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 June, 2015

Primary Care Corner with Geoffrey Modest MD: Atrial fibrillation incidence in Framingham Study

29 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

The Lancet just published an article on the trends of atrial fibrillation (afib) over the past 50 years, based on the meticulous Framingham Study database (see They looked at 9511 study participants from 1958-2007, assessing afib prevalence, incidence, risk factors and mortality over 10-year blocks, stratified by sex, with 202,417 person-years of observation.  Results:

–There were 1544 cases of new-onset afib (821 men and 723 women)

–Between the first and last 10-year groupings

–The age-adjusted prevalence of afib quadrupled from 20.4 to 96.2 per 1000 person-years in men, and from 13.7 to 49.4 cases per 1000 person-years in women (p<0.001)

–The age-adjusted incidence of afib went from 3.7 to 13.4 per 1000 person-years in men, and from 2.5 to 8.6 cases per 1000 person-years in women​ (p<0.001)

–For EKG-diagnosed afib, found on routine exams during the study, the age-adjusted prevalence per 1000 person-years increased from 12.6 to 25.7 in men (p=0.007)and 8.1 to 11.8 in women (p=0.009) — of note, the pickup of afib over time decreased for those EKG-diagnosed during the routine exams and increased in those diagnosed by reviewing outside medical records (ie, more afib cases were diagnosed outside of the study and fewer during the study’s routine exams)

–But, there was no significant change in the age-adjusted incidence of afib by EKGs in the routine exams over time (increase was from 1.83 to 3.75 in men (p=0.06) and from 1.31 to 1.58 in women (p=0.23)

–The prevalence of most risk factors changed over time in those with new-onset afib, as follows:

–smoking decreased from 40.9% to 12.7%

–moderate/heavy alcohol decreased from 10.2% to 5.4%

–BMI>30 increased from 27.3% to 35.4%

–systolic BP>160mmHg decreased from 38.6% to 16.9%

–hypertension treatment increased from 22.1% to 59.8%

–diabetes increased from 5.7% to 19.6%

–prevalent heart failure decreased from 5.7% to 3.5%

–significant heart murmur decreased from 20.0% to 8.1%

–no significant change in prevalent MI

–overall summary of the effects of these changes in risk factors: the population attributable risks for incident afib plummeted for systolic blood pressure (47.3 to -2.1), heart failure (7.8 to 1.4), significant heart murmur (21.9 to 3.1) and EKG-LVH (from 10.4 to 1.8), but increased for BMI (12.0 to 16.9) and diabetes (3.2 to 5.9). though systolic blood pressure decreased so dramatically, the attributable risk for treated hypertension (from 9.8 to 19.5) suggests that there is a residual effect of hypertension even if treated, but much less so than untreated hypertension (and,presumably, hypertension prevention would be even better, eg by diet and exercise, ….)

–over a 20-year period after the development of afib, there was a 74% decrease in stroke (p=0.001) and a 25.4% decrease in mortality (p=0.003)

So, this study brings out some important points: many of the risk factors for afib are modifiable, and there have been some very impressive changes over the past 5 decades in several of them (especially the identification and treatment of hypertension, profound decreases in smoking, decreases in alcohol consumption), though these improvements have been somewhat overcompensated by the worsening in BMI and diabetes. The difference between prevalence and incidence are likely the result (for prevalence) of improved detection and the fact that people are living longer with afib (impressive decreases in strokes and mortality). For example, I do remember 20-30 years ago that it was not so clear that paroxysmal atrial fibrillation was as bad, and in need to treatment, as permanent afib. Or that it was safe to use anticoagulants in older and somewhat frail people.  Or that afib is as common as we now know it to be. The ascertainment of afib by the screening EKGs during the study is probably the most consistent marker of real changes in incidence, and these were largely nonsignificant over time (in men it was pretty close to significant, but the point is that this confirms that the increase in BMI and diabetes makes up for the improvements in the other risk factors in the incidence of afib — though, of course, there could be unidentified risk factors that play a role as well). And it is important to remember that the Framingham Study was based on white European immigrants.

Primary Care Corner with Geoffrey Modest MD: PPIs associated with MIs???

25 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

There was a recent article published online and hitting the press regarding a possible association between the use of proton-pump inhibitors (PPIs) and subsequent MIs (see DOI:10.1371/journal.pone.0124653​). The article was unusual in that it used a new technique for medical research: data-mining.  The bottom line is that if there were a real association between PPI use and MIs, this would have profound medical implications, since over 113 million PPI scripts are filled annually around the world with over $13 billion in sales, and in the US in 2009 21 million people filled at least one PPI script (the 3rd highest seller in the US).


–they looked at over 16 million clinical documents on 2.9 million people to assess PPI use and cardiovascular risk. There were 2 large data sources for the data mining (Stanford, and Practice Fusion, Inc) and one prospective source for a survival analysis. For the data-mining, they electronically evaluated clinic notes to search for patients with a GERD diagnosis, use of meds, and subsequent notes with diagnoses of cardiovascular disease

–overall, 32,363 patients were identified (mean age 55, 44% male, 50% white/27% unknown, 6% on clopidogrel). They identified a similar number of propensity score-matched controls. Mean followup 2.1 years


–overall, patients with GERD and on PPIs had an adjusted odds ratio of 1.16 (1.09-1.24) for association with MI. H2 blockers were not associated with increased cardiovasc risk

–survival analysis in the prospective cohort followed 5.2 years found a 2-fold increased association with cardiovascular mortality [HR=2.00 (1.07-3.78, p=0.031)]. No association was found for H2-blockers

–the PPI association was independent of the concommitant use of clopidogrel (used in only 5.9% of the population)

Although this data-mining is potentially a powerful tool to look at possible real associations, it is much less persuasive than the usual organized epidemiologic studies in terms of the types and quality of the data collected.  I bring up this article for several reasons.

