Treating PE in the ED

Outpatient treatment of pulmonary embolism on the rise.


If a patient shows up in the emergency department with a pulmonary embolism (PE), is it safe to send him home? The protocol in many hospitals says absolutely not: The vast majority of PE patients are routinely admitted for several days to monitor their condition and supervise the start of anticoagulants. However, some hospitals are cautiously exploring ED treatment and discharge for PE.

Caution is understandable, since the overall 30-day mortality rate for PE is between 10% and 30%, according to published studies. But a growing body of evidence suggests that low-risk PE patients can safely skip the inpatient stay, or at least make it shorter, if they have access to appropriate anticoagulant medications, home support, and outpatient follow-up.

Photo by Thinkstock
Photo by Thinkstock

If hospitals can accurately identify the low-risk patients, and put appropriate supports in place to care for them as outpatients, experts said, they may be able to cut the cost of these patients' care significantly and save them the wear and tear of a hospital stay, without jeopardizing their health in any way.

“If you're not going to do better in the hospital, we want you to be at home,” said Margaret Fang, MD, FACP, a hospitalist and director of the anticoagulation clinic at University of California San Francisco Medical Center. She says outpatient treatment of PE patients is made possible by advances in anticoagulants. “Ten or 15 years ago, many patients had to be hospitalized for IV heparin,” she said, but now low-risk patients can be sent home with oral or injectable medications.

Overall, U.S. emergency departments see thousands of PE patients every year, but no more than 15% of them are discharged without an inpatient stay, according to an analysis of 10 years of national emergency department data published in March in Respiratory Research.

The number of PE diagnoses in the ED is growing, primarily because the condition can now be readily confirmed via CT scan, the study concluded. Inpatient treatment for PE patients averages $8,000 per stay, according to the analysis, and the authors estimated that half the patients in the study could have been safely discharged for outpatient follow-up.

Scoring risk

That estimate is based on the patients' scores on the Pulmonary Embolism Severity Index (PESI). The PESI, a list of 11 weighted criteria based on age, sex, medical history, and clinical findings like blood pressure and temperature, predicts the 30-day mortality rate for PE patients, stratifying them into 5 categories. The low-risk categories, I and II, have an estimated 30-day mortality rate of no more than 3.5%. Category V, the highest risk, includes patients whose 30-day mortality rate is between 10% and 24.5%.

The PESI, first published in 2005 by researchers from the U.S., Switzerland, and France, is currently the gold standard for predicting risk in PE patients. It is endorsed by the American College of Chest Physicians and the European Society of Cardiology. Its effectiveness as a tool to identify low-risk patients was validated in a randomized clinical trial published in the July 2, 2011, issue of The Lancet. In that study, 344 patients in PESI categories I and II were assigned to either initial outpatient treatment or inpatient treatment. The study found no significant differences in outcome between the 2 groups.

Practice patterns for PE vary widely, according to David R. Vinson, MD, a PE researcher and emergency physician with the Permanente Medical Group in Sacramento, Calif., and co-chair of Kaiser Permanente's Clinical Research in Emergency Services and Treatment Network (CREST).

For example, the hospital of the University of Ottawa, in Ontario, Canada, discharges more than half of its acute PE patients from the ED directly to outpatient care, Dr. Vinson said. In comparison, the rate at U.S. hospitals is more like 1%, according to an analysis of almost 2,000 patients in 22 U.S. emergency departments, published in 2011 in the Journal of the American College of Cardiology. Dr. Vinson's own group averaged about 8% in 2013, although he said that rate is on the rise.

An increase in outpatient PE treatment was also observed by Dr. Fang in a recent analysis of PE patients treated in 4 integrated health networks, published as a research letter in the June 1 JAMA Internal Medicine. While only 8.3% of the PE patients seen in the EDs were discharged to outpatient treatment, that percentage almost doubled over the time of the study, from 5.6% in 2004 to 11.1% in 2010.

