Development and Validation of a Natural Language Processing Model to Identify Low-Risk Pulmonary Embolism in Real Time to Facilitate Safe Outpatient Management.

STUDY OBJECTIVE: This study aimed to (1) develop and validate a natural language processing model to identify the presence of pulmonary embolism (PE) based on real-time radiology reports and (2) identify low-risk PE patients based on previously validated risk stratification scores using variables extracted from the electronic health record at the time of diagnosis. The combination of these approaches yielded an natural language processing-based clinical decision support tool that can identify patients presenting to the emergency department (ED) with low-risk PE as candidates for outpatient management. METHODS: Data were curated from all patients who received a PE-protocol computed tomography pulmonary angiogram (PE-CTPA) imaging study in the ED of a 3-hospital academic health system between June 1, 2018 and December 31, 2020 (n=12,183). The "preliminary" radiology reports from these imaging studies made available to ED clinicians at the time of diagnosis were adjudicated as positive or negative for PE by the clinical team. The reports were then divided into development, internal validation, and temporal validation cohorts in order to train, test, and validate an natural language processing model that could identify the presence of PE based on unstructured text. For risk stratification, patient- and encounter-level data elements were curated from the electronic health record and used to compute a real-time simplified pulmonary embolism severity (sPESI) score at the time of diagnosis. Chart abstraction was performed on all low-risk PE patients admitted for inpatient management. RESULTS: When applied to the internal validation and temporal validation cohorts, the natural language processing model identified the presence of PE from radiology reports with an area under the receiver operating characteristic curve of 0.99, sensitivity of 0.86 to 0.87, and specificity of 0.99. Across cohorts, 10.5% of PE-CTPA studies were positive for PE, of which 22.2% were classified as low-risk by the sPESI score. Of all low-risk PE patients, 74.3% were admitted for inpatient management. CONCLUSION: This study demonstrates that a natural language processing-based model utilizing real-time radiology reports can accurately identify patients with PE. Further, this model, used in combination with a validated risk stratification score (sPESI), provides a clinical decision support tool that accurately identifies patients in the ED with low-risk PE as candidates for outpatient management.

Association of race and in-hospital outcomes following acute pulmonary embolism: A retrospective cohort study.

BACKGROUND: Racial disparities in health care are well established, with Black patients frequently experiencing the most significant consequences of this inequality. Acute pulmonary embolism (PE) is increasing in incidence and an important cause of morbidity and mortality in the United States, but little is known about racial disparities in the inpatient setting. HYPOTHESIS: Black and White patients admitted with acute PE will have different in-hospital outcomes. METHODS: All PE patients from January 1, 2016 to June 30, 2017 were retrospectively identified using ICD-10 codes. Data were abstracted by manual chart review for all image-confirmed PEs. RESULTS: A total of 782 patients with acute PE were identified, of which 319 (40.8%) were Black and 463 (59.2%) were White. Black patients had higher BMI (median [Q1-Q3]: 30.3 [25.4-36.6] vs. 29.3 [24.5-33.8] kg/m2 , p = .017), were younger (61 [48-74] vs. 67 [54-75] years, p = .001), and were more likely to have a history of heart failure (16.0 vs. 7.1%, p < .001), while White patients had higher rates of malignancy (46.9 vs. 34.5%, p = .001) and recent surgery (29.6 vs. 18.2%, p < .001). Black patients were more likely to receive systemic thrombolysis (3.1% vs. 1.1%, p = .040), while White patients had numerically higher rates of surgical embolectomy (0.3% vs. 1.1%, p = .41). No difference in inpatient mortality was observed; however, Black patients had longer hospital length of stay (5.0 [3-9] vs. 4.0 [2-9] days, p = .007) and were more likely to receive warfarin (23.5 vs. 12.1%, p < .001). CONCLUSIONS: Similar in-hospital mortality rates were observed in Black and White patients following acute PE. However, Black patients had longer hospital stays, higher warfarin prescription, and fewer traditional PE-related risk factors.

Medical Management of Peripheral Artery Disease.

Peripheral artery disease is a highly morbid yet undertreated atherosclerotic disease. The cornerstones of peripheral artery disease therapy consist of smoking cessation, lipid-lowering therapy, and hypertension treatment. More recently, clinical trials have demonstrated that novel antiplatelet and antithrombotic therapies reduce the risk of both cardiovascular and limb events in this patient population. In this review, we highlight the components of optimal medical therapy of peripheral artery disease and the evidence base for these therapies.

Use of hospital resources in the care of patients with intermediate risk pulmonary embolism.

