There are no facts

Musings from a professional life driven by curiosity and skepticism. Thoughts and opinions expressed here are solely my own and do not reflect the positions of my employer.

[paulberglmd]

Don’t just blame me or the patients for the opioid epidemic

Much attention has been drawn to the opioid epidemic as of late: Obama posed for pictures with music artist and recovered addict Macklemore. The House of Representatives passed a series of (unfunded) legislation to curb the problem. The CDC recently issued new guidelines on opioid use. Questions abound about the role of opioids in the untimely death of Prince. The list goes on.

Lots of solutions are batted around, but one of the biggest problems that is still not adequately discussed is the horrendously distorted reimbursement system for medical care in this country. 

First of all, I can prescribe a number of opioid medications without any fight from insurance companies. On the other hand, any time I try to prescribe highly effective adjunct treatments like topical NSAID’s, I am nearly certain to get a prior authorization request from my ole pal the insurance company! Sure, some topical treatments -- like lidocaine -- are of less certain value for pain issues, but isn’t use of chronic opioids just as dubious? I need to be able to easily prescribe effective topical treatments without a fight with insurance or Medicare; then I could prescribe fewer opioids.

The reimbursement system also provides a disincentive for other effective treatments for chronic painful disorders. What if insurance plans actually paid for supervised exercise plans for painful conditions that are known to improve with exercise like osteoarthritis of the hip and knee, peripheral arterial disease, or fibromyalgia? You may argue that most plans pick up the tab for physical therapy (PT), but PT is a short-term fix. I would love to see Medicare and  insurance plans subsidize water exercise classes for arthritis patients or an aerobic exercise plan for my fibromyaglia patients.

Finally, there are simply not enough resources in primary care settings to screen for opioid addiction or dependence. I personally believe the most common path to opioid addiction starts with a somewhat legitimate use for opioids in a compliant patient. Unfortunately some of these patients then become dependent or addicted but continue to “fly under the radar” until it’s too late. Sure, many primary care and pain clinics have pain medication agreements, more sophisticated urine drug tests, and other ways to “weed out” those abusing the medications. But it takes a bit of time to actively screen and seek out those with abusive behaviors that are no longer deriving benefit from opioids. Who is funding these aggressive preventive efforts?

As an aside -- and not all related to reimbursement -- I don’t think the FDA is doing us any favors with strengthening warnings about heart attacks and strokes with the use of NSAID’s. I think these warnings have rendered physicians NSAID-a-phobic to an unreasonable degree. I see patients that have been given incredibly lame excuses about why they shouldn’t use NSAID’s. 

Sorry, I’ve had a lot of pent-up angst about opioids that I needed to vent!

[paulberglmd]

3D mammography is good, but is it worth it?

I recently looked into 3D mammography -- also known as breast tomosynthesis. Our organization recently began offering 3D mammography, and one of the major medical groups in town routinely advertises messages like, "Every woman deserves the best; every woman deserves 3D mammography.” So I wondered how valid these claims were. I was pointed to this really nice piece from JAMA in 2014. With these data, I'm pretty convinced of the benefit in reducing recall rate and the rate of false positives. These are a couple of the so-called harms of mammography that are rather common. My biggest qualm with this JAMA piece -- and really anything that purports an increased rate of cancer detection -- are the very familiar issues of lead-time bias and frank overdiagnosis. You might say, "Well, in this JAMA study, the rate of detection for in situ tumors was equal, so that's not very likely. But there was an increase in detection of early invasive cancers, so that must be a good thing!" To these arguments, I would point to this fine study in JAMA IM in 2015 that shows higher-intensity screening mammography (not 3D necessarily) is associated with a higher rate of detecting small tumors but has no observable effect on 10-year mortality. Ultimately, it becomes a question of value. Do the additional costs of the technology and additional radiation dose justify the results? Or are groups that hawk 3D mammography simply offering a fancier way to detect cancers slightly earlier with no impact on how often women die from breast cancer? Earlier detection intuitively feels good to patients and doctors. Early detection seems to offer piece of mind. But if women are dying from breast cancer with or without the 3D mammogram, then what’s the real impact?

