Molecular Phenotyping of Patients with Sepsis and Kidney Injury and Differential Response to Fluid Resuscitation

Molecular Phenotyping of Patients with Sepsis and Kidney Injury and Differential Response to Fluid Resuscitation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Molecular Phenotyping of Patients with Sepsis and Kidney Injury and Differential Response to Fluid Resuscitation Elizabeth Kiernan, Leila R. Zelnick, Ayesha Khader, Taylor D. Coston, and 11 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4523416/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Purpose Previous work has identified two AKI sub-phenotypes (SP1 and SP2) characterized by differences in inflammation and endothelial dysfunction. Here we identify these sub-phenotypes using biospecimens collected in the emergency department and test for differential response to restrictive versus liberal fluid strategy in sepsis-induced hypotension in the CLOVERS trial. Methods We applied a previously validated 3-biomarker model using plasma angiopietin-1 and 2, and soluble tumor necrosis factor receptor-1 to classify sub-phenotypes in patients with kidney dysfunction (AKI or end-stage kidney disease [ESKD]). We also compared a de novo latent class analysis (LCA) to the 3-biomarker based sub-phenotypes. Kaplan-Meier estimates were used to test for differences in outcomes and sub-phenotype by treatment interaction. Results Among 1289 patients, 846 had kidney dysfunction on enrollment and the 3-variable prediction model identified 605 as SP1 and 241 as SP2. The optimal LCA model identified two sub-phenotypes with high correlation with the 3-biomarker model (Cohen’s Kappa 0.8). The risk of 28 and 90-day mortality was greater in SP2 relative to SP1 independent of AKI stage and SOFA scores. Patients with SP2, characterized by more severe endothelial injury and inflammation, had a reduction in 28-day mortality with a restrictive fluid strategy versus a liberal fluid strategy (26% vs 41%), while patients with SP1 had no difference in 28-day mortality (10% vs 11%) ( p-value-for-interaction = 0.03). Conclusion Sub-phenotypes can be identified in the emergency department that respond differently to fluid strategy in sepsis. Identification of these sub-phenotypes could inform a precision-guided therapeutic approach for patients with sepsis-induced hypotension and kidney injury. acute kidney injury sepsis volume vasopressors Figures Figure 1 Figure 2 Figure 3 Take Home Message We applied a 3 biomarker prediction model to the CLOVERS analysis and showed feasbility of identifying sub-phenotypes of patients with sepsis. These sub-phenotypes are clinically distinct and respond differently to fluid resuscitation in sepsis. Identification of sub-phenotypes could inform precision-guided therapy for patients with sepsis-associated hypotension and kidney injury. INTRODUCTION Acute kidney injury (AKI) is the most common form of organ failure in sepsis developing in as many as 60% in septic patients with hypotension 1,2 . Sepsis-associated AKI is associated with prolonged hospitalization, need for kidney replacement therapy, future chronic kidney disease (CKD) and death 3–8 . At present, providers base treatment decisions in AKI on parameters such as urine output and serum creatinine, yet these biomarkers fail to differentiate the highly heterogenous clinical syndrome of AKI and cannot reliably predict response to common interventions such as volume resuscitation and vasopressors in sepsis 9,10 . We previously applied latent class analysis (LCA) to two cohorts to identify two distinct sub-phenotypes in patients with sepsis-associated AKI, and subsequently developed a parsimonious 3-variable model for clinical application 11 . This 3-variable model included plasma biomarkers of endothelial function (angiopoietin-1 (Ang-1) and angiopoietin-2 (Ang-2)) and inflammation (soluble tumor necrosis factor receptor-1 [sTNFR-1]) 11 . Ang-1 and -2 are vascular endothelial growth factors that have opposing activities. Ang-1 is an agonist for the Tie-2 receptor and is protective by stabilizing the endothelium, promoting vessel maturation, and preventing microcirculatory capillary leakage. In contrast, Ang-2 typically acts as an antagonist for binding to the Tie-2 receptor and promotes endothelial dysfunction in the form of vessel destabilization, endothelial cell apoptosis, capillary leak, and deregulation of inflammation 12–14 . Independently, sTNFR-1 functions in the early innate inflammatory response in sepsis and has been associated with AKI and mortality 15 . Our model established two distinct AKI sub-phenotypes: those with a lower ratio of Ang-2/Ang-1 and low sTNFR-1, designated SP1 and characterized by endothelial stability and improved short and long-term outcomes. In contrast, those with higher ratio of Ang-2/Ang-1 and high sTNFR-1, or SP2 phenotype, were characterized by high inflammatory state and leaky endothelium and worse clinical outcomes 11,16–18 . The Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis (CLOVERS) clinical trial randomized patients with sepsis to an initial restrictive or liberal fluid resuscitation strategy and demonstrated no differences in clinical outcomes between treatment arms or in the sub-groups with AKI or end-stage kidney disease (ESKD) 19 . Early and aggressive fluid resuscitation is a cornerstone of sepsis management which aims to improve hypoperfusion caused by increased capillary leak and endothelial permeability. In the setting of kidney injury or disease, however, the benefit of fluid resuscitation is counterbalanced by impaired volume management. We hypothesized that liberal fluid resuscitation would be preferentially harmful to septic patients with evidence of both endothelial injury and kidney disease, and that application of the SP1 or SP2 sub phenotypes may enable identification of patients most likely to benefit from early vasoconstrictive therapy. To address these questions, we first identified sub-phenotypes with known kidney injury (AKI or ESKD) using plasma samples collected in the emergency department, representing an early and critical time for treatment implementation in sepsis. Second, in a sensitivity analysis, we verified sub-phenotypes by applying LCA to 23 variables collected at study enrollment and compared sub-phenotype identification to the 3-variable model. Third, since serum creatinine can lag up to 48 hours and is insensitive for kidney injury 20 , we conducted a sensitivity analysis including patients without AKI on study enrollment to test whether the treatment effect by sub-phenotype was preserved. Fourth, to understand the overlap and differences between different sub-phenotypes in critical illness, we cross-tabulated our sub-phenotypes with a previously established hypo- and hyperinflammatory acute respiratory distress syndrome (ARDS) sub-phenotypes. METHODS Study Design and Oversight We conducted a secondary analysis of data from CLOVERS, a multicenter, prospective, phase 3, randomized, non-blinded trial. Patients enrolled in CLOVERS were allocated to either a restrictive or liberal fluid resuscitation strategy in the setting of sepsis. All patients were randomized within 4 hours of meeting inclusion criteria, and duration of protocol for restrictive or liberal fluid resuscitation was 24 hours. Patients randomized to a restrictive fluid protocol had vasopressors prioritized as the primary treatment for sepsis-induced hypotension, with “rescue fluids” being permitted for prespecified indications that suggested severe intravascular volume depletion. The liberal fluid protocol consisted of a recommended initial 2 L IV infusion of isotonic crystalloid, followed by fluid boluses administered based on clinical triggers with “rescue vasopressors” permitted for prespecified indications. A more detailed description of methods can be found in the Supplement of the primary manuscript 19 . AKI Definition AKI was defined using the KDIGO criteria as an increase in serum creatinine at randomization of ≥50% or ≥0.3 mg/dL above a baseline serum creatinine. The lowest outpatient or inpatient serum creatinine value within the last year prior to the index hospitalization was used for the baseline when available (N=909). If no pre-hospitalization serum creatinine was available and there was no known history of CKD, then we imputed a baseline serum creatinine based on an eGFR of 90 ml/min/1.73 m2 (N=517) using the 2021 creatinine-based CKD-EPI equation 11 . If no pre-hospitalization serum creatinine was available and there was a known history of CKD, then we used the lowest serum creatinine value during the index hospitalization as the baseline (N=29). Patients were subsequently staged according to the serum creatinine at randomization using the KDIGO guidelines. Seventy-five patients were on maintenance hemodialysis prior to study enrollment, (ESKD). Since ESKD patients represent severe impairment in kidney-mediated volume management and guidelines do not offer guidance on distinct therapeutic approaches across the spectrum of ESKD, we included them in the analysis. AKI Sub-phenotype Identification Methods for measurement of Ang-1, Ang-2 and sTNFR-1 can be found in the online supplement. We used plasma biomarker concentrations of Ang-1, Ang-2 and sTNFR-1 to identify two sub-phenotypes (SP1 and SP2). Since the publication of our prior works, Meso Scale Discovery changed the sTNFR-1 assay from a R-Plex to a U-Plex platform. In analyses provided in the supplement we found the correlation of these two assays to be high (Pearson’s correlation, r = 0.94 ), and we created a calibration equation ( Figure S1) . We updated the 3-variable prediction model that initially used R-Plex measured sTNFR-1 to include U-Plex measured values to account for the updated sTNFR-1 biomarker platform. The model to identify AKI sub-phenotypes is TNFR_R-Plex = -101.3387 +0.3617*TNFR_U-Plex followed by the previously published prediction equation Logit(P(SP2)) = -35.44+1.99*log(Ang-2/Ang-1) + 3.41*log(TNFR_R-Plex). Participants with a probability of SP2 > 0.5 were then classified as having the SP2 sub-phenotype. Methods of LCA applied to the baseline variables can be found in the Methods in the supplement. Previously described ARDS sub-phenotypes were identified using a published four variable model that included plasma concentrations of interleukin-6, sTNFR-1, serum bicarbonate and clinical documentation of receipt of vasopressors prior to randomization 11,21 . Outcome Measures The primary outcome for this analysis was 28-day and 90-day mortality. Additional outcomes included need for invasive mechanical ventilation (IMV), need for new KRT, and ICU-free days through 28 days of follow-up. Statistical Analysis We compared the risk of death by sub-phenotype over 28 and 90 days of follow-up using Cox regression, adjusting in nested models for potential confounders. To test for heterogeneity of treatment effect, we completed an intention-to-treat comparison of the treatment effect of restrictive or liberal fluid resuscitation strategy on 28-day and 90-day mortality by sub-phenotype. For this analysis, we used Kaplan-Meier 28-day and 90-day mortality point estimates involving all patients who were discharged home or were still alive at day 90, with data censored at day 91. Patients who were lost to follow-up prior to day 90 were censored at the time they were last confirmed to be alive. We compared 28-day and 90-day mortality point estimates and evaluated the evidence for a treatment by sub-phenotype interaction using a non-parametric bootstrap approach with 10,000 replicates. Secondary outcomes were the time to IMV among patients who were not receiving IMV at randomization and the time to new KRT among patients not on dialysis at the time of randomization. For these outcomes, we used a Fine-Gray sub-distribution hazard model to estimate the sub-distribution hazard ratio of each outcome accounting for the competing risk of death. Finally, we used linear regression to estimate the difference in ICU-free days through day 28 between the two treatment groups. All analyses were conducted using the R 4.2.3 software environment (R Foundation for Statistical Computing, Vienna, Austria). Two-sided p-values < 0.05 were