Machine learning-based prediction of abdominal aortic aneurysms for individualized patient care.

OBJECTIVE: The United States Preventative Services Task Force guidelines for screening for abdominal aortic aneurysms (AAA) are broad and exclude many at risk groups. We analyzed a large AAA screening database to examine the utility of a novel machine learning (ML) model for predicting individual risk of AAA. METHODS: We created a ML model to predict the presence of AAAs (>3 cm) from the database of a national nonprofit screening organization (AAAneurysm Outreach). Participants self-reported demographics and comorbidities. The model is a two-layered feed-forward shallow network. The ML model then generated AAA probability based on patient characteristics. We evaluated graphs to determine significant factors, and then compared those graphs with a traditional logistic regression model. RESULTS: We analyzed a cohort of 10,033 patients with an AAA prevalence of 2.74%. Consistent with logistic regression analysis, the ML model identified the following predictors of AAA: Caucasian race, male gender, advancing age, and recent or past smoker with recent smoker having a more profound affect (P < .05). Interestingly, the ML model showed body mass index (BMI) was associated with likelihood of AAAs, especially for younger females. The ML model also identified a higher than predicted risk of AAA in several groups, including female nonsmokers with cardiac disease, female diabetics, those with a family history of AAA, and those with hypertension or hyperlipidemia at older ages. An elevated BMI conveyed a higher than expected risk in male smokers and all females. The ML model also identified a complex relationship of both diabetes mellitus and hyperlipidemia with gender. Family history of AAA was a more important risk factor in the ML model for both men and women too. CONCLUSIONS: We successfully developed an ML model based on an AAA screening database that unveils a complex relationship between AAA prevalence and many risk factors, including BMI. The model also highlights the need to expand AAA screening efforts in women. Using ML models in the clinical setting has the potential to deliver precise, individualized screening recommendations.

Evaluating the prevalence of abdominal aortic aneurysms in the United States through a national screening database.

OBJECTIVE: The U.S. Preventative Services Task Force guidelines for abdominal aortic aneurysm (AAA) screening are based mainly on studies of older Caucasian males from non-U.S. POPULATIONS: This study was designed to analyze the findings of a large, all-inclusive AAA screening program in the United States. METHODS: Screening events were held nationally by a U.S. nonprofit organization between 2001 and 2017. AAA screening was offered regardless of risk profile. Participants filled out a demographics form with known comorbidities. Significant risk factors were determined using logistic regression with backward stepwise variable selection. Odds ratios (OR) are reported with 95% confidence intervals (CIs). RESULTS: A total of 9457 screened participants (47% male) were analyzed. The mean age was 67 ± 9 years with 40.8% between 65 and 75 years old. Most participants were Caucasian (83.4%), followed by African American (13.1%). Screened risk factors included hypertension (58.1%), hyperlipidemia (54.9%), smoking (52.0%), cardiac disease (29.2%), diabetes mellitus (18.4%), a family history of AAA (22.4%) or brain aneurysms (8.6%), and body mass index (26.9 ± 5.28). Overall, 267 participants (2.82%) were found to have an AAA (>3 cm). Those ages 65 to 75 had a prevalence of 2.98%. In a fully adjusted, multivariate logistic regression, there was an increased risk of AAA in males (OR, 3.24; 95% CI, 2.39-4.40), current smokers (OR, 3.28; 95% CI, 2.36-4.54), previous smokers (OR, 1.86; 95% CI, 1.41-2.47), cardiac disease (OR, 1.30; 95% CI, 1.01-1.68), family history of AAA (OR, 1.60; 95% CI, 1.20-2.14), and advancing age (P < .0001). Female ever smokers 65 to 75 years old had a prevalence of 1.7%. Male smokers 45 to 54 and 55 to 64 years old had a prevalence of 3.37% and 4.43%, respectively. There was an increased risk of AAA in females with morbid obesity (OR, 5.54; 95% CI, 1.34-22.83 in never smokers and OR, 5.61; 95% CI, 1.04-30.15 in smokers), female smokers with hypertension (OR, 3.22; 95% CI, 1.21-8.58), males with cardiac disease (OR, 2.06; 95% CI, 1.08-3.90 in never smokers and OR, 1.48; 95% CI, 1.05-2.09), male smokers with a family history of AAA (OR, 1.69; 95% CI, 1.61-2.46), and current smokers (OR, 6.33; 95% CI, 2.62-15.24 for females and OR, 2.50; 95% CI, 1.70-3.65 for males). CONCLUSIONS: This study shows that there remain high-risk groups outside the current guidelines that would likely benefit from AAA screening. Risk factors for AAA include male gender, smoking, cardiac disease, family history of AAA, and advancing age. The most significant risk factor is current smoking status.