  1. As I have highlighted in many of my older blogs (prior to the BMJ posts), I am very concerned about the over-use of PPIs. It used to be said that treatment for GERD (the most common indication for PPIs) could be either step-up (start with calcium, then go to H2 blocker, then to PPI as needed for symptom control) or step-down therapy (hit hard with PPI, then wean down to H2 blocker or calcium as tolerated). Not so surprisingly, it is rare that we in primary care actually do the step-down (after all, the patient is doing well on the PPI, there are lots of other issues to address in the confines of a quick primary care visit, so dealing with stepping down therapy just isn’t the priority). But PPIs are very powerful drugs, and there are known and potential problems associated with their use. The documented adverse effects include those associated with infections (since neutralizing the gastric acid eliminates one of the barriers to infection), malabsorption (since stomach acidity is important for some nutrient absorption), and potentially other issues. In terms of infections, there are pretty good data that prolonged PPI use is associated with increased risk of C. difficile infections, other enteric infections (salmonella, campylobacter), and community-acquired pneumonia (though data here are a bit mixed). Malabsorption has been found for iron, vitamin B12, magnesium and calcium, and the possible association of PPIs with hip, spine and wrist fractures (and an FDA warning about this). And one of the other concerns is the profound hypergastinemia, with potential risk of colon cancer (not found so far) and atrophic gastritis (which may be more common in those who are H Pylori positive, and could potentially lead to gastric cancer). And potential drug interactions (eg with clopidogrel) or other unusual adverse effects (acute interstitial nephritis)
  2. There are many plausible explanations of an association between PPI use and MI, including:

–PPIs inhibit the activity of dimethylarginine dimethylaminohydrolase (DDAH), which is responsible for 80% of the clearance of asymmetric dimethylarginine (ADMA), which is an endogenous molecule which inhibits the activity of nitric oxide synthase (NOS), which might impair endothelial function, increase vascular resistance, and promote inflammation and thrombosis. (This is the explanation offered by the authors of  the study, though the data are based on animal studies and cultured human endothelial cells)

–perhaps there is an associated chronic inflammatory state created by the increase in GI infections from PPI use, and this inflammatory state is associated with CAD (as found with rheumatoid arthritis, in a couple of new studies on psoriasis, perhaps in HIV??)

–or, the explanation I favor, is that a small but significant proportion of patients being treated for GERD in fact have GERD-like symptoms from CAD which are falsely ascribed to GERD. I did just see a patient who epitomized this: he had GERD-like symptoms, treated by H2 blockers which did not work, augmented to a PPI which still did not work. And on further questioning, his GERD symptoms were exertional. He will be getting worked-up for this, but he presents the case of approximately 10% of angina patients present with predominantly GI symptoms, including typical GERD symptoms (even more classical than his, with postprandial and not exertional symptoms). Therefore, the relationship of PPI to MI may be really from the error of using PPIs to treat GI symptoms which really come from the heart…

So, I bring this up mostly as a means to reinforce my sense that we use too many PPIs (my other big concern is overuse of NSAIDs, but that is a side issue here…). And also to highlight a new mechanism in medical research for doing quick and dirty epidemiologic associations through data-mining.​


Primary Care Corner with Geoffrey Modest MD: Tight diabetes control and cardiovasc disease– followup of the VA study

25 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

The Veterans Affairs Diabetes Trial (VADT) was one of the triumvirate published around 2009, along with the ACCORD and ADVANCE trials, which looked at intensive glycemic control and cardiovascular outcomes. These trials basically found that intensive control did not help and perhaps hurt: the ACCORD trial achieved an A1c of 6.4% in the intensive group vs 7.5% in the standard group, but had no benefit overall for nonfatal MI, nonfatal stroke and cardiovascular disease (CVD) deaths and actually had a 22% increased mortality after 3.5 years; the ADVANCE trial achieved an A1c separation of 6.3% vs 7.0% and found no macrovascular or mortality benefit after 5 years; and the VADT achieved an A1c separation of 6.9% vs 8.5% and found no CVD or mortality benefit after 5.6 years.  The current study looked at VADT outcomes in the 1791 military veterans after 11.8 years of followup, up to 5 years after the study ended (see N Engl J Med 2015;372:2197-206).


–followup data available for 92.4% of the participants, with 77.7% agreeing to additional data collection: annual surveys and periodic chart reviews

–participants in the followup study: mean age 61, 97% male, 63% white/15% hispanic/16% black, diabetes for 12 years, 43% with prior cardiac event, 73% hypertensive, BMI 31, BP 132/76, LDL 106 mg/dl, HDL 36 mg/dl, 15% current smokers

–the A1c separation between the groups decreased from 1.5 percentage points (6.9% vs 8.5%) during the trial, to 0.2-0.3 percentage points (8.0% vs 8.3%) by 3 years after the trial ended .


–intensive group had lower risk of primary outcome (time to first CVD event: heart attack, stroke, new or worsening heart failure, amputation for ischemic gangrene, CVD death) with HR 0.83 (0.70-0.99), p=0.04, after median followup of 9.8 years.   Absolute risk reduction of 8.6 major CVD events per 1000 person-years, though no reduction in CVD mortality

​–no reduction in total mortality, with HR 1.05 (0.89-1.25) after median followup of 11.8 years

Conclusion: no evidence of any mortality benefit (similar results in long-term followup of the ADVANCE trial). And, the benefit (significant 17% decrease in major CVD events with intensive therapy) did come at some expense (increased severe hypoglycemic events in 21.2% vs 9.9% in the standard therapy arm, and likely increased medicalization in the intensive arm). Of note, posthoc analysis of the ACCORD trial (see Diabetes Care 33:983–990, 2010) did find that the achieved A1c still had the strongest relationship with mortality, with a 1% increase in A1c being associated with a 20% increase in mortality.  But they found that higher mortality  was largely explained by which group the patients were assigned to, and not by the achieved A1c — specifically, those patients in the intensive control arm who could not achieve an A1c<7% had the highest mortality. This analysis suggested that the issue was flogging those in the intensive group who had hard-to-control diabetes with more meds (and, they happened to use a lot of rosiglitazone…).