BACKGROUND: Risk stratification and management of hemodynamically stable pulmonary embolism remains challenging. Professional societies have published stratification schemes, but little is known about the management of patients with intermediate risk pulmonary embolism. We describe the care of these patients at an academic health system. METHODS: Patient encounters from 1 January 2016 to 30 June 2017 were retrospectively identified utilizing a multihospital, electronic health record-based data warehouse. Using the 2019 European Society of Cardiology criteria, differences in hospital resource utilization, defined as intensive care unit admission, use of invasive therapies, and length of stay, were examined in patients with intermediate risk characteristics. RESULTS: A cohort of 322 intermediate risk patients, including 165 intermediate-low and 157 intermediate-high risk patients, was identified. Intermediate-high risk patients more often underwent catheter-directed therapy (14.0% vs. 1.8%; P<0.001) compared to intermediate-low risk patients and had a 50% higher rate of intensive care unit admission (relative risk 1.50; 95% confidence interval 1.06, 2.12; P=0.023). There was no difference in median intensive care unit length of stay (2.7 vs. 2.0 days; P=0.761) or hospital length of stay (5.0 vs. 5.0 days; P=0.775) between intermediate-high risk and intermediate-low risk patients. Patients that underwent invasive therapies had a 3.8-day shorter hospital length of stay (beta -3.75; 95% confidence interval -6.17, -1.32; P=0.002). CONCLUSION: This study presents insights into the hospital resource utilization of patients with intermediate risk pulmonary embolism. The 2019 European Society of Cardiology risk stratification criteria are a clinically relevant scheme that identifies patients more often treated with intensive care unit admission and advanced therapies.

Use of hospital resources in the care of patients with intermediate risk pulmonary embolism.

BACKGROUND: Risk stratification and management of hemodynamically stable pulmonary embolism remains challenging. Professional societies have published stratification schemes, but little is known about the management of patients with intermediate risk pulmonary embolism. We describe the care of these patients at an academic health system. METHODS: Patient encounters from 1 January 2016 to 30 June 2017 were retrospectively identified utilizing a multihospital, electronic health record-based data warehouse. Using the 2019 European Society of Cardiology criteria, differences in hospital resource utilization, defined as intensive care unit admission, use of invasive therapies, and length of stay, were examined in patients with intermediate risk characteristics. RESULTS: A cohort of 322 intermediate risk patients, including 165 intermediate-low and 157 intermediate-high risk patients, was identified. Intermediate-high risk patients more often underwent catheter-directed therapy (14.0% vs. 1.8%; P<0.001) compared to intermediate-low risk patients and had a 50% higher rate of intensive care unit admission (relative risk 1.50; 95% confidence interval 1.06, 2.12; P=0.023). There was no difference in median intensive care unit length of stay (2.7 vs. 2.0 days; P=0.761) or hospital length of stay (5.0 vs. 5.0 days; P=0.775) between intermediate-high risk and intermediate-low risk patients. Patients that underwent invasive therapies had a 3.8-day shorter hospital length of stay (beta -3.75; 95% confidence interval -6.17, -1.32; P=0.002). CONCLUSION: This study presents insights into the hospital resource utilization of patients with intermediate risk pulmonary embolism. The 2019 European Society of Cardiology risk stratification criteria are a clinically relevant scheme that identifies patients more often treated with intensive care unit admission and advanced therapies.

A Pilot Study of a Standardized Smoking Cessation Intervention for Patients with Vascular Disease.

BACKGROUND: The goal of this study is to evaluate the efficacy of a smoking cessation intervention performed by a vascular surgery provider compared with current smoking cessation practices. METHODS: Patients with peripheral arterial and aneurysmal disease who presented to the vascular surgery service at a tertiary care center over a 9-month period were randomized to either control or intervention group. Both control and intervention groups received 2 weeks of free nicotine patches and referral to an outpatient smoking-cessation program. The intervention group additionally received a brief presentation by a vascular surgeon regarding the benefits of smoking cessation, with a focus on vascular complications. At enrollment and at follow-up, patients underwent carbon monoxide breath testing and completed a survey. The primary outcome was smoking cessation or reduction among control and intervention groups in patients who underwent medical management, endovascular procedures, or open surgical procedures. Fisher's exact test was used to assess the primary outcome among groups. RESULTS: Fifty-nine patients were enrolled in the trial initially, but 55 had 1-month follow-up (control n = 28, intervention n = 27) and 52 had long-term follow-up (control n = 28, intervention n = 24). By long-term follow-up, 40 patients (77%) had reduced smoking by at least 50% and 16 patients (31%) had quit completely. At long-term follow-up, 88% of patients in the intervention group and 68% of patients in the control group reduced smoking (P = 0.1). CONCLUSIONS: A large proportion of vascular patients who received 2 weeks of nicotine replacement with or without the addition of brief smoking cessation counseling delivered by a vascular surgery provider were able to reduce smoking and maintain reduction after 6 months. Delivery of a brief standardized smoking cessation counseling session by a vascular surgery provider is safe and feasible. Additional randomized controlled trials with large enrollment periods and long follow-up are needed to determine the efficacy of this intervention in comparison to standard care.

