Author + information
We thank Drs. Nieto and Blumenthal for their thoughtful and intriguing comments regarding our recent article (1). The major findings in our study were that asymptomatic black subjects had a lower prevalence of coronary calcium compared to white subjects, but nevertheless suffered more coronary heart disease events during a 70-month follow-up period. After adjusting for age, gender, and coronary risk factors, black race was associated with an odds ratio of 2.16 for a coronary event (95% CI limits, 1.34–3.48). Drs. Nieto and Blumenthal advance two possible explanations of our findings:
1. Blacks suffered higher rates of co-morbid conditions, risk factors, and diminished access to health care treatment, collectively impacting event-free survival deleteriously.
2. The duration from disease onset to event occurrence differed in blacks compared to whites, secondary perhaps to a higher case-fatality rate among our black subjects (incidence-prevalence bias).
Drs. Nieto and Blumenthal suggest that our findings may at least in part be explained by higher co-morbidities and risk factors in blacks, combined with lower access to health care. This possibility cannot be excluded, as we stated in our article. Blacks underwent revascularization at rates similar to whites, yet it could conceivably be argued that the rates of revascularization should have in fact been higher in blacks, commensurate with their event rate. Blacks in our study had roughly equivalent coronary risk factors to whites; however, it is not at all clear that standard coronary risk factors derived from epidemiologic investigations comprised almost exclusively of white subjects are applicable to blacks to the same extent that they are to whites. In fact, as we pointed out, there is evidence that some risk factors such as smoking, hypertension, and cholesterol have a different impact on black subjects compared to white subjects. Although black and white subjects in our study had equivalent overall Framingham risk, black subjects had significantly higher systolic blood pressure and body mass index and a higher incidence of diabetes mellitus and a history of hypertension. Conversely, white subjects were older, and they had a higher incidence of positive family history of coronary heart disease. As we pointed out, it is possible that the variable impact of standard risk factors on differing ethnic groups could have affected our results.
Drs. Nieto and Blumenthal suggest that the inverse relation between incidence (I) and disease duration (D) might explain the lower prevalence of calcification and the higher incidence of clinical events. If the proposed relation Prevalence = I × D were valid, I could be higher for blacks even though Prevalence was lower only if the disease duration, D, were much shorter for blacks. However, this was not the case: the duration between screening and events was the same for both ethnic groups (Table 1).
In addition, prevalence of calcification at the time of screening is not the same as prevalence of clinical disease or even of atherosclerosis. Calcification is only imperfectly related to atherosclerosis—and the latter only imperfectly related to clinical events. Our findings regarding higher coronary event rates but lower prevalence of calcium in blacks thus cannot be explained in the manner proposed by Drs. Nieto and Blumenthal, and they pose intriguing and important questions for further investigation.
- American College of Cardiology