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.

Archive for April, 2016

Primary Care Corner with Geoffrey Modest MD: Nonstatin Lipid Lowering Drugs, and Statin Myopathy

29 Apr, 16 | by EBM

By Dr. Geoffrey Modest

The ACC/AHA has just released a therapy guideline for patients who are on a statin but do not achieve the goal LDL (see http://content.onlinejacc.org/article.aspx?articleID=2510936#tab1 for full guideline). The context here is that recent guidelines have suggested that we do not add meds on top of statins, but there are a couple of studies which led them to change these recommendations). Main points:

  • If a patient is on a high-intensity statin but not achieving the goal LDL (which in the former guidelines was >50% decrease in LDL, though they are noting that one “may consider LDL <100” as a target)
    • Address statin adherence and intensify lifestyle changes (consider phytosterols)
    • Increase to high-intensity statin if not already on one [as pointed out in prior blogs, there are really pretty broad differences in statin intensity within their groupings: rosuvastatin 40 gets more LDL reduction than atorvastatin 40, so I definitely move within the high-intensity group as my first med change. As a side-issue, I do see many patients who have a really low LDL, in the 40-50 range, even on “low-intensity” statins, and I do not move to more intense ones — see my many prior blogs on this: I am still pretty convinced of the data on treating to a goal LDL and still do so. And I think their “consider LDL<100” suggests that they are moving more in that direction as well]
    • Consider adding non-statin meds:
      • Consider ezetimibe first, or consider adding or replacing with PCSK9 inhibitor second
    • Especially in patients with clinical CAD, and especially in diabetics, it is better to look at non-HDL levels (i.e., total cholesterol minus HDL), which has more predictive value than LDL levels [there are a few studies I have seen on this. In general, for those just on statins, the non-HDL is a better predictor of further clinical events (see JAMA 2012; 307:1302), and the target is non-HDL <100. as per above, they do include as a “may consider” the LDL goal of <70 or non-HDL goal of <100]
    • In high risk patients on a statin who do not achieve the goal, consider adding ezetimibe [though they acknowledge there is an evidence gap there, which is a euphemism for the fact that this has not been studied], or a bile-acid sequestrant if triglycerides <300 as second line [also no clinical outcome data on this….]. And in those with really high risk (documented CAD plus baseline LDL>190 and not achieve >50% LDL reduction or LDL<70), consider adding PCKS9 inhibitor even as a first step
    • They even consider PCKS9 inhibitors in patients without clinical CAD but baseline LDL>190 and not achieve >50% reduction in LDL on high-intensity statin. But no clear role for them for primary prevention or LDL <190, with or without diabetes
    • Can consider a nonstatin add-on in adults 40-75 yo without clinical ASCVD or diabetes, LDL 70-189 and estimated 10-year risk of >7.5% and on a statin for primary prevention but who do not achieve a >=30% reduction in LDL. consider ezetimibe or bile-acid sequestrant
  • They are not so clear on what to do if the patient is statin-intolerant. They do suggest that there may be less problem with lower statin dose [and the data are pretty clear that the majority of LDL reduction is with 5-10mg of a statin] or less frequent dosing [I have seen some data long ago suggesting similar lipid effects if take atorvastatin only a few times a week as with every day.]

Interestingly, at about the same time as the AHA came out with their guidelines for using non-statin lipid lowering therapy, JAMA published the GAUSS-3 RCT which found that the commonly-reported statin myopathy is very often not really true, and also found that evolocumab (a PCSK9 inhibitor) is much more effective than ezetimibe (see doi:10.1001/jama.2016.3608).

Details (in pretty brief):

  • 511 patients with uncontrolled high LDL levels and a history of intolerance to 2 or more statins (e.g. atorvastatin 10mg or any dose of other statins)
  • 2 phases of the study:
    • Phase A: a 24-week crossover trial with atorvastatin 20mg vs placebo, on each drug for 10 weeks
      • 491 patients (mean age 61, 50% women, 35% with CAD, mean LDL of 212. BMI 28, 95% white, 35% with CAD, 10% smokers, 13% diabetic, 63% high risk with 10-yr risk Framingham risk score of >20%. more than 80% were intolerant to 3 or more statins)
      • Results: “intolerable” muscle symptoms in 43% (209 of the 491) in those put on atorvastatin but not on placebo; 130 (27%) had muscle symptoms only on placebo; 48 (10%) had symptoms on both placebo and atorvastatin; and 85 (17%) had symptoms to neither drug
    • Phase B: those with muscle symptoms only on atorvastatin were randomized to ezetimibe 10mg/d vs evolocumab 420 mg subcutaneously per month.
      • Results: mean LDL 220 initially, decreasing to:
        • Ezetimibe: 183 mg/dL, absolute decrease of 31.0 [-16.7% (-20.5 to -12.9%)]; 0% achieved LDL <70 at 24 weeks
        • Evolocumab: 103.6 mg/dL, absolute decrease of 106.8 [-52.8% (-55.8 to -49.8%)]; 27% achieved LDL <70 at 24 weeks
      • Muscle symptoms reported in 28.8% on ezetimibe and 20.7% on evolocumab.

So, several issues:

  • In this last study, the presence of “intolerable” muscle symptoms actually from statins was confirmed in a minority of patients who had failed >= 3 statins in 80% of them. And a large retrospective study looked at people who had discontinued a statin due to adverse effects but were then rechallenged, finding a 92% success in restoring therapy, although not necessarily with the same statin or dose. This brings up the “nocebo” effect, where placebo gives patients either a perceived adverse effect (e.g., myalgias, in the above study) or even a profound measurable physiologic effect (e.g. hypotension in a patient who “overdosed” on placebo) — see prior blog: http://blogs.bmj.com/ebm/2013/11/25/primary-care-corner-with-dr-geoffrey-modest-nocebo/ . And, for better or worse, patients are more focused on adverse effects of meds than before (?? why: related to TV advertising, more distrust of meds given publicity of drug company malfeasance, deterioration in clinician/patient relationship….). But, in my experience, the bottom line is that if a patient is convinced that a med has a bad adverse effect, there is a pretty low probability that changing to another of the same class of drug will work.
  • Unfortunately the new guidelines do not really answer a pretty common question: what do I do with patients who are unable to take statins, even after trying several different ones at low doses?  The ACC/AHA guidelines punt on this one, with some vague recommendation that these patients “should be evaluated for statin intolerance and considered for referral to a lipid specialist”. The reality is that there really are limited possibilities, as follows:
    • Ezetimibe: the reason for including ezetimibe as an “add-on” drug in the ACC/AHA guidelines above is based on the IMPROVE-IT trial (see http://blogs.bmj.com/ebm/2015/06/23/primary-care-corner-with-geoffrey-modest-md-improve-it-trial-ezetimibe/for the blog on the IMPROVE-IT trial of simvastatin plus ezetimibe vs just simvastatin on patients with acute coronary syndrome, finding small but significant benefit with the addition of ezetimibe, though this drug-company sponsored trial and, per the blog, raises many questions about how useful ezetimibe really is as an adjuvant (e.g., it might be more useful to raise the statin dose maximally first). There are no studies on other cardiac conditions (primary prevention, those with stable CAD, etc.). And the data on ezetimibe is pretty clear: those with genetic mutations which presumably create the same situation as by ezetimibe (inhibiting the Niemann-Pick C1 like 1 protein) have lower ASCVD risk, ezetimibe lowers LDL cholesterol by about 17% alone, and provides an additional 14% reduction beyond that of simvastatin, but the trials of using ezetimibe are pretty bad (other than IMPROVE-IT): a 14-month trial of people with CAD on a statin were randomized to ezetimibe or 2 g of niacin, finding an increase in carotid intima-media thickness (CIMT) with ezetimibe, vs decrease with niacin. a 2 year study (ENHANCE trial) in patients with familial hypercholesterolemia put on simvastatin 80mg with or without ezetimibe found that ezetimibe lowered LDL a lot, but there was again an increase in CIMT with the addition of ezetimibe and a nonsignificant trend to more cardiac events with ezetimibe. bottom line: I am not so excited about this drug, and in general would not prescribe it alone (without a statin) and would first really push the statin to the maximally effective one (eg rosuvastatin 40) before considering it as an add-on
    • PCSK9-inhibitors: pretty powerful LDL lowering (though in the above study, the 52.8% is on a par with the high-intensity statins). The only clinical study to date is the pretty short 11 month OSLER study (N Engl J Med 2015; 372:1500), which found that evolocumab plus standard therapy (mostly statin, some on just ezetimibe) had a 61% decrease in LDL, and 0.95% cardiovascular events (vs 2.18% in the standard-therapy group). So, really expensive drug, reasonable at this point in patients at very high risk of a clinical event with suboptimal LDL lowering on a maximal statin, bigger studies are ongoing, and no clinical data yet on monotherapy.
    • And the old standbys, with some reasonable clinical data:
      • Cholestyramine: lipid research trial: lowered LDL 10-20%, 2% decrease in ASCVD events for each 1% lowering of LDL. No mortality benefit but study was not powered to achieve than endpoint. I have been using more colesevalam as my preferred bile acid sequestrant, since it is a little stronger in terms of LDL reduction and has additional benefit on LDL when added to a statin, though there are no studies showing it specifically has clinical benefit (and there are some studies showing improved glycemic control in diabetics)
      • Gemfibrozil: Helsinki study: 1% decrease in cholesterol was associated with 4% decrease in ASCVD events. Also underpowered for mortality benefit; VA-HIT trial: 22% decrease in ASCVD in those with mean LDL 111 but low HDL of 32.
      • Niacin (which I have not used for years) at high doses in the Coronary Drug Project was associated with 27% decrease in ASCVD over 6 years, with decreased mortality 9 years after the study stopped

