low blood pressure goal is better

 

A new Chinese study found that lowering the systolic blood pressure (SBP) to under 120mmHg in patients with high cardiovascular risk was associated with fewer vascular events and only a minor excess in adverse events (see htn BP lower than 120 better ESPRIT study Lancet2024 in dropbox, or doi.org/10.1016/ S0140-6736(24)01028-6)

 

Details:

-- 11,255 hypertensive participants with high cardiovascular risk were enrolled in an open-label, blinded-outcome study from 116 hospitals or communities in China and randomized to a systolic blood pressure target of <120mmHg or <140mmHg from 2019-2020

-- participants were at least 50yo with high cardiovascular risk, defined as established cardiovascular disease or at least 2 major cardiovascular risk factors defined as being at least 60yo for men or 65yo for women, diabetes, dyslipidemia (defined as total cholesterol >200, or LDL >130, or HDL<40), and current smoker (defined as smoking >1 cigarette/day in past 12 months)

    -- 4359 had diabetes, 3022 had a prior stroke

-- exclusion criteria: known secondary cause of hypertension, one-minute standing systolic blood pressure <110mmHg, left ventricular ejection fraction <35% or symptomatic heart failure in past 6 months, scheduled revascularization within next 6 months, eGFR <45, >2+ proteinuria in past 6 months, and active alcohol or substance abuse in past 12 months

-- blood pressure was assessed by a trained investigator using an electronic blood pressure monitor which was connected to a computer (to avoid transcription errors from writing down the numbers), with BP measured 3 times at an interval of 1 minutes after the participant had a quiet rest of 5 minutes. The mean value of these 3 SBP was then recorded.

-- after participant randomization to the BP target groups, they were followed at months 1, 2, and 3, and then every 3 months, with additional visits as needed to titrate to the appropriate target goal

-- ten types of free drugs were provided for participants: ACE inhibitors (enalapril), ARBs (valsartan or irbesartan), calcium channel blockers (nifedipine or amlodipine), thiazide-like diuretics (hydrochlorothiazide or indapamide), b-blockers (metoprolol); other meds could be used but were not provided for free in the trial

-- mean age 65 (25% were >70yo, 50% 60-69yo), 41% female, 31% current smoker/19% former/50% never, excessive drinking 12%/moderate drinking 20%/not drinking 68%, BMI 26 (27% had BMI >28, the cutpoint defined as “obesity” in Chinese individuals)

-- hypertension had been diagnosed a mean of 10 years; diabetes in 39%, with average diagnosis for 7yrs; coronary heart disease in 29%; stroke 27%

-- mean systolic BP 147 mmHg, mean diastolic BP 83 mmHg; number of BP meds were 1 in 43%/2 in 39%/3 in 14%/more in 2%; eGFR (creatinine-based) 83, total cholesterol 155/LDL 89/HDL 35/triglycerides 151; statin use in 47%; aspirin use in 43%

 

--primary outcome: composite of MI, revascularization, hospitalization for heart failure, stroke, or death from cardiovascular causes, per intention-to-treat analysis

-- prespecified subgroups: age, sex, systolic BP tertiles, time from hypertension diagnosis, diabetes and time from diagnosis, coronary heart disease, stroke, peripheral arterial disease, atrial fibrillation, smoking status, alcohol, consumption, BMI, LDL, HDL, number of BP meds, statin use, antiplatelet treatment

-- prespecified secondary outcomes: components of the primary composite outcome, death from any cause, a composite of the primary outcome or death from any cause, and composite kidney outcome (end-stage kidney disease, a sustained decline in eGFR to <10, death from renal causes, or a sustained decline of >40% in eGFR from baseline). Sustained decline in eGFR required 2 measurements during 2 planned consecutive follow-ups

 

-- follow-up 3.4 years

 

Results:

-- achieved systolic blood pressure:

    -- intensive treatment group: 119.1 mmHg (stabilization at this level took 9 months), requiring 2.7 meds on average

