Many studies have established that mortality from heart attacks is related to diastolic blood pressure in a J-curve with increased mortality at low as well as high pressures. This has been observed in untreated as well as treated patients and a similar phenomenon has been found in large population studies. Two hypotheses to account for this curve have been advanced. The direct causation hypothesis attributes the increased mortality at low pressures to low coronary perfusion. The reverse causation hypothesis attributes the curve to a subgroup of patients with low blood pressures as a result of pre-existing disease. This controversy is still unresolved and needs a prospective trial to decide the issue.