Statin-associated polymyalgia rheumatica. An analysis using WHO global individual case safety database: a case/non-case approach

• Published in: PloS one 2012-07, Vol.7 (7), p.e41289
• Main Authors: de Jong, Hilda J I, Saldi, Siti R F, Klungel, Olaf H, Vandebriel, Rob J, Souverein, Patrick C, Meyboom, Ronald H B, Passier, J L M Anneke, van Loveren, Henk, Cohen Tervaert, Jan Willem
• Format: Article
• Language: English
• Subjects: Arthritis; Case reports; Case studies; Confidence intervals; Data bases; Data mining; Databases, Pharmaceutical; Dictionaries; Drug dosages; Female; Gender; Humans; Hydroxymethylglutaryl-CoA Reductase Inhibitors - adverse effects; Immune system; Laboratories; Male; Medicine; Middle Aged; Multiple sclerosis; Pharmaceutical industry; Pharmaceutical sciences; Pharmacology; Polymyalgia rheumatica; Polymyalgia Rheumatica - chemically induced; Public health; Regression analysis; Reporting; Rheumatology; Safety; Statins; Statistical analysis; Vitamin D; Vitamin deficiency; World Health Organization
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0041289

To assess whether there is an association between statin use and the occurrence of polymyalgia rheumatic (PMR) in the spontaneous reporting database of the World Health Organisation (WHO). We conducted a case/non-case study based on individual case safety reports (ICSR) in the WHO global ICSR database (VigiBase). Case reports containing the adverse event term polymyalgia rheumatica (WHOART or MedDRA Preferred Term) were defined as cases. Non-cases were all case reports containing other adverse event terms. Each case was matched to five non-cases by age, gender, and time of reporting. Case reports regarding a statin as suspected or concomitant drug were identified using the Anatomical Therapeutic Chemical (ATC) classification. Multivariate logistic regression was used to calculate reporting odds ratios (RORs) with 95% confidence intervals (CI). We identified 327 reports of PMR as cases and 1635 reports of other ADRs as non-cases. Among cases, statins were more frequently reported as suspected agent (29.4%) compared to non-cases (2.9%). After adjustment for several covariates, statins were significantly associated with reports of PMR (ROR 14.21; 95% CI 9.89-20.85). The results of this study lends support to previous anecdotal case reports in the literature suggesting that the use of a statin may be associated with the occurrence of PMR. Further studies are needed to study the strength of the association in more detail and to elucidate the underlying mechanism.