Largest Area Under The Curve Research Articles

Association of endothelial function and limb artery indices with coronary artery stenosis severity in patients with hypertension

Background Hypertension is one of the major risk factors for cardiovascular events. This study aims to analyse the association of endothelial function and limb artery indices with coronary artery stenosis (CAS) severity in hypertension based on easily accessible and detailed clinical information, and to help accurately identify high-risk groups and avoid missed diagnosis and misdiagnosis. Methods Admission data of 1,375 consecutive hypertensive patients complicated with suspected coronary atherosclerotic heart disease (CHD) from September 2020 to August 2021 in China-Japan Friendship Hospital were retrospectively assessed. All candidates underwent coronary angiography for screening. A total of 600 eligible patients were classified in the CHD group (n = 359) and non-CHD group (n = 241) based on their coronary angiography results. Subjects in the CHD group were further assigned to ‘high stenosis’ (n = 178) and ‘low stenosis’ (n = 181) subgroups based on the median value of Gensini score. Endothelial function and limb artery indicators, including brachial artery flow-mediated vasodilatation (FMD), ankle-brachial index (ABI) and brachial-ankle pulse velocity (baPWV), were examined and compared between subgroups. Multivariate logistic regression analysis and multiple linear regression analysis were carried out to select independent risk factors of CAS severity in hypertension. A predictive equation was conducted according to the results of multivariate logistic regression analysis to make clinical practice easier. As the receiver operating characteristic (ROC) curve had been plotted, the predictive ability of endothelial function and limb artery indicators in CAS severity in hypertension was detected by the area under the curve (AUC). Results In patients with hypertension, the FMD (p = 0.023), ABI (p < 0.001) and baPWV (p < 0.001) of CHD patients appeared substantially different from the non-CHD patients. Furthermore, the ABI (p < 0.001) and baPWV (p = 0.032) both independently associated with CAS severity in hypertensive patients with CHD. Based on the results of multivariate logistic regression analysis with CAS severity as a dependent variable, a predictive equation of baPWV, ABI and FMD was developed: combined coefficient = Logit(p)=5.531–0.218*FMD–7.019*ABI + 0.244*baPWV. From the combined coefficients of baPWV, ABI and FMD, the largest AUC was 0.800, suggesting a powerful predictive value of CAS severity in hypertensive patients, followed by ABI (AUC = 0.747, 95%CI 0.693–0.796), baPWV (AUC = 0.704, 95%CI 0.648–0.756) and FMD (AUC = 0.588, 95%CI 0.529–0.645). Conclusion This study shows that baPWV, ABI and FMD are independent risk factors for CHD, of which, baPWV and ABI are strongly associated with CAS severity in hypertensive patients. The predictive ability of CHD in hypertensive patients may be enhanced through combining the three endothelial function and limb artery indicators. The results may help to facilitate clinical decision-making during treatment and management of coronary artery disease.

Expression of Serum DCLK1 in Gastric Cancer Patients and Its Relationship with CEA, CA19-9, and CA72-4

Objective: To investigate the expression of Doublecortin-like kinase-1 (DCLK1) in the serum of gastric cancer patients and its relationship with carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), and carbohydrate antigen 72-4 (CA72-4). Methods: Fifty patients diagnosed with gastric cancer at the hospital from January to December 2021 were selected as the gastric cancer group, and 50 patients diagnosed with chronic atrophic gastritis during the same period were selected as the control group. The serum concentrations of DCLK1, CEA, CA19-9, and CA72-4 were measured in both groups. Cut-off values and AUC (Area Under the Curve) were determined based on the ROC curve, and the expression of DCLK1 and its relationship with CEA, CA19-9, and CA72-4 were analyzed. Results: The average concentrations of DCLK1, CEA, CA19-9, and CA72-4 in the serum of gastric cancer patients were significantly higher than those in the control group (P &lt; 0.05). CA72-4 had the highest sensitivity (62%), CEA had the highest specificity (98%), and DCLK1 had the largest AUC (0.709). The combined diagnosis of gastric cancer using DCLK1, CEA, and CA19-9 resulted in the largest AUC (0.826), with a sensitivity of 82% and a specificity of 76%. Conclusion: The expression of DCLK1, CEA, CA19-9, and CA72-4 in the serum of gastric cancer patients is significantly higher than that in the control group. The combined detection of DCLK1, CEA, and CA19-9 offers better sensitivity and specificity for the diagnosis of gastric cancer.

A Dunnett-Type Test and Its Sample Size Calculation for Comparing K ROC Curves with a Control.