Internal Carotid Artery Agenesis in a Trauma Patient with Ipsilateral Neck GSWs.

A 38-year-old man presented to the emergency room in the trauma bay for multiple ballistic injuries to the right neck. He was hemodynamically stable, protecting his airway, and neurologically intact. Computed tomography angiography (CTA) revealed absent filling the right internal carotid artery from its origin to the circle of Willis, which was intact, as well as absent petrous carotid canal on the right. The patient was diagnosed with right internal carotid artery (ICA) agenesis and discharged in several days. This report demonstrates the importance of an in-depth knowledge of vascular embryology and anatomy. The patient has agreed to have images and case details published.

Health literacy and abdominal aortic aneurysms.

OBJECTIVE: Little is known about the public's knowledge of abdominal aortic aneurysms (AAA). Although preventive screening is available, millions of Americans remain unaware of their risk. Improved health literacy has been associated with increased screening and improvement in health outcomes. This study assessed the level of AAA literacy among respondents who participated in a free AAA screening event. METHODS: Thirteen key words used by vascular surgeons to describe the risk, diagnosis, and treatment options for AAA were extracted from the screening tool used by the nation's largest provider of free AAA diagnostic services, AAAneurysm Outreach. The National Institutes of Health recommends readability of patient education materials to be at the sixth-grade level, but a readability analysis of these words placed them at a grade level of 14.6. A self-administrated questionnaire was developed that allowed respondents to compare each of the extracted words with a definitionally correct or incorrect word that reflected a sixth-grade readability score. These scores were then compared with the available demographics. RESULTS: There were 570 completed questionnaires. Of the participants, 57.6% were female, 61.4% were 60 and above, and 32.6% were veterans. The average number of correct answers was 9.31 out of 13 (72% correct). Only 4.7% answered all questions correctly, with 29.1% missing five or more answers. The most frequently missed words were asymptomatic, screening, and cholesterol (56.5%, 44%, and 41.4% incorrect, respectively). The most frequently known terms were abdominal, diagnosis, and genetic (96%, 95.3%, and 91.9% correct, respectively). The remaining words fell between these extremes. Those aged 60 and above scored significantly lower than younger respondents (P < .0001). A post hoc power analysis indicated that the power to detect the obtained effects of age at the .05 level was greater than 0.95. Gender and veteran status did not produce any significant differences. CONCLUSIONS: These data suggest an important communication gap between the words used by clinicians to describe the risks, diagnostic results, and treatment options of AAA and the targeted at-risk population, especially those 60 years and older.

Statistical and machine learning methodology for abdominal aortic aneurysm prediction from ultrasound screenings.

