Prospective Experience of Pulmonary Embolism Management and Outcomes

Abstract: Objective. We sought to evaluate the impact of pulmonary embolism (PE) response teams (PERTs) on all consecutive patients with PE. Background. Multidisciplinary PERTs have been promoted for the management and treatment of (PE); however, the impact of PERTs on clinical outcomes has not been prospectively evaluated. Methods. We prospectively studied 220 patients with computed tomography (CT)-confirmed PE between January, 2019 and August, 2019. Baseline characteristics, as well as medical, interventional, and operational care, were captured. The total population was divided into 2 groups, ie, those with PERT activation and those without PERT activation. PERT activation was left at the discretion of the primary team. Our primary outcome was 90-day composite endpoint (rate of readmission, major bleeds, and mortality). Using 2:1 propensity-matched and multivariable-adjusted Cox proportional hazard analyses, we examined the impact of PERT activation on primary outcome, treatment approach, and length of stay. Results. Of the total 220 patients, PERT was activated in 47 (21.4%). The PERT cohort, as compared with the non-PERT cohort, was more likely to present with dyspnea, syncope, lower systolic blood pressure, higher heart rate, higher respiratory rate, lower oxygen saturation, higher troponin levels, and higher right ventricular to left ventricular ratio. PERT activation was associated with increased use of advanced therapies (36.2% vs 1.2%; P<.001) and catheter-directed inventions (25.5% vs 0.6%; P<.001). In multivariable-adjusted analysis of propensity-matched cohorts, PERT activation was associated with lower 90-day outcomes (hazard ratio, 0.40; 95% confidence interval, 0.21-0.75; P<.01). Conclusion. At our institution, PERT had a clinically significant impact on therapeutic strategies and 90-day outcomes in patients with PE. 

J INVASIVE CARDIOL 2021;33(3):E173-E180. Epub 2021 February 11. 

Key words: advanced therapies, outcomes, PE, PERT, pulmonary embolism, pulmonary embolism response team

Pulmonary embolism (PE) is the third leading cause of cardiovascular death in the United States, with up to 100,000 deaths annually.^1-3 The majority of these deaths are caused by intermediate-high risk PEs, with reported mortality rates as low as 5% to as high as 65%, depending on severity.^4,5 

Over the past 15 years, time trend analyses of diverse populations show decreasing rates of case fatalities from PE despite increasing incidence rates.⁶ The reason for improved outcomes is unclear. During this time, multidisciplinary pulmonary embolism response teams (PERTs) and effective advanced therapies were developed. PERTs, which are composed of a spectrum of specialists, provide the potential to optimize treatment strategies and identify high-intermediate risk PEs that may benefit from advanced therapies — interventions the literature suggests are underutilized.⁵ However, to our knowledge, the impact of PERT on clinical outcomes has not yet been prospectively evaluated. Therefore, we conducted a prospective study to evaluate the clinical outcome of PERT vs non-PERT activation among consecutive all comers with PE. 

Methods

Study population. This was an institutional review-board approved prospective cohort study of all consecutive patients with computed tomography (CT)-confirmed PE admitted to our institution, which is a tertiary-care center with capabilities for advanced therapies, between January 2019 and August 2019. Informed consent was waived.

Multidisciplinary PERT was implemented at our institution in mid-2018. PERT activation was left at the discretion of the treating physician using the institutional PERT pager system and was not mandated. Our PERT is composed of the activating physician, a critical care intensivist, a vascular medicine specialist, an interventional cardiologist, and a cardiothoracic surgeon. Upon activation, the team convenes in real time via telephone and a treatment plan is created. Patients were later stratified by PERT activation into 2 groups, ie, PERT and non-PERT groups. We also categorized severity of PE by the severity index (PESI), a validated marker of PE severity.^7,8

Data collection and endpoints. Every CT finding of acute or subacute PE was automatically flagged and triggered a confidential data collection through an independent research nurse, irrespective of PERT activation. Information was collected using electronic medical record and stored in Health Insurance Portability and Accountability Act (HIPAA)-compliant database (RedCaps). The treating physician was blinded to the data-collection process. Recorded data included presenting symptoms, initial vitals, demographics, medical comorbidities, PE risk factors, biomarkers, imaging data (right ventricular to left ventricular ratio [RV:LV]), and treatment plans (medical and/or interventional). 

