Transdermal fentanyl in the management of children with chronic severe pain

Results from an international study

Authors

  • Julia C. Finkel M.D.,

    Corresponding author
    1. Department of Anesthesiology, George Washington University, Washington, DC
    2. Department of Pediatrics, George Washington University, Washington, DC
    3. Anesthesia Pain Management Service, Children's National Medical Center, Washington, DC
    4. Pediatric Pain Program, David Geffen School of Medicine, University of California–Los Angeles, Los Angeles, California
    • Anesthesia Pain Management Service, Children's National Medical Center, 111 Michigan Avenue, NW, Washington, DC 20010===

    Search for more papers by this author
    • Fax: (202) 884-5999

  • Allen Finley M.D.,

    1. Department of Pediatrics, George Washington University, Washington, DC
    2. Department of Anesthesia, Dalhousie University, Halifax, Nova Scotia, Canada
    3. Department of Psychology, Dalhousie University, Halifax, Nova Scotia, Canada
    Search for more papers by this author
  • Christine Greco M.D.,

    1. Pediatric Pain Management, IWK Health Centre, Halifax, Nova Scotia, Canada
    Search for more papers by this author
    • Dr. Greco acts as a consultant for Janssen Pharmaceuticals.

  • Steve J. Weisman M.D.,

    1. Pain Treatment Services, Children's Hospital, Boston, Massachusetts
    2. Pain Management, Children's Hospital of Wisconsin, Milwaukee, Wisconsin
    3. Department of Anesthesiology, Medical College of Wisconsin, Milwaukee, Wisconsin
    Search for more papers by this author
  • Lonnie Zeltzer M.D.

    1. Department of Pediatrics, Medical College of Wisconsin, Milwaukee, Wisconsin
    2. Department of Pediatrics, David Geffen School of Medicine, University of California–Los Angeles, Los Angeles, California
    3. Department of Anesthesiology, David Geffen School of Medicine, University of California–Los Angeles, Los Angeles, California
    4. Department of Psychiatry, David Geffen School of Medicine, University of California–Los Angeles, Los Angeles, California
    5. Department of Biobehavioral Sciences, David Geffen School of Medicine, University of California–Los Angeles, Los Angeles, California
    Search for more papers by this author
    • Dr. Zeltzer has acted in the past as a consultant for Janssen Pharmaceuticals.


Abstract

BACKGROUND

The current study was conducted to assess the safety and tolerability of a transdermal fentanyl delivery system for the relief of chronic pain in a pediatric population, and also to validate titration recommendations and conversion to transdermal fentanyl from oral opioid therapy.

METHODS

This 15-day (with 3-month extension), single-arm, open-label trial was conducted at 66 sites in 10 countries. A total of 199 pediatric patients (ages 2–16 years) with both malignant and nonmalignant conditions who were receiving oral or parenteral opioids for moderate to severe chronic pain were enrolled. Transdermal fentanyl doses were titrated upward according to the rescue medication consumed during the previous application period. Degree of pain was assessed by patients and parents/guardians using visual and numeric scales. Level of play and quality of life were assessed using the Play Performance Scale (PPS) and the Child Health Questionnaire (CHQ). Adverse events were monitored on Days 1–15. Hypoventilation and sedation were monitored every 4 hours during the first 72 hours of the study.

RESULTS

A total of 173 patients completed the primary treatment period and 130 entered the extension phase. The average daily pain intensity scores were reported to have decreased by Day 16 and improvements in the mean PPS scores were observed to the end of the extension period. The CHQ scores demonstrated improvements in 11 of 12 domains after Month 1 of the extension period.

CONCLUSIONS

Transdermal fentanyl was found to be a safe and well tolerated alternative to oral opioid treatment for children ages 2–16 years who were previously exposed to opioid therapy. Cancer 2005. © 2005 American Cancer Society.

In both malignant and nonmalignant conditions, persistent pain and fear of pain is a recognized source of anxiety to children and their families. Nevertheless, severe pain in children frequently goes untreated.1

Pure μ-opioid agonists have demonstrated unquestioned analgesic efficacy in adults and are often the only effective method with which to relieve a patient's chronic pain2; however, to our knowledge, few studies have been conducted to date to provide evidence of their safety and efficacy in children.3, 4 In addition, routes of drug administration that are acceptable for acute pain or for use in adults may be distressing to children with chronic pain.1

Past use of analgesia in pediatric patients suggests that young children were believed to feel pain less severely than adults,5 and also that they would be predisposed to addiction.6, 7 Young children are often more sensitive to pain than adults.8 Chronic pain in children is now known to be a direct cause of cognitive and emotional disturbance, as a result of anxiety, sleep disturbances, physical disability, restriction of daily activities, and social withdrawal.1, 9–11 Furthermore, it is now recognized that children are not more easily addicted to opioids than adults.12 In addition, children are perhaps more vulnerable than adults to the negative effects of pain. There is evidence that pediatric stress can influence the behavior and physiologic function of individuals for life8, 13, 14 and that sensitization of nociceptive pathways early in life can lead to increased pain perception in the future.1, 15–17