A method of analysis of a database of patients (n = 10 329) screened for an abdominal aortic aneurysm (AAA) is presented. Self-reported height, weight, age, gender, ethnicity, and parameters "Heart Problems," "Hypertension," "High Cholesterol," "Diabetes Mellitus," "Smoker Past 2 Years," "Ever Smoked?," "Family History AAA," and "Family History Brain Aneurysm" were provided. Incidence of a AAA (defined as 3 cm diameter) was calculated as a function of age and body mass index (BMI) of greater than or less than a BMI 25 for various patient groups. Age was grouped into one of three categories in 15-year intervals (35-50 years, 50-65 years, and 65 to 80 years). Most patients were Caucasian (n = 8575) and the largest group of patients with a AAA was the Caucasian male (198 of 279 total detected AAAs). A machine learning algorithm was written, with learning inputs from the acquired patient database. Of all groups, Caucasian males were found to have the highest incidence of AAA, with males in general higher than females. Smoking within the past two years was highly associated with AAA incidence, and a past history of smoking to a lesser extent. The incidence of AAA increased with age. When dividing groups into two cohorts by a BMI of 25, generally middle-aged patients with a BMI > 25 had a higher incidence of a AAA. However, in general, the older age group with a BMI < 25 had a higher incidence of AAA. The addition of machine learning allows one to note the effect of an input keeping other input parameters constant. This helps identify a parameter that may be an independent predictor of a particular outcome. When using BMI as the single changing input, an increasing BMI was associated with an increased probability of a AAA, most significantly in middle-aged patients, and then narrowing to similar probabilities in older age. This AAA screening program is ongoing. As data continues to be collected with particularly those patient groups presently underrepresented, questions as to an association of AAA with BMI as a function of age, and also an improvement in machine learning algorithm accuracy for various patient populations will continue.

Surgical and endovascular central venous reconstruction combined with thoracic outlet decompression in highly symptomatic patients.

BACKGROUND: Subclavian vein stenosis or occlusion at the thoracic outlet is a problem associated with certain anatomic and environmental stresses (venous thoracic outlet syndrome [VTOS]), the presence of central venous catheters, and the high flows associated with arteriovenous (AV) access in the limb. We describe our experience with open and endovascular techniques for restoring patency in highly symptomatic patients. METHODS: A prospectively collected database of patients was queried for patients treated for central venous obstructive disease in the setting of highly symptomatic VTOS and ipsilateral AV access from October 2011 to August 2016. RESULTS: During the study period, 54 procedures were performed in 53 patients (68% male; mean age, 50.1 years). Indications for operation were venous outflow obstruction in patients with conventional VTOS (n = 19) or costoclavicular junction stenosis associated with ipsilateral dialysis access (n = 34). All patients had significant symptoms of swelling or pain. Eight patients underwent on-table pharmacomechanical thrombolysis for acute occlusion. All patients underwent costoclavicular junction decompression, 48 by infraclavicular first rib resection and 5 by claviculectomy; 6 patients underwent sternoclavicular rotation (Molina procedure) in addition to rib resection for further exposure. Surgical reconstruction of the vein was employed in 18 patients (33%); 9 underwent interposition grafting, 1 had jugular turndown, and 8 had patch angioplasty. The one patient undergoing two procedures suffered acute occlusion after patch repair followed by jugular turndown. Four patients underwent surgical reconstruction after thrombolysis. Endovascular procedures were performed in 36 patients (67%); 23 underwent venous angioplasty alone, and 13, all with hemodialysis access-associated stenosis, underwent stenting. Mean operative time was 135 (±63.5) minutes, and mean estimated blood loss was 238 (±261) mL. Median length of stay was 4 days. Perioperative complications were noted in 14 (26.4%) patients, including wound complications (n = 6), cardiac complications (n = 4), reocclusion (n = 3), and hemothorax requiring chest tube placement (n = 1) in a patient undergoing on-table thrombolysis. Mean follow-up was 13.6 (0.6-58.5) months. Initial clinical symptom relief was experienced in 100% of patients at the time of hospital discharge. During follow-up, 5 (9.4%) patients developed recurrent symptoms, 6 (11.3%) had reocclusion of the central system, and 16 (30.2%) required reintervention for restenosis, all but 2 in patients with ipsilateral hemodialysis access. Mean time to reintervention was 134 (±285) days. CONCLUSIONS: Given our decision-making threshold, both open and endovascular procedures are associated with relatively low morbidity and high efficacy for treatment of central venous occlusion in both symptomatic VTOS and AV access-associated subclavian vein disease. Restenosis is common in patients with a patent ipsilateral hemodialysis access.

