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 November, 2014

Primary Care Corner with Geoffrey Modest MD: Paget disease clinical guidelines

25 Nov, 14 | by EBM

By: Dr. Geoffrey Modest

The Endocrine Society just released clinical practice guidelines for the diagnosis and management of Paget disease of the bone.


Some background: there is a genetic component with family history in 10-20% and autosomal dominant transmission pattern, clinically rare prior to age 40, men and women affected.

Basic recommendations:


–In patient with suspected Paget, Get plain xray of suspected area. If diagnosed with Paget, then get radionuclide bone scan to assess extent of disease

–If diagnosed with Paget by xray, get biochemical evaluation: alkaline phosphatase (ALP) or more specific marker of bone formation (especially if there is hepatobiliary disease which could affect the ALP) – can check amino-terminal propeptide of type 1 collagen (P1NP, the best option but expensive) or bone-specific ALP (though there is about 20% cross-reactivity between antibodies to liver and bone ALP) and bC-terminal propeptide of type 1 collagen (bCTx) or urinary N-terminal propeptide of type 1 collagen (NTx).


–Bisphosponates if at risk of future complications

–“Suggest” a single 5-mg dose of IV zolendronate as treatment of choice, since it usually lasts >6 years and requires less costly followup. other options include alendrontate 40mg/d for 6 months (may need to retreat in 2-6 years) or risedronate 30 mg for 2 months (may need to retreat in 1-5 years)

–If symptoms need to be controlled urgently (see below), assess markers of short-term response to assess if there is adequate response to therapy (here it is best to use b​CTx as marker of bone resorption, since it responds more rapidly to treatment than ALP). Assess before and shortly after treatment initiation

–If osteolytic lesions present, repeat xray in 1 year

–Maintaining remission: check ALP or other baseline disease markers at 6-12 weeks after therapy, though maximal suppression may take 6 months. follow bone turnover markers (should be below midpoint of assay reference range) as marker of need for retreatment. for zolendronate, check every 1-2 years. for other regimens, check every 6-12 months.

–Monostotic Paget:  all of the biochemical markers of bone turnover may be normal even in symptomatic patients

Complication Management

–Hearing loss: treat with potent bisphosphonate (eg IV zolendronate) to prevent worsening

–Osteoarthritis: if severe, bisphosphonates (even before joint replacement surgery).  when mild, use analgesics. often hard to know if it is the paget’s or underlying osteoarthritis causing the pain. can consider bisphophonates as well as analgesics, though data sparse.

–Bowing of lower extremity: potent bisphosphonate (eg IV zolendronate)​ prior to surgery

–Paralysis (with disease of the spine): bisphosphonate with neurosurgical evaluation. usually medical therapy is sufficient (unless severe structural damage)

–Neoplasm: if osteosarcoma or giant cell tumor, pretreat with  potent bisphosphonate prior to surgery. interesting that neoplasms seem to be less frequent in this era of bisphosphonates.

In reviewing financial disclosures, of the 7 members of the taskforce, 2 without conflict, 1 with declared conflicts, and 4 with financial interests in several drug companies but stated that there were no conflicts.


Primary Care Corner with Geoffrey Modest MD: HDL a negative risk factor? Or cholesterol efflux??

24 Nov, 14 | by EBM

By: Dr. Geoffrey Modest 

Although the data on HDL being protective of atherosclerotic disease is pretty consistent, there are negative studies, and the data on medications which dramatically improve HDL levels (eg, the cholesteryl ester transfer protein –CETP –inhibitors, such as torcetrapib) are clinically disappointing. there have been several explanations for this. for example, some HDL particles are “pro-inflammatory” and elicit an atherosclerotic response (which may be related to apolipoprotein C-III). A new article in New England Journal of Medicine highlights another angle — that the issue is not the HDL number, but HDL’s functionality in the reverse transport of cholesterol, as measured by the cholesterol efflux capacity (see DOI: 10.1056/NEJMoa1409065). This article looks at the Dallas Heart Study, where they measured HDL levels, HDL particle concentrations, and cholesterol efflux capacity in 2924 adults free of cardiovascular disease and followed people a median of 9.4 years, assessing the primary endpoint of nonfatal MI, nonfatal stroke, coronary revascularization, or cardiovascular death.  Untitled


