Easix score predicts the need for vasopressor support in critically ill pediatric hematology oncology patients

Background Sepsis is a common and serious complication in pediatric hematology oncology (PHO) patients. The Endothelial Activation and Stress Index (EASIX) score offers a potentially accessible tool for risk stratification in septic patients. Our

was to evaluate the utility of the EASIX score in predicting adverse clinical outcomes among septic PHO patients.

Retrospective review of all PHO patients admitted to the intensive care unit (ICU) with sepsis from July 2022 till December 2024. Results 53 patients with 65 sepsis events were included in the analysis. The median age was 14.9 [IQR 9.9] y and the most common disease was hematologic malignancy (71%). In our cohort, 60% needed vasopressor support, 36% required IMV, and 22% underwent renal replacement therapy (RRT). Log2-EASIX >2.5 was associated with higher vasopressor requirements (87% versus 45% in the low log2-EASIX group, p=0.001), and increased need for RRT (39% versus 12%, p=0.024. PHO patients with log2-EASIX >2.5 were 6.9 more likely to require vasopressor support [(95% CI 1.7-27.8) p=0.007]. In addition, PHO patients with log2-EASIX >2.5 had longer ICU stay (7 d versus 2 d in the low log2-EASIX group, p=0.024), and extended overall hospitalization (33 d versus 25 d, p=0.029). Conclusion Higher EASIX score was associated with adverse outcomes in critically ill septic PHO patients. Our findings suggest that EASIX score can be used as a tool for identifying septic patients at increased risk of clinical deterioration and poor outcomes. Prospective studies in larger cohorts are warranted to validate and expand upon these findings.

Sepsis is a common complication in pediatric hematology oncology (PHO) patients with a particularly high mortality rates of 41%-46% when they require ICU admission (1). These patients are especially vulnerable to infections due to their immunocompromised status and require immediate evaluation and treatment when sepsis is suspected. Rapid recognition and reversal of shock are crucial to reduce the risk of multiorgan dysfunction and death. Current Sepsis guidelines emphasize the urgency of early interventions, including fluid resuscitation and timely initiation of antimicrobial therapy (2). Screening tools- such as vital sign-based alerts and clinical assessments- are increasingly utilized to identify patients at risk. With the widespread adoption of electronic health records (EHR), automated alerts incorporating vital signs and risk indicators for severe illness have become integral in clinical workflows. In the context of sepsis, endothelial activation and injury contribute to the development of septic shock and the progression to multiorgan dysfunction. Early identification of septic patients at risk for poor outcomes supports timely interventions. Recently, the Endothelial Activation and Stress Index (EASIX) score, intro- duced by Luft et al, emerged as a potential biomarker for endothelial injury (3). Given that the EASIX score reflects endothelial injury, it may serve as a tool for identifying patients at an increased risk of developing 1 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. severe sepsis and experiencing poorer outcomes. Additionally, the EASIX score is a straightforward tool that relies on three commonly available laboratory values (creatinine, lactate dehydrogenase (LDH), and platelet count), making it an easy and practical screening tool to identify patients at risk of severe sepsis and adverse outcomes. Therefore, our hypothesis was that EASIX score at the time of sepsis could predict the need for vasopressor support in critically ill PHO patients. Additionally, this exploratory study aimed to investigate the association of the EASIX score with survival in this high-risk population. A secondary goal was to compare the predictive performance of the EASIX score with the