Obesity-related dysregulation of the tryptophan-kynurenine metabolism: role of age and parameters of the metabolic syndrome.

OBJECTIVE: Obesity-related immune mediated systemic inflammation was associated with the development of the metabolic syndrome by induction of the tryptophan (TRP)-kynurenine (KYN) pathway. The study aimed to assess whether this holds true across the lifespan from juvenility to adulthood. DESIGN AND METHODS: Five hundred twenty-seven participants aged between 10 and 65 years were analyzed. Standard anthropometric measures, carotid ultrasound, and laboratory analysis including interleukin-6, ultra-sensitive C-reactive protein, lipids, glucose metabolism, neopterin, TRP, KYN levels, and the KYN/TRP ratio were performed. RESULTS: Overweight/obese (ow/ob) adults had significantly increased KYN serum levels and a significantly increased KYN/TRP ratio. In sharp contrast, ow/ob juvenile males aged ≤18 years showed decreased, females similar KYN and KYN/TRP ratio in comparison to their control counterparts. Also, adult ow/ob subjects with metabolic syndrome showed markedly increased KYN/TRP ratios contrary to decreased KYN/TRP ratios in ow/ob juveniles. Abdominal fat content, characterized by age normalized waist circumference, and not body mass index, had the strongest effect for an increase of the KYN/TRP ratio in adults. CONCLUSIONS: TRP metabolism and obesity-related immune mediated inflammation differs markedly between juveniles and adults. While childhood obesity seems to be dominated by a Th2-driven activation, an accelerated production of Th1-type cytokines may pave the way for later atherosclerotic endpoints.

Interleukin 10-coated nanoparticle systems compared for molecular imaging of atherosclerotic lesions.

Atherosclerosis (AS) is one of the leading causes of mortality in high-income countries. Early diagnosis of vulnerable atherosclerotic lesions is one of the biggest challenges currently facing cardiovascular medicine. The present study focuses on developing targeted nanoparticles (NPs) in order to improve the detection of vulnerable atherosclerotic-plaques. Various biomarkers involved in the pathogenesis of atherosclerotic-plaques have been identified and one of these promising candidates for diagnostic targeting is interleukin 10 (IL10). IL10 has been shown to be a key anti-inflammatory responding cytokine in the early stages of atherogenesis, and has already been used for therapeutic interventions in humans and mice. IL10, the targeting sequence, was coupled to two different types of NPs: protamine-oligonucleotide NPs (proticles) and sterically stabilized liposomes in order to address the question of whether the recognition and detection of atherosclerotic-lesions is primarily determined by the targeting sequence itself, or whether it depends on the NP carrier system to which the biomarker is coupled. Each IL10-targeted NP was assessed based on its sensitivity and selectivity toward characterizing atherosclerotic-plaque lesions using an apolipoprotein E-deficient mouse as the model of atherosclerosis. Aortas from apolipoprotein E-deficient mice fed a high fat diet, were stained with either fluorescence-labeled IL10 or IL10-coupled NPs. Ex vivo imaging was performed using confocal laser-scanning microscopy. We found that IL10-targeted proticles generated a stronger signal by accumulating at the surface of atherosclerotic-plaques, while IL10-targeted, sterically stabilized liposomes showed a staining pattern deeper in the plaque compared to the fluorescence-labeled IL10 alone. Our results point to a promising route for enhanced in vivo imaging using IL10-targeted NPs. NPs allow a higher payload of signal emitting molecules to be delivered to the atherosclerotic-plaques, thus improving signal detection. Importantly, this allows for the opportunity to visualize different areas within the plaque scenario, depending on the nature of the applied nanocarrier.