A Randomised Controlled Trial of Acetaminophen Ibuprofen and
A Randomised, Controlled Trial of Acetaminophen, Ibuprofen, and Codeine for Acute Pain relief in Children with Musculoskeletal Trauma Clark et al, Paediatrics 2007 Tom Larsen, ST 4, Addenbrookes
Objectives �Clearly stated objective – to determine which of the three analgesics, given as a single dose, provides the most efficacious analgesia for children presenting to the ED with musculoskeletal injury. ◦ Common clinical problem ◦ Lack of pre-existing ED based evidence (equivocal evidence from post operative studies referenced)
Design/Methods �Single centre – Children’s Hospital of Eastern Ontario with ED attendance of 55000 per annum �Powered adequately (although power calculation subsequently found to have been done incorrectly) �Randomised – computer generated sequence �Blinded – patients, parents and researcher blinded �Inclusion criteria: 6 to 17 years old, presenting with musculoskeletal injury which occurred within 48 hrs �Exclusion criteria: Contraindication to a study drug, requiring resuscitation, open fracture, IV line in situ, recent use of a study drug, significant cognitive impairment
Recruitment �Recruitment 8 hrs/day by research assistant � 780 eligable children during this time ◦ 336 enrolled ◦ 324 refused consent ◦ 48 not approached as research assistant with another patient ◦ 38 missed ◦ 34 other reasons �? Recruitment bias
Interventions � 3 arms ◦ 15 mg/kg acetaminophen (max 650 mg) ◦ 10 mg/kg ibuprofen (max 600 mg) ◦ 1 mg/kg codeine (max 60 mg) �? Bias introduced by dosing – dose of ibuprofen seems closer to maximum safe dose.
Data collection �Children reported pain on visual analogue scale at baseline and at 30 min intervals for 120 mins �Asked at 60, 90 and 120 minutes if requiring further analgesia �Questioned for adverse events with open questioning in ED and with open questioning and checklist at 2 day telephone follow up �Patients discharged before observation period complete given materials to complete remaining scores and mail in results ◦ ? Source of bias
Follow up - ? Attrition bias
Outcome measures �Primary outcome: Change in patient’s self-reported pain from baseline at 60 minutes �Secondary outcomes ◦ Change in pain score from baseline at 30, 90, 120 mins ◦ Requirement for additional analgesia ◦ Number achieving adequate analgesia at 60 and 120 minutes (defined as VAS score <30 mm) �Unclear if outcome measures defined prospectively
Results �From 60 minutes and onwards the ibuprofen group had significantly greater improvement in pain score (no significant difference between codeine and acetaminophen at any time point) �At 60 minutes more patients in ibuprofen group achieved adequate analgesia �No significant difference in number of patients requiring additional analgesia across the groups �No significant adverse events �No significant difference in rate of minor side effects
Results - subgroups �Subgroup analysis showed ibuprofen produced greater improvement in patients with fractures �No significant difference between the three medications in patients with soft tissue injuries
Analysis �Appropriate statistical analysis of baseline and outcome data �Data analysed on a per protocol basis ◦ Included if received dose, had baseline data and had primary outcome data ◦ ? attrition bias �Subsequent intention to treat analysis not presented although stated to be similar to the per protocol analysis presented
Discussion �Despite ibuprofen providing greater reduction in pain score only 52% were defined as having adequate analgesia with ibuprofen alone and there was no significant difference in the requirement for further analgesia between the groups �Concerns regarding effects of NSAIDS on fracture healing �Possible that more adequate dosage of paracetamol may have improved outcomes in this group, particularly in older children
Conclusion �Well designed study providing good evidence for the superiority of ibuprofen as a single agent for analgesia after musculoskeletal injury in children �Difficulties with recruitment, follow up, dosing and per protocol analysis may introduce some bias.
- Slides: 13