No evidence for a J-shaped curve in treated hypertensive patients with increased cardiovascular risk: The VALUE trial

Sverre E. Kjeldsen, Eivind Berge, Sripal Bangalore, Franz H. Messerli, Giuseppe Mancia, Björn Holzhauer, Tsushung A. Hua, Dion Zappe, Alberto Zanchetti, Michael A. Weber, Stevo Julius

Research output: Contribution to journalArticlepeer-review


Previous studies have debated the notion that low blood pressure (BP) during treatment, particularly diastolic (DBP), is associated with increased risk of cardiovascular disease. We evaluated the impact of low BP on cardiovascular outcomes in a high-risk population of 15,244 hypertensive patients, almost half of whom had a history of coronary artery disease (CAD). In the prospective Valsartan Antihypertensive Long-term Use Evaluation (VALUE) trial, patients were randomized to valsartan or amlodipine regimens and followed for 4.2 years (mean) with no difference in the primary cardiovascular endpoint. A Cox proportional hazards model was used to evaluate the relationship between average on-treatment BP and clinical outcomes. The relationship between BP and cardiovascular events was adjusted for age, gender and body mass index, and baseline qualifying risk factors and diseases (smoking, high total cholesterol, diabetes mellitus, proteinuria, CAD, previous stroke and left ventricular hypertrophy). DBP ≥ 90 mmHg, compared with <90 mmHg, was associated with increased incidence of the primary cardiovascular endpoint (all cardiac events); however, DBP <70 mmHg, compared with ≥ 70 mmHg, was not associated with increased incidence after covariate adjustment (no J-shaped curve). Similar results were observed for death, myocardial infarction (MI), heart failure and stroke, considered separately. Nadir for MI was at DBP of 76 mmHg and for stroke 60 mmHg. The ratio of MI to stroke increased with lower DBP. In CAD patients the MI to stroke ratio was more pronounced than in patients without CAD but there was no significant J-curve in either group. Systolic BP ≥ 150 but not <130 mmHg, compared with 130-149 mmHg, similarly was associated with increased risk for primary outcome. In conclusion, patients in BP strata ≥ 150/90 mmHg, but not patients in BP strata <130/70 mmHg, were at increased risk for adverse outcomes in this hypertensive, high-risk population. Although benefit in preventing MI in relation to preventing stroke levels off for the lowest BPs, these data provide no support for a J-curve in the treatment of high-risk hypertensive patients. The increase in the ratio of MI to stroke with lower DBP indicates target organ heterogeneity in that the optimal on-treatment DBP for cerebroprotection is below that for cardioprotection.

Original languageEnglish
Pages (from-to)83-92
Number of pages10
JournalBlood Pressure
Issue number2
Publication statusPublished - Mar 3 2016


  • Blood pressure-lowering treatment
  • cardiovascular morbidity
  • cardiovascular mortality
  • clinical trial
  • hypertension

ASJC Scopus subject areas

  • Internal Medicine
  • Cardiology and Cardiovascular Medicine


Dive into the research topics of 'No evidence for a J-shaped curve in treated hypertensive patients with increased cardiovascular risk: The VALUE trial'. Together they form a unique fingerprint.

Cite this