Category Archives: Literature Reviews

Must-Read Resuscitation

Posted on by
Reply

Folks,

I know I am incessantly encouraging people to follow Dr. Smith’s EKG blog, but the most recent post is too good to not tell you all about. In it, there is a write up of a fantastic resuscitation.

Highlighted in it are double-defibrillation, Head up CPR, hypokalemia bolus dosing, ultrasound for stemi and VFib (!), esmolol bolus for refractory vfib,  cath lab availability and MDM, hypothermia discussion, and prognostics in cardiac arrests.  I mean, this case report has it all, and it is referenced and discussed for our enjoyment.  Read it.

The resuscitation you wish you had run IS HERE!

Read it. I’ll buy you a drink if you do.

Zach

Head injury patients with delayed presentation to ED

Posted on by
1

Another pearl from EM Lit of Note. Bottom line: Retro review of CT head obtained for trauma divided into <24 hours and >24 hours time of presentation. The delayed presenters had a HIGHER percentage of positive CT and had a similar amount of patients requiring NES intervention.

We discussed this the other day. First Care obtained a head CT on a patient several days after a minor head injury. We presumed it was not indicated. Then I read this paper.

Unfortunately I cannot find in the paper a description of time to presentation. It is grouped into less than 24 hours and greater than. I wonder if the likelihood of positive CT scan decreases as time from injury to presentation increases.

In any event, this poses difficult questions. Should we obtain more CTs on the delayed presenters? They are as likely to have positive findings. In addition, the NICE guideline is 70% sensitive, versus its comforting 98% in the less than 24 hour group.

Would be a good article to discuss for journal club. Would love to see some comments.

Zofran

Posted on by
Reply

Well here is the study that has had everyone talking for a while. Good old ondansetron category B now may cause congential malformations in the fetus. This study in the journal Reproductive Toxicology retrosp of 1349 infants concludes a risk for CV defect (septum specifically) is increased to a statistically significant degree. My favorite prior study looked at 608,000 pregnancies and concluded no malformation risk. This is in the NEJM. So I am much more likely to follow the recs of the huge NEJM study.

Now we could do more prospective studies of course to see if ondansetron is causing any true effects. And we can always use the promethazine, prochlorperazine class of drugs in these patients while in the ED or even for discharge. I still love good old extremely evidence-based ginger pills, also B6, diphenhydramine, not eating crap food, etc.

Discussion is welcome.

72 Hour Returns

Posted on by
Reply

Much has been written about ED recidivism. Ryan Radecki at EM Lit of Note has just reviewed another article about returns, this one looking at bounce backs from inpatient services as well. His conclusion is that 72 hours returns should not be used as a quality measure for ED metrics. His complaint is the nonspecific analysis of these returns. When we discuss 72 returns in our didactics, we attempt to focus on the cases from which we can learn how to prevent discharge of inappropriate patients. We are discussing shifting more toward discussing less cases with well defined take home points as we enhance didactics. Take a look at the Radecki commentary and if interested, the growing body of literature. If anyone is especially interested, a solid admin project / scholarly project would be to analyze our 72 returns over a longer period than a month. Drs Coleman and Platt presented a poster at CORD when we presented the Room9ER poster years ago.

Finally

Posted on by
Reply

I am quite excited about this study but sad I was not the one to publish it. Out of South Florida, a study of 72 patients who received a NONCONTRAST CT Abdomen for various locations of abdominal pain.

Excluded were: trauma, pregnant, surgery in past 30 days, pain more than 3 days, MD discretion (thought usually to be suspicion for vascular emergency), unable to be reached for 7 day follow up, would not consent, unstable, suspected renal stone (as this is already the standard of care), known cancer, BMI < 18. Sounds like a lot of exclusions but this pared the population down to the general population of patients in whom I would consider the noncontrast CT abd for nonspecific abdominal pain. Notice suspicion for appendicitis was not an exclusion, many studies have shown minimal benefit to oral contrast in appendicitis. They also did not exclude very acute abdominal pain (ie using a minimum number of hours of pain).

The results: 0/72 patients had a missed diagnosis on the initial noncontrast CT. 3/72 had repeat contrasted studies which added little. Of the 31 patients discharged, no one had surgery, death, or repeat CTs done.

