Association between Glycemic Variability and Short- and Long-Term Mortality in Critically Ill Patients with Cardiovascular-Kidney-Metabolic Syndrome: A Cohort Study from the MIMIC-IV Database

Association between Glycemic Variability and Short- and Long-Term Mortality in Critically Ill Patients with Cardiovascular-Kidney-Metabolic Syndrome: A Cohort Study from the MIMIC-IV Database | 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 Article Association between Glycemic Variability and Short- and Long-Term Mortality in Critically Ill Patients with Cardiovascular-Kidney-Metabolic Syndrome: A Cohort Study from the MIMIC-IV Database Kangxing Wang, Yijie Yin, Huaiyu Xiong, Yukun Zhu, Yongfang Zhou, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8337456/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Background: The newly defined Cardiovascular-Kidney-Metabolic (CKM) syndrome represents a complex phenotype characterized by extreme physiological vulnerability. While Glycemic Variability (GV) is a recognized stressor in critical illness, its specific prognostic significance within this multimorbid CKM population remains uncharacterized. We aimed to determine the independent association of GV with mortality in CKM patients and to investigate the mediating role of serum lactate. Methods: We conducted a large-scale retrospective cohort study involving 46,958 critically ill adults with CKM syndrome (stages 1-4) using the MIMIC-IV database. GV was quantified as the coefficient of variation (CV) of all glucose measurements during the ICU stay. The primary outcomes were 28-day and 180-day all-cause mortality, analyzed using multivariable Cox proportional hazards models. Mediation analysis was employed to quantify the proportion of the association statistically attributable to serum lactate levels. Results: High GV was strongly associated with increased mortality risk. In fully adjusted models, patients in the highest GV tertile faced a significantly higher risk of 28-day mortality (Adjusted Hazard Ratio [aHR], 1.23; 95% CI, 1.15–1.31) and 180-day mortality (aHR, 1.31; 95% CI, 1.24–1.37) compared to the lowest tertile (both P < 0.001). Mediation analysis suggested that serum lactate statistically mediated this association, accounting for 10.5% of the relationship with 28-day mortality (P < 0.001). Notably, interaction analyses demonstrated that the adverse association of high GV with mortality was significantly more pronounced in non-diabetic patients (P-interaction = 0.001) and in those with lower baseline illness severity (SOFA score < 5, P-interaction = 0.002). Conclusion: In critically ill patients with CKM syndrome, elevated GV is a significant, independent predictor of both short- and long-term mortality. Analysis suggests this relationship may be partially mediated by lactate-associated metabolic stress. Our findings highlight GV as a crucial prognostic marker, particularly in non-diabetic CKM patients and those presenting with seemingly lower disease severity, suggesting a potential benefit of stricter glycemic stewardship in these subgroups. Health sciences/Biomarkers Health sciences/Diseases Health sciences/Endocrinology Health sciences/Medical research Health sciences/Nephrology Health sciences/Risk factors Glycemic Variability Cardiovascular-Kidney-Metabolic Syndrome Lactate MIMIC-IV Figures Figure 1 Figure 2 Figure 3 Figure 4 Background In 2023, the American Heart Association (AHA) introduced the concept of Cardiovascular-Kidney-Metabolic (CKM) syndrome to formally recognize the multisystemic nature of metabolic disorders and their profound impact on cardiovascular and renal health [1]. This new framework highlights the progressive interplay between obesity, diabetes, chronic kidney disease (CKD), and cardiovascular disease (CVD) [2]. Critically ill patients with CKM syndrome represent an exceptionally vulnerable population within the intensive care unit (ICU), characterized by depleted physiological reserves and facing exceedingly high mortality rates [3]. Therefore, identifying modifiable risk factors specific to this high-risk cohort is a clinical and research priority. Abnormal glycemic excursions are ubiquitous in the ICU and represent a more profound metabolic derangement than simple hyperglycemia [4]. A robust body of evidence now establishes that high glycemic variability (GV), a measure of blood glucose fluctuations, is a superior predictor of adverse outcomes compared to mean glucose levels in general critically ill populations [5,6]. For example, studies using large ICU databases have demonstrated that high GV is an independent predictor of in-hospital mortality in patients with cerebrovascular disease and in general medical-surgical ICU patients [7,8]. The primary mechanisms thought to link GV to poor outcomes include the promotion of oxidative stress, endothelial dysfunction, and systemic inflammation [9,10]. While the detrimental impact of GV is increasingly recognized in various ICU cohorts, its prognostic significance within the newly conceptualized, extremely high-risk population of critically ill patients with CKM syndrome has not yet been investigated. CKM syndrome is not merely an aggregation of comorbidities but a unique physiological state of profoundly depleted reserves across interconnected organ systems [1,2]. It is therefore plausible that the metabolic instability reflected by high GV may exert a disproportionately severe impact in this specific population. This represents a critical knowledge gap in both critical care medicine and the emerging field of CKM health. Given this knowledge gap, the present study was designed to leverage a large, contemporary cohort of critically ill patients with CKM syndrome from a high-resolution ICU database. Our primary objectives were threefold: (1) to determine the independent association between GV and both short- and long-term mortality; (2) to explore the role of serum lactate as an objective indicator of metabolic distress in this association; and (3) to identify specific patient subgroups in whom the prognostic impact of GV is most pronounced. Methods Data sources This retrospective cohort study utilized data from the Medical Information Mart for Intensive Care IV (MIMIC-IV, v3.1), a large, de-identified public database containing high-granularity data from patients admitted to the intensive care units of a tertiary academic medical center in Boston, MA, between 2008 and 2022 [11].The use of the database was approved by the Institutional Review Boards of the Massachusetts Institute of Technology and Beth Israel Deaconess Medical Center, which waived the requirement for individual patient consent; a study author (Certification no. 13024213) has completed the required training and obtained access. Study population We identified a cohort of 81,452 adult patients (≥18 years) with Cardiovascular-Kidney-Metabolic (CKM) syndrome (stages 1-4) at their first ICU admission from the MIMIC-IV (v3.1) database. The CKM diagnosis was operationalized based on the 2023 American Heart Association (AHA) criteria [2]. We sequentially excluded patients with an ICU length of stay < 24 hours (n = 16,124), those with multiple ICU admissions (n = 6,874), and those with fewer than three glucose measurements during their ICU stay (n = 11,496). This resulted in a final analytical cohort of 46,958 patients (Figure 1), who were subsequently stratified into tertiles based on their glycemic variability. Definition of CKM Syndrome We defined and staged Cardiovascular-Kidney-Metabolic (CKM) syndrome at ICU admission according to the criteria outlined in the 2023 American Heart Association (AHA) scientific statement [2]. To operationalize these criteria for the MIMIC-IV database, we established a detailed, rule-based classification protocol. This protocol systematically integrated a comprehensive combination of International Classification of Diseases (ICD-9/10) codes, specific laboratory thresholds (e.g., HbA1c, lipids, eGFR), and anthropometric data. Our final study cohort included patients meeting criteria for CKM stages 1 through 4. A complete, line-by-line breakdown of this classification protocol, detailing the specific codes, thresholds, and logic used to apply each AHA criterion, is provided in Supplementary Table S1. Data Extraction and Definitions All study data were extracted from the MIMIC-IV database (v3.1) using Structured Query Language (SQL). The primary exposure, glycemic variability (GV), was defined as the coefficient of variation (CV). The CV was calculated as the ratio of the standard deviation (SD) to the mean of all glucose measurements throughout the entire ICU stay, expressed as a percentage (CV = [SD / mean] × 100) [12]. We collected a comprehensive set of baseline covariates, defined within the first 24 hours of ICU admission unless otherwise specified. These covariates included: (1) Demographics: age, gender, race, and body mass index (BMI); (2) Severity of illness scores: Sequential Organ Failure Assessment (SOFA), Oxford Acute Severity of Illness Score (OASIS), Glasgow Coma Scale (GCS), and the Charlson Comorbidity Index; (3) Baseline diagnoses and comorbidities: hypertension, diabetes, chronic obstructive pulmonary disease (COPD), heart failure (HF), stroke, and CKM stage were identified using International Classification of Diseases (ICD-9/10) codes or predefined criteria. Sepsis was defined according to the Sepsis-3 criteria [13]. We also collected data on laboratory values, vital signs, and interventions. For laboratory values, we used the first available measurement. For vital signs, we used the most deranged values recorded. Interventions, including mechanical ventilation and vasopressor use, were documented if they occurred within this 24-hour window. Study Endpoints The primary endpoints were 28-day and 180-day all-cause mortality. Vital status was determined by linking electronic health records with the U.S. Social Security Death Index, a method that accurately captures both in-hospital and out-of-hospital deaths. Statistical analysis Baseline characteristics were presented as mean (SD) or median (IQR) for continuous variables and as frequency (percentage) for categorical variables, with comparisons across glycemic variability (GV) tertiles (T1-T3) performed using appropriate statistical tests. Multivariable Cox proportional hazards regression models were employed to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between GV tertiles and mortality. Three sequential models were constructed: Model 1 (unadjusted), Model 2 (adjusted for age, gender, race), and Model 3 (fully adjusted for baseline demographics, comorbidities, severity of illness scores, vital signs, laboratory values, and interventions). The dose-response relationship was visualized using restricted cubic splines with three knots, and survival probabilities were illustrated with Kaplan-Meier curves (log-rank test). Subgroup analyses were conducted across strata of age, BMI, SOFA score, diabetes, sepsis, and CKM stages, with interaction terms tested. A mediation analysis was performed to