doi: 10.1161/01.CIR.0000141257.61498.6A
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2004;110:e49-e50.)
© 2004 American Heart Association, Inc.
Correspondence |
8-Iso-Prostaglandin F2
as a Risk Marker in Patients With Coronary Heart Disease
Institute of Clinical Physiology, CNR, Pisa, Italy, cristina.vassalle{at}ifc.cnr.it
To the Editor:
We read with great interest the manuscript by Schwedhelm et al1 reporting that urinary levels of 8-iso-prostagladin F2
(8-isoprostane) represent an independent and cumulative risk factor of coronary artery disease (CAD) together with different cardiovascular risk factors.
We agree that elevated levels of oxidative stress may be a risk factor for the onset and progression of the atherosclerotic process. However, we would like to draw attention to some assertions in this article. As quoted by the authors, we have recently observed that elevated levels of 8-isoprostane are associated with the extent and severity of CAD and with the occurrence of different atherogenic risk factors.2 Schwedhelm and colleagues did not find a similar association between 8-isoprostane levels and the severity of CAD. Nevertheless, we measured 8-isoprostane levels in plasma and not in urinary samples, as the authors incorrectly reported. The measurements in different biological samples may account, at least in part, for the differences observed.
Differences in population characteristics may also account for the lack of correlation with severity of CAD. In our study, we included patients referred for coronary angiography. Coronary stenosis was considered significant if there was luminal diameter narrowing of 
50% in at least one epicardial coronary artery. Schwedhelm et al1 recruited patients who had ischemia symptoms and controls who were asymptomatic for CAD. However, it is not clear whether all patients had undergone coronary angiography. Thus, the authors could clarify this point and inform Circulation readers on which criteria to determine CAD severity were adopted. Finally, the gas chromatography–tandem mass spectrometry (GC/MS) method is extremely reliable but also elaborate, requiring complex instrumentation and specialized operators. This fact likely limits the application of this knowledge for routine clinical diagnosis. Recent results indicate that the evaluation of F(2)-isoprostane with enzyme immunoassay (EIA) correlates well with values obtained by the GC/MS method3. The use of simple biological markers of oxidative stress may be relevant in adding a new predictive and prognostic factor of CAD. Thus, EIA could represent a straightforward and rapid procedure for evaluating oxidative stress particularly suitable for routine application, because it allows the use of equipment commonly present even in biochemical laboratories.
References
1. Schwedhelm E, Bartling A, Lenzen H, et al. Urinary 8-iso-prostaglandin F2 as a risk marker in patients with coronary heart disease: a matched case-control study. Circulation. 2004; 109: 843–848.
2. Vassalle C, Botto N, Andreassi MG, et al. Evidence for enhanced 8-isoprostane plasma levels, as index of oxidative stress in vivo, in patients with coronary artery disease. Coron Artery Dis. 2003; 14: 213–218.[CrossRef][Medline] [Order article via Infotrieve]
3. Devaraj S, Hirany SV, Burk RF, et al. Divergence between LDL oxidative susceptibility and urinary F(2)-isoprostanes as measures of oxidative stress in type 2 diabetes. Clin Chem. 2001; 47: 1974–1979.
Response
Clinical Pharmacology Unit, Institute of Experimental and Clinical Pharmacology, University Hospital Hamburg-Eppendorf, Hamburg, Germany
Institute of Clinical Chemistry, University Hospital Hamburg-Eppendorf, Hamburg, Germany
Institute of Mathematics and Data Processing in Medicine (J.C.B.),, University Hospital Hamburg-Eppendorf, Hamburg, Germany
Institute of Clinical Pharmacology, Hannover Medical School, Hannover, Germany
We agree with Drs Vasalle and Andreassi that data from our recent study1 evaluating urinary isoprostane excretion in patients with coronary heart disease are not easily comparable with their previous observations in plasma.2 Our case-control study was designed to assess differences between 8-iso-prostaglandin F2 (8-iso-PGF2) levels in coronary heart disease patients and healthy controls rather than to search for an association between urinary 8-iso-PGF2 and severity of disease. Hence, 54% of our patients had had a myocardial infarction, but not all patients underwent coronary angiography or myocardial scintigraphy. Secondly, urinary 8-iso-PGF2 may not reflect blood plasma levels of 8-iso-PGF2 because it is extensively metabolized before renal elimination. However, the metabolite of 8-iso-PGF2, ie 2,3-dinor-5,6-dihydro-8-iso-prostaglandin F2, was also quantified in our study and correlated excellently with its precursor. Finally, differences in the analytical methodologies used should be considered.
