LETTER TO THE EDITOR
Reply
Fause Attie, MDa
a National Institute of Cardiology of Mexico,Juan Bodiano #1,Seccion XVI,Tlapan,Mexico CityMexico 14080
We thank Dr. Bassan very much for his interest about our recent article concerning surgery for atrial septal defect (ASD) in adults (1). Our explanation follows.
A compound index is a common strategy recently used for developing some clinical trials in which the sample size for a specific "end point" is either not possible or very difficult to obtain (2). When the sample size is calculated for a compound index, and it is defined as the primary "end point," the analysis at the conclusion of the study should be directed to this "end point" as a "package" (2,3). Of course, the curiosity to see what happened in the compound index is unavoidable, but caution needs to be considered in the interpretation of the results. The way that Dr. Bassan analyzed the data, we can see that sudden death was almost significant, but the small number of cases probably did not permit this conclusion. The high incidence of pneumonia was certainly unexpected. We do not have a clear explanation for this feature; however, pulmonary infection was responsible for 28.7% of the total mortality. Recurrent pneumonia is a difficult diagnosis to be easily differentiated from pulmonary infarct in these types of patients. We believe that the pulmonary infarct was probably underestimated. Our study is the first randomized clinical trial published about this specific topic, and no complex statistical manipulation was done (4,5), because the data are the data!
 |
References
|
|---|
- Attie F, Rosas M, Granados N, Zabal C, Buendía A, Calderón J. Surgical treatment for secundum atrial septal defect in patients >40 years old: a randomized clinical trial. J Am Coll Cardiol. 2001;38:2035–2042[Abstract/Free Full Text]
- Braunwald E, Cannon CP, McCabe CH. An approach to evaluating thrombolytic therapy in acute myocardial infarction: The ‘unsatisfactory outcome’ end point. Circulation. 1992;86:683–687[Free Full Text]
- Feinstein AR. Clinical EpidemiologyThe Architecture of Clinical Research. Philadelphia, PA: WB Saunders; 1985.
- Walter SD, Feinstein AR, Wells CK. A comparison of multivariable mathematical methods for predicting survival: II. Statistical selection of prognostic variable. J Clin Epidemiol. 1990;4:349–359[CrossRef]
- Altman DG, Stavola BL. Practical problems in fitting a proportional harzards model to data with updated measures of the covariates. Stat Med. 1994;13:301–341[Medline]
Top
References