Pulmonary Embolism in the United States: A Report From the National PERT Consortium Authors: William Brent Keeling, Steven Horbal, Kenneth Rosenfield, Robert Lookstein, Richard Channick, Rachel P. Rosovsky, John M. Moriarty, Matthew D. Langston, Seth Sokol, Thomas M. Todoran, Eric A. Secemsky Published: 2026 ABSTRACT BackgroundPulmonary embolism (PE) remains the third leading cause of cardiovascular-related mortality, and few efforts have been made to study it exclusively through real-world registry data. This research shows demographic and outcomes data from a real-world, PE-specific database, yielding a contemporary examination of PE care in the United States.MethodsAll data were extracted from the National PERT Consortium registry. This study included consecutive patients diagnosed with PE from October 2015 to June 2024 at 54 US sites. All patients were assigned a PE-specific risk stratification score based on available data. Descriptive statistics were used to analyze the data.ResultsDuring the study period, 11,612 unique patients were included for analysis, with the majority (73.7%) classified as intermediate risk. Patients tended to be slightly obese (mean body mass index, 31.1 kg/m2) and were more frequently men (51.1%). A patient history of deep vein thrombosis was observed in 1 in 3 patients, and 1 in 4 had a history of malignancy. Patients with high-risk PE underwent intervention more frequently. Furthermore, patients with PE on average spent 4 days in the hospital. When evaluating survival by treatment types, 30-day and 1-year survival in patients who underwent catheter-based intervention were 95.1% and 92.3%, respectively. One-year survival was 67.8% for patients with high-risk PE.ConclusionReal-world data from the National PERT Consortium database demonstrate current practices in the care of patients with PE. Opportunities exist for better risk stratification and improvement in care across the spectrum of PE. Introduction Pulmonary embolism (PE) remains a significant cause of morbidity and mortality for both hospitalized and nonhospitalized patients. PE is the third leading cause of cardiovascular-related death, and it continues to be a leading cause of in-hospital mortality.1 Prior epidemiological research in PE showed that the overall incidence has been increasing for hospitalized patients.2 These same researchers detailed an increase in in-hospital mortality for patients with venous thromboembolism (VTE), highlighting the need for further research into prevention and intervention. The largest PE-specific registry created to date was the iCOPER registry, and the results from that were published in 1999.3 This registry included >2400 unique patients from >50 centers in 7 countries and served as an important benchmark for PE-specific research and data gathering; however, both the US patient population and treatment strategies for managing PE have dramatically changed since then. Considering the ongoing need for increased awareness and research into PE, the PERT Consortium was founded in 2015, and soon thereafter, the National PERT Consortium database was initiated. This database was initially conceived as a quality assurance tool not only to ensure member institutions could compare outcomes with national standards but also to facilitate large-scale biomedical research because of the extensive volume of collected data. Multiple research projects have grown out of the National PERT Consortium database, including important contributions to the medical literature on patients with high-risk PE,4 differences in outcomes following PE based on sex,5 and the cost of PE treatment.6 Nevertheless, the results of all patients within the National PERT Consortium database have not been profiled in totality or published to date. Hence, we sought to detail the demographics, interventions, and outcomes of >12,000 unique patients with PE contained within the National PERT Consortium database from 2015 to 2024, making this the largest study with a granular, prospective, PE-specific database. In so doing, this unique PE study gives broader insight into the modern treatment of PE and identifies ongoing areas for further research and quality improvement. Methods Study Sites and PERT Registry All data were extracted from the National PERT Consortium registry, maintained by the National PERT Consortium (N = 12,346). The National PERT Consortium registry is a prospective registry that follows patients throughout their hospitalization and follow-up. Data are collected on each patient for the duration of the inpatient stay, and both 30-day and 1-year vital, and hospital readmission status is recorded. Each of the datum elements was reviewed and fitted into the appropriate section of our registry (ie, demographics were all gathered in Section 1—Initial Assessment and Acute Treatment) (Central Illustration). This study included consecutive patients diagnosed with PE enrolled into the registry from October 16, 2015, to June 1, 2024, at 54 participating sites (Supplemental Table S1). Contributing institutions were trained in the use of the dataset prior to data collection. The Research Electronic Data Capture (REDCap) was used to collect deidentified data from each participating site. Either central (Advarra) or local institutional review board approval was required for all sites. As data