Artificial intelligence-based pulmonary embolism classification: Development and validation using real-world data.

This paper presents an artificial intelligence-based classification model for the detection of pulmonary embolism in computed tomography angiography. The proposed model, developed from public data and validated on a large dataset from a tertiary hospital, uses a two-dimensional approach that integrates temporal series to classify each slice of the examination and make predictions at both slice and examination levels. The training process consists of two stages: first using a convolutional neural network InceptionResNet V2 and then a recurrent neural network long short-term memory model. This approach achieved an accuracy of 93% at the slice level and 77% at the examination level. External validation using a hospital dataset resulted in a precision of 86% for positive pulmonary embolism cases and 69% for negative pulmonary embolism cases. Notably, the model excels in excluding pulmonary embolism, achieving a precision of 73% and a recall of 82%, emphasizing its clinical value in reducing unnecessary interventions. In addition, the diverse demographic distribution in the validation dataset strengthens the model's generalizability. Overall, this model offers promising potential for accurate detection and exclusion of pulmonary embolism, potentially streamlining diagnosis and improving patient outcomes.

Novel Risk Stratification and Hemodynamic Profiling in Acute Pulmonary Embolism: A Proposed Classification Inspired by Society for Cardiovascular Angiography and Intervention Shock Staging.

Treatment options for patients with acute pulmonary embolism (PE) and right ventricular shock (RVS) have grown exponentially. Therapy options include anticoagulation, systemic thrombolysis, catheter-based thrombolysis/ thrombectomy, and may include short-term mechanical circulatory support. However, the incidence of short-term morbidity and mortality has not changed despite the emergence of several advanced therapies in acute PE. This is possibly due to the inclusion of heterogenous populations in research studies without differentiation based on the acuity/severity of presentation. We propose a novel classification for PE-RVS to allow for standardizing appropriate therapy escalation and better communication of the severity among cardiovascular critical care, and emergency health care professionals.

Renal dysfunction improves risk stratification and may call for a change in the management of intermediate- and high-risk acute pulmonary embolism: results from a multicenter cohort study with external validation.

BACKGROUND: Renal dysfunction influences outcomes after pulmonary embolism (PE). We aimed to determine the incremental value of adding renal dysfunction, defined by estimated glomerular filtration rate (eGFR), on top of the European Society of Cardiology (ESC) prognostic model, for the prediction of 30-day mortality in acute PE patients, which in turn could lead to the optimization of acute PE management. METHODS: We performed a multicenter, non-interventional retrospective post hoc analysis based on a prospectively collected cohort including consecutive confirmed acute PE stratified per ESC guidelines. We first identified which of three eGFR formulae most accurately predicted death. Changes in global model fit, discrimination, calibration and reclassification parameters were evaluated with the addition of eGFR to the prognostic model. RESULTS: Among 1943 patients (mean age 67.3 (17.1), 50.4% women), 107 (5.5%) had died at 30 days. The 4-variable Modification of Diet in Renal Disease (eGFRMDRD4) formula predicted death most accurately. In total, 477 patients (24.5%) had eGFRMDRD4 < 60 ml/min. Observed mortality was higher for intermediate-low-risk and high-risk PE in patients with versus without renal dysfunction. The addition of eGFRMDRD4 information improved model fit, discriminatory capacity, and calibration of the ESC model. Reclassification parameters were significantly increased, yielding 18% reclassification of predicted mortality (p < 0.001). Predicted mortality reclassifications across risk categories were as follows: 63.1% from intermediate-low risk to eGFR-defined intermediate-high risk, 15.8% from intermediate-high risk to eGFR-defined intermediate-low risk, and 21.0% from intermediate-high risk to eGFR-defined high risk. External validation in a cohort of 14,234 eligible patients from the RIETE registry confirmed our findings with a significant improvement of Harrell's C index and reclassification parameters. CONCLUSION: The addition of eGFRMDRD4-derived renal dysfunction on top of the prognostic algorithm led to risk reclassification within the intermediate- and high-risk PE categories. The impact of risk stratification integrating renal dysfunction on therapeutic management for acute PE requires further studies.

Current Controversies in Caring for the Critically Ill Pulmonary Embolism Patient.

