Aims: Etiopathogenesis of Osteoarthritis (OA) and the role of Prostaglandins in OA is not yet clear. We conducted this study to elucidate the levels of Prostaglandin E2 (PG E2) in different stages of OA of the knee joint and correlate this with the treatment of OA. Methods: Five controls (with symptoms after knee injury few months ago but normal by radiography and arthroscopy) and fifteen cases (diagnosed to have OA by radiography and confirmed and graded arthroscopically) had arthroscopic lavage with debridement as necessary. NSAIDs were stopped at least 2 weeks before sampling. Articular cartilage specimens from margins or from osteophytes and synovial tissue specimens adjacent to degenerated cartilage were obtained during arthroscopy. Plasma was obtained from ipsilateral femoral vein just before arthroscopy. The samples were stored in liquid nitrogen[−8° C] and extracted with organic solvent. PG E2 levels measured in duplicate by enzyme immunoassay. Wilcoxon rank sum test and Kauskal Willis one way analysis of variance test were used. Results: Higher PG E2 levels were found in articular cartilage samples from osteoarthritic knee as compared to controls. This was more evident in early and moderate OA than in late OA. Similar trend was seen in case of synovial tissue and plasma. However it reached statistical significance only in case of synovial tissue.[p=0.025]Conclusion: PG E2 levels are altered in case of OA [with statistically sigificant changes in the synovial tissue]. The changes are similar to the changes in activity of chondrocytes found by other workers. While this change can explain the symptomatic relief by NSAIDs in OA, it also raises the possibility of serious damage to the articular damage by NSAIDs as PG E2 maybe protective rather than destructive [Dingle et al 1991]