Classification of anatomic patterns of peripheral artery disease with automated machine learning (AutoML).

AIM: The aim of this study was to investigate the potential of novel automated machine learning (AutoML) in vascular medicine by developing a discriminative artificial intelligence (AI) model for the classification of anatomical patterns of peripheral artery disease (PAD). MATERIAL AND METHODS: Random open-source angiograms of lower limbs were collected using a web-indexed search. An experienced researcher in vascular medicine labelled the angiograms according to the most applicable grade of femoropopliteal disease in the Global Limb Anatomic Staging System (GLASS). An AutoML model was trained using the Vertex AI (Google Cloud) platform to classify the angiograms according to the GLASS grade with a multi-label algorithm. Following deployment, we conducted a test using 25 random angiograms (five from each GLASS grade). Model tuning through incremental training by introducing new angiograms was executed to the limit of the allocated quota following the initial evaluation to determine its effect on the software's performance. RESULTS: We collected 323 angiograms to create the AutoML model. Among these, 80 angiograms were labelled as grade 0 of femoropopliteal disease in GLASS, 114 as grade 1, 34 as grade 2, 25 as grade 3 and 70 as grade 4. After 4.5 h of training, the AI model was deployed. The AI self-assessed average precision was 0.77 (0 is minimal and 1 is maximal). During the testing phase, the AI model successfully determined the GLASS grade in 100% of the cases. The agreement with the researcher was almost perfect with the number of observed agreements being 22 (88%), Kappa = 0.85 (95% CI 0.69-1.0). The best results were achieved in predicting GLASS grade 0 and grade 4 (initial precision: 0.76 and 0.84). However, the AI model exhibited poorer results in classifying GLASS grade 3 (initial precision: 0.2) compared to other grades. Disagreements between the AI and the researcher were associated with the low resolution of the test images. Incremental training expanded the initial dataset by 23% to a total of 417 images, which improved the model's average precision by 11% to 0.86. CONCLUSION: After a brief training period with a limited dataset, AutoML has demonstrated its potential in identifying and classifying the anatomical patterns of PAD, operating unhindered by the factors that can affect human analysts, such as fatigue or lack of experience. This technology bears the potential to revolutionize outcome prediction and standardize evidence-based revascularization strategies for patients with PAD, leveraging its adaptability and ability to continuously improve with additional data. The pursuit of further research in AutoML within the field of vascular medicine is both promising and warranted. However, it necessitates additional financial support to realize its full potential.

Compliance assessment regarding the PVC management on normal wards of a university hospital.

Objective: The risk of peripheral venous catheter (PVC) infections in inpatients is often underestimated, even if it is lower than that for central venous catheters. Guidelines for the prevention of PVC-associated infections describe the evidence-based management of PVCs. The aims of this study were the development of standardized methods for compliance assessment regarding PVC management and the evaluation of self-reported knowledge and implementations among healthcare providers regarding PVC care. Method: We developed a checklist based on the recommendation of the Commission of Hospital Hygiene and Infection Prevention at the Robert Koch Institute (KRINKO) Berlin for the standardized evaluation of PVC management. The following parameters were collected and evaluated: condition of the puncture site, condition of the bandage, presence of an extension set, presence of a plug, and documentation. The checklist was applied in 14 normal wards in 2019. After feedback of the ward staff on the results, it was applied again in 2020 in the same wards. For retrospective data analysis, we used a newly developed PVC-quality index. After the second evaluation in 2020, we carried out an anonymous survey among the healthcare providers. Results: The evaluation of 627 indwelling PVCs showed a significant increase in compliance related to the presence of an extension set (p=0.049) and documentation (p<0.001) in the 2nd year. The quality index increased in 12 out of 14 wards. The participants of the survey were aware of the in-house standard "Prevention of vascular catheter-associated infections", with a mean score of 4.98 on a Likert scale (1=not aware, 7=completely aware). The main barrier to implementation of the preventive measures was the time factor. Survey participants were more aware of PVC placement than PVC care. Conclusion: The PVC quality index is a valuable tool for the assessment of compliance regarding PVC management in daily practice. Feedback from the ward staff on the results of compliance assessment improves PVC management, but the outcome is very heterogeneous.

