Written by: Jon Andereck, MD (EM Resident Physician, PGY-2, NUEM); Edited by: Meghan Quigley, MD (EM Resident Physician, PGY-4, NUEM); Expert Commentary by: Scott Dresden, MD
A 68 year-old man rolls in on a stretcher with EMS, looking just fine. You stroll over to room 12 as the nurses help shift him over to the bed and EMS gives you a brief report. “He was watching the Bears destroy the Packers this afternoon, stood up to go to the bathroom, made it about halfway when he started to feel woozy. Got down to his knees but then passed out before he made it all the way down. Witnessed by his wife, no head trauma, was out for about 30 seconds, no seizure activity. Vitals were all normal, blood sugar 107, stable en route. History of hypertension, hyperlipidemia, borderline diabetic.”
After thanking the paramedics for their excellent work, you get to work yourself. You ask the nurse to grab basic labs with a troponin, place him on the monitor, and get an EKG. The rest of the history is fairly unremarkable, but he does make it clear that if you don’t let him get home to watch the end of the Bears game, he will curse you, and your children, and their children.
Crossing your fingers for a pristine workup, you are reassured by his labs – CBC normal, initial troponin negative, basic chemistry normal. His EKG is not wildly abnormal, but does have some new flattened t-waves inferiorly. What the heck do you do with this? You start sweating just a touch, fearing that these slight EKG abnormalities are going to doom you to admitting this patient and cursing your kin.
The San Francisco Syncope rule tells you he is not low risk due to the presence of new EKG changes, but while this rule is 98% sensitive for detecting people at risk of serious adverse outcomes it is only 56% specific . The Evaluation of Guidelines in Syncope Study (EGSYS) Score tells you he is low risk for cardiogenic syncope, with a 1-2% risk of cardiogenic syncope given his nonspecific EKG changes [i] . The ACEP Clinical Practice Guidelines are both outdated and vague, and do not necessarily force your hand one way or the other regarding the optimal disposition of this patient . Your gut tells you he’s fine but that you should probably admit him anyway. His gut tells you he’s fine and that you should let him go home. To admit or not to admit – if only there were more data to help out! Drumroll please…
Short-term prognosis and current management of syncopal patients at intermediate-risk: Results from the IRiS (Intermediate-Risk Syncope) Study
Study: Numeroso F, G Mossini et al. “Short-term prognosis and current management of syncopal patients at intermediate-risk: Results from the IRiS (Intermediate-Risk Syncope) Study.” Acad Emerg Med. Accepted 4 May 2016. Accepted article; doi: 10.1111/acem.13013 .
Study Design: Single-center, prospective, observational cohort study (Academic Hospital of Parma, Italy; 90,000 ED visits/yr)
Population: 347 adult patients between the ages of 18-90 who presented to the emergency department (ED) with undetermined syncope over the 6-month period from May – Oct 2013. Using criteria from the European Society of Cardiology (ESC), American College of Emergency Physicians (ACEP), and the Syncope Evaluation in the ED Study (SEEDS), the patients were stratified into intermediate- and high-risk groups as detailed in Table II. Low-risk syncope (Table II), patients with clear etiology of their syncope, and patients requiring hospitalization for other medical indications were excluded.
Figure 1: Patient Selection Flow Chart & Table II: Risk stratification of patients with syncope
Intervention: No interventions – this was an observational study.
Primary: Cumulative event-free survival. Events include:
- All-cause mortality
- Acute cardiovascular events
- Major acute bleeding requiring transfusion
- Major interventional procedures
- Syncopal recurrences causing major trauma
Secondary: Details about hospital admission for intermediate risk patients:
- Admission rate and destination
- Length of stay
- Final diagnosis
- Estimated costs associated with admission
Results: Patients identified as being intermediate risk had a very low rate (0.8%, n=2) of major adverse events at 30 days, and this risk was significantly lower than the rate of adverse events in the high risk group (27.8%, n=27, p<0.01). The two adverse events in the intermediate-risk group were one embolic stroke in a patient with a-fib and one pacemaker placement in a patient with sick sinus syndrome. In the high-risk group, 6 of the 27 adverse events occurred after hospital discharge.
