Cardiology, Critical Care, Improving Outcomes, Improving Throughput, Radiology

POCUS, Aneurysms, and Mortality Rates.

If you’re a #FOAM follower, you have probably seen the pleas for bedside ultrasound more than once. This paper takes an interesting approach to try and demonstrate its value in the diagnosis of aortic dissection: Over a two year period and 386,547 patient visits, there was a review of 123 medical reports and 194 autopsy reports, of which 32 patients were identified for inclusion. 16 received EP POCUS, 16 did not.

Median time to diagnosis – 80 minutes in the POCUS group vs 226 minutes in the non POCUS group. Misdiagnosis was 0% in the POCUS.

Mortality adjusted for DNR status: 15.4% vs 37.5%, POCUS vs non-POCUS.

Time to dispo? 134 minutes vs 205 minutes, POCUS vs non-POCUS. (and probably a much greater difference in time to *appropriate* disposition.)

[note that neither mortality or time to dispo was statistically significant] 

With that said, I agree with the authors conclusions, (particularly in light of this previous post): “Patients who receive EP FOCUS are diagnosed faster and misdiagnosed less compared with patients who do not receive EP FOCUS. We recommend assessment of the thoracic aorta be performed routinely during cardiac ultrasound in the emergency department.”

Standard
Cardiology, Mythbusting

Holiday Heart is Real

I wanted to put this one out early this year, even though the paper was published in 2014.

An analysis of 12,195 cases of first MI (spanning 52 countries!) was compared to 15,583 age and sex-matched controls. While they found that frequent alcohol use in moderation was associated with a reduced risk compared to non-drinkers (OR 0.87), heavy episodic drinking (six or more drinks) within a 24 hour period was associated with an increased risk of MI (OR 1.4).

This risk was significantly elevated in those over age 65 with an OR of 5.3 !

Take home – Holiday Heart is real, don’t blow it off as GERD if they show up in your ED, and don’t let your elders drink then go shoveling this holiday season!

Enjoy the holidays safely!

Standard
Cardiology, Improving Outcomes, Improving Throughput

The Lean ED in the Obs Unit

The “lean ED” has been embraced in multiple ways – keep your vertical patients vertical, a disposition area where patients await paperwork & final signatures, bedside sono, the list goes on.

It would stand to reason that lean principles would trickle down the halls and up the elevator to other departments as well.  Along these lines, I’ve started to see ultrasound trickle into the hands of internal medicine residents with good results:

This study (http://www.ncbi.nlm.nih.gov/pubmed/25492052), after just 9 hours of training, focused cardiac ultrasound allowed for “substantial” agreement between trainee and tutor for identification of pericardial effusion (k=0.71) and global cardiac systolic function (k= 0.77), “moderate” for marked right and left ventricular enlargement (k= 0.56-0.64).  After an additional 9 hours of training, residents were only proficient in obtaining adequate images in the parasternal window without mastering the apical and subcostal views.

While here (http://www.ncbi.nlm.nih.gov/pubmed/26179460), ten internal medicine attending physicians performed focused cardiac ultrasound on178 inpatients and compared to formal echo interpreted by cardiologists.  Sensitivity and specificity for any degree of left ventricular systolic function was 91% and 88% respectively.  The interrater agreement between the internist and cardiologist was “good/substantial” with K = 0.77.

These principles can and should be applied to observation units, where many chest pain rule outs and CHF exacerbations will be placed – and along with them, orders for an echo (I’ll leave the discussion of whether or not they all need an echo for another day).  I’ve waited until the next day for an echo to be done and read in an observation unit so I can dispo a patient.  If you are working in observation regularly, and chest pain and CHF is your bread and butter, I see no reason why echo competency can not be obtained rather quickly.  By learning bedside cardiac sono as an observation unit provider, you’ll make yourself an invaluable team member and likely decrease length of stay without harming patients.