Ideally, we would have large prospective studies that document this impact on death rates (cancer-specific mortality) before universally adopting 3D mammography. Unfortunately, I think the train has already left the station. 

[paulberglmd]

Should med students just “follow” patients?

I regularly attend on the inpatient general medicine services and also frequently facilitate Physical Examination Teaching Rounds for our third-year medicine clerks. In these situations, I so often hear students use the phrase, "I'm following Mr. Jones." I personally have tried to move away from such language when I attend. Following implies a very low level of participation by third- and fourth-year students. In the context of the RIME evaluation scheme, following probably best correlates with the reporter level... or maybe an interpreter at best. The term also suggests that other patients on the service are less important to the students' collective education. If a student is only following a certain subset of patients, then what even lower level activity are they using to learn from the other patients... Passively observing? Overhearing? Ignoring? It is time to ditch following if we are truly engaging our students as active participants in patient care. Students should proudly declare a more active role! But what verb best describes what a third-year clerk or a fourth-year sub-I does? Managing is fine for the sub-I (and what we are hoping for!), but the term seems a bit too advanced for most third year students. Guiding is a potential option but probably overstates the role of the student and -- since it implies some level of supervision -- could be very deceptive to the patient. There is a certain appeal to looking after since it maintains the explicit truth that students are not yet doctors. Yet looking after still connotes a deeper level of understanding and engagement than following does.

I'm open to suggestions. Maybe the best compromise is as simple as the student saying, "Hi Mr Jones. I'm a medical student, and I am taking care of you this hospital stay." There is plenty of truth in that.

[paulberglmd]

Do you really need that specialist anymore?

"Well, I'll talk to my cardiologist about that first."

Ah, yes. There is a specialist involved in your care. I have no problem with that, but you have to realize that I'm well positioned to manage many of the same conditions that your specialist handles.

As a general internist that takes pride in knowing a fair bit about a lot of things, I really am not sure what to tell patients who want to keep seeing a specialist for a well controlled problem. I particularly struggle when that problem is right in my wheel house and something I manage on my own on a daily basis. In the words of this 2014 JGIM article, I really don't when or how to ask patients to allow me to "pull back" their care from a specialist? 

You have to understand that managing conditions like cardiovascular disease, hypertension, diabetes, and COPD are the “bread and butter” of my training as a general internist. I don't have a huge chip on my shoulder, but I can take care of these conditions just fine by myself. 

Previous coronary stent? I'm happy to refill your high-intensity statin and blood pressure medicines. And I probably am more likely to consider the benefits and risk of indefinitely continuing both your Plavix and aspirin (dual anti-platelet therapy).

History of diabetes that is controlled on just pills? Yes, my clinic and I are very well equipped to manage your blood pressures, check your hemoglobin A1c, check your feet, and make sure you get your eyes checked every 1-2 years.

When I see a specialist's note to me say, "Condition stable. No changes in medications. Follow-up 6 months," I wonder what value that person is really adding to the patient's care. In fact, I regularly manage many of these diseases without specialists' assistance every day, so does a specialist really need to be involved any more?

It's hard to fault patients in this uncomfortable love triangle. Patients often feel loyal to their specialists... or feel obligated to see them... or perceive that "more care is better"... or have any number of reasons why they want to continue with a specialist. I really think generalists and specialists alike are doing a disservice to patients when ongoing specialty care is no longer needed but allowed to continue in perpetuity. And sadly, specialists and generalists don't always agree on when transitions to a primary care setting are appropriate. 

So what's a lowly generalist to do? In scenarios where I feel knowledgeable and comfortable, I usually suggest to patients that they may not need to see Dr. Specialist. Yes, I know the specialists like stable patients with stable problems, but we have far too many people with medical problems that no one is going out of business any time soon.