Diagnostic biomarkers are key components of diagnostics. In this paper, we consider diagnostic biomarkers taking continuous values that are associated with a dichotomous disease status, called malignant or benign. The performance of such a biomarker is evaluated by the area under the curve (AUC) of its receiver operating characteristic curve. We assume that, together with the disease status, one control and multiple experimental biomarkers are collected from each subject to test if any of the experimental biomarkers have a larger AUC than the control. In this case, each experimental biomarker will be compared with the control so that a multiple testing issue is involved in the comparisons. In this paper, we propose a simple non-parametric statistical testing procedure to compare K(≥2) experimental biomarkers with a control, adjusting for the multiplicity and its sample size calculation method. Our sample size formula requires the specification of the AUC values (or the standardized effect size of each biomarker between the benign and malignant groups) together with the correlation coefficients between the biomarkers, the prevalence of the malignant group in the study population, the type I error rate, and the power. Through simulations, we show that the statistical test controls the overall type I error rate accurately and the proposed sample size closely maintains the specified statistical power.

The Use of Platelet Count and Indices as Prognostic Factors for Mortality in Children with Sepsis.

Sepsis is still one of the leading causes of mortality and morbidity in children worldwide. Consumptive coagulopathy and suppression of thrombopoiesis in the bone marrow resulting from immune dysregulation are pathological mechanisms that cause thrombocytopenia in sepsis. Platelet count (PLT) and indices, such as mean platelet volume (MPV), platelet distribution width (PDW), and plateletcrit (PCT) are markers of platelet activation and are strongly influenced by platelet morphology and proliferation kinetics. We aimed to study the use of platelet count and indices as predictors of mortality in children with sepsis. This is a cross-sectional study of 45 children diagnosed with sepsis on admission at Haji Adam Malik Hospital, Medan, North Sumatra, Indonesia, between October and November 2022. Blood samples were drawn upon admission, and platelet count and indices were then determined for all children. Subjects were followed up till discharge from hospital or death. Receiver Operating Characteristic (ROC) curve analysis of platelet parameters was done to determine the area under the curve (AUC), optimal cut-off value, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) in predicting mortality in children with sepsis. Using the cut-off values from ROC curve analysis, the odds ratio with 95% confidence interval was calculated using multiple logistic regression analyses. A P value less than 0.05 was considered statistically significant. MPV, PDW, and PDW/PLT were significantly higher in non-survivors than survivors (P=0.04, P=0.02, and P=0.04, respectively). ROC curve analysis showed that PDW had the largest AUC (0.708 [95% CI=0.549-0.866]) with a cut-off value of 14.1%, sensitivity of 63.6%, and specificity of 82.6%. PDW was also the only parameter that significantly affected the outcome of children with sepsis. PDW≥14.1% at admission increases the risk of mortality by 5.7 times. Admission PDW is a fast and specific tool to predict the outcome of children with sepsis.

Predictive value of thrombin-antithrombin III complex and tissue plasminogen activator-inhibitor complex biomarkers in assessing the severity of early-stage acute pancreatitis.

The development of acute pancreatitis (AP) is strongly linked to blood clotting and fibrinolysis issues. Modern clinical practices now utilize advanced blood markers like thrombin-antithrombin III complex (TAT), plasmin-α2-plasmin inhibitor complex, thrombomodulin (TM), and tissue plasminogen activator-inhibitor complex (t-PAIC) to assess thrombosis risk. Our study used a highly sensitive chemiluminescence technique to measure these markers in AP patients, aiming to determine their early predictive value for AP severity. There were 173 patients with AP, all of whom developed symptoms within 72h; 102 individuals had onset symptoms within 48h. The biomarkers were measured upon admission before determining the severity of AP. The levels of TAT, plasmin-α2-plasmin inhibitor complex, TM, and t-PAIC were significantly higher in the severe acute pancreatitis (SAP) group compared with the mild acute pancreatitis and moderate severe acute pancreatitis groups. For the patients within 72h of onset, TAT, TM, and t-PAIC predicted the occurrence of SAP. For the patients within 48h of onset, TAT and t-PAIC predicted the occurrence of SAP. The area under the curve (AUC) of prediction models is similar to Bedside Index for Severity in Acute Pancreatitis (BISAP) but significantly higher than C-reactive protein (P<0.05). Notably, t-PAIC had a larger AUC than TAT, BISAP, and C-reactive protein. In the initial 48h, plasma TAT and t-PAIC levels may predict the development of SAP. Within 72h, plasma levels of TAT, TM, and t-PAIC may predict the development of SAP, and the TAT+TM+t-PAIC prediction model achieved a maximum AUC of 0.915, comparable to BISAP.

Sample size calculation for comparing two ROC curves.

Biomarkers are key components of personalized medicine. In this paper, we consider biomarkers taking continuous values that are associated with disease status, called case and control. The performance of such a biomarker is evaluated by the area under the curve (AUC) of its receiver operating characteristic curve. Oftentimes, two biomarkers are collected from each subject to test if one has a larger AUC than the other. We propose a simple non-parametric statistical test for comparing the performance of two biomarkers. We also present a simple sample size calculation method for this test statistic. Our sample size formula requires specification of AUC values (or the standardized effect size of each biomarker between cases and controls together with the correlation coefficient between two biomarkers), prevalence of cases in the study population, type I error rate, and power. Through simulations, we show that the testing on two biomarkers controls type I error rate accurately and the proposed sample size closely maintains specified statistical power.