taken as evidence of statistical significance. RESULTS Baseline Characteristics of Participants by Sepsis-associated AKI Sub-phenotype Among 1563 patients enrolled in CLOVERS, 1289 had biospecimens collected prior to randomization and available for analysis ( Figure 1 ). Among the 1289 patients, 846 (66%) had kidney dysfunction (771 had AKI and 75 had ESKD) at study enrollment and 443 (34%) did not have AKI or ESKD. Among the 846 patients with kidney dysfunction the previously published sub-phenotype classification model, incorporating plasma Ang-1, Ang-2, and sTNFR-1 measurements at randomization, classified 605 (72%) patients as SP1 and 241 (28%) as SP2 (Table 1) . As expected by randomization, the distribution of SP1 and SP2 were similar between treatment arms. Patients with SP2 had more severe KDIGO stage of AKI on study enrollment compared to SP1 ( Figure S2 ). To determine whether SP2 is enriched in the AKI population, we applied the 3-biomarker prediction model to patients without AKI or ESKD on study enrollment. Among 443 patients, we found that 411 (93%) were classified as SP1 and 32 (7%) as SP2. The 32 patients with SP2 were characterized by high rates of CKD and ten patients subsequently developed AKI within 48 hours of study enrollment based on changes in serum creatinine ( Table S1 ). In a sensitivity analysis, we applied LCA to 23-variables collected at study enrollment and found that a 2-class model best separated the data ( Table S3 ). We also found high correlation between the LCA derived sub-phenotypes and the 3-variable biomarker sub-phenotypes, Cohen’s Kappa of 0.80. With the high correlation and to support prospective identification of these sub-phenotypes, we used the 3-variable derived sub-phenotypes for the subsequent analyses. Clinical Outcomes by Sub-Phenotype To evaluate the association of sub-phenotypes with clinical outcomes, patients without AKI were compared against those with SP1 and SP2 and then patients with SP1 and SP2 were directly compared ( Table S4 ). In models adjusting for demographics, comorbidities, KDIGO stage of AKI or ESKD and baseline sequential organ failure assessment (SOFA) scores, patients with SP1 had a similar risk of new KRT, ICU length of stay, receipt of IMV, 28-day and 90-day mortality as patients with no AKI. Patients with SP2 had greater risk of 28-day (HR = 2.33 (95% CI: 1.60, 3.38) and 90-day mortality (HR = 1.98 (95% CI: 1.45, 2.71) than SP1. Patients with SP2 were more likely than those with SP1 to require new KRT (sHR=3.03 (95% CI: 1.42, 6.47) or IMV after study enrollment (sHR=1.54 (95% CI: 1.00, 2.37) ( Table S5) . Those with SP2 had ICU stays that were 3 days longer than those with SP1 (adjusted difference 3.1 days (95% CI: 1.8 to 4.5 days) ( Table S7 ). Difference in Volume of Fluid and Timing of Vasopressors by CLOVERS Randomization In the 24-hour protocol following randomization, patients with SP1 randomized to the liberal resuscitation strategy received on average 2 L more of IV fluids than patients with SP1 randomized to the restrictive resuscitation strategy (3,684 ± 1,644 mL vs 1,684 ± 1657 mL) ( Table S6 ). Similarly, patients with SP2 randomized to the liberal resuscitation strategy received on average 4,091 (± 1793) mL compared to 2,244 (± 2,305) mL in the restrictive strategy. Consistent with the trial protocol, patients randomized to the restrictive strategy received vasopressors more often than patients in the liberal strategy across both sub phenotypes. Interaction Between Sub-phenotype and Resuscitation Strategy Next, we determined whether sub-phenotypes identified using biospecimens collected prior to randomization respond differently to the fluid resuscitation strategy. In SP1, a similar proportion of patients died at 28 days with liberal and restrictive fluid strategies (11% vs 10%, respectively) ( Figure 2 ). In SP2 a liberal strategy compared with a restrictive strategy led to greater 28-day mortality (41% vs 26% respectively) and the difference in treatment effects between SP1 and SP2 was 14% (95% CI: 2%, 27%); p-value for interaction = 0.03 . This effect was not observed in the patients without AKI ( Table 2 ). Among patients with SP1, 90-day mortality with a liberal and restrictive fluid strategies was 18% versus 17%, while SP2 was 48% versus 26% (p-value for interaction = 0.06 ). We did not observe a difference for alternative outcomes among sub-phenotypes by treatment ( Table S8 ). Next, we evaluated the continuous probability of belonging to SP2 and the interaction of a liberal versus a restrictive fluid resuscitation. We observed a significant interaction between the probability of sub-phenotype and fluid resuscitation strategy and 28-day mortality ( p=0.008 ) ( Figure 3 ). As patients had a greater certainty of the SP2 phenotype (i.e. higher probability of SP2 based on the 3-biomarker model), we observed a larger difference in 28-day mortality between a liberal versus restrictive fluid resuscitation strategy. In the primary CLOVERS publication, there was no observed heterogeneity of treatment effect based on SOFA scores or ESKD status 19 . In sensitivity analyses, we evaluated sub-phenotypes restricted to the AKI population and excluded ESKD. Among patients with SP1, 28-day mortality with liberal and restrictive fluid strategies was 10% versus 10% and in patients with SP2 was 39% versus 29% (p-value for interaction = 0.16) ( Table S9 ). In another sensitivity analysis, we used the 3-biomarker model to classify all patients with plasma available in CLOVERS as SP1 and SP2. Among patients with SP1, 28-day mortality with a liberal and restrictive fluid strategies was 9% versus 9% and in patients with SP2 was 41% versus 27% (p-value for interaction = 0.02 ). We also observed a similar decrease in 90-day mortality in SP2 ( Table 2 ). Cross-Tabulation of AKI and ARDS Sub-phenotypes We directly compared the cross-tabulation of SP1 and SP2 and ARDS sub-phenotypes (hypoinflammatory and hyperinflammatory). Cross-tabulation demonstrated mild to moderate overlap between hypoinflammatory and SP1 and hyperinflammatory and SP2 but also demonstrated distinct differences (Cohen’s Kappa for Agreement of 0.42) ( Table S10 ). Out of 846 patients, 77.0% were concordant (i.e. hypoinflammatory and SP1 or hyperinflammatory and SP2), and 23.0% were classified discordantly. We observed no differential response to a restrictive or liberal fluid resuscitation strategy between ARDS sub-phenotypes for 28-day (p-value for interaction = 0.25) or 90-day mortality (p-value for interaction = 0.60) ( Table S11 ). DISCUSSION The heterogeneity of AKI has limited insight into clinical trajectory and hindered development of interventions that target an individual patient’s pathophysiology 22–25 . We completed a secondary analysis of the CLOVERS trial and demonstrated that sub-phenotypes previously defined in ICU patients were also present in the emergency department, an earlier and critical window for therapeutic decisions in sepsis treatment. These sub-phenotypes predicted differential treatment response to liberal versus restrictive resuscitation strategies among patients with kidney dysfunction; revealing a treatment effect which was not seen in the original CLOVERS study. Taken together, these data offer a precision medicine treatment strategy in sepsis-induced hypotension and kidney dysfunction based on a 3-biomarker prediction model which may identify populations with differential response to fluid resuscitation strategy. Despite a long-held belief by clinicians that a rise in serum creatinine early during sepsis should result in additional fluid therapy 9 , patients with SP2 and kidney dysfunction had a greater mortality with a liberal fluid strategy compared to a restrictive fluid strategy. A difference in clinical outcomes between fluid strategy was not observed in patients without AKI/ESKD or patients with SP1. While these findings are hypothesis-generating, it suggests that early in sepsis the combination of kidney dysfunction and endothelial injury may identify a sub-phenotype that is harmed by a liberal fluid therapy. In addition, the observed heterogeneity of treatment effect was not present when identifying sub-groups based on SOFA scores, ESKD or AKI status 26,27 , suggesting that the sub-phenotypes capture information that is not readily available by clinical definitions. In this manner, application of the 3-biomarker prediction model to all patients with sepsis-induced hypotension may improve generalizability and time to sub-phenotype identification to facilitate future clinical trials without misclassification of patients with SP2. Our prior work leveraged these AKI sub-phenotypes in the Vasopressin and Septic Shock Trial (VASST), a randomized control trial investigating early addition of vasopressin to norepinephrine in sepsis. In that analysis, we showed that in SP1, early addition of vasopressin compared to norepinephrine alone was associated with improved 90-day mortality, but in SP2, vasopressin showed no significant treatment difference. Extrapolating these findings with the current CLOVERS analysis, we hypothesize that patients with the SP1 would derive benefit from either a restrictive or liberal fluid resuscitation and early addition of vasopressin with norepinephrine for persistent hypotension, while those with the SP2 may benefit from an initial restrictive fluid strategy and monotherapy with norepinephrine. Our study has several strengths. First, a majority of patients had a pre-hospitalization serum creatinine value available to approximate baseline kidney function and to accurately determine AKI status early after hospital presentation. Leveraging this data highlighted that approximately two-thirds of patients in the trial had AKI on hospital presentation for sepsis-induced hypotension. These findings are consistent with prior studies of AKI in septic shock and highlight that prevention of AKI in sepsis may be infeasible and therapies that target recovery should be prioritized 28,29 . Second, we were able to leverage the randomization in the CLOVERS trial to address limitations of indication bias of fluid administration in observational studies to demonstrate that patients with SP2, characterized by greater endothelial dysfunction and inflammation, have an improved mortality with a restrictive resuscitation strategy compared to a liberal strategy. Third, the CLOVERS trial included patients with ESKD, a group that is often excluded from large randomized clinical trials 30 . While limited by small group size, our data suggest that most patients with ESKD have a sub-phenotype consistent with SP2 (79%) and may benefit from a restrictive fluid resuscitation strategy. This analysis should be interpreted in the context of its limitations. First, there was variation in the amount of fluid received in the CLOVERS trial in the restrictive and liberal fluid groups and overlap of total volumes between groups may dampen a potential therapeutic signal. Second, this was a retrospective study of a completed clinical trial. However, the study design and prediction model for sub-phenotype identification were defined a priori . Third, we underscore that our work does not preclude that alternative AKI sub-phenotypes are present. Fourth, elevations of angiopoietins and sTNFR-1 are not specific to AKI. However, a large body of preclinical and clinical studies have demonstrated the importance of these pathways and biomarkers in kidney diseases, such as AKI and CKD 13,31,32 . Fifth, serum creatinine concentrations were not collected after day 3 of participation in CLOVERS and so we are unable to draw conclusions about differential rate of kidney recovery between sub-phenotypes. In summary, this work posits that early identification of AKI sub-phenotypes can inform clinical decision making in patients with sepsis-induced hypotension. We observed that SP2 had improved 28-day mortality with a restrictive fluid resuscitation strategy. Identification of AKI sub-phenotypes may facilitate development of targeted treatment interventions in critically ill patients. Declarations Author Contribution : Conception and design: AK, LZ, CH, PB Enrollment of subjects, collection of data and completion of measurements: NJ, NS, ID, CH, PB. Analysis and interpretation: All authors Drafting the manuscript for important intellectual content: All authors All authors read and approved this manuscript. Funding Support: Funding was received from the NIDDK K23DK116967 and R01DK133177 (PKB) Data Sharing Statement: Data available within the article or its supplementary materials DISCLOSURE STATEMENT: The author declares that they had no relevant or material financial interests that relate to the research described in this paper . References Bhatraju PK, Robinson-Cohen C, Mikacenic C, et al. Circulating levels of soluble Fas (sCD95) are associated with risk for development of a nonresolving acute kidney injury subphenotype. Crit Care . 