I don't want patients to think I'm egotistical or have an inferiority complex. And I know I'm expected to keep patients satisfied. Yet I also know that some patients' regular visits to specialists are truly unnecessary. I just wish there was a way to communicate that information without insulting the patients or spiting my specialist colleagues.

[paulberglmd]

Distracted doctoring kills one’s diagnostic abilities... and that’s scary

In my quest to maintain a healthy balance of patient care, teaching, scholarship -- and oh yeah, that little thing called my personal life -- I have been getting in the routine of writing my clinic notes while in the room with the patient. Otherwise, I’m certain to find myself virtually buried in the contemporary surrogate for paperwork, the EHR.

Now much has been said about how much we physicians hate the EHR, how we feel like typists rather than healers, and how we have to click a mouse probably up to 100 times in a patient’s chart in a single patient visit.

But I recently have realized how much scribing my own visits is killing my ability to make diagnoses. 

These days, I’ve abandoned the time-tested tradition of how physicians used to make diagnoses. We’d take the history, carefully examine the patient, and then simply think.

I now find myself in a whir of typing and organizing my note on the fly, adding new items to the problem list, adding visit diagnoses, ordering tests as I talk, and dumping content into the patient instructions. And in this tedious yet cognitively taxing routine, I am unable to use my specially trained doctor brain to the best of its abilities. 

Now one might simply say, “Well, just slow down and don’t practice distracted doctoring.” OK, fine. But then I’ll be up all night writing patient notes, and I’ll become an irritable, unfulfilled, and (even more) sleep-deprived doctor instead.

I can’t offer any fix except to say that paying scribes should be seen as one of the most fruitful investments in medicine today. If I could give my patients more undistracted face time with me and make less diagnostic mistakes... Man, that has to be worth a scribe’s salary to someone.

[paulberglmd]

Screening for lung cancer with low-dose chest CT is nowhere near prime time

I feel compelled and inspired to comment on low-dose chest CT (LDCT) for lung cancer screening after reading this nice narrative review called Beyond the Guidelines in the most recent Annals and a recent virtual discussion I had with some third-year students. 

Despite the USPSTF grade B rating, I believe that routine LDCT for lung cancer screening is NOWHERE near prime time.

The narrative in this week's Annals of Internal Medicine covers a lot of the issues, so I won’t recapitulate the article completely. But here are my big problems with this blanket recommendation from the USPSTF and other professional societies endorsing widespread use.

Clearly, not all patients are of equivalent risk, and not all CT results are of equivalent risk. It really needs to be easy for primary care physicians like me to access and use risk prediction models and radiographic scoring systems like the Lung-RADS model to help me understand my patient’s risk.

Screening results in many harms including radiation, cost, and probably most importantly false positives. False positives can actually cause serious harm -- INCLUDING DEATH -- to patients. Yes, in the main trial from which data are derived, LDCT saved lives, but there were people that appear to have died from their diagnostic work-up too. As Dr Schwartzstein points out in the Annals article, the absolute risk reduction (ARR) for screening LDCT is pretty minuscule:

Screening reduced the chance of lung cancer death by only 0.3% and exposed the patient to possible false-positive results with subsequent unnecessary tests and interventions and anxiety during the wait for a repeated scan.

Very few primary care physicians in this country -- including me -- are adequately informed, have adequate decision support, and have adequate time and clinical support to implement this recommendation on a wide scale. Who has time to fully counsel patients on the risks and benefits? Who is going to follow-up abnormal CT scans? How will we manage all the follow-up testing that may be needed such as results from CT-guided biopsy? At what point do competing causes of death outweigh cancer risk, and thus when will screening stop?

USPSTF grades change as more evidence emerges! The medical community once had greater enthusiasm for prostate cancer screening. Of course, in 2002, mammography in 40-49 year-old women was still considered a grade B recommendation (and is covered by the Affordable Care Act as if it were still grade B). The medical community, and primary care physicians in particular, are still struggling with how to screen for these cancers -- despite having much more data on screening trials for these two other cancers.

