| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1997;63:300-301
© 1997 The Society of Thoracic Surgeons
Department of Cardiac Surgery, University Hospital Antwerp, Wilrijkstraat 10, B-2650 Edegem-Antwerp, Belgium
To the Editor:
Due to the aging population in western countries, the number of aortic valvular replacements-combined with coronary bypass grafting, if necessary-is increasing rapidly. The procedure is being performed more often even in octogenarians and older patients. To analyze costs versus effectiveness of the operation, survival, and quality of life, the limited expected additional survival in the very old must be taken into account [1].
It was with great interest that we read the article by He and associates [1] on the survival of 877 patients more than 70 years of age who, over a period of 30 years, underwent aortic valvular replacement, combined in 41.6%, with coronary bypass grafting. We believe that the survival of these old patients should not be reported as raw data. Ages ranged from 70 to 94 years, which implies that expected survival would vary greatly. In addition, someone operated on at age 80 years in, say, 1964 would, probably due to a secular trend, have a shorter life expectancy than someone of that age in 1984.
We have presented our findings in 643 Dutch patients who had undergone aortic valvular replacement between 1966 and 1986. Twenty-four percent also had coronary artery disease. One hundred thirty-eight were 70 years old or older. Mean follow-up was 73 months, for a total of 3,603 patient-years [2].
We compared raw survival data with those provided by the Dutch Central Bureau of Statistics. Every patient of ours was matched for age, date of birth, sex, and calendar year of operation with a group of individuals from the general population. The methods were extensively explained in another article [3].
We concluded that after perioperative mortality older patients survived almost as long as matched individuals. Younger patients, however, did less well than the younger general cohort. We can only speculate on the causes of this discrepancy. In contrast to the findings of He and associates, we found no difference in corrected survival between men and women.
In light of our experience, we believe that any analysis of survival patterns of older patients must take the background mortality, ie, the non–disease-related mortality, into account.
References
Division of Cardiothoracic Surgery, Department of Surgery, University of Hong Kong Grantham Hospital, Aberdeen, Hong Kong
Providence St. Vincent Hospital and Medical Center, 9155 Sw Barnes Rd Portland, Or 97225
We thank Dr Moulijn and associates for their interesting comments on our article. They suggest that we should use "corrected" (ie, relative) rather than "raw" (ie, observed) survival estimates. Although it may be indicated under certain circumstances, relative survival is not the only method that can be used in medical research and data analyses.
Relative survival is the observed survival divided by the expected survival of a series from the general population matched for age, sex, and time on risk. Relative survival can be useful for comparing groups with different mean ages. For example, an older group may have lower survival than a younger group, but that difference may be more or less than would be expected due to the different age alone.
However, the use of relative survival may be misleading, because these surgical patients are not a random sample from the general population. We have found that the observed survival for patients undergoing coronary operations was better than that expected from the general population, especially in the older age groups [1]. Does that mean that operation eliminates the cardiac risk, and even bestows additional life-protection?
The answer is "No," because these patients are selected not to have other life-threatening illnesses. They are sufficiently poised sociologically to gain access to specialty medical care. And perhaps, because of their high-level medical care, they are more concerned than average with modifying their coronary artery risk factors and in pursuing a healthy lifestyle.
For the above-mentioned reasons, we usually use observed survival rather than relative survival in our studies [1–3].
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
