Related Experiment Videos
Valve data collection: problems and pitfalls
P J Drury1, M M Black, C J Ashman
1Institute for Biomedical Equipment Evaluation and Services, Lodge Moor Hospital, Sheffield, UK.
Journal of Medical Engineering & Technology
|January 1, 1992
Summary
This study analyzes over 16,000 heart valve implants using proportional hazards analysis to compare valve performance and patient outcomes. It reveals comparative event-free survival rates for major heart valve types.
Area of Science:
- Cardiovascular Surgery
- Biomedical Engineering
- Medical Statistics
Background:
- The University of Sheffield manages the UK's largest multicentre heart valve implant follow-up study.
- The database contains data on over 16,000 valve implants from 57 surgeons across 22 UK centers.
- Over 30 valve models, categorized into ball, disc, porcine, pericardial, and homograft types, are included.
Purpose of the Study:
- To analyze the performance of different heart valve types.
- To compare patient survival and event-free survival rates.
- To address limitations of traditional survival analysis by incorporating prognostic factors.
Main Methods:
- Descriptive statistical analysis of heart valve implant data.
- Actuarial methods for generating survival and event-free survival graphs.
- Proportional hazards analysis to account for patient prognostic factor variations.
Main Results:
- Valuable insights into the performance of various heart valve models.
- Identification of comparative event-free survival rates among major valve types.
- Demonstration of proportional hazards analysis for detailed patient group comparisons.
Conclusions:
- Proportional hazards analysis offers a more nuanced understanding of heart valve performance.
- This method allows for in-depth study of valve types concerning thromboembolic complications and dysfunction.
- Comparative patient outcomes can be better assessed by accounting for prognostic factors.