Are retinal examinations useful in assessing cardiovascular risk?

To the Editor: The results from the VAMPAHICA Study Group presented in the article “Usefulness of Optic Fundus Examination With Retinography in Initial Evaluation of Hypertensive Patients.” provide evidence that the use of retinography would increase the proportion of people classified as being at high risk of developing a major cardiovascular event within ten years.1 However, the study does not answer an important question: does the use of retinography result in more accurate cardiovascular risk stratification? To do this, the authors should examine whether people have been correctly reclassified. In the World Health Organization/ International Society of Hypertension (WHO-ISH, 2003) guidelines, the presence of target organ damage is used to assign people to the highest risk category.2 Therefore, use of retinography as a marker of target organ damage will assign people to a higher risk category as seen in Table 3 of the article by the VAMPAHICA Study Group. However, reclassification of people to a higher category alone is not sufficient evidence that retinography “provides a more accurate cardiovascular risk stratification”. Consideration needs to be given to whether these people are reclassified correctly or incorrectly. When people are reclassified, some will be reclassified correctly (e.g., people who develop a major cardiovascular event are reclassified to a higher risk category) whereas others will be reclassified incorrectly (e.g., people who do not develop a major cardiovascular event are reclassified to a higher risk category). The Net Reclassification Improvement combines the proportion of people classified correctly and incorrectly into a summary measure that can be used to determine whether the reclassification results in improved risk stratification.3 If data on the cardiovascular events are available, we encourage the authors to examine whether the use of retinography improves the proportion of people correctly and incorrectly reclassified. This would provide a more comprehensive assessment of the clinical utility of retinography in assessing cardiovascular risk.