Wellness

Journal Club Synopsis-Wellness.   November 29, 2012


Kudos to the resident presenters:  Mark Hemming, Maggie Putman, Sola Balogun, Steve Walchuk, Brad Kutka and Bret Negro.  Also, many thanks to Dr. John Principe and his staff at the WellBeingMD Center. 

 

 

Article 1:  Mitrou PN, et al: Mediterranean dietary pattern and prediction of all-cause mortality in a US population: results from the NIH-AARP Diet and Health Study.  Arch Intern Med. 2007 Dec 10;167(22):2461-8.


This prospective observational study was a 1995 survey of 380,296 people with no history of chronic disease as part of the National Institutes of Health-AARP Diet and Health Study.  The questionnaire asked about conformity with the Mediterranean dietary pattern (emphasis on vegetables, fruits, nuts, legumes, whole grains, fish, olive oil).  Conformity with the Mediterranean diet was associated with significantly reduced all-cause and cause-specific (cardiac and cancer) mortality at 5 and 10 years.   In men, hazard ratio comparing high to low diet conformity for all-cause mortality was 0.79 (95% CI, 0.76-0.83).   For women, high conformity with the diet was associated with a 20% decreased risk for all-cause mortality.  Multivariate models attempted to adjust for potential confounding mortality risk factors.  Results were maintained in participants who had never smoked, and for both genders.

 

Both Mark and Maggie mentioned issues of recall bias (how accurately the diet survey was filled out), and Dan stressed that although the trial tried to correct for confounders, there are still likely additional positive choices that people eating a healthy diet make that also contribute to their overall state of health.  Chintan pointed out that participants were only surveyed once, and diet popularity often changes over time; no way of knowing if they complied with the Mediterranean diet for 10 years.  Bottom line for the presenters-the results are consistent with other studies of the Mediterranean diet.  The diet is easy to remember and worth sharing with patients, family.


 

Article 2:  Kodama S, et al: Cardiorespiratory Fitness as a Quantitative Predictor of All-Cause Mortality
and Cardiovascular Events in Healthy Men and Women; A Meta-analysis.  JAMA. 2009;301(19):2024-2035.

This laborious meta-analysis attempts to quantify the relationship between cardiorespiratory fitness (CRF) and all-cause mortality as well as coronary heart disease (CHD)/cardiac events in healthy participants.  Data were obtained from 33 eligible studies of > 185,000 participants.  Compared with participants with high CRF, those with low CRF had a relative risk for all-cause mortality of 1.70 (95% CI , 1.51-1.92; p<0.001) and for CHD/cardiac events of 1.40 (95% CI, 1.32-1.48; p<0.001).   A 1-MET (1 km/hour higher jogging speed) increased level of maximal aerobic capacity was associated with 13% and 15% decreases in risk for all-cause mortality and CHD/cardiac events respectively.  This study suggests that cardiorespiratory fitness could be integrated into the risk factor profile for CHD, and establishes the minimal walking speeds or Bruce protocol performance for men and women at various ages that may help prevent CHD. 

 

Harwood liked the outcome measure of all-cause mortality:  you don’t want to decrease cardiac mortality but increase the mortality from falling off treadmills.

 

Stats pearl:  There was a lively discussion of the applications of various study methodologies.  As Barounis stated, a RCT evaluates one variable and one outcome and in general provides the highest level of evidence.  Observational cohort studies don’t examine cause and effect, and can’t account for all confounders.  So, an excellent RCT will trump a meta-analysis and certainly trump an observational trial.   At the same time, for this evening’s clinical questions (diet, exercise), prospective observational cohort trials may be the most appropriate and ethical trial design.  Randomizing anybody to the “high saturated fat sessile lifestyle” arm isn’t going to make it past an IRB.  Another great example from Harwood-you’re not going to perform a RCT on the safety of motorcycle helmets-it’s going to be a population study.

 

Or from a Bayesian (E. Kulstad):  it’s situational-for these questions, our pre-test probability is already weighted.  For diet, if the alternative hypothesis is that a McDonald’s diet is the same or better than Mediterranean, well nobody’s saying that.  Even before seeing the data, you have an established pre-test probability. 

