You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.


Primary Care Corner with Geoffrey Modest MD: Monitor BP effects by ABPM?? SPRINT trial

14 Jun, 17 | by gmodest

by Dr Geoffrey Modest

The SPRINT hypertension study assessed different blood pressure targets on clinical outcomes, finding the lower the pressure, the better. A subsequent ambulatory blood pressure substudy of SPRINT looked at the relationship between the achieved ambulatory versus clinic-based blood pressures (see DOI: 10.1161/HYPERTENSIONAHA.116.08076.)



— the SPRINT study randomized patients to aggressive vs less aggressive blood pressure control, achieving a systolic of 121 mmHg versus 136 mmHg, respectively. The primary clinical outcome (MI, acute coronary syndromes, stroke, heart failure, or death from cardiovascular causes) was 25% less frequent in those having tighter control. (see here for details)

— in the current ancillary study on ambulatory blood pressure monitoring (ABPM), they performed ABPM at the 27-month study visit in a subgroup of 897 SPRINT participants

— for this ancillary study, mean age 71, 29% female, 66% white/28% black/3% Hispanic, BMI 30, 46% never smoker, 38% nondrinker/26% moderate drinker/10% heavy drinker, 21% CVD at baseline, eGFR 67, LDL 108/HDL 53/triglycerides 100, number of antihypertensive meds at the 27 month visit = 2.9 with 78% on ACE-I/ARB, 60% on calcium-blocker, 76% diuretics, 40% beta-blocker

— at the 27 month clinic visit:

— clinic-based systolic 120 mmHg

— nighttime ABPM systolic 116 mmHg

— daytime ABPM systolic 127 mmHg

— 24-hour ABPM systolic 123 mmHg



— for those on intensive therapy:

— decreased clinic-based systolic BP by 16.0 mmHg

— decreased nighttime systolic BP by 9.6 mmHg

— decreased daytime systolic BP by 12.3 mmHg

— decreased 24 hour systolic BP by 11.2 mmHg

–there was poor agreement in participants between clinic-based systolic BP and daytime systolic ABPM



— the role of ABPM continues to evolve, with studies documenting its strong association with cardiovascular and renal clinical events. Data for the last decade or so have pretty convincingly shown that ambulatory blood pressure is more predictive of cardiovascular events than clinic-based blood pressure, leading ultimately to the USPSTF promoting ABPM in their most recent guidelines (see , and my blog on it )

— One issue with the SPRINT study has been the rather eclectic way they measured the clinic-based blood pressure, reducing its generalizability to our clinical practice (see here which comments on their approach). One advantage of ABPM is that it is an equalizer, where the target blood pressure is pretty reproducible from one ambulatory monitor to another, and does not depend on measuring blood pressure in a clinic setting based on a very structured and likely unreproducible methodology (even without the strict regimen in SPRINT, how many of us check blood pressures after the patient has been resting quietly in a dark room for 5 minutes??)

— Unfortunately, there are relatively limited data on ambulatory blood pressure monitoring as a means to follow up patients with hypertension on treatment. There are some studies which do suggest that it is useful: a Brazilian study (see Salles GF. AchInternMed.2008;168:2340), which looked prospectively at 556 patients with resistant hypertension for 4.8 years, found that ambulatory blood pressure monitoring predicted clinical endpoints, and that clinic-based blood pressure did not. And in fact 40% of those with clinic-based “resistant hypertension” did not even have hypertension on ABPM (adding to the studies finding that even “appropriately” documented clinic-based blood pressure is really a poor predictor, and sometimes pretty irrelevant, especially as compared to ABPM: see here for an array of blogs on this, including the quite impressively documented and  prescient recommendations from NICE in the UK in their 2011 guidelines


So, this study adds further to the imperative to use ABPM, especially in those patients with blood pressures near the goal (ie, the patient with 200/110 mmHg can just be treated….). And, though much less well documented (but easier for many) to use home-based measurements. This study helps reinforce the utility of ABPM in those being treated (again, I would assume, especially so if the achieved blood pressure is pretty close to goal: eg the Brazilian study above on refractory hypertension finding no benefit for clinic-based blood pressures did not have granular data, but I would assume that those way above goal continued to need adjustment of their meds independent of the ABPM reading, and those near goal were more likely to be in-range with ABPM even though the clinic-based pressures were high.


And, again, this study does raise a pretty basic concern: how often do our accepted clinical approaches (in this case clinic-based blood pressures) not really reflect the reality of actual clinical outcomes, yet are passed down over time and accepted? Which all reinforces the importance of constantly challenging our existing models. Another example which I have commented on recently is with A1c as a surrogate marker for diabetes and macrovascular complications, suggesting that the issue may be less with the A1c achieved than with the medications we choose to put patients on, and it is probably most important to choose ones that have documented cardioprotection (see here )

Primary Care Corner with Geoffrey Modest MD: Low-dose hypertension therapy

7 Jun, 17 | by gmodest

by Dr Geoffrey Modest

A recent meta-analysis assessed 42 trials comparing low-dose (one-quarter dose) antihypertensive therapy versus placebo, finding some efficacy and essentially no adverse events at this low dose (See DOI: 10.1161/HYPERTENSIONAHA.117.09202).


— 42 trials were identified involving 20,284 participants

— 37 comparisons evaluated quarter-dose therapy versus placebo

— 7 comparisons were of dual quarter-dose therapy versus placebo

— 1 comparison of quadruple quarter-dose therapy versus placebo

— standard full-dose therapy for some of the commonly prescribed medications were:

— atenolol 50 mg

— metoprolol 100 mg

— carvedilol 25 mg

— amlodipine 5 mg

— verapamil 240 mg

— lisinopril 20 mg

— enalapril 20 mg

— valsartan 80 mg

— candesartan 8 mg

— hydrochlorothiazide 25 mg



— quarter-dose therapy versus placebo led to a 4.7/2.4 mmHg improvement (p<0.001)

— dual quarter-dose therapy was associated with a 6.7/4.4 mmHg improvement versus placebo (p<0.001), as effective as standard-dose monotherapy

— quadruple quarter-dose therapy versus placebo (only one study) found a blood pressure reduction of 22.4/13.1 mmHg (p<0.001)

— adverse events in both the single and dual quarter-dose therapies were not significantly different from placebo and had significantly fewer adverse events compared to standard-dose monotherapy



— the studies used in this meta-analysis were quite old, the average being published to 17 years ago, and 85% had eligibility criteria based on diastolic blood pressure alone.

— It is important to realize that this study only looked at office-based blood pressure targets, not clinical outcomes. For example, beta-blockers are now out of favor because of a few meta-analyses showing little benefit in terms of clinical outcomes. And some commonly used meds are not included (eg losartan).

— Some of the older JNC recommendations, perhaps based on some of these studies, did raise the issue of combining low-dose medications versus increasing a single medication, given the synergy in blood pressure lowering of the combination and the significant decrease in adverse events

— there was also an interesting meta-analysis from 2009 (see Wald DS. Am J Med 2009; 122: 290), looking at 42 trials with 10,968 participants, finding that doubling the dose of thiazides, beta blockers, ACE inhibitors, calcium channel blockers, and the summation of all of these classes of antihypertensives led to only 22% of the equivalent incremental effect of adding a 2nd agent (though calcium channel blockers did a little better than others, especially ACE inhibitors). This study really reinforces my own anecdotal experience, that doubling the dose of lisinopril, amlodipine, etc seems to have much less blood pressure lowering effect than adding a 2nd agent (e.g going from lisinopril 10 mg to 20 mg has pretty consistently a lower incremental value than adding low-dose hydrochlorothiazide to the lisinopril 10 mg, which has the added benefit of being available as a single combination pill). Sort of similar to statins where the most significant cholesterol-lowering is at the lowest dose (I have a patient on 1/3 of atorvastatin 10mg, ie 3.3 mg, with impressive lipid lowering). There was also another study finding that half-dose therapy was about 80% as effective as compared to full standard-dose therapy.