Emergency physicians must be prepared to rapidly diagnose and resuscitate patients with pulmonary embolism (PE). Certain aspects of PE resuscitation run counter to typical approaches. A specific understanding of the pathophysiology of PE is required to avoid cardiovascular collapse potentially associated with excessive intravenous fluids and positive pressure ventilation. Once PE is diagnosed, rapid risk stratification should be performed and treatment guided by patient risk class. Although anticoagulation remains the mainstay of PE treatment, emergency physicians also must understand the indications and contraindications for thrombolysis and should be aware of new therapies and models of care that may improve outcomes.

Balloon pulmonary angioplasty for chronic thromboembolic pulmonary hypertension.

Chronic thromboembolic pulmonary hypertension (CTEPH) is classified as Group-4 pulmonary hypertension caused by organized thrombi in pulmonary arteries. Pulmonary endarterectomy (PEA) has been shown to improve the symptoms and prognoses of patients with proximal CTEPH. The soluble guanylate cyclase stimulator (riociguat) is the sole FDA-licensed drug for the treatment of CTEPH, and guidelines recommend its use for patients with inoperable CTEPH or residual or recurrent pulmonary hypertension following PEA. Balloon pulmonary angioplasty (BPA) is a new procedure, but it is a promising alternative to PEA, especially in patients with inoperable CTEPH. This review summarizes the history, indications, procedures and complications of BPA. Finally, we discuss the future perspective of BPA for better management of CTEPH.

Definitions, adjudication, and reporting of pulmonary embolism-related death in clinical studies: A systematic review.

BACKGROUND: Pulmonary embolism (PE)-related death is a component of the primary outcome in many venous thromboembolism (VTE) studies. The absence of a standardized definition for PE-related death hampers study outcome evaluation and between-study comparisons. OBJECTIVES: To summarize definitions for PE-related death used in recent VTE studies and to assess the PE-related death rate. PATIENTS/METHODS: A systematic literature search was conducted on 26 April 2018 from 1 January 2014 up to the search date in MEDLINE, Embase, and CENTRAL. Cohort studies and randomized trials in which PE-related death was included in the primary outcome were eligible. Screening of titles, abstracts, and full-text articles, and data extraction were independently performed in duplicate by two authors. Study outcomes included the definition for PE-related death, VTE case-fatality rate, and death due to PE rate. Descriptive statistics were used to analyze the data. RESULTS: Of the 6807 identified citations, 83 studies were included of which 27% were randomized trials, 31% were prospective, and 42% retrospective cohort studies. Thirty-five studies (42%) had a central adjudication committee. Thirty-eight (46%) reported a definition for PE-related death of which the most frequently used components were "autopsy-confirmed PE" (50%), "objectively confirmed PE before death" (55%), and "unexplained death" (58%). Median VTE case-fatality rate was 1.8% (interquartile range, 0.0-13). CONCLUSIONS: Only half of the included studies reported definitions for PE-related death, which were very heterogeneous. Case-fatality rate of VTE events varied widely across studies. Standardization of the definition and guidance on adjudication and reporting of PE-related death is needed.

Frequency and intensity of pulmonary bone marrow and fat embolism due to manual or automated chest compressions during cardiopulmonary resuscitation.

Iatrogenic consequences of cardiopulmonary resuscitation (CPR) include sternal or rib fractures, pulmonary bone marrow embolisms (BME) and fat embolisms (FE). This report aimed to analyze the frequency and intensity of pulmonary BME and FE in fatal cases receiving final CPR efforts with the use of automated chest compression devices (ACCD) or manual chest compressions (mCC). The study cohort (all cardiac causes of death, no ante-mortem fractures) consisted of 15 cases for each group 'ACCD', 'mCC' and 'no CPR'. Lung tissue samples were retrieved and stained with hematoxylin eosin (n = 4 each) and Sudan III (n = 2 each). Evaluation was conducted microscopically for any existence of BME or FE, the frequency of BME-positive vessels, vessel size for BME and the graduation according to Falzi for FE. The data were compared statistically using non-parametric analyses. All groups were matched except for CPR duration (ACCD > mCC) but this time interval was linked to the existence of pulmonary BME (p = 0.031). Both entities occur in less than 25% of all cases following unsuccessful CPR. BME was only detectable in CPR cases, but was similar between ACCD and mCC cases for BME frequency (p = 0.666), BME intensity (p = 0.857) and the size of BME-affected pulmonary vessels (p = 0.075). If any, only mild pulmonary FE (grade I) was diagnosed without differences in the CPR method (p = 0.624). There was a significant correlation between existence of BME and FE (p = 0.043). Given the frequency, intensity and size of pulmonary BME and FE following CPR, these conditions may unlikely be considered as causative for death in case of initial survival but can be found in lower frequencies in autopsy histology.

