Risk of gout flare after medication: prescription symmetry sequence analysis

Abstract

Objectives: The research aimed to study the following questions: (1) five well-known gout-related medications were selected to test the validity of the prescription symmetry sequence analysis in Taiwan; (2) four exploratory medications were selected to test their relation to gout flares.

Methods: We utilized the 2003-2017 dataset of the Taiwan National Health Insurance Program containing all claims data with 2 million beneficiaries as a data source. In order to explore the temporal association, we designed a scenario of medication-induced gout flares. Nine medications were selected as the index agent, including aspirin (low-dose), thiazide diuretics, loop diuretics, ethambutol, pyrazinamide, metformin, pioglitazone, fenofibrate, and losartan. The gout flare was defined as subjects with use of the marker agent for treatment of gout flares. The observation-window period between initiation of the index agent and initiation of the marker agent was 1 year. Subjects who used an index agent and a marker agent on the same day were excluded. The prescription symmetry sequence analysis was carried out to compare the observed number of persons who took an index agent prior to starting a marker agent with the observed number of persons who took a marker agent before starting an index agent. The adjusted sequence ratio (adjusted SR) with 95% confidence interval was applied to estimate the relation between an index agent and the marker agent.

Results: Among five medications including aspirin (low-dose), thiazide diuretics, loop diuretics, ethambutol, and pyrazinamide, the adjusted sequence ratio ranged from 1.15 to 3.35 and all reached statistical significance. Fenofibrate use and losartan use were associated with a lower probability of gout flares, with reaching statistical significance (adjusted SR = 0.60 for fenofibrate and adjusted SR = 0.92 for losartan). Metformin use was associated with a greater probability of gout flares, with reaching statistical significance (adjusted SR = 1.14). Pioglitazone use did not reach statistical significance.

Conclusion: Based on the confirmatory analysis including five well-known gout-related medications, this study supports that the prescription symmetry sequence analysis can be used to detect an adverse drug event associated with one potential offending agent. The exposure to fenofibrate or losartan might be a protective factor against gout flares. Metformin use could be associated with a greater probability of gout flares, but this finding should be validated by other studies.

Key points: • What is already known about this subject? 1. The prescription symmetry sequence analysis is a useful method for detecting an adverse drug reaction associated with one potential offending drug. 2. Numerous medications are found to induce gout flares. • What does this study add? 1. The prescription symmetry sequence analysis supports the evidence that aspirin (low-dose), thiazide diuretics, loop diuretics, ethambutol and pyrazinamide are associated with a greater probability of gout flares. 2. The exposure to fenofibrate or losartan might be a protective factor against gout flares. 3. Metformin use could be associated with a greater probability of gout flares. • How might this impact on clinical practice or future developments? 1. Clinicians should always consider the possibility of medication-induced gout flares. If gout flares develop, discontinuation of risky medications is the first step. Then prescribing cascades can be eliminated.

Keywords: Aspirin; Gout flare; Loop diuretics; Prescription symmetry sequence analysis; Thiazide diuretics.