In a prior blog on the increased efficacy of chlorthalidone over hydrochlorothiazide, I lamented the fact that chlorthalidone was only available as 25 mg in the United States, that that dose was associated with significant hypokalemia, and that the pill was too small to cut easily. One of my readers commented that he actually took one quarter of the pill, did not find it difficult to cut the pill, and his blood pressure was well-controlled with that.


So, I do realize these meta-analyses are based on old studies, but it does seem to have validity, and is reinforced by my own clinical experience. Based on the study comparing doubling the dose versus combination medications, my practice has been to double the dose of the medication only if there is a compelling other reason (such as trying to decrease the level of proteinuria by maximizing ACE inhibitor dose) or if the medication is well-tolerated and the patient is pretty close to the target blood pressure.

Primary Care Corner with Geoffrey Modest MD: creatinine increases after ACE/ARB may not be so good

15 Mar, 17 | by EBM

By Dr. Geoffrey Modest

A recent article in the BMJ challenged the long-held belief that ACE inhibitor/ARB related increases in creatinine were actually renoprotective (see Schmidt M. BMJ 2017;356:j791).


  • Observational study of 122,363 patients starting treatment with ACE inhibitor (ACE-I) or ARB from 1997 to 2014
  • They assessed the rates of end-stage renal disease, myocardial infarction, heart failure, and all-cause death among patients whose creatinine increased 30% or more after starting treatment, and also assessed the effect of each 10% increase in creatinine above the patient’s baseline.


  • 2078 patients (1.7%) had a creatinine increase of 30% or more.
  • Comparing those with a creatinine increase of > 30%, vs those with < 30%:
    • 56% female vs 46%
    • Median age 68 vs 63
    • Myocardial infarction in 10.5 vs 4.5%
    • Heart failure in 19 vs 4.8%
    • Arrhythmia in 17.2 vs 6.8%
    • Peripheral arterial disease in 6 ​vs 2.5%
    • Stage 3b CKD in 6.9 vs 3.7%/stage 4 in 2.0 vs6%
    • Beta blockers in 23.7 vs 17%
    • Loop diuretics in 28.6 vs 7.2%
    • Potassium sparing diuretics in 8.8 vs 2.0%
    • NSAIDs in 34.0 vs 23.5%
    • Underweight in 2.3 vs 0.9%, healthy weight in 26.9 vs8%, overweight in 34.5 vs38.4%, and obesity 29.0 vs  33.4%
    • So, basically those with greater creatinine increases had more baseline characteristics associated with increased morbidity/mortality
  • Creatinine increases of 30% or more were associated with (adjusted for age, sex, calendar period, socioeconomic status, lifestyle factors, chronic kidney disease, diabetes, cardiovascular morbidities, and use of other antihypertensive drugs and NSAIDs):
    • 43 times the rate of end-stage renal disease, incidence rate 3.43 (2.40-4.91)
    • 46 times the rate of myocardial infarction, IR 1.46 (1.16-1.84)
    • 37 times the rate of heart failure, IR 1.37 (1.14-1.65)
    • 84 times the rate of all-cause death, IR 1.84 (1.65-2.05)
  • There was a greater increase for all outcomes as creatinine went from an increase of <10%, to 10-19%, to 20-29%, to 30-39%, and to > 40%
  • These results were consistent across calendar periods, subgroups of patients, and among those continuing to use ACE inhibitor/ARB’s
  • There were much more dramatic increases in the rate of renal failure during the 1st year after starting ACE-I/ARBs (12.2-fold increase) versus in the 2nd year (3.7-fold) versus 2nd to 5th year (1.7- fold), but then increase to 2.5-fold from 5 to 10 years. However the numbers were small and the trend was nonsignificant. However, there were similar trends for heart failure and mortality which were significant. Heart failure was initially 1.9-fold increase in the 1st year but then settled in at 1.5-fold increase.


  • One major concern is that only about 10% of patients receive the recommended monitoring of serum creatinine soon after starting ACE-I/ARBs and only 20% of those with an increase of >30% discontinue the drugs as is recommended.
  • It has been widely held that larger increases of creatinine after taking these medications (up to 30%) were in fact renoprotective, supported theoretically/mechanistically that by decreasing intra-glomerular pressures, we were sparing the fragile glomeruli frombarotrauma. These data were not terribly rigorous. For example, there was a small study (Apperloo AJ. Kidney Intl 1997; 51(3): 793), which did find that in 40 nondiabetic patients with impaired renal function prior to therapy, those with a greater GFR decline after ACE-I had more stable renal function, and this decline was completely reversible after stopping ACE-I therapy at 4 years.) And this study has been cited in subsequent reviews as clear evidence that the higher the creatinine increase (up to 30%) the better…
  • The benefits of the study are its huge size, its real-world outcomes data and the fact that it represented the general UK population in terms of age/sex/ethnicity, the fact that they only looked at patients who had at least one year of being continuously in the registry before they were started on an ACE-I/ARB, and that they had long-term follow-up until the 1st diagnosis of end-stage renal disease, myocardial infarction, heart failure, and all-cause mortality.
  • However, the negatives of the study are that the patients who had more significant creatinine increases were clearly sicker and had more inherent likelihood of getting these clinical endpoints. Specifically, these patients were older, had more underlying chronic kidney and heart disease, and had more drugs that were potentially nephrotoxic. In particular the use of potassium sparing diuretics suggests the possibility that these patients had more severe hypertension or heart failure requiring these drugs (and the study did not stratify the degree of these conditions at baseline). The increased use of NSAIDs might signal that these patients had more pain, were less ambulatory, perhaps more overweight (though the BMI of those with more a bump in creatinine was actually lower, the specific individuals who went on to renal failure may well have been those with a higher BMI and on NSAIDs, but these data not available), and were less able to have a healthy lifestyle, which has repeatedly been associated with increased morbidity/mortality. And, though they did mathematically model to compensate for the array of potential adverse biases, there was such a divergence in the baseline characteristics of the 2 groups (>30% versus <30% creatinine increase) that I do not trust this mathematical manipulation to compensate for the real potential biases between the groups. (In addition, they did not comment on the underlying clinical conditions of the patients comparing those with <10% increase in creatinine vs those with >30%, but only for those <30% vs >30%)
  • It was also notable that pretty much all of these outcomes were much more dramatic in the 1st year after starting ACE-I/ARBs, suggesting that we should be doing increased surveillance particularly in that 1st
  • There are some perhaps relevant prior studies which suggest that increased creatinine is associated with cardiovascular disease, found in patients with mild to moderate renal dysfunction. However, these were patients with intrinsic renal disease as opposed to medication-induced increases of creatinine. So, not sure this is directly applicable to ACE-I/ARB-induced creatinine increases, but the above results are consistent with this.
  • We also did not know the levels of proteinuria of these patients, so there could be an important unaccounted for bias here. The studies suggesting the renoprotective effect of ACE-I/ARBs in diabetics found renal protection in those with proteinuria, even at low levels of albuminuria.

There have been other studies showing that patients with very significant proteinuria, especially those with greater than 1 g of albumin per day, do have renal protection by ACE inhibitors (see GISEN group. Lancet 1997; 349: 1857, which found that in 352 nondiabetic patients with proteinuria of 3 g of more than 24 hours, ramipril led to significant decreases in proteinuria as well as the rate of GFR decline, and a subsequent study of 186 patients with 1 to 3 g of proteinuria also had renal protection, but to a lesser degree, see Ruggenenti P. Lancet 1999; 354: 359.)