assess the role of serum lactate. To handle missing data, we performed multiple imputation using the chained equations (MICE) package on all covariates with less than 40% missing values; variables with missingness above this threshold were excluded. To ensure the robustness of our findings, sensitivity analyses were performed by extending the follow-up period to 90 and 365 days, and by repeating the primary analysis on a dataset imputed using the random forest method. All analyses were performed using R software (version 4.4.2), and a two-sided P-value < 0.05 was considered significant. Results Baseline characteristics The baseline characteristics of the 46,958 included patients are detailed in Table 1 . Patients in the highest GV tertile (T3) presented with a significantly greater burden of critical illness compared to those in the lowest tertile (T1). This was evidenced by higher severity scores (median SOFA: 5.00 vs. 4.00; median OASIS: 33.00 vs. 30.00), a greater prevalence of comorbidities like diabetes (47.89% vs. 20.46%) and sepsis (65.12% vs. 47.98%), and consequently, a greater need for intensive organ support, including mechanical ventilation (50.17% vs. 39.64%) and vasopressors (42.74% vs. 29.62%; all P < 0.001). Crucially, this gradient of severity directly translated to worse unadjusted outcomes: 28-day mortality in the T3 group was nearly double that of the T1 group (20.48% vs. 11.38%; P < 0.001), a stark trend that persisted for 180-day mortality (33.02% vs. 19.79%; P < 0.001). Survival analysis The multivariable Cox regression analyses revealed a significant, dose-dependent association between GV and mortality (Table 2). While this association was pronounced in the unadjusted model, it remained robust after comprehensive adjustment (Model 3). In this final model, patients in the highest GV tertile (T3) faced a significantly increased risk of both 28-day (aHR, 1.23; 95% CI, 1.15–1.31; P < 0.001) and 180-day mortality (aHR, 1.31; 95% CI, 1.24–1.37; P < 0.001) compared to the lowest tertile (T1). Although attenuated, a significantly increased risk for 180-day mortality was also evident for the medium tertile (T2) (aHR, 1.08; 95% CI, 1.03–1.13; P = 0.002). These findings establish high GV as a potent, independent predictor of both short- and long-term mortality in this high-risk population. Association with outcomes Visual inspection of the dose-response relationship confirmed a continuous, positive association between GV and mortality risk (Figure 2A-B). Restricted cubic spline analysis revealed a near-linear relationship, with the adjusted hazard for both 28-day and 180-day mortality progressively increasing with higher GV (both P for nonlinearity > 0.05). This dose-dependent effect was further corroborated by the Kaplan-Meier survival analysis (Figure 2C-D). The survival curves for the three GV tertiles separated early and continued to diverge throughout the follow-up period, with the highest GV tertile (T3) exhibiting significantly lower survival probability compared to the other groups (P for log-rank < 0.01 for overall comparison). Subgroup analyses We performed prespecified subgroup analyses to test the consistency of the association between high GV (T3 vs. T1) and mortality (Figure 3). The adverse prognostic effect of high GV was broadly consistent across most patient strata, including age, sepsis, and CKM stage (all P for interaction > 0.05). However, we identified significant effect modification by both diabetes status and baseline illness severity. The detrimental impact of high GV was substantially more pronounced in patients without pre-existing diabetes (aHR for 28-day mortality: 1.42 [95% CI, 1.31–1.53]; P for interaction = 0.001) and in those with a lower initial severity of illness (SOFA score < 5; aHR: 1.32 [95% CI, 1.18–1.47]; P for interaction = 0.002). A similar pattern of interaction was observed for 180-day mortality. While a weaker interaction was noted for BMI with 28-day mortality (P for interaction = 0.026), this was not significant for the 180-day outcome and may represent a chance finding. Mediation analysis To formally test our hypothesis, a statistical mediation analysis was conducted to quantify the mediating role of serum lactate in the association between GV and mortality (Figure 4). For the primary endpoint of 28-day mortality, serum lactate was a statistically significant mediator across all three adjustment models. In the fully adjusted model (Model 3), the proportion of the total effect mediated by lactate was 10.54% (P < 0.001). A similar significant mediating effect was observed for 180-day mortality, with a mediated proportion of 5.33% in the final model (P < 0.001). For both 28-day and 180-day outcomes, both the average causal mediation effect (ACME, the indirect path through lactate) and the average direct effect (ADE, the direct path of GV) were statistically significant (all P < 0.001). Sensitivity Analyses The robustness of our findings was confirmed through two sensitivity analyses. Repeating the primary analysis with an alternative random forest-based imputation method yielded highly consistent results (Supplementary Table S2). Furthermore, the prognostic association of high GV (T3 vs. T1) proved durable, remaining significantly associated with mortality in the fully adjusted model at both 90 days (aHR, 1.32; 95% CI, 1.25–1.39) and 365 days (aHR, 1.29; 95% CI, 1.24–1.35) (Supplementary Table S3). Collectively, these results establish that our primary findings are robust to the choice of imputation method and are sustained over long-term follow-up. Discussion In this large-scale retrospective cohort study of critically ill patients with Cardiovascular-Kidney-Metabolic (CKM) syndrome, we have observed for the first time that high glycemic variability (GV) is a potent and independent predictor of both short- and long-term mortality. This association persisted even after extensive adjustment for a wide range of demographic, clinical, and severity-of-illness confounders. Notably, our analysis uncovered two critical nuances: the mortality risk associated with high GV was significantly more pronounced in patients without a prior diagnosis of diabetes and in those with a lower initial severity of illness. Furthermore, we identified serum lactate as a significant, albeit partial, mediator in the pathway linking GV to adverse outcomes, providing a potential mechanistic link for this association. Our primary finding—that high GV is associated with increased mortality—is broadly consistent with a substantial body of literature in general and mixed ICU populations [8,14,15]. Previous studies have established GV as a superior prognostic marker compared to mean glucose levels in diverse cohorts, including patients with sepsis, cerebrovascular disease, and post-cardiac surgery [16–18]. Our results reinforce this paradigm, extending its validity to the unique and extremely vulnerable population of critically ill patients with CKM syndrome. However, the novelty of our study lies in its specific focus on this newly defined syndrome. The CKM framework conceptualizes a progressive state of depleted physiological reserves across interconnected organ systems [1,2]. Patients with CKM syndrome are characterized by chronic inflammation, endothelial dysfunction, and metabolic inflexibility, creating a physiological milieu that may amplify the deleterious effects of acute metabolic instability reflected by high GV [2,19,20]. Perhaps the most striking finding of our study is the significant effect modification by diabetes status, where the prognostic harm of high GV was substantially greater in patients without pre-existing diabetes. This seemingly paradoxical result is not an isolated observation but aligns with a growing body of evidence suggesting that stress-induced hyperglycemia carries a graver prognosis in individuals without a history of diabetes across various critical illnesses [21,22]. Two primary hypotheses may explain this phenomenon. First, the concept of cellular "preconditioning" posits that chronic exposure to hyperglycemia in diabetic patients may induce adaptive mechanisms, such as an upregulation of endogenous antioxidant defenses, making them relatively more resilient to the acute metabolic insults posed by high GV [23]. In contrast, the glucose-naïve tissues of non-diabetic patients are more susceptible to the glucotoxicity from acute glycemic excursions, which are known to trigger more severe oxidative stress and endothelial dysfunction than sustained hyperglycemia [24]. Second, the etiology of high GV likely differs between the two groups. In non-diabetic patients, high GV serves as a more direct and unadulterated marker of an extreme endogenous stress response, reflecting a profound dysregulation of counter-regulatory hormones [25]. In diabetic patients, however, GV represents a complex interplay between this endogenous stress and exogenous factors like insulin administration, potentially confounding its interpretation as a pure indicator of acute physiological stress and thus diluting its prognostic strength. Another key and seemingly paradoxical finding was that the prognostic association between high GV and mortality was strongest in patients with a lower initial burden of organ dysfunction (SOFA score < 5). This suggests that in patients with overwhelming multi-organ failure (high SOFA score), their trajectory towards mortality is likely governed by a confluence of established and often irreversible pathophysiological processes. In this context, the independent prognostic contribution of any single variable like GV is diminished, a phenomenon that can be understood in prognostic modeling where the "signal" of one biomarker is obscured by the overwhelming "noise" of global illness severity [26]. In stark contrast, for patients who are less critically ill at baseline, this "lower-noise" environment allows acute metabolic dysregulation, as captured by high GV, to emerge as a more significant prognostic factor. Here, it may function as a critical "second hit" that potentially precipitates a cascade of clinical deterioration and subsequent organ injury leading to death [27]. Therefore, our finding highlights the potential role of GV not merely as a prognostic marker, but as a crucial and early indicator of impending physiological collapse, particularly in patients who might otherwise be stratified into a lower-risk category. Our mediation analysis provides novel, quantitative evidence for a specific biological pathway linking high GV to mortality, finding that serum lactate partially mediates this relationship. This is highly biologically plausible, as high GV is known to induce excessive reactive oxygen species (ROS) production, leading to profound mitochondrial dysfunction [9,10]. This impaired mitochondrial respiration can force cells into anaerobic glycolysis, even in the presence of adequate oxygen—a state of cytopathic hypoxia or Type B lactic acidosis—resulting in lactate accumulation [28]. Thus, in this context, serum lactate transcends its role as a simple marker of tissue