Standard
Cardiology, Cardiology, Improving Throughput, Mythbusting

Surprise! You still do not need to treat asymptomatic hypertension in the ED!

Surprise! We have yet another paper, in addition to the ones mentioned here that demonstrates acutely managing hypertension without end organ damage is unnecessary.

 

This multicenter study retrospectively looks at 1,016 patients, age 18 years or older, with an initial blood pressure greater than or equal to 180/100, with no evidence of target organ damage, that were discharged with a primary diagnosis of hypertension.  435 patients (42.8%) received antihypertensive therapy (88.5% received clonidine!), while the rest did not receive antihypertensive treatment in the ED.   Those who received antihypertensives often had a higher mean initial systolic and diastolic BP (systolic 202 vs 185, and diastolic 115 vs 106). Otherwise, all other measures were similar & not statistically significant- this includes repeat ED visits within 24 hours (4.4% vs 2.4% antihypertensives vs none), repeat visit within 30 days (18.9% vs 15.2%), mortality at 30 days (0.2% for both groups), as well as mortality at one year (2.1% vs 1.6%).

Unfortunately, this is a retrospective study, so perhaps they are not catching the patients with a chief complaint of hypertension, and perhaps ultimately diagnosed as a bleed.  It is reassuring to see that 30 day and mortality and 1 year mortality are similar.  However, with 94.5% of the total patient population being African American, these results may not necessarily be broadly applicable to all ED patients.  It would have been nice to see them analyze all patients over a 12 month period who presented with a triage BP > 180/100, and blind reviewers to disposition and outcomes.  With only ~500 patients in each arm, I am not certain this adequately screens for rare events, but I still think that this is one of the better attempts to date to back up ACEP policy on asymptomatic hypertension

(1) Initiating treatment for asymptomatic hypertension in the ED is not necessary when patients have follow-up; (2) Rapidly lowering blood pressure in asymptomatic patients in the ED is unnecessary and may be harmful in some patients; (3) When ED treatment for asymptomatic hypertension is initiated, blood pressure management should attempt to gradually lower blood pressure and should not be expected to be normalized during the initial ED visit.

Standard
Cardiology, Cardiology, Improving Outcomes, Improving Throughput, Mythbusting

Chest Pain Statistics That’ll Make Obs Bump Trops

I have been lucky enough to attend ACEP over the last few years, and even luckier to watch David Newman speak. Specifically, I have attended his “Is One Troponin Enough?” lecture, which was practice changing for me.  It also encouraged me to drench both my patient discussions and MDM’s in evidence prior to discharging patients. Among others, Newman is at it again in today’s article, found here.

At 3 institutions over 5 years, from July 1, 2008 to June 30, 2013, encompassing over a million ED visits in total at these three institutions combined, the authors sought to determine the incidence of clinically relevant adverse cardiac events in patients hospitalized for chest pain with two negative troponins, normal vital signs at time of arrival to the ED, and nonischemic EKGs throughout their stay.  Clinically relevant events were defined as life-threatening arrythmia, inpatient STEMI, cardiac or respiratory arrest, or death during hospitalization.

Essentially, during the 5 year study period, over one million ED visits, and over 11,000 patients admitted with two negative troponins, only 20 patients had an adverse outcome (0.18%). When you exclude abnormal vitals at presentation, ischemic EKG findings, left bundle branch block or a paced rhythm, only 4 events were seen out of 7266 patients (0.06%), with two being non-cardiac, and two (possibly) being iatrogenic.

Wow.

Let’s process this for a minute.  When taking all-comers – not just the low risk observation patient- with two negative troponins (drawn between 60-240 minutes in this study) your risk is quite low at 0.18%, now exclude non-ischemic EKGs, patients with abnormal vitals, a paced rhythm, or a left bundle branch block, and your risk is 0.06% of an adverse outcome – and more likely to be iatrogenic than cardiac!