    -- standard treatment group: 134.8 mmHg (stabilization at this level took 2 months), requiring 2.0 meds on average

 

-- meds used at final follow-up, comparing intensive vs standard care treatment arms:

    -- ACEi or ARB: 79% vs 65%

    -- betablocker: 58% vs 46%

    -- calcium channel  blocker: 88% vs 72%

    -- diuretic: 43% vs 15%

 

-- number of meds used, comparing intensive vs standard care treatment arms:

    -- 0: 2% vs 3%

    -- 1: 9% vs 27%

    -- 2: 30% vs 41%

    -- 3: 38% vs 23%

    -- more: 22% vs 5%

 

-- primary outcome:

   -- intensive treatment group: 547 participants (9.7%)

   -- standard treatment group: 623 participants (11.1%)

        -- 12% decrease in intensive treatment group, hazard ratio (HR) 0.88 (0.78-0.99), p=0.028

            -- and, if revascularization is excluded in the primary outcome analysis (often not included in the composite of major adverse cardiovascular events), there was a 16% decrease with intensive treatment, HR 0.84 (0.74-0.96)

 

    -- per exploratory analysis, benefit after at least one year of treatment:

        -- 22% decrease in intensive treatment group, HR 0.78 (0.67-0.90), p=0.028

    --number-needed-to-treat (NNT) with intensive BP control (which was independent of whether they had diabetes or had a stroke):

        -- primary outcome: 75 patients for 3 years to prevent one primary outcome

        -- cardiovascular death: 148 patients for 3 years to prevent one death

 

Here is graph of the outcomes: the bottom graph (inset) in A is the same data plotted on a larger scale; B is a plot of hazard ratios by lengths of follow-up time

 

 

--  causes of death, comparing intensive vs standard care treatment arms:

    -- all causes of death: 160 vs 203, 21% decrease with intensive treatment, HR 0.79 (0.64-0.97), p<0.0082

    -- cardiovascular causes: 59 (1.1%) vs 97 (1.7%), 39% decrease with intensive treatment, HR 0.61 (0.44-0.84), with major differences:

        -- coronary death (including sudden death): 37 vs 62

        --stroke: 16 vs 28

    -- non cardiovascular deaths: 92 vs 99

-- Composite of primary outcome or death from any cause 11% decrease, HR 0.89 (0.80-0.89) p=0.0049

-- none of the individual outcomes reached statistical significance

--composite kidney outcomes, comparing intensive vs standard care treatment arms:

    -- 169 (3.0%) vs 102 (1.8%), 70% more with the intensive med group, HR 1.70 (1.33-2.17), though per the table below, essentially no very serious complications

 

 

-- Serious adverse events: 2366 (42.1%) in those in the intensive treatment group vs 2378 (42.2%) in the standard treatment group, no statistically significant difference

    -- syncope: 24 (0.4%) vs 8 (0.1%) was significantly different: HR 3.00(1.35-6.68); all were hospitalized and 2 had a fracture (one in each group)

    -- no significant difference in hypotension (0.1% vs 0.1%), electrolyte abnormalities (0.2% vs 0.2%), injurious falls (0.5% vs 0.4%), or acute kidney injury (0.1% vs <0.1%)

 

Commentary:

-- the above study benefited from being large, using multiple sites of care (both hospital and community setting), high adherence to meds, few people lost to follow-up, and lots of clinical outcomes to compare

-- this study found that patients at high cardiovascular risk did better with a goal systolic BP<120 mmHg than <140mmHg

    -- and, the NNT for 3 years was 75 for a primary outcome and 148 for cardiovascular death

        --but the perspective here is that at 3 years, the curves above are splaying apart (ie the benefit of intensive BP therapy is greater at the 3-year mark than before that). and hypertension is typically a lifelong disease requiring lifelong treatment, so the NNT may in fact be much lower to show benefit of the intensive therapy over more time

    -- this study also found no difference in patients with either a history or by duration of diabetes; similarly there was no outcome difference in those with a history of stroke