Pulmonary Embolism for the Cardiologist: Emphasis on Diagnosis.

PURPOSE OF REVIEW: Pulmonary embolism (PE) is a common condition with high morbidity and mortality particularly if misdiagnosed or untreated. It has non-specific clinical manifestations, often presenting similarly to other cardiovascular conditions. The aim of this review is to summarize the clinical presentation, diagnostic algorithms, and imaging studies utilized to efficiently make or exclude the diagnosis of pulmonary embolism. RECENT FINDINGS: Recent cohort studies have raised questions about the classic presentations of PE (particularly with regard to syncope) or lack thereof. Diagnosis of PE continues to evolve with new diagnostic algorithms, use of age-adjusted D-dimer cutoffs, validation of older algorithms, and emerging data on multimodality ultrasound and ventilation-perfusion (V/Q) single-photon emission computed tomography (SPECT) imaging in the diagnosis of PE. Optimizing clinical outcomes of PE depend on correct diagnosis of the condition. Given significant variability in clinical presentation, use of a diagnostic algorithm is essential. Use of a clinical decision rule and D-dimer testing can risk stratify patients to allow for judicious use of diagnostic imaging. V/Q scan, CT pulmonary angiography, and lower extremity ultrasound remain the diagnostic imaging modalities of choice with other promising imaging modalities requiring further study.

Incidence and risk factors of chronic thromboembolic pulmonary hypertension in patients with diagnosis of pulmonary embolism for the first time in real world.

BACKGROUND: The incidence and risk factors of chronic thromboembolic pulmonary hypertension (CTEPH) in patients with acute pulmonary embolism (PE) have been well reported. However, in real world, patients diagnosed with PE for the first time were usually composed of acute PE, sub-acute PE, and chronic PE, and the cumulative incidence and risk factors of CTEPH in this cohort were still unknown. METHODS: A prospective, long-term, follow-up study was conducted to assess the incidence of symptomatic CTEPH in consecutive patients with PE diagnosed for the first time. Patients with unexplained persistent dyspnea during follow-up underwent transthoracic echocardiography and, if the findings indicated pulmonary hypertension, ventilation-perfusion lung scanning and right heart catheterization. CTEPH was confirmed if perfusion defects were present, mean pulmonary artery pressure (mPAP) ≥25 mmHg and pulmonary artery wedge pressure (PAWP) ≤15 mmHg. RESULTS: The cumulative incidence of CTEPH in patients with PE diagnosed for the first time was 11.2% at 3 months, 12.7% at 1 year, 13.4% at 2 years, and 14.5% at 3 years. The following factors increased the risk of CTEPH: time from symptoms to treatment of PE ≥1 month (odds ratio (OR), 14.77), intermediate (OR, 37.63) to high risk PE (OR, 39.81), segmental and sub-segmental branch location of embolism (OR, 8.30) and PE-related primary risk factors (OR, 5.01). 9.4% of CTEPH patients developed from acute PE, and 90.6% from sub-acute and chronic PE. CONCLUSIONS: In real world, CTEPH is a relatively common and serious complication in PE patients diagnosed for the first time. Early diagnosis and treatment of PE will decrease the incidence of CTEPH in these unspecified patients.

Positive predictive values of peripheral arterial and venous thrombosis codes in French hospital database.

French hospital database, called Programme de Médicalisation des Systèmes d'Information (PMSI), covers all hospital stays in France (>66 million inhabitants). The aim of this study was to estimate the positive predictive values (PPVs) of primary diagnosis codes of peripheral arterial and venous thrombosis codes in the PMSI, encoded with the International Classification of Diseases, 10th revision. Data were extracted from the PMSI database of Toulouse University Hospital, south of France. We identified all the hospital stays in 2015 with a code of peripheral arterial or venous thrombosis as primary diagnosis. We randomly selected 100 stays for each category of thrombosis and reviewed the corresponding medical charts. The PPV of peripheral arterial thrombosis codes was 83.0%, 95% confidence interval (CI): 73.9-89.1, and the PPV of correct location of thrombosis was 81.0%, 95% CI: 72.2-87.5. The PPV of pulmonary embolism was 99.0%, 95% CI: 93.8-99.9. The PPV of peripheral venous thrombosis was 95.0%, 95% CI: 88.2-98.1, and the PPV of correct location of thrombosis was 85.0%, 95% CI: 76.7-90.7. Primary diagnoses of peripheral arterial and venous thrombosis demonstrated good PPVs in the PMSI.