Now let’s couple the above study with this study, where they examined almost 700,000 private-insurance ED patients in 2011 presenting with chest pain. They followed patients that both did and did not receive additional diagnostic testing (exercise stress test, stress ECHO, myocardial perfusion scintigraphy, or coronary CTA). Essentially, the rate of MI at 7 days and 190 days was low overall (0.11% and 0.33% respectively). Most importantly – patients who did not undergo initial non-invasive testing were no more likely to experience a myocardial infarction than those who did not receive additional testing. Compared to no testing, additional testing was associated with significantly higher odds of cardiac catheterization and revascularization procedures without a concomitant improvement in the odds of experiencing an MI.

So why are we consulting cardiology for observation patients with two negative troponins? Why are we ordering stress testing for inpatient or observation patients? This is yet another example of why it is important to provide well-informed consent for your patient, and a great example of a well-intentioned hospitalization and consultation providing (potentially) more harm than good.

It is high time we cease and desist the scare tactics we employ to patients to strong arm them into either an AMA or admission, rather than providing them a look at the current data before making their decision.

Standard
Cardiology, Cardiology, Improving Outcomes, Improving Throughput, Mythbusting, Pulmonary, Pulmonary, Twelve Trials of Christmas!

Day Five of Christmas- where’s the Obs EBM?

Welcome to the Twelve Trials of Christmas series on EMinFocus!  This is the fifth of twelve posts in a series where I ramble on various topics for which I would love to see an EM study done.  I’ve taken morsels of prior studies (case series, small trials, etc) and highlight reasons on why I believe this study would benefit EM.  Some may pan out, some may not.  All of them I would be highly interested in assisting with in any way possible to continue to advance our fine specialty.

I believe that the longer a patient stays in the hospital, the more likely they are to develop new and exciting pathology (pulmonary emboli, VRE, general deconditioning, an unhealthy affinity for tuna sandwiches, etc).  t’s not that I want to throw people out so to so to speak, it’s that their risk of developing hospital acquired badness is higher than their risk of adverse outcome from their diagnosis (like, say, low / no risk chest pain).  As such, I’ve recently taken an affinity towards our observation unit to see which, if any, of the classically admitted COPD’ers / CHF’ers, etc, can be placed in observation for <24 hours (actually need to be there), and do well.

We need to see some studies for who is a candidate for observation – and perhaps some data on why ED providers put no risk chest pain and other similar low yield findings in observation. There is little to no data on observation units, despite over a third of hospitals in the US having them.  Despite guidelines for observation, there are few evidence based guidelines.  We know that increasing age is associated with an increased rate of admission from observation status (about 26% vs 18 %); and there are also cellulitis guidelines for observation units, but that is about it.  While there are Ottawa guidelines for admission for CHF and COPD, and the famous PORT scores, these rules were designed to identify low risk patients, not necessarily those that needed a day or two in the hospital.

 So, perhaps, much like we have recently begun to risk stratify PE’s into going home and being lysed, so we will have to do the same for utilization of observation & inpatient resources.  What delineates which patients in the grey area between obvious discharge home and ICU admission can go to observation – not just the ED provider saying “yeah, they look ok for observation” – some actual evidence that suggests the patient can be turned around in under 24-36 hours.  I suspect for CHF, we’ll see that most of these patients, once properly cared for in the ED (cough, nitro nitro nitro, cough), can go to observation, which will likely surprise most ED & inpatient providers – thus, hopefully decreasing the risk of hospital acquired badness for these patients.

Standard
Cardiology, Cardiology, Improving Outcomes, Improving Throughput, Mythbusting

“So Doc, what are we doing about this patients blood pressure?”

 

The next patient in the rack was sent by WebMD after their new iHealth Wireless Wrist Blood Pressure Monitor told them to come to the ED because their blood pressure was at “stroke level.”

