Association between the dead space fraction and mortality in critically ill patients with acute respiratory failure: an analysis of the MIMIC-IV database

Association between the dead space fraction and mortality in critically ill patients with acute respiratory failure: an analysis of 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 Research Article Association between the dead space fraction and mortality in critically ill patients with acute respiratory failure: an analysis of the MIMIC-IV database feifei wang, Xingping Lv, Wei xu, yunhe Zhang, chun Wang, shaolin ma, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7972231/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background : Few studies have investigated the association between V D /V T and mortality in critically ill patients with acute respiratory failure (ARF). Objective : The primary objective of this study is to clarify the association between V D /V T and mortality in critically ill patients with ARF, as well as to determine its optimal predictive threshold for clinical application. Methods : This study was based on the MIMIC-IV database. The analysis using multivariable logistic regression, cox proportional hazards model, and restricted cubic splines (RCS) was used to explore the association between V D /V T and mortality. Results : A total of 2,254 patients were included in the study, of whom 1,709 (75.82%) died during their ICU stay. Both multivariable logistic regression and cox proportional hazards model identified V D /V T as the most significant predictor of ICU mortality (OR = 16.90, 95%CI: 7.38-38.69, P˂0.001; HR=9.60, 95%CI: 4.95-18.61, P˂0.001). After adjusting for covariates, the RCS analysis continued to demonstrate a significant overall association and a non-linear relationship between V D /V T and ICU mortality, with an identified inflection point at 0.222 for V D /V T . In addition, the results of the subgroup and stratified analyses were robust, and the univariate forest plot demonstrated that V D /V T exhibited an extremely strong association with ICU mortality, accompanied by an overall OR of 2.34 (95% CI: 1.91-2.86, p˂0.001). Conclusion : V D /V T is positively associated with ICU mortality in critically ill patients with ARF. And the V D /V T of 0.222 may serve as a potential prognostic threshold for predicting clinical outcomes in critically ill patients with ARF. dead space fraction acute respiratory failure mortality MIMIC-IV Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Acute respiratory failure (ARF) stands as a predominant trigger for critical care consultations, representing a cardinal manifestation of numerous cardiopulmonary disorders and carrying a substantial burden of morbidity and mortality globally [ 1 , 2 ]. A pivotal pathological feature underlying ARF is the augmentation of alveolar dead space—defined as alveolar regions receiving ventilation without corresponding perfusion—which directly compromises effective gas exchange. The dead space fraction (V D /V T ), calculated as the ratio of dead space volume to tidal volume, serves as a central metric for quantifying dead space ventilation. Beyond merely reflecting the volume of non-functional alveoli, V D /V T integrates multiple dimensions of lung dysfunction in acute hypoxemic respiratory failure, including the magnitude of intrapulmonary shunting and the extent of pulmonary vascular injury. Clinically, V D /V T is commonly derived at the bedside using arterial partial pressure of carbon dioxide (PaCO₂) and end-tidal carbon dioxide pressure (PETCO₂), via the formula: V D /V T = (PaCO₂ – PETCO₂)/PaCO₂. In healthy adults, V D /V T typically remains below 0.30; however, this threshold is markedly exceeded in patients with acute respiratory distress syndrome (ARDS), with values surpassing 0.55–0.60 consistently linked to heightened mortality in this subset [ 3 ]. Prior investigations have underscored the clinical relevance of V D /V T : Sivan et al. observed significant discrepancies between PETCO₂ and PaCO₂ when the arterial-to-alveolar oxygen ratio dropped below 0.3, highlighting the interplay between increased alveolar dead space and impaired oxygenation in severe lung disease [ 4 ]. Additionally, V D /V T has been shown to independently predict mortality in pediatric acute hypoxemic respiratory failure, even after accounting for illness severity [ 5 ]. In ARDS specifically, V D /V T exhibits a robust association with mortality across both early and intermediate disease phases—each 0.05 increment in VD/VT correlates with a 59% increase in the odds of death in the early phase and a 186% increase in the odds in the intermediate phase [ 6 ]. Despite these insights, the existing literature predominantly focuses on ARDS—the most severe phenotype of ARF—with limited attention directed toward characterizing V D /V T and its prognostic value in the broader population of critically ill patients with ARF. This gap is notable, as ARF encompasses a heterogeneous spectrum of conditions beyond ARDS, and understanding V D /V T role in this larger cohort could enhance risk stratification and clinical decision-making. To address this limitation, the present study aimed to: (1) investigate the distribution and clinical correlates of V D /V T in a diverse cohort of critically ill patients with ARF; (2) clarify the independent association between V D /V T and ICU mortality in this population; and (3) identify an optimal V D /V T threshold that may serve as a practical prognostic tool for clinical application. By leveraging data from the MIMIC-IV database, this study seeks to expand the evidence base for V D /V T as a prognostic marker in ARF, ultimately supporting more targeted and effective management of critically ill patients. Methods Data Source The core analysis of this study was based on the Medical Information Mart for Intensive Care IV (MIMIC-IV) database (Version 2.2). This database contains comprehensive clinical data of patients admitted to the Intensive Care Unit (ICU) of Beth Israel Deaconess Medical Center in Boston from 2008 to 2019 [7,8], serving as a rich and reliable data source for medical research. The use of the database has been approved by relevant institutions and complies with research data usage regulations. Qualified researchers can obtain the data free of charge for research purposes after completing the corresponding data use agreement. Patient Selection Adult patients (aged ≥ 18 years) diagnosed with ARF via ICD-9 and ICD-10 codes were identified from the database. The inclusion criterion was that patients had available PaCO₂ and PETCO₂ data within 24 hours of ICU admission. The exclusion criteria included: patients with an ICU length of stay of less than 24 hours, and patients with a calculated V D /V T equal to 1. These criteria were set to ensure the validity of the study sample and the reliability of the data. Data Extraction Key data were extracted from the clinical records of patients within 24 hours of ICU admission, including the following categories:Demographic characteristics: Age, gender; Vital signs: Mean arterial pressure (MAP), respiratory rate (RR), percutaneous arterial oxygen saturation (SPO₂); Laboratory test indicators: White blood cell count (WBC), platelet count, creatinine, alanine aminotransferase (ALT), aspartate aminotransferase (AST), albumin, glucose; Comorbidity status: Hypertension, diabetes mellitus, myocardial infarction, heart failure, chronic kidney disease, acute kidney injury, malignant tumor, pneumonia, stroke, chronic obstructive pulmonary disease (COPD); Disease severity scores: Sequential Organ Failure Assessment (SOFA) score, Acute Physiology and Chronic Health Evaluation Ⅲ (APACHE Ⅲ) score, Simplified Acute Physiology Score Ⅱ (SAPS Ⅱ) score; Arterial blood gas analysis parameters: pH, PaO₂, PaCO₂, lactate; Other relevant data: ICU length of stay, PETCO₂. The primary outcome indicator of this study was ICU mortality, defined as death from any cause during the patient's ICU stay. V D /V T was calculated using the simplified Enghoff modification of the Bohr equation [9], with the specific formula: V D /V T = (PaCO₂ - PETCO₂)/PaCO₂. Statistical Analysis All statistical analyses were performed using R software (Version 4.5.1), with the significance level set at α = 0.05. Descriptive statistics: Continuous variables, which typically follow a non-normal distribution in clinical data, were expressed as median (interquartile range, IQR), and inter-group comparisons were conducted using the Mann-Whitney U test; Categorical variables were presented as frequency (percentage), and inter-group comparisons were performed using the chi-square test or Fisher's exact test (when applicable). Correlation analysis: Univariate and multivariate logistic regression analyses were used to screen for factors independently associated with ICU mortality, and the results were presented as adjusted odds ratios (OR) with their 95% confidence intervals (CI); Cox proportional hazards regression models were employed to evaluate the association between V D /V T groups and ICU mortality outcomes, and the results were expressed as hazard ratios (HR) with their 95% CI. Non-linear relationship analysis: Restricted cubic spline (RCS) analysis was performed based on the Cox proportional hazards model to assess the non-linear relationship between V D /V T and ICU mortality. The RCS model was adjusted for the aforementioned potential confounding factors, and the optimal cut-off value of V D /V T was determined through this analysis. Survival analysis: Patients were divided into four groups according to the quartiles of V D /V T , and the Kruskal-Wallis test was used to compare differences in ICU mortality among the four groups; Patients were further divided into high and low V D /V T groups based on the cut-off value determined by RCS analysis. Kaplan-Meier (KM) curves were plotted to visualize survival outcomes, and the log-rank test was used to compare survival differences between the two groups. Subgroup analysis: A forest plot was generated to present the results of univariate logistic regression analysis regarding the association between V D /V T and ICU mortality in key patient subgroups. The OR value and its 95% CI for each subgroup were displayed to verify the robustness of the association. Results Participants Characteristics This study enrolled 2,254 patients in total. Of the study cohort, 1,709 patients (75.82%) had an outcome of death during ICU hospitalization, while 545 patients (24.18%) survived. Significant differences were observed between the two groups in the following variables: WBC, albumin, glucose, pH, PaO 2 , lactate, ALT, AST, creatinine, RR, SpO₂, hospital length of stay, ICU length of stay, scoring systems, PETCO₂, V D /V T , hypertension, myocardial infarction, chronic kidney disease, mechanical ventilation, and acute kidney injury. No statistically significant differences were found between the two groups with respect to age, PaCO 2 , or other variables not specified above. Table 1 presents the baseline characteristics of all participants. Logistic regression and Cox proportional hazards models analysis Univariable logistic regression analysis indicated that V D /V T was the most significant predictor of ICU mortality (OR=23.78, 