2017;21(1). doi:10.1186/S13054-017-1807-X Kellum JA, Chawla LS, Keener C, et al. The Effects of Alternative Resuscitation Strategies on Acute Kidney Injury in Patients with Septic Shock. Am J Respir Crit Care Med . 2016;193(3):281-287. doi:10.1164/RCCM.201505-0995OC Joannidis M, Metnitz PGH. Epidemiology and natural history of acute renal failure in the ICU. Crit Care Clin . 2005;21(2):239-249. doi:10.1016/J.CCC.2004.12.005 Hoste EAJ, Bagshaw SM, Bellomo R, et al. Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. Intensive Care Med . 2015;41(8):1411-1423. doi:10.1007/S00134-015-3934-7 Clermont G, Acker CG, Angus DC, Sirio CA, Pinsky MR, Johnson JP. Renal failure in the ICU: comparison of the impact of acute renal failure and end-stage renal disease on ICU outcomes. Kidney Int . 2002;62(3):986-996. doi:10.1046/J.1523-1755.2002.00509.X Bellomo R, Kellum JA, Ronco C. Acute kidney injury. Lancet (London, England) . 2012;380(9843):756-766. doi:10.1016/S0140-6736(11)61454-2 Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care . 2004;8(4). doi:10.1186/CC2872 Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: a multinational, multicenter study. JAMA . 2005;294(7):813-818. doi:10.1001/JAMA.294.7.813 Moledina DG, Parikh CR. Phenotyping of Acute Kidney Injury: Beyond Serum Creatinine. Semin Nephrol . 2018;38(1):3-11. doi:10.1016/J.SEMNEPHROL.2017.09.002 Bellomo R, Kellum JA, Ronco C, et al. Acute kidney injury in sepsis. Intensive Care Med . 2017;43(6):816-828. doi:10.1007/s00134-017-4755-7 Bhatraju PK, Zelnick LR, Herting J, et al. Identification of Acute Kidney Injury Subphenotypes with Differing Molecular Signatures and Responses to Vasopressin Therapy. Am J Respir Crit Care Med . 2019;199(7):863-872. doi:10.1164/RCCM.201807-1346OC Bhatraju PK, Cohen M, Nagao RJ, et al. Genetic variation implicates plasma angiopoietin-2 in the development of acute kidney injury sub-phenotypes. BMC Nephrol . 2020;21(1). doi:10.1186/S12882-020-01935-1 Mansour SG, Bhatraju PK, Coca SG, et al. Angiopoietins as Prognostic Markers for Future Kidney Disease and Heart Failure Events after Acute Kidney Injury. J Am Soc Nephrol . 2022;33(3):613-627. doi:10.1681/ASN.2021060757 Samuel CS. Targeting angiopoietin-2 as a novel treatment option for kidney fibrosis. Kidney Int . 2022;102(4):691-694. doi:10.1016/J.KINT.2022.07.023 Mikacenic C, Hahn WO, Price BL, et al. Biomarkers of Endothelial Activation Are Associated with Poor Outcome in Critical Illness. PLoS One . 2015;10(10). doi:10.1371/JOURNAL.PONE.0141251 Li Y, Liu P, Zhou Y, et al. Activation of Angiopoietin-Tie2 Signaling Protects the Kidney from Ischemic Injury by Modulation of Endothelial-Specific Pathways. J Am Soc Nephrol . 2023;34(6):969-987. doi:10.1681/ASN.0000000000000098 Star BS, Boahen CK, van der Slikke EC, et al. Plasma proteomic characterization of the development of acute kidney injury in early sepsis patients. Sci Rep . 2022;12(1). doi:10.1038/S41598-022-22457-W Wiersema R, Jukarainen S, Vaara ST, et al. Two subphenotypes of septic acute kidney injury are associated with different 90-day mortality and renal recovery. Crit Care . 2020;24(1). doi:10.1186/S13054-020-02866-X NI S, IS D, RG B, et al. Early Restrictive or Liberal Fluid Management for Sepsis-Induced Hypotension. N Engl J Med . 2023;388(6):499-510. doi:10.1056/NEJMOA2212663 Zhou H, Hewitt SM, Yuen PST, Star RA. Acute Kidney Injury Biomarkers-Needs, Present Status, and Future Promise NIH Public Access. Nephrol Self Assess Progr . 2006;5(2):63-71. Sinha P, Delucchi KL, Chen Y, et al. Latent class analysis-derived subphenotypes are generalisable to observational cohorts of acute respiratory distress syndrome: a prospective study. Thorax . 2022;77(1):13-21. doi:10.1136/THORAXJNL-2021-217158 Joannidis M, Metnitz PGH. Epidemiology and natural history of acute renal failure in the ICU. Crit Care Clin . 2005;21(2):239-249. doi:10.1016/J.CCC.2004.12.005 Hoste EAJ, Bagshaw SM, Bellomo R, et al. Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. Intensive Care Med . 2015;41(8):1411-1423. doi:10.1007/S00134-015-3934-7 Ronco C, Bellomo R, Kellum JA. Acute kidney injury. Lancet (London, England) . 2019;394(10212):1949-1964. doi:10.1016/S0140-6736(19)32563-2 Clermont G, Acker CG, Angus DC, Sirio CA, Pinsky MR, Johnson JP. Renal failure in the ICU: Comparison of the impact of acute renal failure and end-stage renal disease on ICU outcomes. Kidney Int . 2002;62(3):986-996. doi:10.1046/j.1523-1755.2002.00509.x Khader A, Zelnick LR, Sathe NA, et al. The Interaction of Acute Kidney Injury with Resuscitation Strategy in Sepsis: A Secondary Analysis of a Multicenter, Phase 3, Randomized Trial (CLOVERS). Am J Respir Crit Care Med . October 2023. doi:10.1164/RCCM.202308-1448LE NI S, IS D, RG B, et al. Early Restrictive or Liberal Fluid Management for Sepsis-Induced Hypotension. N Engl J Med . 2023;388(6):499-510. doi:10.1056/NEJMOA2212663 Reynolds PM, Stefanos S, MacLaren R. Restrictive resuscitation in patients with sepsis and mortality: A systematic review and meta-analysis with trial sequential analysis. Pharmacotherapy . 2023;43(2):104-114. doi:10.1002/phar.2764 Manrique-Caballero CL, Del Rio-Pertuz G, Gomez H. Sepsis-Associated Acute Kidney Injury. Crit Care Clin . 2021;37(2):279-301. doi:10.1016/J.CCC.2020.11.010 Chewcharat A, Chang YT, Sise ME, Bhattacharyya RP, Murray MB, Nigwekar SU. Phase-3 Randomized Controlled Trials on Exclusion of Participants With Kidney Disease in COVID-19. Kidney Int reports . 2021;6(1):196-199. doi:10.1016/J.EKIR.2020.10.010 Coca SG, Vasquez-Rios G, Mansour SG, et al. Plasma Soluble Tumor Necrosis Factor Receptor Concentrations and Clinical Events After Hospitalization: Findings From the ASSESS-AKI and ARID Studies. Am J Kidney Dis . 2023;81(2):190-200. doi:10.1053/J.AJKD.2022.08.007 Niewczas MA, Pavkov ME, Skupien J, et al. A signature of circulating inflammatory proteins and development of end-stage renal disease in diabetes. Nat Med . 2019;25(5):805-813. doi:10.1038/S41591-019-0415-5 Tables Table 1. Baseline characteristics stratified by sub-phenotypes. No AKI (N = 443) SP1 (N = 605) SP2 (N = 241) Age (years), average ± SD 56.1 ± 17.0 61.7 ± 14.6 62.7 ± 15.1 Male, n (%) 217 (49%) 332 (55%) 133 (55%) Race, n (%) White 327 (74%) 435 (72%) 158 (66%) Black 51 (12%) 100 (17%) 55 (23%) Asian 18 (4%) 15 (2%) 8 (3%) Other 6 (1%) 5 (1%) 2 (1%) Not reported 41 (9%) 50 (8%) 18 (7%) Ethnicity, n (%) Hispanic/Latino 74 (17%) 88 (15%) 28 (12%) Not Hispanic/Latino 353 (80%) 493 (81%) 207 (86%) Not reported 16 (4%) 24 (4%) 6 (2%) Randomization location, n (%) ED 412 (93%) 566 (94%) 209 (87%) ICU 21 (5%) 38 (6%) 30 (12%) Other 10 (2%) 1 (0%) 2 (1%) Invasive mechanical ventilation, n (%) 26 (6%) 35 (6%) 21 (9%) Source of infection, n (%) Pneumonia 152 (34%) 163 (27%) 55 (23%) Urinary tract infection 83 (19%) 168 (28%) 57 (24%) Intra-abdominal infection 43 (10%) 55 (9%) 31 (13%) Skin or soft tissue infection 57 (13%) 54 (9%) 21 (9%) Other or unknown 108 (24%) 165 (27%) 77 (32%) Weight (kg), average ± SD 75.1 ± 26.9 80.7 ± 24.9 80.9 ± 24.4 BMI (kg/m2), average ± SD 26.3 ± 8.2 28.1 ± 7.8 28.2 ± 8.6 SOFA score, average ± SD 2.2 ± 2.2 3.4 ± 2.3 5.7 ± 3.0 History of CKD, n (%) 19 (4%) 46 (8%) 74 (31%) Baseline serum creatinine (mg/dL), average ± SD 0.9 ± 0.8 0.8 ± 0.4 0.9 ± 0.4 Randomization serum creatinine (mg/dL), average ± SD 0.9 ± 0.8 2.0 ± 1.5 3.6 ± 2.5 Volume of fluid administered before randomization (mL), median (IQR) 2000 (1400-2400) 2100 (1525-2500) 2100 (1450-2498) Biomarker Concentrations, median (IQR) sTNFR-1 3207 (2062-5464) 5391 (3713-8333) 18553 (11899-27238) Ang-1 5442 (2450-9743) 5477 (2997-9360) 1768 (801-3953) Ang-2 6053 (4093-10531) 8518 (5315-13138) 23311 (15949-38385) Ang-2/Ang-1 ratio 1.2 (0.6-3.1) 1.6 (0.8-3.4) 13.6 (7.4-27.8) Entries are mean (SD) for continuous variables and N (%) for categorical variables, except as noted. Abbreviations: BMI, body mass index; SOFA, sequential organ failure assessment scores; CKD, chronic kidney disease; sTNFR-1, soluble tumor necrosis factor receptor-1; Ang-1, angiopoietin-1; Ang-2, angiopoietin-2 Table 2. Treatment effects by sub-phenotype status among patients with kidney dysfunction and all patients in CLOVERS CLOVERS Populations % who died, liberal % who died, restrictive Difference, restrictive – liberal (95% CI) Difference of Difference (95% CI) p-value for interaction Patients with AKI and ESKD (n=846) 28-day mortality SP-1 11% 10% -1% (-6%, 4%) 14% (2%, 27%) 0.03 SP-2 41% 26% -15% (-27%, -3%) 90-day mortality SP-1 18% 17% -1% (-7%, 5%) 13% (-1%, 27%) 0.06 SP-2 48% 36% -13% (-24%, -1%) Patients without AKI or ESRD (n=443) 28-day mortality SP-1 6% 8% 2% (-3%, 6%) 2% (-33%, 38%) 0.89 SP-2 36% 35% -1% (-36%, 34%) 90-day mortality SP-1 11% 16% 5% (-2%, 11%) 7% (-30%, 43%) 0.72 SP-2 43% 41% -2% (-38%, 34%) All Patients (n=1289) 28-day mortality SP-1 9% 9% 0% (-3%, 4%) 14% (2%, 25%) 0.02 SP-2 41% 27% -13% (-25%, -2%) 90-day mortality SP-1 15% 17% 1% (-3%, 6%) 14% (1%, 27%) 0.03 SP-2 48% 36% -13% (-24%, -1%) Supplementary Files CloversSubphenotypeSupplementICMFinal.docx STROBEchecklistcohortICM.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4523416","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":313128317,"identity":"613cd6eb-9538-498c-b6bc-67f5b6c1c402","order_by":0,"name":"Elizabeth Kiernan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAApElEQVRIiWNgGAWjYDACHsYGIGkDZkuQoiWNJC1g8jAJWvh5Djc+rtxxPnHDAeaDt3mI0SLZ29hsePbMbaAWtmRrorQYnGdsk2xsu5247QCPmTRRWuzPM7b/bGw7B9TC/404LQa8jW2MjW0HQLawEadF4szBZqDDko33H2YztpxDjBb+nvSHHxvb7GRntjc/vPGGGC0IwEya8lEwCkbBKBgF+AAAW+sylXjFU7kAAAAASUVORK5CYII=","orcid":"https://orcid.org/0009-0009-1061-1108","institution":"UW: University of Washington","correspondingAuthor":true,"prefix":"","firstName":"Elizabeth","middleName":"","lastName":"Kiernan","suffix":""},{"id":313128318,"identity":"32ddd11e-21bf-4792-a03f-9c5a8cb359e6","order_by":1,"name":"Leila R. Zelnick","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Leila","middleName":"R.","lastName":"Zelnick","suffix":""},{"id":313128319,"identity":"62c2faca-43c6-42c9-aa95-fd2c1bb06104","order_by":2,"name":"Ayesha Khader","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Ayesha","middleName":"","lastName":"Khader","suffix":""},{"id":313128320,"identity":"0344b8fd-d3d1-49d9-b3ab-6ba73922d120","order_by":3,"name":"Taylor D. Coston","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Taylor","middleName":"D.","lastName":"Coston","suffix":""},{"id":313128321,"identity":"d701adea-519e-4328-902e-a07dd2bf6b2a","order_by":4,"name":"Zoie A. Bailey","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Zoie","middleName":"A.","lastName":"Bailey","suffix":""},{"id":313128322,"identity":"fe7adb3c-6f6b-4875-9de6-6cf5d459fdba","order_by":5,"name":"Sarah Speckmaier","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Sarah","middleName":"","lastName":"Speckmaier","suffix":""},{"id":313128323,"identity":"05d81a95-13d7-48bc-8b04-2063db446226","order_by":6,"name":"Jordan Lo","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Jordan","middleName":"","lastName":"Lo","suffix":""},{"id":313128324,"identity":"58500fff-584f-43c8-9dc5-e22ec7fd3b2b","order_by":7,"name":"Neha Sathe","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Neha","middleName":"","lastName":"Sathe","suffix":""},{"id":313128325,"identity":"4b496884-c05a-4d39-beda-9897c68d0458","order_by":8,"name":"Bryan R. Kestenbaum","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Bryan","middleName":"R.","lastName":"Kestenbaum","suffix":""},{"id":313128326,"identity":"5fdcc869-00b4-47e8-aae4-c6045e10fae3","order_by":9,"name":"Jonathan