So even with their limitations, a prospective observational cohort trial is a satisfactory trial design for certain questions, and as Sola said, a well done meta-analysis of high quality observational trials can provide very useful information.

 

 

Article 3:  Bowen S, Marlatt A: Surfing the urge: brief mindfulness-based intervention for college student smokers. Psychol Addict Behav. 2009 Dec;23(4):666-71.


Mindfulness, or an attentive awareness especially of the present moment, is an important part of Buddhist teachings.  More recently, the concept has been incorporated into western therapies for treatment of mental illness and drug addiction.  This RCT examined effects of a brief mindfulness technique on smoking behavior among nicotine-deprived college student smokers.  A total of 123 participants either received instructions on the technique or were advised to use their own coping skills after being deprived of cigarettes for 12 hours and then taunted with cigarettes.  Primary outcome was the number of cigarettes smoked over the 7-day follow-up period, and the intervention group on average smoked 1.5 fewer cigarettes.  How does it work? Bret-it’s all about acceptance, giving up control.  By observing negative feelings or urges without reacting, participants may learn alternative responses to negative experiences.

 

Lots and lots of problems with the study:  small group sizes, short follow-up, everything was self reported, these were college psychology students with low levels of nicotine addiction and were light smokers (not necessarily externally valid when thinking about our patient population), questionable clinical significance (decrease of 1.5 cigarettes), and they didn't query the intervention group to see if they even used the mindfulness technique.  There is also the well described unconscious desire to please the investigator by providing the “correct” answers on a follow-up survey.  Bottom line-even with the limitations, it’s an intriguing low cost concept, and for our patients, would be interesting to see it studied as part of a more comprehensive smoking cessation program.

 

Finally, a few comments from Dr. Principe.  He encouraged reinforcement of the positive as a means of sustaining behavior change, rather than emphasizing denial and deprivation.  He advocates mindfulness and appreciating/living in the moment, nutritional education, and exercise as components of overall good health. He stressed the importance of “moving along the spectrum”:  we can’t take our population abruptly to an Ornish or Mediterranean lifestyle.  He recommends the book The Blue Zones by Dan Buettner, published by National Geographic, as an exploration of the factors that contribute to certain populations around the world leading extraordinary long, healthy lives.

Low Risk Chest Pain- Accelerated Rule-out

Do you hate chest pain, or is it easy if you have a chest and you have pain you get admitted. Or if you are ultra-low risk (usually 0 risk) you just send them home, no testing. But what about people with a decent story? Is it safe to send someone home, or do these people need further testing as recommended by the AHA (stress within 72 hours?) 

Chest pain represents about 10% of ED visits and a diagnostic strategy to safely identify and discharge low risk patients in a rapid fashion has been debated. A quarter of all hospital admissions are secondary to chest pain, and about 85% do not leave the hospital with a diagnosis of acute coronary syndrome.

Today’s mini JC focuses on the safety of a rapid identification of the “low risk” patient who can be safely discharged from the hospital with a 2 hour-accelerated diagnostic protocol.

1.  Why is this topic important?

Rapid evaluation and discharge of low risk chest pain may substantially reduce both outpatient and inpatient LOS, cost to the patient and health care system, and reduce physician liability. A rapid diagnostic protocol that can discharge patients within two hours of being evaluated represents a superior alternative to admission for serial enzymes, and possible a stress test (which more and more seems to be of limited utility.)

2.  What does this study attempt to show?

The study attempts to demonstrate that in patients whom are low risk defined as two negative troponins, 2 hours apart, a TIMI risk score of 0 (see below) and no new EKG changes (see manuscript, but a LBBB if it was old did not preclude them from enrollment in the study) can be safely discharged home with a 30 day event rate < 1%.  Major adverse cardiac events (MACE) were identified as: death, cardiac arrest, emergent revascularization procedure, high degree AV block, and acute MI. AMI is defined as having ischemic symptoms + a rising biomarker (troponin). In this protocol no physician judgment including (patients whom the provider deemed to be low risk) was included to provide more generalizable results.  It is important to note this a prospective observational study, not an intervention study.