A Novel Risk Score to Predict Thirty-Day Readmissions after Acute Type A Aortic Dissections

Background: Readmissions following acute type A aortic dissections (ATAAD) are associated with potentially worse clinical outcomes and increased hospital costs. Predicting which patients are at risk for readmission may guide patient management prior to discharge. Methods: The National Readmissions Database was utilized to identify patients treated for ATAAD between 2010 and 2018. Univariate mixed effects logistic regression was used to assess each variable. Variables were assigned risk points based off the bootstrapped (bias-corrected) odds ratio of the final variable model according to the Johnson's scoring system. A mixed effect logistic regression was run on the risk score (sum of risk points) and 30-day readmission. Calibration plots and predicted readmission curves were generated for model assessment. Results: A total of 30,727 type A aortic dissections were identified. The majority of ATAAD (66%) were in men with a median age of 61 years and 30-day readmission rate of 19.4%. The risk scores ranging from –1 to 14 mapped to readmission probabilities between 3.5% and 29% for ATAAD. The predictive model showed good calibration and receiver operator characteristics with an area under the curve (AUC) of 0.81. Being a resident of the hospital state (OR: 2.01 [1.64, 2.47], p &lt; 0.001) was the highest contributor to readmissions followed by chronic kidney disease (1.35 [1.16, 1.56], p = 0), discharge to a short-term facility (1.31 [1.09, 1.57], p = 0.003), and developing a myocardial infarction (1.20 [1.00, 1.45], p = 0.048). Conclusions: The readmission model had good predictive capability given by the large AUC. Being a resident in the State of the index admission was the most significant contributor to readmission.

Preoperative radiological indicators for prediction of difficult laryngoscopy in patients with atlantoaxial dislocation

BackgroundDifficult airway remains a great challenge in patients with atlantoaxial dislocation (AAD). Preoperative evaluation and reliable prediction are required to facilitate the airway management. We aimed to screen out reliable radiological indicators for prediction of difficult laryngoscopy in patients with AAD. MethodsA retrospective nested case-control study within a single center longitudinal AAD cohort was conducted to investigate the radiological indicators. All the patients with difficult laryngoscopy from 2010 to 2021 were enrolled as the difficult laryngoscopy group. Others in the cohort without difficult laryngoscopy were randomly selected as the non-difficult laryngoscopy group by individually matching with the same gender, same surgery year, and similar age (±5 years) at a ratio of 6:1. Radiological data on preoperative lateral X-ray images between the two groups were compared. Bivariate logistic regression model was applied to screen out the independent predictive indicators and calculate the odds ratios of indicators associated with difficult laryngoscopy. Receiver operating characteristic curve and area under the curve (AUC) were used to describe the discrimination ability of indicators. ResultsA total of 154 patients were finally analyzed in this study. Twenty-two patients with difficult laryngoscopy and matched with 132 controls. Four radiological parameters showed significant difference between the two groups. Among which, ΔC1C2D (the difference of the distance between atlas and axis in the neutral and extension position), owned the largest AUC. ConclusionsΔC1C2D could be a valuable radiologic predictor for difficult laryngoscopy in patients with AAD.

Examining the Talent Selection Determinants of Ultimate Frisbee Players Selected Into a National Youth Team.

The purpose of the study was to examine whether various athletic performances, anthropometric measures, and playing experience differentiate selected and nonselected ultimate Frisbee players trialing to compete in the world championship. Forty-three Australian male ultimate Frisbee players (age = 21.2 [1.2]y; height = 1.7 [6.8]m; body mass = 69.7 [8.2]kg; playing experience = 3.5 [1.5]y) participated in a 30-m sprint test, single-leg run-up jump approach (both left [JumpLL], and right leg [JumpRL]) and a stationary bilateral vertical jump (JumpBIL), and change-of-direction speed test. Following a selection camp, players were subdivided according to their selection or nonselection into the team. A multivariate analysis of variance revealed that height, 10-m sprint time, acceleration, JumpLL, JumpRL, and JumpBIL were significantly greater for selected players than nonselected players (P < .05). Area under the curve (AUC) was greatest for JumpRL (AUC = 79%; optimal cutoff value of 37.5cm, sensitivity and specificity values of 77% and 71%, respectively), JumpLL (AUC = 74%; optimal cutoff 38.5cm, sensitivity and specificity values 77% and 77%, respectively), and JumpBIL (AUC = 78%; optimal cutoff value of 40.5cm, sensitivity and specificity values 71% and 79%, respectively). The largest AUC (AUC = 81%; 95% CI 0.66-0.97; P = .001) was found when combining the explanatory variables that demonstrated moderate to large effect sizes (ie,height, playing experience, 10-m sprint, acceleration, JumpLL, JumpRL, and JumpBIL), with sensitivity of 93% and specificity of 71%. These athletic performance and anthropometric characteristics differentiating selected and nonselected players may help inform targeted training and player-development strategies.