While indeed, this patient has a blood pressure on initial evaluation of 205/115, but they have not had any chest pain.  They are clearly not altered on eval, and they deny any vision changes or headaches and have a non-focal neurological exam.  When asked why they even checked their blood pressure, it was because all their friends who have doctors were told to check their blood pressure at home.  

So what does this patient need?

If the patient offers some vague complaints that were persistent, I guess you could argue for electrolytes, maybe an EKG / troponin (if dyspneic, diaphoretic, or nausea after exertion then perhaps go the ACS workup route).  If the patient is completely asymptomatic or with fleeting vague symptoms, the first step – if any (after an H&P) – should be a urinalysis.  In June 2010, AJEM published an important study on this.  They screened 5,416 patients just like the ones above.  Essentially, if the patient has neither hematuria nor proteinuria, there is a 98.3% chance that they will not have a serum creatinine above 2.0.  For Creatinine above 1.3 and 1.5, the lack of both proteinuria and hematuria has a negative predictive value of 92.4% and 95.3% respectively. 

What does ACEP say? ACEP recommends, with a level C recommendation, that screening for acute end-organ injury is not required. As long as the patient is not altered, has a normal neurological exam, no vision changes, and no crushing chest pain, you essentially already screened them without doing a single test! A UA will do a great job at ruling out AKI. If a UA is positive for either hematuria or proteinuria, then send a serum creatinine – if at that point you have an elevated creatinine >25% from baseline, then consider admission. With a benign history and exam, the patient does not need electrolytes, troponins, a BNP, an EKG, Chest XRay, and certainly not a brain CT. Here is the data ACEP provides on this:

A Class III study, the VA Cooperative Trial of 1967, was a randomized placebo-controlled trial of 143 male patients with diastolic blood pressure of 115 mm Hg to 130 mm Hg. No adverse outcomes in either group were demonstrated during the initial 3 months of enrollment. Four of 70 patients in the placebo group (6%; 95% CI 2% to 14%) versus 0 of 73 patients in the treatment group (0%; 95% CI 0% to 5%) developed significant complications within 4 months of enrollment, including sudden death, ruptured aortic aneurysm and death, severely elevated blood urea nitrogen level, and congestive heart failure. However, within 20 months, 27 of 70 patients (39%; 95% CI 27% to 51%) treated with placebo and 2 of 73 patients (3%; 95% CI 0.3% to 9.5%) treated with antihypertensive drugs experienced adverse events (absolute risk reduction 36%; number needed to treat = 3.″

So at 3 months, no worries. At 4 months, trouble may start. At 20 months, big trouble. So what are we supposed to do about it in the ED? For the patient who has no symptoms, a normal UA, and isolated hypertension treat based on the locale you’re in. If follow up is easy, leave it alone. If getting into clinic is impossible, consider starting an oral medication in the ED.

Now, what does the legal side of this look like?  Well, I attempted to peruse Google Scholar, using the terms “Emergency Department” “malpractice” and “hypertension.”  I’m far from good at browsing the medico-legal literature, but of the 97 results, I can not find a single case of an asymptomatic hypertensive that was sent home and the provider was later sued.  Of the vaguely pertinent cases where the patient presented with significantly elevated blood pressure (email or tweet me for links), all the patients had some other symptoms that were alarming- “cold symptoms” – ie, misdiagnosed CHF;  chest pain and weakness that was given Ramipril, Bisoprolol / HCTZ that later stroked; syncopal elderly patient that fell twice in the ED and was usually ambulatory; “slow, unsteady, hot & sweat with double vision,” ; severe headache, off balance, blurred vision, dizziness.  Bottom line, you’ve got time to work with outside the ED walls, so there is no need to “buff the chart” and give IV hydralazine prior to discharge.  As ACEP policy notes, you are more likely to do harm than good.

.

.

.

PMID: 20579559

PMID: 24352797

http://emcrit.org/wp-content/uploads/2013/10/Asympt-hypert2-Final-BOD-approved-2013.pdf

Standard