 METHADONE FOR OPIOID WITHDRAWAL 807 Table 1 Signs/symptoms of opioid withdrawal recorded by the nurse Neurological excitability Sleep disturbances Agitation Mild tremors Severe tremors Seizures Choreoathetoid movemements GI disturbances Vomiting Diarrhoea Autonomic dysfunction Hypertension a Tachycardia b Tachypnoea c Fever (>38.5) units Frequent yawning Sweating Goose flesh Mottling a Hypertension, unexplained systolic BP >30 mmhg from baseline for 10 min. b Tachycardia, unexplained heart rate >40 min )1 from base line for 10 min in children <1 year and >30 min )1 in children more than 1 year. c Tachypnoea, unexplained respiratory rate >10 min )1 from base line for 10 min. the second dose of methadone and the fentanyl infusion was stopped at the fourth methadone dose and the route of methadone administration was changed from intravenous to oral. Methadone has a high bioavailability (12) so no dose correction is required for the i.v. to oral conversion. The methadone wean was started the day after oral conversion and the rate of weaning was between 3 and 10% per day depending on the dose and duration of opioid sedation. During the methadone wean if the patient exhibited withdrawal symptoms an additional 10 20% of the methadone dose was administered intravenously by the PICU team on an as needed basis until the symptoms were controlled and the maintenance dose was increased accordingly if necessary. After the patient was transferred to the ward, a clinical pharmacist followed the patients each day. The ICU team was available for consultation to answer any questions. The paediatric residents had been taught from a core lecture, the symptoms, signs and management of opioid withdrawal. Prior to discharge the parents were instructed to contact the hospital paediatrician if there were any concerns about the methadone use or its dosing. After the patients were discharged home, a follow-up phone call was made by the clinical QA coordinator. The parents were asked about symptoms associated with opioid withdrawal, the need for contacting or discussing with a physician about the methadone use and any problems with the methadone protocol. The patients who received opioid infusion for 7 days and then received methadone in our PICU were included from the study. Patients with complicated postextubation course (underlying disease process made withdrawal symptoms unreliable), those who received other withdrawal therapies, patients with severe brain damage, inadequate documentation and those not extubated within 72 h after starting methadone were excluded from the study. Data collection consisted of patient demographics, admission PRISM scores, opioid requirements (fentanyl days, maximum dose, cumulative dose), muscle relaxant requirement, cumulative benzodiazepine (lorazepam) dose, methadone dose, withdrawal symptoms and their duration. Sufentanil and remifentanil are the other opioids used for sedation in our PICU. They were converted to fentanyl equivalents by using the formulas 1 lgækg )1 Æh )1 of sufentanil ¼ 5 lgækg )1 Æh )1 of fentanyl and 0.1 lgækg )1 Æmin )1 of remifentanil ¼ 1 lgækg )1 Æh )1 of fentanyl. Midazolam was considered to be equipotent with lorazepam. Duration of action of lorazepam was considered to be 6 h. This was considered while converting daily dose of midazolam to an equivalent daily dose of lorazepam. Patients were classified into two groups. No or minimal withdrawal (group 1) are patients with <3 symptoms and did not receive additional methadone. Significant withdrawal (group 2) patients are those with 3 symptoms or received additional methadone dose to prevent or treat withdrawal. Data are expressed as mean (±SD) or median (range). Unpaired t-test, Mann Whitney test, Fisher Exact Test and Receiver Operator Characteristics (ROC) were used to analyse the data as appropriate. A P-value of <0.05 was considered significant. Results There were 58 patients during the study period (January 2000 July 2001) who received opioid infusion 7 days and then received methadone. There

4 808 R. SIDDAPPA ET AL. were 28 exclusions from the study. Severe neurological damage (3), complicated postextubation course [nine (six requiring tracheostomy)], continued mechanical ventilation for more than 72 h after instituting methadone (6), concurrent administration of other medications [five (two received clonidine and three received haloperidol)], inadequate documentation (5). There were 20 patients in the minimal or no withdrawal group and 10 patients in significant withdrawal group. Patient characteristics of both groups are summarized in Table 2. There was no significant difference between the groups with respect to patient demographics. Fentanyl dosing and other variables affecting withdrawal are summarized in Table 3. All 30 patients received fentanyl, four in each group were converted to sufentanil, two patients in the minimal withdrawal group and three patients in the significant withdrawal group received remifentanil. There was no significant difference between the groups with respect to opioid dose or duration of use, nor the use of muscle relaxants or benzodiazepines. To determine the optimal dose of methadone that would prevent withdrawal a ROC analysis was performed (Figure 1). The ROC was The dose Table 2 Patient characteristics No withdrawal (n ¼ 20) Withdrawal (n ¼ 10) P-value Age (years) 1.3 ( ) 7.7 ( ) NS Weight (kg) 10 ( ) 24 (4.6 70) NS Male/female 14/6 5/5 NS PRISM Score 7.7 ± ± 8.9 NS NS, not significant. Table 3 Fentanyl dose and other variables affecting withdrawal Variable No withdrawal Withdrawal P-value Sedation days 10 (7 41) 12.5 (7 26) NS Fentanyl max dose 8.2 (4 117) 14 (7 55) NS (lgækg )1 Æh )1 ) Fentanyl cumulative 1.25 ( ) 4.2 (1 10.5) NS dose (mgækg )1 ) Lorazepam cumulative 10 (0 225) 32.6 ( ) NS dose (mgækg )1 ) Muscle relaxant days 3 (0 35) 2 (0 8) NS Withdrawal days 0 (0 1) 3.5 (2 12) <0.001 NS, not significant. Sensitivity Specificity Figure 1 This Receiver Operator Characteristics (ROC) plots the sensitivity against 1-specificity for each datapoint. The datapoints are calculated for each patient based upon