95%CI:13.09-43.20, Table2 ). Among comorbidities, underlying conditions such as kidney disease, and myocardial infarction were identified as risk factors for ICU mortality. Elevated laboratory values including WBC, glucose, lactate, ALT, AST, and creatinine were associated with increased risk of ICU mortality. After adjusting for potential confounders using multivariable logistic regression, V D /V T , glucose, and lactate remained independent risk factors for ICU mortality. Univariate Cox proportional hazards model analysis also revealed that multiple factors were associated with the risk of ICU mortality ( Table 3 ). In terms of underlying diseases, conditions such as myocardial infarction, malignant tumors, and kidney disease were all found to increase the risk of ICU mortality. Additionally, elevated levels of age, WBC, glucose, lactate, AST, ALT, creatinine, and V D /V T were associated with an increased risk of ICU mortality. After adjusting for confounding variables in the multivariate Cox proportional hazards model, V D /V T , malignant tumors, age, WBC, glucose, and lactate were identified as independent factors influencing ICU mortality, among which V D /V T was the most significant predictor of ICU mortality (HR=9.60, 95% CI: 4.95-18.61, p˂0.001). Results from both logistic regression and cox proportional hazards model analyses indicated that V D /V T was strong risk factors for ICU mortality, and both remained significantly independently associated with ICU mortality in multivariate analyses. In contrast, gender showed no significant association with ICU mortality in either model. Analysis of V D /V T Values by Quartile Groups Patients were stratified into quartiles based on V D /V T values, with Group 1 to Group 4 comprising 563 (24.98%), 551 (24.45%), 576 (25.55%), and 564 (25.02%) patients, respectively. Significant differences in multiple clinical characteristics and outcomes were observed across the four groups ( Table 4 ). ICU mortality increased significantly with higher V D /V T quartiles. Trends in laboratory and physiological parameters showed that glucose, lactate, PaCO 2 , ALT, AST, and creatinine increased with rising V D /V T , whereas pH, PaO 2 , and PETCO₂ decreased. Analysis of disease revealed that the incidence of pneumonia and COPD, as well as severity scores (SOFA, APACHE III, SAPS II), were significantly higher in groups with elevated V D /V T . No significant differences were observed across quartiles in age, sex, hypertension, diabetes mellitus, or myocardial infarction. Additionally, Kaplan-Meier survival curve analysis showed significant differences in survival rates among the four groups ( Figure 1 ). Group 1 (with the lowest V D /V T ) consistently had the highest survival rate, and the rate decreased most slowly over time; in contrast, Group 4 (with the highest V D /V T ) had the lowest survival rate, which declined most rapidly as time progressed. Regarding changes in the number of at-risk patients, the number of at-risk individuals in each group gradually decreased with prolonged follow-up, but the decrease was more pronounced in Group 4, consistent with the faster decline in survival rate of Group 4 observed in the survival curve. In summary, patients with higher V D /V T had significantly poorer survival outcomes than those with lower values. Restricted Cubic Splines Analysis Univariable RCS analysis revealed a non-linear association between V D /V T and ICU mortality ( Figure 2 ). The OR exhibited a decreasing trend as V D /V T increased from 0.0 to 0.2. However, beyond the inflection point (0.222), the OR demonstrated a pronounced upward trend with further increases in V D /V T , indicating a significantly elevated risk of ICU mortality at higher V D /V T . These findings suggest that the relationship between V D /V T and ICU mortality is not purely linear but exhibits a distinct non-linear pattern. After adjusting for covariates including sex, age, hypertension, diabetes mellitus, pneumonia, COPD, WBC, and platelets, the RCS analysis continued to demonstrate a significant overall association and a non-linear relationship between V D /V T and ICU mortality ( Figure 3 ). The persistence of this non-linear association after multivariable adjustment further supports that V D /V T is an important independent factor influencing ICU mortality risk. Patients were divided into two groups based on the RCS threshold of V D /V T (0.222): Group 1 (low V D /V T ) and Group 2 (high V D /V T ). Further analysis revealed significant statistical differences in multiple clinical indicators and outcome variables between the two groups ( Table 5 ). In terms of laboratory and physiological parameters, Group 2 showed significantly higher values in glucose, lactate, PaCO 2 , ALT, AST, and creatinine. With regard to disease severity and outcomes, Group 2 had significantly higher scores on the SOFA, APACHE III, and SAPS II, as well as significantly higher incidence rates of pneumonia and COPD, and a markedly elevated ICU mortality. These findings demonstrate that patients with higher V D /V T exhibit more severe illness, worse clinical indicators, and a significantly increased risk of ICU mortality, further validating the close association between elevated V D /V T and adverse outcomes in critically ill patients with ARF. Additionally, Kaplan-Meier survival analysis based on the RCS threshold of V D /V T (0.222) revealed a significant difference in survival between the two groups ( Figure 4 ). The high V D /V T group exhibited a steeper decline in survival over time and demonstrated a significantly higher risk of mortality (HR = 1.984, 95% CI: 1.661–2.369) compared to the low V D /V T group. Furthermore, although the number at risk decreased over time in both groups, high V D /V T group consistently showed lower survival rates than low V D /V T group, and this difference became more pronounced during follow-up. Subgroup Analysis Subgroup analysis results from the univariate forest plot examining the relationship between V D /V T and ICU mortality showed the following: among the 2,254 included patients, V D /V T exhibited an extremely strong statistical association with ICU mortality, with an overall OR of 2.34 (95% CI: 1.91–2.86, p˂0.001, Figure 5 ). After stratification by factors such as gender, hypertension, diabetes mellitus, heart failure, and chronic kidney disease, V D /V T was significantly associated with ICU mortality risk in all subgroups. Furthermore, there was no statistically significant interaction effect between the subgroups. These findings indicate that the positive association between V D /V T and ICU mortality is consistently present across populations with different characteristics. Discussion The current study showed that high V D /V T were associated with increased ICU mortality risk in critically ill patients with ARF who are admitted to the ICU, as per the MIMIC-IV database. The multivariate analysis of logistic regression and cox proportional hazards models revealed that high V D /V T was independently associated with ICU mortality in these patients. Furthermore, following multivariable adjustment, results from the RCS analysis remained robust: they not only confirmed a significant overall association between V D /V T and ICU mortality but also identified a non-linear pattern underlying this relationship. A critical finding was a non-linear cut-point of 0.222 for the V D /V T and mortality relationship identified by RCS analysis, with subgroup analyses based on this grouping robustly corroborated the aforementioned results. Finally, the subgroup forest plot demonstrated an overall OR of 2.34 (95% CI: 1.91–2.86, P˂0.001) for the association between V D /V T and ICU mortality in the overall population. Given the high morbidity and mortality of ARF, it is important to identify reliable prognostic indicators, not only to predict outcomes in individual patients, but also to stratify patients for clinical trials and escalation of supportive measures. In addition to patient characteristics such as age and severity of illness, various physiologic parameters associated with ARF, particularly in the form of ARDS - a distinct form commonly encountered in critically ill patients, have been identified to stratify disease severity and outcomes. The V D /V T represents one such parameter among these multiple variables. Numerous studies have documented a positive correlation between the V D /V T and prognosis in patients with ARDS [ 10 – 12 ]. A prior study reported that increased V D /V T occurs within hours of ARDS onset and independently predicts mortality, even after accounting for overall illness severity, hypoxemia, and compliance, with each 0.05 increase in V D /V T corresponding to 45% higher odds of death [ 3 ]. Data from prior observational studies suggest that a value of 0.60 or higher V D /V T may be associated with more severe lung injury [ 13 , 14 ]. Sustained elevation of V D /V T over the first week additionally identifies patients less likely to survive hospitalization [ 6 , 15 ]. Of note, the association between V D /V T and mortality was independent of the degree of oxygenation impairment in patients with ARDS [ 16 ]. In addition, V D /V T is a powerful predictor of extubation failure in patients on mechanical ventilation [ 17 ]. These findings suggest that the V D /V T might be an even better tool for evaluating the prognosis of patients with ARF. Elevated V D /V T in patients with ARF is likely a reflection of the complex interplay of multiple underlying mechanisms. Dead space refers to the fraction of tidal volume that does not participate in CO₂ elimination, serving as a reliable global indicator of pulmonary functional efficiency. While dead space ventilation exerts negligible clinical impact under physiological conditions, its pathological significance lies in the fact that various pulmonary disorders can trigger its expansion, thereby compromising ventilatory effectiveness. Notably, the magnitude of dead space is predominantly governed by the heterogeneity of pulmonary blood flow distribution [ 18 ]. In the context of acute lung injury, such alterations in blood flow and ventilation arise from two key pathways. First, shock can induce either inadequate pulmonary perfusion or hyperperfusion accompanied by maldistributed blood flow. The latter scenario, which is more common in ARDS, is closely linked to hyperdynamic states (e.g., severe sepsis) that transform large lung regions into areas with extremely high ventilation-perfusion ratios [ 19 ]. Second, pulmonary vascular injury contributes to increased pulmonary vascular resistance through mechanisms including vasoconstriction, microvascular thrombosis, and mechanical compression of pulmonary capillaries due to extravascular inflammatory responses [ 20 , 21 ]. Additionally, intrapulmonary shunt and reduced cardiac output are well-recognized factors associated with elevated V D /V T [ 20 , 22 ], though these primarily reflect perfusion deficits rather than intrinsic ventilatory impairment. Collectively, these mechanisms can augment absolute alveolar dead space and/or convert significant lung segments into zones of marked ventilation-perfusion imbalance, ultimately leading to elevated V D /V T . Thus, elevated V D /V T in ARF