hypoperfusion and becomes a direct indicator of cellular metabolic distress, a phenomenon our study now quantitatively links to GV. Importantly, the mediation was only partial (~10%), indicating that the majority of the association between GV and mortality is likely explained by other pathways. These likely include an exacerbated systemic inflammatory response, as hyperglycemia is known to acutely increase pro-inflammatory cytokines [29], and heightened endothelial injury, as fluctuating glucose levels are particularly damaging to the vascular endothelium [10]. In summary, our findings carry significant clinical implications, advocating for a shift in focus from merely avoiding hyperglycemia—a lesson from the landmark NICE-SUGAR trial [30]—towards achieving glycemic stability [5]. GV should be considered a critical 'metabolic vital sign,' particularly for risk stratification in non-diabetic and less severely ill CKM patients [5]. While our large-scale analysis provides a robust foundation for this view, several limitations must be acknowledged. First, given the retrospective observational design, we can only report associations; causal relationships between GV and mortality cannot be definitively established, and the possibility of residual confounding remains despite extensive adjustment. Second, the reliance on intermittent glucose measurements derived from electronic health records may not capture all glycemic excursions as accurately as continuous glucose monitoring (CGM). Third, this is a single-center study based on the MIMIC-IV database; thus, selection bias is inevitable, and our findings require external validation in multicenter cohorts to ensure generalizability to other populations or healthcare systems. These limitations, however, directly inform a clear path for future research. Prospective, multicenter studies are warranted to confirm our findings and, ultimately, to test whether advanced strategies specifically aimed at minimizing GV—guided by technologies like CGM and closed-loop insulin delivery systems [31]—can translate into improved survival for this exceptionally high-risk population. Conclusion In conclusion, our study Identifies high glycemic variability as a significant, independent risk factor for both short- and long-term mortality in the critically ill Cardiovascular-Kidney-Metabolic (CKM) syndrome population. The prognostic harm of high GV was particularly pronounced in patients without pre-existing diabetes and those with a lower initial severity of illness, identifying key subgroups for heightened clinical vigilance. These findings underscore GV as a crucial marker of profound metabolic decompensation and position glycemic stability, rather than mere glycemic level, as a key therapeutic target for future investigation in this exceptionally vulnerable population. Abbreviations ACME : Average Causal Mediation Effect ADE : Average Direct Effect AHA : American Heart Association aHR : Adjusted Hazard Ratio BMI : Body Mass Index CGM : Continuous Glucose Monitoring CI : Confidence Interval CKD : Chronic Kidney Disease CKM : Cardiovascular-Kidney-Metabolic COPD : Chronic Obstructive Pulmonary Disease CV : Coefficient of Variation CVD : Cardiovascular Disease GCS : Glasgow Coma Scale GV : Glycemic Variability HF : Heart Failure HR : Hazard Ratio ICU : Intensive Care Unit MICE : Multiple Imputation using Chained Equations MIMIC-IV : Medical Information Mart for Intensive Care IV OASIS : Oxford Acute Severity of Illness Score ROS : Reactive Oxygen Species SD : Standard Deviation SOFA : Sequential Organ Failure Assessment SQL : Structured Query Language Declarations Ethics approval and consent to participate: The study was conducted in accordance with the Declaration of Helsinki. Ethical approval and individual patient consent were waived for the MIMIC-IV database because it contains de-identified health information that is publicly available for research purposes. The use of the MIMIC-IV database for this study was specifically approved by the Massachusetts Institute of Technology Institutional Review Board and Beth Israel Deaconess Medical Center. The authorized researcher (certification number 13024213) completed the required data user training, which granted access approval for the database. Consent for publication: Not applicable. Availability of data and materials The dataset utilized in this study was obtained from a large-scale, publicly available critical care database: the Medical Information Mart for Intensive Care IV (MIMIC-IV, version 3.1). Data access was granted and data were extracted by an authorized researcher (Kangxing Wang, certification number: 13024213) after obtaining necessary approvals and completing ethical training. The dataset is publicly accessible for research purposes via PhysioNet. Competing Interests: The authors declare no competing financial interests. Fundings: The study was approved by National Key R&D Program of China（2022YFC2504500）and 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University（ZYGD23012） Author Contributions WKX, XHY, ZYK, and YYJ contributed equally to this work as co-first authors. WKX was the primary contributor, responsible for the study conception, essential data acquisition (coding and extraction), performing the main statistical analysis, and drafting the manuscript. XHY refined the methodology, provided technical support for the analysis, and assisted with data management. ZYK assisted with the statistical modeling, interpretation of results, and critical revision of the manuscript. YYJ contributed to data validation, visualization of results, and substantial revision of the manuscript. KY and ZYF served as co-corresponding authors, secured funding, and supervised the study. Both KY and ZYF critically revised the manuscript for intellectual content. All authors read and approved the final manuscript. Acknowledgments: Not applicable References Ndumele CE, Neeland IJ, Tuttle KR, Chow SL, Mathew RO, Khan SS, et al. A synopsis of the evidence for the science and clinical management of cardiovascular-kidney-metabolic (CKM) syndrome: a scientific statement from the american heart association. Circulation. United States; 2023;148:1636–64. https://doi.org/10.1161/CIR.0000000000001186 Ndumele CE, Rangaswami J, Chow SL, Neeland IJ, Tuttle KR, Khan SS, et al. Cardiovascular-kidney-metabolic health: a presidential advisory from the american heart association. Circulation. 2023;148:1606–35. https://doi.org/10.1161/CIR.0000000000001184 Aggarwal R, Ostrominski JW, Vaduganathan M. Prevalence of cardiovascular-kidney-metabolic syndrome stages in US adults, 2011-2020. JAMA. 2024;331:1858. https://doi.org/10.1001/jama.2024.6892 Service FJ. Glucose variability. Diabetes. 2013;62:1398–404. https://doi.org/10.2337/db12-1396 Krinsley JS. Glycemic variability: a strong independent predictor of mortality in critically ill patients. Crit Care Med. 2008;36:3008–13. https://doi.org/10.1097/CCM.0b013e31818b38d2 Bagshaw SM, Bellomo R, Jacka MJ, Egi M, Hart GK, George C, et al. The impact of early hypoglycemia and blood glucose variability on outcome in critical illness. Crit Care. 2009;13:R91. https://doi.org/10.1186/cc7921 Su Y, Fan W, Liu Y, Hong K. Glycemic variability and in-hospital death of critically ill patients and the role of ventricular arrhythmias. Cardiovasc Diabetol. England; 2023;22:134. https://doi.org/10.1186/s12933-023-01861-0 Wang F, Guo Y, Tang Y, Zhao S, Xuan K, Mao Z, et al. Combined assessment of stress hyperglycemia ratio and glycemic variability to predict all-cause mortality in critically ill patients with atherosclerotic cardiovascular diseases across different glucose metabolic states: an observational cohort study with machine learning. Cardiovasc Diabetol. 2025;24:199. https://doi.org/10.1186/s12933-025-02762-0 Ceriello A, Monnier L, Owens D. Glycaemic variability in diabetes: clinical and therapeutic implications. Lancet Diabetes Endocrinol. 2019;7:221–30. https://doi.org/10.1016/S2213-8587(18)30136-0 Ceriello A, Esposito K, Piconi L, Ihnat MA, Thorpe JE, Testa R, et al. Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients. Diabetes. 2008;57:1349–54. https://doi.org/10.2337/db08-0063 Johnson AEW, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data. 2023;10:1. https://doi.org/10.1038/s41597-022-01899-x Chun K-H, Oh J, Lee CJ, Park JJ, Lee SE, Kim M-S, et al. In-hospital glycemic variability and all-cause mortality among patients hospitalized for acute heart failure. Cardiovasc Diabetol. 2022;21:291. https://doi.org/10.1186/s12933-022-01720-4 Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA. 2016;315:801. https://doi.org/10.1001/jama.2016.0287 Qi L, Geng X, Feng R, Wu S, Fu T, Li N, et al. Association of glycemic variability and prognosis in patients with traumatic brain injury: a retrospective study from the MIMIC-IV database. Diabetes Res Clin Pract. 2024;217:111869. https://doi.org/10.1016/j.diabres.2024.111869 He H, Xie Y, Wang Z, Li J, Zheng S, Li X, et al. Associations of variability in blood glucose and systolic blood pressure with mortality in patients with coronary artery disease: a retrospective cohort study from the MIMIC-IV database. Diabetes Res Clin Pract. 2024;209:111595. https://doi.org/10.1016/j.diabres.2024.111595 Zhou Z, Zhang H, Gu Y, Zhang K, Ouyang C. Relationship between glycemic variability and the incidence of postoperative atrial fibrillation following cardiac surgery: a retrospective study from MIMIC-IV database. Diabetes Res Clin Pract. Ireland; 2024;219:111978. https://doi.org/10.1016/j.diabres.2024.111978 Ali NA, OʼBrien JM, Dungan K, Phillips G, Marsh CB, Lemeshow S, et al. Glucose variability and mortality in patients with sepsis. Crit Care Med. 2008;36:2316–21. https://doi.org/10.1097/CCM.0b013e3181810378 Cai W, Li Y, Guo K, Wu X, Chen C, Lin X. Association of glycemic variability with death and severe consciousness disturbance among critically ill patients with cerebrovascular disease: analysis of the MIMIC-IV database. Cardiovasc Diabetol. England; 2023;22:315. https://doi.org/10.1186/s12933-023-02048-3 Massy ZA, Drueke TB. Combination of cardiovascular, kidney, and metabolic diseases in a syndrome named cardiovascular-kidney-metabolic, with new risk prediction equations. Kidney Int Rep. 2024;9:2608–18. https://doi.org/10.1016/j.ekir.2024.05.033 Baaten CCFMJ, Vondenhoff S, Noels H. Endothelial cell dysfunction and increased cardiovascular risk in patients with chronic kidney disease. Circ Res. 2023;132:970–92. https://doi.org/10.1161/CIRCRESAHA.123.321752 Kosiborod M, Rathore SS, Inzucchi SE, Masoudi FA, Wang Y, Havranek EP, et al. Admission glucose and mortality in elderly patients hospitalized with acute myocardial infarction: implications for patients with and without recognized diabetes. Circulation. 