So, this article does give some pause. Perhaps our model of renoprotection is not so accurate. It is notable that the new JNC 8 guidelines do not recommend using ACE-I/ARBs as the primary treatment for hypertension in diabetic patients. Also the new American Diabetes Association guidelines (blog to come out soon) comment that those with diabetes should be treated with ACE-I, ARBs, thiazides, or dihydropyridine calcium channel blockers (no preference, except they recommend ACE-I/ARBs in those with albumin to creatinine ratio is greater than 300 mg/g, level A recommendation, or in those with 30 to 300 mg/g, level B recommendation). So, what is the bottom line here? I am really not sure, pending other studies (and the best being an RCT). But this study does bring up the thinness of the prior assertions that those with ACE-I/ARB induced creatinine increases do better, and also reinforces the importance of checking creatinine (and lytes) after starting these meds, and stopping them if >30% increase. Otherwise, I am hesitant to change current practice.

Primary Care Corner with Geoffrey Modest MD: Blood Pressure Guidelines for Older Adults

27 Feb, 17 | by EBM

By Dr. Geoffrey Modest

The American College of Physicians and the American Academy of Family Physicians just published guidelines on the pharmacologic treatment of hypertension in adults over 60 yo, with both a systematic review and meta-analysis (see doi:10.7326/M16-1785), and a clinical practice guideline (see )


  • They analyzed 46 publications representing 21 randomized controlled trials and 3 cohort studies
  • 9 trials show that intensive blood pressure treatment substantially improved outcomes in patients with moderate to severe hypertension, with SBP >160 mmHg. The data on lower systolic blood pressures also showed benefit but the results were less consistent.
  • Overall studies of patients achieving SBP <140 mmHg were similar to those that achieved 140, although the reduction in stroke risk was more consistent in the studies where patients achieved the higher SBP
  • In 6 trials comparing different treatment targets with 41,491 patients, treatment targets of SBP <140 mmHg or diastolic blood pressure of <85 mmHg were associated with only marginal benefit, with wide confidence intervals:
    • 14% nonsignificant reduction all-cause mortality, RR 0.86 (0.69-1.06)
    • 21% reduction in stroke, RR 0.79 (0.59-0.99)
    • 18% marginally significant reduction in cardiac events, RR 0.82 (0.64-1.00)
    • Because of their size and the event rates found, these analyses were dominated by the SPRINT and ACCORD trials. SPRINT (which excluded diabetics as well as those with SBP >180, prior stroke, urinary protein excretion >1 g per day or symptomatic heart failure/EF <35%) found marked reductions in mortality in cardiac events, though the ACCORD trial (which included only diabetics, though did achieve an SBP of 119 mmHg, similar to SPRINT) did not find any reduction in mortality or major cardiovascular events with intensive treatment [though other trials found benefit of hypertension treatment to be at least as strong in diabetics]. Also, the SPRINT trial stopped earlier than projected because of benefit, which, as mentioned in my blog on it noted below, will tend to exaggerate benefits and perhaps decrease finding risks.
  • Overall, tighter control “may prevent on average, roughly 10 to 20 events for every 1000 high-risk patients treated over 5 years across a population”
  • Harms of more intensive therapy: in general the evidence was relatively low to moderate strength, but did not find clear evidence of more renal, cognitive impairment, deterioration of quality-of-life/functional status, or increase in fractures or falls, though there was low-quality evidence for increase in syncope.


  • Initiate treatments in adults over 60 years old who have systolic blood pressure persistently at or above 150 mmHg, to reduce the risk for mortality, stroke, and cardiac events (strong recommendation, high quality evidence).
  • Consider initiating or intensifying pharmacologic therapy in patients over 60 years old with a history of stroke or TIA to achieve a targeted systolic pressure of less than 140 mmHg to reduce the risk of recurrent stroke. (Weak recommendation, moderate quality evidence)
  • Consider initiating or intensifying pharmacologic treatment in some adults greater than 60 years old at high cardiovascular risk to achieve a target systolic pressure of less than 140 mmHg to reduce the risk of stroke and cardiac events. (Weak recommendation, low quality evidence)
  • And for all of these recommendations, the risks and benefits should be periodically discussed with the patient.


  • There are really no studies that include the real elderly. The SPRINT elderly subgroup (those patients over 75 at enrollment), still had a mean age of 80, with SD of only 4 years, so really does not inform my practice with lots of people in the 85-100 age range. one might glean from the above trials that the lower blood pressure may well be better, since there was no evidence that age mattered in the groups analyzed (again, not including the very old), subgroup analyses from SPRINT as well as the HYVET trials found that frailty did not matter, and there was more absolute benefit in those with higher cardiovascular risk (and age plays into that). But, at least my practice in the elderly and especially in the very old is to check orthostatics regularly (looking for both initial and standard orthostatic hypotension: See details), try to get home-based BP measurements (and preferentially use these to guide therapy, as long as I have confirmed that the patient measures blood pressure accurately and the cuff is accurate), and assess cognitive function more aggressively (see which is an Italian prospective study in patients mean age of 79 with some baseline cognitive impairment, finding that those in the lowest BP group (SBP<128 mmHg) had more cognitive decline than those with higher pressures)​. So, my guess (without data) is that the benefits will persist in the very old, though I suspect the harms will be greater (patients more frail, more comorbidities, and more sensitive to meds)
  • My major concern with these articles on tighter blood pressure control in general is that there is a tendency in clinical practice to attempt to achieve the goal blood pressure they achieved in the study. However, this brings up a few issues:
    • In general the studies have very specific ways that they measured the blood pressure. The general real-world approach, at least in my experience, is to have a medical assistant bring the patient into the room and measure the blood pressure/record it in the electronic medical record. I have consistently been measuring manual blood pressures myself for the past many years, typically with the patient sitting on the exam table and resting a few minutes while I write my notes in the other room, and often find striking differences from the recorded blood pressure, not uncommonly 30 to 40 mmHg difference. Although most often my recording is much less than that of the medical assistance, at times it is much more (the 118/68, which really is 190/110!!!). So in general I am concerned about relying on automated blood pressure recordings (which in general is less reliable in people with atrial fibrillation and arrhythmias, as well), though my main concern is that the patients, perhaps somewhat deconditioned, walk into the room and sit down without resting and have largely unreliable readings.
    • For example, in the SPRINT study, which did achieve lower blood pressure in the tight control group than often found in other trials (123/62, in the elderly subgroup), they measured the blood pressure as follows: the staff person would tell a patient that they needed to rest for 5 minutes before taking the blood pressure, would leave the room completely, would return but not speak a word with the patient and immediately take the blood pressure. Argument has been raised in the literature that the blood pressure measured in randomized controlled trials is typically 5 to 10 mmHg lower than the clinic-based blood pressure (i.e. a randomized trial with an achieved systolic blood pressure of 123, as above, may be equivalent to a clinic-based blood pressure of 130 or so). For details of the SPRINT trial, see, which reviews the results of the overall trial, as well as which looked at the predesignated subgroup of those greater than 75 years old).
    • There are also significant questions as to the general reliability of office-based blood pressure, both because of whitecoat hypertension as well as masked hypertension (see , as well as the frequent observation that ambulatory blood pressure monitoring is much more predictive of clinical events, leading to the USPSTF and other international groups suggesting this is the preferred mechanism to diagnose hypertension (see
  • So, my real concern is that we may be basing important clinical decisions based on inaccurate data, and that we may be significantly over-treating (predominately) or under-treating hypertension, with their attendant potential adverse outcomes​

By the way, there was a review of intensive lowering of blood pressure in the elderly (defined as >65 yo), essentially simultaneous with the above, in the Journal of the American College of Cardiology, which identified only 4 studies (all included above) with 10,857 patients that met their criteria, finding that intensive blood pressure control with SBP <140 lead to a significant decrease in major cardiovascular events, including cardiovascular mortality and heart failure, but no difference in stroke or MI (see DOI: 10.1016/j.jacc.2016.10.077​). This exemplifies one of the points I made in my blog , that systematic reviews and meta-analyses may well come to different conclusions based on their own inclusion and exclusion criteria, and that we in the trenches (who are responsible for reading and considering implementing important changes in clinical practice) really need to assess how those authors configured their analyses and the relevance of their conclusions to our clinical practice. Not a simple feat.