–Median age 42; 43% male; 49% Black, 30% White; 30% hypertensive; 10% diabetic; 28% smoking; baseline total cholesterol 177, triglycerides 96, LDL 104, non-HDL 127, HDL 47, HDL particle concentration 32.8 mmol/L, Lp(a) 48, hs-CRP 2.7, coronary-artery calcium >10 Agatston units 19%. (ie, a pretty young and healthy crew)

–HDL cholesterol had significant correlations with: BMI, waist-to-hip ratio, visceral and truncal fat, adiponectin, leptin, insulin resistance, and CRP. 35% of variance of HDL was atributable to these other risk factors

–HDL particle concentration correlated with only waist-to-hip ratio, adiponectin and leptin levels; cholesterol efflux capacity correlated only with adiponectin and only to a small degree only. 3% of efflux capacity was attributable to other risk factors.

–132 patients had primary cardiovascular events over the 9.4 years.

–There was an inverse trend for baseline HDL to be associated with cardiovascular events, though even less pronounced when controlling for other risk factors.

–There was an inverse association between HDL particle concentration, adjusting for both the cardiac risk factors and HDL level (HR 0.53, 0.31-0.89)

–There was a graded inverse relationship between cholesterol efflux capacity and cardiovascular endpoints, comparing 4th vs 1st quartlile having HR 0.44 (0.27-0.73), without attenuation by other cardiac risk factors, HDL level, and HDL particle concentration.

–Adding cholesterol efflux capacity to the traditional cardiac risk factors significantly improved the risk-prediction indexes.

So, what does this all mean?

–Function is more important than just being there. it has been shown that there nonfunctional HDLs, as noted above. cholesterol efflux capacity, as shown in this study, is likely the most important functional role of HDL, though HDL also has antioxidant properties, decreases blood viscosity, improves endothelial function, and inhibits platelets.

–As commented on in the NEJM article, there is an assay for cholesterol efflux capacity which is now scalable to large numbers of patients (though i highly doubt it is commercially available, and i would wait for more large studies before ordering it anyway).

–I am always concerned that the newest article that comes out negates the trove of prior findings. the reality is that this Dallas Study was of very healthy people, there were not a lot of cardiac events (132 in 2924 patients followed almost 10 years — only 4.4% !!), and so many previous articles have found that HDL is protective in population studies. and every article I’ve seen which looks at cardiac clinical outcomes has found that the cholesterol/HDL ratio is more predictive of events than the LDL (i’ve seen at least 6 such studies).  so, there is a really good reason that the Framingham Risk Score and others  incorporate the total cholesterol and HDL in their risk models.  The negative press on HDL is largely based on the negative intervention studies (eg with the CETP inhibitors, or with niacin, both of which probably create collateral damage on the vascular system).

–Bottom line for now: until a commercial, inexpensive assay becomes available which reliably predicts atherosclerotic events better than HDL levels, we should continue measuring HDL levels and incorporating them into our cardiac risk assessment.​

Primary Care Corner with Geoffrey Modest MD: New pre-op evaluation recommendations

21 Nov, 14 | by EBM

Two new recommendations for perioperative cardiovascular disease evaluation and management for patients undergoing noncardiac surgery were just released, one from the US and one from Europe (will limit review to common pre-operative primary care issues).

The US guidelines (see doi:10.1016/j.jacc.2014.07.944) reviews the risk factors for perioperative cardiovascular mortality, including the traditional ones (see paper for details):

–MI within the past 6 months, and esp. within the past 2 months

–heart failure, esp. decompensated, and more so in those with reduced ejection fraction

–cardiomyopathies present special management issues, though data on postoperative outcomes is limited and with mixed results

–valvular heart disease, noting the utility of preop echocardiography in those with suspected moderate or greater degrees of valvular disease and that recent advances in preoperative and operative monitoring/therapy have significantly reduced the risk of cardiac events — even those with severe AS and valve area <1cm2 now have only a slight increase in cardiac mortality (2.1% vs 1% at 30 days)

–arrhythmias typically need monitoring but more recent trials suggest they confer less cardiac risk — though their presence should trigger evaluation for underlying cardiopulmonary disease, especially if sustained or are associated with hemodynamic compromise

–high-grade conduction abnormalities such as complete A-V block may require pacing, but even left or right bundle-branch blocks only rarely progress perioperatively.

