New AHA hypertension guidelines

The American College of Cardiology, the American Heart Association, and 9 other societies just released their 2017 guidelines for the prevention, detection, evaluation, and management of high blood pressure in adults (go to https://doi.org/10.1161/HYP.0000000000000065 and download PDF). Below are the main points and changes over prior documents, focusing on primary care (i did not comment on in-hospital or urgent treatments). It is a bit long, but much shorter than the 283 page document:

Details:

-- hypertension is remarkably common in the US, and likely to become more common with the aging population: the Multiethnic Study of Atherosclerosis found that for 45-year-olds without hypertension, the 40-year risk of developing it was: 93% African-American, 92% Hispanic, 86% for white, and 84% for Chinese adults. The Framingham study found 90% likelihood of developing hypertension during one's lifetime if free of hypertension at age 55 to 65.
-- the vast majority of patients with hypertension have stage 1 hypertension: not surprisingly, lowering the cutpoint from 140/90 to 130/80 (see below) will increase prevalence of hypertension from 32% to 46% of the US population
-- proper methods are recommended for the accurate measurement of office blood pressure. They provide a checklist of 20 items they consider important (their table 8), but comment specifically on failure to allow for patient rest, talking with the patient during or immediately before the recording, and improper patient positioning (sitting or lying on an exam table), rapid cuff deflation, and reliance on blood pressures measured at a single occasion. see http://gmodestmedblogs.blogspot.com/2017/09/the-most-important-hypertension-blog.html​ for a study emphasizing the remarkably common inaccuracies in clinic blood pressure measurement.
-- but out-of-office blood pressure measurements are recommended to confirm the diagnosis of hypertension and for titration of BP lowering medication, in conjunction with telehealth counseling or clinical interventions. They note that home blood pressure monitoring (HBPM) is much easier and practical, though the data supporting its use are not as robust as for ambulatory blood pressure monitoring (ABPM). It is important that the patient not smoke/have caffeinated beverages/exercise in the 30 minutes prior to taking a reading, that they rest at least 5 minutes quietly, that they sit with their back straight and supported with their feet flat on the floor and legs uncrossed, that they take at least 2 readings 1 minute apart in the morning before taking medications and in the evening before supper.[I would add that it is important for the patient to bring in their blood pressure cuff, demonstrate its use, and simultaneously compare its value with one taken by the clinician or staff]
-- ABPM/HBPM:
    -- for adults with untreated SBP of 130-160 mmHg or DBP of 80- 100 mmHg, it is reasonable to check daytime ABPM or HBPM before diagnosing hypertension
    -- for those determined to have whitecoat hypertension, it is reasonable to do ABPM or HBPM periodically to monitor for transition to sustained hypertension [though the data support treating hypertension based on ABPM or HBPM, whether whitecoat or not]
    -- it may be useful to confirm hypertension detected by HBPM with ABPM
    -- in untreated adults with office BP of 120-129/75-79 mmHg, it is reasonable to assess HBPM or ABPM to detect masked hypertension [masked hypertension when the blood pressure is actually lower in the office than in the outside world. I suggest either HBPM, or the patient going to the pharmacy a few times to assess their blood pressure, as a means to make this diagnosis]
    -- may be reasonable to check HBPM or ABPM in patients on multiple drug therapies for hypertension and within 10 mmHg above goal, given potential whitecoat effects (see below on refractory hypertension)
    -- may be reasonable to do ABPM for adults treated for hypertension with elevated HBPM readings, prior to intensifying medical therapy
-- environmental risk factors:
    -- overweight/obesity is the big one: these account for between 40 and 80% of the attributable risk of hypertension, and relationship is more profound for waist-to-hip ratio than for BMI. This is bilateral: increases and decreases of weight are associated with increases and decreases the BP
    -- sodium intake: though there has been recent argument in the literature that this is overstated, it is clear that salt sensitivity is related to both blood pressure as well as CVD and all-cause mortality. Salt sensitivity is more common in black adults, older adults, and those with CKD, DM, metabolic syndrome.
    -- high potassium diets, physical fitness, and alcohol consumption play a role (estimates are that alcohol accounts for about 10% of the hypertensive population in the US