Primary Care Corner with Geoffrey Modest MD: Masked Hypertension

12 Dec, 16 | by EBM

By Dr. Geoffrey Modest

A recent study compared clinic blood pressure (CBP) measurements and ambulatory blood pressure monitoring (ABP), finding much more masked hypertension than white-coat hypertension (see White-coat hypertension is when the CBP is higher than the ABP; masked hypertension is the opposite.


  • 888 healthy, employed, middle-aged individuals not on antihypertensive medications, found in a workplace screening program to have a blood pressure of <160/105 mmHg, then had 24 hour ABP.
  • Mean age 45, 89% female, 7.4% black/12% Hispanic
  • They compared the awake ABP (aABP), the CBP, and the difference. CBP was an average of nine readings over three visits after being seated a minimum of five minutes, and the participants had not smoked, eaten or had caffeinated beverages in the prior 30 minutes. Two other blood pressures were recorded 1 to 2 minutes afterwards. Those with CBP >140/90 were defined as having clinic-based hypertension, those with aABP >135/85 were defined as hypertensive. Those with elevated CBP but nonelevated aABP were defined as whitecoat hypertension. Those with nonelevated CBP but elevated aABP were classified as having masked hypertension.


  • Average systolic/diastolic aABP was 123/77 mmHg
  • Average CBP was 116/75 mmHg => average CBP was 7/2 mmHg lower than aABP
  • 3% were hypertensive by CBP; 19.2% were hypertensive by aABP; 15.7% with nonelevated CBP had masked hypertension. specifically,
    • For those with clinic blood pressure higher than ambulatory (white-coat), found overall in 17.8% by systolic pressure and 35.8% by diastolic:
      • Difference of > 5mmHg: 6.9% of subjects by systolic, 14.2% by diastolic
      • Difference of >10 mmHg: 2.5% systolic, 4.2% diastolic
      • Difference of >15 mmHg: 1.1% systolic, 0.9% diastolic
    • For those with ambulatory blood pressure higher than clinic blood pressure (masked), 82.2% of systolic and 64.2% diastolic
      • Difference of >5 mmHg: 63.7% systolic, 32.4% diastolic
      • Difference of >10 mmHg: 34.8% systolic, 9.2% diastolic
      • Difference of >15 mmHg: 14.4% systolic, 1.7% diastolic
    • This difference was most pronounced in young adults and those with normal BMI, decreasing at older ages and higher BMIs but did not disappear
    • No difference between men and women, black patients vs nonblack, Hispanic vs non-Hispanic, cigarette smokers vs past smokers vs nonsmokers


  • I had seen a few studies on masked hypertension with similar findings, but I must admit I assumed there was lots of hyperbole/biases to their conclusions, that white coat hypertension was undoubtedly much more common than masked hypertension. But — just goes to show you: for several patients their daily lives are even more stressful than the calm and relaxing clinician’s office….
  • This is clearly a flawed study in terms of drawing generalizable conclusions:
    • CBP was not really checked in a “clinic”, but at a workplace
    • As a workplace-based study, there is the “healthy worker bias” which not only selects people who tend to be healthier, but also selects people who may have somewhat higher social economic status (which itself seems to confer better health outcomes), as well as having few individuals over the age of 65. Of note only 5% of these people had elevated CBP, likely reflecting this healthier population.
    • The study did not include many nonwhite patients.
  • These biases clearly undercut the generalizability of the study’s results. Also, the high level of masked hypertension raises the question that more fit people (lower BMI) exercise more and have higher ambulatory pressures.
  • BUT, other studies have found masked hypertension in a wide array of patients (see Bobrie G. J Hypertension 2008, 26:1715) which found that the prevalence of masked hypertension was between 8 and 20%, and as high as 50% in treated hypertensive patients. A few studies mentioned in this meta-analysis found that masked hypertension was actually associated with 2 to 3 times the cardiovascular events than either white-coat hypertension or controlled hypertension in treated patients. In untreated patients, the data seems pretty mixed: studies varied between no increased cardiovascular risk to the same risk as untreated sustained hypertension. a couple of the studies:
    • The Jackson Heart Study (Diaz KM. American Journal of Hypertension 28(7) July 2015) looked specifically at African-Americans, finding that the prevalence of masked hypertension was 25.9% (34.4% in people with normal CBP) and that all of the surrogate markers of carotid artery intima-media thickening, left ventricular mass, and microalbuminuria were elevated (vs controls) in those with masked hypertension, and similar to those with sustained hypertension. They also found that male gender, smoking, diabetes, and antihypertensive medication use were independently associated with masked hypertension.
    • And, another study, the Dallas Heart Study (Tientcheu D. JACC. 2015; 66: 2170) assessed masked versus whitecoat hypertension and sustained hypertension in a group with 54% being African-American. They found a 17.8% prevalence of masked hypertension and a 3.3% of whitecoat hypertension. The risk for cardiovascular events over nine years was significantly higher in both the masked hypertension and sustained hypertension groups, but was barely significant in the whitecoat hypertension group, assessing cardiovascular outcomes. Controlling for an array of risk factors including clinic blood pressure measurements, they found that higher 24-hour ambulatory systolic and diastolic pressures were independent risk factors for new cardiovascular events. The adjusted relative risk for cardiovascular events was a 34% increase for each 1-SD increase in the 24 hour blood pressure, a 30% increase for ambulatory systolic blood pressure during the daytime and a 27% increase for ambulatory systolic blood pressure during the nighttime. For diastolic pressure the cardiovascular risk was 21% for each 1-SD increase for each, 24% for ambulatory diastolic pressure during the daytime and 18% for the nighttime. This graphs shows that there was essentially no relationship between clinic systolic pressures and cardiovascular events, that the correlation within each group of office-based blood systolics was only by ABP.


  • So, there are a number of questions that arise from these studies:
    • There is no clear consensus on how to define masked hypertension/what are the cutpoints? They used an ABP cutpoint of 135/85 mmHg as their definition of hypertension.  A consensus guideline suggested a 24-h average of >130/80, a daytime average of >135/85, and a night-time average of >120/70 (see O’Brien E. Hypertension 2013; 62: 988)
    • Are the clinical effects of masked hypertension really just those of increased blood pressure variability (see
    • Is there any real clinical advantage to identifying and treating patients with masked hypertension, and how?
    • How do we look practically for masked hypertension in our patients given that they have normal blood pressure in the office (assuming the studies suggest that identifying and treating these patients actually matters).

So, why, you might ask, am I bringing up masked hypertension, when there are no studies showing that unmasking it and treating it does anything???