Himmelfarb","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Jonathan","middleName":"","lastName":"Himmelfarb","suffix":""},{"id":313128327,"identity":"e6eb6efc-0707-46e8-b211-6cbff5ad5f1a","order_by":10,"name":"Nicholas Johnson","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Nicholas","middleName":"","lastName":"Johnson","suffix":""},{"id":313128328,"identity":"73f3c3ac-69d1-48d6-b1f3-7ebf50a23224","order_by":11,"name":"Nathan Shapiro","email":"","orcid":"","institution":"Harvard Medical School","correspondingAuthor":false,"prefix":"","firstName":"Nathan","middleName":"","lastName":"Shapiro","suffix":""},{"id":313128329,"identity":"b376f8a1-0769-4c8c-81f8-4d490650003b","order_by":12,"name":"Ivor S. Douglas","email":"","orcid":"","institution":"Denver Health Main Campus","correspondingAuthor":false,"prefix":"","firstName":"Ivor","middleName":"S.","lastName":"Douglas","suffix":""},{"id":313128330,"identity":"1ffe319e-a284-4749-b01b-ab8e556f7c53","order_by":13,"name":"Catherine Hough","email":"","orcid":"","institution":"Oregon Health \u0026 Science University","correspondingAuthor":false,"prefix":"","firstName":"Catherine","middleName":"","lastName":"Hough","suffix":""},{"id":313128331,"identity":"579369cc-a8aa-49d9-889f-b0fafc0ae6ad","order_by":14,"name":"Pavan Bhatraju","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Pavan","middleName":"","lastName":"Bhatraju","suffix":""}],"badges":[],"createdAt":"2024-06-03 17:26:43","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4523416/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4523416/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":59524553,"identity":"140013b1-3e39-40f8-ae7d-e429649eafaa","added_by":"auto","created_at":"2024-07-02 20:47:45","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":49025,"visible":true,"origin":"","legend":"\u003cp\u003eMakeup of sub-phenotypes by proportion of KDIGO Stage of AKI and ESKD\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure Legend\u003c/strong\u003e: Provided is the proportion of participants with Kidney Disease Improving Global Outcomes (KDIGO) Stage 1, 2 or 3 AKI and end-stage kidney disease (ESKD) within each sub-phenotype. Participants with SP2 had a higher proportion of participants with more severe AKI compared to participants with SP1.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-4523416/v1/72dbf76ae33a00ae205adbc4.png"},{"id":59525802,"identity":"66f805e7-253b-43e2-82b5-75c938cbc196","added_by":"auto","created_at":"2024-07-02 20:55:45","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":55013,"visible":true,"origin":"","legend":"\u003cp\u003eKaplan-Meier survival curve in sepsis-associated AKI sub-phenotypes stratified by resuscitation treatment group\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure Legend:\u003c/strong\u003e Curves for each sub-phenotype are stratified by liberal or restrictive resuscitation strategy.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-4523416/v1/ba55183c93d5fce65d9e56f0.png"},{"id":59524556,"identity":"9a565560-2526-4767-9bca-bff978fe88c6","added_by":"auto","created_at":"2024-07-02 20:47:45","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":254167,"visible":true,"origin":"","legend":"\u003cp\u003eProbability belonging to SP-2 and differential treatment response to a restrictive and liberal fluid resuscitation strategy\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure Legend: \u003c/strong\u003ePanel A is the probability of belonging to SP2 (x-axis) and the probability of 28-day mortality (y-axis). The lines plot the estimated mortality in either a liberal (red) or restrictive (blue) resuscitation strategy with 95% CIs over a range of probabilities of assignment to SP2. The p-value tests the null hypothesis of no interaction between treatment group and the continuous probability of SP2 membership, and excludes participants who were censored prior to day 28. Plots B, C and D below demonstrate the concentrations of sTNFR-1, Ang-1 and Ang-2 concentrations at different probabilities of belonging to SP2.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-4523416/v1/2ea727964b69cf7e82704d91.png"},{"id":60688605,"identity":"8d627f05-fdba-44cb-a133-92ec1533b32d","added_by":"auto","created_at":"2024-07-19 14:30:48","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1201167,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4523416/v1/7d76d3d1-2171-499e-aa41-b0a2d60f68f5.pdf"},{"id":59524557,"identity":"aaea298c-77f2-4a4d-9f46-246fb60a61eb","added_by":"auto","created_at":"2024-07-02 20:47:45","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":136515,"visible":true,"origin":"","legend":"","description":"","filename":"CloversSubphenotypeSupplementICMFinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-4523416/v1/10fc190839b8cab93a2ff159.docx"},{"id":59524555,"identity":"206447a6-fda6-4ef9-b9c4-7fe424fbc0b1","added_by":"auto","created_at":"2024-07-02 20:47:45","extension":"docx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":33768,"visible":true,"origin":"","legend":"","description":"","filename":"STROBEchecklistcohortICM.docx","url":"https://assets-eu.researchsquare.com/files/rs-4523416/v1/21844999284a3636f313f8bf.docx"}],"financialInterests":"","formattedTitle":"Molecular Phenotyping of Patients with Sepsis and Kidney Injury and Differential Response to Fluid Resuscitation","fulltext":[{"header":"Take Home Message","content":"\u003cp\u003eWe applied a 3 biomarker prediction model to the CLOVERS analysis and showed feasbility of identifying sub-phenotypes of patients with sepsis. These sub-phenotypes are clinically distinct and respond differently to fluid resuscitation in sepsis. Identification of sub-phenotypes could inform precision-guided therapy for patients with sepsis-associated hypotension and kidney injury.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"INTRODUCTION","content":"\u003cp\u003eAcute kidney injury (AKI) is the most common form of organ failure in sepsis\u0026nbsp;developing in as many as\u0026nbsp;60%\u0026nbsp;in\u0026nbsp;septic patients with\u0026nbsp;hypotension\u003csup\u003e1,2\u003c/sup\u003e.\u0026nbsp;Sepsis-associated AKI is associated with prolonged hospitalization, need for kidney replacement therapy, future chronic kidney disease (CKD) and death\u003csup\u003e3\u0026ndash;8\u003c/sup\u003e.\u0026nbsp;At present, providers base treatment decisions in AKI on parameters such as urine output and serum creatinine, yet these biomarkers fail to differentiate the highly heterogenous clinical syndrome of AKI and cannot reliably predict response to common interventions such as volume resuscitation and vasopressors in sepsis\u003csup\u003e9,10\u003c/sup\u003e. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe\u0026nbsp;previously applied latent class analysis (LCA) to two cohorts to\u0026nbsp;identify two distinct sub-phenotypes in patients with sepsis-associated\u0026nbsp;AKI, and subsequently developed a parsimonious 3-variable model for clinical application\u003csup\u003e11\u003c/sup\u003e. This 3-variable model\u0026nbsp;included plasma biomarkers of endothelial function (angiopoietin-1 (Ang-1) and angiopoietin-2 (Ang-2)) and inflammation (soluble tumor necrosis factor receptor-1\u0026nbsp;[sTNFR-1])\u003csup\u003e11\u003c/sup\u003e.\u0026nbsp;Ang-1 and -2 are vascular endothelial growth factors that have opposing activities. Ang-1 is an agonist for the Tie-2 receptor and is protective by stabilizing the endothelium, promoting vessel maturation, and preventing microcirculatory capillary leakage. In contrast, Ang-2 typically\u0026nbsp;acts as an antagonist for binding to the Tie-2 receptor and promotes endothelial dysfunction in the form of vessel destabilization, endothelial cell apoptosis, capillary leak, and deregulation of inflammation\u003csup\u003e12\u0026ndash;14\u003c/sup\u003e. Independently, sTNFR-1 functions in the early innate inflammatory response in sepsis and has been associated with AKI and mortality\u003csup\u003e15\u003c/sup\u003e. Our model established two distinct AKI sub-phenotypes: those with a lower ratio of Ang-2/Ang-1 and low sTNFR-1, designated SP1 and characterized by endothelial stability and improved short and long-term outcomes. In contrast, those with higher ratio of Ang-2/Ang-1 and high sTNFR-1, or SP2 phenotype, were characterized by high inflammatory state and leaky endothelium and worse clinical outcomes\u003csup\u003e11,16\u0026ndash;18\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis (CLOVERS) clinical trial randomized patients with sepsis to an initial restrictive or liberal fluid resuscitation strategy and demonstrated no differences in clinical outcomes between treatment arms or in the sub-groups with AKI or end-stage kidney disease (ESKD)\u003csup\u003e19\u003c/sup\u003e. Early and aggressive fluid resuscitation is a cornerstone of sepsis management which aims to improve hypoperfusion caused by increased capillary leak and endothelial permeability. In the setting of kidney injury or disease, however, the benefit of fluid resuscitation is counterbalanced by impaired volume management. We hypothesized that liberal fluid resuscitation would be preferentially harmful to septic patients with evidence of both endothelial injury and kidney disease, and that application of the SP1 or SP2 sub phenotypes may enable identification of patients most likely to benefit from early vasoconstrictive therapy. \u0026nbsp;To address these questions, we first identified sub-phenotypes with known kidney injury (AKI or ESKD) using plasma samples collected in the emergency\u0026nbsp;department, representing an early and critical time for treatment implementation in sepsis.\u0026nbsp;Second, in a sensitivity analysis, we verified sub-phenotypes by applying LCA to 23 variables collected at study enrollment and compared sub-phenotype identification to the 3-variable model. Third, since serum creatinine can lag up to 48 hours and is insensitive for kidney injury\u003csup\u003e20\u003c/sup\u003e, we conducted a sensitivity analysis including patients without AKI on study enrollment to test whether the treatment effect by sub-phenotype was preserved. Fourth, to understand the overlap and differences between different sub-phenotypes in critical illness, we cross-tabulated our sub-phenotypes with a previously established hypo- and hyperinflammatory acute respiratory distress syndrome (ARDS) sub-phenotypes.