The subcutaneous, intravenous, and even oral administration of opioids can be a source of suffering and distress in some children, particularly those requiring chronic opioid use for severe illnesses.18 In this patient group, transdermal delivery might be preferable, and once titrated and at steady state should demonstrate efficacy equal to more conventional routes of administration. In adults, the transdermal delivery of fentanyl has shown significant advantages over the oral and parenteral routes in the treatment of chronic pain, in terms of ease of delivery, duration of symptom relief, incidence and severity of side effects, and quality of life benefits.19–21

The objectives of the current study were to determine the safety and tolerability of transdermal fentanyl in children with chronic severe pain, and to validate titration recommendations and initial dose conversion from oral opioid therapy.

MATERIALS AND METHODS

Study Design

The current study was a single-arm, nonrandomized, open-label, multicenter trial conducted in accordance with the principles of Good Clinical Practice and the Declaration of Helsinki at 66 sites in 10 countries (Australia, Austria, Brazil, Canada, Costa Rica, Israel, Mexico, Poland, Slovakia, and the U.S.). The primary study was conducted over 15 days and was followed by a 3-month extension period.

Patients

The trial enrolled a total of 199 pediatric patients with moderate to severe chronic pain because of malignant (n = 132 patients) or nonmalignant (n = 67 patients) disease. In patients with malignancy, 45.8% had metastatic disease, 32.1% had localized disease, 17.6% had regional involvement, and 4.6% had an unclassified extent of disease. The most common diagnoses of malignancy were leukemia (27.3%), “other” malignancies (18.2%), sympathetic nervous system malignancies (14.4%), and bone malignancies (12.9%).

Male and female patients were included if they were at least age 2 years and age younger than 16 years. Patients must have received opioids continuously for a minimum of 7 days prior to enrollment, with a projected need for continuous opioid treatment for at least the length of the primary treatment period.

Patients also must have received the equivalent of at least 30 mg of oral morphine on the day before enrollment. The main exclusion criteria were a known sensitivity to fentanyl, other opioids, or adhesives; skin disease that affected the application or local tolerance of or absorption from the fentanyl patch; or a life expectancy of fewer than 15 days. In all cases, the child's parent, guardian, or legal representative was required to sign an informed consent form. In those cases in which a child was able to understand the purpose and implications of the trial, his or her consent also was sought.

Methods

Patients were converted from oral or parenteral opioids to transdermal fentanyl on Day 1. The opioid analgesic requirement was calculated from opioid use during the previous 24-hour period. The equivalent analgesic oral morphine dose was calculated using the equianalgesic potency conversion table (Table 1). The equivalent transdermal fentanyl dose was calculated using a second conversion table (Table 2).22

Table 1. Equianalgesic Potency Conversion Table
NameEquianalgesic dose (mg)
IntramuscularaOral
  • p.r.: per rectum.

  • a

    Based on single-dose studies in which an intramuscular dose of each drug was compared with morphine to establish the relative potency. Oral doses are recommended when changing from a parenteral to an oral route.36

  • b

    The conversion ratio of 10 mg of parenteral morphine equaling 30 mg of oral morphine is based on clinical experience in patients with chronic pain. The conversion ratio of 10 mg of parenteral morphine equaling 60 mg of oral morphine is based on potency in patients with acute pain.37

  • c

    Methadone should be withdrawn the day prior to patch application.

  • d

    It is advisable to apply the transdermal fentanyl patch at the time of the next scheduled dose of tramadol.

Morphine10b30 (60)
Hydromorphone1.57.5
Oxycodone1530
Levorphanol24
Oxymorphone110 (p.r.)
Meperidine75
Codeine130200
Methadonec1020
Tramadold100120
Table 2. Recommended Initial Dose of Transdermal Fentanyl Based on the Daily Oral Morphine Dose
Oral 24-hour morphine dose (mg/day)Transdermal fentanyl dose
30–4412.5
45–13425.0
135–18037.5
181–22450.0
225–27062.5
271–31475.0
315–36087.5
361–404100.0
405–450112.5
451–494125.0
495–540137.5
541–584150.0
585–630162.5
631–674175.0
675–720187.5
721–764200.0
765–810212.5
811–854225.0
855–900237.5
901–944250.0
945–990262.5
991–1034275.0
1035–1080287.5
1081–1124300.0

Dose application

Transdermal fentanyl patches (at doses of 12.5μg/hr, 25μg/hr, 50μg/hr, 75μg/hr, or 100 μg/hr) were applied to nonirritated and nonirradiated skin on a flat surface, such as the chest, back, flank, or upper arm.

