Inverse Association Between Preoperative Cardiac Output and Postoperative Kidney Function in Off-Pump Coronary Artery Bypass Grafting

Inverse Association Between Preoperative Cardiac Output and Postoperative Kidney Function in Off-Pump Coronary Artery Bypass Grafting | 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 Inverse Association Between Preoperative Cardiac Output and Postoperative Kidney Function in Off-Pump Coronary Artery Bypass Grafting Jiange Han, Zhao Zhang, Tao Wang, Yunfei Li, Wenqian Zhai, Jianxu Er, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8225284/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Background The relationship between cardiac output and kidney function following cardiac surgery remains poorly defined. We aimed to evaluate the association between preoperative cardiac output and postoperative acute kidney injury (AKI) in patients undergoing off-pump coronary artery bypass grafting (CABG), a setting without exposure to cardiopulmonary bypass. Methods This cohort study included 1,949 patients aged ≥ 60 years from the Bottomline-CS trial who underwent elective off-pump CABG. Preoperative cardiac output was measured under standardized resting conditions, 1–2 days prior to surgery. The primary outcome was AKI within the postoperative 7 days, defined according to KDIGO serum creatinine criteria. Propensity score-matched analyses were performed to compare the risk of AKI between patients with low and high baseline cardiac output. Restricted cubic spline models were used to examine the continuous relationship between cardiac output and the risk of AKI. The findings were validated in an external cohort to assess their robustness. Results Overall AKI incidence was 11% (213/1,949). In matched analyses, a preoperative cardiac index (CI) ≥ 3.0 L/min/m² was associated with higher AKI than CI < 3.0 (14.3% vs 9.0%; OR 1.69; 95% CI, 1.15–2.48). A similar association was observed for cardiac output ≥ 5.0 L/min versus < 5.0 (13.3% vs 8.9%; OR 1.56; 95% CI, 1.09–2.23). Spline analyses showed a J-shaped relationship, with rising risk above CI about 3.0 L/min/m² and cardiac output about 5.0 L/min. High-output patients had lower systemic vascular resistance with similar mean arterial pressure. Mediation analyses found no explanatory effect of intraoperative hypotension, vasopressor use, oxygenation deficits, fluid balance, or intraoperative cardiac output change. Findings were supported in an external cohort. Conclusions Higher, not lower, preoperative cardiac output is independently associated with increased risk of postoperative AKI in patients undergoing off-pump CABG. cardiac output acute kidney injury coronary artery bypass grafting off-pump surgery hemodynamics risk stratification hyperdynamic circulation propensity score matching Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Acute kidney injury (AKI) is a frequent complication after cardiac surgery, affecting roughly one in five to one in three patients, depending on case-mix and procedure [ 1 – 3 ]. Even small rises in serum creatinine are linked to greater morbidity, longer hospitalization, higher costs, and increased long-term mortality and progression of chronic kidney disease; when renal replacement therapy is required, in-hospital mortality is high [ 1 , 4 , 5 ]. The clinical and economic burden makes better physiological risk characterization a priority [ 1 ]. The pathogenesis of cardiac surgery-associated AKI is multifactorial. Proposed mechanisms include renal hypoperfusion, systemic inflammation, ischemia-reperfusion injury, neurohormonal activation, oxidative stress, hemolysis, and exposure to nephrotoxins, with vulnerability shaped by diabetes, anemia, pre-existing kidney disease, and age [ 1 , 6 , 7 ]. Current preventive approaches emphasize avoidance of nephrotoxins and maintenance of perfusion pressure and volume status within guideline-based hemodynamic care bundles, although their physiological underpinnings in surgical populations remain incompletely validated [ 7 ]. Prior evidence on the relation between cardiac output and AKI comes from different populations, including heart failure cohorts and clinical syndromes described as low output, not from cardiac surgery [ 1 , 8 – 10 ]. In those settings, low output is a composite clinical label that may include hypotension, poor peripheral perfusion, oliguria, and use of inotropes or vasopressors; therefore, cardiac output is usually inferred rather than measured directly [ 1 , 8 – 10 ]. Within cardiac surgery, this association remains uncharacterized. Interpretation in on-pump procedures is confounded by bypass-related factors such as hemolysis, embolization, and systemic inflammatory activation, whereas off-pump coronary artery bypass grafting limits these influences and offers a setting to examine the association with less confounding [ 7 ]. In this cohort study of patients undergoing off-pump CABG, we evaluated the association between preoperative (not intraoperative) cardiac output and postoperative AKI. A priori, we defined a lower pre-operative cardiac output as a predisposing factor, meaning a baseline physiological vulnerability present before anesthesia and surgery. This is distinct from precipitating factors that occur intra-operatively, such as hypotension, bleeding, or nephrotoxin exposure, which can directly trigger renal injury [ 7 ]. Our primary hypothesis was that lower pre-operative cardiac output, as a predisposing factor, would be associated with a higher incidence of post-operative AKI. Methods Study Design This cohort study is a secondary analysis of the Bottomline-CS trial (Better Outcome Through Tissue Oxygenation Monitoring Linked with INtErvention in Cardiac Surgery; ClinicalTrials.gov NCT04896736), a single-center, assessor-blinded, randomized controlled trial that enrolled patients undergoing elective off-pump CABG.[ 11 ] The original trial compared outcomes between patients who received perioperative care guided by multisite tissue oxygen saturation and advanced hemodynamic monitoring versus those managed with standard care. For this secondary analysis, we addressed a distinct research question: whether there is an association between preoperative cardiac output and postoperative AKI, which differs from the primary trial objective. Accordingly, the study design (observational cohort analysis vs. randomized trial), analytical approach, exposures, and outcomes are all substantively different from those of the parent trial. Although the original trial involved randomization and intervention, the present analysis treats the trial population as a single prospective cohort with uniformly measured baseline and intraoperative hemodynamic parameters, enabling an evaluation of physiologic predictors of AKI across the entire study sample. The Institutional Review Board of Tianjin Chest Hospital approved this secondary analysis on February 1, 2024, with a waiver of informed consent due to the use of de-identified data. The reporting adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cohort studies. Setting The Bottomline-CS trial was conducted at Tianjin Chest Hospital, a high-volume tertiary academic center affiliated with Tianjin University, China. As a nationally recognized referral hospital with advanced cardiovascular and pulmonary care infrastructure, the institution provides comprehensive surgical services for a diverse patient population across Northern China. Participants Between June 8, 2021, and December 27, 2023, consecutive patients aged ≥ 60 years scheduled for elective off-pump CABG at Tianjin Chest Hospital were prospectively screened and enrolled. Exclusion criteria included inability or unwillingness to provide informed consent, need for preoperative ventilatory support (invasive or non-invasive), presence of an external cardiac assist device, requirement for urgent or emergent surgery, or an expected life expectancy of less than 30 days. Eligible patients were identified from daily surgical schedules and approached at least 24 hours before surgery for consent. Enrolled participants were randomized to receive either guided perioperative management, based on multisite tissue oxygen saturation and advanced hemodynamic monitoring, or standard care, in which the same monitoring data were collected but not disclosed to the clinical team. All patients underwent uniform perioperative data collection, irrespective of the randomization group. Data acquisition spanned the entire perioperative course, from baseline evaluation through postoperative day 30, and included detailed physiologic measurements, clinical outcomes, and safety assessments. Outcome assessors were blinded to group assignment, ensuring unbiased evaluation. This standardized approach enabled consistent exposure measurement and robust outcome ascertainment across the cohort. Perioperative Care All patients received comprehensive intraoperative monitoring, including pulse oximetry, end-tidal carbon dioxide monitoring, electrocardiography, non-invasive blood pressure, and invasive arterial pressure via a radial artery catheter. Central venous pressure was measured through a catheter inserted into the internal jugular vein. Transesophageal echocardiography was used when clinically indicated. General anesthesia was induced with either propofol or etomidate and maintained with propofol infusion and/or sevoflurane inhalation at the discretion of the attending anesthesiologist. Neuromuscular blockade was achieved using either cisatracurium or rocuronium. All patients were endotracheally intubated and mechanically ventilated, with sufentanil administered as the primary intraoperative analgesic in accordance with institutional standards. Throughout surgery, continuous hemodynamic monitoring was performed using a minimally invasive arterial waveform analysis system (Masimo LiDCO Hemodynamic Monitoring System, Masimo, Irvine, California, USA), which estimated cardiac output, stroke volume, and systemic vascular resistance from the radial arterial pressure waveform. Concurrently, multisite tissue oxygen saturation was assessed using a near-infrared spectroscopy system (Nonin Medical, Inc., Plymouth, Minnesota, USA). Cerebral oxygenation was measured at the left and right forehead, and somatic oxygenation at the forearm brachioradialis. Together, these modalities provided continuous, real-time evaluation of systemic and regional perfusion throughout the procedure. Outcomes The primary outcome was postoperative acute kidney injury (AKI), defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) Clinical Practice Guidelines.[ 12 ] In accordance with the previously published research, urine output criteria were not used due to the frequent use of diuretics and incomplete documentation, which can limit diagnostic reliability.[ 13 ] AKI was identified by postoperative increases in serum creatinine relative to preoperative baseline values. Specifically, an increase in serum creatinine of ≥ 0.3 mg/dL within 48 hours, or a ≥ 50% increase from baseline within 7 days, was considered diagnostic of AKI. Baseline creatinine was defined as the most recent value obtained prior to surgery, and postoperative creatinine was defined as the highest value recorded within the first seven days after surgery.