–also noted: preoperative natriuretic peptide has been found to be an independent risk factor, though requires further prospective studies, and other biomarkers have been incorporated into some of the risk models, including C-reactive protein and creatinine

There are 3 risk calculators noted in the paper, including:




For cardiac testing recommendations:

–first, evaluate functional status, a reliable predictor of cardiac events, with the increased risk if patients unable to perform at least 4 METs of activity (e.g., climbing a flight of stairs or walking up a hill, walking on level ground at 4 mph, or performing heavy work around the house).  There are also formal instruments to evaluate (e.g. Duke Activity Status Index).

–those patients to undergo nonemergent surgery and are unable to perform 4 METs activity should have pharmacologic stress testing, with potential revascularization if abnormal.

–other evaluations include 12-lead EKG if known cardiovascular disease, significant arrhythmia, peripheral arterial disease, cerebrovascular disease, or significant structural heart disease, “except for those undergoing low-risk surgery” and (lower level of certainty) if asymptomatic patients without known  cardiovascular disease. In all recommendations, not necessary if low-risk surgery. If doing EKG, should be within 1-3 months prior to surgery in stable patients. So, current imperative to do EKG within one month of cataract surgery seems quite unnecessary!!

–for patients with potential LV dysfunction on clinical evaluation, reasonable to do echo (in clinically stable, okay if echo done within past year).

–in those with stents: if drug-eluting, best to delay elective noncardiac surgery for 1 year (and at least 180 days). With bare metal stents, delay at least 30 days.

For perioperative therapy:

–only do coronary revasc if indicated according to existing recommendations, don’t do just for perioperative management.

–new recommendations for b-blockers (after discrediting the Dutch DECREASE trials for improprieties, as well incorporating the results of the new POISE-II trial): continue them if they were used chronically before surgery, may be reasonable to begin them in those with intermediate or high risk        myocardial ischemia in preop risk stratification (paying close attention to adverse effects: hypotension, bradycardia, etc.), may be reasonable in those with 3 or more risk factors (diabetes, heart failure, CAD, renal insufficiency, stroke), and if going to use them, begin more than 1 day prior to surgery.         –statins should be continued in patients already on them, and are reasonable if patient undergoing vascular surgery (who should probably have been on them anyway…).

–don’t give alpha-2 agonists (e.g. clonidine) perioperatively. Some studies suggest that calcium blockers (esp. diltiazem) may have benefit, but no recommendation. Can continue ACE-I or ARBs perioperativley.

–for antiplatelet drugs: those with recent stents should continue with dual antiplatelet therapy unless risk of bleeding outweighs benefit; in those without stents, may be reasonable to continue aspirin unless risk outweighs benefit; don’t initiate aspirin prior to surgery if no stent.

–anticoagulants: can continue in surgery with minimal bleeding risk (cataracts, minor derm procedures). For those on warfarin, bridging therapy (heparin) useful if high risk (mechanical mitral valve, or aortic valve plus another risk factor, such as afib, prior thromboemboli, LV dysfunction). Important  ​issue with the newer thrombin inhibitors is that they are not reversible. So stop them at least 48 hours prior to surgery.

The European recommendations (see doi: 10.1093/eurheartj/ehu282​) seem pretty similar on my less intense review. They do suggest a resting echo in asymptomatic patients without signs of cardiac disease but undergoing high-risk surgery. For b-blockers, they specifically recommend atenolol or bisoprolol. indications similar to US recs, though more strongly point out that these should not be fixed dose, but titrated to resting heart rate of 60-70 and systolic bp>100mmHg, and should be started more than one day prior to surgery (as with US recs), though they suggest preferably 7-30 days before. For statins, they suggest trying to start 2 weeks before surgery, with same indications as with US recs. ACE/ARBs, if indicated (e.g. pt with heart failure) should begin at least 1 week prior to surgery. They do seem a bit more aggressive in some issues:  e.g., consider pre-op routine carotid artery imaging in patients undergoing vascular surgery. Most of other recommendations are pretty similar to the US ones.