Dose titration

After the initiation of transdermal fentanyl treatment, doses were titrated upward (with a frequency of no less than every 3 days) using the dose calculated from the conversion table, until steady-state analgesic effectiveness with transdermal fentanyl alone was obtained. Titration was based on the requirement for immediate-release morphine to counter any residual or breakthrough pain experienced by the patient. The dose of transdermal fentanyl was increased by 12.5 μg/hour for every 45 mg of morphine equivalents (ME) of rescue medication consumed on Day 2 or Day 3 of the previous patch application period, up to a maximum increase of 25 μg/hour at any 1 time.

Evaluations

Clinical effectiveness

A global assessment of pain treatment was made using a four-point scale on Days 1 and 16 of the primary treatment period. Treatment was rated as poor, fair, good, or very good by the patient's parent or guardian. The degree of pain was measured by patients age 6 years and older using a colored, vertical Visual Analogue Scale (VAS) ranging from 0 (no pain) to 10 (most pain). Assessments were made by the parent/guardian of each patient using a Numeric Pain Intensity Scale, ranging from 0 (no pain) to 10 (worst possible pain). Scores were recorded twice each day (morning and evening) in a diary. Pain levels also were recorded at the time rescue medication was given and 1 hour later by the parent/guardian only.

The level of play and activity of the patient was monitored every 3 days of the primary treatment period from Day 1 to the endpoint (coinciding with the patch change) by the parent/guardian using the Play Performance Scale (PPS) modified from the Karnofsky Scale for Global Functional Status of Adults.23, 24 Level of activity was described in terms of active play, quiet play, degree of physical limitations, and degree of independence. Ratings ranged from 80–100 (“fully active, normal”) through 50–70 (“mild to moderately restricted”) to 0–40 (“moderate to severe restriction”) and were measured in terms of the individual child's baseline rating.

The overall quality of life of the patients, including their ability to function physically and socially, was monitored using the Child Health Questionnaire (CHQ).25 The CHQ is designed for self-completion by the patient, if age 10 years or older, and their parent/guardian. Questionnaires were completed at baseline and at the end of Months 1 and 3 of the extension period.

The parent portion of the CHQ (CHQ-PF50) contains 50 items in 12 domains and was completed by all parents/guardians of patients ages 5–16 years. The 12 domains are: physical function, role/social emotional behavior, role physical, bodily pain, general behavior, mental health, self-esteem, general health, parent-time impact, parent-emotional impact, family activities, and family cohesion. The child portion of the CHQ (CHQ-CF87) contains 87 items in 10 domains: physical function; role/social functioning because of emotional, behavior, and physical problems scales; bodily pain; general behavior; mental health; self-esteem; general health; and family activities. The questionnaire was completed by patients ages 10–16 years. Scores for both questionnaires ranged from 0–100, with a high score reflecting better functioning and well-being.

Safety

Adverse events were monitored in all patients who had received the trial medication and were reported between the first and last dose of the trial medication. The incidence of adverse events and serious adverse events determined to be related to the study medication were defined by investigators with an assessment of “doubtful,” “possible, ” “probable, ” or “very likely.” Vital signs (including systolic and diastolic blood pressure, heart rate, respiratory rate, and temperature) were obtained and recorded at baseline and each morning on Days 2, 3, 4, 7, and 16. Assessments then were made monthly during the extension period. A physical examination was performed at baseline, if possible on Day 16, and monthly during the extension period.

The patients' respiratory rates and sedation levels were monitored every 4 hours during the first 72 hours after application of the first transdermal fentanyl patch. Any occurrence of bradypnea was reported as a serious adverse event. The Observer's Assessment of Alertness/Sedation Scale26 was used to grade the level of alertness.

Statistical Methods

The current study was an open-label and single-arm trial, so therefore no formal inferential statistical tests were performed. Sample size determination was based on clinical judgment rather than statistical consideration.

An intent-to-treat (ITT) analysis was performed whereby all enrolled patients, regardless of their compliance with the protocol, were included in the analysis unless they received no medication at all. The ITT included all patient data for the primary treatment period and any extension data obtained by the data cutoff date. Calculations were performed using the SAS® software (Version 6.12; SAS Institute, Inc., Cary, NC).27, 28

All effectiveness parameters (i.e., global assessment, pain level, play performance, CHQ, dosing and titration information, and rescue medication) were summarized descriptively by timepoint for the primary treatment period only, overall, and by gender and age category. CHQ, dosing and titration information, and rescue medications were also summarized for the extension period.

Play performance

The Pearson correlation between play performance, pain score, and average daily dose at each timepoint of the primary treatment period (i.e. baseline, Day 4, Day 7, Day 10, Day 13, Day 19, and endpoint) was calculated. In addition, correlation also was examined using change scores at each timepoint from baseline.