[ 13 ] Exposures The primary exposure in this study was baseline cardiac output, measured one to two days prior to surgery using a noninvasive hemodynamic monitoring system. Assessments were performed in a controlled hospital setting with patients in a supine position, awake, calm, with eyes closed, and breathing either room air or their usual home oxygen. Cardiac output, stroke volume, and systemic vascular resistance were estimated using a finger and arm cuff–based monitor (Continuous Noninvasive Arterial Pressure [CNAP™] Module, Masimo LiDCO™ Hemodynamic Monitoring System; Masimo, Irvine, California, USA). This preoperative monitoring approach differs from the minimally invasive arterial waveform analysis employed intraoperatively, enabling physiologic characterization under resting, baseline conditions. Potential Confounders In this cohort study, potential confounders were defined as variables that may be associated with both the exposure (baseline cardiac output measured one to two days prior to surgery) and the outcome (AKI within the first seven postoperative days) without residing in the causal pathway. According to established epidemiologic principles, confounders must precede the exposure, be associated with the outcome, and not be a result of the exposure itself. In contrast, variables on the causal path between exposure and outcome are considered mediators and were addressed separately to avoid overadjustment. The inclusion of mediators or variables affected by the exposure could bias effect estimates by distorting or attenuating the relationship of interest. To preserve the temporal structure and causal interpretability of our analysis, we limited the set of potential confounders to demographic and preoperative characteristics. Intraoperative and postoperative variables were intentionally excluded, as they could be influenced by preoperative cardiac output and, if adjusted for, may introduce collider stratification bias. This decision aligns with the conceptual model that views baseline cardiac output as a determinant, rather than a consequence, of perioperative events. Candidate confounders were selected based on prior literature, clinical reasoning, and their observed associations with exposure and outcome in univariate analyses. These included age, sex, height, body mass index, smoking and alcohol history, prior myocardial infarction, arrhythmia, diabetes, hypertension, and carotid artery disease. The overall comorbidity burden was captured using the age-adjusted Charlson comorbidity index. Preoperative medication use, specifically beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, calcium channel blockers, antidiabetic agents, and diuretics, was also incorporated, given their potential influence on both cardiac function and renal risk. Relevant physiologic and laboratory parameters, such as mean arterial pressure, estimated glomerular filtration rate, and C-reactive protein, were included in addition to comorbid and pharmacologic variables. To account for any residual influence from the original trial design, group assignment in the Bottomline-CS trial was also treated as a covariate. This comprehensive set of baseline covariates was selected to minimize bias and isolate the independent association between cardiac output and postoperative AKI. Potential Mediators To explore the mechanisms underlying the association between preoperative cardiac output and postoperative AKI, we identified intraoperative variables that may function as mediators, i.e., physiologic processes influenced by cardiac output that, in turn, could contribute to renal injury. These variables were evaluated separately from baseline confounders to preserve the integrity of causal inference. Candidate mediators included intraoperative hemodynamic disturbances, oxygenation deficits, vasopressor use, fluid balance, and changes in cardiac output during surgery. Hemodynamic and oxygenation abnormalities were quantified using the area under the curve (AUC) method, which reflects the cumulative duration and magnitude of deviations below predefined thresholds, specifically 10%, 20%, and 30% reductions from ward-based baseline values measured 1–2 days prior to surgery. Vasopressor exposure was defined as the use of vasoactive agents administered to at least 20% but no more than 80% of patients, excluding agents used either rarely or ubiquitously. Fluid balance was assessed by measuring the total intraoperative input (crystalloids, colloids, and autologous blood products) and estimated blood loss. Intraoperative changes in cardiac output were captured using two metrics: (1) relative change in cardiac index (minimum intraoperative value minus preoperative baseline, divided by baseline), and (2) absolute change (direct difference between minimum intraoperative and baseline values). These variables were excluded from the primary analysis to preserve the estimation of the total effect of baseline cardiac output on AKI, avoiding bias introduced by conditioning on post-exposure variables. Instead, they were examined using formal mediation analyses to assess whether intraoperative physiologic disturbances partially explained the relationship between baseline cardiac output and the development of AKI. Data Sources All data for this cohort study were derived from the Bottomline-CS trial, with methodological details previously reported in the published protocol and manuscript. Data collection followed a standardized approach across all phases, preoperative, intraoperative, and postoperative, ensuring internal consistency and minimizing measurement variability. Preoperative data were collected one to two days prior to surgery by trained research staff, who conducted in-person evaluations and physiological measurements in a controlled hospital environment. These assessments included baseline hemodynamic and tissue oxygenation metrics, which were recorded while participants were in a supine, resting state. Intraoperative data, including both continuous and event-based measurements, were collected in real time by designated research team members responsible for recording discrete intraoperative variables and ensuring the completeness of automated monitoring data. Postoperative outcome data were obtained by independent assessors blinded to group assignments and uninvolved in preoperative and intraoperative care, thereby preserving objectivity in outcome ascertainment. Although the original trial involved randomization into two treatment arms, data acquisition was carried out uniformly for all participants in accordance with the predefined protocol, allowing for consistent and unbiased observational analysis across the entire cohort. Bias Multiple strategies were employed to minimize bias across study design, data collection, and analysis. The study protocol was finalized prior to patient enrolment, thereby eliminating the possibility of post hoc design modifications. Eligible patients were identified consecutively from the surgical schedule using predefined criteria, reducing selection bias. Preoperative data, including baseline hemodynamic and tissue oxygenation measurements, were collected by trained personnel not involved in intraoperative care or outcome assessment. Postoperative outcomes were evaluated by assessors blinded to group allocation, thereby minimizing observer bias. Standardized data acquisition procedures, real-time digital entry, and independent third-party audits further reduced the risk of information bias. All physiological measurements were performed using validated devices with prespecified thresholds applied uniformly across study groups. Selection bias and attrition were minimal, as evidenced by the inclusion of 1,960 patients with only 8 missing outcome records (0.4%). Confounding was addressed through the careful selection of clinically and biologically relevant preoperative variables, which were incorporated into a propensity score matching framework to achieve covariate balance without introducing over-adjustment from intraoperative or postoperative factors. Although these measures strengthen internal validity, residual confounding and limited generalizability inherent to single-center observational designs cannot be entirely excluded. Study Size The study size was determined by the number of eligible participants enrolled in the Bottomline-CS trial who met the criteria for this secondary analysis. All patients aged 60 years or older undergoing elective off-pump coronary artery bypass grafting were considered, and those with cancelled surgeries or missing data for AKI classification were excluded. This yielded a final analytical cohort of 1,949 patients. No additional sample size calculation was performed for this observational analysis, as the entire available cohort was included to maximize statistical precision and support the planned propensity score matching approach. Quantitative Variables Continuous variables were assessed for distributional properties using histograms and Q-Q plots. Those following an approximately normal distribution were summarized as means with standard deviations (SD), whereas skewed variables were reported as medians with interquartile ranges (IQR). Categorical variables were presented as frequencies with percentages. For primary analyses, baseline cardiac output was dichotomized using clinically relevant thresholds to facilitate interpretability and support propensity score matching, while preserving the physiological distinction between lower and higher output states. This approach enabled a meaningful evaluation of the relationship between preoperative cardiac output and the risk of AKI in the matched cohorts. Missing Data Missing data were addressed using multiple imputations by fully conditional specification, generating 50 imputed datasets to ensure the stability of estimates. All primary analyses were conducted using the pooled results from the imputed datasets. Statistical Analysis The primary objective of this cohort study was to determine whether preoperative cardiac output is independently associated with the development of AKI. Specifically, we tested the hypothesis that baseline cardiac output levels stratify patients into different risk categories for developing AKI after off-pump coronary artery bypass grafting. Threshold Determination and Exposure Definition To identify a clinically meaningful threshold for cardiac output, we systematically evaluated odds ratios (ORs) for AKI across a range of cutoffs. ORs were calculated across thresholds incremented by 0.1 L/min or 0.1 L/min/m 2 for cardiac index. These were plotted to visualize the relationship between baseline cardiac output thresholds and the risk of AKI. Thresholds were selected based on both statistical significance in the OR-threshold plots and clinical usability. The primary exposure was defined using the selected threshold, with patients categorized into high- or low-cardiac-output groups accordingly. The main analysis utilized cardiac index (cardiac output normalized to body surface area) to account for body size, whereas sensitivity analyses employed absolute cardiac output values. Propensity Score Matching To mitigate confounding, we employed propensity score matching. Propensity scores were derived from logistic regression models incorporating preoperative variables known or suspected to be associated with both cardiac output and AKI. These included demographics, comorbidities, medication use, laboratory values, and randomization group from the parent trial. Nearest-neighbor matching without replacement was used, with a variable matching ratio (1:1 to 1:4) to optimize sample size while maintaining balance. A caliper width of 0.2 standard deviations of the logit of the propensity score was applied. Covariate balance was assessed using absolute standardized differences, with values ≤ 0.1 indicating adequate balance. The incidence of AKI and other categorical outcomes was compared using conditional logistic regression. Linear mixed-effects models were used for continuous outcomes, with matching incorporated as a random intercept. ORs with 95% confidence intervals (CIs) were used to quantify associations between cardiac output and postoperative AKI. Nonlinear Association Analysis To assess the shape of the relationship between cardiac output and AKI, restricted cubic spline regression was performed using continuous cardiac index or cardiac output as predictors. Models were adjusted for the same covariates used in propensity score estimation. These analyses enabled the detection of nonlinear associations, complementing the threshold-based approach. Intraoperative Hemodynamic Trajectory Analysis To determine whether preoperative differences in cardiac output were sustained intraoperatively, we analysed cardiac output trajectories in propensity score–matched cohorts. Measurements before and after anesthesia induction were compared, along with intraoperative median and minimum cardiac index values. Minimum values were calculated by smoothing cardiac index measurements (10-second rolling medians) and extracting the median of the ten lowest values (Supplementary Fig. 1). Between-group differences were assessed using linear mixed-effects models to account for unequal sample sizes and repeated measures. Mediation Analysis To explore whether intraoperative variables mediated the relationship between baseline cardiac output and postoperative AKI, we conducted mediation analyses using a two-step approach.