The Reply

We appreciate the interest Garrett has shown in our work. She points out that while we described several comparisons as being not statistically significant, the magnitude of the differences we reported in mortality attributable to left ventricular hypertrophy by race and ethnicity suggests that the population-wide impact of our findings might be quite large. We acknowledge that we interpreted our results conservatively but agree that the implications of our findings might be greater than we implied. Although our sample might appear to be large enough to have low likelihood of a type II error, dividing it into 3 racial/ethnic groups limits power.

Dr. Garrett also raises the interesting question of more widespread use of echocardiography to detect left ventricular hypertrophy in subjects with prehypertension. Our study does not provide results that can address this question directly, but it suggests that electrocardiography may be useful in this situation given that it is predictive of mortality, is inexpensive, and is widely available. Whether electrocardiography should be part of the initial evaluation of patients with prehypertension is worthy of future consideration; subjects with prehypertension and Novacode scores greater than the thresholds we have previously described1 might benefit from drug treatment without waiting for progression to overt hypertension.