Identification, diagnosis, treatment, and in-hospital outcomes of acute pulmonary embolism: Results from a single integrated health system.

BACKGROUND: Although the high-risk acute pulmonary embolism (PE) population has been described, little is known about the contemporary inpatient experience and practice patterns of the PE population as a whole. METHODS: All patients with a diagnosis of acute PE from January 1, 2016, to June 30, 2017 within our academic, multihospital health system were retrospectively identified using International Classification of Diseases, 10th Revision, codes, and data were manually abstracted by 2 clinical investigators. Descriptive analyses were performed according to clinical risk stratification categories from the European Society of Cardiology. RESULTS: Of 829 total patients, 372 (44.8%) patients had intermediate or high-risk PE. Mean age was 62.1 years old, and 42.1% of patients had a history of malignancy. One hundred fifty-three (18.5%) patients had an acute PE during a hospitalization for another indication. A total of 6.0% underwent invasive PE therapies, 26.1% required intensive care unit admission, and 9.0% experienced in-hospital death or hospice discharge. In a subgroup description, patients who developed acute PE during a hospitalization for another indication had a higher incidence of incomplete risk stratification and a higher mortality (9.8%) than the primary cohort. Mortality was attributed to PE in 48.4% of cases. CONCLUSIONS: This contemporary description of acute PE managed at a single large, multihospital academic health system highlights substantial health care utilization and high mortality despite the available of advanced therapeutics. Additional work is needed to standardize care for the heterogeneous PE population to ensure appropriate allocation of resources and improved outcomes for all PE patients.

Glucose effectiveness in obese children: relation to degree of obesity and dysglycemia.

OBJECTIVE: Impaired glucose effectiveness (GE) plays a role in the deterioration of glucose metabolism. Our aim was to validate a surrogate of GE derived from an oral glucose tolerance test (OGTT) and to assess the impact of degrees of obesity and of glucose tolerance on it. RESEARCH DESIGN AND METHODS: The OGTT-derived surrogate of GE (oGE) was validated in obese adolescents who underwent an OGTT and an intravenous glucose tolerance test (IVGTT). We then evaluated anthropometric determinants of the oGE and its impact on the dynamics of glucose tolerance in a cohort of children with varying degrees of obesity. RESULTS: The correlation of oGE and IVGTT-derived GE in 98 obese adolescents was r = 0.35 (P < 0.001) as a whole and r = 0.51 (P < 0.001) in subjects with normal glucose tolerance. In a cohort of 1,418 children, the adjusted GE was associated with increasing obesity (P < 0.001 for each category of obesity). Quartiles of oGE and the oral disposition index were associated with 2-h glucose levels (P < 0.001 for both). Among 421 nondiabetic obese subjects (276 subjects with normal glucose tolerance/145 subjects with impaired glucose tolerance who repeated their OGTT after a mean time of 28 ± 16 months), oGE changes were tightly associated with weight (r = 0.83, P < 0.001) and waist circumference changes (r = 0.67, P < 0.001). Baseline oGE and changes in oGE over time emerged as significant predictors of the change in 2-h glucose levels (standardized B = -0.76 and B = -0.98 respectively, P < 0.001 for both). CONCLUSIONS: The oGE is associated with the degree of and changes in weight and waist circumference and is an independent predictor of glucose tolerance dynamics.

A low disposition index in adolescent offspring of mothers with gestational diabetes: a risk marker for the development of impaired glucose tolerance in youth.

AIMS/HYPOTHESIS: With the increase in gestational diabetes mellitus (GDM), there is a growing need to understand the effects of intrauterine glucose exposure on the newborn at birth and later in life. The risk of developing impaired glucose tolerance (IGT) in individuals exposed to diabetes in utero has not been adequately investigated. METHODS: We studied 255 obese adolescents with normal glucose tolerance. All of them were investigated for in utero exposure to GDM and underwent an OGTT, which was repeated after approximately 2.8 years. RESULTS: 210 (82.3%) participants were not exposed to GDM (NGDM group), and 45 (17.7%) were exposed to GDM (EGDM group). In the NGDM group, only 8.6% (n = 18) developed either IGT or type 2 diabetes compared with 31.1% (n = 14) of the EGDM group who developed either IGT or type 2 diabetes (p < 0.001). Exposure to GDM was the most significant predictor of developing IGT or type 2 diabetes (OR 5.75, 95% CI 2.19, 15.07, p < 0.001). At baseline and at follow-up, the EGDM group showed a reduction in beta cell function determined by the oral disposition index (p = 0.03 and p = 0.01, respectively), and, at follow-up, they also displayed a reduction in insulin sensitivity compared with the NGDM group (p = 0.05). CONCLUSIONS/INTERPRETATION: Obese youth exposed in utero to GDM show early inability of the beta cell to compensate adequately in response to decreasing levels of insulin sensitivity.