Abstract P256: Myocardial Infarction is not a Risk Factor for Hospital-Acquired Venous Thrombosis: The Medical Inpatients Thrombosis (MITH) Study
Introduction: Venous thrombosis (VT) complicating medical admission is common and the risk factors are poorly understood. Myocardial infarction (MI) is often considered a risk factor for hospital-acquired VT with very little data other than expert opinion to support this assertion. We hypothesized that due to use of antithrombotic medications and early mobilization that MI may not be a hospital-acquired VT risk factor.
Methods: All cases of VT complicating medical admission were found between 01-2002 and 06-2009 at a 500 bed teaching hospital using ICD-9 discharge codes and confirmed by medical record review. Controls were frequency matched 2:1 to cases by admission year and medical service and weighted by the inverse sampling probability. Acute MI was defined as acute MI on the discharge summary, and prior MI as mention of prior MI on the admission note. We used multivariable weighted logistic regression and the Taylor series to calculate odds ratios and 95% confidence intervals (CI) for VT for acute MI and prior MI. Models included age and sex, and a VT risk score previously developed.
Results: Two-hundred ninety-nine cases of VT complicated 64,334 admissions (4.6 per 1000 admissions), 16 VT occurred in the 7,240 patients admitted with acute MI (2.2 per 1000 admissions). The table presents the percent of cases and hospital admissions with prior MI and acute MI, and the age- and sex-adjusted OR and multivariable-adjusted OR for VT. Prior MI was associated with a reduced odds of VT in both models (Table) while acute MI was not associated with VT accounting for VT risk factors (OR 0.83 95% CI 0.34, 2.01).
Discussion: In contrast to ‘conventional wisdom’, prior MI and acute MI were not risk factors for hospital-acquired VT. For acute MI, short hospital stays, use of antithrombotic medications, and early mobilization after the MI may affect hospital-acquired VT risk. For prior MI, statin use and lifestyle changes may reduce hospital-acquired VT risk, however further work is needed to confirm this finding.
|Cases (%) (n = 299)||Hospital Admissions (%) (n = 64,334)||Odds Ratio for VT|
|Age-,Sex- Adjusted||Multivariable Adjusted*|
|Past History of MI||5.8%||25.9%||0.44 (0.27, 0.73)||0.35 (0.18, 0.67)|
|MI on Admission||5.3%||11.3%||0.49 (0.23, 1.02)||0.83 (0.34, 2.01)|
↵* Adjusted for VTE prophylaxis use (mechanical and pharmacologic), full anticoagulation use, heart rate >100 on admission, hypoxic or ventilated on admission, history of VT, fracture in past 3 months, cancer in past 12 months, past history of congestive heart failure, and past history of a rheumatologic or inflammatory disease.
- © 2012 by American Heart Association, Inc.