    -- of note, those on statins with intensive BP therapy in the study had a somewhat stronger benefit in the more intensive group, indicating at least that treatment with statins did not reduce the benefit of the lower achieved blood pressure

    -- one difference in this study vs much of actual clinical practice is the relatively low use of diuretics. they did not provide information on the percent of people on hydrochlorothiazide vs indapamide). There are real concerns about the benefit of HCTZ as the diuretic: see  https://gmodestmedblogs.blogspot.com/2016/04/chlorthalidone-is-better-than-hctz-for.html , as well as a critique of a recent VA study finding no difference between HCTZ and chlorthalidone https://gmodestmedblogs.blogspot.com/2023/01/hypertension-hctz-vs-chlorthalidone.html

 

-- the currentstudy beckons comparison to the US SPRINT trial (https://gmodestmedblogs.blogspot.com/2015/11/tighter-blood-pressure-control-sprint.html ), but has important differences:

    -- SPRINT excluded patients with diabetes and stroke, 2 very important and prevalent groups with hypertension and well-represented in the patients we see regularly

    -- SPRINT used a rather eclectic and largely non-reproducible methodology to assess blood pressure (ie, not practicable in clinical practice), which may well translate into their blood pressure recordings being about 5-10 mmHg lower than what would be expected in a standard clinical setting: see https://gmodestmedblogs.blogspot.com/2017/02/blood-pressure-guidelines-for-older.html for detail. But, in brief, the achieved SBP in SPRINT was 123 mmHg in the tight control group, but that seems to be equivalent to a clinic-based blood pressure of 130mmHg or so

    -- that all being said, though SPRINT did find a relative risk reduction in major vascular events and cardiac deaths lower than the current study, the absolute risk reduction and NNT was similar (likely that the RRR was distorted by the current study including revascularization as part of the primary outcome, which may have diluted the RRR calculation)

 

-- also, both SPRINT and the current study did find that intensive treatment was associated with more patients having sustained declines in renal function, but no renal-related deaths or end stage renal disease

-- other trials of targeted blood pressure reduction:

    -- HOT trial, the 1998 trial "Hypertension  Optimal Treatment" study, a 26 country trial with 18,790 patients aged 50-80, targeting diastolic blood pressure (DBP), decreased the initial mean DBP of 105 mmHg by 20.3 mmHg reduction with a DBP goal <90mmHg, by 22.3 mmHg reduction with a DBP goal of <85 mmHg, and by 24.3 mmHg reduction for a goal DBP of <80mmHg; this study found that the lowest incidence of major cardiovascular events was at a mean DBP of 82.6 mmHg, though patients with diabetes fared better with an achieved DBP target of <80mmHg vs <90mmHg, with a 51% reduction in cardiovascular events (https://pubmed.ncbi.nlm.nih.gov/9635947/ )

    -- the 2010 ACCORD trial did not find a significant benefit of lowering SBP to 119.3 mmHg in patients with diabetes (vs 133.5 mmHg),  (see dm tight bp control ACCORD nejm 2010 in dropbox, or doi 10.1056/NEJMoa1001286)

    -- the 2013 SP3 trial of patients with MRI-defined symptomatic lacunar infarcts found only a trend to benefit in patients achieving an SBP of 127 mmHg vs 138 mmHg (https://pubmed.ncbi.nlm.nih.gov/23726159/ )

    – the STEP study of 9614 patients aged 60-80 found that an achieved systolic blood pressure of 127.5 mmHg vs 135.3 mmHg was associated with a lower incidence of cardiovascular events (see htn elderly STEP study NEJM2021 in dropbox, or DOI: 10.1056/NEJMoa2111437)

-- so mixed findings in the past. i should be noted that these studies did not provide important information regarding the different medications used, and some meds are better than others.... (eg amlodipine has a particularly excellent 24 hour effectiveness and seems to decrease blood pressure variability very well)

 

--other issues that support a lower SBP target:

    -- in many non-Westernized societies, the normal blood pressure is in the 90-110 mmHg range, with no change with age (there have been many studies in the past, including  DOI: 10.1161/01.hyp.14.3.238)