So, bottom line: statins are pretty great overall, with significant reduction in lipids and improvement in clinical events within 6 months or so. And the 2 studies looking at long-term effectiveness have confirmed benefit up to 20 years later. One of my major concerns with statins is the recent observation that those on statins revert to stopping their lifestyle changes (“after all, my cholesterol is so good…..”). And probably we get some blame as well (“your cholesterol is so good”, and then move on to other issues). Lifestyle changes (exercise, healthy diet, losing weight, decreasing stress, stopping smoking, etc. etc.) are so important for many other things besides lipids (diabetes prevention, improved cognition, decreasing ASCVD beyond the effect of decreasing lipids, increasing longevity…) that we as clinicians should make sure to reinforce these issues. And I have had a couple of patients do so well with lifestyle changes that they have come off statins. And these lifestyle changes take on an even more significant role in those who are intolerant of statins

Primary Care Corner with Geoffrey Modest MD: BMI, Height and Socioeconomic Status

28 Apr, 16 | by EBM

By Dr. Geoffrey Modest

One clear limitation of observational epidemiologic studies is that they are only capable of showing an association, not causality. One way to assess possible causality is through a technique called “mendelian randomization”.  There was a recent study looking at the association between BMI or height and socioeconomic status (see BMJ 2016;352:i582​). In this study, one might suspect that the well-known association between BMI and socioeconomic status (SES) could be bidirectional (high BMI might predispose a person, through conscious or unconscious bias, to be less likely to be hired or make much money, hence the lower SES; and, low SES might predispose people to a higher BMI since they are less likely to be able to afford or even have access to more expensive vegetables and fruit, have less ability to exercise because of unsafe neighborhoods, etc.). Mendelian randomization is a process where one looks at genetic differences (which are randomly allocated at conception) as an unconfounded proxy for the risk factor (in this case, BMI and height). So, since the outcome of SES cannot influence the underlying genetic variation which is set at birth, by looking at the genetic code itself instead of the complexities of BMI or height (which may reflect the genes but also be confounded by the social environment), one might see that, for example, the genes associated with increased BMI or height really are the potential drivers of low SES. The ability to do this genetic testing is now greatly enhanced because of the widespread availability of cheap genome-wide studies, currently identifying 10’s of genetic variants associated with BMI and 100’s associated with height.

Details of this study:

  • 119,669 British men and women, aged 37 to 73 in a UK Biobank that had recruited 500,000 people from across the UK from 2006-2010, and included demographics, health status, lifestyle, blood pressure and stored samples of urine, sputum and blood. This analysis was limited to white people with “British ancestry” because those from other ethnic groups did not have enough representation to be adequately powered for a statistical association.
  • Main outcome: the relationship between BMI or height and the following 5 markers of SES: age completed full-time education, level of education, job class, annual household income, and Townsend deprivation index (a composite of unemployment, non-car ownership, non-home ownership, and household overcrowding).

Results:

  • Overall analyses:
    • Pretty much all of the five SES markers were very highly correlated with coronary artery disease, hypertension, type 2 diabetes and “long illness”
    • Tall stature was related inversely to all of the above SES indicators individually, equivalently for men and women; by genetic analysis, these associations were only significant for men
    • Higher BMI was related directly to all of the above SES indicators individually, equivalently for men and women​; by genetic analysis, was only significant for women, and only for the annual household income and the Townsend deprivation index
  • Genetic analysis found that these relationships were partly causal:
    • A 1-SD (6.3cm) taller stature “caused” :
      • 06 year older age of completing full-time education (0.02-0.09, p=0.01)
      • 12 times higher odds of working in a skilled profession (1.07-1.18, p<0.001)
      • $1615 higher annual income ($970-2260, p<0.001), with stronger association in men
    • A 1-SD higher BMI (4.6 kg/m2) “caused”:
      • $4200 lower annual income ($2400-6000, p<0.001) in women
      • 10 higher level of deprivation (0.04-0.16, p=0.001), but only in women

So, what are the implications of this study:

  • We know from many studies that lower SES is association with poorer health and shorter life expectancy
  • And, we know that adult height and BMI are associated with SES, with many studies in richer countries finding that taller people and those with lower BMI have higher SES and better health
  • I cited a previous mendelian randomization trial for alcohol and cardiovascular events, assessing the often-found association between drinking more and less CAD. Mendelian randomization found the opposite: those drinkers with a genetic variant for alcohol dehydrogenase had lower alcohol consumption and lower levels of CAD risk factors (blood pressure, inflammatory markers, etc.) compared to those without this allele, and that the lower risk of CAD events was present in those with this allele who actually were lower alcohol consumers (see BMJ 2014; 349:g4164 4 doi: 10.1136/bmj.g4164,as well as the editorial in Addiction: doi:10.1111/add.12828​). So this suggests that alcohol consumption by itself is not cardioprotective, but there was likely some confounding or bias inherent in the observational studies. And this bias was the cause of the prior held association of increased alcohol consumption and cardioprotection.
  • The current study did find that the observational associations between BMI and height with SES were in fact significantly stronger than the genetic analyses, suggesting that genetics played some role, but actually only a small part in the association (i.e., a significant part of the association between BMI/height and SES may well be social and possibly bidirectional)
  • The authors do acknowledge several limitations: even if there is a genetic predisposition to a higher BMI, do people with higher BMI move to areas where they are more comfortable? (Areas with others having higher BMI, and perhaps these are areas of lower SES, leading to their children being born in areas of lower SES, getting less good education, etc. And the higher BMI itself does have pretty clear adverse health effects, leading to more illness and less ability to be employed and decreasing SES. So, the genetic association may well be dwarfed by the social issues. A little harder to make equivalent arguments about height.
  • Although this is a huge dataset, it seems to me that there are a couple of limitations to the interpretation (I should add that these are my assessments based on what I glean of the mandelian randomization methodology, and would defer to anyone more statistically or biologically sophisticated than I — would welcome feedback and would like to send around contrary interpretations). One is: do we really know all of the important genetic “determinants”, and which really are the important ones? And is there any significant interaction between the many identified genetic markers (i.e., does it make sense to look at individual genetic markers singly as in this study or in combination?).  Second: while it is undoubtedly true that social circumstances cannot change one’s genes fundamentally, it is also true that the function of genes can change at pretty much anytime through epigenetics by DNA methylation and histone modification, leading to changes in the expression of these genes. For example, rats exposed to marijuana have changes in DNA methylation, which can affect function (see http://blogs.bmj.com/ebm/2015/07/29/primary-care-corner-with-geoffrey-modest-md-marijuana-passing-through-the-generations/ ). This blog goes through some of the basic principles of epigenetics, raising the concern that this post-conception phenomenon may have very profound clinical effects. In the case of the rats, where DNA methylation was actually passed to a subsequent generation, those infant mice were more prone to opioid addiction. And there are a variety of epigenetic changes associated with the environment, lifestyle, and chronic diseases themselves. For example, there are animal studies showing that there is a pretty strong association between nutrition and DNA methylation (e.g. J Nutr Biochem 2012; 23: 853); in fact there is a new journal “Environmental Epigenetics” with articles on such things as epigenetics as a mediator between air pollution and preterm birth, chemical exposures and autism, etc.
  • So, bottom line: there may well be genetic variants leading to obesity and short stature, and these may predict a small part in their determining decreased SES. But overall, I think this study reinforces that the relationship between BMI or height with SES is predominantly a social one: there are social biases, conscious or not, which directly affect the overall ability of overweight women and short men to achieve their full human potential, and this is reflected in their lower SES. And that lower SES may well contribute to higher BMI. I bring up this article because there are more and more studies using mendelian randomization, including a recent one on HDL-cholesterol, and that it seemed reasonable to think about the potential value and shortcomings of this technique.

Primary Care Corner with Geoffrey Modest MD: Zika Virus Review

27 Apr, 16 | by EBM

By Dr. Geoffrey Modest

There was a good review of the Zika virus from the CDC in a recent issue of NEJM (see DOI: 10.1056/NEJMra1602113​).

Details:

  • Epidemiology: flavivirus, discovered in 1947 in Zika rainforest in Uganda, transmitted by Aedes africanus mosquit​o, with seroprevalence of 6.1% back then (i.e., lots of human transmission), and with documented wide geographic distribution in Africa and Southeast Asia. Though only rare reports of cases of human illness over the next 57 years
  • 2007: Outbreak in Micronesia (State of Yap) with 5000 infections in population of 6700
  • 2013-4: Big outbreak in French Polynesia involving 32,000 people. Some cases of Guillan-Barre. Other outbreaks in Pacific Islands, not much more in Southeast Asia
  • 2015: outbreak in Brazil, then widely spread with up to 1.3 million suspected cases, with >4300 cases of microcephaly (now officially caused by the Zika virus, per the CDC – i.e., not from insecticides, herbicides, etc.)
  • By March 2016, spread to at least 33 countries and territories in the Americas. The strains of Zika in the Americas is of the Asian genotype, similar to the outbreaks in Yap, French Polynesia, though overall Zika virus has conserved its genetic structure very well over time [which is good in terms of diagnostic testing, development of vaccines, and understanding expected symptoms/sequelae]
  • Transmission: many different Aedes mosquitoes likely involved in nonhuman transmission in nonhuman animals in Africa; A. aegypti and (to lesser extent) A. albopictus, have been involved in nearly all human cases. (? If A. hensilli and A. polynesiensis were involved in Yap)
  • egypti likely to be the bad actor overall: it feeds primarily on humans, often bites multiple people in a single blood meal, has an almost imperceptible bite, and lives in close proximity to humans.
  • Both A. egypti and A. albopictus bite primarily during the daytime, and are widely spread throughout the tropical and subtropical world. A. albopictus is more widely distributed in the eastern US and Hawaii
  • But it seems that A. albopictus does not cause much problem so far, with dengue or Zika, except in Hawaii, which has had some dengue outbreaks.
  • Though Zika has been identified in malaria mosquito vectors, they seem to have low potential for transmission for Zika
  • Non-mosquito transmission– pretty definitely: mother-to-fetus (found in amniotic fluid in fetuses with cerebral abnormalities, found in brain tissue), but in the very few cases of peripartum transmission, not a significant problem; sexual transmission: in one case occurred before the onset of symptoms. Virus identified in sperm up to 62 days after onset of symptoms. Though blood transfusion transmission not reported, it is likely given this happens with other similar flaviviruses. Zika virus has been recovered in breast milk
  • Clinical presentation: unclear incubation period, but likely <1 week (if similar to other flaviviruses). In a volunteer, febrile illness after 82 hours of subcutaneous inoculation, viremia only when symptomatic. In Yap, about 20% of people become symptomatic: macular/pruritic rash (90%), short-term fever (65%), arthralgia/arthritis (65%), conjunctivitis (55%), myalgias (48%), headache (45%), retro-orbital pain (39%), edema (19%), vomiting (10%). Also, hematospermia, dull hearing, swelling of hands/ankles, subcutaneous bleeding. In French Polynesia Guillan-Barre was found in 38 people of 28,000 who sought medical care. Also meningoencephalitis and acute myelitis. In terms of fetal outcomes, much is inferred from other infections (CMV, rubella). Most prominent effects in first trimester infections, though there is evidence that microcephaly can occur with infection late in second trimester or early third. Recent data does suggest that most infections associated with microcephaly occur between 7-13 weeks of gestation. In Brazil, fetal abnormalities were detected by ultrasound in 29% of women infected with Zika (though, note: ultrasound is not a very sensitive method to detect microcephaly). Fetal loss/death has been found in infections from 6-32 weeks gestation. Ocular anomalies in 35%
  • Diagnosis: detection of viral nucleic acid in the blood is definitive, but that may be transient (mostly up to 1 week), though viral RNA has been detected in serum of pregnant women with infected fetus 10 weeks after infection. The virus-specific IgM antibody develops at the end of the first week, but may reflect cross-reaction with other flaviviruses (e.g. dengue, yellow fever). Plaque-reduction neutralization testing provides more specific Zika virus assessment.
  • Prevention: use of mosquito repellent, permethrin treating of clothing, bed nets, window screens help.

 

Primary Care Corner with Geoffrey Modest MD: Life Expectancy and Income

26 Apr, 16 | by EBM

By Dr. Geoffrey Modest

A massive data analysis was done looking at the relationship between income and life expectancy in the US from 2001-2014 (see doi:10.1001/jama.2016.4226).