CHQ

The 12 domain scores from the CHQ-PF50 were standardized using the mean and standard deviation (SD) from the U.S. general population and clinical sample data. The scores then were aggregated using factor score coefficients from the U.S. general population and clinical sample data. The association between CHQ changes from baseline and other effectiveness parameters (pain as reported by parent and child, global satisfaction as reported by the parent, and play performance as reported by parent or guardian) was explored using Pearson correlation coefficients.

Dosing and titration

The following information was collected from patient diaries: average ME opioid taken within the last 24 hours before the initiation of transdermal fentanyl treatment, the duration of treatment with transdermal fentanyl, the average daily dose of transdermal fentanyl, the number of days until the first dose titration was required, the number of days until subsequent titrations were required, and the average titration dose.

RESULTS

Patient Status

A total of 199 patients were enrolled into the study and entered into the ITT population. Of these, 173 patients (86.9%) completed the primary treatment period and 130 patients (65.3%) entered the extension period. At the time of the database cutoff, 26 patients remained in the extension period.

Patient Demographics

Of the 199 ITT patients, 118 (59.3%) were male and 81 (40.7%) were female. The majority of patients were white (54.8%). The mean age of the patients was 10.7 ± 0.28 years and the median age was 12 years (range, 1–16 yrs). Three patients fell outside of the desired age range (1 patient was age 1 year and 2 patients were each age 16 years) and were not included in age category data analyses.

Specification of pain type included nociceptive (70.9%), neuropathic (14.6%), and multiple pain (14.1%) etiologies. An underlying malignancy was present in 132 patients and was found to be related to the cause of pain for 96 patients.

Patient Withdrawal

A total of 26 patients (13.1%) were withdrawn from the study during the primary treatment period: 6 (3.0%) as a result of death, 6 (3.0%) because of adverse events, and an additional 14 patients (7.0%) for various other reasons including a decrease in pain; the patient withdrew; and the patient was ineligible, noncompliant, or demonstrated an insufficient response. Of the 130 patients who entered the extension treatment period, 104 (80.0%) discontinued treatment. The most frequent cause for treatment discontinuation (n = 43 patients [33.1%]) was “other,” mostly comprising patients whose pain had improved or resolved, or who were being weaned off opioids. Additional “other” reasons included increased pain (two patients) and the patient requiring an increase in the frequency the patch was changed (two patients). The next most frequent causes for discontinuation were death (n = 21 patients [16.2%]) and adverse events (n = 11 patients [8.5%]).

A total of 48 patients (24.1%) had death as an adverse event during the combined primary and extension treatment periods. The most common diagnoses leading to death were neuroblastoma (eight patients), malignant neoplasm (six patients), and sarcoma (eight patients). None of these events were considered by investigators to be related to treatment with transdermal fentanyl, apart from one case of neuroblastoma for which the association with the trial medication was considered “doubtful.”

Treatment Initiation

Examination of the medical histories revealed that all patients had previously received at least one opioid analgesic. Morphine, including morphine sulfate, was the opioid previously used by most patients (70.4%). Fentanyl was the second most frequently used opioid (16.1%).

The most common starting dose of transdermal fentanyl was 25 μg/hour, which was required by 90 patients (45.2%). The lowest starting dose, 12.5 μg/hour, was considered appropriate for 59 patients (29.6%). The median recorded initial dose was 25 μg/hour (range, 12.5–175.0 μg/hour). The location of the fentanyl patch varied between applications, although the upper back was used most often (in 63.3% of patients). Tape was used at least once to adhere the patch to the application site in 81 patients (40.7%).

Dose Titration

The average duration of treatment with transdermal fentanyl in the primary treatment period was 14.8 ± 0.25 days in the ITT patient group. Seventy-seven patients (38.7%) required an upward dose titration, at an average time to first dose titration of 5.6 ± 0.30 days. The mean initial daily dose of transdermal fentanyl per kilogram (kg) of body weight was 0.98 ± 0.057 μg/kg/hour, which rose to an average final dose of 1.20 ± 0.091 μg/kg/hour at the end of the primary treatment period.

A total of 84.9% of patients received at least 1 rescue medication, with a mean oral ME of 1.35 ± 0.16 mg/kg during the primary treatment period. Overall, the mean average daily dose of opioid (i.e., transdermal fentanyl plus rescue medication) was 4.72 ± 0.31 mg/kg, expressed as MEs. The average ratio of the total daily transdermal fentanyl dose versus the total daily dose of opioid increased over the primary treatment period from 0.88 ± 0.01 mg/kg to 0.91 ± 0.01 mg/kg. The trend was for the reduced rescue medication to level off after 5 days, suggesting the rapid optimization of the transdermal fentanyl dose. The overall mean dosage of total opioids and rescue medication by body weight during the primary treatment period are summarized in Table 3.