[ 14 ] First, potential mediators, including intraoperative hemodynamic variables, oxygenation parameters, vasopressor use, fluid balance, and cardiac output changes, were modelled using regression analyses adjusted for baseline covariates. Second, logistic regression was used to assess associations between baseline cardiac index, mediators, and AKI. Continuous mediators were standardized (z-scores). Mediation effects were quantified by estimating the average causal mediation effect, average direct effect, total effect, and proportion mediated, using bootstrap resampling (1,000 iterations). A 95% CI excluding zero was considered statistically significant. External Validation To assess generalizability, we replicated the primary analysis in an external cohort drawn from the MIMIC-IV v3.1 database.[ 15 ] This dataset included patients who underwent cardiac procedures with documented preoperative cardiac output and postoperative AKI outcomes. Analyses used the same exposure thresholds and outcome definitions as the primary cohort. Software and Statistical Significance All analyses were performed using R version 3.5.3 (R Foundation for Statistical Computing, Vienna, Austria). Key R packages included: readxl, dplyr, tableone, mice, autoReg, ggplot2, Matching, mediation, survival, and rms. Visualizations were created in Python 3.8.10 using NumPy, Pandas, Matplotlib, Seaborn, and Statsmodels. A two-sided p-value < 0.05 was considered statistically significant. Results Participants The Bottomline-CS randomized controlled trial enrolled a total of 1,960 patients scheduled for elective off-pump CABG between June 8, 2021, and December 27, 2023. For the current cohort study, 11 patients were excluded due to cancelled surgery (n = 5), age younger than 60 years (n = 3), and missing postoperative AKI data (n = 3), resulting in an analytic cohort of 1,949 patients. A detailed flowchart illustrating patient selection criteria and analysis plan is provided in Fig. 1. Cohort Characteristics The study included 1,949 patients with a mean age of 69 ± 5 years; 70% of participants were male, and the mean body mass index was 25 ± 3 kg/m 2 . Preoperative hemodynamic measures revealed a mean cardiac output of 4.5 ± 1.5 L/min and a cardiac index of 2.5 ± 0.8 L/min/m 2 . The mean arterial pressure before surgery was 90 ± 12 mmHg. AKI occurred in 11% (213/1949) of patients. Comprehensive preoperative baseline characteristics, including demographics, medical history, medication use, cognitive assessments, cardiac measurements, hemodynamic data, and laboratory results, are presented in Supplementary Table 1. Preoperative Baseline Cardiac Output Threshold Determination The relationship between preoperative baseline cardiac output thresholds measured at the patient ward 1–2 days prior to surgery and the unadjusted OR of postoperative AKI is presented in Fig. 2. As the threshold for cardiac output increased, the OR for AKI correspondingly increased. Based on this analysis and clinical relevance, we selected a cardiac index threshold of 3.0 L/min/m 2 for the primary analysis. Additionally, thresholds based on a cardiac index of 3.5 L/min/m 2 and absolute cardiac output of 5.0 L/min and 6.0 L/min were chosen for sensitivity analyses to evaluate the robustness of the association. Propensity Score-Matched Analysis of AKI Risk in Patients with High vs. Low Baseline Cardiac Output In the primary analysis, patients were divided into two groups based on a preoperative cardiac index threshold of 3.0 L/min/m 2 : those with a cardiac index of less than 3.0 L/min/m 2 and those with a cardiac index of 3.0 L/min/m 2 or greater. Preoperative baseline characteristics for these two groups are detailed in Supplementary Table 2. Propensity score matching was performed to balance baseline covariates between groups, using variables including age, sex, height, body mass index, smoking status, alcohol consumption, history of myocardial infarction, arrhythmia, diabetes, hypertension, carotid artery disease, age-adjusted Charlson comorbidity index, medications (i.e., beta-blockers, angiotensin-converting enzyme inhibitors, calcium channel blockers, angiotensin receptor blockers, antidiabetic treatment, and diuretics), mean arterial pressure, C-reactive protein, estimated glomerular filtration rate, and randomization group assignment in the Bottomline-CS trial (Table 1). The outcomes of propensity score matching are summarized in Figs. 3A and 3B and Supplementary Table 3. Post-matching absolute standardized differences for all covariates were below 0.1, indicating adequate balance (Fig. 3B). After matching, the mean cardiac index was 2.2 ± 0.8 L/min/m 2 in the lower cardiac index group (n = 942) and 3.6 ± 0.5 L/min/m 2 in the higher cardiac index group (n = 314). The incidence of AKI was significantly higher among patients with a higher baseline cardiac index compared to those with a lower level (14.3% [45/314] vs. 9.0% [85/942]; OR, 1.69; 95% CI, 1.15–2.48). Additional sensitivity analyses using alternative thresholds further supported this finding. Specifically, analyses based on cardiac index thresholds of 3.5 L/min/m 2 (Fig. 3C, Supplementary Table 4), and absolute cardiac output thresholds of 5.0 L/min (Fig. 3D, Supplementary Table 5) and 6.0 L/min (Fig. 3E, Supplementary Table 6), consistently demonstrated that patients with higher baseline cardiac outputs were at significantly greater risk of developing postoperative AKI (Fig. 3A and Supplementary Table 7). Nonlinear Association Between Baseline Cardiac Output and Risk of Postoperative AKI Restricted cubic spline analyses revealed a nonlinear, J-shaped relationship between baseline cardiac output and the risk of postoperative AKI. As shown in Fig. 4A, the odds of AKI began to increase when the baseline cardiac index exceeded approximately 2.0 L/min/m², with a steeper rise beyond 3.0 L/min/m 2 . A similar pattern was observed for absolute cardiac output (Fig. 4B), where the odds ratio started to rise above a cardiac output of roughly 4.0 L/min and became significantly elevated beyond 5.0 L/min. This pattern supports the presence of a threshold effect and reinforces the use of 3.0 L/min/m 2 and 5.0 L/min as clinically meaningful cutoffs for cardiac index and cardiac output, respectively, in risk stratification. Preoperative and Intraoperative Hemodynamic Patterns in Patients with Distinct Baseline Cardiac Output Levels Using propensity score-matched patients (Supplementary Table 5), we compared hemodynamic parameters between groups stratified by a baseline cardiac output threshold of 5.0 L/min. As expected, baseline cardiac output clearly separated the two groups with no overlap (Fig. 5A). Systemic vascular resistance also differed significantly between groups, though values partially overlapped (P < 0.001, Fig. 5B). In contrast, mean arterial pressures were comparable between groups (P = 0.083, Fig. 5C), as anticipated, given that mean arterial pressure was one of the covariates included in the propensity score model. To further explore the relationship between cardiac output and systemic vascular resistance, we generated a scatterplot using the same matched patients (Fig. 5D). An iso-pressure reference line corresponding to a mean arterial pressure of 90 mmHg was overlaid.[ 16 ] The distribution of data points along this line demonstrated that for a given mean arterial pressure, the underlying cardiac output and systemic vascular resistance varied widely among different patients, reflecting the physiologic heterogeneity and compensatory balance between flow and resistance. Cardiac output (rather than cardiac index) was used in these analyses (Figs. 5A–5D) to directly correspond with the following physiologic equation in which cardiac output, not cardiac index, has been conventionally used:​ $$\:Mean\:Arterial\:Pressure=\frac{Cardiac\:Output\times\:Systemic\:Vascular\:Resistance}{80}$$ We then evaluated intraoperative cardiac index trajectories in matched groups stratified by a baseline cardiac index threshold of 3.0 L/min/m 2 (Supplementary Table 3 and Fig. 5E). The cardiac index increased prior to anesthesia induction, likely reflecting anxiety-induced sympathetic activation, and decreased significantly following induction. Throughout the intraoperative period, cardiac index remained consistently higher in patients with a baseline index ≥ 3.0 L/min/m 2 . Specifically, the intraoperative median cardiac index was 2.6 ± 0.6 L/min/m² in the low index group and 2.7 ± 0.7 L/min/m 2 in the high index group (P < 0.001). Minimum intraoperative cardiac index values were 1.5 ± 0.4 L/min/m 2 and 1.6 ± 0.4 L/min/m 2 (P < 0.05), respectively. These findings suggest that preoperative differences in cardiac index persist throughout surgery, despite a remarkable decrease in these differences. Mediation analyses To explore potential mechanisms underlying the association between preoperative cardiac output and postoperative AKI, we conducted mediation analyses focusing on intraoperative variables. These analyses consistently demonstrated a significant total effect of preoperative cardiac output on postoperative AKI yet revealed minimal or negligible indirect effects through intraoperative mediators (Supplementary Tables 8–10). Specifically, variables such as intraoperative hemodynamics, oxygenation metrics, vasopressor use, fluid balance, and decreases in cardiac output did not meaningfully mediate the relationship. This pattern persisted across all modelled pathways and sensitivity analyses. Collectively, these findings suggest that the observed association between higher baseline cardiac output and increased risk of AKI is primarily attributable to direct effects, rather than being mediated by intraoperative physiological disturbances. External Validation in Patients Undergoing Cardiac Procedures To evaluate the generalizability of our findings, we conducted an exploratory analysis using data from the MIMIC-IV v3.1 dataset. We identified 748 adult patients who underwent various cardiac procedures and had pre-procedure cardiac output measurements available (Supplementary Table 11). These procedures encompassed a broad range of cardiac interventions. Thermodilution via pulmonary artery catheter was the predominant method used to measure cardiac output (79.1%), followed by non-invasive bioreactance-based technology (18.2%), arterial pressure waveform analysis (1.5%), and transpulmonary thermodilution with pulse contour analysis (1.2%) (Supplementary Table 12). The heterogeneity of procedures and measurement modalities in the external cohort strengthens generalizability, offering more compelling validation than would be achieved in a uniform population. Using a preoperative cardiac index threshold of 3.0 L/min/m 2 , consistent with the primary analysis, 582 patients were classified as having a low cardiac index and 166 as having a high cardiac index. The incidence of postoperative AKI was 18.8% (109/582) in the low cardiac index group and 27.7% (46/166) in the high cardiac index group. Unadjusted analysis showed that patients with a high preoperative cardiac index had significantly increased odds of developing AKI (OR, 1.66; 95% CI, 1.11–2.47; p = 0.012) (Supplementary Table 13). Due to substantial missingness, i.e., more than 50%, in key covariates used for propensity score modeling, we did not perform matched analyses in this cohort (Supplementary Table 14). Nevertheless, the direction and magnitude of the association were consistent with our findings based on the Bottomline-CS cohort. Importantly, patients in the low cardiac index group were at a higher risk of postoperative complications: they were older (mean age, 69 vs. 60 years), more likely to be male (45% vs. 34%), and had higher rates of myocardial infarction (34% vs. 19%), diabetes (40% vs. 30%), hypertension (79% vs. 57%), and greater comorbidity burden, as indicated by the age-adjusted Charlson Comorbidity Index (median 6 vs. 5) (Supplementary Table 15). Therefore, the findings from the MIMIC-IV cohort, although unadjusted, lend credibility to the observed