 3.  What are the essential findings?

1,975 patients were prospectively enrolled, and ultimately 302 patients (15.3%) developed MACE within 30 days. The accelerated diagnostic protocol (ADP) classified 392 patients (20%) as low risk. One lone patient who was ADP negative developed MACE at 30 days giving the ADP a sensitivity of 99.7%, (95% CI 98.1%-99.9%), and a specificity of 23.4% (95% CI 21.4% to 35.4%).  The lone patient that was missed by the ADP was previously healthy with no risk factors, a normal EKG, and first trop of <0.01, and 2nd of 0.03, with the 3rd troponin being 16.8. Angiography showed right coronary artery and circumflex artery stenosis and he had stenting in both vessels. No cardiac problems at one year follow-up.

In addition, the supplemental data from table’s 2A/B suggests that 75% of ADP negative patients underwent further diagnostic evaluation with stress testing.  7 ultimately had “non-emergent cardiac catheterization with revascularization.” Many patients were exposed to potentially dangerous anti-coagulants. 

Table 2a Use of further investigations in ADP negative patients within 30 days

 

Median # of days post presentation

Inpatient

Outpatient

CT Coronary Angiogram

2.0

17

2

Stress Radionuclide

5.5

11

14

Stress Echo

11.5

5

7

Stress ECG

0.5

265

18

Angiography

2.6

26

3

 

Table 2b Acute treatments on ADP negative patients within 30 days

In-patient Therapy 

Number

Heparin

8

LMW Heparin

20

G2b3a inhibitor

9

Heparin + G2b3a inhibitor

0

LMW Heparin + G2b3a inhibitor

27

Non-emergency Revascularization

7

ADP = Accelerated Diagnostic Protocol, LMW = Low Molecular Weight.

 

4.  How is patient care impacted?

Although this study is not an intervention study, it is strong food for thought. Using a TIMI risk score of 0, no new EKG changes, and 2 negative troponins, (<99%ile for your lab) 2 hours apart suggest these patients are low risk for MACE at 30 days. This study suggests that about 1/5th of our chest pain patients, who make up approximately 10,000 patients visits a year, can be safely discharged within a few hours of their initial ED evaluation. That’s 2,000 patients annually that could avoid admission and further diagnostic testing as an inpatient.

5.  Is this an area of controversy?

Of course. The study is prospective observational, and 75% of ADP negative patients ultimately had further diagnostic testing. In order for us to believe that an accelerated diagnostic protocol is useful the protocol needs to be used on patients in a prospective fashion (patients actually getting assigned the ADP vs. traditional care), and followed for 30-day outcomes.

Others may state that although only one lone patient in the ADP negative group developed MACE at 30 days, many more ADP negative patients received further diagnostic testing (CTCA, stress echo/EKG/radio nucleotide, and angiography with 7 patients ultimately getting non-emergent revascularization.) 

What do I think about this? Well if you were to take a cohort of patients walking around who developed chest pain > 50, many may have coronary artery disease, and if cathed may have a >70% lesion. However, what we have learned from prior mini JC’s is that stable coronary artery disease can be managed medically with equivalent if not superior results to invasively. Therefore, the benefit of an invasive strategy in these patients with NO EKG changes, negative biomarkers and perhaps a positive stress test must be questioned.

6.  Major Limitations of the study?

As stated previously this is a prospective observational study, not an intervention study, although not a limitation of the study itself, it may limit the applicability to your daily practice. However, other studies have suggested similar results including 30 day outcomes in “low risk patients” when evaluating CCTA. Therefore if you have a low risk patient by history, EKG, and a delta troponin at 2 hours it is reasonable to discharge those patients home with follow-up with their primary medical doctor.