So, how does one piece this all together?  My sense is that the data do support aggressive (even very aggressive) control for younger people who have easy-to-control diabetes. Those with harder-to-control diabetes don’t seem to do so well. This may be because of the long-term sequelae of diabetes and the attendant vascular changes, etc. Or the agents we use to improve the A1c (metformin is really great, but as we keep adding other agents, there seems to be less cardioprotection). Or because of an interplay of other comorbidities (eg a 5-year Italian study found that those with increasing numbers of comorbidities such as CAD, lung disease, GI problems, heart failure…. had no improvement in cardiovascular events with an A1C < 7 vs <6.5; those without comorbidities did better with the lower A1C threshold. — see Ann Intern Med. 2009; 151(12):854-860​). Therefore, my bottom line: treat those with long life-expectancy and easy-to-control diabetes as aggressively as I can do safely (as always, as a joint decision with the patient, and largely relying on life-style changes/nonphamacological management), but avoid just adding on more and more meds in those with difficult-to-treat diabetes to lower the A1c below 7.5-8% or so. My caveat on the last point: there are no good intervention trials looking at higher A1c levels than 8% (ie, is a target of 9% or 10% really worse than 8% in difficult-to-treat patients with longstanding diabetes and lots of comorbidities??), and I certainly have many patients who cannot achieve close to a level of 8% (the goal of diabetes management, to me, is not to treat the A1c level, but the patient with diabetes. Although it is important to lower the A1c as much as we can, I have several patients with blood sugars that range from 45-450 with a very high A1c despite many mutual attempts to improve diet and exercise, and where adding more insulin or other agents is in fact dangerous for them).​

Primary Care Corner with Geoffrey Modest MD: NYTimes – more drug company shenanigans

24 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

There was a rather poignant editorial in the NY Times on June 8th, pointing out 2 clever (though, incidentally, illegal) ways the drug companies help boost their profits by preventing/delaying the transition to generics:

  1. Buying off the competition — the example of Provigil: the drug company (Cephalon) first sued 4 generic drug makers for “patent infringement”, then paid them >$300 million to drop their challenge to Provigil’s patent (it turns out that Teva, one of the generic drug makers, ultimately bought out Cephalon, and the Federal Trade Commission made them pay $1.2 billion for Cephalon’s scheming)
  2. Drug company pulling a drug from the market which had an upcoming patent expiration, replacing it with newer formulation, and spending lots of $$ to get doctors and patients to change to the new formulation (example of Namenda for Alzheimer’s, reformulated as a once-a-day drug, so pharmacies cannot substitute a generic for it. And removing the older one from the market prior to generics being available, thereby forcing providers/patients to switch current users to the new formulation. And the new formulation would continue under a patent for years to come). The New York attorney general (Eric Schneiderman) has filed an injunction against the drug company.

see here for the editorial/

It is always striking how many new (and, in these cases, illegal) ways drug companies try to extend their profits. Unfortunately, these shenanigans are happening as many in control in the US government are promoting an agenda of less regulation/oversight. Of note, the BMJ had a remarkable article in 2012 (see BMJ 2012;344:e4348 doi: 10.1136/bmj.e4348) which did a detailed assessment of drug development and the myths of big pharma claims, noting that only 15% of drugs developed are really important new meds (most are yet another ACE-inhibitor, etc), that the stated costs of R&D are drastically overstated by totally inappropriate data and modeling, that marketing costs (included in their R&D costs) dwarf the true costs of drug development, and that drug companies are really effective in corralling the government into providing more subsidies and longer patent protections.


Primary Care Corner with Geoffrey Modest MD: DPP-4 inhibitors and cardiovascular outcomes

24 Jun, 15 | by EBM

By:  Dr. Geoffrey Modest

New Engl J of Med just had a large study on the cardiovascular effects of sitagliptin (see DOI: 10.1056/NEJMoa1501352). Background: sitigliptin is a DPP-4 inhibitor (dipeptidyl peptidase 4 inhibitor), which functions by decreasing the degradation of incretins (and thereby increasing glucose-mediated endogenous insulin secretion and suppressing glucagon levels). But there were concerning studies with other DPP-4 inhibitors being associated with increased risk of hospitalization for heart failure (saxagliptin was associated with 27% increase in the SAVOR-TIMI study, aloglipitin with a nonstatistically significant increase in EXAMINE study).


Details of the new study (drug company sponsored):

–14,671 patients from 673 sites in 38 countries were randomized to adding sitagliptin 100mg/d (50mg if eGFR 30-50) vs placebo to existing therapy. Followed 3 years

–median age 66, 30% female, 68% white/22% asian/12% latino, BMI 30, BP 135/77, eGFR 75, prior MI 43%, 11% current smoker, LDL 91 mg/dl

— results:

–26% of each group discontinued the med or placebo during the study

–at 4 months there was a 0.4% lowering of A1c (from baseline of 7.3%) with sitiigliptin compared  with placebo. By 3 years, there was only a 0.29% difference

–but, during the course of the study, those of sitigliptin were less likely to start long-term insulin therapy (9.7% in those on sitigliptin vs 13.2% on placebo [HR 0.70 (0.63-0.79)]).