modified EASIX (mEASIX) score and the simplified EASIX (sEASIX) score in this patient population. Patients and Methods Study Population All children admitted to the intensive care unit (ICU) at St. Jude Children’s Research Hospital, a specialized pediatric hematologic-oncology hospital, with sepsis between July 2022 and December 2024 were included in the study. Patients were excluded if serum creatinine, LDH, or platelet lab data was unavailable within the first 24 hours of ICU admission, or if a patient experienced multiple sepsis events during the same ICU stay-in which case only the first event was considered. If a patient experienced multiple sepsis events associated with separate ICU admissions, data from each of these events was included in the analysis. Data Collection The following ICU course data were collected: patient demographics, primary oncologic diagnosis, history of hematopoietic cell transplantation (HCT), Pediatric Risk of Mortality (PRISM) score, Pediatric Index of Mortality (PIM) 2, use of invasive mechanical ventilation (IMV) or positive pressure ventilation, use of vasopressor or inotropic infusion ( including norepinephrine, epinephrine, dobutamine, phenylephrine, vasopressin, dopamine, and milrinone), renal replacement therapy (RRT), ICU length of stay, and survival. In addition, laboratory data including serum creatinine, LDH, platelet, and C-reactive protein (CRP) within the first 24 hours of ICU admission were collected. Definitions Sepsis definition was based on the broadly accepted criteria of improving pediatric sepsis outcomes (IPSO) collaborative definitions. IPSO, a collaborative, multicenter quality improvement network sponsored by the Children’s Hospital Association, focuses on improving sepsis outcomes by prompting early recognition and timely treatment. Patients with “IPSO sepsis” were identified either by (1) the presence of a standalone In- ternational Classification of Diseases, 10th edition (ICD - 10) code for severe sepsis/septic shock or (2) the receipt of the following sepsis treatments (blood culture obtained within 72 hours and antibiotics delivered and either two fluid boluses administered OR bolus along with vasoactive agent given PLUS one of the following: ICU admission, lactate measurement, administration of vasoactive agent , use of infectious disease order set usage, or an ICD-10 code for sepsis assigned (4). EASIX score was calculated using the following formula: serum LDH level (U/L) × creatinine level (mg/dL)/platelet count (109/L) (3). Modified EASIX score was calculated based on the formula: serum LDH level (U/L)× CRP (mg/dL)/platelet count (109/L) (5). Simplified EASIX score was calculated using LDH level (U/L)/platelet count (109/L) (6). Statistical Analysis For the primary statistical analysis, all EASIX scores were transformed and analyzed using log2. Two cut points for each log2 EASIX, one for vasopressor classification and one for mortality classification, were derived to maximize the sum of the sensitivity and specificity (Youden method). Categorical variables were presented as frequencies with percentages, whereas median and interquartile range (IQR) were used to describe continuous variables. All continuous variables were analyzed without the assumption of normality usingtheKruskal-Wallisranktest(non-parametricttest).CategoricalvariableswerecomparedusingFisher’s exact test. 2 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. Odds ratios and the corresponding 95% confidence intervals were estimated using multivariable logistic re- gression models with continuous and dichotomized EASIX scores. Model adjustments were made for age, sex, and pre-sepsis HCT. Classification statistics were calculated using a model-estimated probability threshold for assignment to positive outcome when greater than 50%. Receiver operating characteristics (ROC) curves were used to assess the sensitivity, specificity and area under the ROC curve (AUC) of each EASIX score and each outcome (vasopressor use and mortality).