are voluntary and provided at the discretion of the contributing institution, the cohort and study sample should be considered a convenience sample.Central IllustrationData element collection throughout the 5 stages of the PE journey. The National PERT Consortium database splits over 700 data elements into well-defined sections to detail the hospitalization and follow-up of each patient with PE. PE, pulmonary embolism. Data element collection throughout the 5 stages of the PE journey. The National PERT Consortium database splits over 700 data elements into well-defined sections to detail the hospitalization and follow-up of each patient with PE. PE, pulmonary embolism. Baseline Patient and Hospital Characteristics Among the 12,347 patients entered into the National PERT Consortium database, 735 (5.9%) were missing data to determine PERT risk classification and were excluded from analysis. All remaining patients (11,612; 94.1%) in the registry were included in this analysis. Baseline characteristics collected, including sociodemographic and comorbidities, have been previously reported.4,5 Patient age was calculated by subtracting the date of birth from the PE response team (PERT) activation date. Hypertension was present if (1) records indicated the patient was on antihypertensive medication or (2) if a history of hypertensive medication was present, and if multiple systolic blood pressure measurements were recorded, a patient’s average of highest and lowest systolic blood pressure being ≥130 mm Hg was used to approximate hypertensive status. Risk factors were determined based on existing conditions. Intensive care unit (ICU) admission was based on whether the patient spent any time in an ICU level of care during the hospitalization. Advanced therapies were defined as any use of systemic thrombolytics, catheter-directed thrombolysis (including ultrasound assisted and nonultrasound assisted), catheter-based embolectomy, surgical embolectomy, and/or mechanical circulatory support (including extracorporeal membrane oxygenation [ECMO]). PE Risk Classification Patients in this analysis were classified by risk severity using predefined criteria.7,8 Patients classified as low risk were defined by absence of hypotension, right ventricular (RF) dysfunction, or myocardial necrosis/strain as evidenced by cardiac biomarkers. Patients with low-intermediate risk (categorized in this work as intermediate risk) had evidence of RV dysfunction (detected via computed tomography [CT] or echocardiography) or myocardial necrosis/strain (elevated troponin or B-type natriuretic peptide [BNP]/N-terminal pro–B-type natriuretic peptide [NT-proBNP]). Patients with low-intermediate or high-intermediate risk (both categorized in this analysis as "intermediate risk") had both RV dysfunction (by CT and/or echocardiography) and myocardial necrosis/strain (elevated troponin and/or BNP/NT-proBNP) without signs of hypotension or shock. High-risk patients were defined by sustained hypotension (systolic blood pressure <90 mm Hg for >15 minutes), persistent bradycardia (heart rate <40 beats per minute), signs and symptoms of shock, or requiring inotropic support. Binary thresholds for laboratory tests included BNP/NT-proBNP >100, troponin >0.04 ng/mL or documented as elevated per specific hospital assay, lactate >2 mmol/L, and D-dimer >500 ng/mL. Outcomes The primary outcomes in the analysis were in-hospital survival and rates of in-hospital major bleeding. Death during the index hospitalization was determined by the site, as designated on the registry form. The International Society on Thrombosis and Haemostasis (ISTH) major bleeding criteria was used to define major bleeding. ISTH criteria include the following: fatal, intracranial, intraspinal, intraocular, retroperitoneal, intra-articular, or pericardial bleeding; intramuscular bleeding with compartment syndrome; or bleeding causing a fall in hemoglobin level of 20 g/L (1.2 mmol/L) or transfusion of 2 U of whole blood or red cells. Nonmajor bleeding was defined as any bleeding that did not meet ISTH criteria for major bleeding. Secondary outcomes included length of stay, which was estimated by subtracting the initial discharge date from the initial admission date, as well as 30-day and 1-year readmission, major bleeding, nonmajor bleeding, and survival. Statistical Approach Descriptive statistics were used to analyze all included registry participants. Descriptive tables report on the overall aggregated estimates and stratified estimates by PERT risk classification. Mean ± SD are reported for most continuous variables. Length of stay was reported as median and IQR. Frequency and percentages were reported for binary and categorical variables. Specific device types for systemic intravenous thrombolysis and catheter-based interventions were not included. To evaluate the relative use of each device type, reported frequencies and percentages are respective to the therapeutic type and not the total study population. Standardized mean differences (SMDs) were calculated to assess differences in age and race across PERT risk classification. An SMD threshold of 0.1 was used to indicate meaningful imbalance across the different levels of PE risk. Missing data were prevalent among many of the reported factors. Row-specific counts (n) are provided throughout the tables