LETTER TO THE EDITOR
The T–786C endothelial nitric oxide synthase genotype and coronary artery disease: Reply
Gian Paolo Rossi, MD, FACC, FAHA,
Maurizio Cesari, MD and
Achille C. Pessina, MD, PhD
Department of Clinical and Experimental Medicine, Clinica Medica 4 University Hospital, via Giustiniani, 2, 35126 Padova, Italy
gianpaolo.rossi{at}unipd.it
We appreciate the interest of Dr. Batalla and colleagues in the recent study where we reported an association between multivessel coronary artery disease (CAD), which was angiographically assessed, and the -786C allele in the promoter region of the endothelial nitric oxide synthase (eNOS) gene in Caucasians (1). Consistent with our results, Batalla et al. claimed that the same allele was previously reported to be associated with early CAD (e.g., with an episode of CAD defined according to the guidelines of the World Health Organization MONICA project) in male smokers who were younger than 50 years (2). However, in our view, the conclusions drawn in that study were not supported by the results that were actually obtained, for several reasons.
First, although the reported C allele frequency (37% in controls and 45% in CAD patients) was much higher than in Japanese patients (3) and not far from that found in our report and in other studies on Caucasians (1,4), the genotyping at this polymorphism was carried out with the restriction fragment length polymorphism (RFLP) analysis technique, a methodology that is far from optimal. In a pilot study we compared this technique for T-786C eNOS genotyping with the "gold standard" sequencing and found inconsistent enzymatic cleavage that resulted in a substantial rate of misgenotyped patients. Therefore, we abandoned RFLP and replaced it with the more expensive melting curves analysis, which provided accurate results as described in detail (5).
Second, by performing appropriate sample-size calculations (nQuery version 5.0) on the results presented in the study that Batalla et al. quoted (2), it is quite evident that that study was not adequately powered to answer the questions that it posed (e.g., to verify the hypothesis of an association between eNOS polymorphism and CAD). Given the reported sample sizes of 170 and 300 in the CAD patients and control groups, respectively (2), a two-group  2 test with a 0.05 two-sided significance level has 54% power to detect the difference of -786C allele frequency between CAD patients ( 1 of 0.218) and controls (2 of 0.143) that was observed (2). Therefore, in our view, the positive results mentioned by Batalla et al. (2) were either luck or serendipitous findings.
Third, in the aforementioned study, the association between CAD and eNOS -786C allele emerged as significant by comparison, with a 2 analysis, of the CAD patients and the controls, who represented hospital staff, blood bank donors, and eligible residents. However, apart from the self-evident differences of sample size between groups, it is altogether obvious from the results shown that several differences were overlooked. Despite the fact that the investigators carefully selected young male smokers as controls, their groups differed markedly for total cholesterol (220 ± 51 mg/dl vs. 119 ± 42 mg/dl), HDL (high density lipoprotein) cholesterol (32 ± 8 mg/dl vs. 54 ± 13 mg/dl), triglycerides (183 ± 91 mg/dl vs. 120 ± 72 mg/dl), and proportion of diabetic (10% vs. 0%) and hypertensive patients (31% vs. 0%). All differences were highly significant (p < 0.0001) from a statistical standpoint. Therefore, compelling evidence existed for an unbalanced distribution between CAD patients and controls of at least five universally accepted cardiovascular risk factors. We also found a similar unbalanced distribution of these and additional risk factors in our study. Accordingly, we used a logistic regression analysis to account for their confounding effects on the association of -786C allele with CAD. In contrast, no such analysis was performed in the study that Batalla et al. quoted (2).
Thus, in addition to the several limitations that were correctly acknowledged by the investigators themselves (2) other serious issues concerning the methodology used for genotyping, the power of the study, and the way that results were analyzed precluded drawing any conclusion concerning the association of the -786C allele with CAD in the aforementioned study (2).
We carefully avoided any statements of primacy in our report (1); however, it is quite evident at this time that our study was the first to support with sound scientific data the relevance of the -786C allele as a determinant of multivessel CAD in Caucasian patients.
Finally, we certainly agree with the conclusion by Dr. Batalla and colleagues that further studies are necessary in this field. However, the fact that in another study we showed the functional relevance of the same -786C allele in non-Japanese hypertensive patients lends further support to the relevance of this polymorphism for cardiovascular disease in Caucasians (5).
 |
References
|
|---|
1. Rossi GP, Cesari M, Zanchetta M, et al. The T-786C endothelial nitric oxide synthase genotype is a novel risk factor for coronary artery disease in Caucasian patients of the GENICA study. J Am Coll Cardiol. 2003;41:930–937[Abstract/Free Full Text]
2. Alvarez R, Gonzalez P, Batalla A, et al. Association between the NOS3 (-786 T/C) and the ACE (I/D) DNA genotypes and early coronary artery disease. Nitric Oxide. 2001;5:343–348[CrossRef][Medline]
3. Nakayama M, Yasue H, Yoshimura M, et al. T-786 C mutation in the 5'-flanking region of the endothelial nitric oxide synthase gene is associated with coronary spasm. Circulation. 1999;99:2864–2870[Abstract/Free Full Text]
4. Jeerooburkhan N, Jones LC, Bujac S, et al. Genetic and environmental determinants of plasma nitrogen oxides and risk of ischemic heart disease. Hypertension. 2001;38:1054–1061[Abstract/Free Full Text]
5. Rossi GP, Taddei S, Virdis A, et al. The T-786C and Glu298Asp polymorphisms of the endothelial nitric oxide gene affect the forearm blood flow responses of Caucasian hypertensive patients. J Am Coll Cardiol. 2003;41:938–945[Abstract/Free Full Text]
|
Top
References