–primary outcome (composite of cardiovasc death, nonfatal MI, nonfatal stroke, hospitalization for heart failure): 11.4% in those on sitagliptin and 11.6% on placebo, not significantly different

–for the various secondary outcomes (which include the individual components of the primary outcome and more): no difference. Also, no difference by any prespecified subgroups (age, sex, race, region of the world, diabetes duration, baseline diabetes therapy, high vs low A1c, BMI, smoking)

–no difference in hospitalization for heart failure

–no difference in adverse events (though the rate of acute pancreatitis came close to being significant, with 23 cases in those on sitaglipitin and 12 in placebo, p=0.07. Pancreatitis has been found to be increased in several of the DPP-4 agents). No difference in pancreatic cancers (though only 3 year study). There was, however, a statistically significant decrease in eGFR  with sitigliptin of -4.0 vs -2.8 ml/minm/1.73m2.

So, what does this study show? First, this was a very select group of patients (baseline A1c, hypertension, cholesterol very well controlled). They excluded those with significant renal dysfunction (eGFR<30). And…

The good news:

–there was some benefit from sitigliptin in reducing the need for longterm insulin

–there was no increase in cardiac events individually, and there was no  increase in hospitalizations for heart failure

The bad news:

–the difference in A1c with sitigliptin was really pretty small, with people improving their control with other meds (which may explain the lack of benefit from using sitiglipitin)

–there was no cardiac protection by sitigliptin (which is really the most important endpoint, since 70-80% of diabetics die of cardiovasc disease, an order of magnitude greater than mortality from microvascular complications)​

 My bottom line: I have never used DPP-4 inhibitors because they are expensive, they require prior authorizations, they have very small effects on A1C (typically about 0.5%), and they have no documented benefit on actual clinical outcomes (and the decrease in eGFR may presage a worsening of this important microvascular outcome). I also have a baseline unease with medications that block a ubiquitous enzyme (DPP-4) which is on the surface of most cells and deactivates a variety of bioactive peptides, not just GIP (glucose-dependent insulinotropic polypetide) and GLP-1 (glucagon-like peptide-1).  So, I basically plan to continue with my pattern of not using them….



Primary Care Corner with Geoffrey Modest MD: New STD treatment guidelines

24 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

The CDC just updated their guidelines on the treatment of sexually transmitted diseases (see here).

They offer the usual advice to do detailed risk assessment for STDs and HIV, counseling about prevention, making sure immunizations are up to date (eg HPV, hepatitis A and B — they do limit hepatitis A vaccine to MSM, injection drug users, and those with HIV or chronic liver disease; though those of you privy to my prior blogs know that I strongly favor all getting hepatitis A vaccine, given the occasional outbreaks in the US, the increasing travel to countries where hep A is prevalent, the lack of adverse effects from the vaccine, and the fact that it is now being given to all little kids). They also comment on the use of treating HIV aggressively (treatment as prevention, also see here) and the utility of  pre-exposure prophylaxis (the recent CROI meetings had several important papers on this, including the IPERGAY trial, which showed that 2 tabs of tenofovir/emtracitibine (truvada) 2-24 hours before unprotected sex in serodiscordant couples, then 1 tab in 24 hours, and another in 48 hours was stopped early because of efficacy).

Here are some of the changes over previous guidelines:

–sections on transgender men and women (noting the high prevalence of HIV in transgender women — ie, transgender male to female)

–mycoplasma genitalium, in 15-20% of male urethritis, and 30% of men with recurrent urethritis (ie more common than gonorrhea and chlamydia). Pathogenic role in women less clear, though is found in both asymptomatic and symptomatic women (10-30% of cases of clinical cervicitis and 2-22% of PID cases). Does not respond to beta-lactams such as penicillins/cephalosporins, responds inadequately to a 7-day course of doxycycline  (31% cure rate), but does respond to 1-g single dose of azithromycin (though resistance is emerging). If treatment-resistant, can try moxifloxacin (400mg daily for 7,10, or 14 days). Not tested extensively, but small reports of very high cure rates. Consider 14-d course if treatment-resistant PID. Treat sex-partners.

–patients with HIV should have annual screening for hepatitis C in those at high risk for infection

–urethritis: best test for men is NAAT (nucleic acid amplification test) in urine for GC/chlamydia. Same treatment (azithro 1 gm once or doxycycline 100 bid for 7 days), though M gentalium (the most common cause of persistent or recurrent non-gonococcal urethritis) responds better to azithro (see above). Also consider trichomonas. No NAAT is FDA-approved for men, but can still do a urinary test or just treat empirically (and I have seen several cases of symptomatic presumptive trich urethritis in men, responding to treatment with metronidazole)

–cervicitis: test for cervicitis or PID with NAAT of vaginal, cervical, or urine samples (though you should check with your lab. Ours finds lower sensitivity with urine specimens). First catch urines are more sensitive, esp for chlamydia. Or can do self-collected vaginal swabs. Also evaluate women with cervicitis for BV and trichomonas. Consider treating M genitalium, as above, if persistent cervicitis after using azithro or doxycycline.

–gonorrhea: as in prior guidelines, given emergence of resistance, should be treated with both ceftriaxone 250mg IM, and azithro 1 g (though can use cefixime 400mg if ceftriaxone is not available).

–vaginal discharge: no signif changes that i detect, but they do note that BV can be associated with ureaplasma and mycoplasma, and that all women with BV should be tested for other STDs and HIV. For trich infections: 70-85%  have minimal or no symptoms, 2-3x increased risk for HIV acquisition, and NAAT is the preferred diagnostic method (picking up 3-5x more infections than wet mounts). Candida: nonalbicans candida is pretty common, >50% of women with it are minimally or not symptomatic, and treatment remains unclear (?longer duration of nonfluconazole azoles, using 600mg boric acid).