Predictive function of tumor burden-incorporated machine-learning algorithms for overall survival and their value in guiding management decisions in patients with locally advanced nasopharyngeal carcinoma

ObjectiveAccurate prognostic predictions and personalized decision-making on induction chemotherapy (IC) for individuals with locally advanced nasopharyngeal carcinoma (LA-NPC) remain challenging. This research examined the predictive function of tumor burden-incorporated machine-learning algorithms for overall survival (OS) and their value in guiding treatment in patients with LA-NPC. MethodsIndividuals with LA-NPC were reviewed retrospectively. Tumor burden signature-based OS prediction models were established using a nomogram and two machine-learning methods, the interpretable eXtreme Gradient Boosting (XGBoost) risk prediction model, and DeepHit time-to-event neural network. The models' prediction performances were compared using the concordance index (C-index) and the area under the curve (AUC). The patients were divided into two cohorts based on the risk predictions of the most successful model. The efficacy of IC combined with concurrent chemoradiotherapy was compared to that of chemoradiotherapy alone. ResultsThe 1 221 eligible individuals, assigned to the training (n = 813) or validation (n = 408) set, showed significant respective differences in the C-indices of the XGBoost, DeepHit, and nomogram models (0.849 and 0.768, 0.811 and 0.767, 0.730 and 0.705). The training and validation sets had larger AUCs in the XGBoost and DeepHit models than the nomogram model in predicting OS (0.881 and 0.760, 0.845 and 0.776, and 0.764 and 0.729, P < 0.001). IC presented survival benefits in the XGBoost-derived high-risk but not low-risk group. ConclusionThis research used machine-learning algorithms to create and verify a comprehensive model integrating tumor burden with clinical variables to predict OS and determine which patients will most likely gain from IC. This model could be valuable for delivering patient counseling and conducting clinical evaluations.

Establishment and validation of a novel nomogram to predict overall survival after radical nephrectomy

Objective: To establish a nomogram prognostic model for predicting the 5-, 10-, and 15-year overall survival (OS) of non-metastatic renal cell carcinoma patients managed with radical nephrectomy (RN), compare the modelled results with the results of pure pathologic staging, the Karakiewicz nomogram and the Mayo Clinic Stage, Size, Grade, and Necrosis (SSIGN) score commonly used in foreign countries, and stratify the patients into different prognostic risk subgroups. Methods: A total of 1 246 non-metastatic renal cell carcinoma patients managed with RN in Sun Yat-sen University Cancer Center (SYSUCC) from 1999 to 2020 were retrospectively analyzed. Multivariate Cox regression analysis was used to screen the variables that influence the prognosis for nomogram establishment, and the bootstrap random sampling was used for internal validation. The time-receiver operating characteristic curve (ROC), the calibration curve and the clinical decision curve analysis (DCA) were applied to evaluate the nomogram. The prediction efficacy of the nomogram and that of the pure pathologic staging, the Karakiewicz nomogram and the SSIGN score was compared through the area under the curve (AUC). Finally, patients were stratified into different risk subgroups according to our nomogram scores. Results: A total of 1 246 patients managed with RN were enrolled in this study. Multivariate Cox regression analysis showed that age, smoking history, pathological nuclear grade, sarcomatoid differentiation, tumor necrosis and pathological T and N stages were independent prognostic factors for RN patients (all P<0.05). A nomogram model named SYSUCC based on these factors was built to predict the 5-, 10-, and 15-year survival rate of the participating patients. In the bootstrap random sampling with 1 000 iterations, all these factors occurred for more than 800 times as independent predictors. The Harrell's concordance index (C-index) of SYSUCC was higher compared with pure pathological staging [0.770 (95% CI: 0.716-0.823) vs 0.674 (95% CI: 0.621-0.728)]. The calibration curve showed that the survival rate as predicted by the SYSUCC model simulated the actual rate, while the clinical DCA showed that the SYSUCC nomogram has a benefit in certain probability ranges. In the ROC analysis that included 857 patients with detailed pathological nuclear stages, the nomogram had a larger AUC (5-/10-year AUC: 0.823/0.804) and better discriminating ability than pure pathological staging (5-/10-year AUC: 0.701/0.658), Karakiewicz nomogram (5-/10-year AUC: 0.772/0.734) and SSIGN score (5-/10-year AUC: 0.792/0.750) in predicting the 5-/10-year OS of RN patients (all P<0.05). In addition, the AUC of the SYSUCC nomogram for predicting the 15-year OS (0.820) was larger than that of the SSIGN score (0.709), and there was no statistical difference (P<0.05) between the SYSUCC nomogram, pure pathological staging (0.773) and the Karakiewicz nomogram (0.826). The calibration curve was close to the standard curve, which indicated that the model has good predictive performance. Finally, patients were stratified into low-, intermediate-, and high-risk subgroups (738, 379 and 129, respectively) according to the SYSUCC nomogram scores, among whom patients in intermediate- and high-risk subgroups had a worse OS than patients in the low-risk subgroup (intermediate-risk group vs. low-risk group: HR=4.33, 95% CI: 3.22-5.81, P<0.001; high-risk group vs low-risk group: HR=11.95, 95% CI: 8.29-17.24, P<0.001), and the high-risk subgroup had a worse OS than the intermediate-risk group (HR=2.63, 95% CI: 1.88-3.68, P<0.001). Conclusions: Age, smoking history, pathological nuclear grade, sarcomatoid differentiation, tumor necrosis and pathological stage were independent prognostic factors for non-metastasis renal cell carcinoma patients after RN. The SYSUCC nomogram based on these independent prognostic factors can better predict the 5-, 10-, and 15-year OS than pure pathological staging, the Karakiewicz nomogram and the SSIGN score of patients after RN. In addition, the SYSUCC nomogram has good discrimination, agreement, risk stratification and clinical application potential.

