Sunday, September 28, 2008

An agenda to dissolve general Internal Medicine as a unique specialty? Say it isn’t so!

In several previous posts I’ve mentioned the external forces and unintended consequences that threaten Internal Medicine as a specialty. Well now it appears their’s a deliberate agenda afoot to eliminate the specialty of IM. That agenda is stated in an article from a recent issue of the journal Academic Medicine. (You can read the full text here if you’re registered with Medscape). The author has an obvious conflict of interest: He’s a professor and chair of Family Medicine.

Simply stated, the proposal is to “merge” the specialties of Family Practice, Pediatrics and Internal Medicine, a euphemism for the dissolution of Peds and IM as unique, stand alone specialties.

For example:

Because the new discipline would provide continuous, comprehensive health care for an individual through his or her entire lifetime, it would naturally incorporate models and processes of care that integrate the known effects of genetic factors, health risks, lifestyle choices, prenatal and natal factors, and disease processes over a person's complete lifespan. From a professional perspective, additional efficiency and cost saving would also be gained by creating a single set of standards and mechanisms for initial board certification and continued maintenance of certification

So what’s this “new discipline” he’s describing? It’s Family Practice!

How will medical students making career choices react to this? They’ll think long and hard before entering a specialty that’s under consideration for dissolution in 20 years!

I issue this challenge to the American College of Physicians and the Society of General Internal Medicine: Tell us in no uncertain terms that you intend to support general Internal Medicine as a unique specialty for the long haul. If you can’t do that it’s time for us to form a new organization that will represent the highest traditions and professional interests of our specialty.

Saturday, September 27, 2008

Update in selected ID topics

Vancomycin vs metronidazole for severe C. diff.

Mortality concerns with cefepime.

Treatment of Bells palsy.

Posaconazole (Noxafil) vs older azole antifungals for prevention of fungal infections in chemotherapy for MDS and AML.

Anidulafungin (Eraxis) vs fluconazole in invasive candidiasis.

And more. Via CCJM.

Glycemic control in critical illness---yet another negative trial

This one came out after the JAMA meta-analysis. It’s getting monotonous

Friday, September 26, 2008

Diagnosis by You Tube

ER types are fond of saying they operate on sheer adrenaline. Maybe that adrenaline, driven by leftist passion, is why Shadowfax is so quick to suggest that John McCain has had a stroke. Given that the portion of the 7th nerve nucleus representing the upper facial muscles has bilateral cortico-bulbar representation how, pray tell, could a stroke wipe out those muscles? (OK, so is it a brain stem stroke, wiping out the 7th nucleus with no detectable central manifestations? That seems a bit of a stretch). Well, if I present to an ER with Bell’s Palsy please don’t anybody give me TPA.

More from Kevin and Orac.

Update: Many bloggers and commenters have emphasized McCain's ptosis in addition to the peripheral 7th nerve manifestations. As Steve Novella has pointed out:

Isolated ptosis would be an extremely uncommon presentation of a stroke or tumor, other signs would almost certainly be present. Ptosis is most often physiological - meaning normal for that person. This type of ptosis can also come and go, so it may be more noticable at some times than at others.


Novella points out that benign ptosis is very common and may be physiologic. If the chance of a stroke causing upper facial weakness is slim, the chance of one causing this and ptosis at the same time are slim to none.

Thursday, September 25, 2008

Is industry supported CME biased?

According to Daniel Carlat’s spin of a recent ACCME literature review, yes, hopelessly so. But the review concluded: “…to date there is no empirical evidence to support or refute the hypothesis that CME activities are biased.” Here’s what the review actually found:

Pharmaceutical promotions influence attitudes and prescribing---something we’ve known for a long time.

Very little of the literature focuses on CME and almost all the studies predated the first standards (1992) for commercially supported CME. Moreover, no studies directly examined the content of the CME activities in question.

Although not the focus of the review, the authors comment that no studies have assessed the effect on patient outcomes. Those are the outcomes that matter, aren’t they?

Glycemic control in critical illness: we know even less than we thought we did

Glycemic targets for hospitalized patients seem to be moving targets. I tried to make sense of it all with a review of evidence late last year. At that time evidence was on the negative side for strict glycemic control (110 mg/dl or below). The optimal glycemic target was not known, but many experts thought 150 mg/dl was reasonable. Now comes this meta-analysis in JAMA which would seem to be yet another nail in the coffin of tight glycemic control in hospitalized patients. As other studies had shown, tight glycemic control produced no overall benefit beyond “usual care.” What was new in this study was that a “moderate” glycemic target of 150 mg/dl, the consensus of many experts today, was no better than very tight control or usual care.