–PID: now <50% associated with gonorrhea or chlamydia. Lots of “normal” vaginal microorganisms and others (mycoplasma, ureaplasma, CMV, M genitalium) have been implicated. no change in management

–HPV updated section on vaccines and alternative treatments. advised not to use podophyllin (irritation, occ severe systemic toxicity)

–anal cancer screening: data insufficient to recommend routine screening for people with HIV, MSM without HIV, general population

–section on sexual assault and abuse/STDs: pretty complete (to me), includes using NAATs for chlamydia and gonorrhea at sites of penetration, as well as trich. Also check HIV, hep B and syphilis

Overall, this is a very comprehensive, 110-page document (138-page, if include references), with pretty much everything related to STDs including clear and highlighted treatment regimens, and, I think, would be reasonable to download onto your computer desktop if you see patients with sexually transmitted diseases…​

Primary Care Corner with Geoffrey Modest MD: IMPROVE-IT trial ezetimibe

23 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

So, at long last, the IMPROVE-IT trial has been published: the one comparing simvastatin plus ezetimibe vs simvastatin alone in high-risk patients (this was the study that came out as the block-buster, high-profile study at the Am Heart Assn meetings last November –7 months ago!!!, but was notably not published simultaneously, for remarkably opaque reasons….. see DOI: 10.1056/NEJMoa1410489). Ezetimibe decreases absorption of cholesterol from the GI tract and in prior studies lowered LDL by around 23-24%. This was a large double-blind study of patients recently hospitalized for an acute coronary syndrome and already had pretty good lipids initially. details:

–18,144 patients from 1147 sites in 39 countries (mean age 64, 76% male, 84% white, BMI 28, mean LDL was 93.8 mg/dl, 62% had hypertension and 27% diabetes, 33% current smoker, 35% on statin and 42% on aspirin prior to ACS), hospitalized for an acute coronary syndrome (acute MI with or without ST-segment elevation, or high-risk unstable angina) within the past 10 days, whose LDL was 50-100 mg/dl if on lipid-lowering therapy, otherwise up to 125 mg/dl. Randomized to simvastatin 40 vs simvastatin 40 plus ezetimibe 10 (combo group) — though some patients did receive the 80mg dose of simvastatin prior to the FDA warning in 2011, in up to 27% in the simvastatin monotherapy group. Followed median of 6 years.


–achieved LDL in simvastatin group was 69.5 mg/dl; in combo group was 53.7 mg/dl (a 24% reduction, as anticipated). Also decreases in triglycerides, non-HDL cholesterol, apo B, hs-CRP levels)

–primary endpoint (composite of: death from cardiovascular disease; major coronary event such as nonfatal MI, documented unstable angina, or coronary revascularization; or nonfatal stroke): 32.7% with simvastatin-ezetimibe vs 34.7% with simvastatin (absolute risk reduction of 2 percentage points, with HR 0.96 (0.89-0.99, p=0.016), a 6% decrease. Benefit evident after 1 year, curves paralleled after that (ie no increasing ezetimibe effect over time). so, number-needed-to-treat of 50 patients for 7 years to prevent one event.

–in terms of specific individual outcomes (tertiary endpoints), the only ones achieving statistical significance were: any MI, nonfatal MI, ischemic stroke, hemorrhagic stroke, and urgent coronary revascularization

–also, buried in the supplementary materials, subgroup analyses revealed there were statistically significant differences only in the following:

–only in those >=75yo (n=2798, 11% reduced risk with combo therapy)

​–only in females (n=4416, 11% reduced risk with combo therapy)​

​–only in nonsmokers (n=12154, 9% reduced risk with combo therapy)​

–only in those with diabetes (n=4953, 14% reduced risk with combo therapy)

–only in those without prior stroke (n=12154, 9% reduced risk with combo therapy)

 ​–only in those with hypertension (n=11137, 9% reduced risk with combo therapy)

–only in those with creatinine clearance <60 (n=3261, 12% reduced risk with combo therapy)

–adverse event rate was similar in the 2 groups (no statistically significant differences, including myopathy, rhabdo, cancer, LFT changes). 10% in each group discontinued therapy from adverse events.


–this study basically confirms the concept of treating-to-goal (I hate to say that “I told you so”, but will reluctantly do so in this case). But brings up a few issues:

–their figure 2 shows that the achieved LDL with added ezetimibe falls on the line of the achieved LDL line with statins (see following figure). This confirms other empirical observations that, even though statins have many pleiotrophic effects beyond lowering LDL (and some suggest that only 50% of the statin effects can be explained by the LDL effect), it really is the LDL which is predictive of future events, whether achieved by statins or ezetimibe (unless ezetimibe had similar pleiotrophic effects….)


(Cannon et al., 2015)

–what is the correct LDL goal?? the short answer is “who knows??”.

–it is clear that those at higher risk  benefit dramatically by taking statins by lowering their risk 30+%, but they still have a higher residual risk than those at lower initial risk.

–the data from the PCSK9 drugs shows that achieving an LDL of 48 is both clinically efficacious and without significant adverse events (see blog: here)

–so, what makes sense?:  I guess at this point that lowering the LDL to 100 mg/dl in primary prevention is reasonable. In those with evident atherosclerotic disease (CAD, PAD, stroke) or with diabetes, it probably would be useful to lower it more, perhaps to the 70 mg/dl range. and in those at really high risk, such as in the IMPROVE-IT study, or those with evident atherosclerotic disease with lots of ongoing risk factors that they are unable to modify (eg smoking), to shoot for a lower goal.