Primary Care Corner with Geoffrey Modest MD: One-time flexible sigmoidoscopy decrease mortality

21 Nov, 14 | by EBM

JAMA had a population-based clinical trial in Norway of flexible sigmoidoscopy (sig) screening (see doi:10.1001/jama.2014.8266). 100K men and women aged 50-64 were randomized to once-only flex sig (10.2K people), combo of flex sig and fecal occult blood testing (FOBT, also 10.2K people) or control (78.2K), with screening in those 55-64 beginning in 1999-2000 and those 50-54 yo in 2001 and followed until 2012.  Those with positive screen (cancer, adenoma, polyp>1cm, pos. FOBT) were offered colonoscopy.


–Adherence to screening was 63%

–After 10.9 yrs, 71 died from colorectal cancer in screened groups (31.4 deaths/100K person-yrs) and 330 in the control group (43.1 deaths/100K person-yrs), abs diff of 11.7 and HR of 0.73 (0.56-0.94). i.e., there was a difference of about 1 death/1000 individuals by sig screening

–Incidence of colon cancer in 253 in the screened group (112.6 cases/100K person-yrs) and 1086 in the controls (141.0 cases/100K person-yrs), abs diff of 28.4 and HR of 0.80 (0.70-0.92)

–Flex sig was beneficial in both age groups: cancer incidence in 50-54yo had HR of 0.68 (0.49-0.94) and in those 55-64 HR was 0.83 (0.71-0.96).

–Adding the FOBT did not change the results

–No complications of flex sig (done in 20.5K men and women), though of the 2816 colonoscopies done, (19.5% of those screened), there were 6 perforations and 4 were admitted for postpolypectomy bleeding

–Of note, there was no difference between screening and controls until 9 years later (for both mortality and cancer detection), which might suggest that the results would be more impressive with longer follow-up.

So, these hazard ratios are pretty comparable to the existing studies (although this was a truly population-based study, unlike the other trials which consisted of volunteers). Also first study to show benefit in those 50-54 years old. The big issue for us is that we tend to do colonoscopies as our first line, even though the complication rate is much higher (as in this study), and the only good data for cancer-specific mortality rate reductions are based on sigmoidoscopy studies. The issue, of course, is that sigs miss proximal tumors, and around 20% of patients with serious proximal lesions have no distal lesions (and therefore be missed by just doing sigs). The rationale for the FOBT is that proximal lesions bleed more than distal ones and FOBT might help uncover them, though the results above showed no benefit from adding FOBT. Also the clinical benefit of finding/resecting proximal tumors is significantly less than with distal tumors, in several studies. This study did find a 10% reduction in proximal colorectal cancer vs 24% reduction for distal cancers (proximal cancers presumably found on colonoscopy). The advent of stool DNA testing might shift the curve more away from routine colonoscopy screening. So, bottom line: we have largely adopted colonoscopy screening based on what makes sense (and to the delight of the gastroenterologists/hospitals — “scoping for dollars” as one of my GI friends put it), though sigs are reasonable alternatives. We should probably discuss this with patients, though there is an uphill battle given the dominant culture on this. Maybe will change with new tests in the future.


Primary Care Corner with Geoffrey Modest MD: Obstructive sleep apnea

20 Nov, 14 | by EBM


The American College of Physicians (ACP) issued a clinical practice guideline in August (see doi:10.7326/M12-3187), which basically supports performing a sleep study for patients with unexplained daytime sleepiness. There are many adverse outcomes associated with sleep apnea (OSA), including cardiovascular disease (with the degree of risk associated with the degree of disordered breathing), hypertension, cognitive decline, metabolic abnormalities including type 2 diabetes, increased peri-operative mortality, all-cause mortality (esp. with higher levels of disordered breathing) etc. However, some clinicians have been ordering sleep studies in anticipation of improving these associated conditions, which is problematic both because of the remarkable baseline frequency of OSA in the population (on the order of 10-17%) which increases with age, and the data are lacking that finding OSA and treating it leads to any improvement in these associated conditions (e.g. CPAP does not reduce mortality or coronary heart disease in those without daytime sleepiness; there was small reduction in blood pressure in those with daytime sleepiness and using CPAP, but unclear if true in those who are asymptomatic).