Does screening LDCT make sense for some people? Yes, particularly for motivated patients with a very high risk of lung cancer and limited competing health problems. In these situations, I’m confident LDCT can save lives.

But I also don’t want to take a “fools rush in” approach, and I definitely don’t want to be looking back 10 years from now saying, “It looks like LDCT was the PSA of 2015, a failed promise that created more problems than it solved.”

[paulberglmd]

Playing the quality metrics: Increasing numerators for your denominators

As a practicing physician under increasing scrutiny, I’ve become acutely aware of my numerators and denominators in quality metrics. I know colleagues are facing the same pressures of hitting the mark, and I’ve started to wonder how they might try gaming the system to improve their quality scores.

For example, I recently faced the very practical question of, “Should I code ‘type 2 diabetes’ for an isolated hemoglobin A1c of 6.5?”

By the guidelines, such a patient meets a diagnostic criterion for type 2 diabetes. To label said patient “diabetic” would add a “well controlled diabetic” to my numerator and denominator, thus increasing my overall percentage of of well controlled diabetic patients. (The assumption being that my healthcare organization won’t automatically capture this patient into some “diabetes registry” unbeknownst to me.)

But that hardly seems fair to the patient, especially since this patient’s other A1c’s were well below the magic (and controversial) cut-off of 6.5. 

So is there any good reason to label these types of patient “diabetic”? Will it change patient outcomes in the long run?

The short answer to these questions is of course, “No” 

The only good reasons to aggressively diagnose diabetes with a strict cut-off are to (a.) improve one’s own quality metrics and/or (b.) augment the reported complexity of patients to outside agencies like Medicare. 

I think I’m doing the right thing by using discretion in applying diagnostic labels. 

But are other physicians around the country choosing to boost their numerators? Well, why won’t they? And if other docs are indeed gaming the system, then corporate medicine is getting what it paid for. By misappropriating the principles of evidence-based medicine, American medicine has become a culture of metric chasing that often has very little relevance to patient outcomes.

[paulberglmd]

A colleague’s retort to my “no such thing as statin-induced myalgia” claims... and more

I had to share this email retort on my notion that statin-induced myalgias may be a myth. From a colleague, Dr. Marty Muntz, in edited form:

“As Dr. Bergl (and my clinic team) knows, [statin intolerance] is a very interesting topic to me...  

“I hate to lean on anecdotal evidence, but there are some patients who have significant symptoms that completely resolve on discontinuation of statin and that recur with re-introduction.

“Could this adverse effect be due to their expectations and concerns (i.e. more mental than physical)? Sure. But my own experiences suggest that some patients do not tolerate certain statins – especially simvastatin, atorvastatin, and lovastatin. It certainly is odd that some of these patients seem to tolerate pravastatin and rosuvastatin much better.  

“A perspective article in this week’s JAMA references a 2013 study which found that of >11K patients who DC’d statins due to MSK complaints, >6K were rechallenged with 92% success rate – though not necessarily with the same dose or using same statin. This statistic makes me think that... maybe in the future we’ll be able to test for genetic predispositions to understand statin intolerance better.  And it reminds me that intolerance to one statin does not equate to intolerance to all. If patients have a strong indication for statin therapy, we really should re-challenge.”

Well said, Marty. And I couldn’t agree with you more on the following point: 

“The bigger concern I have with recent discussions in our resident clinic conferences is the idea of getting follow-up lipid panels within 3 months of starting a statin and then every 3-12 months thereafter.” 

Editor’s note: This concept was also recently summarized in a JAMA perspective.

“Just think of how many patients you have with an indication to consider statins, then think of the cost of this follow-up lipid testing... And then realize that there is really no good evidence to suggest that this practice makes any difference in cardiovascular outcomes when compared to using a risk-based assessment to start a statin, to decide on intensity at the outset of treatment, and to continue the statin without monitoring.”