The argument against this strategy is that you are missing the “unstable angina people”.  One important factor to remember is that likely the old “unstable angina patient” who came in 20 years ago before ultrasensitive troponins likely was having some trop leak as they exerted themselves with a partially occluded coronary artery. Those patients are likely called NSTEMI nowadays. In addition, plaques that are about 70-80% lesions are much less likely to rupture than are plaques that are say 20,30,40% as these plaques are much more unstable. These type of lesions will not be picked up by a stress test, and wouldn’t get a stent if they were cathed, and maximal medical therapy would be undertaken. Therefore the utility of identifying a plaque that may rupture is inherently flawed.

I reference you to an alternative approach, used at the mayo clinic, for placing the power of decision making back into the patients hands.  

I know two articles in one mini JC, but try using it next time you have a patient with chest pain, and let me know how it works for you! Essentially it is a protocol that after initial evaluation patients are given a pamphlet with 4 options to help them decide what they ultimately want to do. See chest pain decision aid I attached for you. 

Hess E et al. The chest pain choice Decision Aid. Circ Cardiovasc Qual Outcomes. 2012;5:251-259.

Ultimately I think an ADP is useful, and that patients who meet the aforementioned ADP negative criteria can be safely discharged home, and that further diagnostic testing is not warranted from the emergency department. I would recommend having that discussion using the decision aid with your patients so that they can be part of the medical decision making process. 

Until next time!

 

-Dave

 

2-Hour accelerated diagnostic protocol to assess patients with chest pain symptoms using contemporary troponins as the only biomarker: the ADAPT trial. Than M, Cullen L, Aldous S, Parsonage WA, Reid CM, Greenslade J, Flaws D, Hammett CJ, Beam DM, Ardagh MW, Troughton R, Brown AF, George P, Florkowski CM, Kline JA, Peacock WF, Maisel AS, Lim SH, Lamanna A, Richards AM. J Am Coll Cardiol. 2012 Jun 5;59(23):2091-8

CCTA for Low Risk Chest Pain

Mini JC #4, sorry I'm a week off, but it's been a little busier than usual. This week I figured we would cover something recently in EM RAP, and I am pretty sure it will be coming down the pipeline in the future months to come. 