So what’s the current status of glycemic control in hospitalized patients?

The optimal target may be somewhere above 150 mg/dl but we don’t know what that threshold is. At some level, glycemic control matters. In the methods section of the meta-analysis it was implied that in some studies usual care glucose may differ from tight control levels by as little as 20 mg/dl. Thus even usual control in some negative clinical trials may be better than what we achieve in the community. The ongoing NICE SUGAR trial, comparing very tight control with a target of 180 mg/dl, may shed further light.

Background:

Accompanying JAMA editorial.

Med Rants weighs in.

Med Page Today has a synopsis and an interview with the lead investigator.

Tuesday, September 23, 2008

How will the new data on diuretics and hypertension change our practice?

A new study, which I linked Sunday, showed no benefits of thiazide diuretics with respect to coronary mortality or sudden cardiac death (SCD) in the treatment of hypertension. More concerning was a signal of increased SCD attributable to thiazides. Although the trend fell short of statistical significance it was in the same direction as previous findings and had biologic plausibility based on animal models and the known arrhythmic effects of hypokalemia.

That was the bad news about thiazides. The good news was that combinations of thiazide and potassium sparing diuretics were associated with unprecedented reductions in coronary mortality (OR .59, .44-.78) and SCD (OR .60, .38-.94). This lends credence to the long held suspicion that thiazide induced hypokalemia is dangerous and brings a decades old gimmick by the pharmaceutical companies (remember dyazide, maxzide and moduretic?) into the realm of evidence based medicine.

The findings raise many questions. First, we know that placebo controlled investigation conducted in a variety of patient populations demonstrates cardiac mortality benefits for angiotensin converting enzyme inhibitors (ACEI) but not for thiazide diuretics. Why, then, did ALLHAT fail to show superiority of ACEIs over thiazides in a head to head comparison? This can be attributed to flaws in ALLHAT’s design which stacked the deck against ACEI’s. First, the ACEI was started as monotherapy in a population containing large numbers of African Americans, a group known to be resistant to ACEI monotherapy. Second, the step 2 and 3 drugs provided for in the design were arguably more synergistic with thiazide diuretics than with ACEI’s.

Although it remains to be seen whether these findings are in time to influence the soon to be published hypertension guideline update there are lessons we can apply now. Thiazide monotherapy can no longer be considered the starting treatment of choice. If not contraindicated by renal failure or hyperkalemia a thiazide/potassium sparing diuretic combination is favored. While the combination can mitigate against hypokalemia, it can still occur. Moreover, hyperkalemia and hyponatremia remain potential hazards, necessitating frequent laboratory monitoring.

Thiazides are known to work well when prescribed together with ACEI’s and angiotensin receptor blockers (ARB’s). Should this practice change? One option would be to combine an ACEI or ARB with the combination diuretic. This can be safely done with careful monitoring, although widespread adoption of the practice might result in adverse outcomes associated with hyperkalemia as has recently been observed when potassium sparing diuretics are combined with ACEI’s in the treatment of heart failure. Another option would be to continue the use of ACEI or ARB/thiazide combinations on the premise that the ACEI or ARB offsets the danger of hypokalemia. Careful biochemical monitoring is important in any case. Dr. John Oates, senior investigator of the new study, discusses the ins and outs for clinical practice in light of the new data in an interview with Medpage Today.

Sunday, September 21, 2008

Thiazide diuretics, hypokalemia and sudden cardiac death

If this paper had been about some new patented block buster product the New York Times and the Pharma-hating blogs would have been all over it in a matter of hours. Meta-analyses of RCTs in hypertension using thiazide diuretics and thiazide-potassium sparing combination drugs were conducted. The findings:

Significant reductions in both coronary mortality and SCD were observed in the overview of trials in which elderly patients received an ENaC inhibitor/HCTZ combination. The odds ratio (OR) for coronary mortality was 0.59 (95% confidence
interval [CI], 0.44 to 0.78) and for SCD was 0.60 (95% CI, 0.38 to 0.94). In contrast, an overview of the trials using thiazide diuretics alone showed no significant reductions of either coronary mortality (OR, 0.94; 95% CI, 0.81 to 1.09) or SCD (OR, 1.27; 95% CI, 0.93 to 1.75). Use of an ENaC inhibitor combined with HCTZ for treatment of hypertension in the elderly results in favorable effects on coronary mortality and SCD.