Pulmonary Embolism in 2017: How We Got Here and Where Are We Going?

In the 1970s, both the Urokinase Pulmonary Embolism and Urokinase-Streptokinase Pulmonary Embolism trials began the quest to develop thrombolytic therapy for the treatment of acute massive and submassive pulmonary embolism (PE). The goals of these studies were the immediate reduction in clot burden, restoration of hemodynamic stability, and improved survival. Major bleeding became the major barrier for clinicians to employ these therapies. From 1980s to the present time, a number of studies using recombinant tissue-type plasminogen activator for achieving these same above outcomes were completed but major bleeding continued to remain an adoption barrier. Finally, the concept of bringing the thrombolytic agent into the clot has entered the quest for the Holy Grail in the treatment of PE. This article will review all the major trials using peripheral thrombolysis and provide insight into the need for a team approach to pulmonary care (Pulmonary Embolism Response Team), standardization of pulmonary classification, and the need for trials designed for both short- and long-term outcomes using thrombolysis for selected PE populations.

Epidemiology, Pathophysiology, Stratification, and Natural History of Pulmonary Embolism.

Pulmonary embolism (PE) is a common and potentially fatal form of venous thromboembolism that can be challenging to diagnose and manage. PE occurs when there is obstruction of the pulmonary vasculature and is a common cause of morbidity and mortality in the United States. A combination of acquired and inherited factors may contribute to the development of this disease and should be considered, since they have implications for both susceptibility to PE and treatment. Patients with suspected PE should be evaluated efficiently to diagnose and administer therapy as soon as possible, but the presentation of PE is variable and nonspecific so diagnosis is challenging. PE can range from small, asymptomatic blood clots to large emboli that can occlude the pulmonary arteries causing sudden cardiovascular collapse and death. Thus, risk stratification is critical to both the prognosis and management of acute PE. In this review, we discuss the epidemiology, risk factors, pathophysiology, and natural history of PE and deep vein thrombosis.

Risk Stratification for Proven Acute Pulmonary Embolism: What Information Is Needed?

Classification of risk drives treatment decisions for patients with acute symptomatic pulmonary embolism (PE). High-risk patients with acute symptomatic PE have hemodynamic instability (i.e., shock or hypotension present), and treatment guidelines suggest systemically administered thrombolytic therapy in this setting. Normotensive PE patients at low risk for early complications (low-risk PE) might benefit from treatment at home or early discharge, while normotensive patients with preserved systemic arterial pressure deemed as having a high risk for PE-related adverse clinical events (intermediate-high-risk PE) might benefit from close observation and consideration of escalation of therapy. Prognostic tools (e.g., clinical prognostic scoring systems, imaging testing, and cardiac laboratory biomarkers) assist with the classification of patients into these categories.

Role of Survival Time and Injury Severity in Fatal Pulmonary Fat Embolism.

Pulmonary fat embolism (PFE) is frequent in blunt trauma and may occasionally lead to death. A correlation between fracture grade and severity and PFE grade has been described before, but no correlation between PFE and survival time, fat crushing extent, fat crush grade, or number of body regions with fractures could be noted in this small study. To further examine this, we decided to examine the aforementioned points in a far larger study group.Autopsy protocols of 188 nonresuscitated fatalities with blunt trauma and without right heart injury, which underwent whole body dissection, were retrospectively reviewed concerning the presence and the severity of PFE, injuries, survival time, age, sex, and the body mass index.The fracture grade, the fracture severity, and the number of the fractured regions correlated very well with the grade of PFE, but the crushed regions, crush grade, and crush severity did not. We observed a time correlation between survival time and PFE only in the sense that very rapid deaths were often PFE negative. High-grade PFE was observed most often in patients having died less than 6 hours after the incident, and PFE grades of 2 or more were occasionally noted even after 48 hours.