Today's mini JC focuses on what to do with the low risk-intermediate risk chest pain who you are evaluating for outpatient therapy. Is it your policy to do 2 sets and dc, do you obs and do inpt stress, do you admit and let someone else figure it out?
Chest pain in the low-int risk patient becomes the source of angst and much disagreement between physicians depending on your risk tolerance. Today we will look at the NEJM topic on the use of CCTA for the evaluation of patients with chest pain presenting to the emergency department. 
1. Why is this topic important?
The evaluation and safe discharge of patients with low - intermediate risk chest pain is difficult. Missed myocardial infarctions represent the largest proportion of filed law suits against emergency physician and the lack of sensitivity of any one particular test for excluding coronary artery disease makes safe discharge problematic. Currently the recommendations and gold standard test is a stress test "performed within 72 hours", which has a poor sensitivity and specificity, with an undefined time window for how long a "negative stress test" is protective, if one even exists. CCTA is a non-invasive method of evaluating patients for the presence of coronary artery disease, not dissimilar to a cardiac cath, without the risks of periprocedural MI and catheterization related complications. 
2. What does the study attempt to show?
Low to intermediate probability chest pain patients who the physician deemed required admission or further diagnostic testing to evaluate for coronary artery disease were eligible for inclusion. Patients were assigned to two groups one arm was the coronary CT angiography arm, and the other arm was the usual care "whatever you did before arm". The study attempted to show that a negative CCTA was non-inferior to usual care in terms of safety outcomes at 30 days (death, or MI). 
3. What were the findings?
1370 subjects were in enrolled in a 2:1 fashion CCTA and usual care for a total of 908 CCTA and 462 usual care patients. Only 84% of patients who were assigned to the CCTA arm actually got a CT (767/908) mostly due to elevated resting heart rate. In the subset that received a CCTA 83% (640/767) had maximal stenosis < 50% (defined by investigators arbitrarily as the cutoff for non-significant coronary artery disease.) 7% (52/767) had stenosis between 50-69%, and 4%(28/767) had stenosis > 70%, 6% were non-diagnostic. Of the 640 patients who had a negative CCTA examination 0 had death, or MI at 30 days from the index visit. In addition, in the usual care arm there were also no deaths or MIs at 30 days. Overall 5.1% of patients in the CCTA arm got angiography and 4.2% of patients in the usual care arm got cardiac catheterization. Patients in the CCTA arm were more likely to be diagnosed with coronary artery disease than in the usual care arm when discharged from the hospital. 
4. How is patient care impacted?
This is an interesting study and a lot of data and questions are left unanswered. First off almost all of these "low-int" probability patients did well with no deaths at 30 days in any group. There were a few MI's in each group (none in the negative testing group) but the paper does not describe how these were diagnosed, and does not describe what the delta troponin's were. However, I think what can be interpreted from the study is that a negative CCTA is equally as effective at identifying a patient who is safe for discharge home from the emergency department as patients who would have otherwise been admitted. Therefore, in a low to int probability patient with chest pain CCTA is as effective as stress testing (keep in mind you are still doing the delta troponins) for identifying patients at low risk for death or MI at 30 days. 
5. Is this an area of controversy?
Absolutely. There are many questions left unanswered in this study. Almost all of these patients did well, based on being placed by an ED doc into a low risk category based on HPI and PE, and then even stratified to a lower risk group with negative troponin's. Therefore is a test like CCTA which results in radiation exposure, and increased health care costs a necessary test when the pretest probability of disease is so low? For example, what if instead of CCTA we looked at the WBC and if it was between 4-20K (99% of patients) we identified these patients as low risk to go home. It would have performed equally as well as the CCTA in terms of outcomes at 30 days in terms of death, or MI, simply because no one died or had an MI in this group, despite the fact that you know a WBC between 4-20K has no relationship whatsoever to whether or not a patient will have an MI. 
Therefore what we really need to know is what predicted the patients who were going to have an MI, (the very few that did). Was it a positive troponin, or was it a positive CCTA, and if the CCTA predicted an MI prior to a rise in troponin we have learned something valuable through its use. I think most of us know that a stenosis of > 50% doesn't mean its causing your chest pain, however. A ruptured plaque in a patient with a 30% lesion can become a 100% lesion immediately and result in a STEMI, so is a CCTA really predicting who has the MI or was it the positive biomarker like troponin? I think secondary analysis are coming down the line from this very interested novel study. 
6. Major limitations of the study?
Not all patients enrolled in the CCTA arm got a CCTA, and I think the paper is confusing in what happened to the patients who did not get CCTA. Also in the patients with a positive CCTA how many got cathed and what were the results? If they didn't get cathed what was the point of identifying a patient with a >70% lesion on the CCTA? These questions are not delineated in the paper. Another limitation of the study, is what I eluded to earlier. If you test a population with such a low prevalence of disease how can you accurately determine how effective the diagnostic strategy is compared to the gold standard? In this cohort of 1370 patients there were only 13 MI's in total and NO deaths. 

Litt et al. CT Angiography for Safe Discharge of Patients with Possible Acute Coronary Syndromes. NEJM 2012. 