Table 3. Use of Rescue Medication during the Primary Treatment Period, by Body Weight, Expressed in Morphine Eequivalents
  Body weight category (quartiles)
Total1st2nd–3rd4th
n = 199n = 54n = 97n = 47
  1. SE: standard error.

No. assessed198 (99.5%)54 (100%)97 (100%)47 (100%)
Overall mean daily dose in mg/kg (SE) of opioid (fentanyl + rescue medication)4.72 (0.306)6.45 (0.750)4.26 (0.378)3.67 (0.470)
No. of patients requiring rescue medication during the primary treatment period169 (84.9%)45 (83.3%)82 (84.5%)41 (87.2%)
Mean daily dose of rescue medication in mg/kg (SE)1.35 (0.163)1.88 (0.501)1.17 (0.167)1.15 (0.174)

Evaluations

Clinical effectiveness

Global assessment of pain treatment.

The global assessment of pain treatment at the baseline and endpoint is summarized in Table 4. A global assessment was performed at both baseline and Day 16 in a total of 145 patients. Of this group, 54 patients (37.2%) who had prebaseline pain treatment rated as poor or fair had their treatment with transdermal fentanyl rated as good or very good. For 73 patients (50.3%) whose original treatment was rated as good or very good, no reduction in satisfaction with therapy at the end of the primary treatment period was reported. The prebaseline and transdermal fentanyl treatment of 14 patients (9.7%) was rated as poor or fair, and 4 patients (2.7%) had their pain relief rated as worse by Day 16. There appeared to be no age-related or gender-related differences with regard to response to treatment.

Table 4. Summary of Global Assessment of Pain Treatment at the Baseline and Endpoint
ParameterTotal (n = 199) No. (%)
  • Percentages are based on the number of patients assessed at a given timepoint.

  • a

    Baseline: Day 1. The endpoint is defined as the last nonmissing, postbaseline observation through the last day of administration of the trial medication during the primary treatment period.

Baselinea 
No. assessed189
Very good34 (18.0)
Good66 (34.9)
Fair58 (30.7)
Poor31 (16.4)
Endpointa 
No. assessed149
Very good78 (52.3)
Good52 (34.9)
Fair15 (10.1)
Poor4 (2.7)
Pain level.

The average daily pain intensity levels reported by parents/guardians using the numeric pain scale for the ITT population decreased steadily throughout the study period from 3.5 ± 0.23 at baseline to 2.6 ± 0.21 by Day 16, a change of 0.9 points from baseline. The trend was similar for both morning and evening intensity levels, although some convergence of the pain intensity scores was apparent between the two assessment time points (Fig. 1). Comparable changes in daily pain intensity levels were reported by patients using the VAS, with a reduction from 3.7 ± 0.26 at baseline to 3.1 ± 0.26 by the end of the primary treatment period, a change of 0.6 points from baseline.

Figure 1.

Mean (standard error) morning and evening pain intensity levels at baseline (Day 1) and Day 16 as reported by parents/guardians.

PPS.

Parent/guardian-rated improvements in mean PPS scores were observed from baseline (41.22 ± 1.68) to the data collection endpoint (53.80 ± 1.91), representing a mean change of 11.5%. This represents a change from “mostly in bed and participating only in quiet activities” to “gets dressed but lies around much of the day, no active play but is able to participate in quiet play.”

Correlation analysis revealed a positive association between pain (as reported by the patient or parent/guardian) and functional status, such that functional status improved as pain decreased. A positive correlation between PPS scores and the global assessment of pain treatment also was apparent. (However, it should be noted that the correlation analysis was not powered and therefore there is a possibility that these correlations could reflect chance occurrence). At the endpoint, the mean PPS scores of the patients whose parents/guardians rated their transdermal fentanyl patch treatment as “very good” was 61.5 ± 2.84 (n = 75 patients; baseline, 47.7 ± 3.61 [n = 31 patients]), whereas the PPS scores of those patients whose parents/guardians had rated their pain treatment as “poor” at the endpoint was 12.5 ± 2.50 (n = 4 patients; baseline, 34.0 ± 3.56 [n = 25 patients]).

CHQ.

The baseline results of the CHQ parent form were compared with those of a normative population sample (n = 391 participants) of children in the U.S.29 Patients demonstrated significant impairment in their functioning at baseline compared with the accepted norm for children in this age group. At baseline, 78% of patients were rated by their parents/guardians as “limited a lot” in activities requiring “a lot” of energy and 74% were reportedly “limited a lot” in activities requiring “some” energy. Approximately 43% of the parents/guardians reported that their child had “severe pain” during the 4-week period before the baseline assessment, and 63% reported that their child had pain “almost every day.” The results from the CHQ child form were comparable to those of the parent/guardian-completed form, and both forms demonstrated excellent correlation with the PPS scores.