Actually, I personally have argued that for some patients, you don’t even need a baseline lipid panel to make the decision about whether or not to start a statin. I’m sure the lipidologists have a good reason for recommending follow-up testing. But I currently live on the practical end of these guidelines, so like Marty, I have a really hard time seeing the value of follow-up lipids, particularly with respect to the latest treatment guidelines.

Anyways, I’ve said enough about statins for some time... I’ll have to find something else to rail on next.

[paulberglmd]

Urinary catheters for elevated PVR? Hold your hoses!

You gotta love it when your senior resident on service adopts your skepticism for "hospital traditions."

I recently had a patient on service who had several urinary catheters removed and unfortunately replaced for "elevated post-void residual (PVR)." One day, we were prompted by the nursing staff to consider replacing the catheter since our patient had around 300 mL of residual urine after voiding. The patient felt fine, and his creatinine was at his baseline.

And I said something to the effect of, 

"How high of a PVR can we tolerate? There must be plenty of guys with BPH walking around with PVR's of 300 mL that don't even know it. And this guy carries a diagnosis of BPH to boot, so this is probably a chronic problem."

For our patient, we decided to hold off on re-inserting a Foley catheter. And guess what? He continued to have stable renal function and no lower urinary symptoms. He was discharged to acute rehab without incident. 

Well, I got this brilliant email from my senior resident last week following up on this issue. 

Night float is a little slow tonight, and I just got a page about post-void residuals in a guy who we're diuresing pretty strongly. It reminded me of the conversation we had with Doc Bergl. What's the evidence for [defining] a truly high PVR and [what's] the evidence for straight catheterization especially in guys with BPH and likely chronic urinary retention?

Well, there's not much evidence, and at least the urologists are honest enough to admit it.

Kudos Justin (i.e. Justin Adams, superb PGY2 at MCW) for digging into this issue. Here's what he found based on these two nice summaries from 2007 and 2012 (in slightly edited form).

"Urinary retention” itself is a vague term that may refer to lower urinary symptoms of incomplete voiding, hesitancy, urgency, etc... or may refer to objective measures like an elevated PVR. There has been little consistency in the urology literature, particularly in defining chronic urinary retention.

PVR should be measured a few times since there is large interpersonal (person doing the measuring) and intrapersonal (day to day variability in the bladder function) variability.

The cut-offs for the upper limit for defining an “elevated PVR” are highly variable. Most sources and professional organizations in the United States use 300-400 mL as a typical cut-off. But according to Justin, “Apparently we're wimps since our neighbors across the pond may use a PVR up to 1000 mL..." to define urinary retention, again particularly in defining chronic urinary retention. 

In fact, there is very little scientific basis for defining a “cut off” for elevated PVR with the value of 300 mL being related to the point at which urinary bladders usually become palpable on physical exam.

After Justin's email and my own review of these articles, I'm still not sure what to make of the routine hospital practice of measuring PVR's with bladder scanners. This practice is especially problematic in older men who often have chronically elevated PVR.

So unless a patient is complaining of the symptoms of lower abdominal discomfort or subjectively has difficulty emptying his bladder, there is probably no role for checking the PVR.

Bladder scanners are ubiquitous in the modern hospital. But they shouldn’t be abused and shouldn’t prompt us to reflexively (re-)catheterize our patients, many of whom presumably have a chronic problem with “urinary retention” -- whatever that means.

[paulberglmd]

Do statins actually cause muscle aches? Probably not.

Yes, we've all heard so much about myalgias and myopathy occuring in patients on statins. But do these drugs actually cause muscle symptoms any more than placebo?

Last week in our Resident Clinic Conference, I attempted to blow my residents' collective minds when I told them that careful analyses of randomized trials for statins have not shown any increased rate of myalgias in patients taking statins. 

This brief narrative review in this week's JAMA could not be more timely. The JAMA piece joins the growing body of publications that are asking clinicians and patients to reconsider the statin-myalgia myth. 

Last year, a well done systematic review found no increased risk of myalgias in patients taking statins. The 2014 piece is an analysis of all primary and secondary prevention trials in which side effects were adequately reported. The authors found no difference in the frequency of myalgias in patients on statins vs. placebo. 

