Analgesic efficacy of sustained release paracetamol in patients with osteoarthritis of the knee

Abstract

Aims: Paracetamol is widely recommended as the initial treatment for pain associated with osteoarthritis (OA). A sustained release (SR) paracetamol formulation (Panadol Extend) was compared with standard immediate release (IR) paracetamol (Panadol) in patients with knee pain secondary to OA. The primary parameter for assessment of efficacy was patient-assessed global pain relief as determined on day 8 of the treatment period.

Methods: A double-blind, double-dummy, randomized study was conducted. Patients (n=403) were treated for 7 days with paracetamol 4 g day(-1) (SR paracetamol, two 665 mg tablets taken three times daily; IR paracetamol, two 500 mg tablets taken four times daily). Patients completed daily pain measurements and assessed global pain relief at the end of the study. Therapeutic noninferiority was defined on the basis of achieving statistical noninferiority for global pain relief.

Results: Analysis of the primary parameter for the intention to treat population showed that the difference in proportion of patients (SR-IR paracetamol) achieving a successful response on day 8 was -0.7%; 90% CI (-8.82%, 7.45%), P=0.890. For the per protocol population the difference in proportion was -3.0%; 90% CI (-11.61%, 5.66%), P=0.571. As the lower bound of the 90% CI for the treatment difference in each case was greater than the prespecified value (-15%), SR paracetamol was considered to be statistically noninferior to IR paracetamol in terms of pain relief. The treatments were not significantly different for any of the secondary parameters in either populations.

Conclusions: SR paracetamol taken three times daily was statistically and therapeutically noninferior to IR paracetamol taken four times daily in patients with knee pain due to OA. SR paracetamol may be more convenient for patients with chronic pain and has the potential to enhance compliance and therefore pain relief.

Figure 1

Disposition of patients.

Figure 2

Treatment differences (percentage of patients achieving a successful response with SR paracetamol – percentage of patients achieving a successful response with IR paracetamol) and 90% confidence intervals for patient-assessed global pain relief as evaluated for the per protocol and intention to treat populations. The two formulations were concluded to be statistically noninferior since the lower bound of the 90% confidence intervals for the treatment differences was greater than −15% (the prespecified value used to define noninferiority), in each population.