At the end of Month 1 of the extension phase, the parents of the participating patients (n = 36 patients) reported improvement in 11 of the 12 domains assessed. The largest improvement was noted in bodily pain (29.52 ± 4.52; baseline, 18.14). Five other domains demonstrated an improvement of greater than 5 points from baseline: mental health (8.28 ± 2.76; baseline, 54.33), family activities (6.96 ± 3.19; baseline, 43.04), role emotional behavior (12.36 ± 6.08; baseline, 34.72), physical function (7.15 ± 2.71; baseline, 23.65), and role physical (13.82 ± 5.76; baseline, 17.07). The mean change for other domains ranged from 0.28–3.57. At the end of Month 3, participating patients continued to demonstrate sustained improvements in 11 of the 12 domains, with the greatest improvements from baseline being reported in bodily pain (27.14 ± 7.53), role physical (23.53 ± 8.09), physical functioning (13.89 ± 5.74), and role/social behavior (16.01 ± 9.29). Mean score changes from baseline to the end of Months 1 and 3 are shown in Figure 2. Results from the child form demonstrated improvements at Month 1 in the majority of the physical domains, with large improvements noted in bodily pain (26.45 ± 4.74), role physical (10.68 ± 8.33), and physical functioning (8.77 ± 3.41). Such improvements were maintained at the end of Month 3 and confirmed the findings from the parent form and PPS scores.

Figure 2.

Child Health Questionnaire. Parent report domain change from baseline at the end of Month 1 and Month 3. PF: physical functioning; RP: role physical; REB: role emotional behavior; BP: bodily pain; BE: general behavior; MH: mental health; SE: self esteem; GH: general health; PI: parent impact–time health; PE: parent impact–emotional; FA: family activities; FC: family cohesion; PSS: physical summary score; PHS: psychosocial summary score (a higher score indicates better physical or psychosocial functioning and well-being).

Safety (combined primary and extension periods)

Transdermal fentanyl was found to be safe and well tolerated in this population. Age, body weight, and initial dose by body weight had no apparent clinically relevant effects on the overall safety profile of transdermal fentanyl. Evaluation of vital signs and physical examination did not suggest any safety concerns.

One hundred eighty patients (90.5%) reported at least 1 adverse event during treatment. The most frequent adverse events were fever (n = 71 patients), emesis (n = 66 patients), nausea (n = 42 patients), headache (n = 37 patients), and abdominal pain (n = 34 patients). One hundred patients (50.3%) had adverse events with “doubtful” (15.1%), “possible” (16.6%), “probable” (5.5%), or “likely” (13.1%) correlations with the study medication. The most frequent adverse events found to be related to the trial medication were emesis (16.1%), nausea (10.6%), pruritus (7.5%), application site reaction (7.5%), headache (7.0%), constipation (6.0%), and somnolence (5.5%).

The overall incidence of serious adverse events was 43.2% (n = 86 patients). The most common serious adverse events were severe fever (n = 21 patients [10.6%]), neuroblastoma (n = 8 patients [4%]), emesis (n = 8 patients [4%]), malignant neoplasm (n = 7 patients [3.5%]), and pain (n = 7 patients [3.5%]). Serious adverse events considered by the investigator to be related to the study treatment were reported to occur in 19 patients (9.5%) and included emesis (n = 5 patients [2.5%]), pain (n = 4 patients [2%]), dyspnea (n = 3 patients [1.5%]), somnolence (n = 2 patients [1%]), and neuroblastoma (n = 1 patient [0.5%]).

No patient had a respiratory rate that indicated bradypnea occurred within the initial 72 hours of treatment. Bradypnea was defined as fewer than 12 breaths per minute for patients ages 2–6 years, fewer than 10 breaths per minute for patients ages 7–10 years, and fewer than 8 breaths per minute for children ages 11–16 years. A total of 61 patients reported at least 1 adverse event associated with the respiratory system (Table 5). With the exception of four patients, the adverse events experienced were evaluated by the investigator as not being related to treatment.

Table 5. Incidence of Adverse Events within the Respiratory System and Correlation with the Study Medication (in the ITT Population)
CorrelationOverallNoneDoubtfulPossibleProbableLikely
No. (%)No. (%)No. (%)No. (%)No. (%)No. (%)
  1. ITT: intent to treat. Includes all adverse events that occurred during the combined primary and extension treatment periods. Adverse events were summarized using the World Health Organization Adverse Reaction Terminology (WHOART) dictionary by body system and preferred term. Percentages are based on the total number of subjects in the intent-to-treat population.