Intermediate-risk patients were significantly younger (71.9 vs. 75.5, p<0.05) and had significantly fewer comorbidities (42.8% vs. 64.9%, p<0.01) than high-risk patients.
Of the intermediate-risk patients, 62.8% were admitted (39.2% to the ordinary ward, 23.6% to the Emergency Department Observation Unit (EDOU)). Mean hospital LOS was an average of 8.8 days for the ordinary ward and 1.7 days for the EDOU, with an average cost attributable to the syncope evaluation of $2,969 per ordinary ward admission and $431 per EDOU admission. Despite admission, 51% of ordinary ward patients and 29% of EDOU patients were discharged without explanation of the cause of syncope.
This study is meaningful as it directly addresses one of the most difficult disposition decisions we face in the ED: what do we do with those difficult intermediate-risk syncope patients whose disposition is not straightforward? If we discharge them home, will a serious event or death occur? Conversely, if we admit them for telemetry monitoring and echocardiogram, will a meaningful diagnosis even be obtained? Additionally, as healthcare costs continue to mount, will the expense of an admission and work-up benefit the patient?
The key finding of this study was the surprisingly low rate of adverse events (0.8%) for patients presenting to the ED with syncope who fall into the intermediate-risk category. The study authors conclude that it is reasonable to consider discharging these patients home with an outpatient follow-up plan as the additional workup is time consuming, expensive, and often nondiagnostic.
Looking closely at who actually qualifies as “intermediate risk” according to the study design is equally surprising. Unlike the San Francisco Syncope Rule that places any history of heart failure or any new EKG changes into a high risk category, and unlike the EGSYS rule that counts stable EKG changes such as LVH as high risk, the IRiS Study allows all of those patients to potentially fall into the intermediate risk category (see Table II). A patient can have 3 prior MIs and an EF of 25% but still qualify as intermediate risk in this study as long as their EKG does not meet the high-risk criteria and they are not presenting in decompensated heart failure. The focus here is on a patient’s change from baseline; if their chronic disease is stable, they do not necessarily fall into a high-risk category.
Regarding the secondary outcomes of the study, the length of stay and eventual diagnoses are most interesting (financial comparisons between Italy and the US are essentially meaningless). Over half of the patients admitted to the ordinary wards in this study never had an identified cause of their syncope by the time of discharge. This underscores the importance of recognizing that admitting a patient to the hospital does not guarantee a diagnosis and a safe eventual discharge. Is the time and cost of admitting all these intermediate-risk patients to the hospital truly the best thing for them? And do all “intermediate risk” patients carry a similar risk? Ay, there’s the rub.
Prospective, internally valid, and relatively objective criteria for distinguishing between high-risk and intermediate-risk patients.
Single center (low external validity), small patient population (leading to only 2 adverse events in the intermediate risk group), did not use a pre-existing decision rule to risk stratify patients, and some of the major adverse events were up to physician discretion (ie. placing a permanent pacemaker, etc.).
Is this study practice changing?
Probably not, but largely because of its lack of external validity. It is simply too small a study (n = 347) carried out at a single center in Italy, and therefore extrapolating its results to definitively change my practice in Chicago is unlikely. But the results - should they eventually be validated elsewhere – would likely change my practice.
A 0.8% rate of adverse events at 30 days is well within our somewhat arbitrary but highly necessary window of acceptable “miss rate” in emergency medicine. Compared to other common chief complaints, we apply the PERC rule to “rule out” PE by accepting its <1.6% false negative rate , and similarly apply the HEART score for low risk chest pain and accept its 1.7% rate of major adverse cardiac events within 30 days . Should this 0.8% adverse event rate be validated in future studies, this would fall below the “<2%” threshold already established in chest pain and rule-out PE. To be fair, the risk of working up and treating PE and ACS can be more harmful than the monitoring and noninvasive testing of most syncope evaluations, but there are still costs and potential harm associated with these admissions that should be considered. As of yet, I’m not likely to send home the patient with a history of heart failure with reduced ejection fraction who presents with syncope, but that may well change in the coming years.