–does the achieved HDL play a role?

–the TNT (Treating to New Targets trial) found that, controlling for the HDL level, the achieved LDL seemed to determine the cardiovascular risk. but HDL, which seems to be taking a beating in the press, was actually important: the number of clinical events was the same if the achieved LDL were >100 mg/dl but the HDL were >=55 as if the achieve LDL was <70 with an HDL <38. And this was independent of whether the patient were on atorvastatin 10mg or 80mg. see the graph below (see n engl j med 2007:357;13), though this is a secondary analysis.


(Barter et al., 2007)

–does it make sense to generalize from the very high-risk individuals in IMPROVE-IT to the rest of the people we would consider putting on medications to lower lipids??

–again, “who knows?”. But seems to me to be pretty likely.  That being said, the IMPROVE-IT study was a poster-child of a drug company sponsored study. It added an unknown but likely useful lipid-lowering agent to an inferiorly-dosed statin in a combination pill made by the drug company. And as of 11/10/14 (prior to the release of IMPROVE-IT and likely subsequent dramatically increased revenues), the Wall Street Journal reported that the “sales of Zetia were $1.99 billion for the first nine months of 2014, while sales of Vytorin were $1.15 billion for the same period”.  One major concern I have is that in this huge, longterm study, the ezetimibe addition had only a small effect in these patients (who, admittedly, were already at a pretty low LDL level on just simvastatin), but there were pretty dramatic differences by subgroup (ie, no statistically significant effect of adding ezetimibe if nondiabetic, nonhypertensive, smokers, males, <75yo, and with normal renal function.) These were secondary, subgroup analyses, though do raise issues….

–one other comment: despite the above, including the first graph, I am really not fully convinced that ezetimibe has the same efficacy as statins, on an achieved LDL basis. First, the studies on statins typically show significant efficacy within 6-12 months, are usually stopped early because of this efficacy (and not go on for 6-7 years), and have diverging curves showing increasing efficacy over time. Subgroup analysis with statins usually show no differences if the patient has diabetes, is smoking, hypertensive, etc. And there was even a study in patients with ACS (the PROVE-IT TIMI 22 trial, see N Engl J Med 2004; 350:1495-1504) which looked at 4162 patients randomized to pravastatin 40mg vs atorvastatin 80mg, with initial LDL of 106 mg/dl (decreasing to 95 with pravastatin and 62 with atorvastatin), and found benefit in the atorvastatin group within 30 days and increasing benefit — diverging curves — after 2 years.

–given the multitude of data on statins, i personally think that the appropriate approach would be to first maximize the statin (ie, for those under “goal”, use at least atorvastatin 80, but would then move to rosuvastatin 40mg). If still below “goal”, would then consider adding ezetimibe (always reinforcing the appropriate lifestyle changes). Though, based on the TNT trial data above, my LDL “goal” is somewhat modified by the HDL. And we really do need more studies on ezetimibe in different populations to convince me to using it before increasing to the highest potency statin. For more blogs on statins/lipids, see here.

Primary Care Corner with Geoffrey Modest MD: New vs old OCPs and thromboembolism

23 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

The BMJ just published a large analysis of the relationship between combined oral contraceptives OCPs and the risk of venous thromboembolism VTE (see BMJ 2015;350:h2135). They analyzed 2 nested case-control studies from 2 different UK databases, with a combined input from 1340 general practices, assessing the first diagnosis of VTE in women aged 15-49 from 2001-2013, and comparing this group to 5 controls matched for age, practice, and calendar year.


–10562 cases of VTE. mean age 38, 27% smokers in the VTE cases and 21% in controls, obesity in 27% vs 16% in controls. established risk factors for VTE were 47% vs 26% in controls

–odds ratio for incident VTE and use of OCPs in the previous year, adjusted for smoking, alcohol, ethnic group, BMI, comorbidities, and other contraceptives:

–current exposure to any OCP was associated with increased risk of VTE [odds ratio OR 2.97 (2.78-3.17)] vs no OCP exposure over the previous year, with breakdown as follows.

–newer OCPs:

–desogestrel         OR 4.28  (3.66-5.01)

–gestodene            OR 3.64  (3.00-4.43)  –not available in the US

–drospirenone      OR 4.12(3.43-4.96)

–cyproterone        OR 4.27  (3.57-5.11)  –not available in the US

–older OCPs (second generation)

–levonorgestrel     OR 2.38 (2.18-2.59)

​–norethisterone    OR 2.56 (2.15-3.06)    –not available in the US

–norgestimate       OR 2.53 ( 2.17-2.96)  –actually a third generation OCP which is partly metabolized to levonorgestrel, but is less androgenic than levonorgestrel and is actually considered a second generation one in Denmark.

–this translates to: the number of extra cases of VTE per year per women

        –levonorgestrel was lowest at 6 (5-7), along with norgestimate at 6 (5-8)

        ​–desogestrel was highest at 14 (11-17), along with cyproterone at 14 (11-17)

–overall, the risk associated with gestodene was 1.5x higher than levonorgestrel (the most commonly used in the UK), and those of desogestrel, drospirenone, and cyproterone was about 1.8x higher

–desogestrel had a slightly higher odds ratio for VTE with higher doses of estrogen. norethisterone and gestodene actually had slightly higher VTE rates at the lower estrogen dosages (these were not significant, which seems to support the primary association being with the progestin in these women already on lowish doses of estrogen)