Details:

  • Income data was from 1.4 billion de-identified tax records from 1999-2014, using pretax household earnings as measure of income. For those not filing a tax return, they looked at the sum of all wage earnings (W-2 forms) plus unemployment benefits. Income was adjusted to 2012 dollars
  • Mortality data from Social Security Administration tax records.
  • Assessed life expectancy at 40 years of age by household income percentile, sex, geographical area.
  • Evaluated 1,408,287,218 person-year observations for people aged 40-76. Mean age 53. mean household earnings $61,175/yr
  • For people under 63 yo, mortality rates were calculated based on income percentile 2 years earlier (they chose 63 yo, since income rates after age 61 correlate less well with earlier earnings). They used models to estimate mortality rates after age 76, and were adjusted for racial/ethnic composition of income groups. local variations of life expectancy were assessed based on zip code of where their income tax returns were mailed

Results:

  • ​Total of 4,114,380 deaths among men ​(mortality rate of 596.3 per 100K); 2,694,808 deaths among women (mortality rate of 375.1 per 100K)
  • Higher income was associated with greater longevity throughout the income distribution, based on the income distribution at age 40
    • Gap in life expectancy between the richest 1% and the poorest 1% was 14.6 years (14.4-14.8) for men and 10.1 years (9.9-10.3) for women
    • Gap between men and women was more pronounced in the bottom 1% [6.0 years (5.9-6.2)] vs in the top 1% [1.5 years (1.3-1.8)]
    • The overall gap was non-linear by $$: going from the 95th to 100th% ($224K to $1.95M annual income) had much smaller gains in life expectancy than going from the 10th to 15th% ($14K to $20K)
  • Inequality in life expectancy increased over time
    • Between 2001 and 2014, life expectancy increased by 2.34 years for men and 2.91 for women in the top 5% of the income distribution, but only 0.32 years for men and 0.04 years for women in the bottom 5% (p<0.001 for both sexes)
  • ​Life expectancy in low-income people varied substantially across local areas. In the bottom income quartile, life expectancy varied by approx 4.5 years between areas with the highest and lowest longevity
    • Changes in life expectancy between 2001 to 2014 ranged from gains of more than 4 years to losses of more than 2 years across areas
    • The variations formed “belts” in their maps of the US: the states with the lowest life expectancy in the bottom income quartile were from Michigan to Kansas (including Ohio, Indiana, Kentucky, Tennessee Arkansas, Oklahoma); the states with the highest were California, New York and Vermont. In terms of the highest income quartile, the lowest life expectancy (<85.3 yrs) were Nevada, Hawaii,and Oklahoma; and the highest life expectancy (>87.6 yrs) was in Utah, Washington DC, and Vermont
    • Also, trends in life expectancy over time varied: greatest increase was in Massachusetts (>0.19 years annually) for those in the bottom income quartile;  and life expectancy decreased (losing > 0.09 years annually) in Alaska, Iowa and Wyoming
  • Local geographical differences in life expectancy in the lowest income quartile were significantly correlated with health behaviors (e.g., negatively with smoking, r=-0.69, p<0.001 and obesity, r=-0.47, p<0.001; but positively with exercise rates, r=0.32, p=0.004), but not with access to medical care for the lowest income group (assessing health insurance spending, quality of primary care), physical environmental factors (air pollution, lack of access to healthy food; though of note, there was almost statistically significant differences between rich and poor life expectancy in those areas with the most residential segregation), income inequality/social cohesion (income inequality seems to be within income groupings, but they also looked at % religious, % Black), or labor market conditions (unemployment rates, changes in population, changes in size of labor force).
    • Life expectancy for low-income people was positively correlated with the local fraction of immigrants (r=0.72, p<0.001), median home values (r=0.66, p<0.001), fraction of college graduates (r=0.42, p<0.001), population density (r=0.48, p<0.001) and local per-capita government expenditures (r=0.57, p<0.001)
    • Life expectancy for high-income people was less variable, but notable for exercise rates (r=0.46, p<0.001); negatively correlated with Medicare expenditures (r=-0.55, p<0.001) but positively with index of preventive care (r=0.55, p<0.001)
    • Data from the National Center for Health Statistics (NCHS) show that the majority of the variations in those with low SES are related to medical causes (heart disease, cancer) vs external causes (accidents, suicide, homicide)

A few perspective issues:

  • ​As a point of international reference, the life expectancy for men in the bottom 1% income in the US is similar to the mean life expectancy for men in Sudan and Pakistan; men in the top 1% in the US have higher life expectancy than the mean for any other country
  • As an internal point of reference, the 10-year gap in life expectancy between the top and bottom 1% in the US is equivalent to the decrement in life expectancy attributable to lifetime smoking
  • ​From NCHS statistics, the difference in the improved life expectancy for the rich over the poor (around 3 years, in comparing top and bottom 5%) is equivalent to the increase in life expectancy at birth if all cancer deaths were eliminated (3.2 years)

So, some points:

  • This is a quick and dirty, but large and powerful study which gets into some of the details of health inequalities in the US.  The measurements of income, based on tax returns mostly, is undoubtedly missing much, especially from the low income group (typically still very low incomes but from under-the-table jobs, unreported income by undocumented workers, etc.) and from very high income group (offshore shell companies, legal and illegal shenanigans to hide income). Also, the mortality rates in those >76 yo is based on mathematical modeling/estimates. And some of the health outcome differences between the income groups is no doubt affected by local circumstances (even the rich in some areas are exposed to the same environmental exposures as the poor, perhaps some of the same food quality, cultural issues such as types of food eaten or priority to do exercise, or access to the highest tech medical care). Some support for this last argument is that the poor do better if living in areas of higher education and more affluence (higher median home values, etc.), which may represent certain local cultural values, such as less smoking (and, in our poorer communities in Boston, there is clearly less smoking, for example, than in many other areas of the country). Also, the finding that the trend in communities with more geographical separation to have more longevity differences also supports that the more isolation of the different income communities from each other tends to make the longevity lines clearer.
  • Perhaps some of the most poignant and socially-relevant findings are:
    • The gaps are increasing, with some of the poorer communities experiencing a decrease in life expectancy
    • ​One other inequity (noted in other blogs) is the regressive nature of our Social Security system: the rich live longer, so get social security payments for a longer time
    • Although just reading numbers sometimes makes it hard to really understand the impact of the differences, the comments about equivalency of these differences with lifetime smoking or zero deaths from cancer really put this in perspective
    • Also, the international perspective I think is quite useful. again, the US does quite poorly overall in life expectancy, except for the very wealthy, and, per several other blogs (see below), this really does correlate with our remarkably poor social systems (or even “safety nets”) as compared to essentially every other westernized, resource-rich country (as noted in the blog below, the book The American Health Care Paradoxshows quite well that we spend huge amounts on “health care”, but unlike other countries, the vast majority of this is for medical care instead of social or public health programs that promote public health). It is telling in the above JAMA study that longevity was associated with local governmental expenditures, which may reflect more social or public health programs
  • I am pretty surprised that access to care was not so related to life expectancy. Not sure how to explain that, since so much literature has come out, for example, about cancer or heart disease outcome differences by income or by race/ethnicity. One interesting finding was that Medicare coverage at age 65 did not affect longevity, and one might think that there would be improved health care access when getting Medicare; though I certainly see many patients who do not qualify for Medicare (did not work the 40 quarters, or were undocumented workers who may have worked 2 jobs for way more than 40 quarters…). Perhaps some of the issue is that they really only looked at income and did not do a detailed analysis by race/ethnicity (many of their analyses did include race- and ethnicity-adjusted life expectancy, but I’m not really sure what that means, and they did not breakdown any of the more specific data above by these important demographics).

See http://blogs.bmj.com/ebm/2016/02/24/primary-care-corner-with-geoffrey-modest-md-increasing-disparities-in-life-expectancy/​ which goes into detail about the increasing life expectancy disparity paralleling income disparity, with a pretty long list of comments, which I will not repeat here, which also includes an article on mortality in younger people, finding a much higher mortality in the US than in other countries.