A total of 70 episodes of sepsis associated with ICU admission were identified. After applying exclusion criteria, 53 patients with 65 sepsis events were included in the analysis (Supplemental Figure 1). Median age in our cohort was 14.9 [IQR 9.9] y and the most common primary disease was hematologic malignancy (71%) and 48% received HCT (Table 1). In our cohort, 60% needed vasopressor support, 36% required IMV, and 22% underwent RRT. An infectious agent was identified in 69% of the cases, with bacterial infections comprising 53%, viral infections 14%, and fungal infections 2%. The survival to ICU discharge rate in our cohort was 78%. EASIX score and outcome The cut-off value of >2.5 was selected using Youden method to compare high log2-EASIX group versus the low log2-EASIX group for vasopressor use. High log2-EASIX was noted in 35% of the sepsis events (Table 1). The high log2-EASIX group had higher vasopressor requirements (87% versus 45% in the low log2-EASIX group, p=0.001), higher rate of bacterial infection (74% versus 41%, p=0.019), and increased need for RRT (39% versus 12%, p=0.024), along with an unsupported trend towards greater use of IMV (52% versus 27%, p=0.059). In addition, patients in the high log2-EASIX group experienced longer ICU stays (7 d versus 2 d in the low log2-EASIX group, p=0.024), and extended overall hospitalization (33 d versus 25 d, p=0.029). When evaluating log2-EASIX as a continuous variable, children who required vasopressor support had a higher median value compared to those who did not (2.7 versus 1.2, p=0.0002) (Figure 1). Multivariable analysis revealed that a higher log2-EASIX was associated with increased likelihood of vasopressor use with an odd ratio (OR) of 1.7 [(95% CI 1.2-2.4) p=0.005] (Figure 2). Furthermore, log2-EASIX cut-off value support [(95% CI 1.7-27.8) p=0.007]. This threshold demonstrated a sensitivity of 77%, a specificity of 50%, and a positive predictive value of 70% (Table 2). EASIX score and survival ICU survival rates did not differ significantly between the two groups (70% in the high log2-EASIX group versus 83% in the low log2-EASIX group, p=0.221) (Table 1). The median log2-EASIX was 3 among children who did not survive ICU discharge, versus 1.8 in those who survived (p=0.069). Overall survival was not statistically different in the high versus low log-2 EASIX (Figure 3). In addition, the multivariable analysis did not show an association of high log-2 EASIX and mortality (Figure 2). EASIX score versus modified EASIX score Each EASIX score showed comparable performance in predicting the need for vasopressor support, each achieving comparable area under the curve (AUC) (75% EASIX, 73% mEASIX, 76% sEASIX) (Figure 4). Likewise, the scores demonstrated similar AUC when predicting mortality (65% EASIX, 65% mEASIX, 70% sEASIX).

To the best of our knowledge, the present study is the first to investigate the association between EASIX and the need for vasopressor support in critically ill PHO septic patients. In our cohort, elevated EASIX score was significantly associated with the need for vasopressor support. PHO patients with log2-EASIX >2.5 had 6.9-fold increased likelihood of requiring vasopressors. Furthermore, log2-EASIX including a higher RRT requirement, prolonged ICU stays, and extended overall hospitalization duration. Previous studies have 3 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. demonstratedthatpre-transplantEASIXscoreareassociatedwithincreasedmortalityinpatientsundergoing HCT (7). Similarly, in a cohort of 7504 adult ICU patients with sepsis, a higher log2-EASIX was associated with increased risk of 28-day and 90-day mortality (8). In contrast, within our cohort, EASIX score was not significantly associated with ICU survival. Since the EASIX score reflects endothelial injury, it is plausible that higher scores predict severe sepsis