TY - JOUR T1 - Corticosteroid injections: glass half-full, half-empty or full then empty? JF - British Journal of Sports Medicine JO - Br J Sports Med SP - 564 LP - 565 DO - 10.1136/bjsports-2019-101250 VL - 54 IS - 10 AU - John W Orchard Y1 - 2020/05/01 UR - http://bjsm.bmj.com/content/54/10/564.abstract N2 - Cortisone is the most common class of injection in musculoskeletal medicine. Expert guidelines, such as the 2017 edition of Therapeutic Guidelines: Rheumatology (Australia),1 generally recommend corticosteroid injections (CSI) as a treatment for most tendon and joint pain conditions (including those listed in table 1). I quote, ‘local corticosteroid injections should only be given by, or under the supervision of, clinicians with appropriate training and experience’.1 I find it remarkable that the guidelines do not spell out that cortisone injections should only be given where high-quality evidence shows that benefits outweigh the harms; that is, the widespread recommendations to use CSI for multiple joints and multiple tendons are ‘eminence-based’ (who should give them) rather than ‘evidence-based’ (when they should be given).View this table:In this windowIn a new windowTable 1 Comparison of short-term (mainly benefit) versus long-term effects (possible harm) of corticosteroid injection for common indicationsTo be fair, I see that ‘appropriate training’ could cover ‘knowing the indications’ for a treatment but at a time where medical overtreatment is incontrovertible—it exists and no one denies it exists—I argue that the case for ‘indication-based treatment’ should be explicit.Systematic reviews have examined whether cortisone injections are beneficial or harmful. A typical systematic review concludes as per the text in box 1. This conclusion was from a 2002 systematic review on the use of CSI for tennis … ER -