association, as patients with lower baseline cardiac index, despite their higher comorbidity burden, demonstrated a lower incidence of AKI. Discussion In this large, prospective cohort study of patients undergoing elective off-pump coronary artery bypass grafting, we demonstrated that higher preoperative cardiac output is independently associated with a significantly greater risk of postoperative AKI. This association persisted across multiple analytic strategies, including propensity score matching and sensitivity analyses using alternative thresholds. Importantly, this unexpected finding was supported by an independent validation cohort derived from the MIMIC-IV v3.1 database. These results offer a novel and clinically important contribution to our understanding of perioperative hemodynamics and renal risk, and they challenge the conventional assumption that greater cardiac output necessarily implies improved renal perfusion and protection against AKI. To our knowledge, this is the first study to systematically investigate and demonstrate an inverse association between preoperative cardiac output and postoperative AKI in patients undergoing elective cardiac surgery. While reduced cardiac output has traditionally been viewed as a risk factor for renal hypoperfusion and injury, our findings reveal a paradoxical relationship: patients with higher resting cardiac output, measured under tightly standardized preoperative conditions, had a significantly higher risk of developing AKI following surgery. This counterintuitive observation necessitates a re-evaluation of long-standing physiological assumptions and suggests a new potential marker for perioperative risk stratification. Challenging a Traditional Paradigm Historically, the prevailing pathophysiologic paradigm has posited that diminished cardiac output leads to decreased renal blood flow and perfusion pressure, triggering ischemic injury and contributing to the development of AKI.[ 17 – 19 ] This model is particularly prominent in the heart failure literature, where low-output states are commonly associated with renal impairment. Numerous studies on patients with acute and chronic decompensated heart failure have highlighted reduced forward flow as a contributor to worsening renal function and adverse outcomes.[ 6 , 9 , 20 – 22 ] Based on this premise, higher cardiac output would be expected to confer a protective effect on renal function. However, our findings directly contradict this assumption. In this cohort of well-characterized, hemodynamically stable patients undergoing elective, off-pump CABG, those with higher cardiac output exhibited increased, rather than decreased, rates of postoperative AKI. This association remained evident despite balanced baseline characteristics, including mean arterial pressure, across comparison groups. Cardiac output was measured 24 to 48 hours prior to surgery under standardized, resting conditions, with patients in a supine position, calm, awake, breathing room air, and free from sedation or anesthesia. This provided a reliable snapshot of baseline physiology, independent of procedural or pharmacological influences. These data suggest that higher cardiac output in this context may not reflect superior cardiovascular function or more effective perfusion, but rather a maladaptive physiologic state. Potential explanations include systemic vasodilation, which requires compensatory increases in cardiac output to maintain perfusion pressure, impaired renal autoregulatory responses, or neurohormonal activation reflective of underlying physiologic stress. The fact that blood pressure was similar between groups further supports the interpretation that it is not perfusion pressure alone, but rather the broader hemodynamic configuration, that may predispose patients to AKI. Importantly, caution is warranted when interpreting these findings in the context of prior studies. Unlike earlier investigations that were primarily based on patients with heart failure, our study population consisted of elective surgical patients without decompensated cardiac function. The mechanisms linking cardiac output to renal outcomes in heart failure, where forward flow is pathologically reduced and venous congestion plays a dominant role, may not fully translate to the perioperative setting of preserved cardiac function and controlled hemodynamics. Nonetheless, the paradoxical association we observed invites a re-evaluation of how cardiac output is conceptualized in relation to renal risk across diverse clinical populations. Hemodynamic and Mechanistic Insights To better understand this relationship, we examined the distribution of systemic vascular resistance and cardiac output in relation to mean arterial pressure. Our scatterplots revealed that, for comparable mean arterial pressures, patients exhibited a broad range of cardiac output and systemic vascular resistance combinations (Fig. 5D). This finding reinforces the concept that mean arterial pressure alone is insufficient to characterize hemodynamic state or predict renal perfusion adequacy. Notably, patients with higher cardiac output tend to exhibit lower systemic vascular resistance, supporting the hypothesis that high-output states may reflect compensatory responses to peripheral vasodilation, which in turn may affect renal perfusion heterogeneity or glomerular filtration dynamics. Despite exploring a range of potential mediating factors, including intraoperative hypotension, cardiac index trajectories, tissue desaturation indices, vasoactive drugs, and fluids, we were unable to identify a definitive explanation for the observed association. Importantly, the cardiac output-AKI relationship appeared to be preoperatively determined and was not explained by intraoperative events. These findings underscore the complexity of the interplay between flow, pressure, resistance, and renal autoregulation and suggest that resting preoperative cardiac output may serve as a physiologic marker of latent renal vulnerability. Consistency with Heart Failure Literature Although our study is the first to demonstrate this association in the context of cardiac surgery, our findings are consistent with several investigations in heart failure populations, which have questioned the traditional model linking reduced cardiac output with renal dysfunction. For example, in patients diagnosed with heart failure undergoing pulmonary artery catheterization, no positive correlation was found between cardiac index and renal function.[ 23 ] On the contrary, a weak but statistically significant inverse relationship was observed, such that a higher cardiac index was associated with a lower estimated glomerular filtration rate.[ 23 ] Additional studies have similarly reported a lack of correlation, or even paradoxical associations, between cardiac output and renal function in patients with advanced heart failure.[ 24 – 29 ] In one study of patients undergoing right heart catheterization for decompensated heart failure, central venous pressure, but not cardiac output, correlated with renal function.[ 27 ] Others have found that passive venous congestion may play a more critical role than forward flow in determining renal outcomes.[ 29 ] Collectively, these data suggest that renal dysfunction may be more closely tied to pressure dynamics, venous outflow resistance, or neurohormonal dysregulation than to absolute cardiac output. Our study extends these insights to a new clinical context: elective cardiac surgery. Unlike heart failure cohorts, our patients were not in decompensated states and underwent off-pump procedures with proactive, prospective hemodynamic monitoring. Thus, our results provide novel evidence that the non-linear or inverse relationship between cardiac output and AKI risk may also apply to stable, surgical populations. Validation in an External Cohort The external validation of our findings using the MIMIC-IV database adds credibility and reproducibility to our results. In the diverse cohort of 748 patients undergoing cardiac procedures with available cardiac output data, we observed a similar trend: patients with a pre-procedure cardiac index of ≥ 3.0 L/min/m² had a significantly higher incidence of post-procedure AKI compared to those with a lower cardiac index. Although propensity score matching was not feasible due to missing data, the direction and magnitude of the effect were consistent with our primary analysis. Moreover, the patients in the lower cardiac index group of the MIMIC cohort tended to be older and carried a greater comorbidity burden, including a higher prevalence of myocardial infarction, diabetes, hypertension, and elevated Charlson comorbidity scores. Despite these higher baseline risks, they experienced lower rates of AKI, suggesting that higher cardiac output may reflect a more complex or maladaptive physiology not fully captured by conventional risk stratifications. Clinical Implications These findings carry important clinical implications. First, they challenge the traditional paradigm that greater cardiac output is invariably beneficial for renal perfusion. Second, they suggest that preoperative cardiac output, when measured under resting conditions, may serve as a valuable tool for risk stratification in AKI following cardiac surgery. Third, our results underscore the limitations of using mean arterial pressure alone to assess renal perfusion risk, highlighting the need for a more integrated hemodynamic assessment. Given that AKI is a major contributor to perioperative morbidity, prolonged hospitalization, and long-term adverse outcomes, improving prediction and prevention strategies remains a clinical priority. Incorporating cardiac output into risk assessment models could enhance predictive accuracy and inform preoperative optimization strategies. Moreover, our findings prompt a re-evaluation of whether interventions aimed at increasing cardiac output, such as the use of inotropic agents or fluid loading, are universally appropriate in the perioperative period, especially in patients who already exhibit high-output states. Strengths and Limitations This study has several notable strengths. It utilized high-quality, prospectively collected hemodynamic data acquired under standardized, controlled conditions. Cardiac output was measured with patients in a supine position, breathing room air, awake, and calm, minimizing physiologic variability and reducing the risk of confounding. Propensity score matching allowed for rigorous adjustment of baseline characteristics, and the consistency of findings across multiple thresholds and analytic strategies enhanced the internal validity of the observed association. Furthermore, external validation using an independent cohort supports the broader applicability of our results. Nonetheless, several limitations should be acknowledged. First, as an observational cohort study, the potential for residual confounding cannot be completely ruled out, despite robust statistical adjustments. Second, while cardiac output was measured preoperatively in a consistent manner, intraoperative hemodynamic monitoring employed different methodologies, and we were unable to establish causality or delineate precise mechanistic pathways. Third, we were unable to validate our findings in an external cohort with an identical surgical population and covariate set, largely due to the rarity of routine preoperative cardiac output measurement in non-research settings. This limitation highlights a broader challenge in integrating physiologic data into large-scale clinical datasets, underscoring the need for further research in comparable populations. Future Directions Further research is needed to clarify the mechanisms linking higher preoperative cardiac output to increased risk of AKI. Prospective studies that incorporate simultaneous assessments of renal perfusion, oxygenation, and biomarkers of tubular injury could help elucidate the underlying pathophysiology and identify potential mediators. In addition, randomized trials evaluating hemodynamic management strategies in patients with elevated cardiac output may help determine whether targeted modulation of flow or vascular resistance can mitigate the risk of AKI. Broader implementation of non-invasive cardiac output monitoring during preoperative assessment could also enable more accurate, physiology-based risk stratification and inform individualized perioperative care. Conclusion In conclusion, this study provides the first evidence that higher preoperative cardiac output is associated with an increased risk of AKI after elective off-pump CABG. This counterintuitive finding challenges conventional assumptions and introduces a novel physiologic marker of renal risk. By integrating high-fidelity hemodynamic measurements, robust statistical analyses, and external validation, our work advances the understanding of AKI pathophysiology and offers a foundation for future research into personalised perioperative hemodynamic management. These findings inform a more nuanced approach to assessing and mitigating AKI risk in cardiac surgical patients. Declarations Funding: This study was funded by the Tianjin Science and Technology Project (No. 20JCZDJC00810). Availability of data, code, and other materials: Patient-level data will be available collaboratively with approval by the corresponding authors. Ethics approval and consent to participate: This secondary analysis was approved by the Institutional Review Board of Tianjin Chest Hospital (approval date: February 1, 2024), and informed consent was waived due to the use of de-identified data. Consent for publish: Not applicable. Competing interests: The authors declare no competing interests. Clinical trial number: Not applicable (secondary analysis of the Bottomline-CS trial; original registration: NCT04896736). References Cheruku SR, Raphael J, Neyra JA, Fox AA. Acute Kidney Injury after Cardiac Surgery: Prediction, Prevention, and Management. Anesthesiology. 