All I can say is wow. The point estimate for increased sudden cardiac death attributable to thiazide diuretics was 27%! Why the media silence? Because there’s no Phama-bashing opportunity. Recall that the media outlets have a love affair with ALLHAT, a large hypertension trial which they overhyped as saying thiazides were the clear first choice as starting agents for hypertension, and which they trumpeted as a victory of science over evil drug company marketing.

Although the results fell short of statistical significance the finding (in today’s parlance a “signal”) was concerning because it lined up with what was previously known on the subject. The paper adds to what we knew before because some of the data on SCD were previously unpublished. For me it was a see I told you so moment.

John Oates, M.D, senior author of the paper, interviewed in Vanderbilt Medical Center’s Reporter, discussed the strength of the finding when taken into account with previous information:

Even though the increase was not statistically significant, it was “going in the direction in which you didn't want to go,” Oates said.

Observational studies previously had found an increase in sudden cardiac death in patients taking a thiazide diuretic alone, and one showed that sudden death was greater at higher doses of thiazides, he said.

Studies in animal models of heart attacks also have demonstrated that low potassium levels (caused by thiazide diuretics) can spark the abnormal heart rhythms that lead to sudden death.

Do thiazide diuretics given alone have an adverse effect of increasing the risk of sudden cardiac death in patients with high blood pressure? It's possible.
“There's biologic plausibility for an adverse effect of the thiazides,” Oates said.

Oates is a giant in hypertension and for decades has done bench-to-bedside work in the field. Given his normally reserved manner this statement was startling concerning the finding of increased SCD:

“If it's true, it's probably the largest adverse effect in the history of modern pharmacology. The number of individuals affected over the last 50 years would be staggering.”

Too bad there’s no corporate interest to bat down. It’d be headline making stuff. But the implications for public health are huge. It’s really a mix of good news and bad news for thiazides. When combined with potassium sparing diuretics the benefits are unprecedented.

The promoters of ALLHAT said thiazides were the clear starting agents of choice for treatment of hypertension. They turned it into dogma. They spun it. They promoted it. They were wrong.

Will the soon to be released update of the JNC guidelines incorporate the new findings? How should they change our management of hypertension? I’ll post my thoughts on that topic in the next few days.

Saturday, September 20, 2008

Ophthalmology for the generalist: optic neuritis

The most important signs are a monocular decrease in acuity, monocular decrease in color vision and an afferent papillary defect (Marcus Gunn pupil). Here is a case discussion and general review from the American Journal of Emergency Medicine.

Friday, September 19, 2008

Meet MRFIT

Who’s he? He, or, rather, it, was the Multiple Risk Factor Intervention Trial, an all too often forgotten landmark study. Although conducted before the age of EBM this trial, its registry and its derivative works are among the most important in all of medicine. So it’s “fitting” that, as we approach the 22nd anniversary of its publication, JAMA ran a commentary and a free link to the original article. In this commentary JAMA reported the 25 year follow up data on the MRFIT cohort. This follow up analysis verified the findings of the original study:

The main finding of our report was that the relationship between serum cholesterol and CHD mortality is continuous, graded, and strong; ie, CHD risk is progressively higher at every cholesterol level from 160 mg/dL and higher levels, with no threshold. This finding prevails with 5-, 10-, 15-, 20-, and 25-year follow-up and for the first, second, third, fourth, and fifth 5-year follow-up periods. These robust results, controlled for age, systolic blood pressure, number of cigarettes smoked per day, diabetes status, race and ethnicity, and study geographic site, prevailed over the 25-year follow-up with only modest attenuation in quantitative strength of relative risk from higher serum cholesterol levels and with increase over time in absolute excess risk from higher serum cholesterol levels as the CHD death rate increased annually.

The commentary points out that MRFIT debunked some myths which, unfortunately, remain popular today. One such myth is that there is a threshold value of serum cholesterol for cardiovascular risk. In the 60’s and 70’s it was 300. Later it was 200 + age, then 200. Today, the idea that lower and lower targets are better is dismissed as “disease mongering.” Yet, as MRFIT makes clear there’s no threshold. Or if there is it’s lower than anything we’ve accomplished in the Western world. The Framingham study showed the same thing. Now those were observational studies, but innumerable diet and drug trials, lowering cholesterol by various means, also showed it.

