Meverden, Ryan A.

Background: Risk stratification for suspected pulmonary emboli (PE) remains an area of active research. Current risk stratification algorithms have moderate discriminatory capabilities and are guideline-recommended, but they are poorly utilized in practice, with most clinicians relying predominantly on their clinical training and personal “gestalt.” Our group recently derived an Artificial Intelligence (AI) algorithm using electrocardiograms (ECG) to predict the presence of PE (AUC 0.69) on computed tomography scans of the pulmonary arteries (CTPA) in a cohort of 79K patients. Aim: Validate the AI-ECG algorithm in a contemporary dataset of patients with suspected PE. Methods: Emergency Department (ED) visits between 9/1/2022 and 12/31/2023 across the Mayo Clinic Enterprise were searched using a unified data platform. ED visits in patients over 18 years of age with a discretely documented Wells' PE or PERC Rule, a D-dimer result, or a CTPA were included. Only CTPA imaging and D-dimers performed within 48 hours of ED admission were eligible to be included. Patients were included in the final cohort if an ECG was performed within 6 hours of ED admission, D-dimer testing, or CTPA. Patients were considered negative for PE if CTPA was negative, and when this wasn't performed, a negative D-dimer (high sensitivity) or PERC rule of 0 was considered negative. A previously derived Artificial Neural Network algorithm to analyze the ECG to predict PE was applied to the dataset for validation. Results: A total of 27,028 ED visits in 24,288 unique patients meet inclusion criteria. Among this group, the first ED visit with a CTPA and an ECG within 6 hours was included, leaving 18,936 patients for the final validation cohort. The mean age of the cohort was 54.7 (SD 18.7), and 57% were male. Among the entire cohort, 13,573 underwent D-dimer testing for which 4,895 (36%) were positive at standard thresholds. A total of 10,486 patients underwent CTPA (55% of patients evaluated), of which 455 were positive for PE (4.3%). Using the AI-ECG model from the derivation cohort applied to this validation cohort resulted in an AUC of 0.69, identical to the derivation cohort. Using ECG model estimates alone, low (0.74%, n=2447, NPV 99.3%), elevated (2.27%, n=15404, NPV, 97.72%), and high-risk (8.02%, n=1085, NPV 91.98%) categories for PE can be created. A negative D-dimer in the elevated risk group increased the negative predictive value (NPV) to >99.9% and could be used to reduce imaging in this group further. Using the proposed ECG to D-dimer cascade risk stratification could further reduce the need for PE imaging for an overall NPV of 99.9% and sensitivity of 95.2%. We then used additional clinical parameters to train a new algorithm with additional clinical data and the AI-ECG algorithm estimates. Multiple machine learning algorithms were then evaluated in a dataset divided into training (2/3) and testing (1/3) datasets, after downsampling (to address class imbalance). The best results were obtained with AI-ECG results and D-dimer: Ada Boost (AUC 0.93, F1 score 0.895) compared to standard logistic regression (AUC 0.79, F1 score 0.746). The addition of age, sex, and oxygen saturation did not improve model performance. Conclusion: We have now validated the previously derived AI-ANN ECG algorithm for PE in a contemporary and independent cohort and demonstrate potential for integration into clinical PE assessment algorithms. Unlike prior risk stratification processes, ECGs are invariably performed in patients with cardiopulmonary symptoms and have the potential to prevent missed or delayed PE diagnosis by proactively identifying high-risk patients. Additionally, we demonstrate that the results of this algorithm can be combined with D-dimer values, resulting in a simple, powerful, predictive algorithm (AUC 0.93) that pairs cardiovascular electrophysiology with laboratory biomarkers and could lead to a new paradigm in clinical risk stratification for PE.