2005;111:3078–86. https://doi.org/10.1161/CIRCULATIONAHA.104.517839 He H-M, Wang Z, Xie Y-Y, Zheng S-W, Li J, Li X-X, et al. Maximum stress hyperglycemia ratio within the first 24 h of admission predicts mortality during and after the acute phase of acute coronary syndrome in patients with and without diabetes: a retrospective cohort study from the MIMIC-IV database. Diabetes Res Clin Pract. Ireland; 2024;208:111122. https://doi.org/10.1016/j.diabres.2024.111122 Ceriello A. High glucose induces antioxidant enzymes in human endothelial cells in culture. Monnier L, Mas E, Ginet C, Michel F, Villon L, Cristol J-P, et al. Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. Dungan KM, Braithwaite SS, Preiser J-C. Stress hyperglycaemia. Moore FA, Moore EE. Evolving concepts in the pathogenesis of postinjury multiple organ failure. Surg Clin North Am. 1995;75:257–77. https://doi.org/10.1016/S0039-6109(16)46587-4 Angele MK, Chaudry IH. Surgical trauma and immunosuppression: pathophysiology and potential immunomodulatory approaches. Langenbecks Arch Surg. 2005;390:333–41. https://doi.org/10.1007/s00423-005-0557-4 Nedel W, Deutschendorf C, Portela LVC. Sepsis-induced mitochondrial dysfunction: a narrative review. World J Crit Care Med. 2023;12:139–52. https://doi.org/10.5492/wjccm.v12.i3.139 Esposito K, Nappo F, Marfella R, Giugliano G, Giugliano F, Ciotola M, et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation. 2002;106:2067–72. https://doi.org/10.1161/01.CIR.0000034509.14906.AE Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283–97. https://doi.org/10.1056/NEJMoa0810625 Nimri R, Phillip M, Clements MA, Kovatchev B. Closed-loop control, artificial intelligence–based decision-support systems, and data science. Diabetes Technol Ther. 2024;26:S-68-S-89. https://doi.org/10.1089/dia.2024.2505 Tables Tables are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Table1.xlsx Table 1. Baseline Characteristics of Critically Ill Patients with CKM Syndrome, Stratified by Tertiles of Glycemic Variability. Data are presented as mean (standard deviation) for continuous variables and as n (%) for categorical variables. P-values were calculated using the analysis of variance or Kruskal-Wallis test for continuous variables and the chi-squared test for categorical variables, as appropriate. Abbreviations: BMI, body mass index; CKM, Cardiovascular-Key-Metabolic; COPD, chronic obstructive pulmonary disease; GCS, Glasgow Coma Scale; GV, glycemic variability; HF, heart failure; ICU, intensive care unit; IQR, interquartile range; MAP, mean arterial pressure; OASIS, Oxford Acute Severity of Illness Score; RRT, renal replacement therapy; SOFA, Sequential Organ Failure Assessment. Table2.xlsx Table 2. Association between Glycemic Variability Tertiles and Mortality in Critically Ill Patients with CKM Syndrome. Data are presented as hazard ratios (HRs) and 95% confidence intervals (CIs) derived from Cox proportional hazards models. T1 (lowest glycemic variability tertile) serves as the reference group. Model 1: Unadjusted. Model 2: Adjusted for age, gender, and race. Model 3: Fully adjusted for age, gender, and race (from Model 2), plus the following baseline covariates: Severity of illness scores (Sequential Organ Failure Assessment [SOFA], Oxford Acute Severity of Illness Score [OASIS], Glasgow Coma Scale [GCS], Charlson Comorbidity Index); Comorbidities (hypertension, diabetes, chronic obstructive pulmonary disease, heart failure, stroke, sepsis); Vital signs (minimum mean arterial pressure, maximum heart rate, maximum respiratory rate, maximum temperature); Laboratory values (first measurement of hemoglobin, white blood cell count, platelets, creatinine, lactate, PO2, PCO2, potassium, sodium, chloride, calcium, pH); and Interventions (use of insulin, mechanical ventilation, renal replacement therapy, and vasopressors). Abbreviations: CI, confidence interval; GV, glycemic variability; HR, hazard ratio; CKM, Cardiovascular-Kidney-Metabolic. Supplementary.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 19 Apr, 2026 Editor assigned by journal 19 Apr, 2026 Editor invited by journal 17 Dec, 2025 Submission checks completed at journal 16 Dec, 2025 First submitted to journal 16 Dec, 2025 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-8337456","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":628231742,"identity":"520acd47-ceb0-4b93-9967-41006ed9cb90","order_by":0,"name":"Kangxing Wang","email":"","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Kangxing","middleName":"","lastName":"Wang","suffix":""},{"id":628231743,"identity":"4a696f8f-633b-4b65-94f6-b4c38e7365ed","order_by":1,"name":"Yijie Yin","email":"","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Yijie","middleName":"","lastName":"Yin","suffix":""},{"id":628231744,"identity":"e0720118-4103-4e0e-b469-7c6b40f65641","order_by":2,"name":"Huaiyu Xiong","email":"","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Huaiyu","middleName":"","lastName":"Xiong","suffix":""},{"id":628231745,"identity":"265c2b37-5141-4e52-a7ff-7f27baede9db","order_by":3,"name":"Yukun Zhu","email":"","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Yukun","middleName":"","lastName":"Zhu","suffix":""},{"id":628231746,"identity":"400b3f1a-c9b2-43ea-aee3-6b1ad2320062","order_by":4,"name":"Yongfang Zhou","email":"","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":false,"prefix":"","firstName":"Yongfang","middleName":"","lastName":"Zhou","suffix":""},{"id":628231747,"identity":"ecf85d08-bc5a-4c84-aee9-276c16cc5a88","order_by":5,"name":"Yan Kang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0klEQVRIiWNgGAWjYBACCSgtB6HYSNBiTLqWxAaitUi2nz38mqfmTvr89jMGDB/KDjPwz27Ar0WaJy/NmufYs9wNZ3IMGGecO8wgcecAfi1yDDlmxjxsh3M3SPAYMPO2HWYwkEggoIX/DVDLv8Pp8jOAWv4So0VaIsf4MdDwBIYbQC2MxGiRnPHGjHFu32HDDWfSCg72nEvnkbhBQIvE+RzjD2++HZaXbz+88cGPMms5/hkEtAABmxQPlHUAiHnwqIQD5o8/iFE2CkbBKBgFIxcAAHf0QIqxyCiqAAAAAElFTkSuQmCC","orcid":"","institution":"West China Hospital of Sichuan University","correspondingAuthor":true,"prefix":"","firstName":"Yan","middleName":"","lastName":"Kang","suffix":""}],"badges":[],"createdAt":"2025-12-11 14:08:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8337456/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8337456/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107965757,"identity":"47113e27-265e-4d62-a9d8-7f358a5fd791","added_by":"auto","created_at":"2026-04-28 05:41:24","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":97850,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart for the selection of the study cohort.\u003cbr\u003e\n \u003cem\u003eAbbreviations\u003c/em\u003e: CKM, Cardiovascular-Kidney-Metabolic syndrome; GV, glycemic variability; ICU, intensive care unit; T1, Tertile 1 (Low GV); T2, Tertile 2 (Medium GV); T3, Tertile 3 (High GV)\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/baed2668dfc5df01739507be.png"},{"id":107965872,"identity":"a12c8561-dd94-4cb9-8c7c-65846f57555d","added_by":"auto","created_at":"2026-04-28 05:41:40","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2037954,"visible":true,"origin":"","legend":"\u003cp\u003eAssociation of Glycemic Variability with 28-day and 180-day Mortality.\u003c/p\u003e\n\u003cp\u003e(A-B) Dose-response relationship between glycemic variability (GV) and the adjusted hazard ratio (aHR) for 28-day and 180-day mortality, modeled using restricted cubic splines. Models are adjusted for all covariates in Table 1 (P for nonlinearity \u0026gt; 0.05 for both).\u003c/p\u003e\n\u003cp\u003e(C-D) Kaplan-Meier survival curves for 28-day and 180-day mortality, stratified by GV tertiles. P-value is from the log-rank test. The table indicates the number of patients at risk.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/d54fea36f7ccbbeec8b2085b.png"},{"id":107965924,"identity":"40520cee-bda0-403e-a754-164916ac8372","added_by":"auto","created_at":"2026-04-28 05:42:00","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":6685701,"visible":true,"origin":"","legend":"\u003cp\u003eSubgroup Analysis of the Association Between High Glycemic Variability and Mortality.\u003c/p\u003e\n\u003cp\u003eForest plot showing the adjusted hazard ratios (aHRs) for the association of high glycemic variability (T3 vs. T1) with (A) 28-day mortality and (B) 180-day mortality across prespecified subgroups. Hazard ratios and 95% confidence intervals are derived from the fully adjusted Cox model (Model 3). The P-values for interaction were calculated to assess for effect modification by the subgroup variable.\u003cbr\u003e\n \u003cem\u003eAbbreviations:\u003c/em\u003e BMI, body mass index; CI, confidence interval; CKM, Cardiovascular-Kidney-Metabolic; HR, hazard ratio; SOFA, Sequential Organ Failure Assessment.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/6a28b8d0b9ac6c4a343af913.png"},{"id":107965781,"identity":"82606bac-29e1-4ee5-985e-06348d91fa5c","added_by":"auto","created_at":"2026-04-28 05:41:27","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":2058522,"visible":true,"origin":"","legend":"\u003cp\u003eMediation Analysis of Serum Lactate in the Association Between Glycemic Variability and Mortality.\u003c/p\u003e\n\u003cp\u003eCausal mediation models illustrating the proportion of the total effect of glycemic variability (GV) on mortality that is mediated by serum lactate. The analysis was performed for (A-C) 28-day mortality and (D-F) 180-day mortality across three sequential adjustment models:\u003c/p\u003e\n\u003cp\u003eModel 1 (Panels A, D): Unadjusted.\u003c/p\u003e\n\u003cp\u003eModel 2 (Panels B, E): Adjusted for age, gender, and race.\u003c/p\u003e\n\u003cp\u003eModel 3 (Panels C, F): Fully adjusted for a comprehensive set of baseline covariates, including: Demographics (age, gender, race, BMI); Severity of illness scores (SOFA, GCS, OASIS, Charlson Comorbidity Index); Comorbidities (chronic obstructive pulmonary disease, sepsis); Vital signs (minimum mean arterial pressure, maximum heart rate, maximum respiratory rate, maximum temperature); Laboratory values (first measurement of pH, hemoglobin, platelets, creatinine, PO2, PCO2, potassium, sodium, chloride, calcium); and Interventions (use of insulin in the first 24h, ICU insulin, mechanical ventilation, renal replacement therapy, and vasopressors).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eAbbreviations: \u003c/em\u003eACME (Average Causal Mediation Effect) represents the indirect effect channeled through lactate. ADE (Average Direct Effect) represents the direct effect of GV on mortality not mediated by lactate. All effects were statistically significant (P \u0026lt; 0.001).\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/fb1e4d66c8ab80837147b32e.png"},{"id":107966090,"identity":"05cea210-7d5f-4392-9e8f-35a073d3b598","added_by":"auto","created_at":"2026-04-28 05:42:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":11511070,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/a5d1077b-f63a-4565-987e-074da36303d5.pdf"},{"id":107965839,"identity":"d3e699ad-dce6-4ad7-8208-7635786e09dc","added_by":"auto","created_at":"2026-04-28 05:41:39","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":12766,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable 1. \u003c/strong\u003eBaseline Characteristics of Critically Ill Patients with CKM Syndrome, Stratified by Tertiles of Glycemic Variability.