Another debunked myth, recently resurrected by the Vitorin study (a finding that needs more nuanced understanding) and the pleiotropic effects of statins (yes, they do exist) is that “cholesterol lowering doesn’t matter.”

Why don’t we get it? Because it isn’t that simple and because the people who are leading the popular debate don’t have a clue about the notions of relative risk, absolute risk and population attributable risk. I tried my best to do some debunking last year.

One of the derivative works of MRFIT I mentioned was an analysis of hypertension treatment which demonstrated an increased risk of death in patients treated with thiazides who had electrocardiographic abnormalities. It was probably this concern rather than pharmaceutical company marketing of newer drugs that led to a decline in thiazide prescribing beginning in the 1980’s.

The war on industry supported CME

---is heating up. It’s rather odd that Dr. Daniel Carlat, just yesterday, bemoaned the fact that “We’ve descended to the level of name-calling now” when he was doing some name-calling weeks ago (see this post and its references to dinosaurs and cornered animals).

What’s Dr. Carlat worked up about? He’s taking aim at the Center for Medicine in the public Interest (which he calls a reactionary unthink tank) for their upcoming seminar “Industry Support for Continuing Education of Healthcare Professionals: An evidence-Based Evaluation.” And what, pray tell, is wrong with making an appeal to evidence in this debate? It’s about time somebody did. All the agitators have to offer are personal attacks (did you know that if you favor industry supported CME you’re a dinosaur?) and ad hominem arguments about people’s commercial affiliations (true to form, for the symposium in question Dr. Carlat provides the usual litany). OK, Dr. Carlat, they’re biased. Just like you and me. We get that. But I’m concerned that you’re shedding more heat than light on this debate. Why not attack the arguments instead of the people?

Well, today Drugwonks, the CMPI blog, responded in kind. Go and read the whole exchange. It’s rich. This will be great fun to watch for a while, but eventually I’d like to see a sober evaluation of the question based on evidence along with a little respect for the law of unintended consequences.

Wednesday, September 17, 2008

Benefits and safety of diabetes drugs

In a Perspective piece in this week’s NEJM Allison B. Goldfine, MD offers nuanced suggestions for regulatory policy and research.

If you’re having trouble sorting through the morass of opinions and controversies or don’t have the inclination or the time to read extensively on the subject, don’t worry. I’ve done it for you. Here’s the quickest read I can offer. It’s a bit of an oversimplification but it doesn’t completely distort:

The benefits of drugs to achieve glycemic control in type 2 diabetes must be considered in the perspective of time-----


Days to weeks---> Relief of polyuria, visual blurring, dehydration.

Years---> Microvascular benefit. Magic number for Hgb A1C= 7. Enhanced by blood pressure control, agent specific.

Decades---> Macrovascular benefit. Magic number= 7. Enhanced by blood pressure control, diet, exercise and drugs to manage insulin resistance and dyslipidemia.


There you have it.

Tuesday, September 16, 2008

Thirteen years after the NINDS trial

---we still can’t agree on TPA.

From Emergency Medicine News:

The results of the National Institute of Neurological Disorders and Stroke (NINDS) trial on tissue plasminogen activator were published in 1995, and since then it's been lauded as a landmark study time and again by almost everyone but emergency physicians. (N Engl J Med 1995; 333[24]:1581.) The most recent American Heart Association (AHA) guidelines, supported by the American Academy of Neurology (AAN), recommend intravenous rtPA for selected patients within three hours of ischemic stroke onset. (Stroke 2007;38[5]:1655.)………

From an emergency medicine perspective, the American Academy of Emergency Medicine and the Society of Academic Emergency Medicine unequivocally state that tPA is not standard of care due to insufficient evidence concerning risk and benefit. The American College of Emergency Physicians' position statement also cites insufficient evidence for tPA as standard of care when systems are not in place to ensure that inclusion/exclusion criteria…are followed. ACEP is currently reviewing its clinical policy on stroke, but it won't be finalized until the end of this year or early 2009.

W. Richard Bukata, MD, a clinical professor of emergency medicine at Los Angeles County-USC Medical Center, is quick to point out the discrepancy between the medical groups. So what we have, honestly, is a disagreement between all the emergency physician organizations and the American Heart Association and the American Academy of Neurology, he said.

So 13 years after NINDS, we’re still fighting, despite more recent evidence from community experience validating the use of TPA.