Cardiac response to changes in central venous sonomorphology: Mechanism of impaired heart relaxation in patients with varicose veins of lower extremities.

AIM: This study examines the interactions between central venous and cardiac sonomorphologies to explain the mechanism of impaired heart relaxation in patients with varicose veins of lower extremities. MATERIAL AND METHODS: Part 1: We performed retrospective analysis of influences of inferior vena cava (IVC) diameters on tricuspid and mitral inflow and annular velocities in 64 patients with primary varicose veins. Part 2: We compared IVC diameters and IVC collapsibility index (IVC CI) in patients with varicose veins with normal values. RESULTS: We found a significant positive influence of an increase in maximal expiratory and minimal inspiratory IVC diameters on mitral and tricuspid early annular velocities or e'-waves (p-values < .008), inflow velocities or E-waves (p-values < .05) and early to late inflow E/A ratios (p-values < .01). Less influenced by the changes in venous biometry (p-values > .05) were late mitral and tricuspid inflow, annular and systolic velocities (A-, a'-, s-waves).Expiratory and inspiratory diameters in patients with varicose veins were significantly smaller (p-values < .05 expiratory; < .0001 inspiratory), and IVC CI was significantly higher (p-values < .0001) than the normal values. CONCLUSION: Chronic venous disease impairs the function and the morphology of the entire inferior caval system and the heart. Impaired relaxation of the heart in patients with varicose veins is the result of two factors: (1) impaired venous return resulting in the low central venous pressure and the low early diastolic cardiac inflow; (2) structural changes in the heart resulting in the compensatory increased late diastolic cardiac inflow. Increase in central venous pressure (IVC diameters) and early diastolic cardiac inflow (E-waves), accompanied by unchanged myocardial response (e'-wave) can serve as marker for return to normal physiology.

Earth's gravity field and prevalence of varicose veins and chronic venous disease: Systematic review.

AIM: This study examines the influence of Earth's gravity field on the prevalence of varicose veins in geophysical area. MATERIAL AND METHODS: We performed a systematic review (OVID and Google Scholar) of studies focusing on prevalence of varicose veins to determine the influence of Earth's gravity field-GRACE GGM05S gravity model-on the disease prevalence. PROSPERO: CRD42021279513. RESULTS: 81 studies met inclusion and quality criteria. Areas with stronger gravity have significantly higher prevalence of varicose veins with adjustment for age, gender and body mass index (BMI) (p-values < 0.02). Adjusted for age, prevalence of varicose veins in areas with gravity field +20 mGal and more is 1.37 time higher than in areas with gravity field less than +20 mGal, p-value 0.005 (95% CI: -12.5 to -2.4): mean disease prevalence for gravity field +20 mGal and more-27.5% (mean age, 40.1 years; mean gravity field, +27.1 mGal; 63.9% females, 37 studies, 123,164 participants) vs mean disease prevalence for gravity field less than +20 mGal - 20.1% (mean age, 42.2 years; mean gravity field, +5.7 mGal; 56.8% females, 44 studies, 205,925 participants). Older age is the main risk factor for varicose veins (p-values < 0.005). Female gender and high BMI are insignificantly associated with high prevalence of varicose veins (p-values > 0.4 for gender, p-values > 0.2 for BMI). CONCLUSION: Stronger gravity field is significantly associated with higher prevalence of varicose veins-risk factor. The potential mechanism of this phenomenon is that high gravity field alters systemic venous return, pooling blood and fluid in the peripheral, gravity-dependent regions of the body in upright humans constantly living in the defined geophysical area.

Respiratory changes in biometry of suprarenal inferior vena cava in patients with varicose veins of lower extremities.

