The relationship of all-cause mortality to average on-treatment systolic blood pressure is significantly related to baseline systolic blood pressure: implications for interpretation of the Systolic Blood Pressure Intervention Trial study
Okin, P.M.; Kjeldsen, S.E.; Devereux, R.B.
Journal of Hypertension 36(4): 916-923
The SPRINT study demonstrated that targeting systolic blood pressure (SBP) less than 120 mmHg was associated with lower cardiovascular event and mortality rates. In the LIFE study, however, a lower achieved SBP was associated with increased mortality. Mean baseline SBP in SPRINT was 140 mmHg and a third of the population had a baseline SBP 132 mmHg or less, raising the question of whether the lower baseline SBP in SPRINT could in part account for these differences. All-cause mortality during 4.8 ± 0.9 years follow-up was examined in relation to tertiles of achieved on-treatment average SBP in patients with baseline SBP of 25th percentile or less versus greater than 25th percentile value of 164 mmHg in 7998 nondiabetic hypertensive patients with ECG left ventricular hypertrophy randomly assigned to losartan-based or atenolol-based treatment. Average on-treatment SBP less than 142 mmHg (lowest tertile) and average SBP 142 mmHg to less than 152 mmHg (middle tertile) were compared with average SBP at least 152 mmHg (highest tertile and reference group). In the overall population, there was a significant interaction between baseline SBP 164 mmHg or less and average on-treatment SBP less than 142 mmHg in Cox analysis (χ = 15.48, P < 0.001). Among patients with baseline SBP greater than 164 mmHg, in multivariate Cox analyses adjusting for other potential predictors of mortality and a propensity score for having baseline SBP 164 mmHg or less and compared with average on-treatment SBP at least 152 mmHg, average on-treatment SBP less than 142 mmHg was associated with 32% higher mortality (hazard ratio 1.32, 95% CI 1.01-1.65), whereas average SBP of 142 mmHg to less than 152 mmHg was associated with 24% lower mortality (hazard ratio 0.76, 95% CI 0.59-0.98). In contrast, among patients with baseline SBP 164 mmHg or less, both average on-treatment SBP less than 142 mmHg (hazard ratio 0.60, 95% CI 0.36-0.99) and average SBP of 142 mmHg to less than 152 mmHg (hazard ratio 0.51, 95% CI 0.30-0.89) were associated with significantly lower mortality compared with average SBP of at least 152 mmHg. Achievement of an average SBP less than 142 mmHg was associated with reduced mortality in patients with baseline SBP 164 mmHg or less but with increased mortality in those with higher baseline SBP in LIFE. These findings suggest that the lower mortality associated with a lower targeted SBP in SPRINT may not be applicable to patients with considerably higher baseline SBP than SPRINT patients. Further study is necessary to better understand these findings. http://clinicaltrials.gov/ct/show/NCT00338260?order=1.