Outcomes of interest included inpatient mortality, 30-day and 90-day mortality, readmission, major bleeding, length of intensive care unit (ICU) stay, and total hospital length of stay (LOS). Our primary outcome, a 90-day composite endpoint, was a compilation of the following outcome parameters at 90 days: rate of readmission; rate of major bleeding; and rate of all-cause mortality. All-cause mortality rates were derived from clinical notes and public mortality records from the state of Ohio (Ohio Vital Statistics). Readmissions to our institution and any of the 18 affiliated hospitals within 30 and 90 days from discharge were captured through electronic medical records. Major bleeding was defined according to the International Society of Thrombosis and Hemostasis (ISTH) criteria, ie, intracranial hemorrhage (ICH), fatal bleed, bleed with > 2 g/dL drop in hemoglobin levels (within 48 hours), or requirement of >2 unit blood transfusion.⁹ Advanced therapies included systemic thrombolysis, surgical intervention, and catheter-directed interventions (CDI), ie, catheter-directed thrombolysis using the EKOSonic endovascular system (EKOS Corporation) and catheter-directed thrombectomy using the Flow-Triever system (Inari Medical). 

Statistical analysis. Descriptive statistics were used to examine baseline characteristics and analyze outcomes. Continuous variables, presented as mean ± standard deviation, were compared using t-test (for parametric variables) or Wilcoxon rank-sum test (for non-parametric variables). Categorical variables, presented as counts and percentages, were compared using the Chi-square test unless the frequency was <5, in which case the Fisher’s exact test was utilized. Bar and Sunburst charts were used to display various relation patterns between different data elements.  

Applying the greedy matching algorithm in a 1:2 fashion, propensity-score matching was executed for all baseline characteristics, and outcomes were reanalyzed using the above descriptive statistics. We also used Kaplan-Meier curves and the log-rank comparison test to perform survival analysis of the primary outcomes in the propensity-matched cohort stratified by PERT activation. Lastly, we utilized a multivariable Cox proportional hazard regression model, adjusting for variables that remained significantly different between the propensity-matched cohort, to examine the protective value of PERT activation on 90-day outcomes. 

A P-value of <.05 was considered statistically significant. The R Statistical software, version 3.6 (R Foundation for Statistical Computing) and International Business Machines Statistical Package for the Social Sciences (IBM SPSS) software were used for statistical analyses.

Results

Baseline characteristics. The average age of our patient population was 59.7 years old and 55% were female. Dyspnea was the most common presenting symptom among both groups, and syncope was more frequently observed in the PERT group vs the non-PERT group (Table 1). The PERT cohort appeared more acutely ill, with lower systolic blood pressure, higher heart rate, higher respiratory rate, and lower oxygen saturation on initial presentation (Table 1). In concordance, those patients had higher initial troponins levels and higher RV:LV ratios. Medical comorbidities, such as hypertension, diabetes, smoking, and history of gastrointestinal bleed, were similar in both groups. However, the non-PERT group was more likely to have either active cancer or recent surgery, while family history of venous thromboembolism was more common in the PERT group (Table 1).

Utilizing propensity matching in a 1:2 fashion, we were able to match 47 PERT patients to 94 non-PERT patients. This was successful in eliminating significant differences in baseline characteristics between the 2 groups; however, differences in family history of venous thromboembolism, syncope on presentation, vitals, troponin, and RV:LV ratio remained present (Table 2).

PE severity, PERT activation, management strategies, and follow-up. Of the 220 patients included, only 47 patients (21.4%) had PERT activation. Although roughly two-thirds (146; 66.4%) of all patients had intermediate-high risk PE (PESI >III), only 24.7% of those patients had PERT activation.

Anticoagulation monotherapy alone was the most common therapy (80.9%) and was more frequently used in the non-PERT group (88.4% vs 53.2% in the PERT group; P<.001). The discrepancy in anticoagulation monotherapy use between the 2 groups was greater among those with higher PESI scores. For example, for those with very-high risk PE (PESI ≥V), anticoagulation monotherapy was used in 89.1% of the non-PERT group, while only 31.6% received anticoagulation alone in the PERT group (Figures 1 and 2).