should be interpreted as a hallmark of profound disturbances in pulmonary perfusion, likely resulting from a combination of increased alveolar dead space and severe ventilation-perfusion mismatching. The direct assessment of pulmonary dead space remains technically difficult to perform at the bedside. A potential alternative to the direct measurement of dead space is the use of the simplified Enghoff modification equation ((PaCO 2 –PETCO 2 )/PaCO 2 ), which theoretically should be a reliable estimate of dead space and independently predictive of mortality in critically ill patients with ARDS [ 23 , 24 ]. This equation is known as the Bohr–Enghoff equation and to this day remains the most commonly used method to calculate dead space using mixed-expired CO 2 . The V D /V T , calculated using the Enghoff-Bohr equation, is influenced by the degree of intrapulmonary shunt [ 25 ]. Thus, V D /V T may provide a more comprehensive representation of overall lung injury. In addition, PaCO 2 and PETCO 2 are both measured frequently in critically ill, mechanically ventilated patients. In summary, although the prognostic marker of V D /V T is well-established in critically ill patients, its application in monitoring and guiding ventilatory management offers more immediate clinical benefits. In the present study, we found that patients with a high V D /V T had higher levels of glucose, lactate, ALT, AST, creatinine, and disease severity scores (SOFA, APACHE III, SAPS II), compared to those with a low V D /V T . It demonstrated that patients with a higher V D /V T exhibited more pronounced metabolic disturbances and organ dysfunction. These findings were consistent with a significantly increased risk of ICU mortality, suggesting a strong association between elevated V D /V T and poor prognosis in critically ill patients with ARF. It was found that the relationship between mean blood glucose and ICU mortality was a “J” shape [ 26 ]. And patients with glucose levels ≥ 10 mmol/L had a significantly higher 28-day mortality, with an adjusted odds ratio of 2.06 (95% CI: 1.65–2.57). It is well established that disease severity is positively associated with mortality. Evidence from this study suggests that V D /V T may represent a manifestation of disease severity and serves as a prognostic indicator for risk stratification in critically ill patients with ARF. In addition, an elevated V D /V T is significantly associated with an increased risk of ICU mortality, and this association is consistent across subgroups with different clinical characteristics. This suggests that the V D /V T can serve as a broad-spectrum predictive indicator for the mortality risk of critically ill patients. This study is a retrospective study based on the MIMIC-IV database, and it has limitations. First, this is a retrospective study with no precise timing between different measurements of PETCO₂ and PaCO 2 . Given that PaCO 2 and PETCO2 measurements are largely dependent on temporal hemodynamic and respiratory factors, our study is limited by potential disparities. Second, we only calculated V D /V T for the first admission to the ICU and did not evaluate its dynamic variation, more studies were needed. Finally, although we adjusted for confounding variables, other unknown and underlying factors may still exist, such as respiratory mechanics parameters including tidal volume and pulmonary compliance. Therefore, considering these considerations, further extensive multicenter prospective studies are necessary to verify the role of V D /V T as a prognostic factor for predicting adverse clinical outcomes in critically ill patients with ARF. Conclusion The findings suggest that V D /V T is positively associated with ICU mortality in critically ill patients with ARF. And the V D /V T threshold of 0.222 could represent a critical value for stratifying prognosis in critically ill patients with ARF. Further studies are needed to support the use of V D /V T clinically for early identification of critically ill ARF patients with higher mortality. Abbreviations ARF: acute respiratory failure; RCS: restricted cubic splines; PaCO 2 : arterial partial pressure of CO2; PETCO 2 : end-tidal carbon dioxide pressure; ARDS: acute respiratory distress syndrome; ICU: intensive care unit; MAP: mean artery pressure; RR: respiratory rate; WBC: white blood cell; ALT: alanine aminotransferase; AST: aspartate aminotransferase; COPD: chronic obstructive pulmonary disease; V D /V T : dead space fraction; SOFA: Sequential Organ Failure Assessment; APACHE III: Acute Physiology and Chronic Health Evaluation III; SAPS II: Simplified Acute Physiology Score II; CKD: chronic kidney disease; AKI: acute kidney injury; IQR: interquartile range; OR: odds ratio; CI: confidence interval; HR: hazard ratio; KM: Kaplan-Meier. Declarations Ethics approval and consent to participate This database was approved by Massachusetts Institute of Technology (Cambridge, MA) and Beth Israel Deaconess Medical Center (Boston, MA), with informed consent obtained for original data collection. Thus, ethical approval statement and informed consent are waived for this manuscript. Consent for publication Not applicable. Availability of data and materials The datasets analysed during the current study are available in the MIMIC-IV repository, https://doi.org/10.13026/6mm1-ek67. Competing interests The authors declare that they have no competing interests. Funding This work was supported by the National Key R & D Program of China [grant numbers 2024YFC3505700] and the National Natural Science Foundation of China [grant number 82102288]. Authors' contributions WFF: Conceptualization, Investigation, Writing-Original draft preparation. LXP: Methodology, Formal analysis, Data curation. XW: Conceptualization. ZYH: Investigation. WC: Data curation. MSL: Supervision, Writing-Reviewing and Editing. ZF: Supervision, Writing- Reviewing and Editing. WFF and LXP contributed equally to this work. All authors read and approved the final manuscript. Acknowledgements We are grateful to the Massachusetts Institute of Technology and Beth Israel Deaconess Medical Center for their role in the MIMIC project. Clinical trial number : not applicable. References Vincent JL, Akça S, Mendonça A, et al. The epidemiology of acute respiratory failure in critically ill patients(*). Chest,2002;121(5):1602-9. Linko R, Suojaranta-Ylinen R, Karlsson S, et al. One-year mortality, quality of life and predicted life-time cost-utility in critically ill patients with acute respiratory failure. Crit Care. 2010;14(2):R60. Nuckton TJ, Alonso JA, Kallet RH, et al. Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med. 2002;346(17):1281-6. Sivan Y, Eldadah MK, Cheah TE, et al. Estimation of arterial carbon dioxide by end-tidal and transcutaneous PCO2 measurements in ventilated children. Pediatr Pulmonol. 1992;12:153-7. Ghuman AK, Newth C, Khemani RG, et al. The association between the end tidal alveolar dead space fraction and mortality in pediatric acute hypoxemic respiratory failure. Pediatr Crit Care Med. 2012;13(1):11-5. Raurich JM, Vilar M, Colomar A, et al. Prognostic value of the pulmonary dead-space fraction during the early and intermediate phases of acute respiratory distress syndrome. Respir Care. 2010;55(3):282-7. Johnson AEW, Bulgarelli L, Shen L, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data 10, 1 (2023). https://doi.org/10.1038/s41597-022-01899-x. Johnson A, Bulgarelli L, Pollard T, et al. MIMIC-IV (version 2.2). PhysioNet. 2023. RRID:SCR_007345. Available from: https://doi.org/10.13026/6mm1-ek67. Sinha P, Flower O, Soni N. Deadspace ventilation: a waste of breath!. Intensive Care Med. 2011;37(5):735-46. Zhang YJ, Gao XJ, Li ZB, et al. Comparison of the pulmonary dead-space fraction derived from ventilator volumetric capnography and a validated equation in the survival prediction of patients with acute respiratory distress syndrome.Chin J Traumatol. 2016;19(3):141-5. Morales-Quinteros L, Schultz MJ, Bringué J, et al. Estimated dead space fraction and the ventilatory ratio are associated with mortality in early ARDS. Ann Intensive Care. 2019;9(1):128. Raurich JM, Vilar M, Colomar A, et al. Prognostic value of the pulmonary dead-space fraction during the early and intermediate phases of acute respiratory distress syndrome. Respir Care. 2010;55(3):282-7. Ralph DD, Robertson HT, Weaver LJ, et al. Distribution of ventilation and perfusion during positive endexpiratory pressure in the adult respiratory distress syndrome. Am Rev Res- pir Dis. 1985;131:54-60. Gattinoni L, Bombino M, Pelosi P, et al. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA. 1994;271:1772-9. Kallet RH, Alonso JA, Pittet J-F, et al. Prognostic value of the pulmonary dead-space fraction during the first 6 days of acute respiratory distress syndrome. Respir Care. 2004;49:1008-14. Kallet RH, Zhuo H, Liu KD, et al. The association between physiologic dead-space fraction and mortality in subjects with ARDS enrolled in a prospective multi-center clinical trial. Respir Care. 2014;59(11):1611-8. González-Castro A, Suárez-Lopez V, Gómez-Marcos V, et al. Utility of the dead space fraction (Vd/Vt) as a predictor of extubation success. Med Intensiva. 2011;35(9):529-38. Pontoppidan H, Geffin B, Lowenstein E. Acute respiratory failure in the adult. 3. N Engl J Med. 1972;287(16):799-806. Siegel JH, Giovannini I, Coleman B. Ventilation:perfusion maldistribution secondary to the hyperdynamic cardiovascular state as the major cause of increased pulmonary shunting in human sepsis. J Trauma. 1979;19(6):432-60. Tomashefski JF Jr, Davies P, Boggis C, et al. The pulmonary vascular lesions of the adult respiratory distress syndrome. Am J Pathol. 1983;1983(112):112-26. Zapol WM, Snider MT. Pulmonary hypertension in severe acute respiratory failure. N Engl J Med. 1977;296(9):476-80. Greene R, Zapol WM, Snider MT, et al. Early bedside detection of pulmonary vascular occlusion during acute respiratory failure. Am Rev Respir Dis. 1981; 124(5):593-601. Lecompte-Osorio P, Pearson SD, Pieroni CH, et al. Bedside estimates of dead space using end-tidal CO2 are independently associated with mortality in ARDS. Crit Care. 2021;25(1):333. Kallet RH, Zhuo H, Ho K, et al. Lung Injury Etiology and Other Factors Influencing the Relationship Between Dead-Space Fraction and Mortality in ARDS. Respir Care. 2017;62(10):1241-8. Waltuck BL. The Bohr equation. Anesthesiology. 1970;32(5):472473. Li Y, Li W, Xu B. Between blood glucose and mortality in critically ill patients: Retrospective analysis of the MIMIC-IV database. J Diabetes Investig. 2024;15(7):931-8. Tables Tables 1 to 5 are available in the Supplementary Files section. Additional Declarations No competing interests reported. Supplementary Files Tables.