This is obviously a small sample at one hospital. Selection bias can be a factor, as MD discretion was an exclusion. I would love to see more studies like this, possibly RCTs. As more and more centers get the higher resolution CTs, we should see less and less contrast given for CT of the abdomen and pelvis.

Approach to PE

Posted on by
1

Hey all,
I got the privilege of going to ACEP last week in Boston. When I got the schedule one of the lectures that stood out to be was a PE lecture by Jeffrey Kline. Some of you may recognize the name but if not, he is an attending at IU with a special interest in thromboembolism. He is very active on twitter at @klinelab and wrote the Thromboembolism chapter in Tintinalli’s. After talking about PE last month and specifically approach to PE in the pregnant patient, I thought a summary of the key points would make for a worthwhile post.

The first question in the discussion of VTE should be ‘who actually needs to be tested?’ If someone comes in complaining of recent chest pain or dyspnea, PE needs to be included in the differential. If they are not complaining of those recently and have normal vitals (at all times) then you don’t need to go chasing down a clot that isn’t there. If the patient does complain of those then some sort of documentation is required to show you considered a PE. Even stating ‘I think PE is unlikely because of X, Y and Z’ would likely be enough. Now if your pretest probability is anything other than very low, some combination of wells, perc, Geneva should be applied. I like the following algorithm which I think Kline discussed on EMRap towards the end of last year.
algo

Following that algorithm helps cut down on the number of ct scans you’ll order, cuts down your false-positives, radiation exposure, and contrast induced nephropathy without increasing the number of significant PE’s that you miss.

As far as the pregnant patient, I think everyone knows to start with the lower extremity ultrasounds in hopes of an answer that would let you initiate treatment. However, when that is inevitably negative, there is also an algorithm for that scenario that incorporates a trimester adjusted d dimer.

algopreg

The other main takeaway from this talk was the disposition change on some of the low risk patients. Dr. Kline said he has sent about 70 patients home from the ED after being diagnosed with PE. To stratify who falls into low risk, you can apply the sPESI or HESTIA score as well as who is at low risk of bleeding.

–Simplified PESI-if any +, pt is NOT low risk:
age greater than 80
history of cancer
history of chronic cardiopulmonary disease
pulse greater than 100
BP less than 100
O2 sat less than 90

–Hestia-pt CAN BE considered low risk if
BP greater than 100
No thrombolysis needed
No active bleeding
02 sats greater than 90
Not already anticoagulated
No other medical or social reason for admission
Cr clearance greater than 30
not pregnant, no severe liver disease

For these people they’ll initiate rivaroxaban or apixaban in the ED and send them home with a prescription. The only failures they’ve experienced are people who returned requiring additional pain management. Has anyone done this or considered it? The majority of our patient population would not satisfy these requirements or, frankly, be reliable enough to consider outpatient management, but what about people working in the community with a different population?

Lastly, we all know to look for S1Q3T3 on the ekg to raise suspicion for PE but the odds ratio is only 2.06. Inverted T’s in V2 and V3 have odds ratios of 6.94 and 7.07 respectively, and are the most SPECIFIC ekg finding in pulmonary embolism

Antibiotic review

Posted on by
Reply

Maybe it’s because an overhead projector slide scanned and inserted into a powerpoint presentation comprised my medical school antibiotics curriculum (you know what I’m talking about if you went to U of L), but I’ve never really felt comfortable with the nuances of antibiotics. For those who want to understand them a little better, here’s a great review.

The bar is set, Chrissy!

NEXUS in the Elderly

Posted on by
Reply

Hopefully everyone is using the NEXUS criteria or the Canadian C Spine rule in evaluation of patients who have undergone neck trauma. Those familiar with both know one major difference, age criteria. NEXUS does not use age, Canadian C spine does. Using both rules together, like PERC with Well’s, increases sensitivity at expense of specificity.

Well here is a study on falls in the elderly (i.e. low mechanism which is another difference between NEXUS and Canadian) and application of NEXUS. Turns out, probably shouldn’t be using NEXUS in patients over 65. Liberally scan these folks, radiation is less of a concern, and the cost is justified due to morbidity of missed injuries. And of course do not bother with plain films (in adults).

