Monday, June 19, 2017

Antibiotic stewardship and the coming microbial apocalypse: cognitive factors driving overuse

Is this a “tragedy of the commons?” This is not a conflict between the needs of the individual patient and the good of the commons. There are potential harms to the individual patient from excessive use. From the article:

Our chief moral duty as clinicians is to our individual patients, in defense of physicians who seem to disregard the commons. However, clinicians and patients may be underestimating the individual harms and overestimating the benefits of antibiotics. Although the effects of antibiotics on the host's microbiota are often invisible, evidence that the impact is more deleterious than previously suspected is accumulating (8). Such findings may eventually change our attitude toward individual antibiotic risk to a greater degree than the threat of resistant infections alone. Using antibiotics only when needed is in the best interest of our patients as well as our communities.

According to the editorial, adoption of best practice in the area of overusage is slower than in many other areas of medicine. Why? More from the article:

Long-standing habits are hard to break. Analogous to birth cohort effects, training cohorts may exhibit stable similarities in social practice norms, which are affected by cultural attitudes toward antibiotic benefits versus harms, patient–clinician communication, or perceived expectations, and may result in different thresholds for antibiotic use. Learned practices that are shared, especially between attending physicians and trainees, resist change even when there is no evidence to support the practice. However, physicians are also influenced by their contemporary social networks—the system and social context within which they practice, including the attitudes and behaviors of their surrounding colleagues (10). These networks can be a powerful motivator for change.

Putting it together, accurate weighing of the true risks and benefits of antibiotic prescribing will help to make prudent use more justifiable on a rational level. However, physicians also need to feel that judicious prescribing is the right thing to do on an emotional or intuitive level, which often requires social cues and accountability. Interventions must also be designed with the reality of time pressure in mind, and caution must be taken with procedures that require an expenditure of time or cognitive resources. The correlation in Silverman and coworkers' study between high patient volume and antibiotic prescribing is consistent with the notion that physicians seeing patients with acute respiratory infections are practicing under extremely busy circumstances, which often require rapid decision making and intuition as opposed to deliberate, rational thought.

The last sentence points to a major barrier in the pursuit of evidence based medicine.

Sunday, June 18, 2017

Marathon running might be bad for your kidneys

In this study 82% got some degree of AKI most of whom had microscopic changes of tubular injury. It did not closely correlate with rhabdo.

Saturday, June 17, 2017

Where will artificial intelligence take us?

According to Bob Wachter it’ll be well on its way to taking over the diagnostic role of the clinician, and in as little as 5 years:

In about 5 years, Dr Wachter predicted, a physician will be able to dictate a patient note into a computer, and the computer — using artificial intelligence — will review the chart and the literature and offer a likely diagnosis or care path.

I don’t believe it. The simplest and most formulaic attempt at this, computer interpretation of ECGs, has gotten us nowhere in over 40 years.

But no doubt there will be efforts to implement this sort of thing, thus furthering the epidemic of misdiagnosis.

Friday, June 16, 2017

The coming microbial apocalypse---who’s at fault?

...we would be remiss not to mention the biggest driver of multidrug resistant organisms on a massive scale: antibiotic use in our livestock and crops. Both ID pharmacists and ID physicians know that this culprit is causing far more harm than prolonged antibiotic usage in hospitals.

Thursday, June 15, 2017

ADD medicines may help prevent MVAs

Design, Setting, and Participants For this study, a US national cohort of patients with ADHD (n = 2 319 450) was identified from commercial health insurance claims between January 1, 2005, and December 31, 2014, and followed up for emergency department visits for MVCs. The study used within-individual analyses to compare the risk of MVCs during months in which patients received ADHD medication with the risk of MVCs during months in which they did not receive ADHD medication.

Exposures Dispensed prescription of ADHD medications.

Main Outcomes and Measures Emergency department visits for MVCs.

Results Among 2 319 450 patients identified with ADHD, the mean (SD) age was 32.5 (12.8) years, and 51.7% were female. In the within-individual analyses, male patients with ADHD had a 38% (odds ratio, 0.62; 95% CI, 0.56-0.67) lower risk of MVCs in months when receiving ADHD medication compared with months when not receiving medication, and female patients had a 42% (odds ratio, 0.58; 95% CI, 0.53-0.62) lower risk of MVCs in months when receiving ADHD medication. Similar reductions were found across all age groups, across multiple sensitivity analyses, and when considering the long-term association between ADHD medication use and MVCs. Estimates of the population-attributable fraction suggested that up to 22.1% of the MVCs in patients with ADHD could have been avoided if they had received medication during the entire follow-up.