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7972231","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":558613293,"identity":"9e3661e5-dee9-4661-a123-f9dfe4ab5cab","order_by":0,"name":"feifei wang","email":"","orcid":"","institution":"Shanghai East Hospital","correspondingAuthor":false,"prefix":"","firstName":"feifei","middleName":"","lastName":"wang","suffix":""},{"id":558613299,"identity":"1e482656-37a6-481c-8f83-3b12c59f8605","order_by":1,"name":"Xingping Lv","email":"","orcid":"","institution":"Shanghai East Hospital","correspondingAuthor":false,"prefix":"","firstName":"Xingping","middleName":"","lastName":"Lv","suffix":""},{"id":558613300,"identity":"d1e14280-dcde-4285-9740-a3fe45bb4b88","order_by":2,"name":"Wei xu","email":"","orcid":"","institution":"Shanghai East Hospital","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"xu","suffix":""},{"id":558613301,"identity":"3e295f49-e00f-4533-a4d8-94e70fb4e326","order_by":3,"name":"yunhe Zhang","email":"","orcid":"","institution":"Shanghai East Hospital","correspondingAuthor":false,"prefix":"","firstName":"yunhe","middleName":"","lastName":"Zhang","suffix":""},{"id":558613303,"identity":"b96b6645-4b81-4f83-b223-2321a5df7b5c","order_by":4,"name":"chun Wang","email":"","orcid":"","institution":"Shanghai East Hospital","correspondingAuthor":false,"prefix":"","firstName":"chun","middleName":"","lastName":"Wang","suffix":""},{"id":558613305,"identity":"9315671c-5783-4c8a-a413-26685ca7a0be","order_by":5,"name":"shaolin ma","email":"","orcid":"","institution":"Shanghai East Hospital","correspondingAuthor":false,"prefix":"","firstName":"shaolin","middleName":"","lastName":"ma","suffix":""},{"id":558613306,"identity":"b850ee5e-05cb-4050-9899-348c29a23260","order_by":6,"name":"feng zhu","email":"data:image/png;base64,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","orcid":"","institution":"Shanghai East Hospital","correspondingAuthor":true,"prefix":"","firstName":"feng","middleName":"","lastName":"zhu","suffix":""}],"badges":[],"createdAt":"2025-10-28 17:28:32","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7972231/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7972231/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98180507,"identity":"1f7f71a6-db0b-409f-87cc-fa6c829b6a9a","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"tif","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":205882,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/7abfd621c8d01ac05e297aa1.tif"},{"id":98180510,"identity":"212f558f-48b9-4726-ad53-ca8df01ee769","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47684,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript2.docx","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/7fb6d538f9e7996ab6c46af0.docx"},{"id":98180519,"identity":"56732dc9-58a7-461a-b948-957fa1c28a61","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124734,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/fbabf4a1325cec419d3c0652.tif"},{"id":98180526,"identity":"0545616d-0753-49ba-ad83-5e10b3e26519","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124496,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/1f95752f0dac6573f4b9a2fb.tif"},{"id":98180532,"identity":"13f25084-4278-4ef1-9e2b-2044687a26e9","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":64542,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/b9544975af9d0d7afbfa7432.docx"},{"id":98180530,"identity":"642a0c7a-3b0a-4a6c-968f-4bbb45787193","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"tif","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":175746,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/483b6b4a1494ca5b0cf7b5c7.tif"},{"id":98430693,"identity":"1ecb6998-3153-448c-b976-66dd42bf29e5","added_by":"auto","created_at":"2025-12-17 16:46:03","extension":"tif","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":436196,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/edda1f0c8c39afe5377df241.tif"},{"id":98180527,"identity":"67ce1e0a-2b53-4032-b2fb-5bfe2ab01f91","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"json","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8335,"visible":true,"origin":"","legend":"","description":"","filename":"83586d9e90954cddb07a606b5c7a715e.json","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/edd68f663332a0efedc7285a.json"},{"id":98180516,"identity":"fc7c7a23-2bf8-4980-8c95-11b05e5aa9e2","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":206586,"visible":true,"origin":"","legend":"","description":"","filename":"83586d9e90954cddb07a606b5c7a715e1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/6919556fcede3895ab12fb70.xml"},{"id":98180518,"identity":"cb7ed556-f62b-462a-9d54-1a88608626c0","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"tif","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":205882,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/7a7d03a846d01621e338b3dd.tif"},{"id":98180513,"identity":"b9edfa88-d461-4d4a-b475-a0d45fcd8089","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"tif","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124734,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/d154a1e43a7301f125efc0aa.tif"},{"id":98180523,"identity":"4a19e0fd-4f2d-4364-a8c0-91042b56c0d0","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"tif","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":124496,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/8ffc8aa8a64642a010e00697.tif"},{"id":98180521,"identity":"796a95be-98c1-404d-b44e-c0f8260e6712","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"tif","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":175746,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/57ce26f13d6c2aa4ad87d527.tif"},{"id":98180533,"identity":"85658bce-daf4-4b16-87a3-c36ff9eef3b9","added_by":"auto","created_at":"2025-12-15 01:06:25","extension":"tif","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":436196,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.tif","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/39c3309d2612ba08accafa12.tif"},{"id":98180524,"identity":"7b5ccf7e-3cc2-47b7-85c2-8de3e4cd13f4","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26822,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/149bde4c7677d231b343d945.png"},{"id":98180522,"identity":"a9b5e4e9-60d9-4737-b328-dc2aea647c8f","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23407,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/a56a6e4affc7fb0c2d761ef1.png"},{"id":98180529,"identity":"857a7c31-8bb1-488b-a938-b493cf525afc","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":23190,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/83831a7a55fbd8b1ce4a7b24.png"},{"id":98431985,"identity":"8aeadbde-1253-4d8b-8392-14405c035c9a","added_by":"auto","created_at":"2025-12-17 16:48:43","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25308,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/d1779db56c4a897d21463f12.png"},{"id":98180531,"identity":"37c33ddc-0962-4037-a55c-9b65e1a0b846","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":76537,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/55073b0f8badd5526d64d922.png"},{"id":98431345,"identity":"9480930e-406e-4baa-8da0-58135d65d69b","added_by":"auto","created_at":"2025-12-17 16:47:33","extension":"xml","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":205620,"visible":true,"origin":"","legend":"","description":"","filename":"83586d9e90954cddb07a606b5c7a715e1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/1d2ba95319dde3439af568f6.xml"},{"id":98432335,"identity":"36626133-596d-4959-930a-f74f1e12e736","added_by":"auto","created_at":"2025-12-17 16:49:25","extension":"html","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":204577,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/91b295ce292574e0c81a3a6e.html"},{"id":98180511,"identity":"57b4c618-8e03-44ea-8a49-33f5c64df4f6","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":471320,"visible":true,"origin":"","legend":"\u003cp\u003eKaplan-Meier Survival Curves for ICU Mortality Stratified by Quartiles of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. Patients were stratified into four equal groups (Quartile 1 to Quartile 4) based on the quartile values of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. Quartile 1 represents the lowest V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group, and Quartile 4 represents the highest V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group. The x-axis represents time in days, and the y-axis represents the cumulative survival probability. The Log-rank test was used to compare differences in survival curves among the four groups, with a statistically significant result (Log rank P \u0026lt; 0.001).\u003c/p\u003e","description":"","filename":"Figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/d3f8f8a8a8be0888cb4ef3b9.png"},{"id":98431471,"identity":"e7772a54-3990-4599-91b3-ca70cd103d2a","added_by":"auto","created_at":"2025-12-17 16:47:45","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":105032,"visible":true,"origin":"","legend":"\u003cp\u003eRestricted Cubic Spline (RCS) analysis of the nonlinear association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality (Univariate Model). The x-axis represents the value of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, and the y-axis represents the Odds Ratio (OR) of ICU mortality, with a reference value of 1 (dashed horizontal line, indicating no association with mortality). The shaded area around the OR curve represents the 95% Confidence Interval (CI). Overall P \u0026lt; 0.001, indicating a significant overall association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality in the univariate RCS model. Nonlinear P = 0.032, confirming a statistically significant nonlinear relationship between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality.\u003c/p\u003e","description":"","filename":"Figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/0e07c0827c82d84438313eba.png"},{"id":98180508,"identity":"b82b6bf3-e95a-4c46-bd26-176952905e0b","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":104439,"visible":true,"origin":"","legend":"\u003cp\u003eRestricted Cubic Spline (RCS) analysis of the nonlinear association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality (Multivariate Model). The x-axis represents the value of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, and the y-axis denotes the Odds Ratio (OR) of ICU mortality, with 1.0 (dashed horizontal line) as the reference value (indicating no association with mortality). The shaded area surrounding the OR curve corresponds to the 95% Confidence Interval (CI), reflecting the variability of the OR estimate. Overall P \u0026lt; 0.001, confirming a significant overall association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality even after adjusting for confounding factors. Nonlinear P = 0.040, verifying that the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality remains statistically nonlinear following covariate adjustment—ruling out the possibility that the nonlinear trend observed in the univariate model is driven by confounding variables.