Echocardiographic evaluation of right ventricular diastolic function in pulmonary hypertension.

Right ventricular (RV) diastolic dysfunction may be prognostic in pulmonary hypertension (PH). However, its assessment is complex and relies on conductance catheterisation. We aimed to evaluate echocardiography-based parameters as surrogates of RV diastolic function, provide validation against the gold standard, end-diastolic elastance (Eed), and define the prognostic impact of echocardiography-derived RV diastolic dysfunction. Patients with suspected PH who underwent right heart catheterisation including conductance catheterisation were prospectively recruited. In this study population, an echocardiography-based RV diastolic function surrogate was derived. Survival analyses were performed in patients with precapillary PH in the Giessen PH Registry, with external validation in patients with pulmonary arterial hypertension at Sapienza University (Rome). In the derivation cohort (n=61), the early/late diastolic tricuspid inflow velocity ratio (E/A) and early tricuspid inflow velocity/early diastolic tricuspid annular velocity ratio (E/e') did not correlate with Eed (p>0.05). Receiver operating characteristic analysis revealed a large area under the curve (AUC) for the peak lateral tricuspid annulus systolic velocity/right atrial area index ratio (S'/RAAi) to detect elevated Eed (AUC 0.913, 95% confidence interval (CI) 0.839-0.986) and elevated end-diastolic pressure (AUC 0.848, 95% CI 0.699-0.998) with an optimal threshold of 0.81 m2·s-1·cm-1. Subgroup analyses demonstrated a large AUC in patients with preserved RV systolic function (AUC 0.963, 95% CI 0.882-1.000). Survival analyses confirmed the prognostic relevance of S'/RAAi in the Giessen PH Registry (n=225) and the external validation cohort (n=106). Our study demonstrates the usefulness of echocardiography-derived S'/RAAi for noninvasive assessment of RV diastolic function and prognosis in PH.

1277-P: Prediction of HbA1c Test Eligibility in Emergency Department

Introduction: Diabetes screening using best practice advisory (BPA) alerts integrated into the emergency department (ED) workflow may efficiently identify at-risk patients disproportionally affected by diabetes. The objectives of this study were to 1) validate a BPA algorithm informed by the American Diabetes Association (ADA) screening criteria using elements from the electronic medical record (EMR) to identify eligible patients for hemoglobin A1c (HbA1c) testing, and 2) compare the characteristics of the ADA-based algorithm with one informed by the United States Preventative Services Task Force (USPSTF) recommendations. Methods: This cross-sectional study involved adults admitted to the University of Illinois Health ED in May 2021 with labs ordered (n=2963). Sensitivity, specificity, positive predictive value (PPV), and negative predictive values (NPV) estimated the BPA algorithm’s ability to correctly identify patients eligible for diabetes screening using manual EMR chart review as the reference standard. Receiver operator characteristic (ROC) curves and area under the curve (AUC) were calculated to compare ADA and USPSTF-informed algorithms. Results: The sensitivity, specificity, PPV, and NPV of the ADA model were 0.70 (95% CI: 0.66-0.75), 0.91 (0.90-0.92), 0.76 (0.73-0.79), and 0.88 (0.86-0.89), respectively. Compared to the ADA model, the USPSTF model’s sensitivity was worse (0.25 (0.22-0.27)) and specificity was slightly higher (0.94 (0.93-0.95)), though PPV remained similar (0.75 (0.71-0.79)) and NPV was lower (0.64 (0.62-0.66)). The ROC AUC for the ADA and USPSTF-based models were 0.91 (0.89-0.93) and 0.86 (0.84-0.89), respectively. Conclusion: Despite moderate to low sensitivity, the PPVs and large AUCs are acceptable due to the high prevalence of at-risk populations presenting to the ED. Findings suggest that a BPA algorithm informed by the ADA guideline is an adequate tool to identify eligible patients for diabetes screening in the ED, especially given its ease of workflow integration. Disclosure M.Smart: None. J.Lin: Research Support; Novo Nordisk, Xeris Pharmaceuticals, Inc. T.Lee: None. B.T.Layden: Consultant; Bayer Inc. Y.Eisenberg: None. K.K.Danielson: None. A.Kong: None. Funding Novo Nordisk (G2111)