-dave

Pediatric Fever Control

Fever is often observed in patients with documented infection, and we know from personal experience that fever makes you feel like ass, so we take motrin and tylenol to feel better, but what does this mean for our patients. Our we fighting evolution here?
Egi et al performed a prospective observational study on the association of body temperature and antipyretic treatments with mortality in critically ill patients with and without sepsis, excluding patients with neurologic injury (we know fever is bad when you have a neurologic insult). 
Here is what they found:
1. Why is this topic important?
It is almost reflexive to give febrile patients tylenol or Motrin for fever control. We know that fever creates tachycardia, and is uncomfortable for patients. However, fever is a natural response and multiple studies have demonstrated an increase in heat shock proteins, and a more robust microbicidal activity in patients with fever when compared to patients with euthermia or hypothermia. However, patients who are not infected may also have fever as part of an inflammatory cascade, and this could be detrimental. In this study Egi et al examined the associated between mortality and the treatment of fever in both septic and non-septic patients who were admitted to the intensive care unit with fever. 
2. What does the study attempt to show?
The study attempts to identify associations between fever and outcomes in septic and non-septic patients who were treated with pharmocologic agents like tylenol and Motrin or cooling blankets after being admitted to the intensive care unit. 
3. What were the findings?
1,425 consectuctive adult critically ill patients WITHOUT neurologic injury requiring > 48 hours in the intensive care unit were admitted in 25 ICU's. Every 4 hours body temperature was recorded and antipyretic treatment was monitored until ICU discharge or 28 days after ICU admission, whichever occurred first. Patients were divided into septic and non-septic patients, and the maximum body temperature and use of antipyretic treatment were assessed separately to mortality. 
Treatment with non-steroidal anti-inflammatory drugs (NSAID's) or acetaminophen INDEPENDENTLY increased 28-day mortality for SEPTIC patients (OR:  NSAID's 2.61 p = 0.028 and OR Acetaminophen 2.05 p =0.01), but did not change mortality for non-septic patients (NSAIDs 0.22 P =0.15, Acetaminophen 0.58 P =0.63). Application of physical cooling methods did not associate with mortality in any group. 

4. How is patient care impacted?

Since this is an observational study only association and not causality can be inferred, but this is strong food for thought regarding the aggressive treatment of fever with antipyretics in patients with suspected sepsis. To reiterate the point that fever is a response to stimulate the body's innate immune activity through the production of heat shock proteins and inflammatory mediators that stimulate bactericidal activity of leukocytes. Although fever is associated with increased volume loss, and increased metabolic demand, intravenous fluids, antibiotics, cooling blankets may be a more suitable alternative to antipyresis in the sicker cohort of patients (septic going to the ICU, not febrile kids in the PEC). 
5. Is this an area of controversy?
Since many physicians regard fever as an evil that must be treated, and some won't ever discharge a patient home until the fever has broken I would say yes. Although there are some adverse effects to fever (as previously mentioned and febrile seizures due to rate of rise in children), there advantages to a billion years of evolution. Maybe our bodies have it right. 
6. Major limitations of the study?
The study is observational, and in the ICU patients could be treated with antipyretics at the attending physicians discretion. Maybe sicker patients got tylenol and Motrin (although multivariate analysis should have accounted for this variability). Either way a randomized placebo controlled trial would be necessary to draw practice changing conclusions. 
In summary fever is a common occurrence in septic patients, and this study demonstrated an associated with increased mortality when treating septic patients in the intensive care unit with tylenol and motrin for fever control. This suggests fever may be a host response and suppression of the febrile response might worsen outcomes in septic ICU patients. 

-Dave

To stent or not to stent?

Now onto today's mini JC topic Stents vs medical therapy ALONE for stable coronary artery disease