So, most, but not all, studies over the past 2 decades have often shown increased VTE risk in the 3rd and 4th generation OCPs. The current very large UK groups, given the high quality of reporting, provides more information, though it is retrospective and observational. In terms of bias, one might think that since there have been older studies showing increased VTE risk with the newer agents, there might have been preselection bias to avoiding these agents in women surmised to be at somewhat increased risk, leading to understating the actual risk.  Although I do not prescribe OCPs so often these past years, I must admit that I have been largely avoiding the newer agents because of the issue of increased VTE found in several of the earlier studies, and now confirmed here.​

Primary Care Corner with Geoffrey Modest MD: Melanoma increasing (a lot)

23 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

MMWR just released an analysis of melanoma incidence, mortality trends, and projections in the US from 1982-2030 (see here). Melanoma is common (5th most common cancer in men and 7th in women) and is associated with the most skin cancer deaths, with deaths occurring most frequently in younger people (average of 20.4 years of potential life lost). Costs of treatment are high ($3.3 billion/yr in the US). And melanoma is largely preventable — 90% are attributable to skin damage from UV light exposure, with sunburns being a significant risk factor and 40% of US people reporting sunburn every year. There is clear efficacy of sun-protective behaviors in decreasing UV skin exposure, such as sunscreen, protective clothing. In addition there are suggestive data that cumulative UV radiation exposure over time may also be a culprit. results from MMWR:

–in 2011: 65,647 invasive melanomas in the US, with age-adjusted incidence rate of 19.7/100K, increasing with age and highest in non-Hispanic whites, in women 15-49yo, and in men >50.

–in 2011: 9,128 melanoma deaths, with age-adjusted death rate of 2.7/100K, higher in non-Hispanic whites, increasing with age, and higher in men (4.0) than women (1.7)

–from 1982-2011: melanoma incidence rates increased (doubling), though mortality remained the same.

–without intervention, there are projected to be 112,000 new cases in 2030 with annual cost for treatment projected to increase 252.4% from 2011-2030 (from $454 million to $1.6 billion)

–a comprehensive skin care prevention program (modeled after the one in Australia, which raised community awareness through mass media campaigns, programs in schools and workplaces, provider education, etc, had projected savings of $2.30 for every $1 spent) is estimated to prevent 20% of US melanoma cases from 2011-2030, averting 21,000 cases/yr, and reducing spending $250 million/yr (saving $2.7 billion from 2020-2030).

So, this seems like a largely preventable problem. There are pretty easy individual strategies available, including not using indoor tanning salons and using appropriate sunscreen protection — correct amount, reapplication rates, etc.  But, as in Australia, the major potential impact would be from concerted, community-oriented and community-based public health initiatives. Some of the above gender differences may be attributable to the increased female use of indoor tanning salons (a 2013 report found that approx 1/3 of non-Hispanic white women aged 16-25 use indoor tanning salons each year) and decreased male use of sunscreen protection. It is also important to remember that there still is a melanoma risk in darker skinned individuals. Overall black incidence in 2011 was 1.0/100K, vs 24.6/100K in non-Hispanic whites; but the mortality in black americans was 0.4/100K vs 3.1/100K reflecting a 3-fold increased mortality-to-incidence ratio vs non-Hispanic whites. This higher mortality is perhaps related to data suggesting that darker-skinned people report more frequent sunburns, are less likely to use sun-protection strategies, have a higher incidence of poorer-survival melanomas perhaps related to later diagnosis, a lower perceived risk by patients and providers, and more melanomas in non-sun-exposed areas.

Primary Care Corner with Geoffrey Modest MD: Mass med society opioid prescription guidelines

17 Jun, 15 | by EBM

By: Dr. Geoffrey Modest

The Massachusetts Medical Society just sent out guidelines on opioid therapy prescribing (see here).  A brief summary:

–most of the guidelines are pretty self-evident:

–for initiation of opiates: screen the female patients for pregnancy (and counsel re: risks)

–do some form of screening for risk of opiate abuse [i would add that the website has lots of useful tools for opiate prescribing, including an opioid risk tool to assess for risk of opioid addiction. I would also add that this type of risk tool is a not-completely-accurate instrument].

–inform patients of cognitive issues with taking opiates, and potential risks of operating heavy machinery, driving

–consider consultation if patient has complex pain issues with serious co-morbidities and mental illness, or history of substance abuse disorder

–use opiates at the lowest dose and shortest duration possible

–review other meds being taken, which might increase the risk of opiate harm (eg benzodiazepines)

–regular patient visits to assess progress (function, pain) and reconsider the continuing need for opiates

–chronic opiate usage (>90 days): “consider” consulting with other physicians as part of developing and implementing an ongoing treatment plan

–perform appropriate baseline urine drug testing if risk assessment suggests there may be issues with other drugs [our practice is to do this pretty regularly on almost all patients]

–there should be a re-evaluation of the treatment plan and goals every 60-90 days

–best to have a written treatment agreement, updated annually, signed by the patient and physician, to include goals of therapy (such as improving function and quality of life, control pain, when it is okay to get opiates from other providers, and what the goal is to reduce the opiates to the lowest level possible to control pain and preserve function)

–discuss risks of dependence and addiction; discuss/offer naloxone

and, the more controversial

— “physicians who are not pain management specialists should not initiate treatment plans which call for in excess of 100 milligrams of morphine equivalent opioids per day without a documented consultation with a pain management specialist”

–“if a patient is currently receiving > 100 mg morphine equivalent per day, a plan should be instituted to begin tapering of the dose, and if not possible to do so, consultation with a pain management specialist should be obtained.”