Clinical features of Kawasaki disease complicated by macrophage activation syndrome: an analysis of 27 cases

To investigate the clinical manifestations and laboratory examination results of children with Kawasaki disease complicated by macrophage activation syndrome (KD-MAS), and to provide a basis for identifying early warning indicators for the early diagnosis and treatment of KD-MAS. A retrospective study was performed on 27 children with KD-MAS (KD-MAS group) and 110 children with KD (KD group) who were admitted to Wuhan Children's Hospital, Tongji Medical College, Huazhong University of Science and Technology, from January 2014 to January 2022. Clinical and laboratory data were compared between the two groups. The receiver operating characteristic (ROC) curve was used to investigate the value of laboratory markers with statistical significance in the diagnosis of KD-MAS. Compared with the KD group, the KD-MAS group had significantly higher incidence rates of hepatomegaly, splenomegaly, incomplete KD, no response to intravenous immunoglobulin, coronary artery damage, multiple organ damage, and KD recurrence, as well as a significantly longer length of hospital stay (P<0.05). Compared with the KD group, the KD-MAS group had significantly lower levels of white blood cell count, absolute neutrophil count, hemoglobin, platelet count (PLT), erythrocyte sedimentation rate, serum albumin, serum sodium, prealbumin, and fibrinogen (FIB), a significantly lower incidence rate of non-exudative conjunctiva, and significantly higher levels of C-reactive protein, alanine aminotransferase, aspartate aminotransferase, lactate dehydrogenase (LDH), and serum ferritin (SF) (P<0.05). The ROC curve analysis showed that SF, PLT, FIB, and LDH had high value in the diagnosis of KD-MAS, with areas under the curve (AUC) of 0.989, 0.966, 0.932, and 0.897, respectively (P<0.001), and optimal cut-off values of 349.95 μg/L, 159×109/L, 3.85 g/L, and 403.50 U/L, respectively. The combination of SF, PLT, FIB, and LDH had a larger AUC than PLT, FIB, and LDH alone in the diagnosis of KD-MAS (P<0.05), but there was no significant difference in the AUC between the combination of SF, PLT, FIB, and LDH and SF alone (P>0.05). KD-MAS should be considered when children with KD have hepatosplenomegaly, no response to intravenous immunoglobulin, coronary artery damage, and KD recurrence during treatment. SF, PLT, FIB, and LDH are of high value in the diagnosis of KD-MAS, especially SF is of great significance in the diagnosis of KD-MAS.

Biomarkers as predicting factors for relapse in polymyalgia rheumatica: the importance of alkaline phosphatase.

Dear Editor, PMR is an inflammatory disease that affects older adults and causes pain and stiffness in the neck, shoulders and girdles. PMR is typically accompanied by the elevation of ESR or CRP [1]. Administration of a low–medium dose of steroid is recommended as initial therapy in the treatment of PMR without GCA [1]. Although clinical response is usually favourable, PMR relapses are frequent and have been reported in up to 67% of patients during the first year of follow-up and up to 49% of patients at 5 years [2]. Previous reports have suggested that the elevation of ESR or CRP [2, 3], and female [3], older [4] or heavier [5] patients might be a risk factor for PMR relapse; however, the characteristics underlying PMR relapse require elucidation. Herein, we assessed clinical and laboratory markers of PMR in cases of relapse. Data derived from the initial visit of all adult patients with PMR registered in the database of our hospital from July 2010 to October 2020 were retrospectively collected and analysed. All patients met Bird’s or 2012 ACR/EULAR criteria for PMR. Patients with a history or coexistence of GCA, patients who were suspected of having PMR initially, then whose diagnosis was changed from PMR to another (such as paraneoplastic syndrome), and patients with insufficient data were excluded. Patients who relapsed at least once were classified as the relapse group. We defined ‘relapse’ as a reappearance of clinical symptoms typically as PMR accompanied by an elevation of ESR (≥30 mm/h) or CRP (≥0.5 mg/dl) during the clinical course based on a previous report [2]. The univariate analysis between the relapse and non-relapse groups was performed using Fisher’s exact test (categorical variables) and Mann–Whitney U test (continuous variables). The cut-off for significant difference in the univariate analysis of new continuous variables was calculated using the following method. (i) The minimum to maximum values were separated by 5 and used as cut-off candidates. (ii) We binarized the candidates to above/below the value and added them to the established factors previously measured (age, sex, body weight, CRP and ESR) [2–5]. We created a logistic regression model using all the abovementioned factors as explanatory variables. (iii) Based on the prediction probability calculated from the model, the receiver operating characteristic (ROC) curve related to relapse/non-relapse was drawn, and the area under the curve (AUC) was calculated. (iv) By repeating these steps, the cut-off value candidate with the largest AUC was defined as the final cut-off value. Subsequently, we performed a multivariate logistic regression analysis. A P-value of <0.05 indicated statistical significance, and the adjusted odds ratio and corresponding 95% CI were subsequently estimated. A total of 72 patients were enrolled (Table 1). In total, 51 (70.8%) patients were women, and 31 (43.1%) had experienced a relapse. The univariate analysis revealed that age, sex and body weight did not show significant differences. ESR and serum alkaline phosphatase (ALP) were significantly higher in the relapse group than in the non-relapse group (96.0 vs 75.0 mm/h, P = 0.008; and 314.0 vs 253.0 U/l, P = 0.006, respectively). Although abdominal ultrasonography was performed in 11/72 (15.3%) cases, none revealed cholestatic liver diseases. Serum ALP was extracted as a new factor with a significant intergroup difference.