Take Home Points:
Ultimately, this study highlights the importance of shared decision making. There is no one best answer in syncope management, as in most of the presentations we face in the ED. Such ambiguity can be frustrating, but it is also why many of us chose this field and why we can rest assured that our jobs will not be replaced by robots for at least a few years. Unless you are strong-arming all of your patients into a hospital admission, you are already doing the shared decision-making inherent in the disposition of these patients. This study gives us a few more data points to share in those conversations, and gives me a little bit more confidence that sending these patients home is not a discharge to sudden cardiac death.
Nice overview on medium risk syncope as described in this Italian study.
In reviewing the literature for this expert review, I’ve come to the conclusion that syncope does not lend itself well to clinical prediction rules. There have been multiple attempts at determining who is safe for discharge, many of them have done well in initial tool derivation, however have failed in validation studies. San Francisco Syncope rules have wide variation in results from a 74% sensitivity to a 98% sensitivity and pooled meta-analysis results with a 86% sensitivity (95% CI 83-89). OESIL Risk score seems to better identify low risk patients, however there was one study which found it’s sensitivity was only 88%.
The Canadian Syncope Rule includes points based on history (predisposition to vasovagal symptoms), and physician decision making (ED diagnosis). It has shown good results for detecting low risk patients in 6 Canadian EDs. Perhaps it will have better luck with validation than other studies have.
The problem is the needle in the haystack. These clinical prediction rules are being derived looking for risk factors for the needle, but they have to go through many many many haystacks before they find one. So they are good at ruling out the more common causes of syncope, the rare complications don’t come out until validation.
This finally brings me to the study at hand. IRiS, which essentially is a study of 250 patients with intermediate risk syncope as you defined above. You do a good job of pointing out some of the surprising results considering the existing literature on the topic. Maybe it’s the Italian wine that keeps them safe after syncope.
Additionally, categorizing patients as intermediate in the ED is impractical. I suppose if IRiS has great prospective validation, someone might make an app or an MD calc, otherwise it’s a bit of a cumbersome list to go through.
So unfortunately, while the results of this study seem to be promising, it’s necessary to wait for an external validation study before using it. Given the history of validation of syncope rules, I wouldn’t be surprised to see this list come up short too. For now, continue to rely on your clinical gestalt, and take solace in knowing that as of right now, it’s just as good as any clinical prediction rule for syncope.
If you need data to help you make a decision on admission, I say use the OESIL for now:
Also, if you can’t figure it out in the ED, it’s likely they won’t figure it out on the floor either. So the hospitalization is likely to be “unnecessary” in retrospect, if only we had that retrospectoscope in the ED!
Scott M. Dresden, MD, MS, FACEP
Assistant Professor; Northwestern University; Department of Emergency Medicine; Center for Healthcare Studies
Other Posts You May Enjoy
How To Cite This Blog Post
[Peer-Reviewed, Web Publication] Andereck J, Quigley M (2017, April 4). Intermediate Versus High Risk Syncope [NUEM Blog. Expert Commentary By Dresden S]. Retrieved from http://www.nuemblog.com/blog/syncope
- Quinn, J., et al., Prospective validation of the San Francisco Syncope Rule to predict patients with serious outcomes. Ann Emerg Med, 2006. 47(5): p. 448-54.
- Kariman, H., et al., Validation of EGSYS Score in Prediction of Cardiogenic Syncope. Emerg Med Int, 2015. 2015: p. 515370.
- Huff, J.S., et al., Clinical policy: critical issues in the evaluation and management of adult patients presenting to the emergency department with syncope. Ann Emerg Med, 2007. 49(4): p. 431-44.
- Numeroso, F., et al., Short term prognosis and current management of syncopal patients at intermediate-risk. Results from the IRiS (Intermediate-Risk Syncope) Study. Acad Emerg Med, 2016.
- Kline, J.A., et al., Prospective multicenter evaluation of the pulmonary embolism rule-out criteria. J Thromb Haemost, 2008. 6(5): p. 772-80.
- Backus, B.E., et al., A prospective validation of the HEART score for chest pain patients at the emergency department. Int J Cardiol, 2013. 168(3): p. 2153-8.
[i] In the EGSYS study, abnormal EKG findings were defined as: sinus bradycardia, atrioventricular block greater than first degree, bundle branch block, acute or old myocardial infarction, supraventricular or ventricular tachycardia, left or right ventricular hypertrophy, ventricular pre-excitation, long QT, and Brugada pattern .