The ACP’s specific recommendations:

— ACP recommends a sleep study for patients with unexplained daytime sleepiness (weak recommendation low-quality evidence)

— ACP recommends formal polysomnography in a sleep lab for diagnostic testing in patients suspected of OSA. they recommend portable sleep monitors in patients without serious comorbidities (e.g. chronic pulmonary disease, heart failure, neurologic disorders, where portable monitors have not been evaluated) as an alternative to polysomnography, when polysomnography is not available for diagnostic testing (weak recommendation, moderate-quality evidence). Studies are limited with direct comparisons between the different types of portable monitors, though it makes sense not to use ones that cannot differentiate between central and obstructive sleep apnea, since CPAP may be contraindicated if central. [Although ACP recommends home monitors “when polysomnography not available”, many insurers are requiring the home based monitors unless there are significant comorbidities.]

A couple of comments. First, I really agree with their conclusions. Lacking clear data on clinical benefit, we should be focusing the use of sleep studies on those with functional impairment which might be related to OSA and where CPAP does improve outcomes: i.e., those with daytime somnolence. In this regard it is important to keep in mind that, on the one hand, only 37% of patients with severe OSA had daytime somnolence, and on the other, mortality associated with severe sleep apnea is unrelated to daytime somnolence (but, again, intervention studies are lacking, with no difference in cardiovascular mortality, as noted above). Second, and most importantly, we should continue to focus our attention on the prevention and treatment of obesity, the best-documented risk factor for OSA (though, 15% of OSA diagnoses are in non-obese people –another study confirmed that 32% of men with documented OSA were non-obese with 11% having normal weight, and 12% of women were non-obese with 3% being of normal weight).


Primary Care Corner with Geoffrey Modest MD: Aspirin plus warfarin for afib and cad???

20 Nov, 14 | by EBM

A remarkably​ common clinical situation is the patient with coronary artery disease (CAD) who develops atrial fibrillation (afib). Do you keep the aspirin or other antiplatelet therapy (APT) when starting oral anticoagulants (OAC) such as warfarin??? In theory, these are 2 very different drugs targeting 2 different ongoing mechanisms: aspirin and other APTs inhibit platelets and arterial clots, acute coronary syndromes (ACS), etc; warfarin and other OACs inhibits blood coagulation which interferes with venous clots and venous thrombosis, pulmonary embolism, etc. But there is clearly a higher risk of bleeding if the combo of OAC and APT are used. The ORBIT-AF registry trial ( doi: 10.1161/CIRCULATIONAHA.113.002927), looked at 10126 patients from 176 US practices and found that the combination of ASA and OAC was common (35% of patients), that 39% had no history of documented atherosclerotic disease, and that there was an associated 53% increased risk of major bleeding and 52% increased risk of hospitalization for major bleeding. And (though the numbers were low) rates of ischemic events in the combo group vs the OAC group were the same:  0.48% (11 patients) in the combo group vs 0.38% in the OAC only group (16 patients) after 6 months followup.

The CORONOR study, a prospective French observational study with a really good name, looked at the same issue (​. 4184 consecutive CAD patients who had not had a recent MI or coronary revascularization (for at least one year) were followed for 2 years, also finding increased bleeding with the combo therapy and no difference in cardiac outcomes.


–Baseline characteristics: median 67 years old, 78% men, 62% prior MI, 99% prior cath, 58% multi-vessel disease, 86% prior revascularization procedure, and almost all on statin, b-blocker, ACE/ARB, and 77% on ASA, 40% clopidogrel, 21% both, and 11% on OAC mostly for atrial fib (65% of whom had combo OAC and antiplatelet drug, with n=342).

–Patients who had major bleeding events on multivariate analysis, not surprisingly, were on OAC (HR of 4.69), had diabetes (HR of 2.76), were older (HR of 1.04 per year), and had impaired renal function (HR of 0.98 for each ml/min/1.73 m2 decrease in eGFR).

–Risk of bleeding: when compared to antiplatelet monotherapy, the risk for bleeding from OAC alone was nonsignificantly higher (HR, 1.69)​, but was was much higher when OAC was combined with APT (HR, 7.30)

–Major bleeding sites: GI (55%), intracranial (20%). of the 51 patients with major bleeds (0.6%/yr), 18 events were fatal (35.3%)

–There was no difference in the combined endpoints of cardiovascular death, MI, or nonhemorrhagic stroke in those on combo therapy vs those on OAC alone (HR 1.15, with CI 0.58-2.27 and p=0.697). Their figure 3 shows that the incidence of these endpoints was pretty similar in the 2 groups over a period of 800 days.


A few comments:

–From several studies, aspirin by itself doesn’t work so well for atrial fibrillation.