  • I think it is always useful to hear about different information/perspectives which challenge the predominant ideology. The early studies on “mild hypertension” focused exclusively on clinic-based diastolic blood pressure. Then a few epidemiologic studies documented that systolic blood pressure was an even better predictor of cardiovascular events, moving the clinical target pressures into clinic-based full blood pressures. Then several studies either supporting or debunking the role of white-coat hypertension as important, and now, per my reading, suggesting that it is a little important but much less so than the other forms of hypertension. Then lots of studies on ambulatory blood pressure finding it to be much more predictive of clinical events than clinic-based blood pressure, and other studies showing that blood-pressure variability (either from clinic visit to clinic visit, or over the course of a 24-hour period) as being important. Now, over the past few years, emerges masked hypertension. We still do not know what to do with this, but there are a constellation of studies suggesting that this may be as important as sustained hypertension. But I think the real positive of this evolution in our thinking is that we are now situating hypertension in the realm of a person’s actual life instead of the artificial constructs of the clinic setting, and the data support this…
  • Masked hypertension fits in well with the increasing data on ABP as the predictor-of-choice for clinical events
  • It probably makes sense to think about masked hypertension in certain people, esp those with highish clinical blood pressure, since as in the above study, they are more liklely to have masked hypertension. Or in those with possible hypertension-related damage (e.g. retinal changes, LVH, renal dysfunction/microalbuminuria….) And, I think it makes sense to use the diagnosis of masked hypertension to reinforce the generally-useful-anyway lifestyle changes (diet, exercise, stress reduction…). I would be hesitant to prescribe meds for masked hypertension, lacking any real data on outcomes
  • And, as to how to measure it, it would be great to have ABP monitoring available and inexpensive. But, my completely untested hypothesis (though likely more practicable) is to use home-based or pharmacy-based measurements (which seems to be more accurate than CBP and approach that of ABP in the few studies done), with the clear prescription that the person should sit down/relax for several minutes, then check the numbers.

And, for the complete set of hypertension blogs, see

Primary Care Corner with Geoffrey Modest MD: Optimal blood pressure in patients with atrial fibrillation

30 Nov, 16 | by EBM

By Dr. Geoffrey Modest

It is unclear from the literature what the goal blood pressure should be in patients with atrial fibrillation, and this is not addressed by any of the guidelines. A post hoc analysis of the AFFIRM trial (Atrial Fibrillation Follow-up Investigation of Rhythm Management, a prospective trial assessing the strategy of rate versus rhythm control) looked retroactively at the relationship of achieved blood pressure and outcomes (see Badheka AO. Am J Cardiol 2014; 114: 727)).


  • 3947 patients in the trial were followed 6 years, noting their systolic and diastolic blood pressures (recorded after sitting quietly for at least five minutes) at baseline and at follow-up, divided into 10-mm Hg increments. The follow-up blood pressure was defined as the average of all available blood pressure measurements during each post-baseline visit
  • Mean age 69, and the following were significantly (and much) more frequent in patients who had lower blood pressure, with average percentages overall as follows:  60% in men, 70% hypertension, 40% coronary artery disease, 25% in patients with heart failure, 14% smoking
  • The endpoints assessed were all-cause mortality; the combination of all-cause mortality, ventricular tachycardia or fibrillation, pulseless electrical activity, significant bradycardia, stroke, major bleeding, MI, and PE as a composite secondary outcome.


  • All-cause mortality was observed in 614 people (15.6% of the group)
  • The incidence of all-cause mortality was lowest in those with BP 140/78 mm Hg, with a U-shaped curve. All-cause mortality was:
    • 9 fold higher in the group with systolic blood pressure <110; 1.9 fold higher in those with systolic greater >160
    • 3.9 fold higher in the group with diastolic <60; 1.8 fold higher in the group with diastolic >90
  • There was a similar U-shaped relationship to the composite secondary outcome.
  • Subgroup analyses also found a similar U-shaped curve with an increased all-cause mortality with blood pressure <110/60, including the following subgroups: whether or not they had CAD, hypertension, heart failure, or reduced ejection fraction.


  • One of the complicating factors in assessing the optimal blood pressure in patients with atrial fibrillation is that several of the drugs we use to control rate (e.g. beta-blockers and non-dihydropyiridine calcium channel blockers) also decrease blood pressure. So, one of the complicating factors in interpreting the association between lower blood pressure and increased mortality is inherent in this retrospective observational study: are those who require more medications to control their heart rate at a higher risk of death, just because their harder-to-manage atrial fibrillation is associated with higher mortality? And their higher incidence of hypotension merely reflects their need for more meds (which also lower their blood pressure) to control that rate???
  • Another large issue is the dramatic baseline comorbidities in those with lower blood pressure, reinforcing the fact that these were much sicker patients and raising questions as to whether the study could mathematically adjust for these covariates in their final analysis. The authors did control for age, history of hypertension, history of heart failure, history of MI or revascularization, history of stroke, diabetes, smoking status, use of warfarin, lipid-lowering therapy, diuretics, and which group they were randomized to in the AFFIRM  trial. However, given how apparently sick these patients with low blood pressure were, one wonders if there were other important variables not included (e.g. other medical conditions such as renal failure, or COPD –esp since those with lower BP also had lower BMI, or psychosocial conditions associated with higher mortality such as depression)???
  • And what really is the actual blood pressure in patients with atrial fibrillation? The automated blood pressure cuffs typically use an oscillometric methodology. Studies have shown that many of these cuffs are inaccurate in patients with atrial fibrillation (see DOI: 10.1111/jch.12545). And, a larger issue to me is that there is a large blood pressure variability between measurements in patients with atrial fibrillation, with one measurement picking up a particularly strong, forceful beat, leading to a systolic blood pressure that may be 30 to 40 mm Hg higher than other readings. Some people suggest averaging several recordings (??how many), but I have no idea whether this correlates with clinical events are not, what the best methodology for determining that clinically-relevant BP really should be, or what would be the optimal goal BP (which also requires a good clinical prospective study using a validated methodology. perhaps 24-hour ambulatory blood pressure monitoring with an approved and accurate automated cuff???). As mentioned, i have seen no guidelines address the blood pressure issue: either its measurement or management goals.
  • One issue that is found frequently in observational studies on hypertension is a J-shaped or U-shaped curve. Of note this tends not to be found in controlled trials (e.g. the SPRINT trial), suggesting that there may be a bias in the uncontrolled trials: those with lower blood pressure have higher mortality related to the fact that they have underlying diseases leading to both a lower blood pressure and higher mortality
  • Of note, there are some data suggesting that we do not need to be overly aggressive in controlling rate (see Van Gelder IC. N Engl J Med 2010; 362: 1363), which also raises the interesting question of whether those with aggressive rate control may have had increased mortality because their blood pressure was lower, balancing perhaps a clinical benefit of the lower rate (i.e. there might be some benefit for tighter rate control, but only in those who could achieve this without lowering their blood pressure too much). Unfortunately in this Van Gelder article they did not mention the achieved blood pressures in the 2 groups.
  • This AFFIRM article suggests that the target blood pressure in those with atrial fibrillation may be higher than in the general population. However, the methodologic issues above, to me, simply amplify the issue that there really are no good clinical data guiding us on either how to measure blood pressure or what the goal should be in those with atrial fibrillation….

Primary Care Corner with Geoffrey Modest MD: HTN Goal in Diabetics without CVD

7 Nov, 16 | by EBM

By Dr. Geoffrey Modest

A large Swedish population study found that in diabetics with no previous cardiovascular disease, there were progressively fewer cardiovascular events as the systolic blood pressure was lower (see


  • 187,106 patients in the Swedish national diabetes register for at least 1 year, <= 75 yo, and no known cardiovascular disease (CVD), from 2006-2012 with mean follow-up of 5.0 years. From 861 primary care units and hospital outpatient clinics
  • Most of the demographics got worse as the cohort in each 10-mm group of BP increased: median age was 55 in the lowest SBP group vs 64 in the highest; duration of diabetes 4.8 vs 6.8 years and the higher SBP group was more likely to be on more aggressive diabetes management;  LDL 2.8 vs 3.0 mmol/L but HDL 1.3 in all; more micro/macroalbuminuria in those with the highest SBP; and the mean number of BP meds was 0.7 in the SBP 110-19 cohort vs 1.1 in the 130-139 cohort vs 1.6 in the >160 mmHg cohort


  • ComparingSBP 110-119 mmHg vs those with SBP 130-199:
    • Non-fatal MI, RR 0.76 (0.64-0.91, p=0.003), 24% risk reduction
    • Total acute MI, RR 0.85 (0.72-0.99, p=0.04), 15% risk reduction
    • Non-fatal CVD, RR 0.82 (0.72-0.93, p=0.04), 18% risk reduction
    • Non-fatal coronary heart disease, , RR 0.88 (0.79-0.99, p=0.04),12% risk reduction
    • There was no suggestion of J-shaped relationship, except for heart failure and total mortality, and this was only significant for the lowest SBP group
  • Figure below shows that there was a consistent relationship between SBP and non-fatal CVD events over the course of the study. For all of the CVD endpoints, this relationship held, even after controlling for age, sex, duration of diabetes, type of diabetes treatment, HbA1c, smoking status, LDL, HDL, triglycerides, micro/macroalbuminura; as well as thiazide diuretics, loop diuretics, calcium antagonists, spironolactone, b-blockers, and drugs for heart disease. Of note, they did not control for those on vs not on antihypertensives, which may be important.




