RESULTS
The databases yielded 6,417 studies (3,213 in PubMed, 507 in CAB direct, and 2,697 in Web of Science), and one study was included from other sources. After removing duplicates through a reference manager software (Zotero®), the abstracts of 4,524 titles were screened independently by two authors. Disagreements were among all authors until resolution. We excluded 4,365 articles because they did not investigate corticosteroids applied through the IA route, or not alone, or investigated diseases other than OA, or were not published in English. We subjected 159 articles to full-text analysis. After applying the general inclusion criteria, 95 articles were selected for screening according to the criteria of each study question.
Twenty-three articles fit all inclusion criteria for question 1, and 21 were included for question 2 (Figure 1). Summaries of included studies for each question are available in supplementary material 1 and 2. Okike et al. (2021) included two investigations; however, only the case-control study was included in the results, as the retrospective cohort investigated the effects of corticosteroids plus local anesthetics. Some of the papers also investigated other therapies (e.g., hyaluronic acid [HA] or platelet-rich plasma [PRP]); however, comparison of these to corticosteroids is beyond the scope of this review, so only the corticosteroid and negative control groups of these studies were considered.