\u0026nbsp;\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003e\u003cstrong\u003eStudy Design and Oversight\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a secondary analysis of data from CLOVERS, a multicenter, prospective, phase 3, randomized, non-blinded trial. Patients enrolled in CLOVERS were allocated to either a restrictive or liberal fluid resuscitation strategy in the setting of sepsis. All patients were randomized within 4 hours of meeting inclusion criteria, and duration of protocol for restrictive or liberal fluid resuscitation was 24 hours. Patients randomized to a restrictive fluid protocol had vasopressors prioritized as the primary treatment for sepsis-induced hypotension, with \u0026ldquo;rescue fluids\u0026rdquo; being permitted for prespecified indications that suggested severe intravascular volume depletion. The liberal fluid protocol consisted of a recommended initial 2 L IV infusion of isotonic crystalloid, followed by fluid boluses administered based on clinical triggers with \u0026ldquo;rescue vasopressors\u0026rdquo; permitted for prespecified indications. A more detailed description of methods can be found in the Supplement of the primary manuscript\u003csup\u003e19\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAKI Definition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAKI was defined using the KDIGO criteria as an increase in serum creatinine at randomization of \u0026ge;50% or \u0026ge;0.3 mg/dL above a baseline serum creatinine. The lowest outpatient or inpatient serum creatinine value within the last year prior to the index hospitalization was used for the baseline when available (N=909). If no pre-hospitalization serum creatinine was available and there was no known history of CKD, then we imputed a baseline serum creatinine based on an eGFR of 90 ml/min/1.73 m2 (N=517) using the 2021 creatinine-based CKD-EPI equation\u003csup\u003e11\u003c/sup\u003e. If no pre-hospitalization serum creatinine was available and there was a known history of CKD, then we used the lowest serum creatinine value during the index hospitalization as the baseline (N=29). Patients were subsequently staged according to the serum creatinine at randomization using the KDIGO guidelines. Seventy-five patients were on maintenance hemodialysis prior to study enrollment, (ESKD). Since ESKD patients represent severe impairment in kidney-mediated volume management and guidelines do not offer guidance on distinct therapeutic approaches across the spectrum of ESKD, we included them in the analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAKI Sub-phenotype Identification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMethods for measurement of Ang-1, Ang-2 and sTNFR-1 can be found in the online supplement. We used plasma biomarker concentrations of Ang-1, Ang-2 and sTNFR-1 to identify two sub-phenotypes (SP1 and SP2). Since the publication of our prior works, Meso Scale Discovery changed the sTNFR-1 assay from a R-Plex to a U-Plex platform. In analyses provided in the supplement we found the correlation of these two assays to be high (Pearson\u0026rsquo;s correlation, \u003cem\u003er = 0.94\u003c/em\u003e), and we created a calibration equation (\u003cstrong\u003eFigure S1)\u003c/strong\u003e. We updated the 3-variable prediction model that initially used R-Plex measured sTNFR-1 to include U-Plex measured values to account for the updated sTNFR-1 biomarker platform. The model to identify AKI sub-phenotypes is\u0026nbsp;TNFR_R-Plex = -101.3387 +0.3617*TNFR_U-Plex followed by the previously published prediction equation Logit(P(SP2)) = -35.44+1.99*log(Ang-2/Ang-1) + 3.41*log(TNFR_R-Plex). Participants with a probability of SP2 \u0026gt; 0.5 were then classified as having the SP2 sub-phenotype. Methods of LCA applied to the baseline variables can be found in the Methods in the supplement. Previously\u0026nbsp;described ARDS\u0026nbsp;sub-phenotypes were identified using a published four variable model that included plasma concentrations of\u0026nbsp;interleukin-6, sTNFR-1, serum bicarbonate and clinical documentation of receipt of vasopressors prior to randomization\u003csup\u003e11,21\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOutcome Measures\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe primary outcome for\u0026nbsp;this analysis was 28-day\u0026nbsp;and\u0026nbsp;90-day mortality. Additional outcomes included need for\u0026nbsp;invasive mechanical ventilation\u0026nbsp;(IMV), need for new KRT, and ICU-free days through 28 days of follow-up. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe compared the risk of death by sub-phenotype over 28 and 90 days of follow-up using Cox regression, adjusting in nested models for potential confounders. To test for heterogeneity of treatment effect, we completed an intention-to-treat comparison of the treatment effect of restrictive or liberal fluid resuscitation strategy on 28-day and 90-day mortality by sub-phenotype. For this analysis, we used Kaplan-Meier 28-day and 90-day mortality point estimates involving all patients who were discharged home or were still alive at day 90, with data censored at day 91. Patients who were lost to follow-up prior to day 90 were censored at the time they were last confirmed to be alive. We compared 28-day and 90-day mortality point estimates and evaluated the evidence for a treatment by sub-phenotype interaction using a non-parametric bootstrap approach with 10,000 replicates.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSecondary outcomes were the time to IMV among patients who were not receiving IMV at randomization and the time to new KRT among patients not on dialysis at the time of randomization. For these outcomes, we used a Fine-Gray sub-distribution hazard model to estimate the\u0026nbsp;sub-distribution hazard ratio\u0026nbsp;of each outcome accounting for the competing risk of death. Finally, we used linear regression to estimate the difference in ICU-free days through day 28 between the two treatment groups. All analyses were conducted using the R 4.2.3 software environment (R Foundation for Statistical Computing, Vienna, Austria). Two-sided p-values \u0026lt; 0.05 were taken as evidence of statistical significance.\u0026nbsp;\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003e\u003cstrong\u003e\u003cem\u003eBaseline Characteristics of Participants by Sepsis-associated AKI Sub-phenotype\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong 1563 patients enrolled in CLOVERS, 1289 had biospecimens collected prior to randomization and available for analysis (\u003cstrong\u003eFigure 1\u003c/strong\u003e). Among the 1289 patients, 846 (66%) had kidney dysfunction (771 had AKI and 75 had ESKD) at study enrollment and 443 (34%) did not have AKI or ESKD. Among the 846 patients with kidney dysfunction the previously published sub-phenotype classification model, incorporating plasma Ang-1, Ang-2, and sTNFR-1 measurements at randomization, classified 605 (72%) patients as SP1 and 241 (28%) as SP2 \u003cstrong\u003e(Table 1)\u003c/strong\u003e. As expected by randomization, the distribution of SP1 and SP2 were similar between treatment arms. Patients with SP2 had more severe KDIGO stage of AKI on study enrollment compared to SP1 (\u003cstrong\u003eFigure S2\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo determine whether SP2 is enriched in the AKI population, we applied the 3-biomarker prediction model to patients without AKI or ESKD on study enrollment. Among 443 patients, we found that 411 (93%) were classified as SP1 and 32 (7%) as SP2. The 32 patients with SP2 were characterized by high rates of CKD and ten patients subsequently developed AKI within 48 hours of study enrollment based on changes in serum creatinine (\u003cstrong\u003eTable S1\u003c/strong\u003e). In a sensitivity analysis, we applied LCA to 23-variables collected at study enrollment and found that a 2-class model best separated the data (\u003cstrong\u003eTable S3\u003c/strong\u003e). We also found high correlation between the LCA derived sub-phenotypes and the 3-variable biomarker sub-phenotypes, Cohen\u0026rsquo;s Kappa of 0.80. With the high correlation and to support prospective identification of these sub-phenotypes, we used the 3-variable derived sub-phenotypes for the subsequent analyses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eClinical Outcomes by Sub-Phenotype\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate the association of sub-phenotypes with clinical outcomes, patients without AKI were compared against those with SP1 and SP2 and then patients with SP1 and SP2 were directly compared (\u003cstrong\u003eTable S4\u003c/strong\u003e). In models adjusting for demographics, comorbidities, KDIGO stage of AKI or ESKD and baseline sequential organ failure assessment (SOFA) scores, patients with SP1 had a similar risk of new KRT, ICU length of stay, receipt of IMV, 28-day and 90-day mortality as patients with no AKI. Patients with SP2 had greater\u0026nbsp;risk of\u0026nbsp;28-day (HR = 2.33 (95% CI: 1.60, 3.38) and 90-day mortality (HR = 1.98 (95% CI: 1.45, 2.71) than SP1.\u0026nbsp;Patients\u0026nbsp;with SP2 were more likely than those with SP1 to require new KRT (sHR=3.03 (95% CI: 1.42, 6.47) or IMV after study enrollment (sHR=1.54 (95% CI: 1.00, 2.37) (\u003cstrong\u003eTable S5)\u003c/strong\u003e. Those with SP2 had ICU stays that were 3 days longer than those with SP1 (adjusted difference 3.1 days (95% CI: 1.8 to 4.5 days) (\u003cstrong\u003eTable S7\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eDifference in Volume of Fluid and Timing of Vasopressors by CLOVERS Randomization\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn the 24-hour protocol following randomization, patients with SP1 randomized to the liberal resuscitation strategy received on average 2 L more of IV fluids than\u0026nbsp;patients with SP1 randomized to the restrictive\u0026nbsp;resuscitation strategy (3,684 \u0026plusmn; 1,644 mL vs 1,684 \u0026plusmn; 1657 mL) (\u003cstrong\u003eTable S6\u003c/strong\u003e). Similarly, patients with SP2 randomized to the liberal resuscitation strategy received on average 4,091 (\u0026plusmn; 1793) mL compared to 2,244 (\u0026plusmn; 2,305) mL in the restrictive strategy. Consistent with the trial protocol,\u0026nbsp;patients randomized to\u0026nbsp;the restrictive strategy received vasopressors more often than patients in the liberal strategy across both sub phenotypes.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eInteraction Between Sub-phenotype and Resuscitation Strategy\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNext, we determined whether sub-phenotypes identified using biospecimens collected prior to randomization respond differently to the fluid resuscitation strategy. In SP1, a similar proportion of patients died at 28 days with liberal and restrictive fluid strategies (11% vs 10%, respectively) (\u003cstrong\u003eFigure 2\u003c/strong\u003e). In SP2 a liberal strategy compared with a restrictive strategy led to greater 28-day mortality (41% vs 26% respectively) and the difference in treatment effects between SP1 and SP2 was 14% (95% CI: 2%, 27%); \u003cem\u003ep-value for interaction = 0.03\u003c/em\u003e. This effect was not observed in the patients without AKI (\u003cstrong\u003eTable 2\u003c/strong\u003e). Among patients with SP1, 90-day mortality with a liberal and restrictive fluid strategies was 18% versus 17%, while SP2 was 48% versus 26% (p-value for interaction \u003cem\u003e= 0.06\u003c/em\u003e). We did not observe a difference for alternative outcomes among sub-phenotypes by treatment (\u003cstrong\u003eTable S8\u003c/strong\u003e). Next, we evaluated the\u0026nbsp;continuous\u0026nbsp;probability of belonging to SP2 and the interaction of a liberal versus a restrictive fluid resuscitation. We observed a significant interaction\u0026nbsp;between\u0026nbsp;the probability of sub-phenotype and fluid resuscitation strategy and 28-day mortality (\u003cem\u003ep=0.008\u003c/em\u003e) (\u003cstrong\u003eFigure 3\u003c/strong\u003e). As patients had a greater certainty of the SP2 phenotype (i.e. higher probability of SP2 based on the 3-biomarker model), we observed a larger difference in 28-day mortality between a liberal versus restrictive fluid resuscitation strategy. In the primary CLOVERS publication, there was no observed heterogeneity of treatment effect based on SOFA scores or ESKD status\u003csup\u003e19\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn sensitivity analyses, we evaluated sub-phenotypes restricted to the AKI population and excluded ESKD. Among patients with SP1, 28-day mortality with liberal and restrictive fluid strategies was 10% versus 10% and in patients with SP2 was 39% versus 29% (p-value for interaction = 0.16) (\u003cstrong\u003eTable S9\u003c/strong\u003e). In another sensitivity analysis, we used the 3-biomarker model to classify all patients with plasma available in CLOVERS as SP1 and SP2.\u0026nbsp;Among patients with SP1, 28-day mortality with a liberal and restrictive fluid strategies was 9% versus 9% and in patients with SP2 was 41% versus 27% (p-value for interaction \u003cem\u003e= 0.02\u003c/em\u003e). We also observed a similar decrease in 90-day mortality in SP2\u0026nbsp;(\u003cstrong\u003eTable 2\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eCross-Tabulation of AKI and ARDS Sub-phenotypes\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe directly compared the cross-tabulation of SP1 and SP2 and ARDS sub-phenotypes (hypoinflammatory and hyperinflammatory). Cross-tabulation demonstrated mild to moderate overlap between hypoinflammatory and SP1 and hyperinflammatory and SP2 but also demonstrated distinct differences (Cohen\u0026rsquo;s Kappa for Agreement of 0.42) (\u003cstrong\u003eTable S10\u003c/strong\u003e). Out of 846 patients, 77.0% were concordant (i.e. hypoinflammatory and SP1 or hyperinflammatory and SP2), and 23.0% were classified discordantly. We observed no differential response to a restrictive or liberal fluid resuscitation strategy between ARDS sub-phenotypes for 28-day (p-value for interaction = 0.25) or 90-day mortality (p-value for interaction = 0.60) (\u003cstrong\u003eTable S11\u003c/strong\u003e).