    -- nonpharmacologic interventions suggested (no real changes, but should be emphasized): weight loss (about 1 mmHg for each kg lost), heart-healthy diet such as DASH (about 11 mmHg decrease), sodium restriction (about 5 mmHg decrease), potassium supplementation (about 4 mmHg decrease), increased physical activity (about 5 mmHg decrease), decreased alcohol consumption to no more than 2 drinks per day for men and one for women (about 4 mmHg decrease)
-- workup of hypertension, minor changes: fasting blood glucose, complete blood count, lipid profile, serum creatinine/eGFR, serum sodium/potassium/calcium, TSH, urinalysis, EKG. Optional testing includes echocardiogram, uric acid, urinary albumin-to-creatinine ratio. [I strongly support the uric acid and the urinary albumin-to-creatinine ratio, as this might alter therapy: eg, preferentially prescribing losartan in those with high uric acid (as well as strongly advising a diet low in high-fructose corn syrup), and ACE-I/ARB in those with albuminuria. Though I do think an echocardiogram would be really useful, it is quite expensive and inaccessible, so hard to adopt uniformly]
-- hypertension definition: stage I if blood pressure 130-139/80-89 mmHg; stage II if blood pressure greater than 140/90 millimeters of mercury
-- for patients at lower risk of ASCVD (10-year ASCVD risk less than 10%), they should be managed with nonpharmacologic therapy as above, with repeat blood pressure measurement within 3 to 6 months
--Meds:
    -- the strongest recommendation was to start meds for secondary prevention in those who have already had clinical CVD events, when the average SBP was 130 mmHg or higher or the average DBP was 80 mmHg or higher. For primary prevention, the same but for those with an estimated 10-year ASCVD risk of 10% or higher (e.g. using http://tools.acc.org/ASCVD-Risk-Estimator-Plus/#!/calculate/estimate/ ). A less strong recommendation was to use meds for primary prevention in those of lower ASCVD risk if the SBP were 140 mmHg higher or DBP 90 mmHg or higher.
    --do NOT combine an ACE inhibitor, ARB, and or renin inhibitor. Can be harmful
    -- BP target should be less than 130/80 mmHg
    -- 1^st line agents include thiazide diuretics, CCB's, and ACE inhibitor/ARB's. They do suggest chlorthalidone as the preferred diuretic (see http://gmodestmedblogs.blogspot.com/2016/04/chlorthalidone-is-better-than-hctz-for.html  ). They acknowledge that hydrochlorothiazide should be prescribed at least in the 25-50 milligrams dosages [as noted in this blog, HCTZ 12.5mg works well at lowering clinic-based blood pressure, but is terrible for the 24 hour ABPM, about half as good as other antihypertensives, and argues that this dose should not be used].
    -- for those with stage II hypertension and an average of more than 20/10 mmHg above target, they recommend starting with two 1^st line agents [I have seen enough patients with exaggerated hypotensive response to 2 agents that I really do think it makes sense to start with one agent, see the patient soon, and add another agent later. For unclear reasons, this use of 2 agents was promoted before, then abandoned in JNC8 (see http://gmodestmedblogs.blogspot.com/2013/12/jnc-8.html), but now is rearing its head again]. All medications should be prescribed with lifestyle counseling
    -- patient should be followed monthly until goal blood pressure is reached. And they recommend using HBPM, team-based care, and telehealth strategies as part of the approach
    -- they also  suggest choosing antihypertensive meds based on comorbidities (e.g. angina, heart failure, CKD, non-acute stroke, peripheral arterial disease, diabetes) though the target should still be less than 130/80 mmHg. Patients with atrial fibrillation might benefit more from an ARB to help prevent recurrence of AF [though I would be hesitant to stop anticoagulation…]
    -- pregnancy: preferred meds are methyldopa, nifedipine, and/or labetalol. ACE inhibitors, ARBs, direct renin inhibitors should not be prescribed women who become pregnant
    -- elderly: same goal of 130/80 mmHg  for noninstitutionalized, ambulatory community dwelling adults. Clinical judgment should be used in those with a high burden of comorbidity and limited life expectancy
    -- to improve medication adherence, best to have once-daily medications, and use combination pills whenever possible
-- resistant hypertension: with this new lower goal of 130/80, it is likely that the prevalence of resistant hypertension will increase 4% above the current 13% of the adult hypertensive population (resistant hypertension being defined as taking 3 antihypertensive medications at full dose, with a diuretic as one of the components). They note that the most common 3-drug combination is a CCB, RAS inhibitor, and chlorthalidone. In many studies, spironolactone as a 4th drug often leads to pretty dramatic blood pressure reductions (eg see http://gmodestmedblogs.blogspot.com/2015/09/spironolactone-in-drug-resistant.html  )