\u003c/p\u003e\n\u003cp\u003eData are presented as mean (standard deviation) for continuous variables and as n (%) for categorical variables.\u003cbr\u003e\nP-values were calculated using the analysis of variance or Kruskal-Wallis test for continuous variables and the chi-squared test for categorical variables, as appropriate.\u003cbr\u003e\n \u003cem\u003eAbbreviations: \u003c/em\u003eBMI, body mass index; CKM, Cardiovascular-Key-Metabolic; COPD, chronic obstructive pulmonary disease; GCS, Glasgow Coma Scale; GV, glycemic variability; HF, heart failure; ICU, intensive care unit; IQR, interquartile range; MAP, mean arterial pressure; OASIS, Oxford Acute Severity of Illness Score; RRT, renal replacement therapy; SOFA, Sequential Organ Failure Assessment.\u003c/p\u003e","description":"","filename":"Table1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/48b21b928f25eab9ae40a869.xlsx"},{"id":107965751,"identity":"b27be616-85ff-44b2-a1eb-e4866519acec","added_by":"auto","created_at":"2026-04-28 05:41:24","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":9567,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTable 2. \u003c/strong\u003eAssociation between Glycemic Variability Tertiles and Mortality in Critically Ill Patients with CKM Syndrome.\u003c/p\u003e\n\u003cp\u003eData are presented as hazard ratios (HRs) and 95% confidence intervals (CIs) derived from Cox proportional hazards models. T1 (lowest glycemic variability tertile) serves as the reference group.\u003cbr\u003e\n Model 1: Unadjusted.\u003cbr\u003e\nModel 2: Adjusted for age, gender, and race.\u003cbr\u003e\nModel 3: Fully adjusted for age, gender, and race (from Model 2), plus the following baseline covariates: Severity of illness scores (Sequential Organ Failure Assessment [SOFA], Oxford Acute Severity of Illness Score [OASIS], Glasgow Coma Scale [GCS], Charlson Comorbidity Index); Comorbidities (hypertension, diabetes, chronic obstructive pulmonary disease, heart failure, stroke, sepsis); Vital signs (minimum mean arterial pressure, maximum heart rate, maximum respiratory rate, maximum temperature); Laboratory values (first measurement of hemoglobin, white blood cell count, platelets, creatinine, lactate, PO2, PCO2, potassium, sodium, chloride, calcium, pH); and Interventions (use of insulin, mechanical ventilation, renal replacement therapy, and vasopressors).\u003cbr\u003e\n \u003cem\u003eAbbreviations: \u003c/em\u003eCI, confidence interval; GV, glycemic variability; HR, hazard ratio; CKM, Cardiovascular-Kidney-Metabolic.\u003c/p\u003e","description":"","filename":"Table2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/4496e3549d53437c3b8725dc.xlsx"},{"id":107965742,"identity":"3f17ab81-57e5-4ce5-8abf-56691d039951","added_by":"auto","created_at":"2026-04-28 05:41:16","extension":"docx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":24467,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementary.docx","url":"https://assets-eu.researchsquare.com/files/rs-8337456/v1/a1a4f6b9648f16b7450ecc76.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Association between Glycemic Variability and Short- and Long-Term Mortality in Critically Ill Patients with Cardiovascular-Kidney-Metabolic Syndrome: A Cohort Study from the MIMIC-IV Database","fulltext":[{"header":"Background","content":"\u003cp\u003eIn 2023, the American Heart Association (AHA) introduced the concept of Cardiovascular-Kidney-Metabolic (CKM) syndrome to formally recognize the multisystemic nature of metabolic disorders and their profound impact on cardiovascular and renal health [1]. This new framework highlights the progressive interplay between obesity, diabetes, chronic kidney disease (CKD), and cardiovascular disease (CVD) [2]. Critically ill patients with CKM syndrome represent an exceptionally vulnerable population within the intensive care unit (ICU), characterized by depleted physiological reserves and facing exceedingly high mortality rates [3]. Therefore, identifying modifiable risk factors specific to this high-risk cohort is a clinical and research priority.\u003c/p\u003e\n\u003cp\u003eAbnormal glycemic excursions are ubiquitous in the ICU and represent a more profound metabolic derangement than simple hyperglycemia [4]. A robust body of evidence now establishes that high glycemic variability (GV), a measure of blood glucose fluctuations, is a superior predictor of adverse outcomes compared to mean glucose levels in general critically ill populations [5,6]. For example, studies using large ICU databases have demonstrated that high GV is an independent predictor of in-hospital mortality in patients with cerebrovascular disease and in general medical-surgical ICU patients [7,8]. The primary mechanisms thought to link GV to poor outcomes include the promotion of oxidative stress, endothelial dysfunction, and systemic inflammation [9,10].\u003c/p\u003e\n\u003cp\u003eWhile the detrimental impact of GV is increasingly recognized in various ICU cohorts, its prognostic significance within the newly conceptualized, extremely high-risk population of critically ill patients with CKM syndrome has not yet been investigated. CKM syndrome is not merely an aggregation of comorbidities but a unique physiological state of profoundly depleted reserves across interconnected organ systems [1,2]. It is therefore plausible that the metabolic instability reflected by high GV may exert a disproportionately severe impact in this specific population. This represents a critical knowledge gap in both critical care medicine and the emerging field of CKM health.\u003c/p\u003e\n\u003cp\u003eGiven this knowledge gap, the present study was designed to leverage a large, contemporary cohort of critically ill patients with CKM syndrome from a high-resolution ICU database. Our primary objectives were threefold: (1) to determine the independent association between GV and both short- and long-term mortality; (2) to explore the role of serum lactate as an objective indicator of metabolic distress in this association; and (3) to identify specific patient subgroups in whom the prognostic impact of GV is most pronounced.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eData sources\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis retrospective cohort study utilized data from the Medical Information Mart for Intensive Care IV (MIMIC-IV, v3.1), a large, de-identified public database containing high-granularity data from patients admitted to the intensive care units of a tertiary academic medical center in Boston, MA,\u0026nbsp;between 2008 and 2022 [11].The use of the database was approved by the Institutional Review Boards of the Massachusetts Institute of Technology and Beth Israel Deaconess Medical Center, which waived the requirement for individual patient consent; a study author (Certification no. 13024213) has completed the required training and obtained access.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy population\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe identified a cohort of 81,452 adult patients (\u0026ge;18 years) with Cardiovascular-Kidney-Metabolic (CKM) syndrome (stages 1-4) at their first ICU admission from the MIMIC-IV (v3.1) database. The CKM diagnosis was operationalized based on the 2023 American Heart Association (AHA) criteria [2]. We sequentially excluded patients with an ICU length of stay \u0026lt; 24 hours (n = 16,124), those with multiple ICU admissions (n = 6,874), and those with fewer than three glucose measurements during their ICU stay (n = 11,496). This resulted in a final analytical cohort of 46,958 patients (Figure 1), who were subsequently stratified into tertiles based on their glycemic variability.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDefinition of CKM Syndrome\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe defined and staged Cardiovascular-Kidney-Metabolic (CKM) syndrome at ICU admission according to the criteria outlined in the 2023 American Heart Association (AHA) scientific statement [2]. To operationalize these criteria for the MIMIC-IV database, we established a\u0026nbsp;detailed, rule-based classification protocol. This protocol systematically integrated a comprehensive combination of International Classification of Diseases (ICD-9/10) codes, specific laboratory thresholds (e.g., HbA1c, lipids, eGFR), and anthropometric data. Our final study cohort included patients meeting criteria for CKM stages 1 through 4. A complete,\u0026nbsp;line-by-line breakdown of this classification protocol, detailing the specific codes, thresholds, and logic used to apply each AHA criterion, is provided in\u0026nbsp;Supplementary Table S1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Extraction and Definitions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll study data were extracted from the MIMIC-IV database (v3.1) using Structured Query Language (SQL). The primary exposure, glycemic variability (GV), was defined as the\u0026nbsp;coefficient of variation (CV). The CV was calculated as the ratio of the standard deviation (SD) to the mean of all glucose measurements throughout the entire ICU stay, expressed as a percentage (CV = [SD / mean] \u0026times; 100) [12]. We collected a comprehensive set of baseline covariates, defined within the first 24 hours of ICU admission unless otherwise specified. These covariates included: (1) Demographics: age, gender, race, and body mass index (BMI); (2) Severity of illness scores: Sequential Organ Failure Assessment (SOFA), Oxford Acute Severity of Illness Score (OASIS), Glasgow Coma Scale (GCS), and the Charlson Comorbidity Index; (3)\u0026nbsp;Baseline diagnoses and comorbidities: hypertension, diabetes, chronic obstructive pulmonary disease (COPD), heart failure (HF), stroke, and CKM stage were identified using International Classification of Diseases (ICD-9/10) codes or predefined criteria.\u0026nbsp;Sepsis was defined according to the Sepsis-3 criteria [13]. We also collected data on\u0026nbsp;laboratory values, vital signs, and interventions. For laboratory values, we used the first available measurement. For vital signs, we used the most deranged values recorded. Interventions, including mechanical ventilation and vasopressor use, were documented if they occurred within this 24-hour window.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStudy Endpoints\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe primary endpoints were 28-day and 180-day all-cause mortality. Vital status was determined by linking electronic health records with the U.S. Social Security Death Index, a method that accurately captures both in-hospital and out-of-hospital deaths.