\u003c/p\u003e","description":"","filename":"Figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/8d34d257353e9aa1e61cab2e.png"},{"id":98180517,"identity":"c9f293fa-7b53-4993-b8fd-8f3c8586783a","added_by":"auto","created_at":"2025-12-15 01:06:24","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":356609,"visible":true,"origin":"","legend":"\u003cp\u003eKaplan-Meier survival curves for ICU mortality stratified by the Restricted Cubic Spline (RCS)-derived threshold of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT \u003c/sub\u003e(0.222). Patients were dichotomized into two groups based on the inflection point of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. Group 1 is the low V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group (V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e≤0.222).Group 2 is the high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group (V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e \u0026gt; 0.222). The x-axis represents follow-up time in days, and the y-axis represents cumulative survival probability. The Log-rank test was applied to compare survival differences between the two groups, with a statistically significant result (Log rank P \u0026lt; 0.001). The hazard ratio (HR) for ICU mortality in the high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group versus the low V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group is 1.984 (95% Confidence Interval [CI]: 1.661-2.369).\u003c/p\u003e","description":"","filename":"Figure4.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/83e94e9263ed1ca209812e64.png"},{"id":98180514,"identity":"479ecb1a-f67e-4ff6-9948-9af3436fa101","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1185251,"visible":true,"origin":"","legend":"\u003cp\u003eForest plot of subgroup analyses for the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality. The \"P for interaction\" column evaluates whether the strength of the V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e-ICU mortality association differs across subgroups. All interaction P\u0026gt;0.05, indicating no statistically significant differences in the association across subgroups. 0 = absence of the disease; 1 = presence of the disease.\u003c/p\u003e","description":"","filename":"Figure5.png","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/6b5380908df5aa840faf23bc.png"},{"id":109115014,"identity":"46bdb3db-b70e-466b-b658-ea9241a18947","added_by":"auto","created_at":"2026-05-12 16:13:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3089733,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/5d016b87-cb24-47db-83b5-67ca32a62181.pdf"},{"id":98180509,"identity":"eaf24154-7ba1-4989-9207-40c6f8ff997b","added_by":"auto","created_at":"2025-12-15 01:06:23","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":64542,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-7972231/v1/8c2ccdfadf35791c434515f6.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Association between the dead space fraction and mortality in critically ill patients with acute respiratory failure: an analysis of the MIMIC-IV database","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAcute respiratory failure (ARF) stands as a predominant trigger for critical care consultations, representing a cardinal manifestation of numerous cardiopulmonary disorders and carrying a substantial burden of morbidity and mortality globally [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. A pivotal pathological feature underlying ARF is the augmentation of alveolar dead space\u0026mdash;defined as alveolar regions receiving ventilation without corresponding perfusion\u0026mdash;which directly compromises effective gas exchange. The dead space fraction (V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e), calculated as the ratio of dead space volume to tidal volume, serves as a central metric for quantifying dead space ventilation. Beyond merely reflecting the volume of non-functional alveoli, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e integrates multiple dimensions of lung dysfunction in acute hypoxemic respiratory failure, including the magnitude of intrapulmonary shunting and the extent of pulmonary vascular injury.\u003c/p\u003e\u003cp\u003eClinically, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e is commonly derived at the bedside using arterial partial pressure of carbon dioxide (PaCO₂) and end-tidal carbon dioxide pressure (PETCO₂), via the formula: V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e = (PaCO₂ \u0026ndash; PETCO₂)/PaCO₂. In healthy adults, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e typically remains below 0.30; however, this threshold is markedly exceeded in patients with acute respiratory distress syndrome (ARDS), with values surpassing 0.55\u0026ndash;0.60 consistently linked to heightened mortality in this subset [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Prior investigations have underscored the clinical relevance of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e: Sivan et al. observed significant discrepancies between PETCO₂ and PaCO₂ when the arterial-to-alveolar oxygen ratio dropped below 0.3, highlighting the interplay between increased alveolar dead space and impaired oxygenation in severe lung disease [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Additionally, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e has been shown to independently predict mortality in pediatric acute hypoxemic respiratory failure, even after accounting for illness severity [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. In ARDS specifically, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e exhibits a robust association with mortality across both early and intermediate disease phases\u0026mdash;each 0.05 increment in VD/VT correlates with a 59% increase in the odds of death in the early phase and a 186% increase in the odds in the intermediate phase [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eDespite these insights, the existing literature predominantly focuses on ARDS\u0026mdash;the most severe phenotype of ARF\u0026mdash;with limited attention directed toward characterizing V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and its prognostic value in the broader population of critically ill patients with ARF. This gap is notable, as ARF encompasses a heterogeneous spectrum of conditions beyond ARDS, and understanding V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e role in this larger cohort could enhance risk stratification and clinical decision-making. To address this limitation, the present study aimed to: (1) investigate the distribution and clinical correlates of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e in a diverse cohort of critically ill patients with ARF; (2) clarify the independent association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality in this population; and (3) identify an optimal V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e threshold that may serve as a practical prognostic tool for clinical application. By leveraging data from the MIMIC-IV database, this study seeks to expand the evidence base for V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e as a prognostic marker in ARF, ultimately supporting more targeted and effective management of critically ill patients.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eData Source\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe core analysis of this study was based on the Medical Information Mart for Intensive Care IV (MIMIC-IV) database (Version 2.2). This database contains comprehensive clinical data of patients admitted to the Intensive Care Unit (ICU) of Beth Israel Deaconess Medical Center in Boston from 2008 to 2019 [7,8], serving as a rich and reliable data source for medical research. The use of the database has been approved by relevant institutions and complies with research data usage regulations. Qualified researchers can obtain the data free of charge for research purposes after completing the corresponding data use agreement.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePatient Selection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAdult patients (aged\u0026nbsp;\u0026ge;\u0026nbsp;18 years) diagnosed with ARF via ICD-9 and ICD-10 codes were identified from the database. The inclusion criterion was that patients had available PaCO₂ and PETCO₂ data within 24 hours of ICU admission. The exclusion criteria included: patients with an ICU length of stay of less than 24 hours, and patients with a calculated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e equal to 1. These criteria were set to ensure the validity of the study sample and the reliability of the data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Extraction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eKey data were extracted from the clinical records of patients within 24 hours of ICU admission, including the following categories:Demographic characteristics: Age, gender; Vital signs: Mean arterial pressure (MAP), respiratory rate (RR), percutaneous arterial oxygen saturation (SPO₂); Laboratory test indicators: White blood cell count (WBC), platelet count, creatinine, alanine aminotransferase (ALT), aspartate aminotransferase (AST), albumin, glucose; Comorbidity status: Hypertension, diabetes mellitus, myocardial infarction, heart failure, chronic kidney disease, acute kidney injury, malignant tumor, pneumonia, stroke, chronic obstructive pulmonary disease (COPD); Disease severity scores: Sequential Organ Failure Assessment (SOFA) score, Acute Physiology and Chronic Health Evaluation\u0026nbsp;Ⅲ\u0026nbsp;(APACHE\u0026nbsp;Ⅲ) score, Simplified Acute Physiology Score\u0026nbsp;Ⅱ\u0026nbsp;(SAPS\u0026nbsp;Ⅱ) score; Arterial blood gas analysis parameters: pH, PaO₂, PaCO₂, lactate; Other relevant data: ICU length of stay, PETCO₂.\u003c/p\u003e\n\u003cp\u003eThe primary outcome indicator of this study was ICU mortality, defined as death from any cause during the patient\u0026apos;s ICU stay. V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e was calculated using the simplified Enghoff modification of the Bohr equation [9], with the specific formula: V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e = (PaCO₂ - PETCO₂)/PaCO₂.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll statistical analyses were performed using R software (Version 4.5.1), with the significance level set at \u0026alpha; = 0.05. Descriptive statistics: Continuous variables, which typically follow a non-normal distribution in clinical data, were expressed as median (interquartile range, IQR), and inter-group comparisons were conducted using the Mann-Whitney U test; Categorical variables were presented as frequency (percentage), and inter-group comparisons were performed using the chi-square test or Fisher\u0026apos;s exact test (when applicable). Correlation analysis: Univariate and multivariate logistic regression analyses were used to screen for factors independently associated with ICU mortality, and the results were presented as adjusted odds ratios (OR) with their 95% confidence intervals (CI); Cox proportional hazards regression models were employed to evaluate the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e groups and ICU mortality outcomes, and the results were expressed as hazard ratios (HR) with their 95% CI. Non-linear relationship analysis: Restricted cubic spline (RCS) analysis was performed based on the Cox proportional hazards model to assess the non-linear relationship between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality. The RCS model was adjusted for the aforementioned potential confounding factors, and the optimal cut-off value of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e was determined through this analysis. Survival analysis: Patients were divided into four groups according to the quartiles of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, and the Kruskal-Wallis test was used to compare differences in ICU mortality among the four groups; Patients were further divided into high and low V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e groups based on the cut-off value determined by RCS analysis. Kaplan-Meier (KM) curves were plotted to visualize survival outcomes, and the log-rank test was used to compare survival differences between the two groups. Subgroup analysis: A forest plot was generated to present the results of univariate logistic regression analysis regarding the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality in key patient subgroups. The OR value and its 95% CI for each subgroup were displayed to verify the robustness of the association.