Patients with rheumatic diseases like rheumatoid arthritis (RA) or systemic lupus erythematosus (SLE) remain inherently at risk for cardiac diseases due to the effects of systemic inflammation combined with immunologic activation and antibody formation targeting specific organs. Examples include coronary artery disease, associated ischemic and non-ischemic heart failure, myocarditis, and interstitial lung disease. While treatment with disease modifying anti-rheumatic drugs (DMARDs) adequately controls primary disease activity (clinically, skin and joints; serologically with reduction of inflammatory/disease activity markers), the effects of DMARDs on inciting or ameliorating/worsening pre-existing cardiac diseases associated with rheumatic diseases, are hotly contested.

Hydroxychloroquine (HCQ) is the cornerstone treatment for disease control in SLE and also used in combination with other disease modifying anti-rheumatic drugs (DMARDs) in RA. Long-term use of HCQ has been associated with skin hyperpigmentation, short-term gastrointestinal toxicity, retinal toxicity, and most concerning, cardiac arrhythmic episodes and heart failure, which is irreversible and in many cases associated with mortality. Prior observational/cohort studies demonstrated conflicting results, with some group of studies suggesting increased risk of cardiotoxicity with HCQ use vs. others suggesting no increased risk. The combination of HCQ and azithromycin for treatment of COVID-19 and their association with arrhythmic events have also compounded this concern.

There has yet been a large-scale electronic health record (EHR) based observational study delving deeper into these associations. The Observational Health Data Sciences and Informatics (OHDSI) represents a rich data resource for extraction of this level of information given the multi-institutional and international constituents.

We plan to address:

  1. What proportion of rheumatologic patients prescribed HCQ are receiving ACR/ophthalmology guided dosing (<5 mg/kg) as well as retinal and cardiac screening?
  2. What is the “global” prevalence of HCQ associated cardiotoxicity (arrythmias, bradycardia, myocardial infarction, myocarditis, angina/chest pain, and heart failure)? What proportion of rheumatologic patients exposed to HCQ develop cardiotoxicity?
  3. Among rheumatologic patients taking HCQ long-term, who are likely to develop cardiotoxicities?

Aims 1a. Identify all patients with rheumatoid arthritis, systemic lupus erythematosus, and/or non-specific inflammatory arthritis (age>18) with an exposure to HCQ 1b. Identify all patients who had continued exposure to HCQ (at least 365 days before the onset of outcomes) then developed the pre-defined cardiotoxicity outcomes (arrythmias, bradycardia, myocardial infarction, myocarditis, angina/chest pain, and heart failure)

2a. Develop an automated machine-learning based algorithm, using structured data (co-morbid diagnosis codes, co-administered medications), that identifies novel risk factors for developing HCQ cardio-toxicity Hypothesis: Specific combinations of co-morbidities and/or medications 2b. Develop a machine-learning based algorithm, using unstructured data (e.g. clinical free-text notes), that identifies novel risk factors for developing HCQ cardio-toxicity 2c. Validate this algorithm on cohorts developed from Aim 1a/b

This is an UNPAID research project.

Faculty Advisor

  • Professor: Elizabeth, Park
  • Center/Lab: Rheumatology
  • Location: 630 W 168th St, P&S 3-450, NY, NY 10032
  • Dr Park’s research has focused on cardiovascular complications of rheumatic diseases, specifically predictors of heart failure in rheumatoid arthritis patients, and cardiotoxicity from disease modifying anti-rheumatic drugs (DMARDs) in autoimmune rheumatic disease patients.

Project Timeline

  • Earliest starting date: 11/1/2022
  • End date: 7/1/2023
  • Number of hours per week of research expected during Fall 2022: ~10

Candidate requirements

  • Skill sets: Competency with R/Python, machine learning methods, and NLP
  • Student eligibility: freshman, sophomore, junior, senior, master’s
  • International students on F1 or J1 visa: eligible
  • Academic Credit Possible: Yes
  • Additional comments: None