to characterize the completeness of the data. Owing to rounding, percentages may not add to 100%. Statisticians at the Michigan Clinical Outcomes Research and Reporting Program performed the analysis with R software version 4.4.0 (R Foundation for Statistical Computing). Results Of the 11,612 patients with PE included in this analysis, 11.6% (n = 1342) were categorized as low-risk PE, 73.7% (n = 8569) as intermediate-risk PE, and 14.6% (n = 1701) as high-risk PE. Sociodemographic Variables and Comorbidities Among the total cohort, the mean age was 63.5 ± 15.9 years, average body mass index was 31.7 ± 8.3 kg/m2, and 51.1% were men (Table 1). Among race or ethnicity categories, 65.7% identified as White and 21.8% as Black or African American, while 12.5% of patients were of other ethnicities. A majority of patients (73.7%) had a history of hypertension, and approximately 1 in 4 patients had a history of either an active or prior diagnosis of cancer. Specific risk factors for VTE included a history of a deep vein thrombosis (DVT) in 33.2%, recent surgery in 5.8%, active tobacco use in 15%, prior inferior vena cava filter (IVCF) placement in 1.1%, and active oral contraceptive use in 2.5%. When comparing risk categories, patients presenting with low-risk PE were younger in age, and Black patients presented more often with high-risk PE. SMD for age and race/ethnicity are reported in Supplemental Table S2. Notable differences in SMD were observed in PERT classifications for both age and race. For age, differences between high-low (SMD, 0.40) and intermediate-low (SMD, 0.34) were observed. Regarding race, notable differences were identified for American Indian and Alaska Native (high-low: SMD, −0.14), Hispanic or Latino (high-low: SMD, 0.11), other (intermediate-low: SMD, 0.13), and White (high-intermediate: SMD, 0.16; intermediate-low: SMD, 0.20).Table 1Patient demographics.Overall (N = 11,612)PERT risk assessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)Sociodemographics Age, y (n = 11,148)63.5 ± 15.964.9 ± 15.364.0 ± 15.858.6 ± 16.3 Male sex (n = 11,612)5932 (51.1)885 (52.0)4394 (51.3)653 (48.7) BMI, kg/m2 (n = 10,031)31.7 ± 8.330.2 ± 8.132.2 ± 8.430.6 ± 7.9 Race/ethnicity (n = 11,612) American Indian, Alaska Native99 (0.9)5 (0.3)71 (0.8)23 (1.7) Asian, Pacific Islander112 (1.0)22 (1.3)69 (0.8)21 (1.6) Black or African American2537 (21.8)425 (25.0)1801 (21.0)311 (23.2) Hispanic or Latino597 (5.1)104 (6.1)395 (4.6)98 (7.3) Other638 (5.5)119 (7.0)411 (4.8)108 (8.0) White7629 (65.7)1026 (60.3)5822 (67.9)781 (58.2)Comorbidities Hypertension (n = 11,351)8361 (73.7)1023 (61.6)6413 (76.2)925 (72.3) Active tobacco user (n = 11,607)1742 (15.0)234 (13.8)1264 (14.8)244 (18.2) Recent surgery (n = 11,605)670 (5.8)135 (7.9)438 (5.1)97 (7.2) Active malignancy (n = 11,605)1895 (16.3)322 (19.0)1316 (15.4)257 (19.2) Prior malignancy (n = 11,605)1123 (9.7)165 (9.7)852 (9.9)106 (7.9)VTE risk factors Prior surgery (n = 11,605)1930 (16.6)371 (21.8)1259 (14.7)300 (22.4) Pregnancy (n = 11,605)69 (0.6)8 (0.5)43 (0.5)18 (1.3) DVT (n = 11,605)3857 (33.2)442 (26.0)2951 (34.4)464 (34.7) IVC filter (n = 11,605)133 (1.1)22 (1.3)92 (1.1)19 (1.4)Home medications Anticoagulant (n = 11,605)1599 (13.8)301 (17.7)1069 (12.5)229 (17.1) Antiplatelet (n = 11,605)2731 (23.5)408 (24.0)2029 (23.7)294 (22.0) Statin (n = 11,605)3773 (32.5)582 (34.3)2805 (32.7)386 (28.8) Chemotherapy (n = 11,605)561 (4.8)79 (4.6)410 (4.8)72 (5.4) NSAIDs (n = 11,605)1342 (11.6)152 (8.9)1016 (11.9)174 (13.0) Testosterone (n = 11,605)47 (0.4)4 (0.2)36 (0.4)7 (0.5) HRT (n = 11,605)118 (1.0)14 (0.8)97 (1.1)7 (0.5) Oral contraceptives (n = 11,605)294 (2.5)22 (1.3)230 (2.7)42 (3.1)Values are n (%) or mean ± SD.BMI, body mass index; DVT, deep vein thrombosis; HRT, hormone replacement therapy; IVC, inferior vena cava; NSAID, nonsteroidal anti-inflammatory drug; PERT, pulmonary embolism response team; VTE, venous thromboembolism. Table 1Patient demographics.Overall (N = 11,612)PERT risk assessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)Sociodemographics Age, y (n = 11,148)63.5 ± 15.964.9 ± 15.364.0 ± 15.858.6 ± 16.3 Male sex (n = 11,612)5932 (51.1)885 (52.0)4394 (51.3)653 (48.7) BMI, kg/m2 (n = 10,031)31.7 ± 8.330.2 ± 8.132.2 ± 8.430.6 ± 7.9 Race/ethnicity (n = 11,612) American Indian, Alaska Native99 (0.9)5 (0.3)71 (0.8)23 (1.7) Asian, Pacific Islander112 (1.0)22 (1.3)69 (0.8)21 (1.6) Black or African American2537 (21.8)425 (25.0)1801 (21.0)311 (23.2) Hispanic or Latino597 (5.1)104 (6.1)395 (4.6)98 (7.3) Other638 (5.5)119 (7.0)411 (4.8)108 (8.0) White7629 (65.7)1026 (60.3)5822 (67.9)781 (58.2)Comorbidities Hypertension (n = 11,351)8361 (73.7)1023 (61.6)6413 (76.2)925 (72.3) Active tobacco user (n = 11,607)1742 (15.0)234 (13.8)1264 (14.8)244 (18.2) Recent surgery (n = 11,605)670 (5.8)135 (7.9)438 (5.1)97 (7.2) Active malignancy (n = 11,605)1895 (16.3)322 (19.0)1316 (15.4)257 (19.2) Prior malignancy (n = 11,605)1123 (9.7)165 (9.7)852 (9.9)106 (7.9)VTE risk factors Prior surgery (n = 11,605)1930 (16.6)371 (21.8)1259 (14.7)300 (22.4) Pregnancy (n = 11,605)69 (0.6)8 (0.5)43 (0.5)18 (1.3) DVT (n = 11,605)3857 (33.2)442 (26.0)2951 (34.4)464 (34.7) IVC filter (n = 11,605)133 (1.1)22 (1.3)92 (1.1)19 (1.4)Home medications Anticoagulant (n = 11,605)1599 (13.8)301 (17.7)1069 (12.5)229 (17.1) Antiplatelet (n = 11,605)2731 (23.5)408 (24.0)2029 (23.7)294 (22.0) Statin (n = 11,605)3773 (32.5)582 (34.3)2805 (32.7)386 (28.8) Chemotherapy (n = 11,605)561 (4.8)79 (4.6)410 (4.8)72 (5.4) NSAIDs (n = 11,605)1342 (11.6)152 (8.9)1016 (11.9)174 (13.0) Testosterone (n = 11,605)47 (0.4)4 (0.2)36 (0.4)7 (0.5) HRT (n = 11,605)118 (1.0)14 (0.8)97 (1.1)7 (0.5) Oral contraceptives (n = 11,605)294 (2.5)22 (1.3)230 (2.7)42 (3.1)Values are n (%) or mean ± SD.BMI, body mass index; DVT, deep vein thrombosis; HRT, hormone replacement therapy; IVC, inferior vena cava; NSAID, nonsteroidal anti-inflammatory drug; PERT, pulmonary embolism response team; VTE, venous thromboembolism. Patient demographics. Values are n (%) or mean ± SD. BMI, body mass index; DVT, deep vein thrombosis; HRT, hormone replacement therapy; IVC, inferior vena cava; NSAID, nonsteroidal anti-inflammatory drug; PERT, pulmonary embolism response team; VTE, venous thromboembolism. Presenting Vital Signs, Laboratory Values, and Imaging Results At presentation, the mean maximum heart rate was 107.2 ± 22.7 beats per minute, 57.2% required supplemental oxygen, and 7.6% required vasopressors (Table 2). For patients in whom BNP/NT-proBNP, troponin, and/or lactate were measured, elevated levels were found in 72.9%, 44.6%, and 47%, respectively. Most patients underwent PE-specific imaging (96.3%), which was predominantly CT (92.9%), followed by ventilation–perfusion scan (1.6%). Saddle PE was identified in 18% of patients, while the majority (50.1%) had main pulmonary artery involvement. CT angiography at the index admission demonstrated evidence of RV strain in 64.6%. Among those with echocardiograms, RV strain as noted in 43.4% and McConnell sign in 12.7%. Of all patients, 4.9% also had clots in transit. Patent foramen ovale was identified in only 1.7% of patients. Observed differences across PE risk categories overall represented their disease severity, with a greater degree of cardiac biomarker elevation, elevated lactate, saddle PE, right heart strain, and mechanical ventilation in those with high-risk PE.Table 2Presenting symptoms and diagnoses.Overall (N = 11,612)PERT risk assessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)Vital signs Maximum HR (n = 11,428)107.2 ± 22.7117.2 ± 26.4106.3 ± 21.6100.0 ± 20.0 Minimum SBP (n = 11,200)113.4 ± 23.885.3 ± 23.7118.1 ± 20.4118.4 ± 19.2 Maximum respiratory rate (n = 11,364)25.1 ± 7.828.5 ± 9.724.9 ± 7.322.5 ± 6.5 Minimum O2 saturation (n = 11,166)91.1 ± 7.788.1 ± 11.591.4 ± 6.993.4 ± 5.5 Supplemental O2 (n = 11,293)6458 (57.2)1365 (82.5)4655 (55.6)438 (34.4)Vasopressor (n = 11,534)874 (7.6)874 (51.8)0 (0.0)0 (0.0)Laboratory studies NT-proBNP elevated (n = 8157)5944 (72.9)886 (80.0)5058 (77.4)0 (0.0) Troponin elevated (n = 9849)4391 (44.6)671 (47.4)3720 (49.0)0 (0.0) Lactate elevated (n = 4030)1895 (47.0)632 (62.0)1263 (46.0)0 (0.0) D-dimer elevated (n = 3620)897 (24.8)88 (21.9)733 (25.7)76 (20.8)Diagnostic imaging PE imaging (n = 11,577)11,180 (96.3)1567 (92.1)8370 (97.7)1243 (92.6) CT PA (n = 11,571)10,749 (92.9)1441 (85.1)8121 (94.8)1188 (90.7) Catheter PA (n = 11,577)558 (4.8)98 (5.8)444 (5.2)16 (1.2) Echocardiogram (n = 11,427)8667 (75.9)1299 (77.9)6512 (76.7)866 (68.0) MR angiogram (n = 11,577)9 (0.1)2 (0.1)5 (0.1)2 (0.2) VQ scan (n = 11,570)190 (1.6)27 (1.6)135 (1.6)28 (2.1)CT findings Saddle (n = 11,577)2107 (18.2)311 (18.3)1704 (19.9)92 (7.0) Main (n = 11,577)5803 (50.1)830 (49.0)4600 (53.7)373 (28.4) Distal (n = 11,540)548 (4.7)113 (6.7)388 (4.5)47 (3.6) Right heart strain (n = 10,277)6639 (64.6)978 (69.2)5661 (72.6)0 (0.0)Transthoracic echocardiogram Clot in transit (n = 8450)416 (4.9)102 (8.1)294 (4.6)20 (2.3) Right heart strain (n = 11,577)5029 (43.4)891 (52.6)4138 (48.3)0 (0.0) RV hypokinesis (n = 11,577)1908 (16.5)418 (24.7)1490 (17.4)0 (0.0) Initial RVSP (n = 1515)44.9 ± 16.545.4 ± 16.045.4 ± 16.634.7 ± 13.3 TAPSE (n = 1992)4.3 ± 13.64.0 ± 9.24.5 ± 15.23.4 ± 6.8 Initial EF (n = 3891)57.4 ± 17.755.1 ± 15.657.5 ± 18.860.3 ± 11.4 PFO evidence (n = 8377)144 (1.7)18 (1.4)117 (1.9)9 (1.1) McConnell sign (n = 4086)520 (12.7)116 (17.2)400 (13.5)4 (0.9) Lower DVT (n = 7066)4570 (64.7)541 (60.8)3612 (67.5)417 (50.7) DVT above knee (n = 7066)3476 (49.2)417 (46.9)2752 (51.4)307 (37.3) DVT below knee (n = 7066)1102 (15.6)124 (13.9)866 (16.2)112 (13.6) ICU admission (n = 11,612)4928 (42.4)1302 (76.5)3396 (39.6)230 (17.1) Mechanical ventilation (n = 11,349)635 (5.6)420 (25.2)172 (2.0)43 (3.4)Values are n (%) or mean ± SD.CT, computed tomography; DVT, deep vein thrombosis; EF, ejection fraction; HR, heart rate; ICU, intensive care unit; MR, magnetic resonance; NT-proBNP, N-terminal pro–B-type natriuretic peptide; PA, pulmonary angiogram; PE, pulmonary embolism; PERT, pulmonary embolism response team; PFO, patent forman ovale; RA, right atrial; RV, right ventricular; RVSP, right ventricular systolic pressure; SBP, systolic blood pressure; TAPSE, tricuspid annular plane systolic excursion; VQ, ventilation–perfusion. Table 2Presenting symptoms and diagnoses.Overall (N = 11,612)PERT risk assessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)Vital signs Maximum HR (n = 11,428)107.2 ± 22.7117.2 ± 26.4106.3 ± 21.6100.0 ± 20.0 Minimum SBP (n = 11,200)113.4 ± 23.885.3 ± 23.7118.1 ± 20.4118.4 ± 19.2 Maximum respiratory rate (n = 11,364)25.1 ± 7.828.5 ± 9.724.9 ± 7.322.5 ± 6.5 Minimum O2 saturation (n = 11,166)91.1 ± 7.788.1 ± 11.591.4 ± 6.993.4 ± 5.5 Supplemental O2 (n = 11,293)6458 (57.2)1365 (82.5)4655 (55.6)438 (34.4)Vasopressor (n = 11,534)874 (7.6)874 (51.8)0 (0.0)0 (0.0)Laboratory studies NT-proBNP elevated (n = 8157)5944 (72.9)886 (80.0)5058 (77.4)0 (0.0) Troponin elevated (n = 9849)4391 (44.6)671 (47.4)3720 (49.0)0 (0.0) Lactate elevated (n = 4030)1895 (47.0)632 (62.0)1263 (46.0)0 (0.0) D-dimer elevated (n = 3620)897 (24.8)88 (21.9)733 (25.7)76 (20.8)Diagnostic imaging PE imaging (n = 11,577)11,180 (96.3)1567 (92.1)8370 (97.7)1243 (92.6) CT PA (n = 11,571)10,749 (92.9)1441 (85.1)8121 (94.8)1188 (90.7) Catheter PA (n = 11,577)558 (4.8)98 (5.8)444 (5.2)16 (1.2) Echocardiogram (n = 11,427)8667 (75.9)1299 (77.9)6512 (76.7)866 (68.0) MR angiogram (n = 11,577)9 (0.1)2 (0.1)5 (0.1)2 (0.2) VQ scan (n = 11,570)190 (1.6)27 (1.6)135 (1.6)28 (2.1)CT findings Saddle (n = 11,577)2107 (18.2)311 (18.3)1704 (19.9)92 (7.0) Main (n = 11,577)5803 (50.1)830 (49.0)4600 (53.7)373 (28.4) Distal (n = 11,540)548 (4.7)113 (6.7)388 (4.5)47 (3.6) Right heart strain (n = 10,277)6639 (64.6)978 (69.2)5661 (72.6)0 (0.0)Transthoracic echocardiogram Clot in transit (n = 8450)416 (4.9)102 (8.1)294 (4.6)20 (2.3) Right heart strain (n = 11,577)5029 (43.4)891 (52.6)4138 (48.3)0 (0.0) RV hypokinesis (n = 11,577)1908 (16.5)418 (24.7)1490 (17.4)0 (0.0) Initial RVSP (n = 1515)44.9 ± 16.545.4 ± 16.045.4 ± 16.634.7 ± 13.3 TAPSE (n = 1992)4.3 ± 13.64.0 ± 9.24.5 ± 15.23.4 ± 6.8 Initial EF (n = 3891)57.4 ± 17.755.1 ± 15.657.5 ± 18.860.3 ± 11.4 PFO evidence (n = 8377)144 (1.7)18 (1.4)117 (1.9)9 (1.1) McConnell sign (n = 4086)520 (12.7)116 (17.2)400 (13.5)4 (0.9) Lower DVT (n = 7066)4570 (64.7)541 (60.8)3612 (67.5)417 (50.7) DVT above knee (n = 7066)3476 (49.2)417 (46.9)2752 (51.4)307 (37.3) DVT below knee (n = 7066)1102 (15.6)124 (13.9)866 (16.2)112 (13.6) ICU admission (n = 11,612)4928 (42.4)1302 (76.5)3396 (39.6)230 (17.1) Mechanical ventilation (n = 11,349)635 (5.6)420 (25.2)172 (2.0)43 (3.4)Values are n (%) or mean ± SD.CT, computed tomography; DVT, deep vein thrombosis; EF, ejection fraction; HR, heart rate; ICU, intensive care unit; MR, magnetic resonance; NT-proBNP, N-terminal pro–B-type natriuretic peptide; PA, pulmonary angiogram; PE, pulmonary embolism; PERT, pulmonary embolism response team; PFO, patent forman ovale; RA, right atrial; RV, right ventricular; RVSP, right ventricular systolic pressure; SBP, systolic blood pressure; TAPSE, tricuspid annular plane systolic excursion; VQ, ventilation–perfusion. Presenting symptoms and diagnoses. Values are n (%) or mean ± SD. CT, computed tomography; DVT, deep vein thrombosis; EF, ejection fraction; HR, heart rate; ICU, intensive care unit; MR, magnetic resonance; NT-proBNP, N-terminal pro–B-type natriuretic peptide; PA, pulmonary angiogram; PE, pulmonary embolism; PERT, pulmonary embolism response team; PFO, patent forman ovale; RA, right atrial; RV, right ventricular; RVSP, right ventricular systolic pressure; SBP, systolic blood pressure; TAPSE, tricuspid annular plane systolic excursion; VQ, ventilation–perfusion. PE Treatment Approaches The proportion of patients who received an intervention for PE are detailed in Table 3. Systemic thrombolysis was used in 4.4% of patients, among most of whom were in the high-risk PE group. Of the patients with high-risk PE, 13.9% received systemic thrombolysis. Catheter-based intervention was performed in 22.5% of patients, with the highest proportion of use in patients with intermediate-risk PE (25.4%) and patients with high-risk PE (22.2%). Surgical embolectomy was performed in 1.8% of patients with high-risk PE, and ECMO support was used in 7.4% of patients. Overall, 5.3% of patients underwent IVCF placement, with the greatest use (6.0%) among patients with high-risk PE.Table 3Therapeutic interventions.Overall (N = 11,612)PERT risk assessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)Systemic IV thrombolysis516 (4.4)237 (13.9)270 (3.2)9 (0.7) tPA505 (97.9)230 (97.0)266 (98.5)9 (100.0) Tenecteplase9 (1.7)6 (2.5)3 (1.1)0 (0.0) Other2 (0.4)1 (0.4)1 (0.4)0 (0.0)Catheter-based intervention2614 (22.5)377 (22.2)2175 (25.4)62 (4.6) Ultrasound-facilitated CDT888 (34.0)121 (32.1)748 (34.4)19 (30.6) Nonultrasound CDT125 (4.8)19 (5.0)102 (4.7)4 (6.5) AngioVac35 (1.3)13 (3.4)22 (1.0)0 (0.0) Mechanical thrombectomy1502 (57.5)214 (56.8)1252 (57.6)36 (58.1) Other catheter68 (0.6)11 (0.6)54 (0.6)3 (4.8) Multiple catheter4 (0.0)1 (0.1)3 (0.0)0 (0.0)Surgical embolectomy131 (1.1)31 (1.8)97 (1.1)3 (0.2)ECMO181 (1.6)125 (7.4)52 (0.6)4 (0.3)IVC filter placement612 (5.3)102 (6.0)449 (5.2)61 (4.5)Values are n (%).CDT, catheter-directed thrombolysis; ECMO, extracorporeal membrane oxygenation; IVC, inferior vena cava; PERT, pulmonary embolism response team; tPA, tissue plasminogen activator. Table 3Therapeutic interventions.Overall (N = 11,612)PERT risk assessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)Systemic IV thrombolysis516 (4.4)237 (13.9)270 (3.2)9 (0.7) tPA505 (97.9)230 (97.0)266 (98.5)9 (100.0) Tenecteplase9 (1.7)6 (2.5)3 (1.1)0 (0.0) Other2 (0.4)1 (0.4)1 (0.4)0 (0.0)Catheter-based intervention2614 (22.5)377 (22.2)2175 (25.4)62 (4.6) Ultrasound-facilitated CDT888 (34.0)121 (32.1)748 (34.4)19 (30.6) Nonultrasound CDT125 (4.8)19 (5.0)102 (4.7)4 (6.5) AngioVac35 (1.3)13 (3.4)22 (1.0)0 (0.0) Mechanical thrombectomy1502 (57.5)214 (56.8)1252 (57.6)36 (58.1) Other catheter68 (0.6)11 (0.6)54 (0.6)3 (4.8) Multiple catheter4 (0.0)1 (0.1)3 (0.0)0 (0.0)Surgical embolectomy131 (1.1)31 (1.8)97 (1.1)3 (0.2)ECMO181 (1.6)125 (7.4)52 (0.6)4 (0.3)IVC filter placement612 (5.3)102 (6.0)449 (5.2)61 (4.5)Values are n (%).CDT, catheter-directed thrombolysis; ECMO, extracorporeal membrane oxygenation; IVC, inferior vena cava; PERT, pulmonary embolism response team; tPA, tissue plasminogen activator. Therapeutic interventions. Values are n (%). CDT, catheter-directed thrombolysis; ECMO, extracorporeal membrane oxygenation; IVC, inferior vena cava; PERT, pulmonary embolism response team; tPA, tissue plasminogen activator. Among patients treated with a catheter-based intervention, mechanical thrombectomy was the most frequently used device (57.5% of all catheter-based interventions), followed by ultrasound-facilitated catheter-directed thrombolysis (34.0% of all catheter-based interventions). The distribution of use of both therapeutic interventions was similar among both patients with high-risk and those with intermediate-risk PE. In-Hospital and Longitudinal Outcomes Outcomes for patients with PE are detailed in Table 4. Readmission rates were well distributed across all PE risk categories, with 10.1% at 30 days and 26.8% at 1 year. Most patients (98.7%) survived the first 24 hours. In-hospital survival was lowest among patients with high-risk PE (94.2%) and nearly equivalent between patients with intermediate- and low-risk PE (99.5% and 99.7%, respectively). Overall survival dropped to 91.2% at 30 days and 85.9% at 1 year, primarily due to increased mortality in the high-risk PE group (25.7% at 30 days and 32.2% at 1 year). The overall rate of in-hospital major bleeding in the first 30 days was 4.2% and nonmajor bleeding was 4.3%. On average, patients spent a median of 4 days (IQR, 2-7 days) in the hospital, and patients with high-risk PE had the longest lengths of stay (median, 6 days; IQR, 3-13 days).Table 4In-hospital and longitudinal outcomes.Overall (N = 11,612)PERT Risk AssessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)LOS, d (n = 10,581)4 (2-7)6 (3-13)4 (2-7)2 (3-6)24 h Survival (n = 11,612)11,465 (98.7)1602 (94.2)8525 (99.5)1338 (99.7) Major bleed (n = 11,516)146 (1.3)69 (4.1)64 (0.8)13 (1.0) Nonmajor bleed (n = 11,516)276 (2.4)105 (6.2)143 (1.7)28 (2.2)30 d Survival (n = 11,612)10,587 (91.2)1264 (74.3)8034 (93.8)1289 (96.1) Major bleed (n = 10,132)423 (4.2)142 (9.6)237 (3.1)44 (4.0) Nonmajor bleed (n = 9926)431 (4.3)118 (8.2)261 (3.5)52 (4.8) Readmission (n = 11,612)1178 (10.1)147 (8.6)854 (10.0)177 (13.2)1 y Survival (n = 11,612)9971 (85.9)1153 (67.8)7599 (88.7)1219 (90.8) Major bleed (n = 6382)494 (7.7)152 (15.9)286 (6.1)56 (7.5) Nonmajor bleed (n = 6298)599 (9.5)140 (15.1)387 (8.3)72 (9.8) Readmission (n = 11,612)3115 (26.8)419 (24.6)2293 (26.8)403 (30.0)Values are median (IQR) or n (%).LOS, length of hospital stay; PERT, pulmonary embolism response team. Table 4In-hospital and longitudinal outcomes.Overall (N = 11,612)PERT Risk AssessmentHigh (n = 1701)Intermediate (n = 8569)Low (n = 1342)LOS, d (n = 10,581)4 (2-7)6 (3-13)4 (2-7)2 (3-6)24 h Survival (n = 11,612)11,465 (98.7)1602 (94.2)8525 (99.5)1338 (99.7) Major bleed (n = 11,516)146 (1.3)69 (4.1)64 (0.8)13 (1.0) Nonmajor bleed (n = 11,516)276 (2.4)105 (6.2)143 (1.7)28 (2.2)30 d Survival (n = 11,612)10,587 (91.2)1264 (74.3)8034 (93.8)1289 (96.1) Major bleed (n = 10,132)423 (4.2)142 (9.6)237 (3.1)44 (4.0) Nonmajor bleed (n = 9926)431 (4.3)118 (8.2)261 (3.5)52 (4.8) Readmission (n = 11,612)1178 (10.1)147 (8.6)854 (10.0)177 (13.2)1 y Survival (n = 11,612)9971 (85.9)1153 (67.8)7599 (88.7)1219 (90.8) Major bleed (n = 6382)494 (7.7)152 (15.9)286 (6.1)56 (7.5) Nonmajor bleed (n = 6298)599 (9.5)140 (15.1)387 (8.3)72 (9.8) Readmission (n = 11,612)3115 (26.8)419 (24.6)2293 (26.8)403 (30.0)Values are median (IQR) or n (%).LOS, length of hospital stay; PERT, pulmonary embolism response team. In-hospital and longitudinal outcomes. Values are median (IQR) or n (%). LOS, length of hospital stay; PERT, pulmonary embolism response team. Distributions of procedural outcomes, without controlling for risk stratification, are detailed in Table 5 and Supplemental Tables S2-S4. Median hospital length of stay was observed to be the shortest for the group undergoing catheter-based intervention (median, 4 days; IQR, 2-7 days) and observed to be the longest for the patients who were placed on ECMO (median, 12 days; IQR, 5-20 days). In the first 24 hours, major bleeding rates were observed to be the highest for patients who had surgical embolectomy (8.4%) and observed to be the lowest for patients who underwent catheter-based intervention (1.1%). The 30-day survival rate was observed to be highest in the patients who underwent either catheter-based intervention (95.1%) or surgical embolectomy (93.9%) and observed to be lowest in patients requiring ECMO as a standalone therapy (68.0%); 1-year survival followed similar trends.Table 5Outcomes by advanced therapy type.Systemic IV thrombolysis (n = 516)Catheter intervention (n = 2614)Surgical embolectomy (n = 131)ECMO (n = 181)IVC filter (n = 612)LOS, d (n = 10,581)5 (3-10)4 (2-7)8 (3-17)12 (5-20)7 (3-14)24 h Survival (n = 11,612)478 (92.6)2587 (99.0)129 (98.5)173 (95.6)608 (99.3) Major bleed (n = 11,516)17 (3.3)30 (1.1)11 (8.4)16 (8.8)33 (5.4) Nonmajor bleed (n = 11,516)13 (2.5)39 (1.5)3 (2.3)6 (3.3)33 (5.4)30 d Survival (n = 11,612)408 (79.1)2485 (95.1)123 (93.9)123 (68.0)563 (92.0) Major bleed (n = 10,132)36 (7.8)74 (3.2)15 (12.2)35 (21.9)59 (10.6) Nonmajor bleed (n = 9926)28 (6.2)101 (4.4)5 (4.2)26 (17.0)62 (11.4) Readmission (n = 11,612)30 (5.8)197 (7.5)6 (4.6)4 (2.2)65 (10.6)1 y Survival (n = 11,612)395 (76.6)2412 (92.3)120 (91.6)118 (65.2)520 (85.0) Major bleed (n = 6382)38 (13.3)88 (6.3)15 (18.3)35 (37.6)65 (19.2) Nonmajor bleed (n = 6298)31 (11.0)141 (10.0)6 (7.4)26 (29.9)72 (21.3) Readmission (n = 11,612)103 (20.0)639 (24.4)41 (31.3)25 (13.8)205 (33.5)Values are median (IQR) or n (%).ECMO, extracorporeal membrane oxygenation; IV, intravenous; IVC, inferior vena cava; LOS, length of hospital stay. Table 5Outcomes by advanced therapy type.Systemic IV thrombolysis (n = 516)Catheter intervention (n = 2614)Surgical embolectomy (n = 131)ECMO (n = 181)IVC filter (n = 612)LOS, d (n = 10,581)5 (3-10)4 (2-7)8 (3-17)12 (5-20)7 (3-14)24 h Survival (n = 11,612)478 (92.6)2587 (99.0)129 (98.5)173 (95.6)608 (99.3) Major bleed (n = 11,516)17 (3.3)30 (1.1)11 (8.4)16 (8.8)33 (5.4) Nonmajor bleed (n = 11,516)13 (2.5)39 (1.5)3 (2.3)6 (3.3)33 (5.4)30 d Survival (n = 11,612)408 (79.1)2485 (95.1)123 (93.9)123 (68.0)563 (92.0) Major bleed (n = 10,132)36 (7.8)74 (3.2)15 (12.2)35 (21.9)59 (10.6) Nonmajor bleed (n = 9926)28 (6.2)101 (4.4)5 (4.2)26 (17.0)62 (11.4) Readmission (n = 11,612)30 (5.8)197 (7.5)6 (4.6)4 (2.2)65 (10.6)1 y Survival (n = 11,612)395 (76.6)2412 (92.3)120 (91.6)118 (65.2)520 (85.0) Major bleed (n = 6382)38 (13.3)88 (6.3)15 (18.3)35 (37.6)65 (19.2) Nonmajor bleed (n = 6298)31 (11.0)141 (10.0)6 (7.4)26 (29.9)72 (21.3) Readmission (n = 11,612)103 (20.0)639 (24.4)41 (31.3)25 (13.8)205 (33.5)Values are median (IQR) or n (%).ECMO, extracorporeal membrane oxygenation; IV, intravenous; IVC, inferior vena cava; LOS, length of hospital stay. Outcomes by advanced therapy type. Values are median (IQR) or n (%). ECMO, extracorporeal membrane oxygenation; IV, intravenous; IVC, inferior vena cava; LOS, length of hospital stay. Discussion Pulmonary embolism remains a significant public health concern in the United States and worldwide, and recent data demonstrate that the population-based annual incidence of PE is as high as 1 in 1000 patients.9,10 Efforts to prevent, recognize, and treat PE are increasing worldwide as evidenced by multispecialty groups at leading hospitals devoted to the treatment of patients with PE.11, 12, 13 Moreover, many interested in PE research have formed nascent organizations both in the United States (the National PERT Consortium) and in Europe (EXPERT-PE). As a result of the coalescence around the study of PE by multiple groups, prospectively collected data have been gathered, allowing for the large-scale analysis of patients with PE to further collective knowledge of this disease process. The National PERT Consortium database, the subject of this analysis, is the largest PE-specific series to date in the medical literature. The current report reflects the totality of data from the database as of June 1, 2024. Given the broad representation and number of patients included, there are some important conclusions that can be drawn from these data. Across all risk strata, patients who experienced PE in the National PERT Consortium database were obese with a mean body mass index for the entire cohort of 31.7 kg/m2. This fact is in line with obesity being a known risk factor for VTE.14,15 Moreover, Black patients are overrepresented in the National PERT Consortium database, which further reinforces the known higher incidence of VTE in this population.16 Other known risk factors for VTE were expectedly prevalent among patients within the database, including recent surgery, cancer, and a history of DVT. It is notable that one-quarter of patients in the National PERT Consortium database had a history of cancer, and this likely reflects the enriched nature of the dataset. Pulmonary embolism–specific risk assessment was available for the vast majority of patients. Consistent with reported findings nationwide, the majority of patients fell into either the high-risk or the intermediate-risk category. The National PERT Consortium database undoubtedly represents an enriched PE population as many low-risk patients with PE are never seen by multidisciplinary teams or were not admitted to the hospital. Diagnostic imaging varied according to center, but CT was used in >90% of patients and echocardiography in >70%. Hospital resource utilization increased as anticipated in accordance with increasing levels of risk, as >75% of patients with a high-risk PE went to the ICU. PE treatments also varied by risk strata. Nearly one-quarter of patients within the database underwent catheter-based therapy, while fewer underwent systemic thrombolysis or surgical embolectomy or required ECMO support. Surprisingly, <2% of patients were discovered to have a patent forman ovale, and this likely represents an underreported phenomenon. All these data contribute to an enriched population of patients with PE, which comprise the National PERT Consortium database. The National PERT Consortium database records the vital outcomes of patients with PE out to 1 year, and these data, in conjunction with the shorter-term outcomes, are