Monday, September 15, 2008

Anthony DeMaria defends industry support of CME

Anthony DeMaria, MD is editor of the Journal of the American College of Cardiology (JACC). He’s also chief of Cardiology at UCSD. He probably has better things to do than read this blog. Yet, (if you’ll indulge me in a little self aggrandizement) his recent JACC editorial echoes the exact points I’ve made repeatedly in these pages. This may sound familiar:

Most will agree that the separation between the composition of educational programs and commercial interests has not always been perfect. Perhaps it shouldn't be. It seems to me unreasonable to think that a company that produces angioplasty catheters would be interested in sponsoring a program on obstetrics. As pointed out in a statement from the Council of Medical Specialty Societies, the Gerber Foundation, dedicated to advance the quality of life in infants and young children, would not likely be interested in supporting programs on Alzheimer's Disease.

And this:

…I worry that termination of all commercial support is a major overreaction. The goal that underlies eliminating commercial funding—that is, to insure program objectivity—can be accomplished by existing CME-granting agencies. Certainly medical societies and academic institutions should be capable of excluding inappropriate industry impact. In addition, is not clear how or if the financial support for CME provided by industry could be replaced. Without these funds, important opportunities to increase knowledge might be unavailable to busy clinicians, thereby denying their patients the benefits of this learning.

Dr. Daniel Carlat, author of the Carlat Psychiatry Blog, took this swipe at the last sentence:

By the way, cardiologists pull in an average of $270,000/year. But according to Dr. Demaria, they are still too poor to pay for their own education
:

It’s a misrepresentation of what he said, coming across as little more than a cheap shot. I don’t need to explain what Dr. DeMaria actually meant. The editorial goes on to cite an important consequence of eliminating industry support for CME:

It is not known what industry would do with the money they are currently directing to CME. However, it would likely go to more promotional events and/or direct-to-consumer advertising, neither of which addresses the concerns that have been raised. In fact, such alternate venues would probably only serve to further increase the use of products.

Except for the words likely and probably that’s exactly what I’ve said before. This unintended consequence is easy to spot but has been largely ignored.

He goes on:

I have a further, even more basic, reservation about the proposal to end commercial support for CME. Inherent in such an action is the idea that physicians are like sheep: easily led and without the ability to recognize biased or slanted information. I find this demeaning to the profession. In my experience, physicians are more skeptical than naïve; by nature they are not anxious to accept, but rather are waiting to be convinced. Given the competitive demands entailed in becoming a physician, we are likely intelligent enough to recognize bias when it is present.

Indeed. Physicians are mind numbed idiots. That’s the major premise underlying this inquisition. Proponents of that tired argument are armed with “studies”---soft survey and sales data on the effects of drug rep promotions which have nothing to do with CME or patient outcomes. I’ve previously addressed those arguments, and the evidence behind them is insufficient for the advocates of industry free CME to sustain their burden of proof.

Dr. Roy Poses of Health Care Renewal also weighed in. Both Poses and Carlat accused DeMaria of claiming that disclosure solves everything, then criticized his own failure to disclose. Either they didn’t read the editorial carefully or they erected a straw man so they could ridicule DeMaria’s lack of disclosure. DeMaria makes no assertion that disclosure is a “panacea” or that it resolves anything. Nothing of the sort. I read the editorial three times. It is clearly an opinion piece. It makes no claims. It makes no pretense of objectivity and advocates no policy changes. It merely raises questions and issues a plea that we “think long and hard” before making sweeping changes.

The continuing assault on industry supported CME

This time Dr. Arnold Relman takes a stab in JAMA. For years Relman has been on a crusade to eliminate all forms of industry supported CME and these are his usual talking points. The piece is full of opinion about what should or ought to happen but offers no evidence to support its claims.

Via Carlat Psychiatry Blog.

Friday, September 12, 2008

Rapid response roles spark controversy

What are the indications for calling the rapid response team? What is the role of hospitalists? When should the patient be handed back to the attending physician? What is the attending physician’s responsibility?

These are the current topics of conversation among hospitalist types trying to better define how RRTs should work as reported in Today’s Hospitalist. There were some common threads. Hospitalists are sometimes inappropriately put in the role of “house physicians” for all in hospital emergencies. The function of RRTs and the role of hospitalists need better definition.

I’ve previously expressed my skepticism on this issue, here and here. Joint Commission and IHI claim that RRTs save lives. Hospitals and hospitalist programs all across the country trumpet the success of their programs. Yet, the general concept of the RRT has not been vetted by high level evidence. When applied to limited situations (eg early goal directed therapy) RRTs may save lives, but no one can make the general claim that just because your hospital has a RRT lives will be saved.

