Dose-effect relationships of nonsteroidal anti-inflammatory drugs: A literature review

Background: Many clinicians believe that higher doses of nonsteroidal anti-inflammatory drugs (NSAIDs) are more effective than lower doses for the treatment of rheumatoid arthritis (RA) and osteoarthritis (OA) but are associated with higher rates of adverse events (AEs). However, there is a lack of consensus on dose-effect relationships with the NSAIDs. Objective: The purpose of this review was to investigate evidence for the relationship between NSAID dose, efficacy, and the occurrence of AEs from clinical trials of RA and CIA of the hip and knee. Methods: Relevant English-language publications were identified through a search of EMBASE, MEDLINE, and REFLINE using the terms aceclofenac, diclofenac, etodolac, ibuprofen, isoxicam, lornoxicam, meloxicam, nabumetone, naproxen, piroxicam, tenidap, tenoxicam, arthritis, OA (hip and knee), RA, rheumatic disorders, and musculoskeletal disorders for the period January 1970 to December 1997 (this review was conducted in 1998). Bibliographies of retrieved publications were reviewed for other potentially relevant articles. Selected publications were evaluated for quality (likelihood of bias) based on 4 factors (randomization procedure; completeness of patient and treatment information; standardization and completeness of outcome data; and reporting of attrition data). Results: This review included 99 publications concerning clinical trials conducted in 24 countries and enrolling 28,239 patients. The majority of reports were published in the 1990s, particularly in the latter half of that decade. The average quality of the publications improved over time, with a significant increase in mean quality score from 5.43 in the 1970s to 9.21 during the last half of the 1990s (P < 0.05). Only 8 reports directly compared high and low doses of the same drug in relation to efficacy. Conclusions: Data on the relationship of NSAID dose to efficacy and the incidence of AEs were limited. There is a need for clinical trials directly addressing dose-effect relationships of NSAIDs, as well as reviews of more current literature and reports in languages other than English.