This apparent lack of a difference in rate of myalgias does not appear to be related to patients dropping out of either arm of the studies. In fact, patients tended to drop out of placebo arms more frequently than statin arms. 

There are still very few contrarian publications that actually support the argument that statins do not cause myalgias. If you're interested, here's another article that also documents that myalgias occur with equal frequency in patients taking placebo vs. statin in available trials. 

So what to make of this data? I don't know. Real-life practice suggests that statin-induced myalgias are common, but then again, maybe myaglias are just a convenient excuse for patients and providers to not continue a medicine. As the authors in the JAMA piece put it best, 

[Statin intolerance] is better explained by the expectation that statins can cause harm, particularly to muscle, combined with the background prevalence of muscle symptoms in the population typically prescribed statins.

[paulberglmd]

Do I need a chest x-ray to diagnose and treat pneumonia?

A couple of the residents I supervise recently posed this question to me,

Do you need a CXR to diagnose CAP in outpatient setting?  Some of the residents recalled reading that with good physical exam findings, [one] could treat empirically for CAP without CXR.  However from limited search, IDSA guidelines say yes. There is a line from UpToDate [suggesting] <50% sensitivity comparing exam/history to CXR as gold standard.

Want my opinion? 

No CXR is needed to diagnose CAP in the outpatient setting. I have treated presumptively multiple times for patients that had enough physical findings and abnormal vitals. I actually believe in the "Heckerling Score" that was first published in 1990, was reviewed in this AAFP article from 2007, and is also discussed in Steve McGee's Evidence Based Physical Diagnosis book. 

Frankly, I have seen patients that have met all 5 criteria in the "Heckerling score"... or patients that meet 3-4 criteria who also had focal findings on exam that are very suggestive of PNA such as bronchial breath sounds (which is a reasonably specific finding) or focal wet crackles.

The main issue in this scenario -- and really any clinical scenario -- is what is your threshold for treating presumptively? Is it 50% probability of pneumonia? If so, then just treat. 

If that's not a high enough probability for you, then get the radiograph.

Physicians do this mental exercise every day in medicine. If your probability of disease is high enough, there is no point to doing additional diagnostic testing. Just make the diagnosis and treat it! 

I agree that IDSA defines pneumonia as presence of infiltrate on chest radiograph. But I contend that definition is overly academic. A careful history and exam can help clinicians prioritize who does and doesn't need chest radiography to be treated for pneumonia.

[paulberglmd]

No more fasting lipids, please!

I was recently discussing "fasting lipids" in my clinic; unfortunately, this old habit just won't die.

The background and discussion in this article from 2014 says it all.

Fasting lipids are no better than non-fasting lipids in prognosticating about CV disease.

Fasting lipids are inconvenient! They may require a patient to make a return visit. Worse yet, physicians may completely miss the opportunity to get labs when the patient is actually in the office.

Only triglycericdes (TG's) really depend on the last meal, and they contribute very little to the total cholesterol level. And even TG's don't vary too much -- unless you eat something like a Big Mac right before getting your blood drawn.

Fasting lipids do not reflect the typical physiologic state of most people; most Americans have been fed within the last 8 hours most of the time.

It's also worth noting the pooled cohort ASCVD calculator asks for a total cholesterol and HDL-C levels. Both of these values are relatively stable, even in the non-fasting state. 

Rest assured physicians of the world, you can accurately predict your patients' risk levels without a fasting level.

So please, no more fasting lipid panels!

[paulberglmd]

The never-say-die practice of an "annual visit" to the doctor

As a general internist currently accepting new patients, I am still regularly seeing patients for an old practice that has been very slow to die... the annual exam. And many of these patients feel pangs of guilt -- either internal guilt or a loved one who is on their case -- for not having seen a doctor in a few years.

Sadly, the "annual physical" just one of two important myths and potentially wasteful practices that persist even in this era of cost-conscious care and increasing out-of-pocket expenses: (1.) that healthy adults need annual visits to a doctor and (2.) that "routine blood tests" for healthy adults are worthwhile.