Conclusions and Relevance Among patients with ADHD, rates of MVCs were lower during periods when they received ADHD medication. Considering the high prevalence of ADHD and its association with MVCs, these findings warrant attention to this prevalent and preventable cause of mortality and morbidity.

Wednesday, June 14, 2017

Thiamine deficiency and heart failure: evidence for an association is mounting

From the green journal:

Diuretic therapy is a cornerstone in the management of heart failure. Most studies assessing body thiamine status have reported variable degrees of thiamine deficiency in patients with heart failure, particularly those treated chronically with high doses of furosemide. Thiamine deficiency in patients with heart failure seems predominantly to be due to increased urine volume and urinary flow rate. There is also evidence that furosemide may directly inhibit thiamine uptake at the cellular level. Limited data suggest that thiamine supplementation is capable of increasing left ventricular ejection fraction and improving functional capacity in patients with heart failure and a reduced left ventricular ejection fraction who were treated with diuretics (predominantly furosemide). Therefore, it may be reasonable to provide such patients with thiamine supplementation during heart failure exacerbations.

The vitamin C cocktail for severe sepsis and septic shock

I know I’m a little late with this. Here’s the paper published in Chest. Form the paper:


In this retrospective before-after clinical study, we compared the outcome and clinical course of consecutive septic patients treated with intravenous vitamin C, hydrocortisone and thiamine during a 7-month period (treatment group) compared to a control group treated in our ICU during the preceding 7 months. The primary outcome was hospital survival. A propensity score was generated to adjust the primary outcome.


There were 47 patients in both treatment and control groups with no significant differences in baseline characteristics between the two groups. The hospital mortality was 8.5% (4 of 47) in the treatment group compared to 40.4% (19 of 47) in the control group (p less than 0.001). The propensity adjusted odds of mortality in the patients treated with the vitamin C protocol was 0.13 (95% CI 0.04-0.48, p=002). The SOFA score decreased in all patients in the treatment group with none developing progressive organ failure. Vasopressors were weaned off all patients in the treatment group, a mean of 18.3 ± 9.8 hours after starting treatment with vitamin C protocol. The mean duration of vasopressor use was 54.9 ± 28.4 hours in the control group (p less than 0.001).


Our results suggest that the early use of intravenous vitamin C, together with corticosteroids and thiamine may prove to be effective in preventing progressive organ dysfunction including acute kidney injury and reducing the mortality of patients with severe sepsis and septic shock. Additional studies are required to confirm these preliminary findings.

A post at the Skeptics’ Guide to EM has a nice discussion and critical appraisal, and several notables from the FOAM community weighed in. The participants were unanimous in saying that this study is only hypothesis generating and should not change practice at the moment.

The problems with this study are obvious. Issues that concerned me in particular were:

1) There were three interventions. If the effect is true, which one(s) worked?
2) It seems too good to be true.
3) What are we to make of the 40.4% mortality in the control group?

Some would ask why not just give it to septic patients, since it is harmless, right? Others would counter that you could say the same thing about homeopathy. But wait, homeopathy has no plausible mechanism of action. Vitamin C does. Several, in fact.

It’s interesting that the folks at East Virginia don’t feel there is equipoise for a randomized controlled trial. As experience accumulates I expect to see more and more low level evidence published, from that institution and elsewhere. If that experience repeatedly and consistently points toward a therapeutic effect, especially a very large one as suggested in this study, then we may never feel there’s equipoise and it will gradually become accepted into sepsis care. More likely we’ll see results that are not so good, leading eventually to a randomized trial. My bottom line today is that, while it would be difficult to fault someone for incorporating this into sepsis care, the Marik study should be considered hypothesis generating only, in need of further study, and not a mandate for practice change.

ACP puts out its own guideline for hypertension in the elderly

In short, the target is systolic below 150, and consider below 140 if high cardiovascular risk is present.

As to the choice of pharmacologic agents they don’t write anything in stone:

Effective pharmacologic options include antihypertensive medications, such as thiazide-type diuretics (adverse effects include electrolyte disturbances, gastrointestinal discomfort, rashes and other allergic reactions, sexual dysfunction in men, photosensitivity reactions, and orthostatic hypotension), ACEIs (adverse effects include cough and hyperkalemia), ARBs (adverse effects include dizziness, cough, and hyperkalemia), calcium-channel blockers (adverse effects include dizziness, headache, edema, and constipation), and β-blockers (adverse effects include fatigue and sexual dysfunction).

Ace those board exams

Pain is not a vital sign