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eParticipants Characteristics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study enrolled 2,254 patients in total. Of the study cohort, 1,709 patients (75.82%) had an outcome of death during ICU hospitalization, while 545 patients (24.18%) survived. Significant differences were observed between the two groups in the following variables: WBC, albumin, glucose, pH, PaO\u003csub\u003e2\u003c/sub\u003e, lactate, ALT, AST, creatinine, RR, SpO₂, hospital length of stay, ICU length of stay, scoring systems, PETCO₂, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, hypertension, myocardial infarction, chronic kidney disease, mechanical ventilation, and acute kidney injury.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eNo statistically significant differences were found between the two groups with respect to age, PaCO\u003csub\u003e2\u003c/sub\u003e, or other variables not specified above.\u003cstrong\u003e\u0026nbsp;Table 1\u003c/strong\u003e presents the baseline characteristics of all participants.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eLogistic regression and Cox proportional hazards models analysis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnivariable logistic regression analysis indicated that V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e was the most significant predictor of ICU mortality (OR=23.78, 95%CI:13.09-43.20,\u0026nbsp;\u003cstrong\u003eTable2\u003c/strong\u003e). Among comorbidities, underlying conditions such as kidney disease, and myocardial infarction were identified as risk factors for ICU mortality. Elevated laboratory values including WBC, glucose, lactate, ALT, AST, and creatinine were associated with increased risk of ICU mortality. After adjusting for potential confounders using multivariable logistic regression, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, glucose, and lactate remained independent risk factors for ICU mortality.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eUnivariate Cox proportional hazards model analysis also revealed that multiple factors were associated with the risk of ICU mortality (\u003cstrong\u003eTable 3\u003c/strong\u003e). In terms of underlying diseases, conditions such as myocardial infarction, malignant tumors, and kidney disease were all found to increase the risk of ICU mortality. Additionally, elevated levels of age, WBC, glucose, lactate, AST, ALT, creatinine, and V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e were associated with an increased risk of ICU mortality. After adjusting for confounding variables in the multivariate Cox proportional hazards model, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, malignant tumors, age, WBC, glucose, and lactate were identified as independent factors influencing ICU mortality,\u0026nbsp;among which V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e was the most significant predictor of ICU mortality (HR=9.60, 95% CI: 4.95-18.61, p˂0.001).\u003c/p\u003e\n\u003cp\u003eResults from both logistic regression and cox proportional hazards model analyses indicated that V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e was strong risk factors for ICU mortality, and both remained significantly independently associated with ICU mortality in multivariate analyses. In contrast, gender showed no significant association with ICU mortality in either model.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAnalysis of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e Values by Quartile Groups\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePatients were stratified into quartiles based on V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e values, with Group 1 to Group 4 comprising 563 (24.98%), 551 (24.45%), 576 (25.55%), and 564 (25.02%) patients, respectively. Significant differences in multiple clinical characteristics and outcomes were observed across the four groups (\u003cstrong\u003eTable 4\u003c/strong\u003e). ICU mortality increased significantly with higher V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e quartiles. Trends in laboratory and physiological parameters showed that glucose, lactate, PaCO\u003csub\u003e2\u003c/sub\u003e, ALT, AST, and creatinine increased with rising V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, whereas pH, PaO\u003csub\u003e2\u003c/sub\u003e, and PETCO₂ decreased. Analysis of disease revealed that the incidence of pneumonia and COPD, as well as severity scores (SOFA, APACHE III, SAPS II), were significantly higher in groups with elevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. No significant differences were observed across quartiles in age, sex, hypertension, diabetes mellitus, or myocardial infarction.\u003c/p\u003e\n\u003cp\u003eAdditionally, Kaplan-Meier survival curve analysis showed significant differences in survival rates among the four groups (\u003cstrong\u003eFigure 1\u003c/strong\u003e). Group 1 (with the lowest V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e) consistently had the highest survival rate, and the rate decreased most slowly over time; in contrast, Group 4 (with the highest V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e) had the lowest survival rate, which declined most rapidly as time progressed. Regarding changes in the number of at-risk patients, the number of at-risk individuals in each group gradually decreased with prolonged follow-up, but the decrease was more pronounced in Group 4, consistent with the faster decline in survival rate of Group 4 observed in the survival curve. In summary, patients with higher V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e had significantly poorer survival outcomes than those with lower values.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRestricted Cubic Splines Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnivariable RCS analysis revealed a non-linear association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality (\u003cstrong\u003eFigure 2\u003c/strong\u003e). The OR exhibited a decreasing trend as V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e increased from 0.0 to 0.2. However, beyond the inflection point (0.222), the OR demonstrated a pronounced upward trend with further increases in V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, indicating a significantly elevated risk of ICU mortality at higher V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. These findings suggest that the relationship between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality is not purely linear but exhibits a distinct non-linear pattern.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAfter adjusting for covariates including sex, age, hypertension, diabetes mellitus, pneumonia, COPD, WBC, and platelets, the RCS analysis continued to demonstrate a significant overall association and a non-linear relationship between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality (\u003cstrong\u003eFigure 3\u003c/strong\u003e). The persistence of this non-linear association after multivariable adjustment further supports that V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e is an important independent factor influencing ICU mortality risk.\u003c/p\u003e\n\u003cp\u003ePatients were divided into two groups based on the RCS threshold of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e (0.222): Group 1 (low V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e) and Group 2 (high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e). Further analysis revealed significant statistical differences in multiple clinical indicators and outcome variables between the two groups (\u003cstrong\u003eTable 5\u003c/strong\u003e). In terms of laboratory and physiological parameters, Group 2 showed significantly higher values in glucose, lactate, PaCO\u003csub\u003e2\u003c/sub\u003e, ALT, AST, and creatinine. With regard to disease severity and outcomes, Group 2 had significantly higher scores on the SOFA, APACHE III, and SAPS II, as well as significantly higher incidence rates of pneumonia and COPD, and a markedly elevated ICU mortality.\u0026nbsp;These findings demonstrate that patients with higher V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e exhibit more severe illness, worse clinical indicators, and a significantly increased risk of ICU mortality, further validating the close association between elevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and adverse outcomes in critically ill patients with ARF.\u003c/p\u003e\n\u003cp\u003eAdditionally, Kaplan-Meier survival analysis based on the RCS threshold of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e (0.222) revealed a significant difference in survival between the two groups (\u003cstrong\u003eFigure 4\u003c/strong\u003e). The high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group exhibited a steeper decline in survival over time and demonstrated a significantly higher risk of mortality (HR = 1.984, 95% CI: 1.661\u0026ndash;2.369) compared to the low V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group. Furthermore, although the number at risk decreased over time in both groups, high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group consistently showed lower survival rates than low V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e group, and this difference became more pronounced during follow-up.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSubgroup Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSubgroup analysis results from the univariate forest plot examining the relationship between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality showed the following: among the 2,254 included patients, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e exhibited an extremely strong statistical association with ICU mortality, with an overall OR of 2.34 (95% CI: 1.91\u0026ndash;2.86, p˂0.001, \u003cstrong\u003eFigure 5\u003c/strong\u003e). After stratification by factors such as gender, hypertension, diabetes mellitus, heart failure, and chronic kidney disease, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e was significantly associated with ICU mortality risk in all subgroups. Furthermore, there was no statistically significant interaction effect between the subgroups. These findings indicate that the positive association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality is consistently present across populations with different characteristics.