[The authors respond:]
We appreciate Murray Finkelstein's comments about our systematic review; however, we disagree with his conclusion. We carefully chose the wording of our 2 concluding statements. Our first statement, that "low-dose thiazide therapy can be prescribed as the first-line treatment of hypertension with confidence that the risk of death, coronary artery disease and stroke will be reduced," is substantiated by the statistical significance (95% confidence intervals) of the reduction of total mortality, coronary artery disease and stroke with low-dose thiazides, as presented in Table 4. Our second statement was that "the same cannot be said for high-dose thiazide therapy, β-blockers, calcium-channel blockers or ACE inhibitors." A statistically significant reduction in all 3 measures has not been shown for high-dose thiazides, β-blockers or calcium-channel blockers (in Table 4 the confidence intervals include 1.00). Nor has it been shown for ACE inhibitors or any other class of drugs, as they have not been studied in trials meeting the criteria of this review.
We therefore cannot prescribe these other classes as first-line agents with confidence that they will reduce each of these 3 adverse outcomes. We did not conclude, as suggested by Finkelstein, that only low-dose thiazides will prevent death and cardiovascular morbidity. Nor did we conclude, as Finkelstein has, anything about the relative effectiveness of low-dose thiazides and the other classes of drugs; the available head-to-head evidence is insufficient to comment on the relative effectiveness of the different classes of antihypertensive drugs.
We did demonstrate in this review that using thiazides as first-line therapy was associated with a greater reduction in systolic blood pressure and a lower rate of withdrawal for adverse drug effects than that associated with some of the other classes of antihypertensive drugs. We did not comment on the cost advantage of thiazides but are pleased that Finkelstein has made this point.