    -- KDIGO (Kidney Disease: Improving Global Outcomes) in their 2021 guidelines promoted a target SBP of <120mmHg (https://www.kidney-international.org/action/showPdf?pii=S0085-2538%2820%2931269-2)

 

Limitations:

--the standard blood pressure goal in the US for patients with hypertension is <130/90 mmHg. The goal in this study was a systolic <140 mmHg in those on the higher goal target, achieving a systolic of 134.8 mmHg. although their target was <140mmHg instead of <130mmHg, it is likely, but not certain, that the benefit of the lower pressure of <120mmHg would still be more beneficial

-- there was a significant difference in decreases in eGFR in the intensive treatment group. However, there was no information about which BP meds were used. For example, ACEi and ARBs are actually renoprotective by lowering eGFR and decreasing efferent arteriolar resistance and glomerular capillary hydrostatic pressure, and there was a pretty large increase in the use of ACEi/ARBs in the intensive treatment group (79% vs 65%). So, unclear that the increased eGFR was actually bad…

-- though there were impressive results in this study, many patients were not included, including those with eGFR<45 (a very large group of patients with hypertension, especially in the elderly), limiting generalizability of their results

-- also there was a low rate of diuretic prescriptions, which might account for the rare abnormalities in electrolyte abnormalities.  see above for a general concern about HCTZ

-- this study had an open-label design, and as a non-blinded study might introduce bias and non-generalizability of the results

-- this study also was entirely of Chinese participants; given the many cultural differences from other countries (eg, diet, exercise, etc), the results may not be applicable elsewhere

-- all of the participants in the study were at least age 50yo at randomization. we often see younger patients with hypertension, limiting generalizability to them. there was no comment on the older age range, though we know that 25% were at least 70yo. were 80-100yo patients included? How did they do?

-- the definition of high risk varied from a male 60yo with a total cholesterol of 200 mg/dL to one with clinical cardiac disease.  there are large differences between these groups, especially when using a total cholesterol of 200 mg/dL as a cutpoint: the increased cardiovascular risk is much better defined by elevations in apolipoprotein B, total non-HDL cholesterol, or the total cholesterol/HDL ratio

-- as in many studies, other risk factors were not defined in a useful way: eg, "smoking" be defined as smoking >1 cigarette in past 12 months (we now have information that it takes at least 20-30 years of smoking cessation to really lower the cardiovascular risk; and passive smoking was not included ), "diabetes" as a binary variable (though people with A1c in the prediabetes range are at increased cardiovascular risk, those with terrible diabetic control are at much higher risk), ​again limiting applicability of the results to the patients we are seeing

-- another issue that might undercut the benefit of the intensive therapy was that for this 3.4 year trial it took 9 months to reach the steady state of SBP 119mmHg (this long time interval was related to the sudden outbreak of the Covid pandemic), but only 2 months in the standard care group to reach their SBP of 135 mmHg. since it appeared that the longer the individuals were at the 119 mmHg level in the intensive group the better, this 3.4 year study was 7 months shorter for this cohort

 

so,

-- pretty impressive results finding that more intensive blood pressure control down to 119 mmHg is clinically beneficial overall and for the subgroup analyses done (including patients with stroke and diabetes), with very few serious adverse events

    -- undoubtedly, the SBP of 135 mmHg achieved in the standard cohort in this study provides a lot of benefit over higher levels, it was clear that lowering SBP further to the 119 mmHg range added to this benefit

    -- given the short follow-up period of this study, and the increasing benefit over the 3.4 year follow-up that they did find, it is likely that continuing what is most often lifelong treatment will confer even more benefit from the lower target SBP

    -- it does seem reasonable to have a lower target for SBP in general, given the benefit without risk as found in this study. it is still unclear what the best target is. if SBP 110 mmHg better than 119mmHg?

-- the intention of the researchers is to continue to collect data, which should provide important information about the long-term clinical benefits of more intensive blood pressure control.

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would the article, please email me.

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list