2023;139(6):880–98. Karkouti K, Wijeysundera DN, Yau TM, Callum JL, Cheng DC, Crowther M, Dupuis J-Y, Fremes SE, Kent B, Laflamme C, et al. Acute Kidney Injury After Cardiac Surgery. Circulation. 2009;119(4):495–502. O’Neal JB, Shaw AD, Billings FT. Acute kidney injury following cardiac surgery: current understanding and future directions. Crit Care. 2016;20(1):187. Leacche M, Winkelmayer WC, Paul S, Lin J, Unic D, Rawn JD, Cohn LH, Byrne JG. Predicting survival in patients requiring renal replacement therapy after cardiac surgery. Ann Thorac Surg. 2006;81(4):1385–92. Corredor C, Thomson R, Al-Subaie N. Long-Term Consequences of Acute Kidney Injury After Cardiac Surgery: A Systematic Review and Meta-Analysis. J Cardiothorac Vasc Anesth. 2016;30(1):69–75. Ronco C, Bellomo R, Kellum JA. Acute kidney injury. Lancet. 2019;394(10212):1949–64. Zarbock A, Küllmar M, Ostermann M, Lucchese G, Baig K, Cennamo A, Rajani R, McCorkell S, Arndt C, Wulf H, et al. Prevention of Cardiac Surgery–Associated Acute Kidney Injury by Implementing the KDIGO Guidelines in High-Risk Patients Identified by Biomarkers: The PrevAKI-Multicenter Randomized Controlled Trial. Anesth Analgesia. 2021;133(2):292–302. Ronco C, Haapio M, House AA, Anavekar N, Bellomo R. Cardiorenal Syndrome. J Am Coll Cardiol. 2008;52(19):1527–39. Liang KV, Williams AW, Greene EL, Redfield MM. Acute decompensated heart failure and the cardiorenal syndrome. Crit Care Med 2008, 36(1). Sun S, Ma F, Li Q, Bai M, Li Y, Yu Y, Huang C, Wang H, Ning X. Risk model for deaths and renal replacement therapy dependence in patients with acute kidney injury after cardiac surgery. Interact Cardiovasc Thorac Surg. 2017;25(4):548–54. Han J, Zhai W, Wu Z, Zhang Z, Wang T, Ren M, Liu Z, Sessler DI, Guo Z, Meng L, et al. Care guided by tissue oxygenation and haemodynamic monitoring in off-pump coronary artery bypass grafting (Bottomline-CS): assessor blind, single centre, randomised controlled trial. BMJ. 2025;388:e082104. Khwaja A. KDIGO clinical practice guidelines for acute kidney injury. Nephron Clin Pract. 2012;120(4):c179–184. Walsh M, Devereaux PJ, Garg AX, Kurz A, Turan A, Rodseth RN, Cywinski J, Thabane L, Sessler DI. Relationship between intraoperative mean arterial pressure and clinical outcomes after noncardiac surgery: toward an empirical definition of hypotension. Anesthesiology. 2013;119(3):507–15. Baron RM, Kenny DA. The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Personal Soc Psychol. 1986;51(6):1173. Johnson AE, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, Pollard TJ, Hao S, Moody B, Gow B. MIMIC-IV, a freely accessible electronic health record dataset. Sci data. 2023;10(1):1. Meng L. Heterogeneous impact of hypotension on organ perfusion and outcomes: a narrative review. Br J Anaesth. 2021;127(6):845–61. Ray SC, Mason J, O'Connor PM. Ischemic Renal Injury: Can Renal Anatomy and Associated Vascular Congestion Explain Why the Medulla and Not the Cortex Is Where the Trouble Starts? Semin Nephrol. 2019;39(6):520–9. Wagener G, Brentjens TE. Renal Disease: The Anesthesiologist's Perspective. Anesthesiol Clin North America. 2006;24(3):523–47. Prowle JR, Molan MP, Hornsey E, Bellomo R. Measurement of renal blood flow by phase-contrast magnetic resonance imaging during septic acute kidney injury: A pilot investigation*. Crit Care Med. 2012;40(6):1768–76. Bobbio E, Bollano E, Polte CL, Ekelund J, Rådegran G, Lundgren J, Haggård C, Gjesdal G, Braun O, Bartfay S-E, et al. Association between central haemodynamics and renal function in advanced heart failure: a nationwide study from Sweden. ESC Heart Fail. 2022;9(4):2654–63. Carubelli V, Metra M, Lombardi C, Bettari L, Bugatti S, Lazzarini V, Dei Cas L. Renal dysfunction in acute heart failure: epidemiology, mechanisms and assessment. Heart Fail Rev. 2012;17(2):271–82. Damman K, Kalra PR, Hillege H. PATHOPHYSIOLOGICAL MECHANISMS CONTRIBUTING TO RENAL DYSFUNCTION IN CHRONIC HEART FAILURE. J Ren Care. 2010;36(s1):18–26. Hanberg JS, Sury K, Wilson FP, Brisco MA, Ahmad T, Ter Maaten JM, Broughton JS, Assefa M, Tang WHW, Parikh CR, et al. Reduced Cardiac Index Is Not the Dominant Driver of Renal Dysfunction in Heart Failure. J Am Coll Cardiol. 2016;67(19):2199–208. Nohria A, Hasselblad V, Stebbins A, Pauly DF, Fonarow GC, Shah M, Yancy CW, Califf RM, Stevenson LW, Hill JA. Cardiorenal interactions: insights from the ESCAPE trial. J Am Coll Cardiol. 2008;51(13):1268–74. Damman K, Navis G, Smilde TD, Voors AA, van der Bij W, van Veldhuisen DJ, Hillege HL. Decreased cardiac output, venous congestion and the association with renal impairment in patients with cardiac dysfunction. Eur J Heart Fail. 2007;9(9):872–8. Dupont M, Mullens W, Finucan M, Taylor DO, Starling RC, Tang WH. Determinants of dynamic changes in serum creatinine in acute decompensated heart failure: the importance of blood pressure reduction during treatment. Eur J Heart Fail. 2013;15(4):433–40. Guglin M, Rivero A, Matar F, Garcia M. Renal dysfunction in heart failure is due to congestion but not low output. Clin Cardiol. 2011;34(2):113–6. Mullens W, Abrahams Z, Skouri HN, Francis GS, Taylor DO, Starling RC, Paganini E, Tang WH. Elevated intra-abdominal pressure in acute decompensated heart failure: a potential contributor to worsening renal function? J Am Coll Cardiol. 2008;51(3):300–6. Mullens W, Abrahams Z, Francis GS, Sokos G, Taylor DO, Starling RC, Young JB, Tang WHW. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol. 2009;53(7):589–96. Koppie TM, Serio AM, Vickers AJ, Vora K, Dalbagni G, Donat SM, Herr HW, Bochner BH. Age-adjusted Charlson comorbidity score is associated with treatment decisions and clinical outcomes for patients undergoing radical cystectomy for bladder cancer. Cancer. 2008;112(11):2384–92. Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, Crews DC, Doria A, Estrella MM, Froissart M, et al. New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race. N Engl J Med. 2021;385(19):1737–49. Han J, Zhai W, Wu Z, Zhang Z, Wang T, Ren M, Liu Z, Sessler DI, Guo Z, Meng L. Care guided by tissue oxygenation and haemodynamic monitoring in off-pump coronary artery bypass grafting (Bottomline-CS): assessor blind, single centre, randomised controlled trial. BMJ. 2025;388:e082104. Tables Table 1 Preoperative baseline characteristics of 1949 patients Characteristics* Summary Missingness, n Age (year), mean (SD) 69 (5) 0 Male, n (%) 1364 (70.0) 0 Height (cm), mean (SD) 168 (8) 0 Body mass index (kg/m 2 ), mean (SD) 25.3 (3.3) 0 Smoking, n (%)† 1124 (57.7) 0 Drinking, n (%)‡ 498 (25.6) 0 Myocardial infarction, n (%)§ 545 (28.0) 0 Arrhythmia, n (%)¶ 186 (9.5) 0 Diabetes, n (%)# 843 (43.3) 0 Hypertension, n (%)** 1422 (73.0) 0 Carotid artery disease, n (%)†† 482 (24.7) 0 Age adjusted CCI, median (IQR ) ‡‡ 5 (4–7) 0 Beta blocker, n (%) 1664 (85.4) 0 ACEI, n (%) 63 (3.2) 0 Calcium channel blocker, n (%) 801 (41.1) 0 Angiotensin receptor blocker, n (%) 695 (35.7) 0 Antidiabetic, n (%)§§ 899 (46.1) 0 Diuretic, n (%) 201 (10.3) 0 MAP (mmHg), mean (SD)¶¶ 90 (12) 0 C-reactive protein (mg/L), median (IQR)## 1.9 (0.8–5.0) 271 eGFR (mL/min/1.73 m 2 )), median (IQR)*** 93 (81–99) 0 Guided care group, n (%)††† 975 (50.0) 0 SD, standard deviation; CCI, Charlson comorbidity index; IQR, interquartile range; ACEI, angiotensin-converting enzyme inhibitors; MAP, mean arterial pressure; eGFR, estimated glomerular filtration rate * Summary statistics are reported as n (%) for categorical variables and as mean (SD) or median (IQR) for continuous variables. † Smoking was defined as a self-reported history of tobacco use, as obtained during the preoperative interview. ‡ Drinking was defined as a self-reported history of alcohol consumption, excluding occasional or purely social drinking, as obtained during the preoperative interview. § Myocardial infarction was defined as a prior event confirmed by clinical presentation, electrocardiographic changes, elevated cardiac biomarkers, or imaging evidence of myocardial infarction. ¶ Arrhythmia was defined as an abnormal cardiac electrical rhythm that was symptomatic, required treatment, or prompted further diagnostic evaluation. # Diabetes was defined as a documented clinical diagnosis established according to standard care practices. ** Hypertension was defined as a documented clinical diagnosis or recorded use of antihypertensive medication, in accordance with standard medical practice. †† Carotid artery disease was defined as ≥50% stenosis in at least one carotid artery, as identified by preoperative carotid ultrasound screening. ‡‡ The age-adjusted CCI quantifies comorbid conditions based on their number and severity, with additional points assigned for increasing age. The original CCI includes 17 weighted comorbidities, while the age-adjusted version incorporates age to improve risk stratification and predictive accuracy for clinical outcomes.[30] §§ Antidiabetic agents were defined as any pharmacologic treatments used for the management of established diabetes, including both orally administered medications and the use of insulin. ¶¶ MAP was measured 24 to 48 hours before surgery with patients resting in a supine position, eyes closed, and breathing room air in a quiet ward environment. ## C-reactive protein values were missing for 271 patients (13.9%). Multiple imputations by chained equations (50 imputations) were performed using the R package ‘mice’ with the Predictive Mean Matching method. The imputed datasets were combined using Rubin’s rules. *** eGFR was calculated using the 2021 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.[31] ††† Each patient was randomly assigned to either the guided care group or the usual care group as part of the Bottomline-CS randomized controlled trial, which was designed to evaluate the effectiveness of guided care interventions.[32] Additional Declarations No competing interests reported. Supplementary Files COAKISupplementBMCAnesthesiology11262025.