Commentary:
-- The big change in this document is the lowering of the definition of hypertension to >130/80, though in those with 10-year ASCVD risk <10%, not start meds unless above 140/90. And the goal of hypertension therapy is 130/80.
-- There are several positive aspects to this document:
    ​-- lowering the blood pressure cutpoint for defining hypertension seems reasonable in untreated patients, given several epidemiologic studies including the Framingham study showing that the lower the blood pressure the lower the risk of cardiovascular events, down to 110/70.
    -- Though the diagnosis of hypertension will be significantly increased by this lower cutpoint to almost half of the US population, the current document gives strong incentive for both clinicians and patients to focus on healthy lifestyle changes.
    -- There is a big push, more explicitly stated here, to use either HBPM or ABPM to guide therapy. As mentioned in several prior blogs, the data on this are pretty consistent, the UK adopted this approach in 2011 in the NICE guidelines, followed by the USPSTF in 2015 (see  http://gmodestmedblogs.blogspot.com/2015/01/uspstf-recs-on-ambulatory-blood.html  ).
-- The lower BP goal is largely based on a meta-analysis finding that the achieved blood pressure of 133/76 mmHg led to a significant reduction in CVD events, MI, and stroke. This target is somewhat higher than that in the SPRINT trial. But a few comments about this highly touted SPRINT trial:
    -- there were significant structural issues in how blood pressure was monitored in this trial, which are largely non-reproducible in any clinical setting, and probably understated the actual clinic-based blood pressure on the order of 5-10 mmHg (for some reason, the methodology for blood pressure determination was not delineated in the published SPRINT trial and only came out later. see details in http://gmodestmedblogs.blogspot.com/2017/02/blood-pressure-guidelines-for-older.html  )
    -- and on further analysis of the SPRINT trial, they found that ABPM did not correlate well with their clinic blood pressure recordings (see http://gmodestmedblogs.blogspot.com/2017/06/monitor-bp-effects-by-abpm-sprint-trial.html  )
    -- even with a less onerous approach to clinic blood pressure measurement, it is still hard to reproduce what was done in the clinical hyptertension studies (see aforementioned blog http://gmodestmedblogs.blogspot.com/2017/09/the-most-important-hypertension-blog.html​  )
    ​--and, even then, it seems that clinical cardiac events really track with APBM and HPBM dramatically more than with even the best clinic-based recordings
-- although I do suggest HBPM regularly with patients, and there are some data suggesting that HBPM is almost as good as ABPM, it should be remembered that these out-of-office BP studies had a pretty aggressive protocol for HBPM, including discarding the 1^st reading as well as instructing patients to relax for 5 minutes before taking their blood pressure, not take stimulants for 30 minutes, etc. In this regard, it probably would be useful to do ABPM more regularly if it were inexpensive and accessible. And there are ABPM monitors available which are not too expensive and might give us for more accurate and actionable readings.
-- there is a section on screening for secondary causes of hypertension. I did not comment above given that there's not much change, but I would add that I do not routinely screen for primary hyperaldosteronism, in contrast to their suggestions. In larger studies with regular screening, hyperaldo incidence may be in the 5 to 10% range, and as high as 40% in those with resistant hypertension. The issue for me is multifold: most hypertensive patients with hyperaldo do not have hypokalemia, most who have hypokalemia are easily treated, and I usually just add on spironolactone if the patient has hypokalemia or has resistant hypertension, with excellent effect. Hyperaldosteronism is essentially never caused by cancer, and there are studies suggesting that even CT or MRI-documented unilateral adenomas, on surgical inspection are not infrequently bilateral adrenal hyperplasia, for which surgical therapy is not appropriate. Studies suggest that in patients with a family history of hypertension, surgical correction of hyperaldosteronism leads to approximately only 1 fewer blood-pressure medication. And, my bias is to avoid surgery whenever possible, so long as the hypertension can be medically treated reasonably easily, and there is good medication adherence in those with profound hypokalemia.
--for black adults with hypertension requiring therapy, they suggest starting with either a thiazide-type diuretic or CCB. Notably, black patients to have a higher risk of ACE inhibitor-induced cough and angioedema.  Recommendations from NICE were a bit different, suggesting using a CCB for all people of African-Caribbean descent and all white people greater than 55 years old (eg see http://gmodestmedblogs.blogspot.com/2013/12/hypertension-guidelines-european.html  ). My own "go-to" drug overall is amlodipine, which has several studies showing that it is associated with the least blood pressure variability (also a marker for cardiovascular events) as well as the longest duration of action (ie, it works the full 24 hours).  And the higher incidence of strokes in several studies of ACE inhibitors may well be related to their somewhat shorter duration coupled with the higher incidence of strokes in the early morning hours, though the caveat is that many other drugs have not been tested that i have seen (eg nifedipine, other CCBs, ARBs)
--although in several prior posts i have commented on the concern for cognitive decline in elderly patients with lower blood pressure (and there are some data suggesting that those with baseline cognitive impairment can decline further (see http://gmodestmedblogs.blogspot.com/2015/04/too-low-blood-pressure-and-cognitive.html ), most of the recent studies, including SPRINT (mean age of only 80, which is higher than most other studies...) did not find significant cognitive decline. And some earlier studies have shown that lowering the blood pressure with some agents (eg nitrendipine, perindopril) actually improves cognitive function
--for resistant hypertension, there was an impressive article from brazil finding that 40% of those with office-based refractory hypertension were actually well-controlled by ABPM (see commentary in http://gmodestmedblogs.blogspot.com/2017/06/monitor-bp-effects-by-abpm-sprint-trial.html).  Again reinforcing the utility of ABPM, which pretty consistently predicts clinical cardiovasc events better than clinic-based blood pressure measurements.

So, baseline changes are the goal of 130/80, emphasizing aggressive use of lifestyle changes, and reinforcing the utility of out-of-office BP measurements in the diagnosis and treatment of hypertension