phe- notypes, characterized by the need for vasopressor support, RRT, and prolonged ICU stay. This aligns with the central role the endothelium plays in sepsis pathogenesis. The endothelium is a dynamic organ involved in key homeostatic mechanisms including inflammation, coagulation, and vascular tone (9). Endothelial cells express adhesion molecules that facilitate neutrophil transendothelial migration in response to cytokine si- gnaling. They also actively secrete inflammatory cytokines and chemokines, amplifying the immune response. While this amplification can help fight infection, it can also be detrimental due to pathologic inflammation (9). In sepsis, the normal mechanisms that preserve endothelial disruption are compromised, resulting in increased vascular permeability and interstitial edema-key drivers of the organ dysfunction seen in septic patients (9). The endothelium also contributes to the activation of the coagulation system during sepsis, shif- ting the balance toward a pro-coagulant state. This can manifest as disseminated intravascular coagulation, thrombotic thrombocytopenic purpura or thrombocytopenia-associated multi-organ failure, all of which are linked to poor outcomes (10). Additionally, the endothelium plays a crucial role in vascular tone regulation. Reactive oxygen species and reactive nitrogen species produced during sepsis can damage the endothelial glycocalyx, disrupting the endothelium’s ability to properly manage the balance between vasodilatation and vasoconstriction (10). Modified EASIX score has been reported in some studies to be a predictor of cytokine release syndrome (CRS), immune effector cell–associated neurotoxicity syndrome (ICANS) after CD19-directed chimeric an- tigen receptor (CD19-CAR) T-cell therapy (11, 12, 13). The mEASIX score outperformed the EASIX score in these patients (12, 13). In our cohort, however, EASIX, sEASIX, and mEASIX demonstrated comparable performance in predicting the need for vasopressor support. Therefore, any of these scores can help evaluate the risk of requiring vasopressor support, based on the patient’s available laboratory data. While several biomarkers of endothelial disruption have shown strong associations with poor outcomes in sepsis, their clinical utility at the bedside is limited as they are not readily available (14). In contrast, the EASIX score is derived from routinely obtained laboratory, enabling its use at the bedside. This makes it a practical and accessible tool for identifying septic patients at increased risk of clinical deterioration and poor outcomes. The limitations of our study include its retrospective design, relatively small sample size, and the absence of a control group. Nonetheless, this is the first study to evaluate the predictive utility of EASIX score in septic PHO patients. Prospective studies in larger cohorts are warranted to validate and expand upon these findings. In summary, EASIX score patients predicted the need of vasopressor support and RRT in septic PHO. If validated in a larger cohort, incorporating the EASIX score into sepsis screening protocols could aid in the early identification of high-risk patients who may benefit from closer monitoring and potentially help distinguish between distinct sepsis phenotypes. 6 Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relation- ships that could be construed as a potential conflict of interest. 7 Author Contributions LE and MCA contributed to planning, writing, and editing the manuscript; MK contributed to data collec- tion, writing, and editing the manuscript; CB contributed to data analysis and manuscript writing; and SG, SJ, GM, RD contributed to editing the manuscript. All authors approved the submitted version. 4 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. 8 Funding This research was funded by the American Lebanese Syrian Associated Charities (ALSAC). Figure 1.The box plot of the Endothelial Activation and Stress Index (EASIX) Score. Dots represent outlier values of EASIX Figure 2.Endothelial Activation and Stress Index (EASIX) Score Associations with Vasopressor Use and Death Probabilities Figure 3.Kaplan–Meier curve for low vs high EASIX (cut-point Figure4. ROCcurvesforvasopressoruse(panelA)andin-ICUmortality(panelB)ofEndothelialActivation and Stress Index (EASIX) Score and a modified version that replaces measured serum creatinine with C- reactive protein .