Red eyes---how to treat, when to refer

Via Cleveland Clinic Journal of Medicine.

Thursday, September 11, 2008

New findings in 10 year UKPDS follow up

This just came out ahead of print in NEJM. I’ll get it posted now before the news media distort it.

The findings, based on 10 year follow up of the United Kingdom Prospective Diabetes Study (UKPDS), reported in two NEJM papers, are these:

Differences in glycemic control were lost within a year of the end of the original trial. Nevertheless, microvascular benefits persisted at 10 years and, somewhat surprisingly, macrovascular benefits (reduced MI) emerged in the insulin/sulfonylurea group and persisted in the metformin group.

Differences in BP control were lost within two years of trial completion and at 10 years there was no difference in outcomes attributable to blood pressure control.

The popular media are liable to report “blood sugar control prevents heart attacks after all, BP control doesn’t.” For those who care to think beyond the sound bite here’s my take on what’s really significant:

The benefits of glycemic control are realized years later. The benefits of BP control are shorter term, realized more in real time and require ongoing treatment.

The benefits in UKPDS were associated with hemoglobin A1C levels of 7. More aggressive glycemic targets in type 2 diabetes may be associated with macrovascular harm.

This is the first evidence that glycemic control using insulin or sulfonylureas in DM 2 may produce macrovascular benefit.

If we require that drugs for DM 2 be vetted for clinical outcomes some effective drugs may be doomed. Sulfonylureas have carried boxed warnings about macrovascular risk for decades. In today’s regulatory climate they might never have been approved.

The benefits of glycemic control included a reduction in mortality and should put to rest the debate over whether the outcomes associated with glycemic control in DM 2 are “outcomes that matter.”

Tuesday, September 09, 2008

The debate on the hospitalist model

---between Dr. Mark Williams and Dr. Robert Centor just got some more coverage in an interview with Dr. Williams in the September issue of Today’s Hospitalist.

Dr. Williams believes it’s a slam dunk that the hospitalist model is better for patients. Concerning the debate, he said:

That has been settled, and the answer is an unequivocal “yes.” The question we should be debating is not whether to use hospitalists but how to best optimize their use.

Concerning the traditional internist, he believes it’s an anachronism:

I think the concept of a “complete internist” is an anachronism that may have applied 20-plus years ago. While there may be a few internists still like that in rural private practice, it’s clear that optimal care for hospitalized patients is having a hospitalist on-site who’s available to respond to emergencies.

I gave my take in a post I wrote shortly after the debate was published. I believe in the traditional internal medicine model. It’s not for everyone but there are many who can do it well provided the organizational structure is in place. If the debate about the hospitalist model of care is settled it’s not based on evidence. Although the idea of having a hospital based physician on site 24/7 make sense and is appealing the advantages are counterbalanced by the downsides of discontinuity of care, which is why studies have repeatedly failed to show that the hospitalist model improves outcomes for patients.

What has settled the debate is the reality of economics and other external factors, not patient outcome based evidence. The final finishing blow to the traditional model will come if, has been speculated, some version of the Kennedy-Gephardt bill is resurrected.

The Today’s Hospitalist piece, by the way, cited Dr. Centor with an interesting observation: we don’t know the long term outcomes of the hospitalist model. I’ve mentioned this fact a time or two before. One of the larger and more methodologically sound comparative studies, presented at SMH 2005, showed no difference in efficiency or outcomes. The study has not, as far as I’m aware, been published in a Medline indexed journal but the abstract of the SMH presentation, published in this issue of The Hospitalist, tells us that “mortality data up to one year after admission are pending.” That was 3 years ago. Have those data been published or presented anywhere?

Eosinopenia as a predictor of sepsis

Good test characteristics for patients newly admitted to the ICU were demonstrated in this study from Critical Care. (I like to add a manual differential on such patients. That gives you a band count. The combined information may be a more powerful predictor---not evidence based that I’m aware but it makes sense). Think of it as a quick and dirty substitute for the much more expensive and less readily available procalcitonin assay.

Via Medscape.