Primary Care Corner with Geoffrey Modest MD: FDA Changes Metformin Guidelines

26 Apr, 16 | by EBM

By Dr. Geoffrey Modest

The FDA is revising its warnings about the risks of metformin in those with chronic kidney disease (CKD) — See http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm494829.htm , or http://www.fda.gov/downloads/Drugs/DrugSafety/UCM494140.pdf​ for more info), “Requiring manufacturers to revise labeling of metformin-containing drugs to indicate that these products may be safely used in patients with mild to moderate renal impairment”. Although I have sent out many blogs on metformin, I think it is such an important drug that I am sending out this as yet another one. The new FDA labeling recommendations:

  • Before starting metformin, check eGFR
  • Metformin is contraindicated in patients with an eGFR <30
  • Starting metformin in those with eGFR 30-45 is NOT recommended
  • Obtain eGFR at least annually, more frequently if increased risk of renal dysfunction (e.g., elderly)
  • If a patient is on metformin and the eGFR falls to below 45, assess the benefits and risks of continuing treatment [not further defined…]. Discontinue metformin if eGFR falls to <30
  • Discontinue metformin at time, or before, an iodinated contrast imaging procedure in patients with eGFR 30-60, in those with a history of liver disease, alcoholism, or heart failure; and in those getting intra-arterial iodinated contrast. Re-evaluate eGFR 48 hours after the procedure and restart metformin if renal function is stable.

So, a few comments:

  • It’s about time…..   See a few of my many blogs below on metformin, which highlight its incredible importance in diabetics, its current use in the US in those with CKD despite prior strict precautions, its general use in Europe where regulatory agencies suggest using it with decreasing doses depending on the degree of CKD, and even a blog which refers to its effects on the microbiome (and suggesting that these effects may be extremely important in metformin’s mechanism of action).
  • One concern with the above is that they suggest following eGFR more closely in the elderly. This is difficult, since (at least in our lab), eGFR is not reported in those >70: “eGFR values may not be accurate on patients greater than 70 years of age and are not calculated”. My approach is to continue with metformin in the elderly (even those in their 90s) if their creatinine seems “reasonable” — not easy to quantitate, but if their creatinine is below 1.5 or so, and throwing in some gestalt about their likely muscle mass…  But, bottom line, I have been doing this for years on a pretty large patient base and without any adverse effects. Though I often only use 500 mg/d (occasionally 250 mg/d), I find patients still get a huge effect from the med. And, by the way, given some issues with metformin tolerance (mostly GI), I often leave younger people on 500 mg once or twice a day, with really good effect (i.e., I rarely crank it up to the 1000 bid dosage, since my sense from the not-so-great literature, and my experience, is that 500-1000 mg total per day gives about 75-85% of the efficacy of the maximal dose)

http://blogs.bmj.com/ebm/2015/07/23/primary-care-corner-with-geoffrey-modest-md-metformin-ckd-and-death/​ which looked at 813 metformin users with severe CKD (creatinine >6 mg/dL) who did NOT have increased mortality if the dose of metformin was low (<= 500 mg/d)

http://blogs.bmj.com/ebm/2015/01/23/primary-care-corner-with-geoffrey-modest-md-metformin-in-renal-failure/ is a more systematic review of metformin use in those with less severe CKD, suggesting an algorithm for metformin dose adjustment

http://blogs.bmj.com/ebm/2015/01/28/primary-care-corner-with-geoffrey-modest-md-heart-failure-microbiome/ looked at the effect of red meat on the microbiome (increasing TMAO levels, which are strongly pro-atherogenic) and also highlighting that metformin induces positive changes in the microbiome which decrease insulin resistance

Primary Care Corner with Geoffrey Modest MD: Coffee and Decreased Colon Cancer

25 Apr, 16 | by EBM

By Dr. Geoffrey Modest

An interesting case-control study found that coffee consumption is associated with a decreased risk of colorectal cancer CRC (see Cancer Epidemiol Biomarkers Prev. 2016; 25(4): 634).

Details:

  • 5145 colorectal cancer (CRC) cases were compared with 4097 controls from the Molecular Epidemiology of Colorectal Cancer (MECC) study, a population-based study in northern Israel, beginning in 1998.
  • Mean age 70, 52% male, 60% Ashkenazi/20% Sephardi/13% Arab, 7% first-degree relative with CRC, 58% consuming >= 5 vege servings/d, 35% “sports activity”, 10% current/29% former smokers, 25% daily aspirin, 74% colon/23% rectal cancer). Mean coffee intake in controls was 2.0 servings/d. There were significant differences in both quantity and types of coffee drunk by the different ethnic groups
  • Results, comparing coffee drinkers to nondrinkers:
    • Coffee consumption was associated with 26% lower odds of developing CRC [OR 0.74; (0.64–0.86), p<0.001], controlling for known risk factors (age, sex, ethnicity, vegetable consumption, sports participation, statin use, daily low-dose aspirin, smoking status and family history). Additional controlling for total daily liquid consumption or total calorie consumption did not significantly affect these results.
    • Decaffeinated coffee had 18% lower odds [OR 0.82 (0.68–0.99), p=0.04]
    • Boiled coffeehad 18% lower odds [OR, 0.82 (0.71–0.94), p=0.004].
  • There was a dose-response curve, with p<0.001 for the trend. Compared to <1 serving/day:
    • Intake of 1 to <2 servings/dayhad 22% lower odds for developing CRC [OR 0.78 (0.68–0.90), p< 0.001]
    • 2 to 2.5 servings/dayhad 41% lower odds [OR 0.59 (0.51–0.68), p<0.001]
    • ​>2.5 servings/dayhad 54% lower odds [OR, 0.46 (0.39–0.54) p< 0.001]
  • ​Also, there was an overall inverse relationship between CRC and vegetable consumption (p<0.001), daily low-dose aspirin (p=0.03)​, sports participation (p<0.001), and direct relationship with smoking status (p<0.001), and sex (p<0.001).

So, a bright spot for me given my coffee consumption. But does this all make sense?

  • Coffee has many bioactive components, including chlorogenic acids (powerful anti-oxidants, which modify gene expression and inhibit DNA methyltransferase), polyphenols (which have anti-oxidant and antiproliferative effects, and induce cell-cycle arrest in colorectal cell lines), melanoidins (which may increase colon motility), diterpenes (which may be anticarcinogenic by enhancing defense systems against oxidative stress), and caffeine (may also be anti-oxidant, and limit growth of human colon cancer cells). The actual effects of these constituents is not so clear in humans, since much of the above is through in vitro analyses.
  • Coffee also reduces bile acid secretion
  • Coffee also modifies the microbiome: there are some data from small human experiments that coffee does change the fecal microbiome favorably, increasing Bifidobacterium spp. Another study, presented only as an abstract, looked at the microbiome of coffee consumers in an effort to see how coffee might lower the risk of diabetes, finding in rats that those on high fat diet had less weight gain and less insulin resistance, and more lactobacillus in their microbiome (FASEB journal; 2013; 27:951)​
  • And, coffee does change some bowel functions, such as increasing intestinal motility and stool output
  • Though, it should be added that there are some differences in the above components, depending on the actual coffee beans used, the degree of roasting, and the brewing technique
  • And, although the epidemiologic data are not entirely consistent, most suggest the above association (e.g., see Public Health Nutrition 2013; 16: 346, a meta-analysis finding an overall 15% reduction in colon but not in rectal cancer, especially in Europe and in women), and several show a decreased incidence of liver cancer. Of note in the Israeli study, there was a lower incidence of both colon and rectal cancer, though less impressive for rectal cancers.
  • This study did look at aspirin consumption, which is associated with lower colon cancer risk (see: http://blogs.bmj.com/ebm/2015/09/09/primary-care-corner-with-geoffrey-modest-md-low-dose-aspirin-and-colon-cancer-risk/ ), but did not assess NSAID use, also associated in many studies with lower risk of colon cancer
  • Though a case-control study can never be definitive, the constellation of biological plausibility, the consistency with other observational studies, the magnitude of the effect, and the dose-response curve (the more coffee, the better) tends to support the conclusion. There could, however, be reverse causation: did those with GI symptoms from an early cancer, for example, stop drinking coffee?  It would really be hard to have a real randomized controlled study, where thousands of people were randomized to drinking varying amounts of coffee and followed for 5-10 years. So, these case-control studies are about as good as we can get….

