MACHINE LEARNING FOR PREDICTING HEMODYNAMIC DETERIORATION OF PATIENTS WITH INTERMEDIATE-RISK PULMONARY EMBOLISM IN INTENSIVE CARE UNIT.

ABSTRACT: Background: Intermediate-risk pulmonary embolism (PE) patients in the intensive care unit (ICU) are at a higher risk of hemodynamic deterioration than those in the general ward. This study aimed to construct a machine learning (ML) model to accurately identify the tendency for hemodynamic deterioration in the ICU patients with intermediate-risk PE. Method: A total of 704 intermediate-risk PE patients from the MIMIC-IV database were retrospectively collected. The primary outcome was defined as hemodynamic deterioration occurring within 30 days after admission to ICU. Four ML algorithms were used to construct models on the basis of all variables from MIMIC IV database with missing values less than 20%. The extreme gradient boosting (XGBoost) model was further simplified for clinical application. The performance of the ML models was evaluated by using the receiver operating characteristic curve, calibration plots, and decision curve analysis. Predictive performance of simplified XGBoost was compared with the simplified Pulmonary Embolism Severity Index score. SHapley Additive explanation (SHAP) was performed on a simplified XGBoost model to calculate the contribution and impact of each feature on the predicted outcome and presents it visually. Results: Among the 704 intermediate-risk PE patients included in this study, 120 patients experienced hemodynamic deterioration within 30 days after admission to the ICU. Simplified XGBoost model demonstrated the best predictive performance with an area under the curve of 0.866 (95% confidence interval, 0.800-0.925), and after recalibrated by isotonic regression, the area under the curve improved to 0.885 (95% confidence interval, 0.822-0.935). Based on the simplified XGBoost model, a web app was developed to identify the tendency for hemodynamic deterioration in ICU patients with intermediate-risk PE. Conclusion: A simplified XGBoost model can accurately predict the occurrence of hemodynamic deterioration for intermediate-risk PE patients in the ICU, assisting clinical workers in providing more personalized management for PE patients in the ICU.

The effect of central lymph node dissection on the prognosis of cN0 papillary thyroid microcarcinoma: a mid-term follow-up study.

BACKGROUND: To investigate the effect of central lymph node dissection on the prognosis of patients with papillary thyroid microcarcinoma (PTMC) without clinical lymph node metastasis (cN0). METHODS: According to the inclusion and exclusion criteria, 462 patients with cN0 PTMC underwent surgery in the Second Department of General Surgery, Zhongshan City People's Hospital from January 1, 2007, to June 31, 2017. They were divided into two groups: the undissection group (262 cases) and the dissection group (170 cases). A comparison was made between the two groups in terms of postoperative complications, recurrences, metastases, etc., as well aslymph node metastasis risk factors in the central region of cN0 PTMC. RESULTS: There was no lymphatic leakage or death in all patients after the operation. In the dissection group, 64 cases (37.6%) of central lymph node metastasis were found after the postoperative pathological examination. The undissection group was followed up for (92 ± 28.7) months, and the dissection group was followed up for (86 ± 25.4) months (t=-2.165, P = 0.031). In the two groups, there were no lung metastases, bone metastases, or other distant metastases during the follow-up period. In the undissection group, there were 7 cases, while in the dissection group, there were just 2. Recurrence rates between the two groups did not differ significantly (χ2 = 0.126, P = 0.169); Similarly, disease-free survival curves did not differ significantly (χ2 = 2.565, P = 0.708). Hypoparathyroidism and Hypocalcemia also had no difference between the group. In comparison to the undissection group, the capsular invasion rate (P = 0.026), calcification rate(P < 0.001) incidence of postoperative hoarseness (P = 0.017), and hand and foot numbness rate (P < 0.001) were all considerably greater in the dissection group. Multivariate research revealed that capsular invasion (OR = 9.42, P = 0.002), multifocal (OR = 24.57, P < 0.001), and tumor diameter > 5 mm (OR = 5.46, P = 0.019) were the independent risk factors for central lymph node metastasis in cN0 PTMC. CONCLUSIONS: Thyroidectomy alone is safe for cN0 PTMC, but longer-term follow-up is still required for changes in central lymph nodes. For cN0 PTMC patients with tumor diameter > 5 mm, multifocal, and capsular invasion, central lymph node metastasis is more likely to occur. Comprehensive evaluation and individualized and precise treatment are essential.

