| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;72:1795
© 2001 The Society of Thoracic Surgeons
a Departments of Cardio-Thoracic Surgery and Public Health Erasmus Medical Center Rotterdam, PO Box 2040, 3000CA Rotterdam, The Netherlands
e-mail: takkenberg{at}thch.azr.nl
To the Editor
Choosing a prosthetic valve in patients who require aortic valve replacement (AVR) is complex, and dependent on patient characteristics, physician and patient preferences, and prosthetic valve characteristics. With great interest, we read the article by Dr Birkmeyer and colleagues regarding the use of a Markov state-transition simulation model to estimate the preferred valve substitute for patients undergoing AVR [1].
A simulation model creates a large closed cohort of "virtual" patients that is followed over time until all patients have died. It allows detailed insight into the events that cause morbidity and mortality after AVR, unlike standard statistical techniques of survival analysis. Perioperative mortality, prosthetic valve endocarditis, and thrombo-embolism are not considered in Dr Birkmeyer’s model. Although the occurrence of these events did not differ by valve type, including them in the model would allow insight into the considerable impact they may have on patient prognosis and well-being. In this respect, it would also be interesting to investigate quality of life after a valve-related event using the model.
We have developed a microsimulation model to estimate prognosis after AVR with different types of valve substitutes [2–4]. Unlike Markov modeling, microsimulation has memory, ie, it allows for risks to change over time depending on events that happened in the past. For example, operative mortality not only increases with age, but also with each reoperation. In addition to the variables in Dr Birkmeyer’s model, the Rotterdam microsimulation model contains thrombo-embolism, endocarditis, and age-related operative mortality increasing with each subsequent reoperation. We do not agree with Dr Birkmeyer that excess mortality is caused largely by coexisting coronary heart disease that has an additive effect on population-based mortality rates. Previous studies found that excess mortality is multiplicative rather than additive, and decreases from a hazard ratio of 6 at age 35 to a hazard ratio of 1 at age 75 [5, 6]. Excess mortality is mainly related to the heart valve disease and associated cardiomyopathy. A 75-year-old patient after AVR has a similar life expectancy compared with a healthy 75-year-old, largely because of patient selection before operation.
We validated the results from Dr Birkmeyer’s analyses by entering the Markov model input variables into our microsimulation model. We were able to replicate the estimates of life expectancy for patients at different ages with a mechanical versus a tissue valve as published by Dr Birkmeyer. Next, we changed the additive excess hazard into our multiplicative hazard ratio for excess mortality. This resulted in a considerable decrease in life expectancy of 3 years (16%) in patients aged 50 at the time of operation, and a small increase in life expectancy of 0.3 years (6%) in patients aged 80 at the time of AVR. Concomitantly, the actual incidence of hemorrhage causing death decreased by almost 50% in the youngest age group. After adding input variables for operative mortality (increasing with age and with each reoperation), thrombo-embolism, and endocarditis to the model, life expectancy decreased in all age groups by 8% to 10%. As a result, life expectancy in patients aged 60 at the time of AVR was better with a bioprosthesis compared with a mechanical prosthesis while event-free life expectancy was similar. This confirms the preference for a bioprosthesis in 70-year-old patients, but also results in a lower age limit for considering a bioprosthesis.
In conclusion, using either Markov models or microsimulation, similar age- and gender-specific recommendations can be made regarding prosthetic valve choice for AVR. These models are valuable tools, although underlying assumptions, the quality of the input, and a continuous renewal of this input are of the utmost importance before reliable predictions regarding prognosis can be made.
References
Related Article
This article has been cited by other articles:
|
|
 |
|
  M. L. Brown, H. V. Schaff, B. D. Lahr, C. J. Mullany, T. M. Sundt, J. A. Dearani, C. G. McGregor, and T. A. Orszulak Aortic valve replacement in patients aged 50 to 70 years: Improved outcome with mechanical versus biologic prostheses. J. Thorac. Cardiovasc. Surg., April 1, 2008; 135(4): 878 - 884. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