For relevant blogs, see:

http://blogs.bmj.com/ebm/2015/12/07/primary-care-corner-with-geoffrey-modest-md-drink-coffee-and-live-longer/ reporting on the Nurses’ Health Studies, finding a decrease in total mortality in coffee-drinkers

http://blogs.bmj.com/ebm/2015/03/12/primary-care-corner-with-geoffrey-modest-md-coffee-and-decreased-coronary-artery-calcium/ , a Korean observational study, finding a pretty strong association between coffee-drinkers and decreased coronary artery calcifications

There are also several blogs in the BMJ website on chocolate (another personal addiction), also rich in polyphenols. One showed clinical improvement in those with peripheral artery disease (see http://blogs.bmj.com/ebm/2014/09/29/primary-care-corner-with-geoffrey-modest-md-dark-chocolate-helps-with-peripheral-arterial-disease-pad/ )

Primary Care Corner with Geoffrey Modest MD: WHO Global Report on Diabetes

22 Apr, 16 | by EBM

By Dr. Geoffrey Modest

There was a pretty striking and disturbing report by the World Health Organization on the global burden of diabetes (see http://apps.who.int/iris/bitstream/10665/204871/1/9789241565257_eng.pdf?ua=1​ ).

Details:

  • Estimated 422 million adults were living with diabetes in 2014 (vs 108 million in 1980), rising from 4.7% to 8.5% of the adult population in the world (e. 1 in 11).
  • There were 1.5 million deaths in 2012 from diabetes and 2.2 million additional deaths associated with high blood sugar(largely from cardiovascular disease). 43% of these deaths occurred in people <70 yo
  • Presence of diabetes is increasing more rapidly in low- and middle-income countries, including the % of deaths attributable to high blood sugar or diabetes. The most dramatic increase in diabetes from 1980-2014 was in the Eastern Mediterranean region, which increased from 6% to almost 14%, with most other regions going from 4-5% to around 8%. The region with the highest % of deaths from high blood sugar was the Eastern Mediterranean region, followed closely by the region of the Americas, South-East Asia region and Western Pacific region. The lowest were the African region and European region.
  • Much of this change is associated with an increase in risk factors, esp. overweight/obesity. In 2014, more than 1 in 3 adults are overweight and 1 in 10 obese. In 2010 27% of women and 20% of men were “insufficiently physically active”. And in adolescents, 84% of girls and 78% of boys did not meet minimum requirements for the stricter definition of physical activity for their age. Of note, the regions with the most overweight (BMI>25) were the Americas and European. And there was a trend to increasing overweight going from low-income to high-income countries
  • Not surprisingly, the ravages of diabetes lead to huge economic losses to the patients, their families, health systems (often already overburdened) and national economies (loss of work and wages). A study found that expected losses to Growth Domestic Product from 2011 to 2020, including both direct and indirect costs, will be on the order of $1.7 trillion (US $$), with $900 billion for high-income and $800 billion for low- and middle-income countries
  • The WHO has the usual recommendations to prevent diabetes: exercise, eating healthily, not smoking, controlling blood pressure and lipids. But there need governmental public health initiatives/appropriate reinforcing policies, with population-based prevention
  • Effective diabetes management requires early diagnosis, requiring good access for screening for blood sugar, as well as for microvascular and macrovascular damage, and access to appropriate meds
  • Exercise recommendations are: at least 150 min of moderate-intensity aerobic exercise/wk (brisk walking, jogging, gardening) or at least 75 min of vigorous exercise/wk. older adults, same amount but to include balance and muscle strengthening activity. For those 5-17 yo, at last 60 min of moderate to vigorous exercise per day.
  • They highlight Mexico, where the prevalence of overweight/obesity is >33% for kids and 70% for adults, has the highest prevalence of diabetes, and the highest per capita consumption of soft drinks in the world. in Jan 2014, the govt implemented a tax on drinks with added sugar, increasing their price 10%, and preliminary analysis suggests an 11.6% decrease in the quantity consumed
  • Another was in Senegal, where the government used mobile technology to text people and encourage healthy eating during Ramadan, which the next year led to 12000 self-recruited users, highlighting interest and demand. And the WHO document has other examples of creative societal approaches

Why is diabetes increasing in low- and middle-income countries? One of the WHO medical leaders was interviewed on NPR and commented that probably the biggest change in these countries is the dramatic increases in migration from rural areas to cities. This urbanization leads to both decreasing exercise (less manual labor than in rural areas, less walking, more cars), eating more processed foods/fewer basic vegetables/etc., and the resultant increasing obesity.

But, needless to say, this is a huge shift, with noncommunicable diseases overtaking infectious diseases as the major causes of death in middle- to low-income countries, and with truly profound long-term implications for the people and societies overall. There are some good examples of people and governments taking public health initiatives to try to deal with this problem, and there is compelling urgency to develop and expand such initiatives…

See http://blogs.bmj.com/ebm/2015/01/21/primary-care-corner-with-geoffrey-modest-md-community-wide-rural-cardiac-health-program/ for a really interesting blog on a small rural poor county in Maine which has developed a long-lasting and effective multi-dimensional community-wide cardiac health program

Primary Care Corner with Geoffrey Modest MD: CABG for Ischemic Cardiomyopathy?

21 Apr, 16 | by EBM

By Dr. Geoffrey Modest

The NEJM just printed the recent study of patients with ischemic cardiomyopathy and the utility of CABG (see DOI: 10.1056/NEJMoa1602001), the STICHES trial (Surgical Treatment for Ischemic Heart Failure Extension Study).

Details:

  • 1212 patients from 22 countries who had coronary artery disease (CAD), ejection fraction (EF) <35% and with coronary arteries amenable to bypass, were randomly assigned to CABG plus optimal medical therapy vs medical therapy alone. Patients recruited from 2002-2007
  • Mean age 60, 12% female, 36% Latino or nonwhite/64% white, BMI 27, 76% with MI, 59% hyperlipidemia, 59% hypertension, 39% diabetes, 86% NYHA class 2 or 3, mean systolic BP 120 mmHg. EF 27%, left main disease in 3%, proximal LAD 67%. 1 vessel disease 22%/2-vessel 38%/3-vessel 37%. No comment on QRS interval.
  • Median follow-up 9.8 years

Results:

  • Primary outcome (death from any cause):
    • 359 (58.9%) in the CABG group and 398 patients (66.1%) in the medical therapy group, a 16% reduction [HR 0.84 (0.73-0.97), p=0.02]
  • Secondary outcomes:
    • Death from cardiovasc cause: 247 patients (40.5%) in CABG group and 297 (49.3%) in med therapy group, a 21% reduction [HR 0.79 (0.66-0.93), p=0.006]
    • ​Death from any cause or hospitalization from cardiovasc causes: 467 patients (76.6%) in CABG group and 524 (87.0%) in med therapy group, a 28% reduction [HR 0.72 (0.64-0.82), p<0.001]​
  • The above results translate to an incremental median survival benefit of about 18 months (1.44 years), and prevention of 1 death from any cause in 14 patients getting a CABG, or one death from a cardiovasc cause for every 11 patients.
  • About 18% had automatic implantable cardioverter-defibillators (AICDs), same in both groups
  • Adverse events: for CABG: 6% had to return to OR, 2% mediastinitis, 8% other infection, 3.6% death within 30 days (vs 1.2% in med-group), 5% pacemaker (vs 0.5% med-group), 1.8% stroke (vs 0.2% med-group)