But the expedited discharges didn't appear to be based on PESI criteria: More than a quarter of the patients treated as outpatients actually fell into high-risk categories. Despite this, clinical outcomes of patients discharged from EDs were good overall, Dr. Fang noted. Because the analysis was based on administrative data rather than detailed clinical findings, it was impossible to explore the reasons for discharge of higher-risk patients, the authors said.

Many emergency departments of Kaiser Permanente Northern California rely on the PESI to evaluate whether outpatient treatment is appropriate but supplement it with “common sense” criteria, such as significant right ventricular strain on a chest CT or echocardiogram, active cancer, or barriers to adherence like intoxication or homelessness, said Dr. Vinson.

The PESI doesn't rely on radiology or lab results, which makes it easy to apply in any setting, but a paper in the March 2015 CHEST explored using a lab test—a highly sensitive assay of cardiac troponin I—to supplement, and perhaps even replace, a PESI assessment.

The researchers analyzed the records of about 300 patients with confirmed acute PE. Among those who had undetectable levels of cardiac troponin I, no deaths occurred. Among patients with detectable cardiac troponin I, 6% died in the hospital. The test results predicted risk better than the patients' PESI scores.

“If the highly sensitive troponin assay is negative, it indicates that the burden of the PE is small enough that it doesn't cause significant right ventricular strain and injury. Therefore, troponin is a marker of how severe the PE is,” said senior author Rami Doukky, MD, a cardiologist at John H. Stroger Jr. Hospital of Cook County and Rush University Medical Center, both in Chicago.

He cautioned that the study was retrospective, single-center, and relatively small, and he hopes to do a larger prospective study. Meanwhile, his hospital is planning to institute a 24-hour observation protocol for low-risk PE patients, similar to existing protocols for low-risk chest pain.

A short-stay strategy represents a good middle ground between a full inpatient admission and sending a patient directly home, Dr. Vinson agreed, and a stay of less than 24 hours is common practice for many of Kaiser's low-risk PE patients.

Other factors

Choosing an effective risk-assessment tool is only a piece of the puzzle for deciding whether PE patients can be discharged, according to experts.

Hospital staff must assess the patient's home environment and access to medications, do a thorough job of educating the patient and family on how to administer those medications, and make sure there's a physician or clinic ready to provide follow-up care and support, Dr. Fang said. Before sending a PE patient home with anticoagulant medication, she calls the pharmacy to make sure it has the prescription, confirms that it's covered by the patient's insurance, and obtains any required prior authorizations.

In addition to having access to a prescription, patients need to indicate that they will adhere to it. “If you send them home and they don't take their medications, they are likely to experience recurrent, perhaps fatal, PE,” said Dr. Doukky. It may be more cost effective to simply provide the patient free anticoagulants, he said.

Patients discharged on warfarin are contacted within 3 days by Kaiser's anticoagulation services, to monitor and adjust dosages, check for complications, and confirm that they are following the directions received in the hospital.

“Shortening the typical inpatient stay limits the time available for patient education,” Dr. Vinson said. Speedy post-discharge follow-up can help mitigate that loss.

Dr. Vinson is currently analyzing the role that hospitalists play in determining how a PE patient will be handled. There's wide variation in practice: An interim analysis of 96 Kaiser Permanente patients with acute PE discharged directly home from the ED in 2013 showed that about a third had no hospitalist involvement whatsoever, a third received a curbside consult, and about a third received a full assessment with a documented history and physical examination.

Kaiser has developed an electronic decision support tool, based on the PESI and several other factors, that helps its clinicians identify PE patients who might be candidates for outpatient management. “It will be interesting to see if this [hospitalist involvement] pattern changed after we implemented that tool,” Dr. Vinson said.

Whether hospitalists or ED physicians make the choice, the trend toward outpatient treatment of PE will gain momentum as several large studies now underway are completed, he predicted. “There's going to be a spate of publications on this topic in the next 5 years,” Dr. Vinson said. “We suspect clinical practice will change significantly in the coming decade.”