Use of advanced therapies, despite being relatively rare (8.6%), was strongly associated with PERT activation (36.2% vs 1.2% in the non-PERT group; P<.001). Similarly, frequency of CDI was significantly higher in the PERT group (25.5% vs 0.6% in the non-PERT group; P<.001) (Figure 3). Both non-PERT patients who received an advanced therapy (1 EKOS, 1 systemic  tissue plasminogen activator) had intermediate-high risk PE (PESI ≥III). On the other hand, advanced therapy use in the PERT group was not necessarily in parallel with PESI score, as 4 patients (36.4%) with PESI ≤II received an advanced therapy (Figures 1 and 2). There were no intraprocedural complications using advanced therapies. Lastly, among the patients who were alive on discharge, the average number of follow-ups was higher in the PERT group vs the non-PERT group (2.3 vs. 1.4, respectively; P<.01).

Outcomes with and without propensity matching. In our study, lengths of ICU and hospital stays did not differ between the 2 groups. There were also no differences between in-hospital mortality rates, 30-day readmission rates, and 30-day mortality rates between the 2 groups. However, there was a trend toward lower 30-day composite endpoint (27.7% vs 41.6%; P=.08) in the PERT group vs the non-PERT group, respectively (Tables 3 and 4). 

More importantly, at 90 days, the PERT group had significantly lower rates of readmission (19.1% vs 37.6% in the non-PERT group; P=.02) and composite endpoint (36.2% vs 60.1% in the non-PERT group; P<.01). Furthermore, the PERT group had lower rates of 90-day major bleeds (12.8% vs 26.0% in the non-PERT group; P=.06) and 90-day mortality (10.6% vs 23.7% in the non-PERT group; P=.05), however, these did not reach statistical significance (Table 5 and Figure 3). After propensity matching, the 90-day composite endpoint remained statistically significant (36.2% in the PERT group vs 59.6% in the non-PERT group; P<.01) (Table 5). 

After propensity matching and using Kaplan-Meier curves, our study showed PERT activation was associated with a significantly higher proportion of primary-outcome free patients by 90 days (log-rank P=.01) (Figure 3). In multivariable-adjusted analysis of the propensity-matched cohorts, PERT demonstrated a protective value for both major bleeds (adjusted hazard ratio [HR], 0.34; 95% confidence interval [CI], 0.12-0.99; P=.048) and composite endpoint (adjusted HR, 0.40; 95% CI, 0.21-0.75; P<.01), with a trend toward lower mortality (adjusted HR, 0.34; 95% CI, 0.11-1.01; P=.05) at 90 days (Table 6).

Discussion

In this study of 220 prospectively enrolled consecutive patients with CT-confirmed PE in the era of PERT, we found that the treating physician did not take advantage of the multidisciplinary PERT in the majority (~79%) of the study subjects. The non-PERT group was more likely to have active cancer and recent surgery, while those with PERT activation had more abnormal vitals and a larger degree of RV dysfunction. More importantly, after matching and adjusting for these baseline differences, PERT activation was protective against our primary outcome (90-day composite endpoint of readmission, major bleed, and mortality). This finding was primarily driven by a significant reduction in major bleed and a trend toward lower all-cause mortality at 90 days. Lastly, PERT activation was associated with higher use of advance therapies. 

Rosovsky et al were pioneers in exploring PERT outcomes; they studied 440 carefully selected emergency department patients and found no significant increase in major bleed or mortality post PERT.¹⁰ Similarly, Xenos et al, in a retrospective analysis of 77 PERT patients vs 992 non-PERT patients, showed that PERT activation did not influence mortality or readmission rate, yet it did result in shorter ICU stay and total LOS.¹¹ Chaudhury et al were the first to show decreased mortality from pre-PERT to post-PERT eras; however, only 13% of post-PERT era patients had PERT involved in their care.¹² In contrast to historical controls used in the above studies, our study is unique, as we prospectively enrolled all comers with PE and examined the impact of PERT vs no PERT activation on longer-term outcomes. 