–BUT, OACs seem pretty good for those with acute coronary syndromes or recent MIs. The WARIS II study of 3620 patients post-MI found that after 4 years the endpoint of death, nonfatal reinfarction or thromboembolic stroke was lower in those on warfarin than on aspirin alone for secondary prevention after MI.  The SPORTIF III and IV trials with patients with atrial fibrillation randomized to warfarin (INR target 2-3) vs ximelagatran, with low-dose aspirin allowed. Secondary analysis of the 14% of patients on aspirin (69% with CAD, 26% with previous stroke of TIA, 41% with LV dysfunction), found no difference in the combo therapy in terms of stroke or systemic embolism. And the rate of MI in the combo group (0.6%/yr) was not significantly different from warfarin alone (1.0%/yr), but the combo group had more major bleeding (3.9%/yr vs 2.3%/yr).

–The European Society of Cardiology’s 2010 guidelines on atrial fibrillation suggest OAC monotherapy in those with atrial fibrillation and stable vascular disease (ie, >1 year with no acute events) and that antiplatelet therapy “should not be prescribed” in those without a subsequent cardiovascular event. The American College of Chest Physicians clinical practice guidelines — see Chest. 2012 Feb; 141 (2 Suppl):e531S-75S) — similarly recommends that for those with atrial fib and high CHADS(2) score who do not have an intracoronary stent placed and are on OAC plus APT therapy for one year, to stop the APT.​

So, as per usual, would be great to have a well-designed randomized controlled trial of patients with atrial fibrillation plus coronary artery disease, comparing clinical outcomes in those assigned to the combo of OAC and antiplatelet therapy with those on OAC alone. So, what is one to do in the absence of a definitive study? given the increased likelihood of very serious consequences of major bleeds on the combo therapy (35% mortality in these events in the current study), it seems reasonable to me that  in those patients who have stable CAD (eg, 1 year after MI or coronary revascularization), the safest route is to use OAC alone… and in those with atrial fibrillation and no documented CAD, not even to consider APT in addition to OAC.



Primary Care Corner with Geoffrey Modest MD: Doctors questions not covered by 1st amendment

20 Nov, 14 | by EBM

I was surprised to find out this summer that there are significant potential legal constraints on what doctors (and presumably other providers) ask during a patient encounter and that doctors speaking with patients is outside of “free speech” protected by the constitution and instead everything said is considered “treatment”. Turns out that the government has “broad authority to prohibit doctors from asking questions on particular topics” 


There is a case in Florida (Wollschlaeger v Governor of Florida). As they note in the article, this law ‘concerned the constitutionality of the Florida Firearm Owners Privacy Act. That 2011 law threatens doctors with professional discipline if they ask patients whether they own guns or record the resulting information in a patient’s files when doing so is not “relevant” to a patient’s medical care’. ​They do note that there may be reasonable disagreement as to what “relevant” means. not particularly shocking is that the NRA believes that asking if there are guns in the house is none of the doctor’s business (though the Am Acad of Pediatrics does include this as one of the questions that should be asked!!). By a 2-1 vote, the 11th circuit held that doctors asking questions is in fact “treatment” and not protected by the First Amendment. What if another powerful lobby (e.g. tobacco, or alcohol) lobbied successfully in some states to bar doctors from asking relevant questions about these substances?  

This is not just an archaic, marginal, not-particularly-relevant issue. In Texas, the veterinary board used this argument to prohibit licensed veterinarians about giving advice about an animal they did not examine. In Kentucky, the Board of Examiners of Psychology “sent a cease-and-desist letter to a newspaper columnist claiming that his widely syndicated parenting column was unlicensed — hence criminal — practice of psychology” !!!!!!


Primary Care Corner with Geoffrey Modest MD: Angiotensin receptor blockers after ST-elevated MI

19 Nov, 14 | by EBM

There are good data suggesting that ACE inhibitors (ACE-I) are important in patients post STEMI (ST-segment elevated myocardial infarction) and are recommended therapy. However, no data are available on the role of angiotensin receptor blockers (ARBs), which are typically used for other ACE-I indications in the 15-20% of patients intolerant of ACE-I. A prospective Korean cohort study was just published in BMJ looking at their post STEMI treatment, assessing the outcomes of 6698 patients from 53 hospitals, all of whom had primary percutaneous coronary interventions (PCI) and LV ejection fractions >40%, of whom 1185 (17.7%) were given ARBs, 4564  (68.1%) given ACE-I and 949 (14.2%) neither.