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 23 Feb, 2026 Reviews received at journal 20 Feb, 2026 Reviewers agreed at journal 01 Feb, 2026 Reviewers agreed at journal 05 Jan, 2026 Reviews received at journal 22 Dec, 2025 Reviewers agreed at journal 18 Dec, 2025 Reviewers invited by journal 12 Dec, 2025 Editor invited by journal 10 Dec, 2025 Editor assigned by journal 09 Dec, 2025 Submission checks completed at journal 09 Dec, 2025 First submitted to journal 27 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8225284","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":560350858,"identity":"667c07fb-e139-47c8-807e-751a7e4ba18d","order_by":0,"name":"Jiange Han","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jiange","middleName":"","lastName":"Han","suffix":""},{"id":560350859,"identity":"4d8c8bdc-3f82-4685-bed3-f0a73eefba5d","order_by":1,"name":"Zhao Zhang","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhao","middleName":"","lastName":"Zhang","suffix":""},{"id":560350860,"identity":"296d7372-8166-400e-a36c-3a9e095b3472","order_by":2,"name":"Tao Wang","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Tao","middleName":"","lastName":"Wang","suffix":""},{"id":560350861,"identity":"d87108cd-e8d9-4d9d-ac3b-9942126919ee","order_by":3,"name":"Yunfei Li","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yunfei","middleName":"","lastName":"Li","suffix":""},{"id":560350862,"identity":"b66ea7a8-7028-4ad7-a11c-acba06a1f3fc","order_by":4,"name":"Wenqian Zhai","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Wenqian","middleName":"","lastName":"Zhai","suffix":""},{"id":560350863,"identity":"93781f5a-da7a-4770-a135-d7fdb52ed072","order_by":5,"name":"Jianxu Er","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Jianxu","middleName":"","lastName":"Er","suffix":""},{"id":560350864,"identity":"90d2c06f-8c0b-41e6-9367-1cfae16d7ead","order_by":6,"name":"Peng Sun","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Peng","middleName":"","lastName":"Sun","suffix":""},{"id":560350865,"identity":"54458224-cb7d-43aa-8567-7e46cf5210dc","order_by":7,"name":"Ruifang Gao","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Ruifang","middleName":"","lastName":"Gao","suffix":""},{"id":560350866,"identity":"49fccc6a-ec2b-4f38-a12b-b39571a005bc","order_by":8,"name":"Li Li","email":"","orcid":"","institution":"Xiangya Hospital Central South University","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Li","suffix":""},{"id":560350867,"identity":"81cc60ef-cfb8-408b-afd8-1e3d0df42624","order_by":9,"name":"Yangyang Wang","email":"","orcid":"","institution":"Henan Provincial People's Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yangyang","middleName":"","lastName":"Wang","suffix":""},{"id":560350868,"identity":"6548efca-ba22-4303-8998-31e78679c7e6","order_by":10,"name":"Ziyue Liu","email":"","orcid":"","institution":"Indiana University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ziyue","middleName":"","lastName":"Liu","suffix":""},{"id":560350869,"identity":"c3745ab2-100b-4063-b9b5-47bcba7a6f04","order_by":11,"name":"Jing Su","email":"","orcid":"","institution":"Indiana University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Su","suffix":""},{"id":560350870,"identity":"1e3f36b0-1b86-4e9a-8eca-619049c9392a","order_by":12,"name":"Zhigang Guo","email":"","orcid":"","institution":"Tianjin Chest Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhigang","middleName":"","lastName":"Guo","suffix":""},{"id":560350872,"identity":"74abdfe7-add8-4167-9a13-0ec8b7603978","order_by":13,"name":"Lingzhong Meng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAApUlEQVRIiWNgGAWjYLCCBww2PCBagngtCQxppGs5zEC8Ft323sMvEv6clzE4wHzwNg8xWszOnEuzSOC5zWNwgC3ZmjgtN3LMDBIkQFp4zKSJ03L/DVCLwTmgFv5vRGq5wWP8ICHhAMgWNiK1nMkxY0g4kMwjeZjN2HIOUVqOnzH+8OGPnT3f8eaHN94QowUI2CDRwUykcrDaDyQoHgWjYBSMgpEIAFWeL3+Vs3MxAAAAAElFTkSuQmCC","orcid":"","institution":"Indiana University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Lingzhong","middleName":"","lastName":"Meng","suffix":""}],"badges":[],"createdAt":"2025-11-27 23:23:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8225284/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8225284/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98387509,"identity":"a638d27e-5686-4c12-84bc-298de3c8f744","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1205322,"visible":true,"origin":"","legend":"","description":"","filename":"COAKIManuscriptBMCAnesthesiologyR212062025.docx","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/14ef1dadeb92a89a01860a35.docx"},{"id":98441180,"identity":"ac6b62ee-bcce-4437-a9c1-a039499ee1f9","added_by":"auto","created_at":"2025-12-17 17:05:01","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13801,"visible":true,"origin":"","legend":"","description":"","filename":"80c2c74a226b4dd195f71bb1662586b1.json","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/515e80be719a88968aa1f14d.json"},{"id":98440165,"identity":"7213c797-7373-4b9c-82bf-5d80378bcff1","added_by":"auto","created_at":"2025-12-17 17:03:25","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3033144,"visible":true,"origin":"","legend":"","description":"","filename":"COAKISupplementBMCAnesthesiology11262025.docx","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/bc29cad2eadd3842494865f0.docx"},{"id":98387507,"identity":"7621957d-401c-4174-9709-0438b608aad7","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":145197,"visible":true,"origin":"","legend":"","description":"","filename":"80c2c74a226b4dd195f71bb1662586b11enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/6643039cc6b0c7f98576a728.xml"},{"id":98387510,"identity":"29736edd-3d4e-4ccd-8d26-76d63db8a69e","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"emf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":923904,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.emf","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/80df30a60893f265f2d07e04.emf"},{"id":98387514,"identity":"649d9e91-da01-484f-b291-1264d2c0f06e","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"emf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1881308,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.emf","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/ed425d01c492d28a5b375007.emf"},{"id":98441222,"identity":"2f82fdd6-0871-4171-a722-895db7047eb8","added_by":"auto","created_at":"2025-12-17 17:05:05","extension":"emf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3208236,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.emf","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/190b643a6bbb20c570210b2c.emf"},{"id":98387513,"identity":"c97dc840-79b1-451f-91dd-24a5155a1fab","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"emf","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1787904,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.emf","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/fbd2246c0401b1766f8d30e2.emf"},{"id":98387518,"identity":"c2ac0856-860f-4c73-9ece-3112799b5477","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"emf","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2015908,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.emf","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/3b449dfc7ecc8f9815a02aa7.emf"},{"id":98440103,"identity":"ad924874-6984-4d64-b9eb-60bf5b7e8b96","added_by":"auto","created_at":"2025-12-17 17:03:18","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":13200,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/d023fbbdc248f5f94760a251.png"},{"id":98440554,"identity":"9b2c43ea-f996-48fb-a1e0-1ac596470980","added_by":"auto","created_at":"2025-12-17 17:04:00","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":20110,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/487c1a7844816e162b7f4b5f.png"},{"id":98440200,"identity":"51b24209-427b-4c01-bf91-1454b51a82d1","added_by":"auto","created_at":"2025-12-17 17:03:31","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":52015,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/435e7009b217691251333065.png"},{"id":98440593,"identity":"837ca3e2-1813-4760-b41e-df41c453fbbd","added_by":"auto","created_at":"2025-12-17 17:04:04","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":17184,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/50f374be77c8ae3d02582d07.png"},{"id":98440007,"identity":"847b3d61-2252-4a6d-8f2b-af1d3695dc91","added_by":"auto","created_at":"2025-12-17 17:03:12","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35967,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/9252ed5c070f3c30f136cef8.png"},{"id":98387523,"identity":"047721a3-b446-4987-8630-6c0719ca4c7a","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"xml","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":141886,"visible":true,"origin":"","legend":"","description":"","filename":"80c2c74a226b4dd195f71bb1662586b11structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/958283fddccfe4c801e5ebef.xml"},{"id":98387521,"identity":"1a9e9927-f608-4aae-b800-496dea91b8c1","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"html","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":157515,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/a43fb01c4f49f91aaa2892d2.html"},{"id":98387502,"identity":"41345625-fcc3-4bbf-8148-be833c4ab390","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":133957,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFlowchart of the Study Cohort\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis figure outlines the selection of participants from the original Bottomline-CS trial cohort (n=1,960) to the final analytic cohort (n=1,949), with reasons for exclusion specified. Thresholds of baseline cardiac output were subsequently evaluated for their association with the risk of AKI, as detailed in Figure 2. The final threshold was selected based on three criteria: (1) statistical significance in differentiating AKI risk between groups, (2) clinical practicality to favor thresholds that are simple and easily memorable, and (3) broader clinical coverage, prioritizing thresholds that identify a larger proportion of patients at elevated risk to maximize opportunities for prevention.\u003c/p\u003e\n\u003cp\u003eAKI, acute kidney injury\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/a58864d5766ba86670e13857.png"},{"id":98441451,"identity":"a8c47736-9f5c-4d60-8c44-103b1b90865e","added_by":"auto","created_at":"2025-12-17 17:05:26","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":118773,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAssociation Between Baseline Cardiac Output Thresholds and Risk of Postoperative Acute Kidney Injury\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis figure depicts the relationship between baseline cardiac index (A) and absolute cardiac output (B) and the risk of postoperative AKI across a range of thresholds. At each increment, patients were dichotomized into “\u0026lt; threshold” and “≥ threshold” groups, and odds ratios for AKI were calculated. Thresholds ranged from 1.5 to 4.0 L/min/m\u003csup\u003e2\u003c/sup\u003e for cardiac index and from 2.5 to 7.0 L/min for cardiac output, advancing in 0.1-unit increments. A locally estimated scatterplot smoothing (LOESS) curve was fitted to the odds ratio series, with shaded areas indicating the 95% confidence interval. Clinically relevant thresholds, cardiac index of 3.0 and 3.5 L/min/m\u003csup\u003e2\u003c/sup\u003e, and cardiac output of 5.0 and 6.0 L/min, are highlighted in red.\u003c/p\u003e\n\u003cp\u003eOR, odds ratio; AKI, acute kidney injury; CI, confidence interval; LOESS, locally estimated scatterplot smoothing\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/3057dce132ede28bc4420864.png"},{"id":98387504,"identity":"7eef27a6-cc97-4154-9ac9-0c3b3584cc75","added_by":"auto","created_at":"2025-12-17 08:56:45","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":309046,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRisk of Acute Kidney Injury in Patients with High vs. Low Baseline Cardiac Output\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePatients were stratified by baseline cardiac index or cardiac output using four clinically relevant thresholds identified in Figure 2: cardiac index of 3.0 and 3.5 L/min/m\u003csup\u003e2\u003c/sup\u003e, and cardiac output of 5.0 and 6.0 L/min. Propensity score matching was employed to balance baseline characteristics between the high- and low-output groups. For each threshold, the number of patients, AKI incidence, and odds ratios before and after matching are shown (A). Covariate balance was assessed using absolute standardized differences, with crude (orange dots) and matched (blue dots) values compared across variables used in the matching process (B-E).