For patients with osteoarthritis (OA) of the knee, pain is the most debilitating symptom. Although it has been proposed that the chronic phase of the monoiodoacetate (MIA)-induced rodent model of knee joint pain may be superior to other chronic or acute OA models for assessing the analgesic efficacy of novel molecules, relatively few pharmacological studies have been conducted in the chronic phase of this model. Hence this study was designed to use pharmacological methods to characterize the chronic phase of the MIA-induced rat model of knee joint OA pain. Rats received a single intraarticular injection of MIA at 2.5 mg or vehicle (saline) into the left (ipsilateral) knee joint. Pain behaviour was assessed by measuring paw withdrawal thresholds (PWTs) in the hindpaws pre-MIA injection twice-weekly until study completion on day 42. Mechanical allodynia was fully developed in the ipsilateral hindpaws (PWTs ≤ 6g) from day 7 and it persisted until day 42. MIA-injected rats with PWTs ≤ 6g in the ipsilateral hindpaws received single doses of one of four clinically-available drugs that represent four distinct pharmacological classes, viz gabapentin, amitriptyline, meloxicam and morphine, according to a 'washout' protocol with at least 48 h between successive doses. Gabapentin evoked dose-dependent anti-allodynia as did morphine whereas amitriptyline and meloxicam were inactive. Our findings are aligned with clinical data showing that gabapentin and morphine alleviated OA pain in the knee. The lack of efficacy of amitriptyline is consistent with the loss of descending diffuse noxious inhibitory controls reported by others in this model.