\u0026nbsp;\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThe heterogeneity of AKI has limited insight into clinical trajectory and hindered development of interventions that target an individual patient\u0026rsquo;s pathophysiology\u003csup\u003e22\u0026ndash;25\u003c/sup\u003e. We completed a secondary analysis of the CLOVERS trial and demonstrated that sub-phenotypes previously defined in ICU patients were also present in the emergency\u0026nbsp;department, an earlier and critical window for therapeutic decisions in sepsis treatment.\u0026nbsp;These\u0026nbsp;sub-phenotypes predicted differential treatment response to liberal versus restrictive resuscitation strategies among patients with kidney dysfunction; revealing a treatment effect which was not seen in the original CLOVERS study. Taken together, these data offer a precision medicine treatment strategy in sepsis-induced hypotension and kidney dysfunction based on a 3-biomarker prediction model which may identify populations with differential response to fluid resuscitation strategy.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDespite a long-held belief by clinicians that a rise in serum creatinine early during sepsis should result in additional fluid therapy\u003csup\u003e9\u003c/sup\u003e, patients with SP2 and kidney dysfunction had a greater mortality with a liberal fluid strategy compared to a restrictive fluid strategy. A difference in clinical outcomes between fluid strategy was not observed in patients without AKI/ESKD or patients with SP1. While these findings are hypothesis-generating, it suggests that early in sepsis the combination of kidney dysfunction and endothelial injury may identify a sub-phenotype that is harmed by a liberal fluid therapy. In addition, the observed heterogeneity of treatment effect was not present when identifying sub-groups based on SOFA scores, ESKD or AKI status\u003csup\u003e26,27\u003c/sup\u003e, suggesting that the sub-phenotypes capture information that is not readily available by clinical definitions. In this manner, application of the 3-biomarker prediction model to all patients with sepsis-induced hypotension may improve generalizability and time to sub-phenotype identification to facilitate future clinical trials without misclassification of patients with SP2. \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur prior work leveraged these AKI sub-phenotypes in the Vasopressin and Septic Shock Trial (VASST), a randomized control trial investigating early addition of vasopressin to norepinephrine in sepsis. In that analysis, we showed that in SP1, early addition of vasopressin compared to norepinephrine alone was associated with improved 90-day mortality, but in SP2, vasopressin showed no significant treatment difference. Extrapolating these findings with the current CLOVERS analysis, we hypothesize that patients with the SP1 would derive benefit from either a restrictive or liberal fluid resuscitation and early addition of vasopressin with norepinephrine for persistent hypotension, while those with the SP2 may benefit from an initial restrictive fluid strategy and monotherapy with norepinephrine.\u003c/p\u003e\n\u003cp\u003eOur study has several strengths. First, a majority of patients had a pre-hospitalization serum creatinine value available to approximate baseline kidney function and to accurately determine AKI status early after hospital presentation. Leveraging this data highlighted that approximately two-thirds of patients in the trial had AKI on hospital presentation for sepsis-induced hypotension. These findings are consistent with prior studies of AKI in septic shock and highlight that prevention of AKI in sepsis may be infeasible and therapies that target recovery should be prioritized\u003csup\u003e28,29\u003c/sup\u003e. Second, we were able to leverage the randomization in the CLOVERS trial to address limitations of indication bias of fluid administration in observational studies to demonstrate that patients with SP2, characterized by greater endothelial dysfunction and inflammation, have an improved mortality with a restrictive resuscitation strategy compared to a liberal strategy. Third, the CLOVERS trial included patients with ESKD, a group that is often excluded from large randomized clinical trials\u003csup\u003e30\u003c/sup\u003e. While limited by small group size, our data suggest that most patients with ESKD have a sub-phenotype consistent with SP2 (79%) and may benefit from a restrictive fluid resuscitation strategy.\u003c/p\u003e\n\u003cp\u003eThis analysis should be interpreted in the context of its limitations. First, there was variation in the amount of fluid received in the CLOVERS trial in the restrictive and liberal fluid groups and overlap of total volumes between groups may\u0026nbsp;dampen\u0026nbsp;a potential therapeutic signal. Second, this was a retrospective study of a completed clinical trial. However, the study design and prediction model for sub-phenotype identification were defined \u003cem\u003ea priori\u003c/em\u003e. Third, we underscore that our work does not preclude that alternative AKI sub-phenotypes are present. Fourth, elevations of angiopoietins and sTNFR-1 are not specific to AKI. However, a large body of preclinical and clinical studies have demonstrated the importance of these pathways and biomarkers in kidney diseases, such as AKI and CKD\u003csup\u003e13,31,32\u003c/sup\u003e. Fifth, serum creatinine concentrations were not collected after day 3 of participation in CLOVERS and so we are unable to draw conclusions about differential rate of kidney recovery between sub-phenotypes.\u003c/p\u003e\n\u003cp\u003eIn summary, this work posits that early identification of AKI sub-phenotypes can inform clinical decision making in patients with sepsis-induced hypotension. We observed that SP2 had improved 28-day mortality with a restrictive fluid resuscitation strategy. Identification of AKI sub-phenotypes may facilitate development of targeted treatment interventions in critically ill patients. \u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor Contribution\u003c/strong\u003e: Conception and design: AK, LZ, CH, PB\u003c/p\u003e\n\u003cp\u003eEnrollment of subjects, collection of data and completion of measurements: NJ, NS, ID, CH, PB.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAnalysis and interpretation: All authors\u003c/p\u003e\n\u003cp\u003eDrafting the manuscript for important intellectual content: All authors\u003c/p\u003e\n\u003cp\u003eAll authors read and approved this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Support:\u003c/strong\u003e Funding\u0026nbsp;was\u0026nbsp;received from the NIDDK K23DK116967 and R01DK133177 (PKB)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Sharing Statement:\u0026nbsp;\u003c/strong\u003eData available within the article or its supplementary materials\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDISCLOSURE STATEMENT:\u0026nbsp;\u003c/strong\u003eThe\u0026nbsp;\u003cem\u003eauthor\u003c/em\u003e declares that they had no relevant or material financial interests that relate to the research described in this\u003cem\u003epaper\u003cstrong\u003e.\u003c/strong\u003e\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBhatraju PK, Robinson-Cohen C, Mikacenic C, et al. Circulating levels of soluble Fas (sCD95) are associated with risk for development of a nonresolving acute kidney injury subphenotype. \u003cem\u003eCrit Care\u003c/em\u003e. 2017;21(1). doi:10.1186/S13054-017-1807-X\u003c/li\u003e\n\u003cli\u003eKellum JA, Chawla LS, Keener C, et al. The Effects of Alternative Resuscitation Strategies on Acute Kidney Injury in Patients with Septic Shock. \u003cem\u003eAm J Respir Crit Care Med\u003c/em\u003e. 2016;193(3):281-287. doi:10.1164/RCCM.201505-0995OC\u003c/li\u003e\n\u003cli\u003eJoannidis M, Metnitz PGH. Epidemiology and natural history of acute renal failure in the ICU. \u003cem\u003eCrit Care Clin\u003c/em\u003e. 2005;21(2):239-249. doi:10.1016/J.CCC.2004.12.005\u003c/li\u003e\n\u003cli\u003eHoste EAJ, Bagshaw SM, Bellomo R, et al. Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. \u003cem\u003eIntensive Care Med\u003c/em\u003e. 2015;41(8):1411-1423. doi:10.1007/S00134-015-3934-7\u003c/li\u003e\n\u003cli\u003eClermont G, Acker CG, Angus DC, Sirio CA, Pinsky MR, Johnson JP. Renal failure in the ICU: comparison of the impact of acute renal failure and end-stage renal disease on ICU outcomes. \u003cem\u003eKidney Int\u003c/em\u003e. 2002;62(3):986-996. doi:10.1046/J.1523-1755.2002.00509.X\u003c/li\u003e\n\u003cli\u003eBellomo R, Kellum JA, Ronco C. Acute kidney injury. \u003cem\u003eLancet (London, England)\u003c/em\u003e. 2012;380(9843):756-766. doi:10.1016/S0140-6736(11)61454-2\u003c/li\u003e\n\u003cli\u003eBellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. \u003cem\u003eCrit Care\u003c/em\u003e. 2004;8(4). doi:10.1186/CC2872\u003c/li\u003e\n\u003cli\u003eUchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: a multinational, multicenter study. \u003cem\u003eJAMA\u003c/em\u003e. 2005;294(7):813-818. doi:10.1001/JAMA.294.7.813\u003c/li\u003e\n\u003cli\u003eMoledina DG, Parikh CR. Phenotyping of Acute Kidney Injury: Beyond Serum Creatinine. \u003cem\u003eSemin Nephrol\u003c/em\u003e. 2018;38(1):3-11. doi:10.1016/J.SEMNEPHROL.2017.09.002\u003c/li\u003e\n\u003cli\u003eBellomo R, Kellum JA, Ronco C, et al. Acute kidney injury in sepsis. \u003cem\u003eIntensive Care Med\u003c/em\u003e. 2017;43(6):816-828. doi:10.1007/s00134-017-4755-7\u003c/li\u003e\n\u003cli\u003eBhatraju PK, Zelnick LR, Herting J, et al. Identification of Acute Kidney Injury Subphenotypes with Differing Molecular Signatures and Responses to Vasopressin Therapy. \u003cem\u003eAm J Respir Crit Care Med\u003c/em\u003e. 2019;199(7):863-872. doi:10.1164/RCCM.201807-1346OC\u003c/li\u003e\n\u003cli\u003eBhatraju PK, Cohen M, Nagao RJ, et al. Genetic variation implicates plasma angiopoietin-2 in the development of acute kidney injury sub-phenotypes. \u003cem\u003eBMC Nephrol\u003c/em\u003e. 2020;21(1). doi:10.1186/S12882-020-01935-1\u003c/li\u003e\n\u003cli\u003eMansour SG, Bhatraju PK, Coca SG, et al. Angiopoietins as Prognostic Markers for Future Kidney Disease and Heart Failure Events after Acute Kidney Injury. \u003cem\u003eJ Am Soc Nephrol\u003c/em\u003e. 2022;33(3):613-627. doi:10.1681/ASN.2021060757\u003c/li\u003e\n\u003cli\u003eSamuel CS. Targeting angiopoietin-2 as a novel treatment option for kidney fibrosis. \u003cem\u003eKidney Int\u003c/em\u003e. 2022;102(4):691-694. doi:10.1016/J.KINT.2022.07.023\u003c/li\u003e\n\u003cli\u003eMikacenic C, Hahn WO, Price BL, et al. Biomarkers of Endothelial Activation Are Associated with Poor Outcome in Critical Illness. \u003cem\u003ePLoS One\u003c/em\u003e. 2015;10(10). doi:10.1371/JOURNAL.PONE.0141251\u003c/li\u003e\n\u003cli\u003eLi Y, Liu P, Zhou Y, et al. Activation of Angiopoietin-Tie2 Signaling Protects the Kidney from Ischemic Injury by Modulation of Endothelial-Specific Pathways. \u003cem\u003eJ Am Soc Nephrol\u003c/em\u003e. 2023;34(6):969-987. doi:10.1681/ASN.0000000000000098\u003c/li\u003e\n\u003cli\u003eStar BS, Boahen CK, van der Slikke EC, et al. Plasma proteomic characterization of the development of acute kidney injury in early sepsis patients. \u003cem\u003eSci Rep\u003c/em\u003e. 2022;12(1). doi:10.1038/S41598-022-22457-W\u003c/li\u003e\n\u003cli\u003eWiersema R, Jukarainen S, Vaara ST, et al. Two subphenotypes of septic acute kidney injury are associated with different 90-day mortality and renal recovery. \u003cem\u003eCrit Care\u003c/em\u003e. 2020;24(1). doi:10.1186/S13054-020-02866-X\u003c/li\u003e\n\u003cli\u003eNI S, IS D, RG B, et al. Early Restrictive or Liberal Fluid Management for Sepsis-Induced Hypotension. \u003cem\u003eN Engl J Med\u003c/em\u003e. 2023;388(6):499-510. doi:10.1056/NEJMOA2212663\u003c/li\u003e\n\u003cli\u003eZhou H, Hewitt SM, Yuen PST, Star RA. Acute Kidney Injury Biomarkers-Needs, Present Status, and Future Promise NIH Public Access. \u003cem\u003eNephrol Self Assess Progr\u003c/em\u003e. 2006;5(2):63-71.