The enzyme immunoassay (EIA) used by Drs Vasalle and Andreassi for measuring 8-iso-PGF2 in plasma is not as straightforward and accurate as suggested. EIA requires extensive sample preparation as recommended by the supplier (see http://www.caymanchem.com), including use of 3H-labeled 8-iso-PGF2 and application of immunoaffinity column extraction or thin-layer chromatography for isolation of 8-iso-PGF2 before EIA analysis. In addition, a lack of specificity of the EIA at relevant concentrations is obvious. Devaraj et al3 reported a correlation between the EIA (y) and gas chromatography–tandem mass spectrometry (GC-MS) (x), ie, y=0.57x+0.81, r=0.80, n=68, for urine levels of up to 7.4 ng/mg creatinine. Meta-analysis of the data by Devaraj et al3 for pathologically relevant 8-iso-PGF2 levels of up to 4 and 3 ng/mg creatinine revealed a weaker correlation, ie, r=0.63 (n=60) and r=0.40 (n=54), respectively. Furthermore, it is generally accepted that in clinical measurement comparison of methods on the basis of correlation coefficients is inappropriate and misleading.4 Statistical analysis of the data of Devaraj et al3 by the method of Bland and Altman4 resulted in an unacceptable variation of the difference of –0.15±0.99 ng/mg creatinine (mean±SD, n=68) in the values obtained by EIA and GC-MS.
We would like to follow Keaney et al,5 who suggested that "a less precise ELISA (compared with GC-MS) would result in a random misclassification of levels and this would bias us toward the null hypothesis." The desire for simple and rapid analytical methods is understandable. However, reliability in terms of accuracy and precision is considered more important than rapidity in clinical research. Therefore, reliability should have priority over rapidity.
References
1. Schwedhelm E, Bartling A, Lenzen H, et al. Urinary 8-iso-prostaglandin F2 as a risk marker in patients with coronary heart disease. Circulation. 2004; 109: 843–848.
2. Vassalle C, Botto N, Andreassi MG, et al. Evidence for enhanced 8-isoprostane plasma levels, as index of oxidative stress in vivo, in patients with coronary artery disease. Coron Artery Dis. 2003; 14: 213–218.[CrossRef][Medline] [Order article via Infotrieve]
3. Devaraj S, Hirany SV, Burk RF, et al. Divergence between LDL oxidative susceptibility and urinary F2-isoprostanes as measures of oxidative stress in type 2 diabetes. Clin Chem. 2001; 47: 1974–1979.
4. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986; 1: 307–310.[CrossRef][Medline] [Order article via Infotrieve]
5. Keaney JF Jr, Larson MG, Vasan RS, et al. Obesity and systemic oxidative stress: clinical correlates of oxidative stress in the Framingham Study. Arterioscler Thromb Vasc Biol. 2003; 23: 434–439.
This article has been cited by other articles:
 |
|
 |
|
  R. H Boger, E. Schwedhelm, R. Maas, S. Quispe-Bravo, and C. Skamira ADMA and oxidative stress may relate to the progression of renal disease: rationale and design of the VIVALDI study Vascular Medicine, July 1, 2005; 10(1_suppl): S97 - S102. [Abstract] [PDF]
|
|
|
|
|
|
R. H Boger, E. Schwedhelm, R. Maas, S. Quispe-Bravo, and C. Skamira ADMA and oxidative stress may relate to the progression of renal disease: rationale and design of the VIVALDI study Vascular Medicine, May 1, 2005; 10(2_suppl): S97 - S102. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||