Identification of a 3-Gene Prognostic Index for Papillary Thyroid Carcinoma.

The accurate determination of the risk of cancer recurrence is a critical unmet need in managing thyroid cancer (TC). Although numerous studies have successfully demonstrated the use of high throughput molecular diagnostics in TC prediction, it has not been successfully applied in routine clinical use, particularly in Chinese patients. In our study, we objective to screen for characteristic genes specific to PTC and establish an accurate model for diagnosis and prognostic evaluation of PTC. We screen the differentially expressed genes by Python 3.6 in The Cancer Genome Atlas (TCGA) database. We discovered a three-gene signature Gap junction protein beta 4 (GJB4), Ripply transcriptional repressor 3 (RIPPLY3), and Adrenoceptor alpha 1B (ADRA1B) that had a statistically significant difference. Then we used Gene Expression Omnibus (GEO) database to establish a diagnostic and prognostic model to verify the three-gene signature. For experimental validation, immunohistochemistry in tissue microarrays showed that thyroid samples' proteins expressed by this three-gene are differentially expressed. Our protocol discovered a robust three-gene signature that can distinguish prognosis, which will have daily clinical application.

The Distance From the Primary Intimal Tear to the Left Subclavian Artery Predicts Thoracic Aortic Enlargement After Thoracic Endovascular Aortic Repair of Type B Aortic Dissection: A Retrospective Cohort Study.

PURPOSE: The purpose of this study was to evaluate the association between the distance from the primary intimal tear (PIT) to the left subclavian artery (LSA) (PIT-LSA distance) and the risk of aortic enlargement after thoracic endovascular aortic repair (TEVAR). METHODS: This is a retrospective cohort study. A total of 228 patients were reviewed from the database of the Registry Of type B aortic dissection with the Utility of STent graft (ROBUST) study performed from January 1, 2011, to December 31, 2016. Of them, 196 patients were eligible for analysis. The PIT-LSA distance was defined as the length from the distal edge of the LSA orifice to the proximal edge of the PIT along the centerline of the true lumen. According to the border between zone 3 and zone 4 of the Ishimaru classification, patients were divided into group A (n = 117, PIT-LSA distance ≤ 2 cm) and group B (n = 79, PIT-LSA distance > 2 cm). Thoracic aortic enlargement (TAE) was defined as a thoracic aortic volume increase of ≥20%. Multivariate Cox regression was used to estimate the association between the PIT-LSA distance and risk of TAE after TEVAR. RESULTS: The mean age was 52.3 ± 11.6 years, and 88.8% of patients were male. There were no significant differences between groups in demographic and baseline characteristics. The PIT-LSA distance was 1.1 cm (range, -1.6 to 2.0 cm) in group A, and 2.9 cm (range, 2.1-12.6 cm) in group B. TAE occurred in 27 patients in group A, and 6 in group B. The mean follow-up was 12.4 months (range, 0.10-83.1 months) in group A, and 12.63 months (range, 0.10-82.77 months) in group B. The cumulative 12- and 24-month rates of freedom from TAE were 79.0% and 71.3% in group A, versus 92.5% and 92.5% in group B, respectively. Multivariate Cox regression analysis revealed that the PIT-LSA distance was an independent predictor of TAE after TEVAR (adjusted hazard ratio, 0.66; 95% confidence interval, 0.48-0.90; p = 0.009). CONCLUSION: Patients with a more proximal PIT location have a higher incidence of thoracic aortic enlargement after TEVAR. The location of the PIT in relation to the LSA can be used to identify patients who need closed surveillance after TEVAR or early preemptive intervention.

Managing Emergent Surgery for Ruptured Abdominal Aortic Aneurysm during the COVID-19 Pandemic.