A few points:

  • The current study is an extension of the STICH trial, which evaluated the results at 56 months, finding no significant difference between the CABG and medical-therapy group in the rate of death from all causes at that time, though the rates of death from cardiovasc causes and death from any cause or hospitalization were significant.
  • Looking at the curves of event rates, the death-from-any-cause and the death-from-cardiovascular-causes curves were similar, with diverging lines showing increasing benefit of CABG over the first 5-6 years, then parallel curves over the rest of the study, suggesting lasting benefit
  • Was this medical therapy optimal? Approx 90% were on ACE-I or ARB, 85% on statin, 90% of b-blocker, 70% on loop or thiazide diuretic, 50% on potassium-sparing diuretic, 20% on digoxin, 84% on aspirin. So, one could argue that there was not full utilization of an aldosterone antagonist, now part of the optimal medical management (of note, the RALES trial, published in 1999, showed a 30% mortality reduction and a 35% lower frequency of hospitalization for worsening heart failure)
  • Was the surgical therapy optimal? Seems pretty good to me. 91% received at least one arterial conduit. The 3.6% perioperative mortality seems reasonable (per the editorialists), though they do suggest using the Society of Thoracic Surgeons risk calculator, which shows the rate can be as low as 0.7% in healthy/low risk, to >7% or more if multiple comorbidities. See http://riskcalc.sts.org/stswebriskcalc/#/calculate

So, to me, this study raises several issues for us and our patients:

  • Though overall this study seems pretty well done, patients need to understand and place their own value on the trade-off of CABG-related early mortality and a reasonably high morbidity (return to OR, infection, stroke) pretty much right away, but living an average of 18 months longer. From the above study, one in 23 will have a really bad short-term outcome (death or stroke).
  • As with any study with long-term follow-up, optimal therapy usually changes by the time the study is published. So, though it seems that the surgical component is quite good (though minimally-invasive surgery did not happen back then), medically adding an aldosterone antagonist more aggressively might have changed the results some, as above. Also, back then we did not have meds for more aggressive lipid management
  • The other real issue, I think, is that they did not use AICDs much in either of the groups. There have been a slew of studies, but the MADIT-2 study was done in 2002, at the time of recruitment for the STICHES study, and showed that after only 20 months, the mortality rates with an AICD was 14.2% vs 19.8% in the medical group alone, a 31% mortality benefit. (By contrast, the mortality benefit from the CABG group in STICHES was not statistically significant until about 6 years or more). The current recommendations for AICD in those with ischemic cardiomyopathy are an EF <35% and NYHA class 2-3, basically the same as in the STICHES study
  • So, what is one to do? It seems that the most logical course at this point is to really maximize medical therapy, including, I think, using more potent statins if needed to lower the LDL to <70, or even consider the PCSK9 inhibitors; making sure that all patients, as tolerated, are on ACE/ARB, b-blockers, an aldosterone antagonist, aspirin, loop diuretics, and other risk factor management. And, if this is insufficient in terms of EF and symptoms, consider an AICD and/or surgery to be discussed with patient. Given the lack of direct comparison between AICD and CABG, my bias would be to the less invasive one.

Primary Care Corner with Geoffrey Modest MD: Diabetes DPP-4 Inhibitors and the Risk of Heart Failure

20 Apr, 16 | by EBM

By Dr. Geoffrey Modest

The FDA recently came out with a safety alert about 2 DPP-4 inhibitors and the increased risk of heart failure (released 04/05/2016) (see http://www.fda.gov/Safety/MedWatch/SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm494252.htm?source=govdelivery&utm_medium=email&utm_source=govdelivery​ ).

Details:

  • Saxagliptin and alogliptin were singled out because of 2 clinical trials in diabetic patients with heart disease. In each trial more patients taking these meds were hospitalized for heart failure vs those on placebo
    • In the saxagliptin trial: 3.5% were hospitalized for heart failure (vs 2.8% on placebo). Risk factors included prior heart failure or kidney disease
    • In the alogliptin trial: 3.9% were hospitalized for heart failure (vs 3.3% on placebo)
  • Recommendation by the FDA:
    • We should consider stopping these drugs if the patient develops heart failure
  • I believe these trials (not cited specifically by the FDA) were the ones in the blog http://blogs.bmj.com/ebm/2013/09/10/primary-care-corner-with-geoffrey-modest-md-new-diabetes-drugs-dpp-4-is-lower-a1c-not-cardiac-events/​ . The saxagliptin one was clearly the one in the blog, the alogliptin one probably was (reviewing the article, they did not have any breakdown for developing heart failure in the article or the supplementary materials. my guess is that the FDA got their hands on more specific data….)

Ironically, this FDA safety alert was published 2 weeks after the New England Journal of Medicine published a multicenter observational study of these meds and heart failure (N Engl J Med 2016;374:1145), looking at health care data from Canada, UK, and the US, using a nested case-control design (matching each heart failure case to 20 controls from the same cohort), finding:

  • 1,499,650 patients involved, 29,741 hospitalized for heart failure
    • For those without history of heart failure: HR 0.86 (0.62-1.19)
    • For those with history of heart failure: HR 0.82 (0.67-1.00)
    • ​No difference between those on DPP-4 inhibitors or GLP-1 agonists

So, how does this affect us?

  • As I have mentioned many times in the past, I do not see any big benefit from using these drugs: the A1C benefit is not very large (about 0.3% in several studies, including the above 2 studies as well as the sitigliptin one in the additional blog below), DPP-4 is a pretty ubiquitous enzyme which deactivates lots of different bioactive peptides (i.e., DPP-4 inhibitors are hardly magic bullets), and it is really not so surprising that there may be significant collateral damage.
  • The pretty small absolute increases in hospitalizations for heart failure (0.6-0.7%) is likely the tip of the iceberg. Diabetics get lots of heart disease (the vast majority of diabetics, in the 70-80% range, die from heart disease), most heart failure (I think) is treated as an outpatient and therefore not showing up in statistics for hospitalizations, and the mortality from heart failure may well be higher in patients not enrolled in a study, where study personnel tend to follow patients closely and patients have easy access to them (this may be especially true in areas of the country where there is not ready access to any high quality outpatient care or hospitals.)
  • Since so many diabetic patients develop heart failure just because of their diabetes and other risk factors, I think it is important that the FDA brings this drug association to our attention (i.e., one would not think necessarily that the drug caused the heart failure). Not exactly sure what to do with the recent NEJM article, though it was a retrospective matching of patients from many trials (albeit a pretty big one)
  • So, the DPP-4 inhibitors are not on my list of meds to use, even in patients without known underlying heart or kidney disease….. [Though, I should add, that I do use GLP-1 agonists, which have a much more dramatic effect on A1C levels, are more physiologic than exogenous insulin or sulfonylureas, and are very specifically targeted to meal-related endogenous insulin release (the “incretin” effect). So, though it might surprise you, I am really not against all new drugs…]

http://blogs.bmj.com/ebm/2015/06/24/primary-care-corner-with-geoffrey-modest-md-dpp-4-inhibitors-and-cardiovascular-outcomes/ , which looks at sitigliptin (not one of the ones singled out by the FDA), showing very small effects on A1C levels, showing no increase in cardiovascular outcomes, but does bring up the point that for the minimal-effect on A1c, DPP-4 inhibitors block a ubiquitous enzyme on the surface of most cells and deactivates a variety of bioactive peptides.

