In patients with
hypertension, the long-term cardiovascular and all-cause mortality effects of different blood pressure-lowering regimens and
lipid-lowering treatment are not well documented, particularly in clinical trial settings. The Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT) Legacy Study reports mortality outcomes after 16 years of follow-up of the UK participants in the original ASCOT trial.
METHODS: ASCOT was a multicentre randomised trial with a 2 × 2 factorial design. UK-based patients with
hypertension were followed up for all-cause and cardiovascular mortality for a median of 15·7 years (IQR 9·7-16·4 years). At baseline, all patients enrolled into the blood pressure-lowering arm (BPLA) of ASCOT were randomly assigned to receive either
amlodipine-based or
atenolol-based blood pressure-lowering treatment. Of these patients, those who had total
cholesterol of 6·5 mmol/L or lower and no previous
lipid-lowering treatment underwent further randomisation to receive either
atorvastatin or placebo as part of the
lipid-lowering arm (LLA) of ASCOT. The remaining patients formed the non-LLA group. A team of two physicians independently adjudicated all causes of death.
FINDINGS: Of 8580 UK-based patients in ASCOT, 3282 (38·3%) died, including 1640 (38·4%) of 4275 assigned to
atenolol-based treatment and 1642 (38·1%) of 4305 assigned to
amlodipine-based treatment. 1768 of the 4605 patients in the LLA died, including 903 (39·5%) of 2288 assigned placebo and 865 (37·3%) of 2317 assigned
atorvastatin. Of all deaths, 1210 (36·9%) were from cardiovascular-related causes. Among patients in the BPLA, there was no overall difference in all-cause mortality between treatments (adjusted hazard ratio [HR] 0·90, 95% CI 0·81-1·01, p=0·0776]), although significantly fewer deaths from
stroke (adjusted HR 0·71, 0·53-0·97, p=0·0305) occurred in the
amlodipine-based treatment group than in the
atenolol-based treatment group. There was no interaction between treatment allocation in the BPLA and in the LLA. However, in the 3975 patients in the non-LLA group, there were fewer cardiovascular deaths (adjusted HR 0·79, 0·67-0·93, p=0·0046) among those assigned to
amlodipine-based treatment compared with
atenolol-based treatment (p=0·022 for the test for interaction between the two blood pressure treatments and allocation to LLA or not). In the LLA, significantly fewer cardiovascular deaths (HR 0·85, 0·72-0·99, p=0·0395) occurred among patients assigned to
statin than among those assigned placebo.
INTERPRETATION: Our findings show the long-term beneficial effects on mortality of
antihypertensive treatment with a
calcium channel blocker-based treatment regimen and
lipid-lowering with a
statin: patients on
amlodipine-based treatment had fewer
stroke deaths and patients on
atorvastatin had fewer cardiovascular deaths more than 10 years after trial closure. Overall, the ASCOT Legacy study supports the notion that interventions for blood pressure and
cholesterol are associated with long-term benefits on cardiovascular outcomes.
FUNDING: Pfizer.