So, this really brings up a few issues.  Part of the issue is that we in primary care are being pretty rapidly buffeted by opposing  winds. First, the Institute Of Medicine came out with a clear proclamation in the 1990s that we were undertreating pain. This was augmented by the rather seedy/self-serving  Purdue initiative (makers of oxycontin) to make pain the 6th vital sign (which catapulted oxycontin, their drug, into a blockbuster for them). Now we are confronted with the very real prescription drug mortality epidemic, raging out of control. And the response is to put a ceiling on primary care prescribing (stigmatizing both providers and patients), unless we get the approval of a pain clinic. So, here are a few of my random thoughts on this (the issue of opioid use is quite complex, so these points are hardly all-inclusive):

  1. The overdose/death rate for prescription opioids is staggering and a clear social issue, and we as providers should be part of the solution
  2. There are real concerns about prescription opiates now being “gateway” drugs to more high risk behaviors: I have seen and read about many patients taking prescription opioids who can no longer afford them and are switching to street heroin, with its profound risks.
  3. But for an individual patient, there are clearly many who need and benefit from opioids (though there are some interesting genetic issues here, see the blog: here)
  4. It is rather striking that, given how common the opiate-prescribing issues are,  there are a truly remarkable dearth of studies on opioid use – for example, there are NO (ie, zero) randomized controlled studies looking specifically at whether higher dose opioids are more effective than lower dose (and, if higher doses do in fact work significantly better, this might justify using a higher dose in some patients even if there is a higher associatedrisk). This fundamental lack of data makes it difficult to accept a guideline approach of considering 100 morphine-equivalents as the top dose that we should use in primary care.  And, there are not even data on the widely held concept that, for chronic pain, we should use a basal long-acting opiate augmented with a short-acting one for break-through pain (and there even are some data to the contrary).
  5. The concept of needing to get a chronic pain consult after a certain level of opiate use is pretty insulting to us in primary care. We are not forced to get consults around other issues. If someone has hypertension uncontrolled on 4 meds, we do not need a hypertension consult. Or a cardiology consult for a person with atrial fibrillation. And, most of the patients we see on chronic opioids have very complex psychosocial situations, often with underlying psych issues (depression, PTSD, history of domestic or sexual violence, trauma), personality issues, physical trauma (severe accidents) etc. It typically takes us many sessions with a patient to unravel and understand the complexity of care for the patient, for which opioid use is only part. Prescribing opioids has to be understood in this greater context. Although pain clinics might help in some patients in making suggestions about adjuvant therapies or other approaches, it really is us guys in primary care who are in the position to really understand the patients and put all of the pieces together. Hard to replicate this understanding in a single visit with a pain specialist.  So, I would agree that in some cases a pain consult is likely to be helpful with specific questions about medication management or other interventions, to me it makes little sense to mandate one for all patients at some arbitrary medication dosage (especially since this not backed by any evidence-based studies). And, of course, I really doubt there is the capacity in pain clinics to see all patients on >100 morphine-equivalents.
  6. ​Even for issues such as the combined use of opiates and benzodiazepines, there are only observational studies which show a higher mortality. But we know that several of the underlying diseases themselves often treated with benzos have an increased mortality (eg panic disorders). And even if the benzos are associated with more adverse events, there clearly are patients who really need the benzos to function and are willing to take the risk of adverse events (as we do all the time: whenever we prescribe medications, we are balancing the benefits and risks, whether that is the unlikely risk of death from a sulfa drug – which I have seen twice, or the higher likelihood of adverse events with cancer drugs). Clearly we should minimize giving any drugs unless they are really indicated, but the issue of benzos and opiates is not fundamentally different from that of other issues we deal with in primary care on a daily basis.
  7. I think our job in primary care is really to develop relationships with patients. Those who are on or need opiates typically fall into one of 3 categories:

–Those who clearly need opiates to function (eg cancer patients with pain, but also many people with severe knee arthritis who need opiates to be able to walk, and personally, I don’t think we should treat these patients differently even though the cancer patient with pain has a more socially-acceptable diagnosis. And despite geriatric societies putting opiates on their “do not use” list (risk of falls), I do have some quite old patients who really need them to function at all, as well as remain reasonably pain-controlled, as I do with some elderly cancer patients.

–Those who are pretty likely addicted to the opiates, use multiple potentially hazardous illicit drugs, or are likely diverting/selling the opiates. Here, providing opiates may well be harmful to the patient or society. It seems to me that this is the group we need to work with, develop relationships with, and try to help them/support them as best we can without providing opiates (or, if addicted, consider using buprenorphine along with counseling/support)

–Those in the middle. We are not sure if the 30 year old with back pain really has so much pain to require high dose opiates. We can try to give objective tests of function or standardized questionnaires, though pain is fundamentally subjective. We get insights by observing function, pain symptoms as people walk in the hallway or climb onto the examining table or more extensive testing. We get additional information by doing random urine tests, pill counts, checking prescription monitoring programs. Those who screen positive on the opiate risk tool are more likely to be in this category (or the one above). We review and sign controlled med contracts (which we do with others as well, though their clear benefit is not well tested). And, hopefully, through developing a relationship, try to sort out what’s going on in order to best help the patient.

  1. And, one big issue, is that if someone is on >100 morphine-equivalents, and there are guidelines as above, that confers a strong medicolegal aspect to the situation: if someone is on 120 morphine-equivalents and has a bad outcome, we as primary care providers could be sued for not adhering to the “professional society guidelines”. Not that we should just practice defensive medicine, but I have seen a few providers who have been sued (inappropriately, in these cases), and I can tell you it is a really awful experience for them even though they were right.
  2. ​But, with all of this ranting, i do think we as primary care physicians do have the individual patient and broader  social responsibility to try to ensure that our patients who are on opiates really do need them, they are on the minimum amount to allow them to function, we try to avoid using benzos with opiates if possible, and we set up systems to make help prevent opiate diversion and misuse.​

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