AIM: This study examines respiratory biometry of inferior vena cava in patients with varicose veins of lower extremities. MATERIAL AND METHODS: We performed retrospective analysis of clinical and ultrasound data of 67 patients with primary varicose veins. RESULTS: The largest expiratory (mean 16.2 mm, p-value 0.09) and inspiratory (mean 8.2 mm, p-value 0.02) inferior vena cava diameters were in C3 Clinical Etiological Anatomical Pathophysiological clinical class; the smallest expiratory diameters (mean 13.1 mm, p-value 0.5) were in C6 class; the smallest inspiratory diameters (mean 4.6 mm, intercept) were in C2 class. C2 class was associated with highest inferior vena cava collapsibility index (mean 68.2%, intercept); C6 class was associated with lowest collapsibility index (mean 48.3%, p-value 0.04).Recurrent varices in comparison with previously untreated were associated with smaller inspiratory diameters of inferior vena cava (mean 4.4 mm, p-value 0.005), smaller expiratory diameters (mean 13.4 mm, p-value 0.06) and higher collapsibility index (mean 68.5%, p-value 0.005). Patients with recurrent and bilateral varicose veins had identical respiratory biometry of inferior vena cava.Older age was associated with smaller inferior vena cava diameters (p-value <0.01). CONCLUSION: Clinical presentation of varicose veins is associated with different respiratory biometry of suprarenal inferior vena cava. C6 clinical class in comparison with C2 clinical class is associated with lower central venous compliance possible due to the narrowing of inferior vena cava. Smaller inferior vena cava diameters and higher collapsibility index in recurrent subgroup in comparison with previously untreated can be a sign of the significantly altered pressure gradient between the systemic capillaries and the right heart and impaired peripheral venous return. Narrowing of inferior vena cava with age can be a sign of more profound changes in systemic venous return with age in patients with varicose veins in comparison to those without chronic venous disease.

Association between right heart diastolic function and clinical presentation of chronic venous disease and primary varicose veins of lower extremities.

AIM: This study examines the association between right heart diastolic function and clinical presentation of chronic venous disease and primary varicose veins of lower extremities. MATERIAL AND METHODS: We performed retrospective analysis of clinical, peripheral venous Duplex and tricuspid Doppler (early diastolic inflow E-wave, late diastolic inflow A-wave, ratios of E to A waves, early diastolic annular e'-wave, late diastolic annular a'-wave and systolic annular s'- wave) data of 85 patients, 133 legs with primary varicose veins. RESULTS: We found following significant (p-value < 0.05) associations between tricuspid Doppler and clinical presentation of chronic venous disease and primary varicose veins: Clinical Etiological Anatomical Pathophysiological clinical class influenced late diastolic inflow velocities (C6 class A-wave +11.2 cm/s or +27% in comparison with C2), late diastolic annular velocities (C6 class a'-wave +3.3 cm/s or +22% in comparison with C2), systolic annular velocities (C6 class s'-wave +3.7 cm/s or +27% in comparison with C2) and E/A ratios (C6 class E/A ratio -0.22 or -21% in comparison with C2). Recurrent varices in comparison with previously untreated are associated with significantly lower late diastolic inflow velocities (A-wave -4.4 cm/s or -9%) and preserved E/A ratios. Age significantly influenced tricuspid Doppler (E-, A-, e'-, a'-waves and E/A ratios) in patients with chronic venous disease. CONCLUSION: Clinical presentation of primary varicose veins and chronic venous disease can be associated with the right heart diastolic function: C6 Clinical Etiological Anatomical Pathophysiological class in comparison with C2 is associated with increased right ventricular filling and impaired ventricular relaxation - right heart diastolic dysfunction; recurrent varices in comparison with previously untreated are associated with reduced right ventricular filling and preserved right heart diastolic function. Older age is the most important risk factor for varicose veins and chronic venous disease possible due to significant changes in right ventricular filling and in right heart diastolic function with age.