AbstractBackgroundClinical picture and outcome of incidental pulmonary embolism (iPE) compared to symptomatic pulmonary embolism (sPE) remain unclear.MethodsDemographics, recurrent venous thromboembolism (VTE), mortality, major bleeding, and clinically relevant nonmajor bleeding (CRNMB) were compared between iPE and sPE patients who were followed prospectively at Mayo Thrombophilia Clinic (March 1, 2013 to August 1, 2020).ResultsOut of 3576 VTE patients, 1417 (39.6%) had PE: 562 (39.7%) iPE and 855 sPE. Patients with cancer were more likely to have iPE (400 iPE vs. 314 sPE) compared to those without cancer (162 iPE vs. 541 sPE). VTE recurrence rate (all per 100 person‐years) was similar in all iPE and sPE patients (3.34 vs. 3.68, p = .50), with cancer (4.16 vs. 4.89, p = .370), and without cancer patients (0.89 vs. 2.80, p = .25). Higher mortality observed in all patients with iPE compared to sPE (46.45 vs. 23.47, p < .001) and with cancer (56.41 vs. 45.77, p = .03) became not significant after adjustment for age, antiplatelet therapy, metastases, and cancer location. Noncancer iPE patients had higher mortality (15.95 vs. 7.18, p = .006) even after adjustment (p = .05). The major bleeding rate was also higher in all patients iPE compared to sPE (7.10 vs. 3.68, p = .03), but not after adjustment (p = .974); higher major bleeding rate in noncancer patients (6.49 vs. 1.25, p = .007) remained significant after adjustment (.02). CRNMB rate was similar to iPE and sPE patients.ConclusioniPE represents a more serious clinical condition compared to sPE as indicated by the higher mortality and major bleeding but these differences reflect underlying comorbidities rather than the seriousness of the embolic event.

Background: Popliteal fossa cysts (PFCs aka Baker's cysts) are synovial cysts of the knee joint that can be symptomatic or asymptomatic and incidentally identified on ultrasound. Whether PFCs are associated with deep vein thrombosis (DVT) is unknown. Possible mechanisms for an association include direct compression of the popliteal vein, indirect compression on the popliteal vein with leg flexion, adjacent inflammation of the cyst, or relative immobility due to underlying joint disease itself. Methods: Lower extremity venous Duplex ultrasound radiology reports from the inception of electronic archiving through 11/14/2019 were evaluated across the Mayo Clinic Enterprise (Rochester MN, Jacksonville, FL, Scottsdale AZ, and Mayo Clinic Health System) in patients &gt;18 years of age. Natural language processing (NLP) algorithms were created and validated to identify acute DVT (proximal or distal) and PFCs. A random sample of 1,752 ultrasound reports underwent manual review to calculate the sensitivity and specificity of the NLP algorithm. Cases (ultrasounds with acute DVT) were compared to controls (ultrasound without acute DVT) to examine the frequency of PFCs. IRB approval was obtained and patients lacking Minnesota research authorization were excluded. Results: A total of 332,016 lower extremity venous ultrasounds were performed in 223,035 patients; 156,846 unilateral and 175,170 bilateral lower extremities exams. The mean age at ultrasound was 63.3 (SD 16.5) and 54.7% were female. Ultrasound reports were available for analysis starting in 1992 with a significant increase in the number of ultrasounds performed over the study period across the enterprise (Figure 1). Overall, acute DVT was identified in 24,179 (7.3%) of ultrasounds, and PFCs were identified in 32,427 (9.8%) of ultrasounds. The sensitivity and specificity of the NLP algorithm in the full dataset to identify acute DVT was 86.0% and 97.2%, respectfully. The sensitivity and specificity of the NLP algorithm to identify PFCs was 97.8% and 99.5%, respectively. PFCs were present in 9.3% of ultrasounds with acute DVT and 9.8% of ultrasounds without acute DVT (p=0.007), OR 0.94 (95% CI 0.90-0.98). In a multivariate logistic regression model, after adjusting for age and sex, results remained significant (aOR 0.95, 95% CI 0.91-0.995). Comparing ultrasounds before and after 2010, there was a higher percentage of PFCs and acute DVT reported after 2010 (p&lt;0.001 for both). Sensitivity analyses comparing results before or after 2010, by sex, and only in the first ultrasound performed per person, demonstrated similar results. Conclusions: PFCs are negatively associated with the presence of acute DVT on lower extremity venous Duplex ultrasound. This data does not support PFCs as a contributing or causative factor in the development of lower extremity DVT. Figure 1 Disclosures No relevant conflicts of interest to declare.