Association between atherogenic index of plasma and colorectal neoplasms in patients with hypertension.

Association between atherogenic index of plasma and colorectal neoplasms in patients with hypertension.

495. Enterovirus Genotype Specific Immune Response in Cerebrospinal Fluid of Infected Infants

Abstract Background Non-polio Enteroviruses (EV) are important neonatal CNS pathogens. Multiple EV genotypes have been detected in pediatric CSF, e.g. Echovirus (E) 6 and E30. CSF innate immune responses to EV genotypes remain poorly defined. Most data are from EV-A71 or E30 CNS infections and do not compare responses between these or other EV types. We sought to better define innate immune responses to EV genotypes in CSF. Methods Salvaged standard of care CSF samples from ≤6 month olds (real time EV PCR(+) or EV PCR(-) controls) from Jan 2010 - Dec 2020 were tested in duplicate on a 21- cytokine bead panel (MilliporeSigma). EV positive samples were previously genotyped by sequencing the viral capsid gene. Cytokine levels calculated from the standard curve were compared by Kruskal-Wallis and post-hoc analysis (GraphPad Prism 8.4.3). Natural partitioning of participants was explored using principal component analysis and cluster analysis (IBM SPSS v27). The utility of cytokine signatures in predicting EV status was explored using discriminant analysis and ROC analysis (IBM SPSS v27). Results Data from 72 CSF with E6 (N=16), E9 (N= 9), E18 (N=9), and E30 (N=21) showed significant differences among EV genotypes vs. controls for 20 cytokines (IL-17 was excluded). Significant differences in cytokine levels in EV CSF vs controls were seen: E6 for all 20 cytokines; E9 for Fractalkine, IP10, and MCP1; E18 for Fractalkine and MCP1; E30 19 cytokines (not GM-CSF). PCA revealed only minor overlap of controls and EV positives; EV types overlapped, except E30, differing most from E9 and E18 but overlapping E6. The most important type-differentiating cytokines by PCA were MCP1, Fractalkine, IL-8, and IL-10. Patterns in DA resembled PCA; controls clearly separated from EV CSF, E30 being the most distinct. Overall, the discriminant model correctly classified EV type or controls at a 63% rate - highest for controls (94.1%) and E30 (74.1%). In the DA model, the most important cytokines were IP-10, IL-2, IL-1Ra, and Fractalkine. Discriminant scores had the largest area under the curve (AUC), 0.990. Among cytokines, IFNa2 had the largest AUC, 0.987. Conclusion Preliminary data show significant EV-genotype differences in innate immune response to CNS infections. Cytokine patterns may serve as a key predictor for discerning EV genotypes. Disclosures Brian R. Lee, PhD, MPH, CDC: Grant/Research Support|Merck: Grant/Research Support Christopher J Harrison, MD, Astellas: Grant/Research Support|GSK: Grant/Research Support|Merck: Grant/Research Support|Pediatric news: Honoraria|Pfizer: Grant/Research Support Rangaraj Selvarangan, BVSc, PhD, D(ABMM), FIDSA, F(AAM), BioFire: Grant/Research Support|Luminex: Grant/Research Support.