BACKGROUND: Coronavirus disease 2019 (COVID-19) has become a global pandemic which may compromise the management of vascular emergencies. An uncompromised treatment for ruptured abdominal aortic aneurysm (rAAA) during such a health crisis represents a challenge. This study aimed to demonstrate the treatment outcomes of rAAA and the perioperative prevention of cross-infection under the COVID-19 pandemic. METHODS: In cases of rAAA during the pandemic, a perioperative workflow was applied to expedite coronavirus testing and avoid pre-operative delay, combined with a strategy for preventing cross-infection. Data of rAAA treated in 11 vascular centers between January-March 2020 collected retrospectively were compared to the corresponding period in 2018 and 2019. RESULTS: Eight, 12, and 14 rAAA patients were treated in 11 centers in January-March 2018, 2019, and 2020, respectively. An increased portion were treated at local hospitals with a comparable outcome compared with large centers in Guangzhou. With EVAR-first strategy, 85.7% patients with rAAA in 2020 underwent endovascular repair, similar to that in 2018 and 2019. The surgical outcomes during the pandemic were not inferior to that in 2018 and 2019. The average length of ICU stay was 1.8 ± 3.4 days in 2020, tending to be shorter than that in 2018 and 2019, whereas the length of hospital stay was similar among 3 years. The in-hospital mortality of 2018, 2019, and 2020 was 37.5%, 25.0%, and 14.3%, respectively. Three patients undergoing emergent surgeries were suspected of COVID-19, though turned out to be negative after surgery. CONCLUSIONS: Our experience for emergency management of rAAA and infection prevention for healthcare providers is effective in optimizing emergent surgical outcomes during the COVID-19 pandemic.

Prioritizing Susceptible Genes for Thyroid Cancer Based on Gene Interaction Network.

Thyroid cancer ranks second in the incidence rate of endocrine malignant cancer. Thyroid cancer is usually asymptomatic at the initial stage, which makes patients easily miss the early treatment time. Combining genetic testing with imaging can greatly improve the diagnostic efficiency of thyroid cancer. Researchers have discovered many genes related to thyroid cancer. However, the effects of these genes on thyroid cancer are different. We hypothesize that there is a stronger interaction between the core genes that cause thyroid cancer. Based on this hypothesis, we constructed an interaction network of thyroid cancer-related genes. We traversed the network through random walks, and sorted thyroid cancer-related genes through ADNN which is fusion of Adaboost and deep neural network (DNN). In addition, we discovered more thyroid cancer-related genes by ADNN. In order to verify the accuracy of ADNN, we conducted a fivefold cross-validation. ADNN achieved AUC of 0.85 and AUPR of 0.81, which are more accurate than other methods.

A S-Shaped Association of Distal Oversizing Ratio with Distal Stent-Graft-Induced New Entry Following Thoracic Endovascular Aortic Repair for Stanford B Aortic Dissection.

BACKGROUND: This study was performed to evaluate the nonlinear association of the distal oversizing ratio with distal stent-graft-induced new entry (dSINE) following thoracic endovascular aortic repair (TEVAR) for type B aortic dissection (TBAD) and to find the optimal value of the distal oversizing ratio for prevention strategy of dSINE. METHODS: Total of 177 patients who underwent TEVAR for TBAD from the Registry Of type B aortic dissection with the Utility of STent graft were retrospectively investigated. Patients were stratified into two groups on the median distal oversizing ratio: lower group (≤16%, n = 88) and higher group (>16%, n = 89). The Kaplan-Meier method was used to estimate the cumulative incidence of dSINE. The multivariate Cox proportional hazards model was used to identify the association of the distal oversizing ratio with dSINE. Restricted cubic smoothing spline plots and two-piecewise regression were used to analyze the possible nonlinear association. RESULTS: Eleven patients developed dSINE (6.21%) during the median follow-up time of 12.37 months (interquartile range, 8.07-18.17 months). An S-shaped association of the distal oversizing ratio with dSINE was identified. When the distal oversizing ratio was ≤40%, the risk of dSINE increased with distal oversizing ratio, and the risk of dSINE was highest when the distal oversizing ratio reached 40% (adjusted HR, 1.09; 95% CI, 1.02-1.17; P = 0.011). A larger distal oversizing ratio over 40% did not generate a greater risk of dSINE (adjusted HR, 0.95; 95% CI, 0.87-1.05; P = 0.455). CONCLUSIONS: This study substantiated previous findings that the dSINE was associated with the increasing distal oversizing ratio. We also found an S-shaped association between the distal oversizing ratio and dSINE after TEVAR for TBAD. The distal oversizing ratio of 40% can be used for stratified management of patients who underwent TEVAR for TBAD.

The Number of Preoperative Abdominal False Lumen-perfused Small Branches Is Related to Abdominal Aortic Remodeling after Thoracic Endovascular Aortic Repair for Type B Aortic Dissection.