1. Agulnik A. Management of septic shock in children with cancer-Common challenges and research priorities. J Pediatr (Rio J). 2023;99(2):101-4.2. Weiss SL, Peters MJ, Alhazzani W, Agus MSD, Flori HR, Inwald DP, et al. Surviving sepsis campaign international guidelines for the management of septic shock and sepsis- associated organ dysfunction in children. Intensive Care Med. 2020;46(Suppl 1):10-67.3. Luft T, Benner A, Jodele S, Dandoy CE, Storb R, Gooley T, et al. EASIX in patients with acute graft-versus-host disease: a retrospective cohort analysis. Lancet Haematol. 2017;4(9):e414-e23.4. Scott HF, Brilli RJ, Paul R, Macias CG, Niedner M, Depinet H, et al. Evaluating Pediatric Sepsis Definitions Designed for Electronic Health Record Extraction and Multicenter Quality Improvement. Crit Care Med. 2020;48(10):e916-e26.5. Zi´ nczuk A, Rorat M, Simon K, Jurek T. EASIX, Modified EASIX and Simplified EASIX as an Early Predictor for Intensive Care Unit Admission and Mortality in Severe COVID-19 Patients. J Pers Med. 2022;12(7).6. Tomasik J, Avni B, Grisariu S, Elias S, Zimran E, Stepensky P, et al. Endothelial Activation and Stress Index Score as a Prognostic Factor of Cytokine Release Syndrome in CAR-T Patients - A Retrospective Analysis of Multiple Myeloma and Large B-Cell Lymphoma Cohorts. Arch Immunol Ther Exp (Warsz). 2024;72(1).7. Sanchez-Escamilla M, Flynn J, Devlin S, Maloy M, Fatmi SA, Tomas AA, et al. EASIX score predicts inferior survival after allogeneic hematopoietic cell transplantation. Bone Marrow Transplant. 2023;58(5):498-505.8. Xu HB, Ye Y, Xue F, Wu J, Suo Z, Zhang H. Association Between Endothelial Activation and Stress Index and 28-Day Mortality in Septic ICU patients: a Retrospective Cohort Study. Int J Med Sci. 2023;20(9):1165-73.9. Dolmatova EV, Wang K, Mandavilli R, Griendling KK. The effects of sepsis on endothelium and clinical implications. Cardiovasc Res. 2021;117(1):60-73.10. Ince C, Mayeux PR, Nguyen T, Gomez H, Kellum JA, Ospina-Tasc´ on GA, et al. THE ENDOTHELIUM IN SEPSIS. Shock. 2016;45(3):259-70.11. Zhao Y, Zhang X, Zhang M, Guo R, Zhang Y, Pu Y, et al. Modified EASIX scores predict severe CRS/ICANS in patients with acute myeloid leukemia following CLL1 CAR-T cell therapy. Ann Hematol. 2024;103(3):969-80.12. Pennisi M, Sanchez-Escamilla M, Flynn JR, Shouval R, Alarcon Tomas A, Silverberg ML, et al. Modified EASIX predicts severe cytokine release syndrome and neurotoxicity after chimeric antigen receptor T cells. Blood Adv. 2021;5(17):3397-406.13. Zandaki D, Selukar S, Bi Y, Li Y, Zinsky M, Bonifant CL, et al. EASIX and m-EASIX predict CRS and ICANS in pediatric and AYA patients after CD19-CAR T-cell therapy. Blood Adv. 2025;9(2):270-9.14. Fern´ andez-Sarmiento J, Molina CF, Salazar- Pelaez LM, Fl´ orez S, Alarc´ on-Forero LC, Sarta M, et al. Biomarkers of Glycocalyx Injury and Endothelial Activation are Associated with Clinical Outcomes in Patients with Sepsis: A Systematic Review and Meta- Analysis. J Intensive Care Med. 2023;38(1):95-105. Table 1. Characteristics of PHO in the low log-2 EASIX group versus the high log-2 EASIX group log2(EASIX Score)>2.5 Low EASIX High EASIX Total p-value 5 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. log2(EASIX Score)>2.5 n=42 (65%) n=23 (35%) n=65 (100%) Age at ICU Admission (y) 14.5 [10.8] 15.1 [10.1] 14.9 [9.9] 0.641 Sex Female 17 (40%) 15 (65%) 32 (49%) 0.072 Male 25 (60%) 8 (35%) 33 (51%) Race White 27 (64%) 15 (65%) 42 (65%) 0.800 Black 9 (21%) 6 (26%) 15 (23%) Other 6 (14%) 