Objective To investigate 35‐year epidemiological trends in tonsillectomy and adenotonsillectomy. Study Design/Subjects Cross‐sectional survey. Subjects included all patients from birth to age 29 years who had tonsillectomy or adenotonsillectomy from 1970 to 2005. Results Study included 8106 patients (median age 8.0 years; range, 6 months to 29 years; male 3646 patients [45%]). Overall tonsillectomy incidence increased from 126 (95% confidence interval [CI], 111‐140) per 100,000 person‐years in 1970 through 1974 to 153 (95% CI, 139‐166) in 2000 through 2005. A dominant factor, adenotonsillectomy incidence rose sharply from 243 (95% CI, 223‐261) per 100,000 person‐years in 1970 through 1974 to 485 (95% CI, 462‐509) in 2000 through 2005. The indication of upper airway obstruction increased from 12 percent of patients in 1970 to 77 percent in 2005. Conclusions Epidemiological trends in tonsillectomy and adenotonsillectomy have shifted substantially. Overall numbers have increased, and surgical indications have shifted from infection to upper airway obstruction.

Background— Obesity is a risk factor for atrial fibrillation and other cardiovascular conditions. Our objective was to determine whether catheter-based ablation effectively treated atrial fibrillation in obese patients. Methods and Results— Five hundred twenty-three consecutive patients with symptomatic, medication-refractory atrial fibrillation underwent catheter ablation. Patients were grouped by body mass index (lean, <25 kg/m 2 ; overweight, 25 to 29.9 kg/m 2 ; obese, ≥30 kg/m 2 ). Outcome and quality of life were measured with a general health survey (Medical Outcomes Study 36-item Short-Form General Health Survey [SF-36]); patients were assessed before ablation and at 3 and 12 months after the procedure. Two hundred twenty-eight study patients (44%) were overweight, and 201 (38%) were obese. Twelve months after curative ablation, 72% of patients were free of atrial fibrillation without the use of antiarrhythmic agents; 84% were arrhythmia free when those receiving medication were included. Atrial fibrillation was eliminated in 75%, 72%, and 70% of the lean, overweight, and obese patients, respectively, at 12 months ( P =0.41, trend test). SF-36 scores were lower for patients with higher body mass index ( P <0.05) at baseline. SF-36 scores improved in every functional domain for all body mass index groups after ablation. The mean SF-36 total physical score increased from 59±20 at baseline to 77±19 in 12 months ( P <0.001). The total mental health score improved from 66±18 to 79±16 in 12 months ( P <0.001). Conclusions— Catheter ablation of atrial fibrillation was effective in obese patients. Coexistence of atrial fibrillation and obesity indicated lower SF-36 scores, but the improvement in quality of life was consistent across all body mass index categories.

Background— In case series, mitral regurgitation (MR) increased the risk of death after myocardial infarction (MI), yet the prevalence of MR, its incremental prognostic value over ejection fraction (EF), and its association with heart failure and death after MI in the community is not known. Methods and Results— The prevalence of MR and its association with heart failure and death were examined among 1331 patients within a geographically defined MI incidence cohort between 1988 and 1998. Echocardiography was performed within 30 days after MI in 773 patients (58%), and MR was present in 50% of cases, mild in 38%, and moderate or severe in 12%. Among patients with MR, a murmur was inconsistently detected clinically. After 4.7±3.3 years of follow-up, 109 episodes of heart failure and 335 deaths occurred. There was a graded positive association between the presence and severity of MR and heart failure or death. Moderate or severe MR was associated with a large increase in the risk of heart failure (relative risk 3.44, 95% CI 1.74 to 6.82, P <0.001) and death (relative risk 1.55, 95% CI 1.08 to 2.22, P =0.019) among 30-day survivors independent of age, gender, EF, and Killip class. Conclusions— In the community, MR is frequent and often silent after MI. It carries information to predict heart failure or death among 30-day survivors independently of age, gender, EF, and Killip class. These findings, which are applicable to a large community-based MI cohort, suggest that the assessment of MR should be included in post-MI risk stratification.