the dose of methadone given (as a percentage of the predicted dose) and whether or not withdrawal occurred. The higher the ROC, the better the test. A ROC ¼ 0.5 is a random event. The left uppermost point of the curve is the datapoint with the combined highest sensitivity and specificity. This reflects a methadone dose that is 80% of that predicted by our initial protocol. Table 4 Presence and absence of withdrawal symptoms in children if 80% of the predicted methadone dose used Methadone dose No withdrawal (%) Withdrawal (%) 80% predicted 2 (12) 7 (78) >80% predicted 18 (86) 3 (14) P ¼ (Fisher exact test). of methadone that had the highest sensitivity (90%) and specificity (70%) for preventing withdrawal was 80% of that derived from our formula. This was then used to calculate the odds ratio of developing withdrawal because of an inadequate dose of methadone (Table 4). The odds ratio to develop withdrawal with <80% recommended methadone dose is 21. On follow-up of the patients discharged, nine patients were discharged home on methadone. Of the 25 parents contacted, none of the patients had any symptoms of opioid withdrawal, there were no problems associated with the methadone wean. A stigma for methadone was not expressed by any of the parents. Discussion The use of methadone in a PICU for the prevention of opioid withdrawal was reported for the first time in 1990 (14). Opioid weaning protocols with methadone seem to be evolving with respect to dose

5 METHADONE FOR OPIOID WITHDRAWAL 809 and duration (5,7,8). The suggested starting methadone dose varies; earlier articles have suggested a low dose of methadone (0.1 mgækg )1 ) in two to four doses per day (13 15). Tobias recently has suggested 1 : 1 conversion of fentanyl dose/day to methadone dose per day every 12 h (16). Robertson et al. have used four times the fentanyl dose per day as equivalent methadone dose per day given in four divided doses every 6 h to be weaned over 7 days (7). Meyer et al. have utilized 10 times the fentanyl dose per day as equivalent methadone dose given initially in two divided doses every 12 h and then once a day to be weaned over 10 days (8). There is a case report of using methadone 325 lgækg )1 Æh )1 in a 18-month-old, 12 kg child on the first day which is equivalent to 93.6 mg of methadone per day followed by 20 mg i.v. three times a day (17). In our clinical pathway guideline, a methadone dose of three times the fentanyl dose was recommended to prevent opioid withdrawal. The starting dose may be considered empirical because of the variability of reported methadone pharmacokinetics (13) and the prolongation of fentanyl effect following long term high dose fentanyl infusions often used in these at risk patients. It was chosen following recommendations from a series of successful case reports by Tobias (13). This retrospective study shows that 80% of our recommended methadone dose, which corresponds to 2.4 times the fentanyl dose per day (mg) appeared effective in preventing the withdrawal syndrome. When we analysed the patients who received less than the recommended dose, it was apparent that the staff involved in the care of the patient used a lower dose than that recommended because they believed the recommended dose to be excessive. In our initial analysis (n ¼ 10 patients) there was also a significant difference between the no withdrawal group and the withdrawal group with respect to the total cumulative dose and duration of fentanyl infusion. With time the use and dose of methadone became more appropriate to clinical need, especially in those patients with high dose fentanyl exposure for whom initially the predicted methadone dose was not always used. In the final analysis, dose and duration of fentanyl were no longer a significant risk factor for withdrawal. For example: in a 12-year-old boy, the recommended dose of methadone was 2.3 mgækg )1 Æday )1 (60 mgæday )1 ). He received 1.7 mgækg )1 Æday )1 (45 mgæday )1 ) and developed significant withdrawal symptoms. With experience and more familiarity with the protocol recommended doses were used without hesitation. For example: 8-month-old, 6 kg child received recommended dose of 5mgÆkg )1 Æday )1 (30 mgæday )1 ) and did not experience withdrawal symptoms or treatment related side effects. In our institutional recommendation there was no maximum dose of methadone. Others have recommended 40 mg of methadone as maximum dose (7). As stated in the previous example the 12-old-boy developed withdrawal when he was receiving 45 mgæday )1 of methadone. It may be prudent not limit the maximum dose of methadone but rather to titrate it to effect, if the patient experiences withdrawal symptoms. There may be concern about starting high doses of methadone due to potential of opioid dose related side effects. To date there are no published cases of respiratory arrest secondary to methadone use for opioid weaning. However it would appear prudent to initiate the conversion of fentanyl to methadone in the ICU environment in the event that problems arise. In our patients the fentanyl was weaned over 24 h during the initiation of the methadone dosing. Additional methadone doses if required were given under the guidance of PICU physician. Another important component of the methadone weaning involved the post ICU care of the patient. Often the patients are transferred to the floor on methadone. Our practice was to write an order for the complete methadone weaning plan usually over 1 4 weeks. In the ward the residents discussed the patients with the clinical pharmacist who was well acquainted with the patients and the protocol, or contacted the PICU team for clarification or guidance if needed. There are some limitations on this study. First, it is a retrospective study. Secondly, the use of benzodiazepines was uncontrolled and may have confounded our results. Additional benzodiazepines were given on an as needed basis if the patient appeared agitated (usually after supplemental methadone had been tried) and this dosing gradually decreased with time. However benzodiazepines were used to a similar extent in both groups.