of great interest. Patients who underwent surgical embolectomy and survived to discharge had the relative highest observed 30-day and 1-year survival compared with those treated with other treatment modalities. While these findings are descriptive and should not be interpreted as evidence of comparative effectiveness, these data mirror single-center studies that have been previously published.17, 18, 19 They also represent a significant selection bias in which patients are chosen to undergo surgery based on risk severity, as well as other unmeasured factors. Despite this, patients who were appropriate for surgery had evidence of significant right heart dysfunction, with a decreased tricuspid annular plane systolic excursion and increased RV systolic pressure, compared with patients who underwent catheter-based intervention, suggesting a continued role for surgical embolectomy in select patients (Supplemental Table S3). Catheter-based interventions also yielded excellent survival data in both short term and midterm. However, short-term survival among patients with PE who underwent catheter-based therapy was slightly worse than published trials.20, 21, 22 The data contained within the National PERT Consortium database may yield a more real-world observation of broad, unselected patients with PE being treated with catheter-based therapies outside the setting of a clinical trial. When survival was examined based on risk stratification, it was obvious that patients with high-risk PE had decreased short term and midterm survival. Importantly, patients with intermediate-risk PE, when compared to low-risk patients, had similar 24-hour survival but decreased 30-day and 1-year survival, thus reflecting the impact the burden of initial disease on midterm survival. This study is not without limitations. First, this study was limited to patients within the National PERT Consortium database, and broader applicability of these data may be limited, particularly at sites that do not have PERTs. As data were volunteered from each participating PERT institution, the data should be considered a convenience sample and may be subject to unmeasurable selection biases. Further, as these patients are not randomized in any fashion, the chosen treatment strategy certainly could have influenced outcomes. Additionally, patient entry into the database is likely nonsequential and represents a selection bias based on institutional practice. While the database comes with instructions and a data dictionary, imputation errors may still be possible. Moreover, there was unquestionably a collection bias toward sicker patients requiring hospitalization and management by PERTs rather than traditional care pathways and providers. Thus, the true outcomes of patients with lower-risk PE may not be well represented by these data. Other treatment options for PE and VTE are also underrepresented in the National PERT Consortium database, including patients with placement of IVCFs, patients undergoing surgical embolectomy, and patients placed on ECMO. As such, it may be difficult to truly detail outcomes for patients who received these underrepresented treatment strategies. Patients were seen and entered at the discretion of the referral patterns of local centers. Hypertensive status reflects an operational definition and approximation of hypertension owing to available data and may misclassify some patients. Treatment algorithms likely differ by center, and these treatment biases are difficult to account for in a dataset of this size. Risk categorization was inferred based on available data and may not accurately reflect PERT risk categorization. Although sites were requested to include all patients and trained on data entry, this dataset may not be exhaustive of each centers’ full PE population, and no additional adjudication of entered data was performed. Specifically, this analysis may be subject to biases that affect the accuracy of event rates, particularly for events that did not occur. This analysis is descriptive and cross-sectional in nature. Accordingly, the findings should be considered hypothesis generating, susceptible to confounding by indication, and results should not be interpreted as evidence of causal relationships or treatment effects. In summary, data from the National PERT Consortium registry, the largest real-world database of patients with PE accumulated to date, provide critical insights into the management and outcomes across a broad swath of the population in the United States. Importantly, this substantial initiative has provided the foundation for a standardized approach to PE research and data collection.23 The data contained within this initial report will serve as a benchmark for patients with PE, outcomes, and research going forward. CRediT authorship contribution statement William Brent Keeling: Writing – original draft. Steven Horbal: Formal analysis, Writing – original draft. Kenneth Rosenfield: Conceptualization, Writing – original draft. Robert Lookstein: Writing – original draft. Richard Channick: Writing – original draft. Rachel P. Rosovsky: Writing – original draft. John M. Moriarty: Writing – original draft. Matthew D. Langston: Writing – original draft. Seth Sokol: Writing – original draft. Thomas M. Todoran: Writing – original draft. Eric A. Secemsky: Writing – original draft.