Predictors of acute pulmonary embolism-related hospitalizations – an artificial neural network analysis using a nationwide cohort in the United States

Background

Considering a paucity of large-scale data on predictors of pulmonary embolism (PE) and its higher association with complications and worse outcomes, we aimed to determine the predictors of PE in this United States population-based analysis using Artificial Neural Network (ANN) Model in a nationally representative cohort.

Methods

We identified PE-related hospitalizations using 2018's National Inpatient Sample database. The relevant predictive factors for ANN were selected for this cohort. Of all admissions (unweighted n=7,105,498, weighted n=35,527,481), PE cohort (weighted n=387805) consisted of 1.1% of all admissions in 2018. The 2018 cohort was randomly split into training data (unweighted n=4716132, 70.0%) which were used to calibrate ANN and testing data (unweighted n=2019290, 30%) which were used to assess the accurateness of the algorithm. We equated the rate of incorrect prediction between training and testing data and measured the Area under Receiver Operator Curve (AUC) to determine ANN's efficacy in predicting PE hospitalizations.

Results

Patients hospitalized with PE often consisted of older (mean age 62.5±17.1 years), female (51.3%), white (70.5%) patients, and patients from lower-income quartile (0–25% income quartile: 28.8%%), often admitted non-electively (93.7%) with higher rates of cardiovascular disease risk factors. PE admissions revealed significantly higher (6.5% vs. 1.9%, p<0.001) in-hospital mortality, less frequent routine discharges (51.4% vs. 68.1%) and more frequent other facility transfers and requirement of home health care. Normalized Predictors of PE admissions are displayed in Fig. 1. Our ANN model had AUC 0.873 which correlates with an excellent prediction model. Our data demonstrated low levels (0.8%) error in both testing and training models.

Conclusion

Our ANN model showed high performance to predict risk factors for PE admissions in the US population. It will enable clinicians to screen patients at high-risk for PE admissions, curtail complication rate, improve survival and lower the healthcare cost.

Funding Acknowledgement

Type of funding sources: None.

Cardiovascular disease risk factors and outcomes of acute myocardial infarction in young adults in two nationwide cohorts in the united states

Funding Acknowledgements

Type of funding sources: None.

Background

Acute myocardial infarction (AMI) can have considerable morbidity and devastating socioeconomic and psychological consequences in young adults. Previous studies reveal that the decline in mortality in AMI has mainly been in the older population while being comparatively less significant in younger patients. This study compares young adults (18 to 44 years) hospitalized with AMI across two nationwide cohorts, 2007 and 2017, in the United States (US). It examines the burden of AMI hospitalizations, the prevalence of comorbidities, and in-hospital outcomes in young adults a decade apart. It highlights the rise in AMI hospitalizations, lack of decrease in mortality, sex-based and racial disparities, the surge in post-MI complications, and the decline in reperfusion interventions in young AMI patients over a decade.

Purpose

Coronary heart disease prevalence is challenging to ascertain in younger adults because of limited data and frequent silent clinical presentations. AMI and its complications can cause considerable morbidity, psychological trauma, and socioeconomic burden in the young.

Methods

We identified hospitalizations for AMI in young adults in 2007 and 2017 using the weighted data from the National Inpatient Sample (NIS), which covers 20% of stratified data of all non-federal community hospitals in the US. We compared the following data between the two cohorts: admission rates, sociodemographic features, in-hospital morbidity, complications, mortality, rate of coronary interventions, and healthcare utilization between the two cohorts. We used Pearson’s Chi-square test and Mann-Whitney U test to compare categorical and continuous variables, respectively. We also applied multivariable regression analyses to assess and compare the risk of cardiovascular complications and in-hospital mortality while controlling for confounders, including age, sex, race, median household income quartile, primary insurance enrolment, and pre-existing comorbidities.

Results

AMI’s incidence was higher in males in both the cohorts, although with a decline (71.1% vs 66.1%), whereas it rose from 28.9% to 33.9% in females. Hypertension (47.8% vs 60.7%), smoking (49.7% vs 55.8%), obesity (14.8% vs 26.8%), and diabetes mellitus (22.0% vs 25.6%) increased in the 2017 cohort (Table 1). We found no significant difference in all-cause mortality (aOR = 1.01 (0.93-1.10), p=0.749). Post-AMI complications, cardiogenic shock (aOR = 1.16 (1.06-1.27), p=0.001), and fatal arrhythmias increased. Reperfusion interventions decreased in the 2017 cohort (PCI; aOR=0.95 (0.91-0.98), p<0.001; CABG; aOR=0.66 (0.61-0.71), p<0.001) (Table 2).

Conclusion

Our study highlights the rise in AMI hospitalizations, plateauing of mortality, gender disparity, the surge in post-MI complications, and a reassuring decline in the requirement of reperfusion interventions in young AMI patients over a decade.

Frequency, risk and predictors of type 2 myocardial infarction hospitalizations in young obese patients: A nationwide population-based analysis in the United States

Funding Acknowledgements

Type of funding sources: None.

Background

Obesity in the young population is emerging as a challenging health concern. Though there is a rising prevalence of obesity and its potential association with demand ischemia-related myocardial infarction, data remains non-existent to evaluate the association of obesity or higher body mass index (BMI) with type 2 myocardial infarction (T2MI). We aim to identify the frequency, risk and predictors of T2MI in young hospitalized obese patients compared to non-obese patients in this population-based study.