\u003c/p\u003e\n\u003cp\u003eAKI, acute kidney injury; ASD, absolute standardized difference; CI, cardiac index; CO, cardiac output; CCI, Charlson comorbidity index; ACEI, angiotensin-converting enzyme inhibitors; eGFR, estimated glomerular filtration rate.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/50628daf8a43c792eb5292b6.png"},{"id":98440139,"identity":"c6fa6459-ed8b-4636-9de1-5ad3dcc37e8d","added_by":"auto","created_at":"2025-12-17 17:03:20","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":94892,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRelationship Between Baseline Cardiac Output and Postoperative Acute Kidney Injury\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRestricted cubic spline analyses demonstrated a nonlinear, J-shaped relationship between baseline cardiac output and the risk of postoperative acute kidney injury (AKI). The analysis was performed using both baseline cardiac index (A) and absolute cardiac output (B) as continuous predictors. Odds ratios for AKI were plotted relative to the median value of each respective distribution, which served as the reference (dashed line). Light orange bars represent the distribution of patients across the range of cardiac index and cardiac output values. Shaded blue regions denote 95% confidence intervals around the estimated odds ratios. Red dots indicate the predefined thresholds identified in Figure 2: cardiac index values of 3.0 and 3.5 L/min/m\u003csup\u003e2\u003c/sup\u003e, and cardiac output values of 5.0 and 6.0 L/min.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/adf5f2400161e84969a538e9.png"},{"id":98440164,"identity":"66cff43b-8499-4dbb-8194-6849165ec7f4","added_by":"auto","created_at":"2025-12-17 17:03:25","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":129736,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBaseline and Intraoperative Cardiac Output and Its Relationships with Systemic Vascular Resistance and Mean Arterial Pressure\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis figure presents perioperative hemodynamic patterns in propensity score-matched patients. Distributions of baseline cardiac output (A), systemic vascular resistance (B), and mean arterial pressure (C) are shown using violin plots with overlaid boxplots indicating the median and interquartile range; “low” and “high” labels on the x-axis correspond to baseline cardiac output \u0026lt;5.0 L/min and ≥5.0 L/min, respectively. The relationship between baseline cardiac output (x-axis) and systemic vascular resistance (y-axis) is shown (D), with the vertical dashed line marking the 5.0 L/min threshold used for group classification. The black dashed curve represents an iso-pressure reference line for a MAP of 90 mmHg, based on the physiologic equation CO × SVR = 80 × MAP; cardiac output, rather than cardiac index, was used in these analyses to align with this equation. Distribution of cardiac index before and after anaesthesia induction, and the median and minimum cardiac index during surgery, is shown (E), with median and minimum values highlighted, demonstrating that intergroup differences in baseline cardiac index persisted intraoperatively despite an overall remarkable reduction.\u003c/p\u003e\n\u003cp\u003eCO, cardiac output; SVR, systemic vascular resistance; MAP, mean arterial pressure\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/1fe5651c01edbcfbcafeaaea.png"},{"id":98623179,"identity":"762a4188-f92d-474e-a67f-b420e8dcaf81","added_by":"auto","created_at":"2025-12-19 17:05:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2044103,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/8a12f659-6cd4-428c-a217-3676cff3ef38.pdf"},{"id":98441364,"identity":"d156b8d2-8330-40a1-8347-b7ef27e9f100","added_by":"auto","created_at":"2025-12-17 17:05:15","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":3033144,"visible":true,"origin":"","legend":"","description":"","filename":"COAKISupplementBMCAnesthesiology11262025.docx","url":"https://assets-eu.researchsquare.com/files/rs-8225284/v1/85f38eb6e9c6f039f3f3cd56.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Inverse Association Between Preoperative Cardiac Output and Postoperative Kidney Function in Off-Pump Coronary Artery Bypass Grafting","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAcute kidney injury (AKI) is a frequent complication after cardiac surgery, affecting roughly one in five to one in three patients, depending on case-mix and procedure [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Even small rises in serum creatinine are linked to greater morbidity, longer hospitalization, higher costs, and increased long-term mortality and progression of chronic kidney disease; when renal replacement therapy is required, in-hospital mortality is high [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. The clinical and economic burden makes better physiological risk characterization a priority [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe pathogenesis of cardiac surgery-associated AKI is multifactorial. Proposed mechanisms include renal hypoperfusion, systemic inflammation, ischemia-reperfusion injury, neurohormonal activation, oxidative stress, hemolysis, and exposure to nephrotoxins, with vulnerability shaped by diabetes, anemia, pre-existing kidney disease, and age [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Current preventive approaches emphasize avoidance of nephrotoxins and maintenance of perfusion pressure and volume status within guideline-based hemodynamic care bundles, although their physiological underpinnings in surgical populations remain incompletely validated [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003ePrior evidence on the relation between cardiac output and AKI comes from different populations, including heart failure cohorts and clinical syndromes described as low output, not from cardiac surgery [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In those settings, low output is a composite clinical label that may include hypotension, poor peripheral perfusion, oliguria, and use of inotropes or vasopressors; therefore, cardiac output is usually inferred rather than measured directly [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Within cardiac surgery, this association remains uncharacterized. Interpretation in on-pump procedures is confounded by bypass-related factors such as hemolysis, embolization, and systemic inflammatory activation, whereas off-pump coronary artery bypass grafting limits these influences and offers a setting to examine the association with less confounding [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn this cohort study of patients undergoing off-pump CABG, we evaluated the association between preoperative (not intraoperative) cardiac output and postoperative AKI. A priori, we defined a lower pre-operative cardiac output as a predisposing factor, meaning a baseline physiological vulnerability present before anesthesia and surgery. This is distinct from precipitating factors that occur intra-operatively, such as hypotension, bleeding, or nephrotoxin exposure, which can directly trigger renal injury [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Our primary hypothesis was that lower pre-operative cardiac output, as a predisposing factor, would be associated with a higher incidence of post-operative AKI.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy Design\u003c/h2\u003e \u003cp\u003eThis cohort study is a secondary analysis of the Bottomline-CS trial (Better Outcome Through Tissue Oxygenation Monitoring Linked with INtErvention in Cardiac Surgery; ClinicalTrials.gov NCT04896736), a single-center, assessor-blinded, randomized controlled trial that enrolled patients undergoing elective off-pump CABG.[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] The original trial compared outcomes between patients who received perioperative care guided by multisite tissue oxygen saturation and advanced hemodynamic monitoring versus those managed with standard care. For this secondary analysis, we addressed a distinct research question: whether there is an association between preoperative cardiac output and postoperative AKI, which differs from the primary trial objective. Accordingly, the study design (observational cohort analysis vs. randomized trial), analytical approach, exposures, and outcomes are all substantively different from those of the parent trial. Although the original trial involved randomization and intervention, the present analysis treats the trial population as a single prospective cohort with uniformly measured baseline and intraoperative hemodynamic parameters, enabling an evaluation of physiologic predictors of AKI across the entire study sample. The Institutional Review Board of Tianjin Chest Hospital approved this secondary analysis on February 1, 2024, with a waiver of informed consent due to the use of de-identified data. The reporting adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cohort studies.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSetting\u003c/h3\u003e\n\u003cp\u003eThe Bottomline-CS trial was conducted at Tianjin Chest Hospital, a high-volume tertiary academic center affiliated with Tianjin University, China. As a nationally recognized referral hospital with advanced cardiovascular and pulmonary care infrastructure, the institution provides comprehensive surgical services for a diverse patient population across Northern China.\u003c/p\u003e\n\u003ch3\u003eParticipants\u003c/h3\u003e\n\u003cp\u003eBetween June 8, 2021, and December 27, 2023, consecutive patients aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years scheduled for elective off-pump CABG at Tianjin Chest Hospital were prospectively screened and enrolled. Exclusion criteria included inability or unwillingness to provide informed consent, need for preoperative ventilatory support (invasive or non-invasive), presence of an external cardiac assist device, requirement for urgent or emergent surgery, or an expected life expectancy of less than 30 days.\u003c/p\u003e \u003cp\u003e Eligible patients were identified from daily surgical schedules and approached at least 24 hours before surgery for consent. Enrolled participants were randomized to receive either guided perioperative management, based on multisite tissue oxygen saturation and advanced hemodynamic monitoring, or standard care, in which the same monitoring data were collected but not disclosed to the clinical team.\u003c/p\u003e \u003cp\u003eAll patients underwent uniform perioperative data collection, irrespective of the randomization group. Data acquisition spanned the entire perioperative course, from baseline evaluation through postoperative day 30, and included detailed physiologic measurements, clinical outcomes, and safety assessments. Outcome assessors were blinded to group assignment, ensuring unbiased evaluation. This standardized approach enabled consistent exposure measurement and robust outcome ascertainment across the cohort.\u003c/p\u003e\n\u003ch3\u003ePerioperative Care\u003c/h3\u003e\n\u003cp\u003eAll patients received comprehensive intraoperative monitoring, including pulse oximetry, end-tidal carbon dioxide monitoring, electrocardiography, non-invasive blood pressure, and invasive arterial pressure via a radial artery catheter. Central venous pressure was measured through a catheter inserted into the internal jugular vein. Transesophageal echocardiography was used when clinically indicated. General anesthesia was induced with either propofol or etomidate and maintained with propofol infusion and/or sevoflurane inhalation at the discretion of the attending anesthesiologist. Neuromuscular blockade was achieved using either cisatracurium or rocuronium. All patients were endotracheally intubated and mechanically ventilated, with sufentanil administered as the primary intraoperative analgesic in accordance with institutional standards.\u003c/p\u003e \u003cp\u003eThroughout surgery, continuous hemodynamic monitoring was performed using a minimally invasive arterial waveform analysis system (Masimo LiDCO Hemodynamic Monitoring System, Masimo, Irvine, California, USA), which estimated cardiac output, stroke volume, and systemic vascular resistance from the radial arterial pressure waveform. Concurrently, multisite tissue oxygen saturation was assessed using a near-infrared spectroscopy system (Nonin Medical, Inc., Plymouth, Minnesota, USA). Cerebral oxygenation was measured at the left and right forehead, and somatic oxygenation at the forearm brachioradialis. Together, these modalities provided continuous, real-time evaluation of systemic and regional perfusion throughout the procedure.