Monday, September 08, 2008

CME conference---final thoughts

This post will read like a promotion for the Tutorials in the Tetons Update in Cardiovascular Disease. I have a purely personal interest in promoting the meeting. I want it to continue so I can attend future conferences. After first attending in 1981 I was hooked and have returned almost every year since. Traveling to this spectacularly beautiful piece of the planet makes me feel expansive and affords a unique opportunity for learning away from the distractions and pressures of home. It is a time of heightened creative energy and intense personal reflection. It has also been a time for thinking extensively on medicine and life in new ways.









That’s what the late Robert Eliot, MD, had in mind when he started this meeting in 1974. Eliot was chief of Cardiology at the University of Nebraska School of Medicine. Having experienced a myocardial infarction in his 40s without any of the traditional risk factors, he examined his lifestyle and concluded that stress had played a role in the event. He went on to develop an intense personal and professional interest in the relation between mental stress and cardiovascular disease, publishing many papers and two popular books for the lay public. He spent a good portion of his professional life studying and treating stress in physicians. He intended for his annual meeting to be both educational and therapeutic for attendees. Indeed it was. The tradition of a CME course which is both educational and therapeutic for health care professionals has continued since Eliot’s death. A memorial lectureship established in his honor funds one of the speakers at each meeting.


One or two talks each year are devoted to cutting edge topics in their infancy or even years ahead of widespread clinical application. For the last few years these presentations have covered genomics applied to cardiovascular disease. I always feel ahead of the curve in new advances after attending. It was at this conference that I first learned of thrombolysis in myocardial infarction, magnetic resonance imaging, positron emission tomography and Brugada syndrome. It was where I first learned about angioscopic data showing that acute coronary syndromes were caused by thrombi in dynamic stages of formation and lysis, providing a rationale for anticoagulation as a “new” therapeutic approach to acute coronary syndromes.


Each year there is at least one presentation devoted to basic clinical skills such as ECG interpretation, X ray interpretation or auscultation. Over the years I have been exposed to a “parade of stars” at this meeting including Barney Marriott, Proctor Harvey, Michael Debakey, William Casteli and many others.


Not all the presentations are spectacular in this sense. The core material covers the “nuts and bolts” of cardiology of importance to generalist physicians. The conference should be useful to cardiologists, emergency physicians, internists, family practitioners, hospitalists, nurses and physician extenders.








The meeting is ideally situated at one of the spectacular national park lodges, the Jackson Lake Lodge, right in the middle of Grand Teton National Park. It is equidistant and an easy drive from Yellowstone Park to the North and Jackson Wyoming (shopping and excellent restaurants ) to the South. The course, 12-14 CME hours over three days, allows ample time to enjoy the surroundings. Come early, stay late and combine CME and vacation. It’s an ideal place to bring the family.


Consider attending next year. Save the dates for the 2009 conference: August 29-Sept 1. To get on the mailing list or find more information about the conference go to the web site of the sponsoring institution, St. Joseph’s Hospital, Atlanta, Georgia. I’ll be posting updates here about next year’s conference as they become available. Feel free to email me about accommodations, travel and other details. I’d love to see some of you there next year!

Hospitalist comanagement---ask the right questions

I have expressed my concern in previous posts about the recent uncritical promotion of this idea. Amidst the unbridled enthusiasm for comanagement at SHM 2008 there was one voice of healthy skepticism: that of Dr. Eric Siegal of Cogent Healthcare, whose talk is profiled in the August 2008 issue of Today’s Hospitalist. His take? Many of the arguments for comanagement are based more on dogma than science. While comanagement can be a good thing it is not beneficial in some situations. Before embarking on comanagement hospitalst program directors must ask some tough questions.

CME conference day 3

Day 3 of the Tutorials in the Tetons Update in Cardiovascular Disease took place September 1. I’m a little late posting this due to a whirlwind schedule since I returned home.

Dr. Gordon Ewy talked on hypertension, focusing on issues in clinical epidemiology and the treatment of resistant hypertension. Despite greater awareness of the problem of hypertension it remains under treated. This increased awareness has been characterized as disease mongering. Yet, the benefits of treatment to new and more aggressive targets are undisputed. The old definition of normal systolic blood pressure as “100+age” has been thoroughly debunked. He pointed out that hypertension treatment is supported by the greatest number of trials in all of medicine. Evidence is still mounting that hypertension treatment matters, even in the very old. The presentation was scientifically rigorous and free of commercial bias. Drug classes rather than drug names were emphasized. Almost all drug classes for hypertension treatment have generic drugs available.





