Syncope units: One solution to an expensive problem

These multidisciplinary teams function as part of EDs, observation or inpatient units. European studies suggest they may help shorten length of stay.

Syncope accounts for 740,000 emergency department (ED) visits and 460,000 hospital admissions in the U.S. each year. Diagnostic costs can climb into the tens of thousands of dollars, and costs of care are high: $8,700 per admission, with an average length of stay of 3 days, according to data from the U.S. Healthcare Utilization project and an analysis published in the January 2013 Progress in Cardiovascular Diseases.

Yet only about half of all patients admitted for syncope are actually discharged with the condition as their primary diagnosis. The disparity between suspected and actual syncope may occur because clinicians aren't using the most efficient diagnostic tools and procedures.

Photo by Thinkstock
Photo by Thinkstock.

The analysis in Progress in Cardiovascular Diseases indicates that the majority of inpatient tests—chest X-ray, telemetry, cardiac enzymes, echocardiography, cardiac stress tests, electroencephalography, and carotid ultrasonography—are associated with a diagnostic yield of less than 2% for syncope.

The paper's author, Benjamin C. Sun, MD, associate professor in the department of emergency medicine at Oregon Health & Science University in Portland, attributes this diagnostic inefficiency to two things: insufficient guidelines and conservative medicine.

“In my opinion, there haven't been validated diagnostic algorithms to help clinicians approach diagnosis of patients in an evidence-based way. In the absence of such tools, physicians will keep doing what they've always done,” Dr. Sun said. “The other thing that drives this behavior is the nature of syncope, where it's unclear what's going on and a small percentage of patients will have something potentially dangerous. Most people who present with syncope will not have [that kind of] problem, but physicians fear sending home a patient who then experiences sudden death.”

The default, then, is to admit patients with syncope, but this can create a financial burden if the Center for Medicare and Medicaid's recovery audit contractors decide an admission was unnecessary upon retrospective chart review, Dr. Sun added.

One solution some hospitals have turned to is syncope units—multidisciplinary teams that function as part of EDs, observation or inpatient units. While syncope units are fairly rare in the U.S., several European studies of hospital-based syncope teams, such as the Evaluation of Guidelines in Syncope Study (EGSYS) in Italy and an analysis of the Falls and Syncope Services (FASS) model in the U.K., suggest they improve diagnosis and shorten length of stay while using fewer tests.

Staffing the team

Staffing on syncope teams varies widely depending on the size of the hospital, the local patient population and the availability of specialists. In an EGSYS pilot study, 6 hospitals had syncope units that were arranged similarly. Initially, cardiologists managed the units within the hospitals' respective departments of cardiology, and the teams comprised specialized medical and support personnel. Other specialists, such as neurologists, were involved in patient management as necessary.

Eventually these teams evolved into virtual units that relied on a software triaging pathway used by specially trained staff. The EGSYS software is Web-based and was developed to help physicians follow the diagnostic pathway and recommendations of the European Society of Cardiology's guidelines for syncope.

In FASS, the local population was older, so geriatricians and internists with expertise in syncope led the teams. The teams were not isolated management units, but rather specialized teams with clinical expertise. They developed and used algorithms for treating falls and syncope, and distributed the algorithms on pocket-sized laminates to all medical staff in the ED and general internal medicine services.

Hospital generalists are good candidates to lead syncope teams, said Daniel D. Dressler, MD, FACP, professor of medicine at the Emory University School of Medicine in Atlanta.

“It may be helpful to have a specialist involved initially, but this is a high-volume presentation and there may not be enough of them or they may not have the time or availability,” Dr. Dressler said. “It could be most valuable to have generalists see a patient initially, determine if they can make the diagnosis, and involve a specialist like a cardiologist or neurologist when there is diagnostic uncertainty.”

When specialists take the helm, there is a potential risk for narrowed focus and more testing in pursuit of a diagnosis that fits the specialty. “When you're a hammer, everything is a nail,” Dr. Dressler said.

Cardiologists, at least, should be an integral part of a syncope team, countered Salvatore Rosanio, MD, PhD, professor of medicine and cardiology at the University of North Texas Health Science Center (UNTHSC) in Fort Worth.

“The most important thing is ruling out syncope from cardiac causes because that is associated with a high risk of mortality. When you rule that out, then move on to other etiologies that are associated with more benign processes,” Dr. Rosanio said. “Unfortunately, in my experience most of the time cardiologists are consulted late in the process. When the other physicians don't find anything, then they call the cardiologist—after a lot of unnecessary tests.”

Dr. Rosanio, who is leading the development of a syncope team at UNTHSC, noted that physicians who wish to create syncope teams will have to work with hospital administration to get buy-in and decide on staffing. At UNTHSC, team members are chosen based on their expertise in syncope management, their interest in working on the team, and their availability.

Selecting strategies

Once the issue of staffing is settled, the next step is for the team to decide what methodology and strategy to use.

“All clinicians on the team—hospitalists, specialists, ED physicians, and others—should have the same structured approach in their minds. The units that do well approach syncope in a similar way for all patients,” said Dr. Dressler.

An educational program should be in place that follows guidelines and algorithms for managing syncope, added Dr. Rosanio. Such programs are intrinsic to syncope teams in Europe, he said. At UNTHSC, there will be a three-month training involving key physicians on the team such as internists, emergency medicine physicians, neurologists, and cardiologists. Other personnel, such as nurses, will be trained as necessary.

Several organizations offer guidance on the evaluation, diagnosis and management of syncope that a team or unit could use to draw up its protocol (see sidebar, next page).

“You can also program your strategy into the electronic medical record, so that if your patient meets the criteria for admission, it will come up [in an alert],” Dr. Rosanio said.

Several barriers may stand in the way of forming and implementing a syncope team, such as a shortage of staffing, facilities or financial resources. Even if the resources are there, physicians may have to jump through considerable hoops to justify the creation of a team to administrators. Yet hospitals without syncope units still must ensure the condition is evaluated and treated efficiently, starting from the initial encounter with a patient.

Dr. Sun stressed the need for evidence-based protocols to determine whether a patient should be admitted in the first place. “If you don't have an observation unit, you could try to get the emergency medicine physician to use protocols to identify higher-risk versus lower-risk patients” for potentially dangerous outcomes such as sudden death, he said. “The problem is that the existing guidelines are not that good.”

Dr. Rosanio is trying to rectify that. He and his colleagues have proposed an algorithm for diagnosing and managing syncope, which was published in the January-February 2013 International Journal of Cardiology. The algorithm differentiates tests with high diagnostic yield from those that are less efficient and accurate.

“Right now, too many physicians refer patients to useless tests like carotid Dopplers, CT scans, MRIs, etc.” Dr. Rosanio said. “It's a disaster, and we need to put it in order.”

The algorithm rests on 3 key assessments to set the course for diagnosis and management: the patient's cardiac status, the number of syncopal episodes over 2 years and whether a single syncopal episode is associated with secondary trauma. It is based on existing research and takes a stepwise approach.

“If you go through all of the guidelines and evidence, this is just a more organized way of looking at it. If you have a patient with syncope who has one set of symptoms, you do one thing. If the patient has another set of symptoms, you do another thing,” Dr. Rosanio said.

Always consider the patient's history, added Gregory J. Misky, MD, a hospitalist at the University of Colorado at Denver.

“Know the patient you're getting the story from. A 20-year-old without medical problems is different from a 60-year-old with structural heart disease,” Dr. Misky said. “Spend the appropriate time sorting out what happened, what the symptoms were before the episode, and what the symptoms were after. If possible, talk to witnesses who were there.”