\u0026nbsp;\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe current study showed that high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e were associated with increased ICU mortality risk in critically ill patients with ARF who are admitted to the ICU, as per the MIMIC-IV database. The multivariate analysis of logistic regression and cox proportional hazards models revealed that high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e was independently associated with ICU mortality in these patients. Furthermore, following multivariable adjustment, results from the RCS analysis remained robust: they not only confirmed a significant overall association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality but also identified a non-linear pattern underlying this relationship. A critical finding was a non-linear cut-point of 0.222 for the V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and mortality relationship identified by RCS analysis, with subgroup analyses based on this grouping robustly corroborated the aforementioned results. Finally, the subgroup forest plot demonstrated an overall OR of 2.34 (95% CI: 1.91\u0026ndash;2.86, P˂0.001) for the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality in the overall population.\u003c/p\u003e\u003cp\u003eGiven the high morbidity and mortality of ARF, it is important to identify reliable prognostic indicators, not only to predict outcomes in individual patients, but also to stratify patients for clinical trials and escalation of supportive measures. In addition to patient characteristics such as age and severity of illness, various physiologic parameters associated with ARF, particularly in the form of ARDS - a distinct form commonly encountered in critically ill patients, have been identified to stratify disease severity and outcomes. The V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e represents one such parameter among these multiple variables. Numerous studies have documented a positive correlation between the V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and prognosis in patients with ARDS [\u003cspan additionalcitationids=\"CR11\" citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. A prior study reported that increased V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e occurs within hours of ARDS onset and independently predicts mortality, even after accounting for overall illness severity, hypoxemia, and compliance, with each 0.05 increase in V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e corresponding to 45% higher odds of death [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Data from prior observational studies suggest that a value of 0.60 or higher V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e may be associated with more severe lung injury [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Sustained elevation of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e over the first week additionally identifies patients less likely to survive hospitalization [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Of note, the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and mortality was independent of the degree of oxygenation impairment in patients with ARDS [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. In addition, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e is a powerful predictor of extubation failure in patients on mechanical ventilation [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. These findings suggest that the V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e might be an even better tool for evaluating the prognosis of patients with ARF.\u003c/p\u003e\u003cp\u003eElevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e in patients with ARF is likely a reflection of the complex interplay of multiple underlying mechanisms. Dead space refers to the fraction of tidal volume that does not participate in CO₂ elimination, serving as a reliable global indicator of pulmonary functional efficiency. While dead space ventilation exerts negligible clinical impact under physiological conditions, its pathological significance lies in the fact that various pulmonary disorders can trigger its expansion, thereby compromising ventilatory effectiveness. Notably, the magnitude of dead space is predominantly governed by the heterogeneity of pulmonary blood flow distribution [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. In the context of acute lung injury, such alterations in blood flow and ventilation arise from two key pathways. First, shock can induce either inadequate pulmonary perfusion or hyperperfusion accompanied by maldistributed blood flow. The latter scenario, which is more common in ARDS, is closely linked to hyperdynamic states (e.g., severe sepsis) that transform large lung regions into areas with extremely high ventilation-perfusion ratios [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Second, pulmonary vascular injury contributes to increased pulmonary vascular resistance through mechanisms including vasoconstriction, microvascular thrombosis, and mechanical compression of pulmonary capillaries due to extravascular inflammatory responses [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Additionally, intrapulmonary shunt and reduced cardiac output are well-recognized factors associated with elevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e], though these primarily reflect perfusion deficits rather than intrinsic ventilatory impairment. Collectively, these mechanisms can augment absolute alveolar dead space and/or convert significant lung segments into zones of marked ventilation-perfusion imbalance, ultimately leading to elevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. Thus, elevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e in ARF should be interpreted as a hallmark of profound disturbances in pulmonary perfusion, likely resulting from a combination of increased alveolar dead space and severe ventilation-perfusion mismatching.\u003c/p\u003e\u003cp\u003eThe direct assessment of pulmonary dead space remains technically difficult to perform at the bedside. A potential alternative to the direct measurement of dead space is the use of the simplified Enghoff modification equation ((PaCO\u003csub\u003e2\u003c/sub\u003e\u0026ndash;PETCO\u003csub\u003e2\u003c/sub\u003e)/PaCO\u003csub\u003e2\u003c/sub\u003e), which theoretically should be a reliable estimate of dead space and independently predictive of mortality in critically ill patients with ARDS [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. This equation is known as the Bohr\u0026ndash;Enghoff equation and to this day remains the most commonly used method to calculate dead space using mixed-expired CO\u003csub\u003e2\u003c/sub\u003e. The V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e, calculated using the Enghoff-Bohr equation, is influenced by the degree of intrapulmonary shunt [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Thus, V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e may provide a more comprehensive representation of overall lung injury. In addition, PaCO\u003csub\u003e2\u003c/sub\u003e and PETCO\u003csub\u003e2\u003c/sub\u003e are both measured frequently in critically ill, mechanically ventilated patients. In summary, although the prognostic marker of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e is well-established in critically ill patients, its application in monitoring and guiding ventilatory management offers more immediate clinical benefits.\u003c/p\u003e\u003cp\u003eIn the present study, we found that patients with a high V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e had higher levels of glucose, lactate, ALT, AST, creatinine, and disease severity scores (SOFA, APACHE III, SAPS II), compared to those with a low V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. It demonstrated that patients with a higher V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e exhibited more pronounced metabolic disturbances and organ dysfunction. These findings were consistent with a significantly increased risk of ICU mortality, suggesting a strong association between elevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and poor prognosis in critically ill patients with ARF. It was found that the relationship between mean blood glucose and ICU mortality was a \u0026ldquo;J\u0026rdquo; shape [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. And patients with glucose levels\u0026thinsp;\u0026ge;\u0026thinsp;10 mmol/L had a significantly higher 28-day mortality, with an adjusted odds ratio of 2.06 (95% CI: 1.65\u0026ndash;2.57). It is well established that disease severity is positively associated with mortality. Evidence from this study suggests that V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e may represent a manifestation of disease severity and serves as a prognostic indicator for risk stratification in critically ill patients with ARF. In addition, an elevated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e is significantly associated with an increased risk of ICU mortality, and this association is consistent across subgroups with different clinical characteristics. This suggests that the V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e can serve as a broad-spectrum predictive indicator for the mortality risk of critically ill patients.\u003c/p\u003e\u003cp\u003eThis study is a retrospective study based on the MIMIC-IV database, and it has limitations. First, this is a retrospective study with no precise timing between different measurements of PETCO₂ and PaCO\u003csub\u003e2\u003c/sub\u003e. Given that PaCO\u003csub\u003e2\u003c/sub\u003e and PETCO2 measurements are largely dependent on temporal hemodynamic and respiratory factors, our study is limited by potential disparities. Second, we only calculated V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e for the first admission to the ICU and did not evaluate its dynamic variation, more studies were needed. Finally, although we adjusted for confounding variables, other unknown and underlying factors may still exist, such as respiratory mechanics parameters including tidal volume and pulmonary compliance. Therefore, considering these considerations, further extensive multicenter prospective studies are necessary to verify the role of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e as a prognostic factor for predicting adverse clinical outcomes in critically ill patients with ARF.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThe findings suggest that V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e is positively associated with ICU mortality in critically ill patients with ARF. And the V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e threshold of 0.222 could represent a critical value for stratifying prognosis in critically ill patients with ARF. Further studies are needed to support the use of V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e clinically for early identification of critically ill ARF patients with higher mortality.