BACKGROUND: Abdominal aortic enlargement (AAE) is a commonly seen complication after thoracic endovascular aortic repair (TEVAR) for type B aortic dissection (TBAD). This study assessed the relationship between preoperative abdominal false lumen-perfused small branches (pre-AFLSBs) and risk of AAE after TEVAR for TBAD extending through the abdominal aorta. METHODS: One hundred and eighty-four patients with TBAD who had been treated by TEVAR at 4 hospitals in China were stratified in accordance with median pre-AFLSBs: ≤8 (group A) and >8 (group B). AAE was defined as ≥20% increase in abdominal total aortic volume, measured using imaging software featuring centerline analysis, on most recent postoperative computed tomography angiography relative to preoperative examination. Multivariable Cox proportional hazard regression was used to evaluate the relationship between the number of pre-AFLSBs and the risk of AAE after TEVAR. RESULTS: At median 12.4 months imaging follow-up, AAE was present in 65 patients at higher cumulative rate (log-rank test P < 0.001) in group B, which had more frequent partially thrombosed or patent abdominal false lumen than group A at 6-12 months (P < 0.01) and 12-24 months (P = 0.03) of follow-up. In a multivariable Cox proportional hazard regression model, the number of pre-AFLSBs was independently associated with risk of AAE after TEVAR (hazard ratio [for one increase], 1.10; 95% confidence interval, 1.00-1.21; P = 0.04) after adjustment for age, gender, dissection chronicity, Society for Vascular Surgery risk score, preoperative maximum descending aortic diameter, false lumen status of the abdominal aorta, numbers of false lumen-perfused visceral arteries, abdominal intimal larger tears, and preoperative abdominal intimal small tears, and dissection length and descending aortic length ratio. CONCLUSIONS: The number of pre-AFLSBs was positively associated with the risk of AAE after TEVAR for TBAD extending through the abdominal aorta.

Dissection length-to-descending thoraco-abdominal aorta length ratio predicts abdominal aortic enlargement after thoracic endovascular aortic repair for type B aortic dissection involving the abdominal aorta.

OBJECTIVES: This study was performed to assess the association between the dissection length-to-descending thoraco-abdominal aorta length ratio (LLR) and abdominal aortic enlargement (AAE) (≥20% increase in total abdominal aortic volume) after thoracic endovascular aortic repair (TEVAR) in patients with type B aortic dissection. METHODS: We retrospectively analysed data from 184 consecutive patients with type B aortic dissection who underwent TEVAR from January 2011 to December 2016 at 4 hospitals as part of the Registry Of type B aortic dissection with Utility of STent graft study. Preoperative and postoperative computed tomography angiography images were reviewed to assess the LLR and AAE. Patients were stratified into tertiles according to the pre-TEVAR LLR: 0.7 to <1.0 (n = 61), 1.0 to <1.2 (n = 61) and 1.2 to <1.6 (n = 62). The thoracic and abdominal aorta were divided by the celiac trunk. The cumulative incidence of AAE was estimated using the Kaplan-Meier method. A multivariable Cox proportional hazards model was used to assess the independent association between the preoperative LLR and the post-TEVAR risk of AAE. The nonlinear relationship between the LLR and the risk of post-TEVAR AAE was fitted by the restricted cubic smoothing spline, and the inflection point on the fitting curve was determined using a piecewise linear regression model. RESULTS: Baseline demographics, clinical features, preoperative anatomic characteristics and implanted devices were similarly distributed among the pre-TEVAR LLR tertile groups. At 24 months post-TEVAR, the estimated cumulative incidence of AAE significantly differed (P < 0.01) by LLR tertile group: 0.10 [95% confidence interval (CI) 0.00-0.21], 0.65 (95% CI 0.45-0.78) and 0.67 (95% CI 0.40-0.82), respectively. The pre-TEVAR LLR was an independent predictor of post-TEVAR AAE [hazard ratio (per unit increase) 1.03, 95% CI 1.01-1.04] following a nonlinear relationship with an inflection point at LLR = 1.0. CONCLUSIONS: The risk of post-TEVAR AAE is highest when the length of the dissection is greater than or equal to the length of the descending aorta (LLR ≥ 1.0).

Potential five-mRNA signature model for the prediction of prognosis in patients with papillary thyroid carcinoma.