  • So, why is it so difficult to zero-in on a goal blood pressure in diabetics? This study suggests that lower blood pressure is better. But the various guideline groups have been increasing the BP goal lately, though based on no new evidence: the ADA (Am Diabetes Assn) in 2016 set the overall BP guidelines at the higher level of <140/90 (see and explicitly did not recommend it to be <130/70 in older adults, in conformity with JNC8 (which also has a higher goal than JNC7)
  • I think there could be various different explanations:
    • The current study focused on a less-sick population than most of the others: a younger cohort, who had no known baseline CVD, and some did not have treated hypertension
      • Is the diabetes itself different? (perhaps longer-standing diabetes creates end-organ changes which dictate different optimal BP goals)
      • Are we using diabetic medications which make things worse, and using more of them on patients with longer-standing and more treatment-resistant diabetes? Similarly with the antihypertensives?
        • In terms of diabetes control, a case in point here is the ACCORD trial, one of the major studies heralded as a reason to raise the target A1c. Those assigned to the “intensive control wing”, achieved an A1c of 6.4, but 91% were on a thiazolidinedione (TZD), vs an A1c of 7.5 in the less aggressively treated group but with 58% on a TZD. But the TZD of choice was rosiglitazone, which has the unfortunate tendency to increase cardiovascular outcomes (and is one of the reasons that I find it unfortunate that the FDA and most of us accept A1c as an acceptable clinical surrogate).
        • And, this brings up the issue of medication-flogging…. are those patients with easy-to-control diabetes or hypertension different? As in the first point, is there a fundamental difference in their pathophysiology or clinical outcome? A subgroup analysis of this ACCORD study actually found that those who achieved a lower A1c in fact did better, all the way down to an A1c of 6!!, but as the number of meds needed in the attempt to lower the A1c increased, they had worse outcomes (i.e., medication-flogging of patients to improve their A1c led to worse outcomes even at a much higher A1c). See Riddle MC. Diabetes Care; 33:983. An Italian observational study also found that the goal of A1c in terms of clinical outcomes was lower in those with fewer chronic medical conditions (see Greenfield S. Ann of Intern Med 2009; 151: 854).
        • This last point brings up the parallel issue: should the blood pressure goal be different in those with fewer chronic changes from long-standing hypertension (e.g. atherosclerosis, or changes in the local autoregulation of blood pressure at a microvascular level) vs those with perhaps newer onset hypertension with fewer of these changes? should we have different BP goals in those who easily achieve a blood pressure of 110-120 systolic if it can be achieved with 1-2 drugs, vs those with SBP of 140+ systolic, who would be struggling to achieve even close to the lower range with 4 drugs? It was certainly the case in the Swedish study that as the SBP of the cohort increased, there were more meds being used, distributed pretty evenly amongst the different types of meds.
        • The prior observational studies have often found a J-shaped relationship between blood pressure and CVD events in diabetics, though this has been questioned by the potential bias in observational studies that patients who had more bad outcomes at lower pressures did so because they were really sick at the start, and it was this increased morbidity that led to lower blood pressure. It is notable in the above Swedish study that the J-shaped curve did happen in those with lower blood pressure, but only for total mortality and heart failure, and not for the specific CVD outcomes, suggesting that there may have been issues that these patients with lower SBP were indeed sicker. In fact those who died in this Swedish study were likely to have had more comorbidities, since they had higher rates of smoking (32%), use of loop diuretics, spironolactone, and drugs for heart disease.
        • The ACCORD-BP study of diabetic patients (N Engl J Med 2010; 362: 1575), another wing of the above ACCORD study, found no overall benefit in 4733 patients in those achieving a systolic BP of 119 mm Hg vs 133.5 mm Hg, except for the prespecified secondary outcome of stroke, where there was a 41% decrease (p=0.01), but at the expense of an increase in serious adverse events (from 1.3% of the population to 3.3%). The absolute risk of stroke was 0.53%/yr vs 0.32%/yr, which translates roughly to 2.6% vs 1.6% over the 4.7 year study. The serious adverse events were largely hypotension/syncope/bradycardia or arrhythmia/hyperkalemia. The intensive group averaged 3.4 BP meds and the standard group 2.2. But, as opposed to many strokes, all of these serious adverse could be tracked and corrected, and there was no evidence of increased morbidity/mortality from these adverse events. Other trials, such as ONTARGET found a J-shaped curve, and the INVEST trial found no benefit if the SBP were lowered below 130, (though a subgroup analysis of ONTARGET found that it was those with a higher baseline risk who had CVD events, rather than the degree of reduction of the BP). These are the trials cited in JNC8 and the ADA guidelines as the reason to shoot for a higher SBP target.
      • So, my best guess is that lower SBP is better for those who don’t have lots of comorbidities and do have more easily treated hypertension, with the following caveats:

Also,, a recent meta-analysis found benefit of a goal SBP of around 130 to be better overall than 140.


Primary Care Corner with Geoffrey Modest MD: Lower Blood Pressure in Elderly and Decreased Morbidity

20 Oct, 16 | by EBM

By Dr. Geoffrey Modest

A recent study of older community-dwelling high-risk hypertensive patients looked at the relationship between their achieved blood pressure and cardiovascular outcomes (see Myers MG Hypertension. 2016;68:866).


  • 6183 community-dwelling Ontario residents >65yo on antihypertensive therapy, followed mean of 4.6 years (the CHAP study: Cardiovascular Health Awareness Program)
  • Blood pressure measured (as in the SPRINT trial) by AOBP (electronic automated office blood pressure). In Ontario, the protocol was that the person rest seated in a quiet place, undisturbed before and during the readings. The research staff did not speak to the subjects or interact with them. The patient did not wait 5 minutes, but the AOBP recorded the blood pressure each minute for 5 minutes and computed the mean value. The AOBP was recorded in community pharmacies, which a different study found to be similar to AOBP done in the office of the patient’s own family physician. In the SPRINT trial, the initial reading was after the patient seated and wait 5 minutes in quiet, then the blood pressure was measured using a similar automated machine as in the Ontario study
  • Mean age 76, 42% male, 6% self-reported smokers, 48% self-reported high blood cholesterol, 27% diabetic, 12% congestive heart failure
  • Mean AOBP: 134.3/72.9 mmHg. Mean number of antihypertensive meds = 1.8-1.9 for each 10mm blood pressure category


