J Thorac Cardiovasc Surg 2003;126:2118-2119
© 2003 The American Association for Thoracic Surgery
Reply to the Editor
Tirone E. David, MD,
Joan Ivanov, PhD,
Susan Armstrong, MSc,
Harry Rakowski, MD
Division of Cardiovascular Surgery, Toronto General Hospital and University of Toronto, Toronto, Ontario, Canada
Dr Shuhaiber makes some valid points about our study,1 but many of his comments are irrelevant. He ignored the fact that ours was an observational study and did not establish causation but rather an association between functional class and late mortality after mitral valve repair for mitral regurgitation caused by prolapse. The objectives of the study were clearly defined, and in addition to showing the late outcomes of mitral valve repair for floppy valves it also showed associations between certain preoperative variables and late mortality by multivariable analysis with bootstrap validation. Our study was not a prospective randomized trial, and consequently we could not control the variables before the operation as he suggested. This is, of course, a major limitation of all retrospective studies. One can only adjust for those factors that were measured. Residual confounding by unmeasured variables may potentially distort results. However, numerous studies have shown that careful, multivariable analysis of observational data produces treatment effects very similar to those of randomized controlled trials.2-4
Atrial fibrillation and diabetes were included in all multivariable models. Dr Shuhaiber's concerns regarding chronic atrial fibrillation may in fact provide further argument to operate on symptom-free patients. Atrial fibrillation may be a consequence of deterioration in ventricular function in patients with mitral regurgitation, and by the time it first occurs irreversible changes may already be present.
His remarks regarding postoperative functional class and the probability of recurrent mitral regurgitation in our patients are incorrect. One would not expect symptom-free patients to have lessened symptoms after the operation simply because they were already in functional classes 1 and 2 (ceiling effect). The analysis at follow-up was designed to demonstrate the improvement in the patients with symptoms. The mechanism of mitral valve prolapse was similar in both groups of patients, and recurrent mitral regurgitation after mitral valve repair is more dependent on the pathology of the mitral valve than on presence or absence of preoperative symptoms. That is reason why recurrent mitral regurgitation and reoperation rates were similar in the two groups.
We agree that functional class is a soft, subjective variable, especially between classes 2 and 3, but certainly not between classes 1 and 3 or 2 and 4. Moreover, symptoms are usually the reason a patient seeks medical advice. What our study showed is that postponing surgery until symptoms are more obvious (functional classes 3 and 4) is not appropriate if the mitral valve can be repaired. Nowhere in the text of our article will the reader find a sentence stating, "Patients with asymptomatic mitral valve incompetence are candidates for surgery," as written in Dr Shuhaiber's letter. In fact, in the last paragraph in the discussion of our article, the reader will find the following1: "In conclusion, surgical intervention should be considered in asymptomatic patients with severe MR caused by floppy valves if valve repair is feasible, and it can be done with low operative mortality and morbidity because the late survival is identical to that of the general population."
And the paragraph before the last reads as follows1: "This is a retrospective study of a clinical experience of a single surgeon, and the results might to be generalizable. The prevalence of associated cardiac and non-cardiac diseases was relatively small in this series, and statistical values of certain variables might have been altered by chance alone."
We believe our conclusion was far softer than implied in the letter. However, we agree that a controlled randomized trial is needed to determine the appropriateness of mitral valve repair for symptom-free patients with normal left ventricular function.
 |
References
|
|---|
- David TE, Ivanov J, Armstrong S, Rakowski H. Late outcomes of mitral valve repair for floppy valves: Implications for asymptomatic patients. J Thoracic Cardiovasc Surg. 2003;125:1143–1152[Abstract/Free Full Text]
- Concato J, Shah N, Horwitz RI. Randomized, controlled trials, observational studies, and the hierarchy of research design. N Engl J Med. 2000;342:1887–1892[Abstract/Free Full Text]
- Benson K, Hartz AJ. A comparison of observational studies and randomized controlled trials. N Engl J Med. 2000;342:1878–1886[Abstract/Free Full Text]
- Hlatky MA, Califf RM, Harrell FE Jr, et al. Comparison of predictions based on observational data with results of randomized controlled trials of coronary artery bypass surgery. J Am Coll Cardiol. 1988;11:237–245[Abstract]
Related Article
-
Reply to the Editor
- Ray C.-J. Chiu
J. Thorac. Cardiovasc. Surg. 2003 126: 2117.
[Extract]
[Full Text]
[PDF]
Top
References