So why not get an annual physical? Ezekiel Emanuel summarized the issues nicely earlier this year in his NY Times opinion piece, "Skip Your Annual Physical." 

I couldn't agree more. One of my professional homes, the Society of General Internal Medicine, specifically advises against the practice of general health checks. 

In fact, major policy groups have been speaking out for *DECADES* that healthy adults do not need annual physicals! To quote the Canadian Task Force on the Periodic Health Examination in their recommendations from 1979, 

"The routine annual physical examination should be discarded in favour of a selective plan of health protection packages appropriate to the various health needs at the different stages of human life."

As for routine blood tests? Like periodic health checks, they have never been shown to reduce morbidity or mortality.

Now, the Affordable Care Act does require that preventive care be free to patients, but that does not mean all "annual physicals" are free. Many contemporary insurance plans now have a tiered approach by age. I usually tell healthy adults under the age of 50 that a doctor's visit every 5 years is sufficient.

As a general internist who provides primary care, I should be in favor of good preventive care... And I am. It's just that good preventive care does not include annual physical exams or so-called routine blood or urine tests. 

So has it been a couple years since you've seen the doctor? Unless you have a health problem, don't worry... It can wait a few years.

[paulberglmd]

Diagnosing PAD - A case for physical exam

On physical exam teach rounds today, we saw a patient on my service with PAD. When I asked the third-year students how they would diagnose PAD, I was told to order ABI's (ankle brachial indexes). Well, here's a contemporary article that I *LOVE* because it emphasizes the role of physical exam in parallel to sophisticated testing. This study involved diabetic patients with a high risk of PAD. As many clinicians know, ABI's in diabetics are often unreliable given the degree of vascular calcification. This calcification can falsely elevated ABI's. In essence, the study found that normal pedal pulses have very good ability to rule out peripheral arterial disease, and physical exam outperforms the ABI for this purpose. The ability to rule in PAD is improved when ABI's are combined with physical exam findings of weak or absent pedal pulses.

Thus, taking the time to feel pedal pulses is always worthwhile. I've treated patients for presumed PAD and claudication based on that finding alone.

[paulberglmd]

A case for coronary calcium scoring

OK, I have to confess that I have a new favorite test: the coronary calcium score on CT.

I have used it several times as my new go-to test for risk-stratifying patients with low-intermediate risk of CAD, particularly in patients with CAD risk factors and the so-called "atypical, noncardiac" chest pain syndrome. 

By the current standards of care in the US, I may be bending the rules or misappropriating the test. The coronary calcium score is most frequently recommended to further stratify ASCVD risk in patients with low risk by conventional risk factors. Specifically the coronary calcium score is advocated as an additional tool to decide upon statin therapy. See here.

We don't use coronary calcium scoring for chest pain patients. Instead, we usually risk-stratify such patients with a stress test of some sort. 

However, on the other side of the pond, the NICE (UK) recommends using the coronary calcium score in precisely the manner I use it: for low-intermediate risk patients with non-escalating chest pain. Here's a great prospective study on this very issue. 

So here's why I am checking coronary calcium scores. Pull out your calculators and nomograms!

Say you estimate that a patient with chest pain has a 25-30% chance of having CAD (25-30% pretest probability) based on validated risk factors, Framingham score, and clinical gestalt. You send him for a pharmacologic nuclear stress test. Stress test is negative. The patient still has an approximately 10% chance of having symptomatic CAD. Those numbers are too high for my liking.

Send that same patient for a coronary calcium score. If the coronary calcium score is zero, the patient now only has an approximately 2% chance of having symptomatic CAD.

I'll take a single, extremely sensitive test to rule out CAD in these patients any day!

So here's how I use the results:

If the calcium score is really low, then I'm done. No further testing. Chest pain is not related to obstructive CAD; game over. A coronary calcium score of zero is extremely sensitive (>98%) in most studies.