Respiratory system disorders61 (30.7)48 (24.1)10 (5.0)2 (1.0)1 (0.5)0
Dyspnea11 (5.5)7 (3.5)4 (2.0)000
Rhinitis8 (4.0)5 (2.5)3 (1.5)000
Pharyngitis7 (3.5)6 (3.0)1 (0.5)000
Respiratory depression5 (2.5)4 (2.0)001 (0.5)0
Bradypnea2 (1.0)1 (0.5)01 (0.5)00
Bronchospasm2 (1.0)1 (0.5)01 (0.5)00
Respiratory disorder2 (1.0)1 (0.5)1 (0.5)000
Asthma1 (0.5)01 (0.5)000

One of the 4 patients was an 11-year-old male who began experiencing respiratory depression of moderate severity on Day 1 of treatment. He was receiving 12.5 μg/hour (0.28 μg/hr/kg) of transdermal fentanyl at the time of onset of the adverse event. The event was considered serious and most likely related to the trial medication. The respiratory depression resulted in a temporary withdrawal of the trial medication but did not require any other therapy. Vital signs on Day 2 of treatment were a heart rate of 76 beats per minute, a respiratory rate of 18 breaths per minute, systolic/diastolic blood pressure of 132/78 mm Hg, and a temperature of 38.2 °C. The patient recovered from the serious adverse event within 2 days of its onset. The trial medication was resumed on Day 2 of the study at a dose of 12.5 μg/hour (0.28 μg/hr/kg).

A case of moderate respiratory disorder (verbatim term: increased tachypnea) was experienced by a 4-year-old male, beginning on Day 1 of treatment. The patient was receiving a dose of 37.5 μg/hour (2.5 μg/hr/kg) of transdermal fentanyl at the time of onset of the adverse event. The event was not considered serious and the correlation between the event and the trial medication was considered doubtful. The patient recovered from the adverse event within 1 day of onset.

Mild bradypnea (verbatim term: decreased respiration of 10–12 breaths per minute while awake) was reported for a 12-year-old female on Day 6 of treatment with transdermal fentanyl. The patient was receiving a dose of 100 μg/hour (4.35 μg/hr/kg) at the time of the onset of the adverse event. The bradypnea was not considered serious but was considered by the investigator to be possibly related to the trial medication. The patient recovered from the adverse event within 1 day of onset. There was no interruption of the trial medication.

A 15-year-old male experienced a mild bronchospasm (verbatim term: wheezing) on Day 2 of treatment with transdermal fentanyl. At the onset of the event, the patient was receiving a dose of 25 μg/hour (0.34 μg/hr/kg) of transdermal fentanyl. The bronchospasm was not considered serious but was considered by the investigator to be possibly related to the trial medication. The patient recovered from the adverse event within 1 day of onset. There was no interruption of the trial medication.

Analysis of other adverse events commonly associated with opioid use, including gastrointestinal symptoms, raised no unexpected safety concerns (Table 6). One patient whose somnolence was “very likely” related to the study medication withdrew from the study, whereupon she recovered from this adverse event. One patient with pruritus and an abrasion that had a “likely” correlation with the study medication recovered from the pruritus but not the abrasion after study withdrawal. One patient with nausea and one with emesis believed to be “possibly” related to the study treatment withdrew from the study and both patients recovered. One patient with severe nausea and emesis, both of which were considered to “possibly” be related to treatment, withdrew but the adverse events continued.

Table 6. Incidence of Adverse Events Commonly Associated with Opioids (Somnolence, Pruritus, and Emesis and Nausea) and Correlation with the Study Medication (in an ITT Population)
CorrelationOverallNoneDoubtfulPossibleProbableLikely
No. (%)No. (%)No. (%)No. (%)No. (%)No. (%)
  1. ITT: intent to treat. Includes all adverse events that occurred during the combined primary and extension treatment periods. Adverse events were summarized using the World Health Organzation Adverse Reaction Terminology (WHOART) dictionary by body system and preferred term. Percentages are based on the total number of subjects in the intent-to-treat population.

Somnolence14 (7.0)3 (1.5)3 (1.5)4 (2.0)3 (1.5)1 (0.5)
Pruritus24 (12.1)9 (4.5)2 (1.0)5 (2.5)3 (1.5)5 (2.5)
Vomiting66 (33.2)34 (17.1)19 (9.5)10 (5.0)1 (0.5)2 (1.0)
Nausea42 (21.1)21 (10.6)9 (4.5)11 (5.5)01 (0.5)

There were 18 patients who required a change in the dosage of the trial medication because of an adverse event: 6 patients required a dose reduction (for somnolence, pruritus, or nausea), 11 patients required a dose increase (9 of which were because of pain reported as an adverse event), and 1 patient required both a dose increase and a decrease (because of different adverse events).

The mean daily sedation scores were found to decrease over time, but remained above those scores that would indicate a state of moderate alertness. The mean daily sedation scores at Day 1, Day 2, and at the endpoint were 2.65, 2.27, and 2.13, respectively.