  • 904 nonfatal and fatal cardiovascular events during the follow-up period
  • Multivariate adjustment (adjusting for age, sex, coronary artery disease, cerebrovascular disease, congestive heart failure, diabetes, number of anti-hypertensive meds, hypercholesterolemia, smoking, self-reported health status, BMI, number of unique drugs) found that, as compared to an achieved systolic of 110-119 mmHg [n=837]:
    • <110 mmHg: HR 1.38 (1.04-1.81) [n=546]
    • 120-129mmHg: HR 1.30 (1.01-1.66)[n=1308]
    • 130-139mmHg: HR 1.23 (0.96-1.58), nonsignificant [n=1259]
    • 140-149mmHg: HR 1.18 (0.90-1.54), nonsignificant [n=984]
    • 150-159mmHg: HR 1.43 (1.08-1.90) [n=604]
    • 160+ mmHg: HR 1.85 (1.42-2.41) [n=645]
  • Multivariate adjustment for achieved diastolic blood pressure, as compared to referent of 60-69 mmHg [n=1788]:
    • <60 mmHg: HR 1.31 (1.07-1.61) [n=636]
    • 70-79 mmHg: HR 1.02 (0.87-1.21), nonsignificant [n=2212]
    • 80-89 mmHg: HR 1.08 (0.88-1.32), nonsignificant [n=1133]
    • 90+ mmHg: HR 1.14 (0.86-1.52), nonsignificant  [n=414]
  • Multivariate adjustment for pulse pressure (systolic minus diastolic), with 50-59 mmHg being referent [no comment on how many people were in each group]:
    • <50 mmHg: HR 1.06 (0.87-1.29), nonsignificant
    • 60-69 mmHg: HR 1.07 (0.87-1.30), nonsignificant
    • 70-79 mmHg: HR 1.10 (0.88-1.38), nonsignificant
    • 80-89 mmHg: HR 1.33 (1.03-1.72)
    • 90+ mmHg: HR 1.83 (1.42-2.34)
  • No difference in outcome in the 1673 patients with diabetes vs the 4510 nondiabetics


  • So, overall, this study found increased CV outcomes in patients with achieved systolic blood pressure below 110 mmHg or if 120-129 mmHg (vs the nadir of 110-119 mmHg), and not again until >150 mmHg; an increase only with diastolics <60 mmHg; and an increase when the pulse pressure was >80 mmHg (the latter likely reflecting stiff arteries, as with diffuse atherosclerosis). By the way, this nadir of SBP 110-119 mmHg was the same as in the SPRINT trial
  • This observational study, as with many observational studies, did find a J-curve in clinical outcomes in hypertensive patients, with increased morbidity/mortality in those with both lower and higher achieved blood pressure. This was not found in the prospective randomized-controlled SPRINT or ACCORD trials, where the lower blood pressure was better, suggesting that there might be a selection bias here: i.e., are those with lower blood pressure different from the others? Do they have non-blood pressure related terminal illnesses, like cancer or infections or perhaps worse cases of diabetes or heart failure, which leads to lower blood pressure and also to higher morbidity/mortality?)
  • One thing I have commented on in the past: there is a rather strong trend in the medical literature to put many study details in a “supplement”, which i think often includes critical details. In the above study, for example, there were no data on the number of patients in each cell of achieved systolic and diastolic blood pressures, except in the supplement, and none anywhere on the pulse pressure categories. It seems to me to be very important information (was the lack of statistical significance in some groups due to small numbers of patients??). My guess is that pretty few readers go to the trouble of going online to get this information; it is much less of a hassle to just accept the authors’ interpretations in the main article. And if we just got the PDF from the author by email, the supplementary material is not included (at least in my experience). And I pretty often do find very important information buried in the supplements, in the relatively few times I scour them, and this information sometimes changes my interpretation of the article.
  • One of my biggest concerns with clinical management of hypertension is that in the typical rushed clinical environment, most of us simply accept the automated blood pressure readings of the medical assistants as being accurate. This brings up several very important issues in probably what is the most common clinical issue we see in adult patients and one that we can potentially positively affect:
  • The automated machine is not always accurate (e.g. in patients with atrial fibrillation or frequent dysrhythmias)
  • Typically the patient walks into the room and has the blood pressure done quickly in a not-so-restful environment (and many of my patients are pretty unfit physically/deconditioned, so have a much much higher blood pressure than when I retake a manual pressure when they are sitting quietly in the exam room)
  • I do routinely recheck blood pressures in the exam room, typically after leaving the patient sitting on the exam table for about 2 minutes, often with the lights out (I go into the other room to write my note), and find in this not-very-scientific study that about 50% of the medical assistant blood pressures are really pretty close to what I get, 40+% are much lower than the medical assistant pressures (and often by 20-40 mmHg!!!), and the rest are much higher than the medical assistant values (not sure why, but i sometimes get manual pressures of 190/108, when the medical assistants get 118/77…..)
  • There is a significant literature suggesting that clinical outcomes track much better with 24-hour ambulatory blood pressure monitoring (ABPM) or home-based blood pressure (I do ask patients to bring in their cuff and make sure it compares well with the manual recording I get, done at the same time). See the URLs below for more articles on this. and the USPSTF does now (finally) endorse non-clinic based blood pressures (see below)
  • Using AOBP in this Ontario study, similar to the SPRINT study, is an interesting hybrid, which sounds pretty good to me, though i have not seen much in the literature on this. But again, to apply these trials to clinical practice, we need to get blood pressure readings that approximate how it was done in the studies (i.e., at least with having the patient rest in a quiet room for awhile, or using ABPM; though with ABPM, there is only one cutpoint: either above or below 135/85, not with the fine gradations of the SPRINT or other trials assessing treating blood pressure to different targets.
  • One of the strengths of this study is that it was not a formal study with selected patients according to specific inclusion and exclusion criteria, but more of a community approach, likely more reflective of our actual clinical practice (i.e., there is decreased internal validity in that there may have been very uneven distribution of patients in the different categories of achieved blood pressure, perhaps with healthier patients overall able to achieve an SBP in the 110-119 range, or their clinicians were more aggressive in trying to lower their blood pressure; yet more external validity in that it more closely reflects real-world clinical practice. one drawback to the study was that AOBP was measured only one point in time. and they did not have much granular data, such as how long the patient had hypertension or what time of day the AOBP was recorded, or how bad the heart failure was, etc. But it was intriguing that the results (best to have SBP 110-119) was so similar to that of the rigorous SPRINT trial.

See for an array of articles/reviews on hypertension, including a recent one showing that blood pressure variability is associated with more cardiovascular disease

See for a brief review of ambulatory BP monitoring and the USPSTF recommendations

See for the overall SPRINT trial

See for an analysis of the elderly subgroup of the SPRINT trial, finding improvement in cardiac outcomes in those >75 yo with achieved SBP of 123

See for some of the data on home blood pressure monitoring

See shows the benefits of patient self-monitoring blood pressure and self-titrating meds

Primary Care Corner with Geoffrey Modest MD: Orthostatic Hypotension and Increased Heart Failure and Mortality

19 Oct, 16 | by EBM

By Dr. Geoffrey Modest

There was an interesting subgroup analysis of the ACCORD blood pressure wing (Action to Control Cardiovascular Risk in Diabetes) which found that those with orthostatic hypotension (OH) had a significantly higher risk of mortality and heart failure events (see Fleg JL, Hypertension. 2016;68:888 ). Details:

  • The ACCORD trial had 10,251 high-risk patients with type 2 diabetes, hemoglobin A1c >7.5%, and were between 40 and 79 years old with cardiovascular disease or 55 to 79 years with anatomic evidence of subclinical atherosclerosis, albuminurea, LVH, or >= 2 additional cardiovascular risk factors. 4733 were randomly assigned to intensive vs standard blood pressure control in a non-blinded trial, with target systolic blood pressure (SBP) of <120 vs <140 mm Hg, and with no requirements as to what medications to give (clinicians’ decisions)
  • 4266 participants were involved in this analysis, with blood pressure measured at baseline, and at the 12 month, and 48 month follow-up visits. The blood pressure was measured using an automated oscillometric device after the patient had been seated at least five minutes, with the blood pressure determined three times at one minute intervals. The patients then stood up and had their blood pressure measured every minute for three minutes. The patients were asked if they experienced dizziness or felt light-headed.
  • Orthostatic hypotension (OH) was defined as a decline in SBP > 20 mmHg or decline in DBP > 10 mmHg
  • The average difference in blood pressure achieved between the intensive vs standard groups was 14.2/6.1 mmHg, with the mean number of medications being 3.4 for the intensive group and 2.1 for the standard group.
  • Serious adverse events related to the intervention, including hypotension and syncope, were found in 3.3% of the intensive group compared to 1.3% of the standard group, with p<0.001