If the calcium score is high, then the patient should plan on seeing a cardiologist for a diagnostic angiogram. I would never bother with a stress test if the coronary calcium score is off the charts.

That's easy enough. But what about intermediate calcium scores in intermediate risk patients?

 Well, here comes the other beauty of the calcium score: it is a continuous variable, not a dichotomous one. Just like the "gut check" for a mildly elevated ANA titer or a marginally increased WBC count, I can use my brain and experience.

As my experience with this score grows, I will be able to hone my "gut check." Fortunately, I can also view the calcium score in the light of other clinical risk prediction scores (ASCVD calculator or Framingham), so I don't have to rely on intuition alone.

 Anyways, it remains to be seen if coronary calcium scoring becomes a standard of care for the evaluation of low-to-intermediate chest pain in the US. It's very appealing to me.

[medresearch]

Does a doughnut a day equal more sick days? New research suggests obesity is costing the country billions of dollars in lost productivity. (Peter Dazeley/Photographer’s Choice/Getty Images) 

YouTube and Facebook get a lot of flack for distracting employees at the office, but there’s a much larger reason some workers aren’t getting the job done: their weight.

In the United States, obesity is responsible for $8.65 billion in lost workplace productivity each year, according to a new Yale and Columbia University study, published in the Journal of Occupational and Environmental Medicine. 

“Obesity and healthy-living behaviors are often seen as just individual choices,” study co-author Y. Claire Wang, co-director of the Obesity Prevention Initiative at Columbia University, told Yahoo Health. “But our paper really highlights the fact that the burden is beyond just individual choices.”

To calculate exactly how much manpower the U.S. loses each year due to obesity, the researchers analyzed data about height, weight, and missed workdays for health reasons among 14,975 people from the National Health and Nutrition Examination Survey (1998 to 2008); they also considered national data from 2012 examining body mass index (BMI) by state of more than 100,000 people.

Read more »

[paulberglmd]

Unexplained symptoms - When there is no "right answer"

Many of medical students spend a greater part of their career taking exams in which there is always a "correct answer." Usually by some point in the third year, they have realized that clinical medicine isn't so black and white. Nonetheless, assessment of students, residents, and physicians generally occur in the form of multiple choice tests, again in which a "most correct" response is desired.

However, as many who practice medicine -- particularly primary care physicians -- realize, some symptoms are not very easy to pin a specific disease. Some searches are fruitless, a result that is often unsatisfying to the provider and certainly distressing and unnerving for most patients.

Thus, I found this piece in Annals to be most timely and relevant for this (relatively) newly minted general internist. In it, Dr Kurt Kroenke makes several key points that I will liberally reinterpret (in my own words and practice philosophy) here:

Many patient-reported symptoms cannot be clearly ascribed to a "medical illness." However, the inability to attribute a symptom does not mean that it isn't real or isn't worthy of treatment.

Symptoms that suggest a certain disease or syndrome warrant an investigation for those disease. However, extensive work-ups of unexplained symptoms yield diminishing returns.

Most diagnoses that are ultimately made include a diagnosis that was suspected or anticipated in the initial evaluation of the patient.

Even in contemporary medicine, at least 75% of diagnoses are still made from the HISTORY ALONE. The physical exam (in isolation) can probably only make about 5-10% of diagnoses (there is literature to corroborate, but I don't have time to cite here). But the physical exam is an important diagnostic tool to add clarity and confidence to the diagnosis that a good physician has already made by history. Very few diagnoses are made without the help of a good H&P.

In my mind, this piece is a must-read for any generalist hoping to maintain perspective or needing a reminder to focus on both potential causes and potential mitigating factors for symptoms. I also love this piece because it affirms the role of basic skills (the H&P) in an era of fancy blood tests, routine advanced imaging, and highly specialized fields of medicine. I still wonder how I can use the advice and data herein to help me provide better patient care

And it remains to be seen how many patients will accept the advice, "One out of three times, I probably won't be able to find a clear-cut cause for your symptoms."