2 (9%) 8 (12%) Ethnic Group Not Hispanic, Latino/a, or Spanish Origin 30 (71%) 18 (78%) 48 (74%) 0.598 Hispanic, Latino, or Spanish Origin 11 (26%) 4 (17%) 15 (23%) South or Central American 1 (2%) 1 (4%) 2 (3%) Diagnosis Group Hematologic Malignancy 31 (74%) 15 (65%) 46 (71%) 0.608 Benign Hematologic Disease 6 (14%) 3 (13%) 9 (14%) Solid Tumor 5 (12%) 5 (22%) 10 (15%) Bone Marrow Transplant (BMT) No 24 (57%) 10 (43%) 34 (52%) 0.313 Yes 18 (43%) 13 (57%) 31 (48%) BMT After Sepsis No 18 (75%) 13 (87%) 31 (79%) 0.450 Yes 6 (25%) 2 (13%) 8 (21%) Sepsis After ICU Admission No 36 (86%) 17 (74%) 53 (82%) 0.319 Yes 6 (14%) 6 (26%) 12 (18%) Vasopressor Use No 23 (55%) 3 (13%) 26 (40%) 0.001 Yes 19 (45%) 20 (87%) 39 (60%) Death No 35 (83%) 16 (70%) 51 (78%) 0.221 Yes 7 (17%) 7 (30%) 14 (22%) Supplemental Oxygen No 3 (7%) 1 (4%) 4 (6%) 1.000 Yes 39 (93%) 22 (96%) 61 (94%) ECMO No 41 (98%) 21 (95%) 62 (97%) 1.000 Yes 1 (2%) 1 (5%) 2 (3%) Any Infection No 16 (39%) 4 (17%) 20 (31%) 0.095 Yes 25 (61%) 19 (83%) 44 (69%) Bacterial Infection No 24 (59%) 6 (26%) 30 (47%) 0.019 Yes 17 (41%) 17 (74%) 34 (53%) Viral Infection No 33 (80%) 22 (96%) 55 (86%) 0.140 Yes 8 (20%) 1 (4%) 9 (14%) Fungal Infection No 41 (100%) 22 (96%) 63 (98%) 0.359 Yes 0 (0%) 1 (4%) 1 (2%) 6 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. log2(EASIX Score)>2.5 IMV No 30 (73%) 11 (48%) 41 (64%) 0.059 Yes 11 (27%) 12 (52%) 23 (36%) RRT No 37 (88%) 14 (61%) 51 (78%) 0.024 Yes 5 (12%) 9 (39%) 14 (22%) Lactated Ringer’s (LR) No 12 (29%) 6 (26%) 18 (28%) 1.000 Yes 30 (71%) 17 (74%) 47 (72%) ICU Length of Stay (d) 2.0 [10.0] 7.0 [20.0] 2.0 [15.0] 0.024 Hospital Length of Stay (d) 24.5 [36.0] 33.0 [37.0] 27.0 [37.5] 0.029 Platelet Count (103/mm3) 79.0 [119.0] 17.0 [37.0] 45.0 [104.0] <0.001 C-Reactive Protein (mg/dL) 7.7 [13.1] 16.3 [19.1] 11.2 [18.7] 0.022 LDH (U/L) 288.5 [310.0] 363.0 [1615.0] 314.0 [417.0] 0.063 Creatinine (mg/dL) 0.5 [0.4] 0.8 [0.9] 0.5 [0.4] 0.002 PIM2 -3.0 [0.4] -3.0 [1.8] -3.0 [1.5] 0.428 PRISM3 5.5 [7.5] 14.0 [14.0] 8.0 [11.0] 0.004 Cells represent the median [IQR] for continuous data and frequency (%) for categorical data. Continuous data tested using Kruskal-Wallis rank test and Categorical data tested using Fisher’s exact test. Totals in tables refer to number of sepsis cases. PHO, pediatric hematology oncology; EASIX, Endothelial Activation and Stress Index; ICU, intensive care unit; HCT, hematopoietic cell transplant; IMV, invasive mechanical ventilation; RRT, renal replacement therapy; LDH, lactate dehydrogenase; PIM, Pediatric Index of Mortality; PRISM, Pediatric Risk of Mortality Table 2.The classification performance of each EASIX cut point OR (95% CI) p-value AUC Sens Spec PPV NPV Correctly Classified Vasopresser Use Continuous log2(EASIX) 1.7 (1.2,2.4) p=0.0045 75% - - - - - Continuous log2(sEASIX) 1.7 (1.1,2.5) p=0.0092 73% - - - - - Continuous log2(mEASIX) 1.5 (1.1,1.9) p=0.0029 76% - - - - - log2(EASIX)>2.5 6.9 (1.7,27.8) p=0.0070 - 77% 50% 70% 59% 66% log2(sEASIX)>2.5 4.9 (1.5,16.3) p=0.0089 - 82% 54% 73% 67% 71% log2(mEASIX)>7 8.8 (2.1,37.2) p=0.0031 - 74% 69% 78% 64% 72% Death Continuous log2(EASIX) 1.0 (0.7,1.5) p=0.9118 65% - - - - - Continuous log2(sEASIX) 0.9 (0.6,1.4) p=0.7133 65% - - - - - Continuous log2(mEASIX) 1.0 (0.7,1.5) p=0.9776 70% - - - - - log2(EASIX)>3.5 2.3 (0.4,12.2) p=0.3341 - 71% 94% 77% 92% 89% log2(sEASIX)>3.5 0.8 (0.1,4.1) p=0.7684 - 64% 96% 82% 91% 89% log2(mEASIX)>7 2.1 (0.4,10.7) p=0.3515 - 57% 94% 73% 89% 86% EASIX, Endothelial Activation and Stress Index; OR, odds ratio; Sens, sensitivity; Spec, specificity; PPV, positive predictive value; NPV, negative predictive value 7 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. 8 Posted on 2 Jun 2025 | The copyright holder is the author/funder. All rights reserved. No reuse without permission. | https://doi.org/10.22541/au.174884427.71406195/v1 | This is a preprint and has not been peer-reviewed. Data may be preliminary. 9