  • OH occurred at least once in 852 people (20.0%). In the adjusted model, this occurred most commonly in women, current smokers, those with higher baseline SBP, higher A1c, and those on beta blockers, alpha blockers or insulin. Of note, neither age nor assignment to intensive vs. standard BP treatment goals was associated with OH
  • Approximate 5% of all patients, independent of group, felt dizzy on standing. The incidence was slightly more in the intensive group but only at the 48 month examination.
  • There was no significant difference in OH prevalence, incidence, or resolution between those in the intensive vs. control groups.
  • People with OH were about twice as likely to report symptoms of dizziness on standing than those without OH, but this was only in about 17-20% of the patients who were symptomatic.
  • Those with OH had an 85% higher risk for heart failure deaths or hospitalizations (p= 0.01) and 62% higher risk for total mortality (p= 0.02).


  • In my experience, beta blockers are the medication most commonly associated with OH and large decreases in blood pressure on standing, even in those with high sitting pressures.
  • It is not so surprising that those with higher A1c’s and on insulin have more OH, since they may well have more autonomic neuropathy, and there may also be some direct insulin-induced vasodilation, perhaps through endothelium-dependent mechanisms.
  • Or that smokers have more OH (probably more atherosclerotic disease of their large arteries, leading to higher SBP but lower DBP because of their nondistensibility)
  • Or in those who had initially higher SBP (again perhaps related to more atherosclerotic large vessel disease, leading to more isolated systolic hypertension)
  • It is, however, unexpected and striking that they found no correlation of OH with age. All of the studies I have seen have shown a significant increase with age. Perhaps part of the reason is that in this study they had an 80-year-old cut off.
  • And, there was no association with whether the patient was in the more aggressive blood pressure lowering group or not (which also supports checking orthostatics in patients with higher blood pressures)
  • So, in this study, it was unclear whether OH was simply a marker of people at higher risk of morbidity/mortality (e.g. more advanced diabetes, with more autonomic neuropathy, etc. as above), or whether it was the cause. but given the not-so-unlikely possibility of the latter being part of the issue, I think it makes sense to assess OH regularly in patients (and i do so in all elderly hypertensive patients, even if they do not have diabetes, and have found, i think, pretty impressive 30+ mmHg drops in blood pressure even in those with systolics in the 150-160 range, and not so uncommonly…) and customize therapy to avoid excessive falls in blood pressure, whether they are symptomatic or not (the reason being: even if asymptomatic, perhaps there are times when they eat/drink less at home or outside in the heat and they become symptomatic, fall, etc; and perhaps the low flow associated with OH really is not so good for the brain, heart, kidneys, etc in the longterm.)
  • For example, there are some studies showing cognitive decline with low blood pressure: see
  • Another important point from the study was that dizziness is not commonly reported with clear-cut OH by blood pressure measurement (i.e. we should not rely on reported dizziness as a reliable marker of OH), and that OH is not a consistent finding each time it is measured.
  • All of this supports my prior suggestions that we measure orthostatic blood pressures in older patients on a regular basis, even if the SBP is in the 150 range, and adjust meds accordingly (?ACCORDingly).


(See for more studies on orthostatic hypotension, including the finding that initial hypotension on standing is in fact much more common than standard orthostatic hypotension after a couple minutes)

Primary Care Corner with Geoffrey Modest MD: Blood Pressure Variability Increases Cardiovasc Disease

4 Oct, 16 | by EBM

By Dr. Geoffrey Modest

There has been concern about the adverse effects of blood pressure variability on cardiovascular outcomes. A prior blog (see ) reviewed the ALLHAT trial, which found that visit-to-visit blood pressure variability was associated with increased CV events and commented on a 2010 issue of the Lancet that found that hour-to-hour BP variability in individuals was associated with more strokes, and, to a lesser degree, coronary events (see Rothwell PM. Lancet 2010;375:895). BMJ just had a systematic review and meta-analysis confirming the association (see


  • 41 papers representing 19 observational cohort studies and 17 clinical trial cohorts. 24 papers studied long-term variability (monitoring blood pressure in clinics), 4 studies mid-term variability (home monitoring) and 15 short-term (ambulatory blood pressure monitoring).
  • Range of studies: 457 to 122,636 participants; follow-up ranged from 2514 to 490,544 person-years; mean age ranged from 48.5 to 77 yo
  • Increased long term variability in systolic blood pressure was associated with:
    • All-cause mortality: 15% increase, HR 1.15 (1.09 to 1.22)
    • Cardiovascular disease mortality: 18% increase, HR 1.18 (1.09 to 1.28)
    • Cardiovascular disease events: 18% increase, HR 1.18 (1.07 to 1.30)
    • Coronary heart disease: 10% increase, HR 1.10 (1.04 to 1.16)
    • Stroke: 15% increase, HR 1.15 (1.04 to 1.27)
  • Increased mid-term variability (home BP) in daytime systolic blood pressure was associated with all-cause mortality [HR 1.15 (1.06 to 1.26)].
  • Increased short term variability (ambulatory BP) in daytime systolic blood pressure was also associated with all-cause mortality [HR 1.10 (1.04 to 1.16)]. The conclusions are a bit limited, since 2 studies dominated the meta-analysis.


  • As with meta-analyses, they combine different studies with differing methodologies, limiting their conclusions. For example, there is not necessarily any consistency across studies in terms of how the BP was measured, what size cuffs were used, whether using manual or automated devices, etc.)
  • They did not include studies on nocturnal dipping (that’s the normal variation, with lower blood pressure at night on ambulatory monitoring; non-dippers seem to have higher mortality). They did exclude patients on dialysis, since blood pressure variability is basically intrinsic to hemodialysis patients.
  • As a perspective, the 15+% difference in cardiovascular events found still pales in comparison to overall effect of lowering the mean blood pressure. I.e., the primary goal is to decrease the mean blood pressure. That being said, the difference from blood pressure variability in the above meta-analysis did control for the mean blood pressure, revealing an increased risk over the mean BP
  • Blood pressure variability during the day is normal, typically up to the 18/12 mmHg range. This variability is enhanced in those with arterial stiffness (and the above meta-analysis was skewed to older hypertensive patients), which may put these patients at higher CV risk.
  • This all supports the conclusions that:
    • We should be doing more ambulatory or home BP monitoring: several analyses have found that ambulatory or home blood pressure monitoring is superior to office blood pressure at predicting cardiovascular events, perhaps since ambulatory or home measurements are more likely to pick up BP variability.
    • There are likely advantages to using BP meds that produce a more sustained, less variable blood pressure over 24 hours: amlodipine seems to be the best, ACE-I are intermediate (and there are arguments that the increased stroke rate in several studies of ACE-I may be related to the higher blood pressure in the early mornings), and HCTZ up to 25 mg is the worst (for example, see Webb AJS. Lancet 2010; 375: 906). B-blockers are also in the intermediate category.

EBM blog homepage

Evidence-Based Medicine blog

Analysis and discussion of developments in Evidence-Based Medicine Visit site

Creative Comms logo

Latest from Evidence-Based Medicine

Latest from EBM