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBaseline characteristics were presented as mean (SD) or median (IQR) for continuous variables and as frequency (percentage) for categorical variables, with comparisons across glycemic variability (GV) tertiles (T1-T3) performed using appropriate statistical tests. Multivariable Cox proportional hazards regression models were employed to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between GV tertiles and mortality. Three sequential models were constructed: Model 1 (unadjusted), Model 2 (adjusted for age, gender, race), and Model 3 (fully adjusted for baseline demographics, comorbidities, severity of illness scores, vital signs, laboratory values, and interventions). The dose-response relationship was visualized using restricted cubic splines with three knots, and survival probabilities were illustrated with Kaplan-Meier curves (log-rank test). Subgroup analyses were conducted across strata of age, BMI, SOFA score, diabetes, sepsis, and CKM stages, with interaction terms tested. A mediation analysis was performed to assess the role of serum lactate. To handle missing data, we performed multiple imputation using the chained equations (MICE) package on all covariates with less than 40% missing values; variables with missingness above this threshold were excluded. To ensure the robustness of our findings, sensitivity analyses were performed by extending the follow-up period to 90 and 365 days, and by repeating the primary analysis on a dataset imputed using the random forest method. All analyses were performed using R software (version 4.4.2), and a two-sided P-value \u0026lt; 0.05 was considered significant.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eBaseline characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe baseline characteristics of the 46,958 included patients are detailed in \u003cstrong\u003eTable 1\u003c/strong\u003e. Patients in the highest GV tertile (T3) presented with a significantly greater burden of critical illness compared to those in the lowest tertile (T1). This was evidenced by higher severity scores (median SOFA: 5.00 vs. 4.00; median OASIS: 33.00 vs. 30.00), a greater prevalence of comorbidities like diabetes (47.89% vs. 20.46%) and sepsis (65.12% vs. 47.98%), and consequently, a greater need for intensive organ support, including mechanical ventilation (50.17% vs. 39.64%) and vasopressors (42.74% vs. 29.62%; all P \u0026lt; 0.001). Crucially, this gradient of severity directly translated to worse unadjusted outcomes: 28-day mortality in the T3 group was nearly double that of the T1 group (20.48% vs. 11.38%; P \u0026lt; 0.001), a stark trend that persisted for 180-day mortality (33.02% vs. 19.79%; P \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurvival analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe multivariable Cox regression analyses revealed a significant, dose-dependent association between GV and mortality (Table 2). While this association was pronounced in the unadjusted model, it\u0026nbsp;remained robust after comprehensive adjustment\u0026nbsp;(Model 3). In this final model, patients in the highest GV tertile (T3) faced a significantly increased risk of both 28-day (aHR, 1.23; 95% CI, 1.15\u0026ndash;1.31; P \u0026lt; 0.001) and 180-day mortality (aHR, 1.31; 95% CI, 1.24\u0026ndash;1.37; P \u0026lt; 0.001) compared to the lowest tertile (T1). Although attenuated, a significantly increased risk for 180-day mortality was also evident for the medium tertile (T2) (aHR, 1.08; 95% CI, 1.03\u0026ndash;1.13; P = 0.002).\u0026nbsp;These findings establish high GV as a potent, independent predictor of both short- and long-term mortality in this high-risk population.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAssociation with outcomes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eVisual inspection of the dose-response relationship confirmed a continuous, positive association between GV and mortality risk (Figure 2A-B).\u0026nbsp;Restricted cubic spline analysis revealed a near-linear relationship, with the adjusted hazard for both 28-day and 180-day mortality progressively increasing with higher GV (both P for nonlinearity \u0026gt; 0.05). This dose-dependent effect was further corroborated by the Kaplan-Meier survival analysis\u0026nbsp;(Figure 2C-D). The survival curves for the three GV tertiles separated early and continued to diverge throughout the follow-up period, with the highest GV tertile (T3) exhibiting significantly lower survival probability compared to the other groups (P for log-rank \u0026lt; 0.01 for overall comparison).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSubgroup analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe performed prespecified subgroup analyses to test the consistency of the association between high GV (T3 vs. T1) and mortality (Figure 3). The adverse prognostic effect of high GV was\u0026nbsp;broadly consistent\u0026nbsp;across most patient strata, including age, sepsis, and CKM stage (all P for interaction \u0026gt; 0.05).\u003c/p\u003e\n\u003cp\u003eHowever, we identified significant effect modification by both diabetes status and baseline illness severity.\u0026nbsp;The detrimental impact of high GV was substantially more pronounced in patients\u0026nbsp;without\u0026nbsp;pre-existing diabetes (aHR for 28-day mortality: 1.42 [95% CI, 1.31\u0026ndash;1.53]; P for interaction = 0.001) and in those with a\u0026nbsp;lower\u0026nbsp;initial severity of illness (SOFA score \u0026lt; 5; aHR: 1.32 [95% CI, 1.18\u0026ndash;1.47]; P for interaction = 0.002). A similar pattern of interaction was observed for 180-day mortality. While a weaker interaction was noted for BMI with 28-day mortality (P for interaction = 0.026), this was not significant for the 180-day outcome and may represent a chance finding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMediation analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo formally test our hypothesis, a statistical mediation analysis was conducted to quantify the mediating role of serum lactate in the association between GV and mortality (Figure 4).\u003c/p\u003e\n\u003cp\u003eFor the primary endpoint of 28-day mortality, serum lactate was a statistically significant mediator across all three adjustment models. In the fully adjusted model (Model 3), the proportion of the total effect mediated by lactate was 10.54% (P \u0026lt; 0.001). A similar significant mediating effect was observed for 180-day mortality, with a mediated proportion of 5.33% in the final model (P \u0026lt; 0.001). For both 28-day and 180-day outcomes, both the average causal mediation effect (ACME, the indirect path through lactate) and the average direct effect (ADE, the direct path of GV) were statistically significant (all P \u0026lt; 0.001).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSensitivity Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe robustness of our findings was confirmed through two sensitivity analyses. Repeating the primary analysis with an alternative random forest-based imputation method yielded highly consistent results (Supplementary Table S2). Furthermore, the prognostic association of high GV (T3 vs. T1) proved durable, remaining significantly associated with mortality in the fully adjusted model at both 90 days (aHR, 1.32; 95% CI, 1.25\u0026ndash;1.39) and 365 days (aHR, 1.29; 95% CI, 1.24\u0026ndash;1.35) (Supplementary Table S3). Collectively, these results establish that our primary findings are robust to the choice of imputation method and are sustained over long-term follow-up.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this large-scale retrospective cohort study of critically ill patients with Cardiovascular-Kidney-Metabolic (CKM) syndrome, we have observed for the first time that high glycemic variability (GV) is a potent and independent predictor of both short- and long-term mortality. This association persisted even after extensive adjustment for a wide range of demographic, clinical, and severity-of-illness confounders. Notably, our analysis uncovered two critical nuances: the mortality risk associated with high GV was significantly more pronounced in patients without a prior diagnosis of diabetes and in those with a lower initial severity of illness. Furthermore, we identified serum lactate as a significant, albeit partial, mediator in the pathway linking GV to adverse outcomes, providing a potential mechanistic link for this association.\u003cbr\u003eOur primary finding\u0026mdash;that high GV is associated with increased mortality\u0026mdash;is broadly consistent with a substantial body of literature in general and mixed ICU populations [8,14,15]. Previous studies have established GV as a superior prognostic marker compared to mean glucose levels in diverse cohorts, including patients with sepsis, cerebrovascular disease, and post-cardiac surgery [16\u0026ndash;18]. Our results reinforce this paradigm, extending its validity to the unique and extremely vulnerable population of critically ill patients with CKM syndrome. However, the novelty of our study lies in its specific focus on this newly defined syndrome. The CKM framework conceptualizes a progressive state of depleted physiological reserves across interconnected organ systems\u0026nbsp;[1,2]. Patients with CKM syndrome are characterized by chronic inflammation, endothelial dysfunction, and metabolic inflexibility, creating a physiological milieu that may amplify the deleterious effects of acute metabolic instability reflected by high GV\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e[2,19,20].\u003cbr\u003e\u0026nbsp;Perhaps the most striking finding of our study is the significant effect modification by diabetes status, where the prognostic harm of high GV was substantially greater in patients without pre-existing diabetes. This seemingly paradoxical result is not an isolated observation but aligns with a growing body of evidence suggesting that stress-induced hyperglycemia carries a graver prognosis in individuals without a history of diabetes across various critical illnesses [21,22]. Two primary hypotheses may explain this phenomenon. First, the concept of cellular \u0026quot;preconditioning\u0026quot; posits that chronic exposure to hyperglycemia in diabetic patients may induce adaptive mechanisms, such as an upregulation of endogenous antioxidant defenses, making them relatively more resilient to the acute metabolic insults posed by high GV [23]. In contrast, the glucose-na\u0026iuml;ve tissues of non-diabetic patients are more susceptible to the glucotoxicity from acute glycemic excursions, which are known to trigger more severe oxidative stress and endothelial dysfunction than sustained hyperglycemia [24]. Second, the etiology of high GV likely differs between the two groups. In non-diabetic patients, high GV serves as a more direct and unadulterated marker of an extreme endogenous stress response, reflecting a profound dysregulation of counter-regulatory hormones [25]. In diabetic patients, however, GV represents a complex interplay between this endogenous stress and exogenous factors like insulin administration, potentially confounding its interpretation as a pure indicator of acute physiological stress and thus diluting its prognostic strength.