Stable coronary artery disease (or those without EKG signs of ischemia or positive troponin) are often admitted to the hospital, especially if they carry multiple risk factors. Most of us never followup on these or check in care connection the next day to see what happens to our stable chest pain patients. Usually we thought they probably couldn't go home OR  we had hoped that if they had some large coronary artery lesion that needed stenting we would have saved a future MI or maybe a fatality. 
A meta-analysis was recently completed comparing MAXIMAL medical therapy to percutaneous coronary intervention in patients with STABLE coronary artery disease. This study was modern that only RCT who included patients who got modern maximal medical therapy (beta blockers, ACEI, anti-plt regimens) compared to stenting (no angioplasty alone). 
Heres what was found:
1. Why is this topic important?
We admit lots of chest pain patients to the hospital with vague stories with negative troponins and non-ischemic EKG's, which increased LOS, and hospital resource utilization, the question remains where is the benefit?
2. What does the study attempt to show?
This is a meta-analysis of RCT comparing maximal medical therapy to stent placement in patients with stable coronary artery disease, in the modern medical era (ACEI, BB, new fancy anti-plt agents). Outcomes included: Death, MI, or non-planned revascularization and persistent angina
3. What were the key findings?
8 trials for a total of 7229 patients were found.
Death rate for stent implantation vs medical therapy was 8.9 vs 9.1% perspectively (OR 0.98 95% CI 0.84-1.16)
NON-fatal MI for stent 8.9% and medical therapy 8.1% (OR 1.12 95% CI 0.93-1.34)
UN-planned revascularization for stent vs medical 21.4% vs 30.7% (OR 0.78 95% 0.57-1.06)
Persistent angina 29% and 33% (OR 0.8 95% CI 0.6-1.05) 
4. How is patient care impacted?
Patients with stable coronary artery disease should be placed on maximal medical therapy and seen in their doctors office not admitted to the hospital where incentives to have invasive procedures persist. The fact that no benefit is seen in mortality or recurrent MI, AND many of these studies did not use troponin measurements (still using CK-MB's) means that the Nonfatal MI was probably higher in the stent group than actually demonstrated, this is because stenting has a periprocedural MI rate between 5-30% due to distal embolization of plaques, or distal side occlusions and was not detected on the less sensitive CK-MB assays. 
5. IS this an area of controversy?
Yes, cardiologists have a strong incentive to cath patients who likely have disease (09% stenosis, 70% stenosis etc) and place stents, they think and we think this is helping patients live longer and symptom free. But in reality its a costly intervention and we have incomplete evidence of the pathophysiology of plaque rupture. therefore what we end up with is an inability to reject the null hypothesis, PCI in patients with stable CAD does not result in a mortality reduction when compared to maximal medical therapy alone. 
6. Major limitations of the study
This is a meta-analysis of randomized controlled trials, and they tried to select out patients who got MAXIMAL medical therapy and stents + MAXIMAL medical therapy. They had almost 7500 patients, but did not show a statistical or clinically significant benefit to stents over medical therapy for stable CAD. This nebulous outcome of unplanned revascularizations tended to be less frequent in the patients who got stented previously, but I think this is more INHERENT in the fact that they already had PCI and this made a cardiologist less inclined to interrogate the patients coronaries AGAIN. 
Initial Coronary Stent Implantation With Maximal Therapy vs Medical Therapy Alone for Stable Coronary Artery Disease
Stergioplous et al
-Dave

Evaluation for PE

Hey guys,
Its been far too long for the mini JC series, but thats what happens when you are about to have a baby. This JC was stimulated by Elise, Harwood and my discussion about a cancer patient and the evaluation of pulmonary embolism. I'll set it up for you and you think about how you would evaluate the patient.

65 y/o F presents to the ED with shortness of breath. She says this started 2 days ago, and has worsened since that time. She is a breast cancer survivor and is undergoing active chemotherapy. She is not on home O2, but she has smoked and does think she has COPD as evidenced by her inhalers. 
VSS: 95, 95% on RA, 110/70, 37.1 (R), 18 (real) FSG 99
Lungs are reduced BS at bases otherwise unremarkable 
She has no signs or symptoms of DVT, she has never had a blood clot, no recent surgeries. 
What labs and imaging studies do you want?? (US machine is broken sorry!)

SO today's study is the CHristopher study looking at the evaluation of pulmonary embolism in the emergency department. 

Van Belle A et al. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA 2006. 11; 295(2): 172-9.

 

 

1. Why is this topic important?

 

Pulmonary embolism is a commonly feared diagnosis in emergency departments in the United States. The frequency with which pulmonary embolism has been diagnosed has increased over the past ten years, with the number of massive and sub-massive pulmonary embolisms remaining relatively unchanged. This is due in large part to the over zealous testing, and the subsequent diagnosis of likely non-clinically significant pulmonary embolism. A simplified dichotomized clinical decision rule, D-dimer testing and computed tomography may reduce the number of unneeded evaluations for pulmonary embolism in the United States today.