\u003c/li\u003e\n\u003cli\u003eSinha P, Delucchi KL, Chen Y, et al. Latent class analysis-derived subphenotypes are generalisable to observational cohorts of acute respiratory distress syndrome: a prospective study. \u003cem\u003eThorax\u003c/em\u003e. 2022;77(1):13-21. doi:10.1136/THORAXJNL-2021-217158\u003c/li\u003e\n\u003cli\u003eJoannidis M, Metnitz PGH. Epidemiology and natural history of acute renal failure in the ICU. \u003cem\u003eCrit Care Clin\u003c/em\u003e. 2005;21(2):239-249. doi:10.1016/J.CCC.2004.12.005\u003c/li\u003e\n\u003cli\u003eHoste EAJ, Bagshaw SM, Bellomo R, et al. Epidemiology of acute kidney injury in critically ill patients: the multinational AKI-EPI study. \u003cem\u003eIntensive Care Med\u003c/em\u003e. 2015;41(8):1411-1423. doi:10.1007/S00134-015-3934-7\u003c/li\u003e\n\u003cli\u003eRonco C, Bellomo R, Kellum JA. Acute kidney injury. \u003cem\u003eLancet (London, England)\u003c/em\u003e. 2019;394(10212):1949-1964. doi:10.1016/S0140-6736(19)32563-2\u003c/li\u003e\n\u003cli\u003eClermont G, Acker CG, Angus DC, Sirio CA, Pinsky MR, Johnson JP. Renal failure in the ICU: Comparison of the impact of acute renal failure and end-stage renal disease on ICU outcomes. \u003cem\u003eKidney Int\u003c/em\u003e. 2002;62(3):986-996. doi:10.1046/j.1523-1755.2002.00509.x\u003c/li\u003e\n\u003cli\u003eKhader A, Zelnick LR, Sathe NA, et al. The Interaction of Acute Kidney Injury with Resuscitation Strategy in Sepsis: A Secondary Analysis of a Multicenter, Phase 3, Randomized Trial (CLOVERS). \u003cem\u003eAm J Respir Crit Care Med\u003c/em\u003e. October 2023. doi:10.1164/RCCM.202308-1448LE\u003c/li\u003e\n\u003cli\u003eNI S, IS D, RG B, et al. Early Restrictive or Liberal Fluid Management for Sepsis-Induced Hypotension. \u003cem\u003eN Engl J Med\u003c/em\u003e. 2023;388(6):499-510. doi:10.1056/NEJMOA2212663\u003c/li\u003e\n\u003cli\u003eReynolds PM, Stefanos S, MacLaren R. Restrictive resuscitation in patients with sepsis and mortality: A systematic review and meta-analysis with trial sequential analysis. \u003cem\u003ePharmacotherapy\u003c/em\u003e. 2023;43(2):104-114. doi:10.1002/phar.2764\u003c/li\u003e\n\u003cli\u003eManrique-Caballero CL, Del Rio-Pertuz G, Gomez H. Sepsis-Associated Acute Kidney Injury. \u003cem\u003eCrit Care Clin\u003c/em\u003e. 2021;37(2):279-301. doi:10.1016/J.CCC.2020.11.010\u003c/li\u003e\n\u003cli\u003eChewcharat A, Chang YT, Sise ME, Bhattacharyya RP, Murray MB, Nigwekar SU. Phase-3 Randomized Controlled Trials on Exclusion of Participants With Kidney Disease in COVID-19. \u003cem\u003eKidney Int reports\u003c/em\u003e. 2021;6(1):196-199. doi:10.1016/J.EKIR.2020.10.010\u003c/li\u003e\n\u003cli\u003eCoca SG, Vasquez-Rios G, Mansour SG, et al. Plasma Soluble Tumor Necrosis Factor Receptor Concentrations and Clinical Events After Hospitalization: Findings From the ASSESS-AKI and ARID Studies. \u003cem\u003eAm J Kidney Dis\u003c/em\u003e. 2023;81(2):190-200. doi:10.1053/J.AJKD.2022.08.007\u003c/li\u003e\n\u003cli\u003eNiewczas MA, Pavkov ME, Skupien J, et al. A signature of circulating inflammatory proteins and development of end-stage renal disease in diabetes. \u003cem\u003eNat Med\u003c/em\u003e. 2019;25(5):805-813. doi:10.1038/S41591-019-0415-5\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1. Baseline characteristics stratified by sub-phenotypes.\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"714\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo AKI\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(N = 443)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSP1\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(N = 605)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSP2\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e(N = 241)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge (years), average \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e56.1 \u0026plusmn; 17.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e61.7 \u0026plusmn; 14.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e62.7 \u0026plusmn; 15.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eMale, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e217 (49%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e332 (55%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e133 (55%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRace, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eWhite\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e327 (74%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e435 (72%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e158 (66%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eBlack\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e51 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e100 (17%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e55 (23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eAsian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e18 (4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e15 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e8 (3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eOther\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e6 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e5 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e2 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eNot reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e41 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e50 (8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e18 (7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eEthnicity, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eHispanic/Latino\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e74 (17%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e88 (15%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e28 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eNot Hispanic/Latino\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e353 (80%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e493 (81%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e207 (86%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eNot reported\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e16 (4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e24 (4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e6 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRandomization location, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eED\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e412 (93%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e566 (94%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e209 (87%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eICU\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e21 (5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e38 (6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e30 (12%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003eOther\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e10 (2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e1 (0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e2 (1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eInvasive mechanical ventilation, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e26 (6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e35 (6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e21 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSource of infection, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Pneumonia\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e152 (34%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e163 (27%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e55 (23%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Urinary tract infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e83 (19%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e168 (28%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e57 (24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Intra-abdominal infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e43 (10%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e55 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e31 (13%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Skin or soft tissue infection\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e57 (13%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e54 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e21 (9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Other or unknown\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e108 (24%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e165 (27%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e77 (32%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eWeight (kg), average \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e75.1 \u0026plusmn; 26.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e80.7 \u0026plusmn; 24.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e80.9 \u0026plusmn; 24.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eBMI (kg/m2), average \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e26.3 \u0026plusmn; 8.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e28.1 \u0026plusmn; 7.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e28.2 \u0026plusmn; 8.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eSOFA score, average \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e2.2 \u0026plusmn; 2.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e3.4 \u0026plusmn; 2.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e5.7 \u0026plusmn; 3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eHistory of CKD, n (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e19 (4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e46 (8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e74 (31%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eBaseline serum creatinine (mg/dL), average \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e0.9 \u0026plusmn; 0.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e0.8 \u0026plusmn; 0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e0.9 \u0026plusmn; 0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eRandomization serum creatinine (mg/dL), average \u0026plusmn; SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e0.9 \u0026plusmn; 0.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e2.0 \u0026plusmn; 1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e3.6 \u0026plusmn; 2.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eVolume of fluid administered before randomization (mL), median (IQR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e2000 (1400-2400)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e2100 (1525-2500)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e2100 (1450-2498)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\"\u003e\n \u003cp\u003e\u003cstrong\u003eBiomarker Concentrations, median (IQR)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003esTNFR-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e3207 (2062-5464)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e5391 (3713-8333)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e18553 (11899-27238)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eAng-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e5442 (2450-9743)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e5477 (2997-9360)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e1768 (801-3953)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eAng-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e6053 (4093-10531)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e8518 (5315-13138)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e23311 (15949-38385)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"34.78260869565217%\" valign=\"top\"\u003e\n \u003cp\u003eAng-2/Ang-1 ratio\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.141654978962134%\" valign=\"bottom\"\u003e\n \u003cp\u003e1.2 (0.6-3.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"21.458625525946704%\" valign=\"bottom\"\u003e\n \u003cp\u003e1.6 (0.8-3.