Our PERT group had significantly less major bleeding despite the fact that 23% of these patients had systemic or catheter-directed thrombolysis compared with only 1% use in the non-PERT group. It is well known that risk of bleed in patients with venous thromboembolism depends on the choice of anticoagulation drug,^10,11 intensity of anticoagulation,^10,12 and patient characteristics and risk factors.^13,14 Intuitively, a multidisciplinary team, including a vascular medicine specialist, via PERT activation would likely allow thorough evaluation and offer the most balanced therapy, with appropriate monitoring and follow-up. This is even more meaningful, with a greater opportunity to decrease the risk of bleed, in patients who are concomitantly taking antiplatelets therapy.¹⁵ However, these hypotheses should be validated and confirmed in future studies.

Our extended follow-up has allowed us to appreciate the trend toward reduced mortality with PERT activation despite the higher acuity of these patients. While the exact mechanisms remain undetermined, the reduction in major bleed is likely a contributing factor. In addition (although not examined in the current study), Chaudhury et al showed that PERT activation was associated with more rapid inception of anticoagulation therapy;¹⁶ it is well established that the latter confers lower mortality in acute PE.¹⁷ More importantly, the utilization of advanced therapies (36.2% vs 1.2%) especially catheter-based ones (25.5% vs 0.6%) in the PERT group was higher than in the non-PERT group, in concordance with previous reports.^16,18 There have been no adequately powered randomized trials assessing the impact of these catheter-directed therapies on mortality in patients with PE and there is no clear consensus on their use.¹⁹ Hence, multidisciplinary PERT activation has a critical role in precisely selecting potential beneficiaries on a case-by-case basis. 

Cancer is a major risk factor for PE. Indeed, active cancer was the most common risk factor and the second-most common comorbidity after hypertension in our cohort. Despite this, in our cohort, a majority of patients with active cancer did not get PERT activation. Furthermore, 38% of cancer patients in the PERT group were treated with advanced therapies, while none of the patients with cancer in the non-PERT cohort received these interventions. Collectively, activation of PERT had a significant impact on the clinical course of patients with cancer. The intention of advanced therapies is not always to reverse hemodynamic deterioration and prevent death; rather, the purpose is often to improve quality of life and speed recovery, which is the goal of many PE patients, including those with cancer.

Study limitations. There are some limitations to our study. First, it is a single-center study, which limits external generalizability. Second, there were significantly more patients in the non-PERT cohort with active cancer, which could have biased our results; however, we performed propensity and adjusted analyses. Third, we acknowledge the potential for missed endpoints, as patients might have been readmitted to hospitals outside of our healthcare system. Last, although our study had the advantage of blinding the treating physicians and relying on real-time physician decision making, the ideal study design to investigate the impact of PERT on the management and outcomes of patients with PE is randomization, which was not performed.

Conclusion

Multidisciplinary PERT activation was associated with a significant increase in advanced therapies and a reduction in 90-day outcomes. Despite much sicker patients at presentation and use of more advanced therapies, PERT activation was not associated with higher major bleeding or longer ICU stay/hospital LOS.

References

1. Turetz M, Sideris AT, Friedman OA, Triphathi N, Horowitz JM. Epidemiology, pathophysiology, and natural history of pulmonary embolism. Semin Intervent Radiol. 2018;35:92-98.

2. Beckman MG, Hooper WC, Critchley SE, Ortel TL. Venous thromboembolism: a public health concern. Am J Prev Med. 2010;38:S495-S501.

3. Heit JA. Venous thromboembolism: disease burden, outcomes and risk factors. J Thromb Haemost. 2005;3:1611-1617.

4. Belohlavek J, Dytrych V, Linhart A. Pulmonary embolism, part I: epidemiology, risk factors and risk stratification, pathophysiology, clinical presentation, diagnosis and nonthrombotic pulmonary embolism. Exp Clin Cardiol. 2013;18:129-138.

5. Pollack CV, Schreiber D, Goldhaber SZ, et al. Clinical characteristics, management, and outcomes of patients diagnosed with acute pulmonary embolism in the emergency department: initial report of EMPEROR (multicenter emergency medicine pulmonary embolism in the real world registry). J Am Coll Cardiol. 2011;57:700-706.