Median followup of 371 days. This study involved propensity scoring, a technique used in nonrandomized trials to mathematically compensate for covariates that could have influenced treatment allocation (in this case, adjusting for differences in clinical, angiographic and procedural characteristics of the groups).


–In comparing the groups: those on ARBs (vs ACE-I) were older, had higher creatinine levels, had more LAD artery lesions in the area of infarct, though had lower prevalence of smoking. those in the “neither” group tended to have a higher risk profile at the time of the PCI

–Cardiac death or myocardial infarction occurred in 21 patients on ARBs (1.8%), 77 patients on ACE-I (1.7%), but in 33 on neither (3.5%).

–By propensity matching (1175 pairs), no significant difference between ARBs and ACE-I, with those endpoints in 1.8% in ARB group and 2.0% in ACE-Igroup. Propensity matching (803 pairs) between the ARB group and the neither group found the ARB group had 1.7% and the neither group 3.1%, a significant difference (p=0.03)

So, not a randomized controlled trial, but a large trial doing mathematical modeling to try to compensate for underlying potential differences in patients assigned to one of the 3 groups (ARB vs ACE-I vs neither), finding essentially no difference in the hard outcomes of cardiac death or MI between ABR or ACE-I, and that either was much better than neither. There were unanswered questions in this registry cohort, such as the relatively short followup time of 1 year, the dose of meds taken, why patients were assigned to the different groups, and ultimately the numbers of events was pretty small, limiting the statistical power. But, overall this study validates what probably most of us were doing anyway: try an ACE-I, using ARBs with patients who are ACE-intolerant.


Primary Care Corner with Geoffrey Modest MD: Avoid giving steroids with ritonivir!!

19 Nov, 14 | by EBM

The issue came up in terms of adverse effects (i.e., iatrogenic cushings) in patients on HIV cocktail containing ritonivir and use of even local steroids (the issue being that ritonivir is an extremely potent inhibitor of CYP 3A4 , leading to excessive increases in circulating levels of drugs metabolized through this system). I brought up this question with HIV providers in San Francisco (the HIV consult line) and received the following answer:

  1. Lots of problems with inhaled steroids, esp. fluticasone and budesonide. better results with beclomethasone, nasolide and flunisolide (I asked them is that just because fluticasone and budesonide are used more frequently, esp. since prior studies have found that these 2 drugs in particular have the highest tissue-to-blood levels — and they did not know, but said that the medical literature is replete with examples of cushings in using these drugs. of note, a drug company study did find a 350-fold increase of serum fluticasone levels with ritonavir). Probably the best option for many patients is to switch to a non-boosted protease inhibitor HIV cocktail, of which there now are many ones with (my opinion) dolutegravir-containing ones being the best. If unable to switch away from ritonavir, I would preferentially try the lowest dose of beclomethasone.
  1. In terms of intra-articular steroids, they felt there was no problem and that they do them in their HIV clinic. BUT, then Kevin Ard (former Brigham resident who did his continuity clinic at our health center, now ID-trained HIV specialist) chimed in with the following article refuting this claim (see DOI 10.1007/s15010-013-0506-z) — [thanks, Kevin]. This is a case report and review of the literature. major points:

–The case reported involved a patient with periradicular injection of 20mg of triamcinolone acetonide weekly for 6 weeks for lower back pain, who then developed severe cushingoid facies, central obesity, buffalo hump and proximal muscle weakness of her legs, as well as marked hypokalemia. Her HIV treatment included ritonivir 100mg bid for 6 years, with a CD4 of 820 and suppressed viral load. The symptoms of cushings began 6 week after her first injection. Work-up confirmed suppressed ACTH and cortisol levels.

–15 cases reported in the literature with suppression of hypothalamic-pituitary-adrenal (HPA) axis after injection with triamcinolone. ​Of the 15 reported cases, 9 were women, mean dose of injected triamcinolone was 97mg (40-240) and mean number of injections was 1.6 (1 injection in 9 cases), all were on 100-200 mg ritonavir

​–specifically 4 injections were intra-articular (which presumably has less systemic absorption than IM), 3 of these 4 had total dose of only 40 mg. symptoms began 2 weeks after injections and lasted 4-8 months afterwards. 2 patients with epidural injections had avascular necrosis. (So, a single injection of 40mg triamcinolone can lead to complete suppression of the HPA axis for up to 8 months!!)

