Subjects and Methods
Subjects
The study was conducted in accordance with the Declaration of Helsinki (including the update of Edinburgh, 2000) and Good Clinical Practice guidelines. The study protocol was approved by an independent Ethics Committee. The patients received oral and written information about the intentions and nature of the study. Written informed consent was obtained from all patients.
Patients of either sex aged 18−55 years presenting with nonspecific acute low back pain lasting for no more than 5 days, with intensity rated at least 5 on an 11-point numerical rating scale and requiring medical treatment, were eligible for the study. Exclusion criteria were: pregnant or breastfeeding women; previous episode of low back pain within the last 6 months; specific spinal pathology or symptoms related to other pathologies; intake of analgesics within the 3 hours preceding study inclusion; intake of antiepileptics, antidepressants, barbiturates, anxiolytics or muscle relaxants within the 24 hours preceding inclusion; scheduled daily intake of analgesics, antiepileptics, antidepressants, barbiturates, anxiolytics and muscle relaxants; concomitant treatment with anticoagulants or platelet-aggregation inhibitors; contraindication or allergy to study drugs; concomitant medications interacting with the study drugs; alcohol/drug abuse or dependency; frequent episodes of gastrointestinal disorders, oedema, dizziness or headache; and history of aspirin-induced asthma.
Study Design
The study was conducted double-blind with treatments allocated randomly to two parallel groups of patients at multiple centres (six centres in Russia). The target was to enrol 220 patients (110 per treatment group). A computer-generated randomisation schedule assigned treatments in equal ratio to sequential patients. After providing informed consent, patients were assigned to the next consecutive patient number in a sequential ascending order. A double-dummy technique was used to blind the patients and the personnel conducting the study because of the difference in drug appearance. The following treatments were administered over a 7-day period: lornoxicam 8mg quick-release formulation or diclofenac potassium 50mg. On day 1, patients received in the clinic, no later than 2pm, two tablets of either lornoxicam or diclofenac potassium and two tablets of matching placebo. The second dose was administered in the evening of day 1 as one tablet of either lornoxicam or diclofenac potassium and one tablet of matching placebo. From day 2 to day 7, patients received one tablet of either lornoxicam or diclofenac potassium and one tablet of matching placebo in the morning and in the evening, except on day 7, when only the morning dose was administered. In cases of inadequate pain relief, patients were allowed to receive paracetamol (acetaminophen) as rescue analgesic.
Study Protocol
On day 1, after the first dose of study medications, patients stayed in the pain clinic for an in-house period of 90 minutes. During this period, time to onset of PAR, pain intensity, PAR scores and global evaluation of the study medication were recorded and patients were trained in the use of pain scales and a stopwatch. Patients were then discharged from the pain clinic and continued to record hourly their pain intensity and PAR in a patient diary for a total of 8 hours following the first drug administration. From day 2 to day 6, pain intensity was recorded by the patient every evening, 1 hour after the second daily dose of study medications. Time of study medication administrations, intake of rescue medication, ability to perform daily activities and adverse events were also recorded in the patient diary. Patients returned to the pain clinic on day 7 for a follow-up visit.
During the in-house period, patients were allowed to walk, sit or lie down (sitting in a stooping position was not recommended). From discharge to the end of the study, patients were advised to avoid bed rest, refrain from subjecting their backs to heavy loads or undertake other heavy workloads, and maintain their usual activities as much as possible.
Outcomes
The primary efficacy outcome was time to onset of PAR as measured by the stopwatch method during the 90-minute in-house period. The stopwatch was started simultaneously with the first dose of study medication and placed immediately nearby or given to the patient. Patients stopped the watch when they first detected pain relief. The time elapsed was defined as the time to onset of PAR.
Secondary efficacy outcomes included:
Pain intensity recorded at baseline (T0) and at 15, 30, 45, 60 and 90 minutes post-dose inhouse and hourly from 3 until 8 hours during day 1. From day 2−6 pain intensity was recorded every evening 1 hour after intake of the second daily dose from day 2 to day 6. Pain intensity was assessed using an 11-point numerical rating scale (where 0 = no pain and 10 = unendurable pain) and expressed as pain intensity differences (PID) from baseline and sum of pain intensity differences (SPID) from baseline calculated over the following time intervals: T0−T90 min, T3h−T8h, T0−T8h and day 1−day 6, peak PID and time to peak PID over the T0- to T90-min interval.