Machine learning‐based prediction of cognitive outcomes in de novo Parkinson's disease

AbstractBackgroundCognitive impairment is a debilitating symptom in Parkinson’s disease (PD), with high variability in onset and course of progression. This present study aimed to establish an accurate multivariate machine learning model to predict cognitive outcome in newly diagnosed PD.MethodUsing baseline measures from the Parkinson’s Progression Markers Initiative (PPMI) cohort we subset clinical, biofluid and genetic/epigenetic variables. Annual cognitive assessments over an eight‐year time span were used to define two outcomes of i) dementia conversion, and ii) cognitive impairment. Outcomes and variable subsets were tested using multiple machine learning (ML) algorithms (random forest, elasticnet, SVM‐linear and cforest) to predict prognosis of individual cognitive decline in PD. Prediction was assessed using multiple measures including Area under the Curve (AUC) and Mathews Correlation Coefficient (MCC).ResultFor both cognitive outcomes, irrespective of the of ML algorithm, models consisting of clinical variables alone performed best, with high specificity and the largest AUC's (0.88 ‐ 0.92) and MCC's (0.57‐0.80). Notably, cognitive impairment outcome showed better sensitivity than dementia conversion outcome (0.72 – 0.81 vs 0.29 – 0.64, respectively). Addition of biological and genetic variables did not largely improve the model performance. However, a number of cerebrospinal fluid (CSF) proteins and epigenetic markers showed high predictive weighting in multiple models, when included alongside clinical variables.ConclusionMachine learning algorithms can accurately predict cognitive impairment development in PD. Within the generated models, clinical predictors appear to play a more prominent role than biological predictors. We further present additional future study data, expanding epigenetic research into PD dementia.

Best Fit Sphere Back and Adjusted Maximum Elevation of Corneal Back Surface as Novel Predictors of Keratoconus Progression.

We evaluated the Maximum Elevation of Corneal Back Surface adjusted to the same Best Fit Sphere Back (BFSB) between timeline measurements (AdjEleBmax) and the BFSB radius (BFSBR) itself as new tomographic parameters for documentation of ectasia progression and compare them with the most recent and reliable parameters used on keratoconus (KC) progression. We evaluated the performance and the ideal cutoff point of Kmax, D-index, posterior radius of curvature from the 3.0 mm centered on the thinnest point (PRC), EleBmax, BFSBR, and AdjEleBmax as isolated parameters to document KC progression (defined as a significant change in two or more variables), we found a sensitivity of 70%, 82%, 79%, 65%, 51%, and 63% and a specificity of 91%, 98%, 80%, 73%, 80%, and 84% to detect KC progression. The area under the curve (AUC) for each variable was 0.822, 0.927, 0.844, 0.690, 0.695, 0.754, respectively. AdjEleBmax presented a greater specificity, larger AUC, and better performance compared to EleBmax without any adjustment, with similar sensitivity. Although AdjEleBmax and BFSB demonstrated smaller AUC and specificities comparing with Kmax and D-Index, AdjEleBmax still presented a good performance with a reasonable AUC. Since the shape of the posterior surface, more aspheric and curved than the anterior, may facilitate detection of change, we suggest the inclusion of AdjEleBmax in the evaluation of KC progression in conjunction with other variables to increase the reliability of our clinical evaluation and early detection of progression.

Prognostic prediction by a novel integrative inflammatory and nutritional score based on least absolute shrinkage and selection operator in esophageal squamous cell carcinoma.

This study aimed to establish and validate a novel predictive model named integrative inflammatory and nutritional score (IINS) for prognostic prediction in esophageal squamous cell carcinoma (ESCC). We retrospectively recruited 494 pathologically confirmed ESCC patients with surgery and randomized them into training (n = 346) or validation group (n = 148). The least absolute shrinkage and selection operator (LASSO) Cox proportional hazards (PH) regression analysis was initially used to construct a novel predictive model of IINS. The clinical features and prognostic factors with hazard ratio (HRs) and 95% confidence intervals (CIs) grouped by IINS were analyzed. Nomogram was also established to verify the prognostic value of IINS. According to the LASSO Cox PH regression analysis, a novel score of IINS was initially constructed based on 10 inflammatory and nutritional indicators with the optimal cut-off level of 2.35. The areas under the curve (AUCs) of IINS regarding prognostic ability in 1-year, 3-years, and 5-years prediction were 0.814 (95% CI: 0.769-0.854), 0.748 (95% CI: 0.698-0.793), and 0.792 (95% CI: 0.745-0.833) in the training cohort and 0.802 (95% CI: 0.733-0.866), 0.702 (95% CI: 0.621-0.774), and 0.748 (95% CI: 0.670-0.816) in the validation cohort, respectively. IINS had the largest AUCs in the two cohorts compared with other prognostic indicators, indicating a higher predictive ability. A better 5-years cancer-specific survival (CSS) was found in patients with IINS ≤ 2.35 compared with those with IINS > 2.35 in both training cohort (54.3% vs. 11.1%, P < 0.001) and validation cohort (53.7% vs. 18.2%, P < 0.001). The IINS was then confirmed as a useful independent factor (training cohort: HR: 3.000, 95% CI: 2.254-3.992, P < 0.001; validation cohort: HR: 2.609, 95% CI: 1.693-4.020, P < 0.001). Finally, an IINS-based predictive nomogram model was established and validated the CSS prediction (training set: C-index = 0.71 and validation set: C-index = 0.69, respectively). Preoperative IINS is an independent predictor of CSS in ESCC. The nomogram based on IINS may be used as a potential risk stratification to predict individual CSS and guide treatment in ESCC with radical resection.