Tuesday April 26th 9 pm Eastern
Note: *Thursday* April 28th 8 pm BST
2016 Apr 2. [Epub ahead of print]
Blood-Pressure Lowering in Intermediate-Risk Persons without Cardiovascular Disease.
Lonn EM, Bosch J, López-Jaramillo P, Zhu J, Liu L, Pais P, Diaz R, Xavier D, Sliwa K, Dans A, Avezum A, Piegas LS, Keltai K, Keltai M, Chazova I, Peters RJ, Held C, Yusoff K, Lewis BS, Jansky P, Parkhomenko A, Khunti K, Toff WD, Reid CM, Varigos J, Leiter LA, Molina DI, McKelvie R, Pogue J, Wilkinson J, Jung H, Dagenais G, Yusuf S; HOPE-3 Investigators.
- Free full text available at NEJM website, until formal publication (~ May)
- Supplementary material (PDFs of protocol, appendix and disclosures)
- Accompanying NEJM editorial
- Also, HOPE-3 statin study and
- HOPE-3 protocol paper at CJC ($wall alert)
Hot on the heels of the SPRINT and ACCORD trials comes HOPE-3, an important study examining the potential benefits of statins and anti-hypertensives for primary prevention in patients at intermediate risk for cardiovascular (CV) events. For the purposes of this #NephJC, we will be focusing on the results of the blood pressure (BP) analysis only, as reported by Lonn et al.
HOPE-3 employed a randomized, double-blind, 2x2 factorial design (Figure above). Participants were included if they were
- male over the age of 55
- female over the age of 65
- AND whilst also possessing at least one CV risk factor:
- Exclusion criteria included subjects with documented CV disease and those with an eGFR <45ml/min/1.73m^2.
How does this compare with SPRINT and ACCORD?
Those who proceeded to randomization received either
- a combination of 16mg candesartan and 12.5mg hydrochlorothiazide or
- As part of the overall study, subjects were also randomized to receive either 10mg rosuvastatin or placebo (Figure S3).
BP was taken as the mean of *2* measurements taken after 5 minutes’ rest, with an Omron device. Median follow-up was 5.6 years.
- 1st co-primary outcome: Composite of CV death, non-fatal myocardial infarction (MI) or non-fatal stroke.
- 2nd co-primary outcome: As 1st co-primary, plus resuscitated cardiac arrest, heart failure or coronary revascularization
- 1st co-secondary outcome: As 2nd co-primary, plus angina with evidence of ischaemia
- 2nd co-secondary outcome: Fatal or non-fatal stroke
Of the 14,682 participants included in the study, 12,705 were randomized (see Figure S1). Patients were well age- and sex-matched. Mean initial BP readings were also similar: 138.2±14.7 mmHg and 137.9±14.8 mmHg for the intervention and placebo groups, respectively. How does the patient population compare with SPRINT and ACCORD? see below for comparative table 1s:
There was a clear separation in mean BP readings (a mean difference of 6.0±13.0 mmHg) (see figure 1 below).
Before we go into outcomes, lets again compare with SPRINTand ACCORD
The headline results from this study are displayed in Table 2. There was no significant difference in the incidence of either primary or secondary outcomes between the treatment and placebo arms of the study.
Figures 2 A, B and C display these findings in graphical form, and clearly demonstrate no significant reduction in the incidence of the 1st co-primary outcome, stroke or MI, respectively.
When participants were grouped according to baseline BP readings (a prespecified sub-group analysis), anti-hypertensive therapy was found to benefit those whose initial BP reading was >143.5mmHg; a significant reduction in the incidence of the first and second co-primary outcomes was observed (Forest plots 3 A and B, below).
Note the effect on stroke. The difference may *look significant* in a subgroup - but the trend was not significant (p 0.217 for heterogeneity).
Significantly more patients in the intervention group withdrew from the study due to symptomatic hypotension (3.4% vs. 2.0%, p<0.001). There was no reported difference in adverse renal outcomes (0.5% vs. 0.3%, p=0.13). And lastly, how does this compare with SPRINT and ACCORD?
Just 6 months have passed since the #NephJC community debated the relative merits of the SPRINT trial. SPRINT randomised 9361 participants with increased cardiovascular risk (but not diabetes mellitus) to a target systolic BP of either <120mmHg or <140mmHg. Take-home messages included a 25% reduction in the primary endpoint and, remarkably, a 27% reduction in all-cause mortality at 2 years. Key points raised during the lively #NephJC chat were the failure of intensive BP treatment to reduce the incidence of stroke and the lack of benefit for patients with CKD.
The ACCORD trial randomised 4733 participants with Type II diabetes mellitus to the same target BPs as SPRINT. A 12% reduction in the incidence of the primary outcome was reported, although this was not statistically significant. However, intensive BP treatment did significantly lower the risk of stroke (heart failure reduction drove much of the outcome) between study groups.
Crucial differences between these two trials were the greater statistical power of SPRINT, which was related to sample size (9361 vs. 4733 participants), the older population of SPRINT (mean age 68 vs. 62), and the exclusion of patients with CKD in ACCORD. Nonetheless, their results highlight some similarities and trends, as can be seen in the following forest plot:
In HOPE-3, the mean age of the study population falls between SPRINT and ACCORD, at 65 years. However, the population is fundamentally different from those recruited for either of SPRINT or ACCORD. Participants in HOPE-3 were a much *lower risk* population, primarily because those with clinical CV disease or diabetes mellitus were excluded. Nonetheless, patients recruited into HOPE-3 who were at higher risk, such as those with a high baseline BP reading (>143.5mmHg), did derive significant benefit from anti-hypertensive therapy with regards to the 1st and 2nd co-primary outcomes.
Another important point to note is that HOPE-3 did not target a specific systolic BP, as was the case in both SPRINT and ACCORD. In HOPE-3, mean BP achieved was 6.0±13.0 mmHg lower in the intervention arm than those treated with placebo, compared to mean reductions of approximately 14mmHg in the intensive treatment arms of SPRINT and ACCORD. It remains possible that a more marked BP reduction in the HOPE-3 population could have resulted in significant differences between intervention groups. However, the more intensive BP lowering strategies of SPRINT and ACCORD also require closer monitoring and more regular follow-up; something which is harder to justify in a lower risk cohort.
Finally, how do we interpret the null effect on incident stroke observed in SPRINT and HOPE-3? It is possible that being diabetic confers an independently increased risk of stroke. The truth is, however, that some of the published data is conflicting. A meta-analysis by Law et al. found broadly similar results to ACCORD, yet another more recent Lancet paper found that lowering systolic BP to <130mmHg was not associated with a significant reduction in recurrent stroke. Moreover, it’s worthwhile to note that in SPRINT (and presumably also in HOPE-3 as stroke = clinical CV disease), patients with prior stroke or TIA were excluded from the trial.
Where does this latest huge BP RCT leave us regarding goals of hypertension management? Is there now more clarity or more confusion? Please join the discussion in the upcoming #NephJC.
Summary by Peter Gallacher, Royal Infirmary of Edinburgh,
Additional comparison tables by Swapnil Hiremath
Tuesday April 26th 9 pm Eastern
Note: *Thursday* April 28th 8 pm BST