Why Are We Undertreating So Many Pulmonary Embolisms?

LONG BEACH, California — A small fraction of patients with pulmonary embolism (PE) who are eligible for advanced therapies are actually getting them, reported investigators who conducted a big data analysis.

“Advanced PE therapy seems to be vulnerable to disparate use, and perhaps underuse,” Sahil Parikh, MD, a cardiovascular interventionalist at the Columbia University Medical Center in New York City, said when he presented results from the REAL-PE study at the Society for Cardiovascular Angiography and Interventions (SCAI) 2024 Scientific Sessions.

The underuse of advanced PE therapies is “the controversy,” Parikh said after his presentation. “It remains unclear what the role of invasive therapy is in the management of so-called high-intermediate-risk people. There isn’t a Class 1 guideline recommendation, and there is a very rapidly evolving trend that we’re increasingly treating these patients invasively,” he said.

“However, if you come to these meetings [such as SCAI], you might think everyone is getting one of these devices, but these data show that’s not the case,” Parikh said.

The analysis mined deidentified data from Truveta, a collective of health systems that provides regulatory-grade electronic health record data for research. The database included 105 million diagnoses made from January 1, 2018 to May 5, 2023; according to the diagnosis codes, 435,296 of these were for pulmonary embolism, and according to the procedure codes, 2072 patients — 0.48% of all patients with a PE diagnosis — received advanced therapy.

The researchers accessed data on patients treated with ultrasound-assisted catheter-directed thrombolysis or mechanical thrombectomy, identified from claims codes. Patient characteristics — age, race, ethnicity, sex, comorbidities, and diagnoses — were also accessed for the analysis. Earlier results were published in the January issue of the Journal of the Society for Cardiovascular Angioplasty Interventions. 

Less Intervention for Black Patients and Women

White patients were more likely to receive advanced therapy than were Black patients (0.5% vs 0.37%; P = .000), Parikh reported, and women were less likely to receive advanced therapy than were men (0.41% vs 0.55%; P = .000).

The only discernable differences in outcomes were in major bleeding events in the 7 days after the procedure, which affected more White patients than it did Black patients (13.9% vs 9.3%) and affected more women than it did men (16.6% vs 11.1%).

What’s noteworthy about this study is that it demonstrates the potential of advanced data analytics to identify disparities in care and outcomes, Parikh said during his presentation. “These analyses provide a means of evaluating disparities in real clinical practice, both in the area of PE and otherwise, and may also be used for real-time monitoring of clinical decision-making and decisional support,” he said. “We do think that both novel and established therapies can benefit equally from similar types of analyses.

Big Data Signaling Disparities

“That’s where these data are helpful,” Parikh explained. They provide “a real snapshot of how many procedures are being performed and in what kinds of patients. The low number of patients getting the procedure would suggest that there are probably more patients who would be eligible for treatment based on some of the emerging consensus documents, and they’re not receiving them.”

The data are “hypotheses generating,” Parikh said in an interview. “These hypotheses have to be evaluated further in more granular databases.”

REAL-PE is also a “clarion call” for clinical trials of investigative devices going forward, he said. “In those trials, we need to endeavor to enroll enough women and men, minority and nonminority patients so that we can make meaningful assessments of differences in efficacy and safety.”

This study is “real proof that big data can be used to provide information on outcomes for patients in a very rapid manner; that’s really exciting,” said Ethan Korngold, MD, chair of structural and interventional cardiology at the Providence Health Institute in Portland, Oregon. “This is an area of great research with great innovation, and it’s proof that, with these type of techniques using artificial intelligence and big data, we can generate data quickly on how we’re doing and what kind of patients we’re reaching.”

Findings like these may also help identify sources of the disparities, Korngold added. 

“This shows we need to be reaching every patient with advanced therapies,” he said. “Different hospitals have different capabilities and different expertise in this area and they reach different patient populations. A lot of the difference in utilization stems from this fact,” he said.

“It just underscores the fact that we need to standardize our treatment approaches, and then we need to reach every person who’s suffering from this disease,” Korngold said.