No added value of duloxetine for patients with chronic pain due to hip or knee osteoarthritis

Purpose: The currently prescribed analgesics for OA related pain do not always sufficiently reduce pain and can be associated with side-effects or can be contra-indicated. Other treatment options are therefore needed. An option may be duloxetine, a serotonin and norepinephrine reuptake inhibitor (SNRI). Duloxetine is hypothesized to reduce chronic pain by central inhibition of pain. Several placebo-controlled studies showed small to moderate effect of duloxetine. However, the effectiveness in a primary care setting is unknown. Neither is known whether the presence of symptoms of centrally sensitized pain alters the response to duloxetine. Therefore, our aim was to assess the effectiveness of duloxetine added to usual care provided by the general practitioner (GP) for patients with chronic OA pain. Secondary objectives were to assess cost-effectiveness and to assess whether the presence of symptoms of centralized pain alters the response to duloxetine. Methods: A multicentre, open label, cluster randomised trial was conducted in primary care. Patients were eligible when ≥18 years, had hip and/or knee OA based on the clinical ACR criteria, had chronic pain on most days in the past 3 months and insufficient benefit to of NSAIDs. Patients were excluded when knee or hip replacement was scheduled, when using antidepressants or neuropathic pain medication, if they had rheumatoid arthritis, or contra-indications for duloxetine, or were unable to sign informed consent. All eligible patients were identified before randomisation of the GP practice was performed. Patients received duloxetine 60mg/day added to usual care, or usual care alone. Usual care was provided according to the Dutch GP guidelines. The primary outcome was WOMAC pain (0-20) at 3 months. We aimed to detect a difference of 1.9 points (effect size 0.4) between the two groups and a difference of 2.9 points (effect size 0.6) for the subgroup of patients with symptoms of centralized pain (modified painDETECT >12). Secondary outcomes were WOMAC pain at 12 months, WOMAC function, quality of life, side-effects, cost-effectiveness, patients satisfaction, perceived improvement, and OMERACT-OARSI responder criteria. A linear mixed model with repeated measurements was used to assess the differences in continuous outcomes and generalised estimating equations analyses were performed for dichotomous outcomes. The analyses were adjusted for age, sex, modified painDETECT score, HADS depression score and the presence of two or more comorbidities. Results: In total, 133 patients were included and 132 were randomised. 66 patients (31 practices) were randomised to duloxetine added to usual care and 66 patients (34 practices) to usual care alone. Characteristics of the GP practices were similar in both groups. Some characteristics of the patients differed between the two groups. The duloxetine group consisted of fewer women (59.1% vs 75.8%), patients were younger (63.2 years vs 65.4 years) and had fewer comorbidities (15.2% vs 33.2% had ≥ 2 comorbidities). Most patients included had knee OA (77.3% in duloxetine group and 86.4% in the usual care group) and 40% of the patients had symptoms of centralized pain. No differences were found for WOMAC pain at 3 months (adjusted difference -0.58 95% confidence interval (CI) [-1.80; 0.63]) or at 12 months (adjusted difference -0.26 95% CI[-1.86; 1.34]). The secondary outcomes quality of life, patient satisfaction and the OMERACT-OARSI responder criteria also showed small, non-significant differences (Table 1). For the subgroup of patients with symptoms of centralized pain no effect of duloxetine was found either (-0.32 95% CI[-2.32; 1.67]. Furthermore, the intervention was not cost-effective; a small positive difference of 0.04 QALY’s was found for the duloxetine group. The costs per gained QALY were €13.000,- from health care perspective. The uncertainty analysis showed 80% probability that these costs were lower than €50.000,- per QALY gained, which is the threshold for moderate disease severity. From societal perspective, these costs were €54.000,- with a 48% probability of costs being lower than €50.000,-. In the duloxetine group, 56 patients (85%) started using duloxetine. After three months 61% of the patients and at 1 year 35% of the patients were still using duloxetine. At 3 months 89.3% of the patients in the duloxetine group reported at least one side-effect compared to 72.5% in the usual care group. Nausea, weight loss, constipation, yawning and hyperhidrosis were reported significantly more frequently by patients in the duloxetine group. Patients in the duloxetine group contacted their GP more frequently (51.8% vs 30.8% at 3 months) and were more often referred to an orthopaedic surgeon (10.7% vs 3.8% at 3 months). In the total follow-up time, 5 patients in the duloxetine group had a total hip replacement (THR) or total knee replacement (TKR) while none of the patients receiving usual care had a THR or TKR. Conclusions: We can rule out a clinically relevant effect (1.9 points on WOMAC pain scale) of duloxetine added to usual care compared to usual care alone in patients with chronic OA pain. For patients with symptoms of centralized pain we can rule out a larger effect of duloxetine (2.9 points), but the presence of a smaller difference (1.9 points) cannot be ruled out and our Results in this subgroup need to be confirmed in another trial.