Aspects of blood pressure and lipid lowering drug treatment in individuals stratified by absolute cardiovascular disease risk

Background Primary prevention of cardiovascular disease (CVD) based on an absolute risk approach is more cost-effective than the traditional individual risk factor approach based on blood pressure (BP) or lipid thresholds alone. Although BP lowering drug treatment is recommended in high-risk individuals even for those below traditional BP thresholds (e.g. <140/90 mmHg) in many guidelines, withholding BP lowering drug in individuals with mildly elevated BP and low-moderate risk is controversial as it is contrary to historical recommendations and there are concerns regarding legacy effects. Similarly, lipid-lowering drug treatment (LLT) is widely prescribed in high-risk individuals based on evidence from large randomised controlled trials. However, these trials lacked elderly participants who are at high baseline risk due to their age. The effects of lipid-lowering drug treatment in the healthy elderly remain unclear. Aim To investigate the effect of BP and LLT on short- and long-term mortality outcomes stratified by absolute CVD risk. Methods: Two post-hoc observational studies of the Australian National Blood Pressure study (ANBP), the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT) and a systematic review and meta-analysis were conducted to investigate the effects of BP lowering drug treatment in individuals stratified by absolute CVD risk on short- and long-term outcomes. ANBP was a community-based randomised placebo-controlled trial in 'mild hypertension' defined as diastolic BP ranging from 95-109 mmHg and systolic BP lower than 200 mmHg conducted in Australia in the 1970s. ALLHAT was a double-blind, randomised active-controlled trial conducted in multiple centres in North America that included participants with untreated systolic BP lower than 180 or treated systolic BP lower than 160 mmHg. Based on the BP treatment status at baseline, participants were re-stratified to 'treatment na‚àövòve' and 'previous treatment' group. The systematic review and meta-analysis included randomised controlled trials with a post-trial phase and which enrolled participants who were middle-aged and had mildly elevated BP. Similarly, to investigate the effects of LLT in the elderly, a post-hoc observational study on the Second Australian National Blood Pressure study (ANBP2) was conducted that re-stratified participants into 'LLT' and 'no LLT' group based on their treatment status at baseline. All four studies excluded participants with previous CVD events. Treatment effects were estimated by hazard ratio with 95% confidence interval (CI) using a Cox proportional hazard model. Subgroup analyses by baseline estimated CVD risk score were performed. Results Generally, the three studies on the effects of BP lowering drug treatment did not record any clinical harms as a result of delayed drug treatment in low-moderate risk individuals or in middle-aged adults with mildly elevated BP. The post-hoc study on the ANBP population (median follow-up 4.4. years) observed a substantial beneficial effect of BP lowering drug treatment regarding absolute risk reduction in any trial endpoint, all-cause mortality and major CVD events in the highest risk tertile group, whereas the low or moderate risk tertile was unlikely to benefit. In the ALLHAT trial with a longer follow-up period (up to 14.1 years), delayed BP lowering drug treatment was not associated with any significantly increased risk of all-cause or CVD mortality at any level of CVD risk stratification when drug therapy was closely monitored by a clinician. Similarly, in the systematic review and meta-analysis in middle-aged adults with mildly elevated BP, we found non-significant effects of delayed BP lowering drug treatment in short- and long-term all-cause and CVD mortality regardless of the CVD risk stratification. In the study looking at the effects of LLT in those aged 65 years or over stratified by absolute CVD risk, LLT was associated with a reduced long-term all-cause mortality suggesting that the mortality benefit of LLT for the elderly may take longer to become evident at any level of CVD risk. High-risk participants also obtained further benefits for short-term all-cause mortality. Conclusions In terms of the effects of BP lowering drug treatment, our analysis provides further justification that an absolute risk strategy is superior to management based on the BP level alone in identifying those who are most likely to benefit from drug therapy. The results support using absolute CVD risk to determine when to initiate BP lowering drug treatment for primary prevention of CVD. These studies found no long-term adverse risk of all-cause or CVD mortality in low-moderate risk individuals or middle-aged adults with mildly elevated BP, thus partly addressing clinician concerns of 'legacy effects' when BP lowering drug therapy is delayed. In contrast, LLT in the elderly may require long-time follow-up (e.g. 10 years) to become evident.