Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Treated diastolic blood pressure levels below 85 mm Hg are associated with an increased risk for cardiac events but not for stroke

ACP J Club. 1991 May-June;114:70. doi:10.7326/ACPJC-1991-114-3-070

Source Citation

Farnett L, Mulrow CD, Linn WD, Lucey CR, Tuley MR. The J-curve phenomenon and the treatment of hypertension. Is there a point beyond which pressure reduction is dangerous? JAMA. 1991;265:489-95.



To review the medical literature to answer the question: Is there a point beyond which blood pressure reduction in patients with hypertension is no longer beneficial and possibly even harmful?

Data sources

English-language studies published from 1966 to 1989 were identified from MEDLINE, references in pertinent articles, and written communications with experts on hypertension.

Study selection

Of 134 articles judged by 2 independent reviewers to be potentially relevant, 13 met the selection criteria. Studies were selected if they included patients with hypertension who received treatment with follow-up of ≥ 1 year; the end points of stroke, myocardial infarction, and mortality from all causes; outcomes stratified by treated blood pressure levels; and if the study design was a retrospective or prospective cohort or randomized trial. 1small study was excluded from further analysis.

Data extraction

Data were abstracted by 4 independent reviewers who assessed features of content and quality of research (overall weighted kappa for scoring articles was 0.62). A J-shaped relation was held to be shown if the relation between blood pressure and cardiovascular outcomes was not linear and if the lowest stratification of achieved level of blood pressure was associated with no decrease in cardiovascular events.

Main results

48 473 patients were included in the 12 studies. Several studies did not measure blood pressure before cardiovascular events occurred, adjust for cardiovascular risk factors and treatment co-interventions, adjust for length of follow-up for well and poorly controlled patients, or consider the metabolic effects of treatment. No consistent J-shaped relation between treated blood pressure level and stroke was seen. However, a J-shaped relation between treated diastolic blood pressure and cardiac events was shown by all but 1 study. Studies often did not correlate systolic blood pressure with cardiac events, and those that did gave conflicting results. A mathematical model of the data estimated that cardiovascular events were least frequent at a diastolic blood pressure of 85 mm Hg and that event rates were twice as high at 75 mm Hg (10.6 events per 1000 person-years) as at 85 mm Hg (5.2 events per 1000 person-years). Thus, the beneficial therapeutic threshold was a diastolic blood pressure of 85 mm Hg.


Treated diastolic blood pressure levels below 85 mm Hg are associated with increased risk for cardiac events but not for stroke.

Source of funding: Not stated.

Address for article reprint: Dr. C.D. Mulrow, Ambulatory Care (11C), Audie L. Murphy Memorial Veterans Hospital, 7400 Merton Minter Boulevard, San Antonio, TX 78248, USA.


The conventional wisdom that the lowest blood pressures are associated with the lowest risks is under serious attack. After Stewart reported that the frequency of myocardial infarction increased when diastolic pressure was reduced below 100 mm Hg (phase IV) among patients with severe hypertension (1), others have also found evidence for a J-shaped relation between diastolic blood pressure and cardiac, but not cerebral, mortality. The question is whether a lower bound can be established for blood pressure control during antihypertensive treatment, a level that minimizes overall risk.

Over the past few years, this matter has been subject to intense debate among hypertension specialists (2). The thorough review of available data by Farnett and colleagues suggests that the J-curve is a reality rather than a product of selection bias. The phenomenon occurs both in controls and treated patients, and a number of mechanisms relating to (pre-existing) impaired coronary flow reserve may be responsible. As the authors point out, conclusive evidence that it is unwise to lower diastolic pressure below 85 mm Hg, particularly in patients with coronary disease or left ventricular hypertrophy, has to await the outcome of trials in which patients are randomly assigned to different treatment goals. In the meantime, most practitioners will not readily add medication once the diastolic pressure has fallen to 85 mm Hg, knowing that below that threshold hardly any additional benefit can be expected, even for stroke prevention.

Gert A van Montfrans, MD
University of AmsterdamAmsterdam, The Netherlands


1. Stewart IM. Relation of reduction in pressure to first myocardial infarction in patients receiving treatment for severe hypertension. Lancet. 1979;1:861-5.

2. Zanchetti A. Clinical round table: how much should blood pressure be lowered? The problem of the J-shaped curve. J Hypertension. 1989;7(Suppl 6):S338-48.