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"20.61711079943899%\" valign=\"bottom\"\u003e\n \u003cp\u003e13.6 (7.4-27.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eEntries are mean (SD) for continuous variables and N (%) for categorical variables, except as noted. Abbreviations: BMI, body mass index; SOFA, sequential organ failure assessment scores; CKD, chronic kidney disease; sTNFR-1, soluble tumor necrosis factor receptor-1; Ang-1, angiopoietin-1; Ang-2, angiopoietin-2\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTable 2. Treatment effects by sub-phenotype status among patients with kidney dysfunction and all patients in CLOVERS\u003c/strong\u003e\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"863\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd width=\"27.083333333333332%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eCLOVERS Populations\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.574074074074074%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e% who died, liberal\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.847222222222221%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e% who died, restrictive\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.055555555555557%\" valign=\"bottom\"\u003e\n \u003cp\u003e\u003cstrong\u003eDifference, restrictive \u0026ndash; liberal (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.939814814814813%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDifference of Difference (95% CI)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.5%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-value for interaction\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"27.083333333333332%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePatients with AKI and ESKD (n=846)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.574074074074074%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.847222222222221%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.055555555555557%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.939814814814813%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.5%\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.13888888888889%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e28-day mortality\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.944444444444445%\" valign=\"top\"\u003e\n \u003cp\u003eSP-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.574074074074074%\" valign=\"top\"\u003e\n \u003cp\u003e11%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.847222222222221%\" valign=\"top\"\u003e\n \u003cp\u003e10%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.055555555555557%\" valign=\"top\"\u003e\n \u003cp\u003e-1% (-6%, 4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.939814814814813%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e14% (2%, 27%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.5%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.05152224824356%\" valign=\"top\"\u003e\n \u003cp\u003eSP-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.4192037470726%\" valign=\"top\"\u003e\n \u003cp\u003e41%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.995316159250585%\" valign=\"top\"\u003e\n \u003cp\u003e26%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"36.53395784543326%\" valign=\"top\"\u003e\n \u003cp\u003e-15% (-27%, -3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.13888888888889%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e90-day mortality\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.944444444444445%\" valign=\"top\"\u003e\n \u003cp\u003eSP-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.574074074074074%\" valign=\"top\"\u003e\n \u003cp\u003e18%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.847222222222221%\" valign=\"top\"\u003e\n \u003cp\u003e17%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.055555555555557%\" valign=\"top\"\u003e\n \u003cp\u003e-1% (-7%, 5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.939814814814813%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e13% (-1%, 27%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.5%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.05152224824356%\" valign=\"top\"\u003e\n \u003cp\u003eSP-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.4192037470726%\"\u003e\n \u003cp\u003e48%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.995316159250585%\"\u003e\n \u003cp\u003e36%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"36.53395784543326%\"\u003e\n \u003cp\u003e-13% (-24%, -1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePatients without AKI or ESRD (n=443)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e28-day mortality\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSP-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e8%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2% (-3%, 6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\"\u003e\n \u003cp\u003e2% (-33%, 38%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.05152224824356%\" valign=\"top\"\u003e\n \u003cp\u003eSP-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.4192037470726%\"\u003e\n \u003cp\u003e36%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.995316159250585%\"\u003e\n \u003cp\u003e35%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"36.53395784543326%\"\u003e\n \u003cp\u003e-1% (-36%, 34%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e90-day mortality\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSP-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e11%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e16%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e5% (-2%, 11%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\"\u003e\n \u003cp\u003e7% (-30%, 43%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.05152224824356%\" valign=\"top\"\u003e\n \u003cp\u003eSP-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.4192037470726%\"\u003e\n \u003cp\u003e43%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.995316159250585%\"\u003e\n \u003cp\u003e41%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"36.53395784543326%\"\u003e\n \u003cp\u003e-2% (-38%, 34%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"27.083333333333332%\" colspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAll Patients (n=1289)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.574074074074074%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.847222222222221%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.055555555555557%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.939814814814813%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.5%\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.13888888888889%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e28-day mortality\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.944444444444445%\" valign=\"top\"\u003e\n \u003cp\u003eSP-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.574074074074074%\" valign=\"top\"\u003e\n \u003cp\u003e9%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.847222222222221%\" valign=\"top\"\u003e\n \u003cp\u003e9%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.055555555555557%\" valign=\"top\"\u003e\n \u003cp\u003e0% (-3%, 4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.939814814814813%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e14% (2%, 25%)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.5%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.05152224824356%\" valign=\"top\"\u003e\n \u003cp\u003eSP-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.4192037470726%\" valign=\"top\"\u003e\n \u003cp\u003e41%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.995316159250585%\" valign=\"top\"\u003e\n \u003cp\u003e27%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"36.53395784543326%\" valign=\"top\"\u003e\n \u003cp\u003e-13% (-25%, -2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"20.13888888888889%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e90-day mortality\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"6.944444444444445%\" valign=\"top\"\u003e\n \u003cp\u003eSP-1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"11.574074074074074%\" valign=\"top\"\u003e\n \u003cp\u003e15%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.847222222222221%\" valign=\"top\"\u003e\n \u003cp\u003e17%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"18.055555555555557%\" valign=\"top\"\u003e\n \u003cp\u003e1% (-3%, 6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"17.939814814814813%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e14% (1%, 27%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"12.5%\" rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd width=\"14.05152224824356%\" valign=\"top\"\u003e\n \u003cp\u003eSP-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"23.4192037470726%\" valign=\"top\"\u003e\n \u003cp\u003e48%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"25.995316159250585%\" valign=\"top\"\u003e\n \u003cp\u003e36%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd width=\"36.53395784543326%\" valign=\"top\"\u003e\n \u003cp\u003e-13% (-24%, -1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"acute kidney injury, sepsis, volume, vasopressors","lastPublishedDoi":"10.21203/rs.3.rs-4523416/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4523416/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003ePurpose\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrevious work has identified two AKI sub-phenotypes (SP1 and SP2) characterized by differences in inflammation and endothelial dysfunction. Here we identify these sub-phenotypes using biospecimens collected in the emergency department and test for differential response to restrictive versus liberal fluid strategy in sepsis-induced hypotension in the CLOVERS trial.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe applied a previously validated 3-biomarker model using plasma angiopietin-1 and 2, and soluble tumor necrosis factor receptor-1 to classify sub-phenotypes in patients with kidney dysfunction (AKI or end-stage kidney disease [ESKD]). We also compared a de novo latent class analysis (LCA) to the 3-biomarker based sub-phenotypes. Kaplan-Meier estimates were used to test for differences in outcomes and sub-phenotype by treatment interaction.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong 1289 patients, 846 had kidney dysfunction on enrollment and the 3-variable prediction model identified 605 as SP1 and 241 as SP2. The optimal LCA model identified two sub-phenotypes with high correlation with the 3-biomarker model (Cohen’s Kappa 0.8). The risk of 28 and 90-day mortality was greater in SP2 relative to SP1 independent of AKI stage and SOFA scores. Patients with SP2, characterized by more severe endothelial injury and inflammation, had a reduction in 28-day mortality with a restrictive fluid strategy versus a liberal fluid strategy (26% vs 41%), while patients with SP1 had no difference in 28-day mortality (10% vs 11%) (\u003cem\u003ep-value-for-interaction\u003c/em\u003e = 0.03).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSub-phenotypes can be identified in the emergency department that respond differently to fluid strategy in sepsis. Identification of these sub-phenotypes could inform a precision-guided therapeutic approach for patients with sepsis-induced hypotension and kidney injury.\u003c/p\u003e","manuscriptTitle":"Molecular Phenotyping of Patients with Sepsis and Kidney Injury and Differential Response to Fluid Resuscitation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-02 20:47:40","doi":"10.21203/rs.3.rs-4523416/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b6ebd536-4158-4974-84cb-3980b11d95f2","owner":[],"postedDate":"July 2nd, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-07-19T14:22:41+00:00","versionOfRecord":[],"versionCreatedAt":"2024-07-02 20:47:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4523416","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4523416","identity":"rs-4523416","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}