 –methylprednisolone may be a reasonable therapeutic option (reduced dose methylprednisolone — 20-40 mg– was suggested in the radiology literature (ref 9 in the article) and “may present lower risk”, but with the caveat that there are insufficient cases to know for sure if this is safe. in that article they suggested checking an am cortisol level 2 weeks after the injection.

Note: this becomes more of an issue given the (rather remarkable and fortunate) aging of the HIV population, as HIV has evolved into a chronic condition, and living longer, as we know, is associated with more musculoskeletal problems.  But, the good news is that for most patients, there are great non-protease inhibitor options. So, seems that the safest thing would be to change the HIV regimen prior to putting someone on inhaled steroids or injectable ones. if that were not feasible, makes sense to use the ones not implicated above and at the lowest dose you can (with the rather large caveat that the identified cases are possibly found because of the more frequent use of inhaled fluticasone or injected triamcinolone in the population, instead of their really being the bad actors here).


Primary Care Corner with Geoffrey Modest MD: Placebo for coughs in kids, and an adult perspective

18 Nov, 14 | by EBM

An interesting article found that placebo was more effective than no treatment in kids with cough (see doi:10.1001/jamapediatrics.2014.1609). This was a 13 month study in 2 outpatient pediatric practices of kids aged 2-47 months who had nonspecific acute cough of less than 7 days. Parents were given surveys the day before and after the allotted treatment, which was either pasteurized agave nectar, caramel-colored placebo (each given 30 minutes prior to bedtime), or “no treatment”. All parents were instructed in routine care of the child: hydration, saline nasal spray, and use of acetaminophen/ibuprofen as needed (ie, the “no treatment” group did get treatment, just not a teaspoon of “meds”). The survey assessed cough frequency, cough severity, congestion severity, rhinorrhea severity and the cough effect on child and parent sleep.


–119 children completed the study, mean age 23 months, 50% female, 87% white non-Hispanic.

–No difference in any of these outcomes between administering agave nectar or placebo (though agave nectar was nonsignificantly better on all outcomes)

–Significant differences between either one of these two treatments and “no treatment” for all of the above outcomes, all with p<0.05

–Subgroup analysis suggested that the effects of agave nectar over placebo were somewhat more pronounced in those under 1 year old, though nonsignificant (but only 30 kids <1yo)


So, background is that acute cough is a remarkably common outpatient complaint in kids and a common reason for a healthcare visit, with nocturnal cough disturbing sleep in 88% of children and 72% of parents. Given current valid concerns about adverse events with medications as well as their unclear efficacy in young children (eg antihistamines, decongestants, antitussives), providers have limited medication options. There are some studies showing efficacy of honey, but that should not be used in kids <1yo for concerns about botulism. in above study, they chose agave nectar, since it sort of tastes like honey (though no data that it really works), does have some anti-inflammatory properties (as does honey), but lacks honey’s anti-oxidant effects.​

So, this article does suggest that placebo has a role for cough in kids (there was an analysis in 2002 suggesting that 85% of the treatment response in adults with cough was attributable to placebo effect). This is really not so surprising. There is an obvious interconnection between an individual’s clinical condition and their psychosocial state. to me, perhaps the most obvious example is the therapeutic effect of the provider-patient relationship itself. Many patients derive tremendous benefit without specific treatment. There are a slew of articles on the placebo effect, showing both the dramatic efficacy of placebo, the relatively frequent adverse effects as well, and effect even studies where the patients benefited even if aware that they were taking a placebo. there was an interesting article I ran across last year on this but, one of my favorite articles in the medical literature was an analysis of the CHARM trial, a placebo controlled trial of candasartan in patients with congestive heart failure (see Lancet 2005; 366: 2005–11), which found that those patients who adhered well to their treatment did well and had a lower all-cause mortality, and it didn’t matter whether they were assigned to candasartan or to placebo — ie, people who took their assigned medicine (who were probably more open to having a good effect from that medicine and perhaps more open to “taking care of themselves”) had similarly lower mortality independent of whether their assignment was to “active” medication or not…..


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