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eARF: acute respiratory failure; RCS: restricted cubic splines; PaCO\u003csub\u003e2\u003c/sub\u003e: arterial partial pressure of CO2; PETCO\u003csub\u003e2\u003c/sub\u003e: end-tidal carbon dioxide pressure; ARDS: acute respiratory distress syndrome; ICU: intensive care unit;\u0026nbsp;MAP:\u0026nbsp;mean artery pressure; RR: respiratory rate; WBC: white blood cell; ALT: alanine aminotransferase; AST: aspartate aminotransferase; COPD: chronic obstructive pulmonary disease; V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e: dead space fraction; SOFA: Sequential Organ Failure Assessment; APACHE III: Acute Physiology and Chronic Health Evaluation III; SAPS II: Simplified Acute Physiology Score II; CKD: chronic kidney disease; AKI: acute kidney injury; IQR: interquartile range; OR: odds ratio; CI: confidence interval; HR: hazard ratio; KM: Kaplan-Meier.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis database was approved by Massachusetts Institute of Technology (Cambridge, MA) and Beth Israel Deaconess Medical Center (Boston, MA), with informed consent obtained for original data collection. Thus, ethical approval statement and informed consent are waived for this manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets analysed during the current study are available in the MIMIC-IV repository, https://doi.org/10.13026/6mm1-ek67.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Key R \u0026amp; D Program of China [grant numbers 2024YFC3505700] and the National Natural Science Foundation of China [grant number 82102288].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWFF: Conceptualization, Investigation, Writing-Original draft preparation. LXP: Methodology, Formal analysis, Data curation. XW: Conceptualization. ZYH: Investigation. WC: Data curation. MSL: Supervision, Writing-Reviewing and Editing. ZF: Supervision, Writing- Reviewing and Editing. WFF and LXP contributed equally to this work.\u0026nbsp;All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe are grateful to the Massachusetts Institute of Technology and Beth Israel Deaconess Medical Center for their role in the MIMIC project.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number\u003c/strong\u003e: not applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eVincent JL, Ak\u0026ccedil;a S, Mendon\u0026ccedil;a A, et al. The epidemiology of acute respiratory failure in critically ill patients(*). Chest,2002;121(5):1602-9.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLinko R, Suojaranta-Ylinen R, Karlsson S, et al. One-year mortality, quality of life and predicted life-time cost-utility in critically ill patients with acute respiratory failure. Crit Care. 2010;14(2):R60.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eNuckton TJ, Alonso JA, Kallet RH, et al. Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med. 2002;346(17):1281-6.\u003c/li\u003e\n \u003cli\u003eSivan Y, Eldadah MK, Cheah TE, et al. Estimation of arterial carbon dioxide by end-tidal and transcutaneous PCO2 measurements in ventilated children. Pediatr Pulmonol. 1992;12:153-7.\u003c/li\u003e\n \u003cli\u003eGhuman AK, Newth C, Khemani RG, et al. The association between the end tidal alveolar dead space fraction and mortality in pediatric acute hypoxemic respiratory failure. Pediatr Crit Care Med. 2012;13(1):11-5.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eRaurich JM, Vilar M, Colomar A, et al. Prognostic value of the pulmonary dead-space fraction during the early and intermediate phases of acute respiratory distress syndrome. Respir Care. 2010;55(3):282-7.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eJohnson AEW, Bulgarelli L, Shen L, et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci Data 10, 1 (2023). https://doi.org/10.1038/s41597-022-01899-x.\u003c/li\u003e\n \u003cli\u003eJohnson A, Bulgarelli L, Pollard T, et al. MIMIC-IV (version 2.2). PhysioNet. 2023. RRID:SCR_007345. Available from: https://doi.org/10.13026/6mm1-ek67.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSinha P, Flower O, Soni N. Deadspace ventilation: a waste of breath!. Intensive Care Med. 2011;37(5):735-46.\u003c/li\u003e\n \u003cli\u003eZhang YJ, Gao XJ, Li ZB, et al. Comparison of the pulmonary dead-space fraction derived from ventilator volumetric capnography and a validated equation in the survival prediction of patients with acute respiratory distress syndrome.Chin J Traumatol. 2016;19(3):141-5. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eMorales-Quinteros L, Schultz MJ, Bringu\u0026eacute; J, et al. Estimated dead space fraction and the ventilatory ratio are associated with mortality in early ARDS. Ann Intensive Care. 2019;9(1):128. \u0026nbsp;\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eRaurich JM, Vilar M, Colomar A, et al. Prognostic value of the pulmonary dead-space fraction during the early and intermediate phases of acute respiratory distress syndrome. Respir Care. 2010;55(3):282-7.\u003c/li\u003e\n \u003cli\u003eRalph DD, Robertson HT, Weaver LJ, et al. Distribution of ventilation and perfusion during positive endexpiratory pressure in the adult respiratory distress syndrome. Am Rev Res- pir Dis. 1985;131:54-60. \u0026nbsp; \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eGattinoni L, Bombino M, Pelosi P, et al. Lung structure and function in different stages of severe adult respiratory distress syndrome. JAMA. 1994;271:1772-9.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKallet RH, Alonso JA, Pittet J-F, et al. Prognostic value of the pulmonary dead-space fraction during the first 6 days of acute respiratory distress syndrome. Respir Care. 2004;49:1008-14. \u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKallet RH, Zhuo H, Liu KD, et al. The association between physiologic dead-space fraction and mortality in subjects with ARDS enrolled in a prospective multi-center clinical trial. Respir Care. 2014;59(11):1611-8.\u003c/li\u003e\n \u003cli\u003eGonz\u0026aacute;lez-Castro A, Su\u0026aacute;rez-Lopez V, G\u0026oacute;mez-Marcos V, et al. Utility of the dead space fraction (Vd/Vt) as a predictor of extubation success. Med Intensiva. 2011;35(9):529-38.\u003c/li\u003e\n \u003cli\u003ePontoppidan H, Geffin B, Lowenstein E. Acute respiratory failure in the adult. 3. N Engl J Med. 1972;287(16):799-806.\u003c/li\u003e\n \u003cli\u003eSiegel JH, Giovannini I, Coleman B. Ventilation:perfusion maldistribution secondary to the hyperdynamic cardiovascular state as the major cause of increased pulmonary shunting in human sepsis. J Trauma. 1979;19(6):432-60.\u003c/li\u003e\n \u003cli\u003eTomashefski JF Jr, Davies P, Boggis C, et al. The pulmonary vascular lesions of the adult respiratory distress syndrome. Am J Pathol. 1983;1983(112):112-26.\u003c/li\u003e\n \u003cli\u003eZapol WM, Snider MT. Pulmonary hypertension in severe acute respiratory failure. N Engl J Med. 1977;296(9):476-80.\u003c/li\u003e\n \u003cli\u003eGreene R, Zapol WM, Snider MT, et al. Early bedside detection of pulmonary vascular occlusion during acute respiratory failure. Am Rev Respir Dis. 1981; 124(5):593-601.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eLecompte-Osorio P, Pearson SD, Pieroni CH, et al. Bedside estimates of dead space using end-tidal CO2 are independently associated with mortality in ARDS. Crit Care. 2021;25(1):333. \u0026nbsp;\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eKallet RH, Zhuo H, Ho K, et al. Lung Injury Etiology and Other Factors Influencing the Relationship Between Dead-Space Fraction and Mortality in ARDS. Respir Care. 2017;62(10):1241-8.\u003c/li\u003e\n \u003cli\u003eWaltuck BL. The Bohr equation. Anesthesiology. 1970;32(5):472473.\u003c/li\u003e\n \u003cli\u003eLi Y, Li W, Xu B. Between blood glucose and mortality in critically ill patients: Retrospective analysis of the MIMIC-IV database. J Diabetes Investig. 2024;15(7):931-8.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 to 5 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"dead space fraction, acute respiratory failure, mortality, MIMIC-IV","lastPublishedDoi":"10.21203/rs.3.rs-7972231/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7972231/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: Few studies have investigated the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and mortality in critically ill patients with acute respiratory failure (ARF).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjective\u003c/strong\u003e: The primary objective of this study is to clarify the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and mortality in critically ill patients with ARF, as well as to determine its optimal predictive threshold for clinical application.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: This study was based on the MIMIC-IV database. The analysis using multivariable logistic regression, cox proportional hazards model, and restricted cubic splines (RCS) was used to explore the association between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and mortality.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: A total of 2,254 patients were included in the study, of whom 1,709 (75.82%) died during their ICU stay. Both multivariable logistic regression and cox proportional hazards model\u0026nbsp;identified V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e as the most significant predictor of ICU mortality (OR = 16.90, 95%CI: 7.38-38.69, P˂0.001; HR=9.60, 95%CI: 4.95-18.61, P˂0.001). After adjusting for covariates, the RCS analysis continued to demonstrate a significant overall association and a non-linear relationship between V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e and ICU mortality, with an identified inflection point at 0.222 for V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e. In addition, the results of the subgroup and stratified analyses were robust, and the univariate forest plot demonstrated that V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e exhibited an extremely strong association with ICU mortality, accompanied by an overall OR of 2.34 (95% CI: 1.91-2.86, p˂0.001).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e is positively associated with ICU mortality in critically ill patients with ARF. And the V\u003csub\u003eD\u003c/sub\u003e/V\u003csub\u003eT\u003c/sub\u003e of 0.222 may serve as a potential prognostic threshold for predicting clinical outcomes in critically ill patients with ARF.\u003c/p\u003e","manuscriptTitle":"Association between the dead space fraction and mortality in critically ill patients with acute respiratory failure: an analysis of the MIMIC-IV database","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-15 01:06:18","doi":"10.21203/rs.3.rs-7972231/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"bead4cd2-64fd-4b54-ac68-c123fb3417f2","owner":[],"postedDate":"December 15th, 2025","published":true,"recentEditorialEvents":[{"type":"decision","content":"Rejected","date":"2026-05-12T15:56:11+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-05-12T16:11:52+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-15 01:06:18","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7972231","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7972231","identity":"rs-7972231","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}