Diagnostic validity of hematologic parameters in evaluation of massive pulmonary embolism.

BACKGROUND: The aim of this study was to determine the hematologic parameter with the highest diagnostic differentiation in the identification of massive acute pulmonary embolism (APE). METHODS: A retrospective study was performed on patients diagnosing with APE between June 2014 and June 2016. All radiological and laboratory parameters of patients were scanned through the electronic information management system of the hospital. PLR was obtained from the ratio of platelet count to lymphocyte count, NLR was obtained from the ratio of neutrophil count to lymphocyte count, WMR was obtained from white blood cell in mean platelet volume ratio, MPR was obtained from the ratio of mean platelet volume to platelet count, and RPR was obtained from the ratio of red distribution width to platelet count. RESULTS: Six hundred and thirty-nine patients consisting of 292 males (45.7%) and 347 females (54.3%) were included in the research. Independent predictors of massive risk as compared to sub-massive group were; pulmonary arterial systolic pressure (PASP) (OR=1.40; P=.001), PLR (OR=1.59; P<.001), NLR (OR=2.22; P<.001), WMR (OR=1.22; P<.001), MPR (OR=0.33; P<.001), and RPR (OR=0.68; P<.001). Upon evaluation of the diagnostic differentiation of these risk factors for massive APE by employing receiver operating characteristic curve analysis, it was determined that PLR (AUC±SE=0.877±0.015; P<.001), and NLR (AUC±SE=0.893±0.013; P<.001) have similar diagnostic differentiation in diagnosing massive APE and these two parameters are superior over PASP, MPR, WMR, and RPR. CONCLUSION: We determined that the levels of NLR and PLR are superior to other parameters in the determination of clinical severity in APE cases.

Novel Angiographic Classification of Each Vascular Lesion in Chronic Thromboembolic Pulmonary Hypertension Based on Selective Angiogram and Results of Balloon Pulmonary Angioplasty.

BACKGROUND: Balloon pulmonary angioplasty (BPA) is an alternative therapy for patients with chronic thromboembolic pulmonary hypertension who are ineligible for standard therapy, pulmonary endarterectomy. Although there are several classifications of vascular lesions, these classifications are based on the features of the specimen removed during pulmonary endarterectomy. Because organized thrombi are not removed during balloon pulmonary angioplasty, we attempted to establish a new classification of vascular lesions based on pulmonary angiographic images. We evaluated the success and complication rate of BPA in accordance with the location and morphology of thromboembolic lesions. METHODS AND RESULTS: We reviewed 500 consecutive procedures (1936 lesions) of BPA in 97 patients with chronic thromboembolic pulmonary hypertension and investigated the outcomes of BPA based on the lesion distribution and the angiographic characteristics of the thromboembolic lesions, as follows: type A, ring-like stenosis lesion; type B, web lesion; type C, subtotal lesion; type D, total occlusion lesion, and type E, tortuous lesion. The success rate was higher, and the complication rate was lower in ring-like stenosis and web lesions. The total occlusion lesions had the lowest success rate. Tortuous lesions were associated with a high complication rate and should be treated only by operators with extensive experience with BPA. CONCLUSIONS: We modified the previous angiographic classification and established a new classification for each vascular lesion. We clarified that the outcome and complication rate of the BPA are highly dependent on the lesion characteristics.

Prognostic value of computed tomography in acute pulmonary thromboembolism. / Valor pronóstico de la tomografía computarizada en la tromboembolia pulmonar aguda.

In addition to being the standard reference for the diagnosis of acute pulmonary thromboembolism, CT angiography of the pulmonary arteries can also provide valuable information about the patient's prognosis. Although which imaging findings are useful for prognosis remains controversial, signs of right ventricular dysfunction on CT are now included in clinical algorithms for the management of pulmonary thromboembolism. However, the optimal method for obtaining these measurements while maintaining a balance between the ease of use necessary to include their evaluation in our daily activity and the loss of precision in its predictive capacity remains to be determined. Moreover, other variables associated with pulmonary thromboembolism that often go unobserved can complement the prognostic information we can offer to clinicians. This review aims to clarify some of the more controversial aspects related to the prognostic value of CT in patients with pulmonary embolisms according to the available evidence. Knowing which variables are becoming more important in the prognosis, how to detect them, and why it is important to include them in our reports will help improve the management of patients with pulmonary embolism.