Nonsteroidal anti-inflammatory drugs in chronic and relapsing musculoskeletal pain: pearls for clinical practice

Abstract Nonsteroidal anti-inflammatory drugs (NSAIDs) are the principal pharmacological option for inflammatory musculoskeletal pain and are frequently treated as an interchangeable class from which any agent may be selected. This narrative review examines the pharmacological basis of that assumption and organises the evidence around three questions: which molecule, at what dose, and in which formulation. Four observations follow from the available data. COX selectivity is an incomplete predictor of clinical risk, since a non-selective agent may carry a cardiovascular profile comparable to that of the coxibs while a coxib may reduce events across the whole gastrointestinal tract; risk is therefore better assigned to individual molecules at defined doses than to selectivity classes. Comparative efficacy is dose-dependent rather than uniformly equivalent, so equivalence between agents applies only at comparable analgesic doses, and apparent non-response is frequently sub-therapeutic dosing. Organ-specific risk differs in kind as well as in degree across the cardiovascular, gastrointestinal, renal, hepatic and haemostatic axes, and proton pump inhibitors protect the upper gastrointestinal tract only. Modified-release formulations alter onset and peak-to-trough fluctuation, but no evidence indicates that they attenuate class-level organ toxicity. Contraindications lying outside the cardiorenal and gastrointestinal axes are stated, together with the clinically relevant drug interactions and the regulatory restrictions applying to individual agents.