6 810 R. SIDDAPPA ET AL. In conclusion, our highest risk factor for opioid withdrawal in methadone protocol patients was an inadequate dose of methadone. Methadone is effective in minimizing the withdrawal symptoms. A methadone dose of 2.4 times the fentanyl dose day is effective in minimizing the withdrawal symptoms. We recommend a methadone dose of 2.5 times the daily fentanyl dose (mg) as the starting dose of methadone for opioid withdrawal protocols. References 1 Marx CM, Rosenberg DI, Ambuel B et al. Pediatric intensive care sedation: survey of fellowship training programs. Pediatrics 1993; 91: Quinn MW, Wild J, Dean HG et al. Randomised double-blind controlled trial of effect of morphine on catecholamine concentrations in ventilated pre-term babies. Lancet 1993; 342: Anand KJS, Hickey PR. Halothane-Morphine compared with high-dose sufentanil for anesthesia and postoperative analgesia in neonatal cardiac surgery. N Engl J Med 1992; 326: Puntillo K, Casella V, Reid M. Opioid and benzodiazepines tolerance and dependence: application of theory to critical care practice. Heart & Lung 1997; 26: Yaster M, Sabine KB, Berde C et al. The management of opioid and benzodiazepine dependence in infants, children and adlolescents. Pediatrics 1996; 98: Katz R, Kelly HW, His A. Prospective study on the occurrence of withdrawal in critically ill children who receive fentanyl by continuous infusion. Crit Care Med 1994; 22: Robertson CR, Darsey E, Fortenberry JD. Evaluation of an opiate-weaning protocol using methadone in pediatric intensive care unit patients. Pediatr Crit Care Med 2000; 1: Meyer MT, Berens RJ. Efficacy of an enteral 10-day methadone wean to prevent opioid withdrawal in fentanyl-tolerant pediatric intensive care unit patients. Pediatr Crit Care Med 2001; 2: Holder EL, Leckman JF, Ehrenkranz R et al. Clonidine in neonatal narcotic-abstinence syndrome. N Eng H Med 1981; 305: Deutsch ES, Nadkarni VM. Clonidine prophylaxis for narcotic and sedative withdrawal syndrome following laryngotracheal reconstruction. Arch Otolaryngol Head Neck Surg 1996; 122: Tobias JD. Subcutaneous administration of fentanyl and midazolam to prevent withdrawal after prolonged sedation in children. Crit Care Med 1999; 27: Deshpande JK, Tobias JD. The Pediatric Pain Handbook. St. Louis: Mosby-Year Book, Inc, 1996; Tobias JD, Deshpande JK, Gregory DF. Outpatient therapy of iatrogenic drug dependency following prolonged sedation in the pediatric intensive care unit. Intensive Care Med 1994; 20: Joseph JD, Schleien CL, Haun SE. Methadone as treatment for iatrogenic opioid dependency in pediatric intensive care unit patients. Crit Care Med 1990; 18: Anand KJS, Arnold JH. Opioid tolerance and dependence in infants and children. Crit Care Med 1994; 22: Tobias JD. Withdrawal, and physical dependency after longterm opioid analgesia of children the pediatric intensive care unit. Crit Care Med 2000; 28: Williams PI, Sarginson RE, Ratcliffe JM. Use of methadone in the morphine-tolerant burned paediataric patient. Br J Anaesth 1998; 80: Accepted 15 May 2003