Hot off the Press, Droperidol is Still Safe

Posted on by
Reply

Yet another quality article illustrating the safety of the most magical drug, droperidol. Not sure where people are getting it, as it is not being manufactured currently. We have none at any of my hospitals in Louisville. It is sad but perhaps someday we will get it back. In the same Annals issue an indictment of professional societies, journal editorial boards, and government advisory committees with their misinterpretation of “so-called facts.” Great reading, Dr Newman is the man.

Chest Pain Admission Dilemma

Posted on by
Reply

Hey guys here is an interesting article with actual patient oriented outcomes related to admission for chest pain.

Several take home points:

1. From highest quintile of admission rate to lowest (81% to 38%) the rate of MI and death went up by 3.6 per 1000 and 2.8 per thousand. This correlation implies that when you admit more patients you save lives.

2. It is VERY IMPORTANT to note the patient population. These are Medicare patients with average age of 71 years. So we ARE NOT talking about low risk chest pain ED patients.

3. Even though it looks impressive to save these lives, the NNT or number needed to admit to prevent one MI is 250 and to prevent one death is 333. Thats a lot of admissions. And admitting geriatric patients is often not a good thing for them. May be why the decrease in deaths is less than the decrease in MIs. They were dying because of a hospital acquired infection or deconditioning or something else.

4. It is striking to see how different the practice patterns are at different hospitals regarding admission of a fairly homogenous cohort of patients.

Appreciate any further comments.

Just get a walking O2 sat

Posted on by
1

In patients with some suspicion for PE, even with a negative d dimer, I have often ordered a walking O2 sat and HR. This was not really evidence based, but maybe now could be. Below is the abstract for a prospective cohort study of patients known to be with and without PE. Interesting data even if only 114 patients. Cannot get full text yet.

Take home point. Combined sensitivity of HR increase of 10 BPM AND Sat decreased of >/= 2% was 100%.

ie if HR does not go up by 10 or more AND sat does not drop by 2 or more they are very unlikely, based on this small study, to have a PE.

 

 

CJEM. 2015 May;17(3):270-8. doi: 10.1017/cem.2014.45.

Ambulatory vital signs in the workup of pulmonary embolism using a standardized 3-minute walk test.

Amin Q1, Perry JJ1, Stiell IG1, Mohapatra S1, Alsadoon A1, Rodger M2.

Author information

Abstract

OBJECTIVE:

Diagnosing pulmonary embolism can be difficult given its highly variable clinical presentation. Our objective was to determine whether a decrease in oxygen saturation or an increase in heart rate while ambulating could be used as an objective tool in the diagnosis of pulmonaryembolism.

METHODS:

This was a two-site tertiary-care-centre prospective cohort study that enrolled adult emergency department or thrombosis clinic patients with suspected or newly confirmed pulmonary embolism. Patients were asked to participate in a standardized 3-minute walk test, which assessedambulatory heart rate and ambulatory oxygen saturation. The primary outcome was pulmonary embolism.

RESULTS:

We enrolled 114 patients, including 30 with pulmonary embolism (26.3%). A ≥2% absolute decrease in ambulatory oxygen saturation and an ambulatory change in heart rate >10 beats per minute (BPM) were significantly associated with pulmonary embolism. An ambulatory heart rate change of >10 BPM had a sensitivity of 96.6% (95% confidence interval [CI] 83.3 to 99.4) and a specificity of 31.0% (95% CI 22.1 to 45.0) forpulmonary embolism. A ≥2% absolute decrease ambulatory oxygen saturation had a sensitivity of 80.2% (95% CI 62.7 to 90.5) and a specificity of 39.3% (95% CI 29.5 to 50.0) for pulmonary embolism. The combination of both variables yielded a sensitivity of 100.0% (95% CI 87.0 to 100.0) and a specificity of 11.0% (95% CI 6.6 to 21.0).

CONCLUSION:

In summary, our study found that an ambulatory heart rate change of >10 BPM or a ≥2% absolute decrease in ambulatory oxygen saturation from baseline during a standardized 3-minute walk test are highly correlated with pulmonary embolism. Although the findings appear promising, neither of these variables can currently be recommended as a screening tool for pulmonary embolism until larger prospective studies examine their performance either alone or with pre-existing rules.