Although the mortality rate of papillary thyroid carcinoma (PTC) is relatively low, the recurrence rates of PTC remain high. The high recurrence rates are related to the difficulties in treatment. Gene expression profiles has provided novel insights into potential therapeutic targets and molecular biomarkers of PTC. The aim of the present study was to identify mRNA signatures which may categorize PTCs into high-and low-risk subgroups and aid with the predictions for prognoses. The mRNA expression profiles of PTC and normal thyroid tissue samples were obtained from The Cancer Genome Atlas (TCGA) database. Differentially expressed mRNAs were identified using the 'EdgeR' software package. Gene signatures associated with the overall survival of PTC were selected, and enrichment analysis was performed to explore the biological pathways and functions of the prognostic mRNAs using the Database for Visualization, Annotation and Integration Discovery. A signature model was established to investigate a specific and robust risk stratification for PTC. A total of 1,085 differentially expressed mRNAs were identified between the PTC and normal thyroid tissue samples. Among them, 361 mRNAs were associated with overall survival (P<0.05). A 5-mRNA prognostic signature for PTC (ADRA1B, RIPPLY3, PCOLCE, TEKT1 and SALL3) was identified to classify the patients into high-and low-risk subgroups. These prognostic mRNAs were enriched in Gene Ontology terms such as 'calcium ion binding', 'enzyme inhibitor activity', 'carbohydrate binding', 'transcriptional activator activity', 'RNA polymerase II core promoter proximal region sequence-specific binding' and 'glutathione transferase activity', and Kyoto Encyclopedia of Genes and Genomes signaling pathways such as 'pertussis', 'ascorbate and aldarate metabolism', 'systemic lupus erythematosus', 'drug metabolism-cytochrome P450 and 'complement and coagulation cascades'. The 5-mRNA signature model may be useful during consultations with patients with PTC to improve the prediction of their prognosis. In addition, the prognostic signature identified in the present study may reveal novel therapeutic targets for patients with PTC.

Association Between Extent of Stent-Graft Coverage and Thoracic Aortic Remodeling After Endovascular Repair of Type B Aortic Dissection.

Purpose: To examine the association between the extent of stent-graft coverage and thoracic aortic expansion after thoracic endovascular aortic repair (TEVAR) for type B aortic dissection. Materials and Methods: A retrospective analysis was conducted of 201 patients (mean age 52.4±11.5 years; 178 men) with acute (135, 67.2%) or chronic (66, 32.8%) type B aortic dissection who underwent TEVAR at 4 medical centers. The mean stent-graft length was 157.1±33.3 mm. The percentage of stented descending aorta (PSDA) represented the extent of stent-graft coverage. After using restricted cubic smoothing spline plots to confirm the roughly linear relationship between PSDA and the risk of thoracic aortic expansion, patients were stratified into 2 groups on the median PSDA: the lower group (≤31.3%) and the higher group (>31.3%). Thoracic aortic expansion was defined as a ≥20% increase in the total thoracic aortic volume on the most recent postoperative computed tomography angiography scan compared with the preoperative measurement. The Kaplan-Meier method was used to estimate the cumulative freedom from thoracic aortic expansion after TEVAR; estimates are given with the 95% confidence interval (CI). A multivariable Cox proportional hazards model was used to analyze any independent association of the PSDA as a continuous or categorical variable with the risk of thoracic aortic expansion; results are presented as the hazard ratio (HR) and 95% CI. Results: No patients developed symptoms of spinal cord ischemia during hospitalization. Over a median 12.4 months of imaging follow-up, 34 (16.9%) patients developed thoracic aortic expansion. The estimate of freedom from thoracic aortic expansion at 12 months for the overall PSDA was 84.0% (95% CI 77.8% to 88.6%); between the groups, the freedom from thoracic aortic expansion estimate for the PSDA ≤31.3% group was significantly lower than in the higher group (p=0.032). Regression analysis showed no significant association between the risk of thoracic aortic expansion and the PSDA as a continuous variable (HR 0.97, 95% CI 0.91 to 1.03, p=0.288); however, analyzing the PSDA as a categorical variable indicated a significantly lower risk of thoracic aortic expansion for the PSDA >31.3% group (HR 0.46, 95% CI 0.22 to 0.95, p=0.036) after adjusting for a variety of demographic and anatomical characteristics. Conclusion: More extensive stent-graft coverage appears to improve thoracic aortic remodeling after TEVAR. However, the clinician should balance the benefit of extensive stent-graft coverage and its related risk of spinal cord ischemia.