\u003cbr\u003e\u0026nbsp;Another key and seemingly paradoxical finding was that the prognostic association between high GV and mortality was strongest in patients with a lower initial burden of organ dysfunction (SOFA score \u0026lt; 5). This suggests that in patients with overwhelming multi-organ failure (high SOFA score), their trajectory towards mortality is likely governed by a confluence of established and often irreversible pathophysiological processes. In this context, the independent prognostic contribution of any single variable like GV is diminished, a phenomenon that can be understood in prognostic modeling where the \u0026quot;signal\u0026quot; of one biomarker is obscured by the overwhelming \u0026quot;noise\u0026quot; of global illness severity [26]. In stark contrast, for patients who are less critically ill at baseline, this \u0026quot;lower-noise\u0026quot; environment allows acute metabolic dysregulation, as captured by high GV, to emerge as a more significant prognostic factor. Here, it may function as a critical \u0026quot;second hit\u0026quot; that potentially precipitates a cascade of clinical deterioration and subsequent organ injury leading to death [27]. Therefore, our finding highlights the potential role of GV not merely as a prognostic marker, but as a crucial and early indicator of impending physiological collapse, particularly in patients who might otherwise be stratified into a lower-risk category.\u003c/p\u003e\n\u003cp\u003eOur mediation analysis provides novel, quantitative evidence for a specific biological pathway linking high GV to mortality, finding that serum lactate partially mediates this relationship. This is highly biologically plausible, as high GV is known to induce excessive reactive oxygen species (ROS) production, leading to profound mitochondrial dysfunction\u0026nbsp;[9,10]. This impaired mitochondrial respiration can force cells into anaerobic glycolysis, even in the presence of adequate oxygen\u0026mdash;a state of cytopathic hypoxia or Type B lactic acidosis\u0026mdash;resulting in lactate accumulation\u0026nbsp;[28]. Thus, in this context, serum lactate transcends its role as a simple marker of tissue hypoperfusion and becomes a direct indicator of cellular metabolic distress, a phenomenon our study now quantitatively links to GV. Importantly, the mediation was only partial (~10%), indicating that the majority of the association between GV and mortality is likely explained by other pathways. These likely include an exacerbated systemic inflammatory response, as hyperglycemia is known to acutely increase pro-inflammatory cytokines\u0026nbsp;[29], and heightened endothelial injury, as fluctuating glucose levels are particularly damaging to the vascular endothelium\u0026nbsp;[10].\u003c/p\u003e\n\u003cp\u003eIn summary, our findings carry significant clinical implications, advocating for a shift in focus from merely avoiding hyperglycemia\u0026mdash;a lesson from the landmark NICE-SUGAR trial [30]\u0026mdash;towards achieving glycemic stability [5]. GV should be considered a critical \u0026apos;metabolic vital sign,\u0026apos; particularly for risk stratification in non-diabetic and less severely ill CKM patients [5].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWhile our large-scale analysis provides a robust foundation for this view, several limitations must be acknowledged. First, given the retrospective observational design, we can only report associations; causal relationships between GV and mortality cannot be definitively established, and the possibility of residual confounding remains despite extensive adjustment. Second, the reliance on intermittent glucose measurements derived from electronic health records may not capture all glycemic excursions as accurately as continuous glucose monitoring (CGM). Third, this is a single-center study based on the MIMIC-IV database; thus, selection bias is inevitable, and our findings require external validation in multicenter cohorts to ensure generalizability to other populations or healthcare systems.\u003c/p\u003e\n\u003cp\u003eThese limitations, however, directly inform a clear path for future research. Prospective, multicenter studies are warranted to confirm our findings and, ultimately, to test whether advanced strategies specifically aimed at minimizing GV\u0026mdash;guided by technologies like CGM and closed-loop insulin delivery systems [31]\u0026mdash;can translate into improved survival for this exceptionally high-risk population.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, our study Identifies high glycemic variability as a significant, independent risk factor for both short- and long-term mortality in the critically ill Cardiovascular-Kidney-Metabolic (CKM) syndrome population. The prognostic harm of high GV was particularly pronounced in patients without pre-existing diabetes and those with a lower initial severity of illness, identifying key subgroups for heightened clinical vigilance. These findings underscore GV as a crucial marker of profound metabolic decompensation and position glycemic stability, rather than mere glycemic level, as a key therapeutic target for future investigation in this exceptionally vulnerable population.\u003c/p\u003e\n"},{"header":"Abbreviations","content":"\u003cp\u003e\u003cstrong\u003eACME\u003c/strong\u003e: Average Causal Mediation Effect\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eADE\u003c/strong\u003e: Average Direct Effect\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAHA\u003c/strong\u003e: American Heart Association\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eaHR\u003c/strong\u003e: Adjusted Hazard Ratio\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBMI\u003c/strong\u003e: Body Mass Index\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCGM\u003c/strong\u003e: Continuous Glucose Monitoring\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCI\u003c/strong\u003e: Confidence Interval\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCKD\u003c/strong\u003e: Chronic Kidney Disease\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCKM\u003c/strong\u003e: Cardiovascular-Kidney-Metabolic\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCOPD\u003c/strong\u003e: Chronic Obstructive Pulmonary Disease\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCV\u003c/strong\u003e: Coefficient of Variation\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCVD\u003c/strong\u003e: Cardiovascular Disease\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGCS\u003c/strong\u003e: Glasgow Coma Scale\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGV\u003c/strong\u003e: Glycemic Variability\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHF\u003c/strong\u003e: Heart Failure\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHR\u003c/strong\u003e: Hazard Ratio\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eICU\u003c/strong\u003e: Intensive Care Unit\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMICE\u003c/strong\u003e: Multiple Imputation using Chained Equations\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMIMIC-IV\u003c/strong\u003e: Medical Information Mart for Intensive Care IV\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eOASIS\u003c/strong\u003e: Oxford Acute Severity of Illness Score\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eROS\u003c/strong\u003e: Reactive Oxygen Species\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSD\u003c/strong\u003e: Standard Deviation\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSOFA\u003c/strong\u003e: Sequential Organ Failure Assessment\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSQL\u003c/strong\u003e: Structured Query Language\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was conducted in accordance with the Declaration of Helsinki. Ethical approval and individual patient consent were waived for the MIMIC-IV database because it contains de-identified health information that is publicly available for research purposes. The use of the MIMIC-IV database for this study was specifically approved by the Massachusetts Institute of Technology Institutional Review Board and Beth Israel Deaconess Medical Center. The authorized researcher (certification number 13024213) completed the required data user training, which granted access approval for the database.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe dataset utilized in this study was obtained from a large-scale, publicly available critical care database: the Medical Information Mart for Intensive Care IV (MIMIC-IV, version 3.1). Data access was granted and data were extracted by an authorized researcher (Kangxing Wang, certification number: 13024213) after obtaining necessary approvals and completing ethical training. The dataset is publicly accessible for research purposes via PhysioNet.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing financial interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFundings:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was approved by National Key R\u0026amp;D Program of China（2022YFC2504500）and 1.3.5 project for disciplines of excellence, West China Hospital, Sichuan University（ZYGD23012）\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWKX, XHY, ZYK, and YYJ contributed equally to this work as co-first authors. WKX was the primary contributor, responsible for the study conception, essential data acquisition (coding and extraction), performing the main statistical analysis, and drafting the manuscript. XHY refined the methodology, provided technical support for the analysis, and assisted with data management. ZYK assisted with the statistical modeling, interpretation of results, and critical revision of the manuscript. YYJ contributed to data validation, visualization of results, and substantial revision of the manuscript. KY and ZYF served as co-corresponding authors, secured funding, and supervised the study. Both KY and ZYF critically revised the manuscript for intellectual content. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eNdumele CE, Neeland IJ, Tuttle KR, Chow SL, Mathew RO, Khan SS, et al. A synopsis of the evidence for the science and clinical management of cardiovascular-kidney-metabolic (CKM) syndrome: a scientific statement from the american heart association. Circulation. United States; 2023;148:1636\u0026ndash;64. https://doi.org/10.1161/CIR.0000000000001186\u003c/li\u003e\n\u003cli\u003eNdumele CE, Rangaswami J, Chow SL, Neeland IJ, Tuttle KR, Khan SS, et al. Cardiovascular-kidney-metabolic health: a presidential advisory from the american heart association. Circulation. 2023;148:1606\u0026ndash;35. https://doi.org/10.1161/CIR.0000000000001184\u003c/li\u003e\n\u003cli\u003eAggarwal R, Ostrominski JW, Vaduganathan M. Prevalence of cardiovascular-kidney-metabolic syndrome stages in US adults, 2011-2020. JAMA. 2024;331:1858. https://doi.org/10.1001/jama.2024.6892\u003c/li\u003e\n\u003cli\u003eService FJ. Glucose variability. Diabetes. 