Methods

We used National Inpatient Sample (2018, ICD-10 codes) to identify T2MI in young (18-44 years) hospitalized patients. Obesity was identified from comorbidities or using diagnostic codes for BMI>30 kg/m2. We performed multivariable regression analysis for the primary outcome of odds of T2MI in young obese patients compared to non-obese patients. The frequency of T2MI was compared between obese vs non-obese patients in overall and subgroup populations. Sociodemographic characteristics and comorbidities in T2MI-obese vs. T2MI-non-obese cohorts were also compared. A p<0.05 was considered a threshold for statistical significance.

Results

Out of 1,268,255 young hospitalized patients with obesity, 555 had T2MI. T2MI was significantly higher in young obese than non-obese (44 T2MI/100000 hospitalizations in young obese patients vs. 17 T2MI/100000 hospitalizations in young non-obese patients, overall 0.04% in obese vs. 0.02% in non-obese, p<0.001). Multivariate analysis revealed higher odds of T2MI in obese than nonobese when adjusted for demographics (aOR 2.65, 95% CI:2.42-2.90, p<0.001) and social demographics with comorbidities (aOR 1.60, 95% CI:1.24-2.07, p<0.001). In young obese, higher risk was found with advancing age (OR 1.07, 95% CI 1.03-1.11, p=0.001), in males than females (aOR 2.70, p<0.001), and blacks (aOR 2.22, p=0.011) and Native Americans (OR 3.91, 95% CI: 1.13-13.49, p=0.011) vs whites. Comorbidities including chronic obstructive pulmonary disease (OR 1.86), chronic kidney disease (CKD, OR 2.36), rheumatoid arthritis/collagen vascular disease (RA/CVD, OR 3.04) Iin young obese patients independently increased the risk of T2MI hospitalizations [Table 1]. The T2MI-obese cohort had a significantly higher rate of hyperlipidemia, hypertension, diabetes, COPD, and prior history of MI and TIA/stroke compared to the T2MI-nonobese cohort [Table 2].

Conclusion

This nationwide analysis revealed a significantly higher risk of T2MI in young obese patients compared to nonobese after excluding patients with concomitant diagnoses of T1MI. Males, blacks compared to females and whites, and comorbidities including COPD, CKD and RA/CVD predicted a higher risk of T2MI in young obese patients. Future studies are warranted to evaluate the role of higher body mass index in myocardial oxygen demand-supply mismatch and short-term/long-term risk and outcomes of T2MI.

Nationwide frequency, risk and outcomes of type-2 myocardial infarction in patients with versus without previously revascularized myocardial infarction (type 1)

Funding Acknowledgements

Type of funding sources: None.

Background

Type 2 myocardial infarction (T2MI), due to a mismatch between myocardial oxygen demand and supply, is being increasingly recognized with improved diagnostics. The upsetting concern of developing T2MI in patients with prior revascularized occlusive acute myocardial infarction (AMI) or type 1 MI (T1MI) makes it crucial to define the clinical profile and outcomes of T2MI in revascularized patients of ACS.

Purpose

To determine the risk and prognosis of T2MI in patients who had previously had coronary revascularization (PCI or CABG)

Methods

We used the National Inpatient Sample (2018) dataset from the United States to identify T2MI adult hospitalizations using ICD-10 codes and define our study arm as T2MI excluding secondary T1MI diagnoses but having prior revascularized (with percutaneous coronary intervention or coronary artery bypass grafting) AMI. We then compared demographics and comorbidities in T2MI cohort with vs without personal history of revascularized AMI. We used multivariate analysis to study the odds of T2MI hospitalizations with prior revascularized AMI and in-hospital outcomes (all-cause mortality, cardiogenic shock and resource utilization) adjusting for confounders.

Results

There were 33155 T2MI adult hospitalizations after excluding AMI (median age 71 years, 50.6% male, 67.3% white); 1435 (4.3%) had previously revascularized AMI. T2MI in the study arm had higher chances of hospitalization with prior revascularized AMI when adjusted for socio-demographics (aOR 6.92, 95% CI:6.50-7.36, p<0.001) and socio-demographics with comorbidities (aOR 5.70, 95%CI: 5.48-5.94, p<0.001) (Table 1). Study arm often had elderly (≥65 years old, 78.4% vs 65.8%), male (66.6% vs 49.9%), white (76.7% vs 66.9%), upper socio-economic class (20.2 vs 16.8%), patients who were often admitted to non-electively (99.3 vs 97.1%) and to rural (10.5 vs 9.3%) hospitalizations compared to control arm. The study arm had a significantly higher prevalence of diabetes mellitus, hyperlipidemia, peripheral vascular disease, chronic obstructive pulmonary disease, renal failure, deficiency anemias, prior TIA/stroke, depression and smoking. T2MI cohort with prior revascularized AMI did not show any significant association with in-hospital all-cause mortality (1.7 vs 3.0%, aOR 0.49, 95%CI 0.18-1.34, p=0.164) and cardiogenic shock (1.7% vs 2.1%, p=0.399) however, had lower hospital expenditure (median USD 31273 vs 36567) and fewer transfers to other facilities (19.5 vs 22.1%) than those without prior revascularized AMI (Table 2).

Conclusion

Population-based analysis of this nationally representative sample revealed up to six times higher risk of developing T2MI in patients with prior history of AMI (revascularized) but without any significant impact on all-cause in-hospital mortality or cardiogenic shock. Future studies are warranted to assess the short-term/long-term outcomes of T2MI in high risk patient population with previously revascularized AMI.