Non-steroidal anti-inflammatory drugs (NSAIDs) are among the most widely used painkillers in the world, either as prescription or over-the-counter (OTC) drugs. Cardiovascular risks of NSAIDs remain a major safety concern after the NSAID rofecoxib was withdrawn from the market due to serious adverse cardiovascular events, including stroke and myocardial infarction (MI), also known as a heart attack. MI occurs when blood flow to a portion of the heart muscle is decreased or completely cut off, injuring the heart muscle because of lack of oxygen. This is generally the result of blockage of the coronary arteries by plaques, cholesterol, fat deposits and blood cells. Symptoms of MI include chest pain or discomfort that can radiate to the neck, jaw, shoulder, arm and back, shortness of breath, abnormal heart beating, fatigue, nausea, sweating and heart burn. Epidemiological studies investigating whether the use of traditional NSAIDs (e.g., diclofenac, ibuprofen and naproxen) increases the risk of MI have so far given inconclusive results. Therefore, in this study we will evaluate whether the use of NSAIDs increases the risk of MI in patients with no major risk factors. To further investigate the effect of NSAIDs on MI, we will categorize results according to the number of prescriptions and type of NSAID prescribed.
The risk of myocardial infarction (MI) associated with NSAIDs was first observed during the long-term clinical trials of rofecoxib (Vioxx), a selective COX-2 inhibitor. Risk of cardiovascular disease in association with NSAID use has since remained a major safety concern. Several epidemiological studies have been conducted on selective and nonselective NSAIDs, and an imbalance of COX-1/COX-2 selectivity and predominant COX-2 inhibition have been proposed as an explanation for the increased risks of MI found among NSAID users. It is generally acknowledged that certain COX-2 inhibitors can increase the risk of MI, but results for traditional NSAIDs remain inconclusive.
Our primary aim is to evaluate the association between NSAID use and risk of MI among patients without major risk factors in the CPRD GOLD. We plan to conduct a nested case-control analysis in a population of NSAID users to reduce protopathic and indication bias. Firstly, we will select all patients aged 40-79 years old who had received at least one NSAID prescription. From this base population, we will select all cases without major risk factors who had a first MI between 2006 and 2019. Cases will be matched to up to four controls based on age, sex, general practice, and calendar year. We will use conditional logistic regression to estimate crude and adjusted odds ratios with 95% confidence intervals of MI among current and recent NSAID users, compared with past NSAID users. Odds ratios derived from this nested case-control study will estimate the risk ratio. As a secondary aim we will estimate crude and adjusted odds ratios of MI according to the number of NSAID prescriptions and types of NSAIDs prescribed among current and recent users of each individual NSAID, compared to past users of any NSAID.
The outcome of interest in this study will be incident idiopathic MI, which will be identified through Read codes. We will only select idiopathic MI cases, because by including non-idiopathic MI cases one could dilute the potential effect of NSAID use on MI. Idiopathic MI cases will be defined as cases that do not have a code for a proximate cause (e.g., major surgery) or risk factor of MI (e.g., ischemic heart disease, diabetes type 1, see L. Definition of Study Population). MI codes used to select cases will be MI diagnosis codes and MI ECG codes. If patients had a potential MI code within 30 days before the MI diagnosis code or MI ECG code, the date when the potential MI code first occurred will be set as the index date. MI possibly related to surgery or to mural, atrial or ventricular thrombosis will not be considered as an acute MI. Patients who had one of these codes within 30 days after the MI diagnosis or MI ECG code will not be considered as cases.
Susan Jick - Chief Investigator - BCDSP - Boston Collaborative Drug Surveillance Program
Susan Jick - Corresponding Applicant - BCDSP - Boston Collaborative Drug Surveillance Program
Brenda Baak - Collaborator - BCDSP - Boston Collaborative Drug Surveillance Program
HES Admitted Patient Care