6. Konstantinides SV, Meyer G, Becattini C, et al. 2019 ESC guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS): the task force for the diagnosis and management of acute pulmonary embolism of the European Society of Cardiology (ESC). Eur Respir J. 2019;54:1901647.

7. Aujesky D, Obrosky DS, Stone RA, et al. Derivation and validation of a prognostic model for pulmonary embolism. Am J Respir Crit Care Med. 2005;172:1041-1046.

8. Donze J, Le Gal G, Fine MJ, et al. Prospective validation of the pulmonary embolism severity index. A clinical prognostic model for pulmonary embolism. Thromb Haemost. 2008;100:943-948.

9. Schulman S, Kearon C, Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost. 2005;3:692-694.

10. van der Hulle T, Kooiman J, den Exter PL, Dekkers OM, Klok FA, Huisman MV. Effectiveness and safety of novel oral anticoagulants as compared with vitamin K antagonists in the treatment of acute symptomatic venous thromboembolism: a systematic review and meta-analysis. J Thromb Haemost. 2014;12:320-328.

11. van Dongen CJ, van den Belt AG, Prins MH, Lensing AW. Fixed dose subcutaneous low molecular weight heparins versus adjusted dose unfractionated heparin for venous thromboembolism. Cochrane Database Syst Rev. 2004:CD001100.

12. van der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briet E. Bleeding complications in oral anticoagulant therapy. An analysis of risk factors. Arch Intern Med. 1993;153:1557-1562.

13. Kuijer PM, Hutten BA, Prins MH, Buller HR. Prediction of the risk of bleeding during anticoagulant treatment for venous thromboembolism. Arch Intern Med. 1999;159:457-460.

14. Ruiz-Gimenez N, Suarez C, Gonzalez R, et al. Predictive variables for major bleeding events in patients presenting with documented acute venous thromboembolism. Findings from the RIETE registry. Thromb Haemost. 2008;100:26-31.

15. Kurlander JE, Gu X, Scheiman JM, et al. Missed opportunities to prevent upper GI hemorrhage: the experience of the Michigan Anticoagulation Quality Improvement Initiative. Vasc Med. 2019;24:153-155.

16. Chaudhury P, Gadre SK, Schneider E, et al. Impact of multidisciplinary pulmonary embolism response team availability on management and outcomes. Am J Cardiol. 2019;124:1465-1469.

17. Smith SB, Geske JB, Maguire JM, Zane NA, Carter RE, Morgenthaler TI. Early anticoagulation is associated with reduced mortality for acute pulmonary embolism. Chest. 2010;137:1382-1390.

18. Rosovsky R, Chang Y, Rosenfield K, et al. Changes in treatment and outcomes after creation of a pulmonary embolism response team (PERT), a 10-year analysis. J Thromb Thrombolysis. 2019;47:31-40.

19. Giri J, Sista AK, Weinberg I, et al. Interventional therapies for acute pulmonary embolism: current status and principles for the development of novel evidence: a scientific statement from the American Heart Association. Circulation. 2019;140:e774-e801.

From the ¹Department of Medicine, University Hospitals Cleveland Medical Center, Cleveland, Ohio; ²Harrington Heart & Vascular Institute, University Hospitals Cleveland Medical Center, Cleveland, Ohio; ³Division of Pulmonary Medicine, Department of Medicine, University Hospitals Cleveland Medical Center, Cleveland, Ohio.

Funding: Inari Medical provided an unrestricted education grant that funded the research nurse who collected the data for this project. Inari Medical had no involvement in the design, data collection, analysis, or writing of this manuscript. 

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Li reports consulting for Abbott, Medtronic, Terumo, and Abiomed. Dr Shishehbor reports global advisory board membership and consultancy to Abbott Vascular, Medtronic, Boston Scientific, Terumo, and Philips. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript accepted August 21, 2020.

Address for correspondence: Mehdi H. Shishehbor, DO, MPH, PhD, Professor of Medicine, Case Western Reserve University School of Medicine, University Hospitals, 11100 Euclid Avenue, Lakeside 3rd floor, Cleveland, OH 44106. Email: Mehdi.Shishehbor@UHhospitals.org; Twitter: @shishem