A Retrospective Study of the Role of Thiazolidinediones to Reduce the Incidence of Atrial Arrhythmias

1.0 Background

Cardiac surgery is associated with many perioperative complications, including supraventricular arrhythmias.(1-4) Atrial fibrillation occurs in 15 to 40% of patients after coronary artery bypass grafting (CABG) and up to 60% after CABG and valve repair. It can lead to significant complications such as stroke, prolonged hospitalizations, increasing mortality and morbidity.(1;5)

Post-operative atrial arrhythmias seem to be related to the degree of inflammation and oxidative stress induced during surgery.(2-4) Our group and others have shown that atrial arrhythmias are associated with increased oxidative stress.(6-10) Moreover, increased oxidative stress associated with angiotensin II-mediated NADPH oxidase activation leads to electrical abnormalities likely to propagate arrhythmias.(11;12)

Thiazolidinediones (TZDs) are antidiabetic agents that enhance insulin sensitivity through activating peroxisome proliferators-activated receptor (PPAR) gamma. Beside their glucose lowering effects, TZDs are shown to exhibit anti-inflammatory and antioxidant properties.(13-17)

2.0 Objectives

Based on the discussion above, we hypothesize that pre-operative use of TZDs will decrease post-operative AF or Aflut incidence. This study will be a retrospective cohort analysis of all patients that underwent elective cardiac surgeries between 2000-2005 at Emory University Hospital, Crawford Long Hospital, or the Atlanta Veterans Affairs Medical Center to evaluate the relationship of pre-operative TZD use to the incidence of post-operative atrial arrhythmias.. This study is undertaken in preparation for a funding proposal for a prospective trial.

3.0 Patient Selection

3.1 Eligibility criteria:

1. All patients that patients who underwent elective cardiac surgery in the Emory University Hospital, Crawford Long Hospital, or the VA Medical Center between 2000-2005.

3.2 Ineligibility criteria:

1. Emergent operations.
2. The presence of AF or Aflut at the time of surgery.
3. Patients without adequate documentation of atrial rhythm following surgery.
4. New York Heart Association (NYHA) class IV heart failure at the time of surgery.
5. Hyperthyroidism
6. Implanted devices for designed for active management of atrial arrhythmias by pacing or defibrillation
7. Known illicit drug use
8. Known ethanol abuse
9. Electrophysiological ablation for atrial tachycardia within 6 months of the operation.

4.0 Registration and randomization: none

5.0 Therapy: none

6.0 Pathology: none

7.0 Patient assessment: none

8.0 Data collection: Data will be collected from review of the patient's hospital charts, from telemetry recordings, and ECGs to document the presence or absence of AF or Aflut during the hospital admission for cardiac surgery. The presence or absence of atrial fibrillation will be diagnosed on the basis of an electrocardiographic recording and confirmed by a cardiologist. Demographic data collected will include: age, race, sex, body mass index, blood pressure, NYHA classification, Killip classification, and the history of previous myocardial infarction, hypertension, diabetes, smoking, alcohol use, antiarrhythmic drug use, presence and type of pacemaker if any, history of AF or Aflut, previous cardioversions, type of operation, and length of operation. Patients enrolled in this study will be given unique study numbers. No follow up data will be required from patients.

9.0 Statistical considerations: This study design is a retrospective chart review of a cohort of patients undergoing cardiac surgery to determine if there is an association between pre-operative TZD use and the incidence of post-operative atrial fibrillation or Aflut. The null hypothesis is that there is no association between the preoperative use of TZDs and postoperative AF and Aflut. Within the cohort, those patients with AF or Aflut will be compared to those without these atrial arrhythmias in the post-operative period. Baseline data for the two groups will be expressed as mean ± standard deviation for continuous variables, and frequencies for categorical variables. Differences in baseline characteristics between the groups will be examined by use of Fisher exact and Mann-Whitney tests for categorical and continuous variables, respectively. Quantification of any association between pre-operative TZD and the presence of atrial fibrillation will modeled by logistical regression analysis after adjusting for other variables that may affect arrhythmia frequency.