Primary Care Corner with Geoffrey Modest MD: Hypertensive Treatment in Patients at Intermediate Risk

19 Apr, 16 | by EBM

By Dr. Geoffrey Modest

Recently 2 articles were in the NEJM from the HOPE-3 study, looking at cholesterol and blood pressure lowering in people at “intermediate-risk”, using a 2×2 factorial design. The blog Friday looked at the lipid arm of the study. This one will evaluate the hypertension group (see DOI: 10.1056/NEJMoa1600175). Study funded by Canadian Instit of Health Research and a drug company)

Details:

  • 12,705 people in 21 countries/6 continents (eligibility criteria: men ≥​ 55 yo/women ≥​ 65, with at least one of: elevated waist-to-hip ratio, history of low HDL, current or recent smoking, dysglycemia, family history of premature CVD, mild renal dysfunction; also women ≥​​60 yo with 2 or more of these criteria), without baseline cardiovascular disease (CVD) and intermediate risk (defined at annual risk of approx 1%, per the INTERHEART risk score, a score from 0-49, where the low-risk group had a score ≤ 9), randomized to candesartan 16mg plus hydrochlorthiazide 12.5mg vs placebo.
  • Median age 66, 46% female, 87% with elevated waist-to-hip ratio, mean BMI 27, 27% current/recent smokers, 6% diabetes, mean HDL 44.7 mg/dl, mean LDL 127.8, triglycerides 128.8, hs-CRP 2.0, INTERHEART risk score of 14.5, 29% Chinese/27% Hispanic/20% White/15% South Asian/2% Black, 22% on a BP med and 11% on aspirin)
  • Initial 4 week run-in period where people got both rosuvastatin and the BP med (candesartan/HCTZ) to make sure they were tolerated. the most common adverse effect was hypotension in 2%
  • First coprimary outcome: composite of death from cardiovascular causes, nonfatal MI, or nonfatal stroke; second coprimary outcome: the first one plus revascularization, heart failure, and resuscitated cardiac arrest. and, they added a secondary outcome: the second coprimary outcome plus angina with evidence of ischemia
  • Follow-up 5.6 years
  • Results:
    • Mean blood pressure at baseline was 138.1/81.9 mmHg, achieving decrease of 6.0/3.0 mmHg with active meds over placebo. By the end of the study 76.8% were taking the active med and 75.7% the placebo
    • Candesartan/HCTZ (vs placebo) resulted in:
      • First coprimary outcome: in 260 people (4.1%) vs 279 (4.4%), a nonsignificant 7% reduction [HR 0.93 (0.79-1.10, p=0.40)]
      • Second coprimary outcome: in 312 people (4.9%) vs 328 (5.2%), a nonsignificant 5% reduction [HR 0.95 (0.81-1.11, p=0.51)]​
    • ​The prespecified subgroup of those in the upper third of systolic pressure (>143.5/mean 154.1) did have significantly lower reductions:
      • The newly added secondary outcome: in 129 people (6.0%) vs 172 (8.3%), a 28% reduction [HR 0.72 (0.57-0.90)], with significant decreases in stroke (42%)
      • First coprimary outcome: 27% decrease (HR 0.73, 0.56-0.94)
      • Second coprimary outcome: 24% decrease (HR 0.76, 0.60-0.96)​
      • And, the trend of benefit in the upper third over the middle and lower thirds had p=0.02 for the first coprimary outcome and p=0.009 for the second.
    • And, there actually was a trend to harm in lowering the blood pressure in the lowest tertile (<131.5 mmHG; mean 122.2)
    • The curves pretty completely matched until about year 7, when there may have been some splaying of the curves [hard to interpret: there would need to be longer follow-up to see if that curve separation continued or this was a blip]
    • As with the lipid arm, there was no evidence of any significant interaction of the combination of meds: the results were essentially the same in those on BP meds plus placebo vs BP meds plus statin
    • No difference in new-onset diabetes, or pretty much any of the other adverse outcomes other than symptomatic hypotension, dizziness or lightheadedness in 3.4% vs 2.0% in placebo; but no difference in syncope, renal dysfunction, serum potassium levels, and no difference in serious unexpected reactions or permanent discontinuation of meds.

So, a few thoughts:

  • This population was at lower actual risk of CVD events than most prior trials [e.g. 2.2% in the SPRINT trial (see blog below) vs 0.8% for the first coprimary outcome and 0.9% for the second in the above HOPE-3 study]. These other trials also achieved more blood pressure reduction [e.g. 15/8 in comparing the intensive vs standard arms of the SPRINT trial, vs 6/3 in this study]
  • So, there may be several explanations for the lack of benefit in the HOPE-3 trial:
    • They looked at a relatively healthy group, who experienced fewer events than in most of the prior studies. So, it may be that a longer follow-up period would yield a different clinical outcomes. supporting evidence includes:
      • The fact that those at higher risk on subgroup analysis, the tertile with the highest initial blood pressure, did the best
      • The likely increasing benefit over time (at least to my looking at the graphs) could support achieving statistically significant benefit in extended follow-up
      • And, they used lower doses of meds than in other studies and achieved much lower blood pressure decreases, so the effect might well take longer to see
    • Again, my conclusion from this study, as with the prior blog on the lipid arm, is that lifestyle changes are really crucial. This population was overweight, had higher lipids than what probably really is “normal” (at least normal for a non-CVD prone population), had many smokers, how much exercise they did/not reported. And the actual achieved lowering of blood pressure was pretty small (and, per other studies, an amount achievable with pretty much any of the lifestyle interventions individually, let alone in combination)
    • In terms of meds, the study does support using meds (though secondary analysis, albeit pre-specified) for those with systolic blood pressure >143.5/mean 154.1​ (though I would still really promote the lifestyle changes first)
    • I think one of the confusing issues with blood pressure goals is that they may be different in patients taking meds vs through life-style changes. So, though the target BP in those on meds is perhaps around 140/80 in diabetics and somewhat lower in others, the targets may be even lower in those doing lifestyle changes (losing weight, lower sodium DASH-type diet, exercise, etc.). And, it is not just the adverse events/collateral damage from the meds. My guess is that the lifestyle changes actually reverse the abnormal physiology creating hypertension and are not just lowering the blood pressure through vasodilation, decreasing angiotensin, decreasing intravascular/interstitial fluid, etc., as is done by the meds.
    • So, bottom line, I think it is important not to be swayed by the guidelines promoting treatment only in those with pretty high blood pressures (>140 or >150 systolic), or be inured to the fact that so many of our patients do have supraphysiologic blood pressures but below medication-treatable. We should be pretty aggressive about even low levels of increased blood pressure (e.g., the old “mild” or “prehypertensive” groups, which as noted by a few epidemiologic studies, comprise the largest group with hypertension-related morbidity/mortality). And meds still play a role at least in intermediate risk patients with systolic BP >143 (or, mean of 154), if that level remains after optimizing lifestyle changes. And it still may be useful in those with lower levels, though we need a more extended follow-up period in those in this HOPE-3 study.

For other relevant blogs, see:

http://blogs.bmj.com/ebm/category/hypertension/ for a review of several blogs on hypertension

http://blogs.bmj.com/ebm/2015/11/19/primary-care-corner-with-geoffrey-modest-md-tighter-blood-pressure-control-the-sprint-trial/​ for an assessment of the SPRINT trial, which found significant benefit to a lower blood pressure target.

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