Dr. Barry Greenberg, Professor of Medicine and Director of the Advanced Heart Failure Treatment Program at UCSD, gave a presentation on heart failure. There were several points of interest to hospitalists. Much of the new information on in patient treatment and outcomes comes from two large databases, Optimize-HF and ADHERE. From these studies we have learned that bad outcomes occur late. Whereas in patient mortality is 4%, the combined readmission and mortality rate at 60-90 days is almost 40%. In these registries the average length of hospital stay was around 4 days. That’s lower than some reports from previous years and seems awfully low to me for such desperately ill patients. These facts, combined with data from older studies that late heart failure outcomes are worse when fluid overload is incompletely treated during hospitalization raise a provocative question: are we discharging heart failure patients too soon?

A related issue is the extent to which hospitalists should adjust patients’ long term medications before discharge. We are under intense pressure to reduce length of stay. The rule of thumb is stabilize, discharge and let the out patient PCP deal with long term medications. While this approach may improve our utilization stats it may not produce the best outcomes. In an earlier presentation Greenberg cited this study showing that if beta blockers are not started in the hospital they are unlikely to be started at all. Although it was a study of patients with myocardial infarction its lessons apply to heart failure and other conditions.

Dr. Greenberg made brief mention of the Neseritide controversy and suggested we reserve judgment pending completion of the ASCEND-HF study which aims to resolve questions regarding Neseritide’s effect on 30 day mortality.

Dr. Greenberg’s presentation was balanced and free of commercial bias. No suggestion was made favoring one member of a class of drugs over another. My review of primary sources indicates that the content was scientifically rigorous.

The conference ended with bench-to-bedside discussions by Dr. Nicolas A. F. Chronos and Dr. J. Jeffrey Marshall on platelet disorders, spanning pathophysiology, aspirin and clopidogrel resistance, and a review of the new guidelines. I’ll be posting final impressions about the conference shortly.



Wednesday, September 03, 2008

Tuesday, September 02, 2008

CME conference day 2

I’m really behind in blogging about the CME conference, having spent much of the last couple of days sightseeing (spent all day today in Yellowstone Park). Day 2 of the conference took place August 31. Every year one special lecture (the Robert Eliot memorial lecture---more on Dr. Eliot in a future post) is devoted to lifestyle issues in the prevention of cardiovascular disease for health care professionals to apply to their patients’ lives and their own lives. This year’s featured speaker was Dr. Arthur Agatston of South Beach diet fame. He had no financial conflicts of interest (he did not mention South Beach and his lecture was funded by a foundation in memory of Dr. Eliot) and discussed the obesity epidemic as well as some initiatives he’s involved in in the public schools to combat childhood obesity.

Every year there are a couple of lectures devoted to cutting edge topics which are in their infancy in clinical application. For the last several years these presentations have covered topics in the genomics of cardiovascular risk assessment and drug therapy. Dr. H. Robert Superko presented some exciting developments on two novel heart disease genetic variants, KIF6 and 9p21. The fields of cardiovascular genomics and pharmacogenomics are a few years away from direct widespread clinical application, but they’ll soon turn EBM upside down. Gone will be the era of “one size fits all” drug therapy and risk assessments based on population studies. Get ready! (Actually, the era of pharmacogenomics is already here for limited applications. The package insert for warfarin, for example, says:

The lower initiation doses should be considered for patients with certain genetic variations in CYP2C9 and VKORC1 enzymes as well as for elderly and/or debilitated patients and patients ….


Dr. Gordon Ewy, chief of cardiology at the University of Arizona and pioneer of CPR, gave an update on his research in favor of the “new CPR.” I have blogged about Ewy’s work many times, starting here. Ewy is years ahead of his time in CPR. The guideline writers have been slow to adopt his work. If you’re going to collapse from a heart attack, do it in Arizona!

None of the presentations on day 2 emphasized pharmaceutical products. The content was relevant, scientifically rigorous and free of commercial bias.

I’m flying home tomorrow, and back to a demanding hospitalist schedule Thursday. I’ll blog about day 3 and post some general observations about the conference (along with a few more images from Grand Teton and Yellowstone) in a few days!

Fluconazole prophylaxis

---in non-neutropenic ICU patients who remain febrile despite broad spectrum antibiotics is not beneficial according to this study.

Monday, September 01, 2008

Distinguishing benign from serious causes of back pain

Physical therapy, analgesics and tincture of time may not be the correct approach. This review helps explain how to distinguish benign from serious causes.

Via Cleveland Clinic Journal of Medicine.