2013;62:1398\u0026ndash;404. https://doi.org/10.2337/db12-1396\u003c/li\u003e\n\u003cli\u003eKrinsley JS. Glycemic variability: a strong independent predictor of mortality in critically ill patients. Crit Care Med. 2008;36:3008\u0026ndash;13. https://doi.org/10.1097/CCM.0b013e31818b38d2\u003c/li\u003e\n\u003cli\u003eBagshaw SM, Bellomo R, Jacka MJ, Egi M, Hart GK, George C, et al. The impact of early hypoglycemia and blood glucose variability on outcome in critical illness. Crit Care. 2009;13:R91. https://doi.org/10.1186/cc7921\u003c/li\u003e\n\u003cli\u003eSu Y, Fan W, Liu Y, Hong K. Glycemic variability and in-hospital death of critically ill patients and the role of ventricular arrhythmias. Cardiovasc Diabetol. England; 2023;22:134. https://doi.org/10.1186/s12933-023-01861-0\u003c/li\u003e\n\u003cli\u003eWang F, Guo Y, Tang Y, Zhao S, Xuan K, Mao Z, et al. Combined assessment of stress hyperglycemia ratio and glycemic variability to predict all-cause mortality in critically ill patients with atherosclerotic cardiovascular diseases across different glucose metabolic states: an observational cohort study with machine learning. Cardiovasc Diabetol. 2025;24:199. https://doi.org/10.1186/s12933-025-02762-0\u003c/li\u003e\n\u003cli\u003eCeriello A, Monnier L, Owens D. Glycaemic variability in diabetes: clinical and therapeutic implications. Lancet Diabetes Endocrinol. 2019;7:221\u0026ndash;30. https://doi.org/10.1016/S2213-8587(18)30136-0\u003c/li\u003e\n\u003cli\u003eCeriello A, Esposito K, Piconi L, Ihnat MA, Thorpe JE, Testa R, et al. Oscillating glucose is more deleterious to endothelial function and oxidative stress than mean glucose in normal and type 2 diabetic patients. Diabetes. 2008;57:1349\u0026ndash;54. https://doi.org/10.2337/db08-0063\u003c/li\u003e\n\u003cli\u003eJohnson AEW, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data. 2023;10:1. https://doi.org/10.1038/s41597-022-01899-x\u003c/li\u003e\n\u003cli\u003eChun K-H, Oh J, Lee CJ, Park JJ, Lee SE, Kim M-S, et al. In-hospital glycemic variability and all-cause mortality among patients hospitalized for acute heart failure. Cardiovasc Diabetol. 2022;21:291. https://doi.org/10.1186/s12933-022-01720-4\u003c/li\u003e\n\u003cli\u003eSinger M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA. 2016;315:801. https://doi.org/10.1001/jama.2016.0287\u003c/li\u003e\n\u003cli\u003eQi L, Geng X, Feng R, Wu S, Fu T, Li N, et al. Association of glycemic variability and prognosis in patients with traumatic brain injury: a retrospective study from the MIMIC-IV database. Diabetes Res Clin Pract. 2024;217:111869. https://doi.org/10.1016/j.diabres.2024.111869\u003c/li\u003e\n\u003cli\u003eHe H, Xie Y, Wang Z, Li J, Zheng S, Li X, et al. Associations of variability in blood glucose and systolic blood pressure with mortality in patients with coronary artery disease: a retrospective cohort study from the MIMIC-IV database. Diabetes Res Clin Pract. 2024;209:111595. https://doi.org/10.1016/j.diabres.2024.111595\u003c/li\u003e\n\u003cli\u003eZhou Z, Zhang H, Gu Y, Zhang K, Ouyang C. Relationship between glycemic variability and the incidence of postoperative atrial fibrillation following cardiac surgery: a retrospective study from MIMIC-IV database. Diabetes Res Clin Pract. Ireland; 2024;219:111978. https://doi.org/10.1016/j.diabres.2024.111978\u003c/li\u003e\n\u003cli\u003eAli NA, OʼBrien JM, Dungan K, Phillips G, Marsh CB, Lemeshow S, et al. Glucose variability and mortality in patients with sepsis. Crit Care Med. 2008;36:2316\u0026ndash;21. https://doi.org/10.1097/CCM.0b013e3181810378\u003c/li\u003e\n\u003cli\u003eCai W, Li Y, Guo K, Wu X, Chen C, Lin X. Association of glycemic variability with death and severe consciousness disturbance among critically ill patients with cerebrovascular disease: analysis of the MIMIC-IV database. Cardiovasc Diabetol. England; 2023;22:315. https://doi.org/10.1186/s12933-023-02048-3\u003c/li\u003e\n\u003cli\u003eMassy ZA, Drueke TB. Combination of cardiovascular, kidney, and metabolic diseases in a syndrome named cardiovascular-kidney-metabolic, with new risk prediction equations. Kidney Int Rep. 2024;9:2608\u0026ndash;18. https://doi.org/10.1016/j.ekir.2024.05.033\u003c/li\u003e\n\u003cli\u003eBaaten CCFMJ, Vondenhoff S, Noels H. Endothelial cell dysfunction and increased cardiovascular risk in patients with chronic kidney disease. Circ Res. 2023;132:970\u0026ndash;92. https://doi.org/10.1161/CIRCRESAHA.123.321752\u003c/li\u003e\n\u003cli\u003eKosiborod M, Rathore SS, Inzucchi SE, Masoudi FA, Wang Y, Havranek EP, et al. Admission glucose and mortality in elderly patients hospitalized with acute myocardial infarction: implications for patients with and without recognized diabetes. Circulation. 2005;111:3078\u0026ndash;86. https://doi.org/10.1161/CIRCULATIONAHA.104.517839\u003c/li\u003e\n\u003cli\u003eHe H-M, Wang Z, Xie Y-Y, Zheng S-W, Li J, Li X-X, et al. Maximum stress hyperglycemia ratio within the first 24 h of admission predicts mortality during and after the acute phase of acute coronary syndrome in patients with and without diabetes: a retrospective cohort study from the MIMIC-IV database. Diabetes Res Clin Pract. Ireland; 2024;208:111122. https://doi.org/10.1016/j.diabres.2024.111122\u003c/li\u003e\n\u003cli\u003eCeriello A. High glucose induces antioxidant enzymes in human endothelial cells in culture. \u003c/li\u003e\n\u003cli\u003eMonnier L, Mas E, Ginet C, Michel F, Villon L, Cristol J-P, et al. Activation of oxidative stress by acute glucose fluctuations compared with sustained chronic hyperglycemia in patients with type 2 diabetes. \u003c/li\u003e\n\u003cli\u003eDungan KM, Braithwaite SS, Preiser J-C. Stress hyperglycaemia. \u003c/li\u003e\n\u003cli\u003eMoore FA, Moore EE. Evolving concepts in the pathogenesis of postinjury multiple organ failure. Surg Clin North Am. 1995;75:257\u0026ndash;77. https://doi.org/10.1016/S0039-6109(16)46587-4\u003c/li\u003e\n\u003cli\u003eAngele MK, Chaudry IH. Surgical trauma and immunosuppression: pathophysiology and potential immunomodulatory approaches. Langenbecks Arch Surg. 2005;390:333\u0026ndash;41. https://doi.org/10.1007/s00423-005-0557-4\u003c/li\u003e\n\u003cli\u003eNedel W, Deutschendorf C, Portela LVC. Sepsis-induced mitochondrial dysfunction: a narrative review. World J Crit Care Med. 2023;12:139\u0026ndash;52. https://doi.org/10.5492/wjccm.v12.i3.139\u003c/li\u003e\n\u003cli\u003eEsposito K, Nappo F, Marfella R, Giugliano G, Giugliano F, Ciotola M, et al. Inflammatory cytokine concentrations are acutely increased by hyperglycemia in humans: role of oxidative stress. Circulation. 2002;106:2067\u0026ndash;72. https://doi.org/10.1161/01.CIR.0000034509.14906.AE\u003c/li\u003e\n\u003cli\u003eIntensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360:1283\u0026ndash;97. https://doi.org/10.1056/NEJMoa0810625\u003c/li\u003e\n\u003cli\u003eNimri R, Phillip M, Clements MA, Kovatchev B. Closed-loop control, artificial intelligence\u0026ndash;based decision-support systems, and data science. Diabetes Technol Ther. 2024;26:S-68-S-89. https://doi.org/10.1089/dia.2024.2505\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables are available in the Supplementary Files section.\u003c/p\u003e\n"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Glycemic Variability, Cardiovascular-Kidney-Metabolic Syndrome, Lactate, MIMIC-IV","lastPublishedDoi":"10.21203/rs.3.rs-8337456/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8337456/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eThe newly defined Cardiovascular-Kidney-Metabolic (CKM) syndrome represents a complex phenotype characterized by extreme physiological vulnerability. While Glycemic Variability (GV) is a recognized stressor in critical illness, its specific prognostic significance within this multimorbid CKM population remains uncharacterized. We aimed to determine the independent association of GV with mortality in CKM patients and to investigate the mediating role of serum lactate.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eWe conducted a large-scale retrospective cohort study involving 46,958 critically ill adults with CKM syndrome (stages 1-4) using the MIMIC-IV database. GV was quantified as the coefficient of variation (CV) of all glucose measurements during the ICU stay. The primary outcomes were 28-day and 180-day all-cause mortality, analyzed using multivariable Cox proportional hazards models. Mediation analysis was employed to quantify the proportion of the association statistically attributable to serum lactate levels.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003eHigh GV was strongly associated with increased mortality risk. In fully adjusted models, patients in the highest GV tertile faced a significantly higher risk of 28-day mortality (Adjusted Hazard Ratio [aHR], 1.23; 95% CI, 1.15–1.31) and 180-day mortality (aHR, 1.31; 95% CI, 1.24–1.37) compared to the lowest tertile (both P \u0026lt; 0.001). Mediation analysis suggested that serum lactate statistically mediated this association, accounting for 10.5% of the relationship with 28-day mortality (P \u0026lt; 0.001). Notably, interaction analyses demonstrated that the adverse association of high GV with mortality was significantly more pronounced in non-diabetic patients (P-interaction = 0.001) and in those with lower baseline illness severity (SOFA score \u0026lt; 5, P-interaction = 0.002).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eIn critically ill patients with CKM syndrome, elevated GV is a significant, independent predictor of both short- and long-term mortality. Analysis suggests this relationship may be partially mediated by lactate-associated metabolic stress. Our findings highlight GV as a crucial prognostic marker, particularly in non-diabetic CKM patients and those presenting with seemingly lower disease severity, suggesting a potential benefit of stricter glycemic stewardship in these subgroups.\u003c/p\u003e","manuscriptTitle":"Association between Glycemic Variability and Short- and Long-Term Mortality in Critically Ill Patients with Cardiovascular-Kidney-Metabolic Syndrome: A Cohort Study from the MIMIC-IV Database","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-28 05:40:09","doi":"10.21203/rs.3.rs-8337456/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-04-20T02:49:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-20T02:48:17+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-12-17T14:22:00+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-16T18:15:47+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-12-16T13:35:54+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f6b59d3f-7591-4c03-8cb3-107ceff11720","owner":[],"postedDate":"April 28th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":66865873,"name":"Health sciences/Biomarkers"},{"id":66865874,"name":"Health sciences/Diseases"},{"id":66865875,"name":"Health sciences/Endocrinology"},{"id":66865876,"name":"Health sciences/Medical research"},{"id":66865877,"name":"Health sciences/Nephrology"},{"id":66865878,"name":"Health sciences/Risk factors"}],"tags":[],"updatedAt":"2026-04-28T05:40:10+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-28 05:40:09","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8337456","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8337456","identity":"rs-8337456","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}