Fifty-three patients died during the study. Apart from one patient, none of the deaths were considered to be related to treatment with transdermal fentanyl, but instead were the result of the patient's underlying disease. One patient with metastatic neuroblastoma developed severe progression of the neuroblastoma on Day 12 of the study and died the same day. The correlation between the progression of neuroblastoma and the trial medication was assessed by the investigator as “doubtful. ”

DISCUSSION

Chronic pain in children may be undertreated, leading to unnecessary suffering by the patient.1, 30, 31 The use of opioids in children has been limited by caution over other possible ill effects. The ethical dilemmas involved in properly testing opioid formulations in children have resulted in a lack of defined dosing procedures for this patient group.

In the current study, a combination of direct conversion from current treatments and upward titration depending on rescue medication requirements has produced favorable efficacy, safety, and tolerability profiles in patients who have been switched to transdermal fentanyl from oral or parenteral opioids. The data presented herein suggest that transdermal fentanyl is a well tolerated treatment for chronic pain experienced by children previously exposed to opioid therapy. The conclusions regarding dose titration should be restricted to the duration used in this study. Furthermore, the observations made in the current study were of a heterogeneous population and therefore only a generalized recommendation can be made because no specific patient population was examined.

The use of transdermal fentanyl was associated with improved scores on the patients' PPS and CHQ, compared with baseline. On the CHQ, 11 of 12 domains demonstrated some improvement at the end of Month 1 or Month 3. The greatest change was observed in the bodily pain and physical domains. Substantial improvements also were noted in the emotional behavior and mental health domains. Although such comprehensive improvements might be attributable to the confounding factors of an open-label, single-arm trial and, for some measures, small sample size, these positive results nevertheless indicate a satisfactory quality of life for these pediatric patients receiving transdermal fentanyl. This finding is especially important for those patients in the study who were in palliative care and nearing end of their life. Many children near the end of life have difficulty swallowing and therefore oral medication as the sole means of analgesia can limit good pain control. Similarly, good standards for palliative care suggest limiting invasive procedures that provide analgesia through intravenous, intramuscular, or subcutaneous routes of delivery.32 For these reasons, a transdermal delivery system is ideal for this seriously ill pediatric population.

Approximately 60% of the patients in the current study did not require upward titration during the primary treatment period, suggesting the conversion table used in the current study was appropriate and that the converted doses would allow a relevant evaluation of safety.

Greater than 80% of the total daily opioid requirement was delivered by transdermal fentanyl, immediately after conversion to the patch. The use of rescue medication reached a stable daily total within two patch applications, indicating that the proposed titration algorithm can be used successfully. The majority of patients requiring dose adjustments achieved adequate analgesia within 6 days. By the end of the primary treatment period, 90% of the total daily opioid dose was being provided by transdermal fentanyl.

The adverse events observed in the current study are consistent with those of a potent opioid and a study population with serious and often progressive and life-threatening disease. There were no adverse events associated with transdermal fentanyl that indicated a specific risk in this pediatric population. The pattern of medication-related adverse events was similar to that observed in adult patients and is the expected profile of an opioid analgesic. Because the protocol had a minimum requirement for baseline opioid treatment of 30 mg/day ME, young children qualifying for the study on this basis received a high dose of opioid per kg of body weight compared with some older children in the study (body weight categories ranged from 6.45 mg/kg3 in the 1st quartile to 3.67 mg/kg3 in the 4th quartile). However, conversion to transdermal fentanyl at correspondingly high doses appeared to produce no adverse consequences in younger patients who had demonstrated tolerance to equivalent opioid doses.

There was an overall improvement in pain relief and treatment satisfaction associated with the administration of transdermal fentanyl over the primary treatment period. The convergence of pain intensity scores between morning and evening may suggest that blood levels of fentanyl remained fairly constant during the treatment period. This possibility would have to be examined specifically. Continuous pain relief may reduce the need for additional analgesics and also reduce sleep disturbances caused by the recurrence of pain.33

The transdermal delivery of fentanyl has the potential advantages of being a noninvasive approach and being able to be administered in cases in which oral administration is difficult because of advancing disease or poor compliance. Use of the transdermal route also has resulted in a decreased incidence of constipation and nausea in adult patients being treated for cancer pain.21 A previous study of transdermal fentanyl in 13 pediatric patients with opioid-dependent and stable chronic cancer pain demonstrated a decrease in gastrointestinal disturbance, with some patients able to stop taking laxatives or antiemetics.18

Results from global assessments of pain treatment, safety, and quality of life indicate that transdermal fentanyl is an acceptable alternative to oral opioid therapy in children. Transdermal fentanyl is especially useful for good pain management in children with life-limiting conditions in whom oral or injectable routes of drug delivery are difficult to administer or would add further distress. The results also suggest that the dose conversion and titration guidelines proposed are appropriate for use in this population. The data produced by this study also may be applicable to adult patients in clinical practice, particularly when there is a requirement for a low starting dose of transdermal fentanyl. The sustained duration of pain relief and the low frequency of dosing with transdermal fentanyl both serve to improve patient compliance and increase acceptability of the treatment.34, 35

Ancillary