 

2. What does the study attempt to show?

 

Consecutive patients in whom the treating physician suspected the possibility of pulmonary embolism, defined as sudden onset of dyspnea, sudden deterioration of existing dyspnea or sudden onset of pleuritic chest pain, without another apparent cause, could be potentially included in this study. The study then applied the wells score, dichotomizing patients into pulmonary embolism unlikely (score <4) and pulmonary embolism likely (score >/= 4). In all patients with a wells score of less than 4, a d-dimer test was applied, with the cutoff of >/= 500ng/mL being considered positive. If the d-dimer was negative, no further investigation was obtained, but if the d-dimer was positive, or if the wells score was >/= 4 a CT was obtained to evaluate for the diagnosis of pulmonary embolism. The study attempts to demonstrate that using this simplified dichotimzed approach to the wells score <4, and a negative d-dimer assay patients can be safely discharged home without further diagnostic evaluation.

 

3.  What are the key findings?

 

3306 study patients were eligible for inclusion, and 2206 patients (66.7%) were classified as unlikely as having a pulmonary embolism using the dichotomized wells score and underwent d-dimer testing. 1057 patients had a normal d-dimer, and 1028 (32%) were not treated with anti-coagulation.  Subsequent nonfatal venous thromboembolism (VTE) occurred in 5 patients (0.5% [95% CI 0.2%-1.1%]) no patients in this group died from a pulmonary embolism at 3-month follow-up.

 

In the high-risk wells group (wells score >/= 4) 674 patients (20.4%) were included. CT excluded pulmonary embolism in 1505, of which 1,436 patients were not treated with anti-coagulants; with a 3-month incidence of VTE of 1.3% (95% CI, 0.7%-2.0%). In patients in whom the initial CT scan was negative, PE was considered a possible cause of death in 7 (0.5% [95% CI, 0.2%-1.0%]). Overall 674 patients (20.3%) were diagnosed with a pulmonary embolism in patients included in this study cohort. 

 

4. How is patient care impacted?

 

This study demonstrates that in patients in whom a physician suspects the diagnosis of pulmonary embolism, the use of a dichotomized wells score < 4 and a negative d-dimer assay negate further evaluation for pulmonary embolism, with a < 1% incidence of VTE at 3-months, and no fatal pulmonary embolisms identified at follow-up.  

 

 

5. Is this an area of controversy?

 

The evaluation for pulmonary embolism continues to produce anxiety amongst emergency physicians. However, one of the most striking finding in this study is that the overall incidence of pulmonary embolism was 20%. Compare this to the PERC study by Dr. Kline in which the incidence was 7%, this is almost a 3-fold increase in the number of patients with pulmonary embolism. In addition, just 0.15% of patients in the Kline cohort ultimately died from pulmonary embolism, where in the Christopher Study mentioned here 0.5% died from fatal pulmonary embolism. This suggests that emergency physicians, particularly in the United States, evaluate for pulmonary embolism in a lower risk cohort than physicians in other countries. This occurs despite the fact that there is no mortality benefit, and there is a potential for harm.

 

6. What are the major limitations of the study?

 

 Several patients received anti-coagulation despite being low risk for pulmonary embolism, or having been excluded for pulmonary embolism for other pathologic causes. This could have reduced the number of VTE identified at three months in the negative, or low risk cohorts.  Also, the type of CT detector used was not standardized, and the d-dimer assay varied by site. These represent potential confounders of the study results. However, the generalizability of the study and the large number of patients included with few exclusion criteria make this an important piece of the literature for emergency medicine physicians. 

 

Reviewed by David Barounis


So I know this was written more research-like and I apologize, but the end result is what are you going to do about evaluation of PE in the above patient. 

 

I think in our patient  her wells score was between 1.5 (active cancer) and a 4 (if you said diagnosis is PE or at least as likely). If you got a trop and a BNP and they were both normal (< 0.02, and a BNP< 100) would you feel better? I think I would, and a diagnosis of PE in my mind is < 2%. However I don't think its unreasonable to get a d-dimer, and if you have a negative d-dimer, a neg trop and a neg BNP in this patient population I think the above study suggests you really do not need to CT scan these patients to look for PE. I think many people would not even get a d-dimer either. Far too often I think d-dimer is used in the ultra low risk patient or "0" risk patient, instead of the patient it was meant for, which is our wells score <4 patients. 

 

 

 

 

-Dave