PAR recorded 15, 30, 45, 60 and 90 minutes post-dose and hourly from 3 until 8 hours during day 1 using a 5-point verbal rating scale (where 0 = none, 1 = a little, 2 = some, 3 = a lot, 4 = complete). From day 2−6 PAR was recorded every evening 1 hour after intake of the second daily dose. PAR is expressed as total pain relief (TOTPAR) calculated over the following time intervals: T0−T90 min, T3h−T8h and T0−T8h, peak PAR and time to peak PAR over the T0- to T90-min interval.
Use of rescue medication as assessed by the number of patients requiring rescue medication, number of paracetamol tablets used, and time to intake of rescue medication.
Ability to perform daily activities assessed every evening after pain assessments.
Global evaluation of the study medication assessed on day 1 at the end of the in-house period and on day 7 using a 5-point verbal rating scale (where 1 = poor and 5 = excellent).
Adverse events and concomitant use of medication were recorded throughout the study. An adverse event was defined as any untoward medical occurrence irrespective of causal relationship to the treatments. For each adverse event, the severity, outcome and causal relationship to treatment were recorded.
Statistics
The primary objective was to demonstrate non-inferiority of lornoxicam compared with diclofenac potassium in terms of time to onset of PAR. The null hypothesis was that time to onset of PAR with lornoxicam would be ≥20% longer than with diclofenac potassium. With an expected time to onset of PAR of approximately 30 minutes, the 20% limit (non-inferiority margin) corresponded to approximately 6 minutes. Assuming no difference in time to onset of PAR between lornoxicam and diclofenac potassium, the power calculations indicated that a total of 220 patients (110 per group) was required to ensure a power of 85−90% with type I error rate (α) = 5%.
Since the objective of the study was to demonstrate non-inferiority, the primary analysis (time to onset) was performed on both the intent-to-treat (ITT) and per-protocol (PP) population according to statistical guidelines.[18] All other analyses were performed on the ITT population only. Statistical programming was performed in SAS, software version 8.2.
The null hypothesis was tested based on inversion of the log-rank test according to the following procedure: (1) non-inferiority of lornoxicam was defined as the time to onset of PAR being at most 20% slower than that of diclofenac potassium; (2) a standard survival approach was used with the treatment described as a covariate and centre as stratum; (3) assumed proportional time to onset for the treatments [let Slornoxicam (k*t) = Sdiclofenac potassium(t), where k is a proportionality constant, and Slornoxicam(t) and Sdiclofenac potassium(t) are the survival functions for time to onset for the two treatments]; (4) a 95% CI for the proportionality constant was found as the set of k values where the log-rank significance test is not significant; (5) lornoxicam was declared non-inferior if the CI for k was >0.80.
Superiority testing was performed using a standard log-rank test if lornoxicam was first proved non-inferior to diclofenac potassium.
The primary analysis was supported by an analysis using an accelerated failure time model of time to onset of PAR. The choice among candidate distributions (Weibull, Gamma, Log-logistic or Log-normal) was determined based on the best fit as assessed by the log-likelihood. The model included treatment and centre as terms in the model. A 95% CI for the parameter describing the lornoxicam time to onset of PAR relative to diclofenac potassium time to onset was computed based on the estimated parameter and standard error for treatment. For non-inferiority, the lower limit of this CI had to be >0.8.
If a patient received rescue analgesic within the first 90 minutes, the observation was censored at the time of intervention. If a patient did not experience onset of PAR within the first 90 minutes, time to onset was right-censored at 90 minutes.
The SPID endpoints were calculated using the trapezoidal rule, with pain intensity before first administration of treatment subtracted at all subsequent time-points, and with reversal of the scale so that high values indicated a positive development. Furthermore, SPID was divided by the time interval and thus expressed a mean PID value. The linear model used for SPID endpoints included treatment and centre as factors and pain intensity at baseline as a covariate. Time-profile plots of PID compared treatments over the entire trial period. The same methodology was used in relation to PAR. Ability to perform daily activities and global evaluation of the study medication were compared using the Wilcoxon rank sum tests.
Adverse events were tabulated per treatment group, system organ class, preferred term, severity and relationship to study medication.
Clin Drug Invest. 2006;26(5):267-277. © 2006 Adis Data Information BV
Cite this: Analgesic Efficacy and Safety of Lornoxicam Quick-Release Formulation Compared With Diclofenac Potassium - Medscape - Jun 01, 2006.
Comments