\u003c/p\u003e\n\u003ch3\u003eOutcomes\u003c/h3\u003e\n\u003cp\u003e The primary outcome was postoperative acute kidney injury (AKI), defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) Clinical Practice Guidelines.[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] In accordance with the previously published research, urine output criteria were not used due to the frequent use of diuretics and incomplete documentation, which can limit diagnostic reliability.[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] AKI was identified by postoperative increases in serum creatinine relative to preoperative baseline values. Specifically, an increase in serum creatinine of \u0026ge;\u0026thinsp;0.3 mg/dL within 48 hours, or a\u0026thinsp;\u0026ge;\u0026thinsp;50% increase from baseline within 7 days, was considered diagnostic of AKI. Baseline creatinine was defined as the most recent value obtained prior to surgery, and postoperative creatinine was defined as the highest value recorded within the first seven days after surgery.[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eExposures\u003c/h2\u003e \u003cp\u003eThe primary exposure in this study was baseline cardiac output, measured one to two days prior to surgery using a noninvasive hemodynamic monitoring system. Assessments were performed in a controlled hospital setting with patients in a supine position, awake, calm, with eyes closed, and breathing either room air or their usual home oxygen. Cardiac output, stroke volume, and systemic vascular resistance were estimated using a finger and arm cuff\u0026ndash;based monitor (Continuous Noninvasive Arterial Pressure [CNAP\u0026trade;] Module, Masimo LiDCO\u0026trade; Hemodynamic Monitoring System; Masimo, Irvine, California, USA). This preoperative monitoring approach differs from the minimally invasive arterial waveform analysis employed intraoperatively, enabling physiologic characterization under resting, baseline conditions.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePotential Confounders\u003c/h3\u003e\n\u003cp\u003eIn this cohort study, potential confounders were defined as variables that may be associated with both the exposure (baseline cardiac output measured one to two days prior to surgery) and the outcome (AKI within the first seven postoperative days) without residing in the causal pathway. According to established epidemiologic principles, confounders must precede the exposure, be associated with the outcome, and not be a result of the exposure itself. In contrast, variables on the causal path between exposure and outcome are considered mediators and were addressed separately to avoid overadjustment. The inclusion of mediators or variables affected by the exposure could bias effect estimates by distorting or attenuating the relationship of interest.\u003c/p\u003e \u003cp\u003eTo preserve the temporal structure and causal interpretability of our analysis, we limited the set of potential confounders to demographic and preoperative characteristics. Intraoperative and postoperative variables were intentionally excluded, as they could be influenced by preoperative cardiac output and, if adjusted for, may introduce collider stratification bias. This decision aligns with the conceptual model that views baseline cardiac output as a determinant, rather than a consequence, of perioperative events.\u003c/p\u003e \u003cp\u003eCandidate confounders were selected based on prior literature, clinical reasoning, and their observed associations with exposure and outcome in univariate analyses. These included age, sex, height, body mass index, smoking and alcohol history, prior myocardial infarction, arrhythmia, diabetes, hypertension, and carotid artery disease. The overall comorbidity burden was captured using the age-adjusted Charlson comorbidity index. Preoperative medication use, specifically beta-blockers, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, calcium channel blockers, antidiabetic agents, and diuretics, was also incorporated, given their potential influence on both cardiac function and renal risk.\u003c/p\u003e \u003cp\u003eRelevant physiologic and laboratory parameters, such as mean arterial pressure, estimated glomerular filtration rate, and C-reactive protein, were included in addition to comorbid and pharmacologic variables. To account for any residual influence from the original trial design, group assignment in the Bottomline-CS trial was also treated as a covariate. This comprehensive set of baseline covariates was selected to minimize bias and isolate the independent association between cardiac output and postoperative AKI.\u003c/p\u003e\n\u003ch3\u003ePotential Mediators\u003c/h3\u003e\n\u003cp\u003eTo explore the mechanisms underlying the association between preoperative cardiac output and postoperative AKI, we identified intraoperative variables that may function as mediators, i.e., physiologic processes influenced by cardiac output that, in turn, could contribute to renal injury. These variables were evaluated separately from baseline confounders to preserve the integrity of causal inference.\u003c/p\u003e \u003cp\u003eCandidate mediators included intraoperative hemodynamic disturbances, oxygenation deficits, vasopressor use, fluid balance, and changes in cardiac output during surgery. Hemodynamic and oxygenation abnormalities were quantified using the area under the curve (AUC) method, which reflects the cumulative duration and magnitude of deviations below predefined thresholds, specifically 10%, 20%, and 30% reductions from ward-based baseline values measured 1\u0026ndash;2 days prior to surgery. Vasopressor exposure was defined as the use of vasoactive agents administered to at least 20% but no more than 80% of patients, excluding agents used either rarely or ubiquitously.\u003c/p\u003e \u003cp\u003eFluid balance was assessed by measuring the total intraoperative input (crystalloids, colloids, and autologous blood products) and estimated blood loss. Intraoperative changes in cardiac output were captured using two metrics: (1) relative change in cardiac index (minimum intraoperative value minus preoperative baseline, divided by baseline), and (2) absolute change (direct difference between minimum intraoperative and baseline values).\u003c/p\u003e \u003cp\u003eThese variables were excluded from the primary analysis to preserve the estimation of the total effect of baseline cardiac output on AKI, avoiding bias introduced by conditioning on post-exposure variables. Instead, they were examined using formal mediation analyses to assess whether intraoperative physiologic disturbances partially explained the relationship between baseline cardiac output and the development of AKI.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eData Sources\u003c/h2\u003e \u003cp\u003eAll data for this cohort study were derived from the Bottomline-CS trial, with methodological details previously reported in the published protocol and manuscript. Data collection followed a standardized approach across all phases, preoperative, intraoperative, and postoperative, ensuring internal consistency and minimizing measurement variability.\u003c/p\u003e \u003cp\u003ePreoperative data were collected one to two days prior to surgery by trained research staff, who conducted in-person evaluations and physiological measurements in a controlled hospital environment. These assessments included baseline hemodynamic and tissue oxygenation metrics, which were recorded while participants were in a supine, resting state. Intraoperative data, including both continuous and event-based measurements, were collected in real time by designated research team members responsible for recording discrete intraoperative variables and ensuring the completeness of automated monitoring data.\u003c/p\u003e \u003cp\u003ePostoperative outcome data were obtained by independent assessors blinded to group assignments and uninvolved in preoperative and intraoperative care, thereby preserving objectivity in outcome ascertainment. Although the original trial involved randomization into two treatment arms, data acquisition was carried out uniformly for all participants in accordance with the predefined protocol, allowing for consistent and unbiased observational analysis across the entire cohort.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eBias\u003c/h2\u003e \u003cp\u003eMultiple strategies were employed to minimize bias across study design, data collection, and analysis. The study protocol was finalized prior to patient enrolment, thereby eliminating the possibility of post hoc design modifications. Eligible patients were identified consecutively from the surgical schedule using predefined criteria, reducing selection bias. Preoperative data, including baseline hemodynamic and tissue oxygenation measurements, were collected by trained personnel not involved in intraoperative care or outcome assessment. Postoperative outcomes were evaluated by assessors blinded to group allocation, thereby minimizing observer bias. Standardized data acquisition procedures, real-time digital entry, and independent third-party audits further reduced the risk of information bias. All physiological measurements were performed using validated devices with prespecified thresholds applied uniformly across study groups.\u003c/p\u003e \u003cp\u003eSelection bias and attrition were minimal, as evidenced by the inclusion of 1,960 patients with only 8 missing outcome records (0.4%). Confounding was addressed through the careful selection of clinically and biologically relevant preoperative variables, which were incorporated into a propensity score matching framework to achieve covariate balance without introducing over-adjustment from intraoperative or postoperative factors. Although these measures strengthen internal validity, residual confounding and limited generalizability inherent to single-center observational designs cannot be entirely excluded.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eStudy Size\u003c/h2\u003e \u003cp\u003eThe study size was determined by the number of eligible participants enrolled in the Bottomline-CS trial who met the criteria for this secondary analysis. All patients aged 60 years or older undergoing elective off-pump coronary artery bypass grafting were considered, and those with cancelled surgeries or missing data for AKI classification were excluded. This yielded a final analytical cohort of 1,949 patients. No additional sample size calculation was performed for this observational analysis, as the entire available cohort was included to maximize statistical precision and support the planned propensity score matching approach.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eQuantitative Variables\u003c/h2\u003e \u003cp\u003eContinuous variables were assessed for distributional properties using histograms and Q-Q plots. Those following an approximately normal distribution were summarized as means with standard deviations (SD), whereas skewed variables were reported as medians with interquartile ranges (IQR). Categorical variables were presented as frequencies with percentages. For primary analyses, baseline cardiac output was dichotomized using clinically relevant thresholds to facilitate interpretability and support propensity score matching, while preserving the physiological distinction between lower and higher output states. This approach enabled a meaningful evaluation of the relationship between preoperative cardiac output and the risk of AKI in the matched cohorts.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eMissing Data\u003c/h2\u003e \u003cp\u003eMissing data were addressed using multiple imputations by fully conditional specification, generating 50 imputed datasets to ensure the stability of estimates. All primary analyses were conducted using the pooled results from the imputed datasets.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eThe primary objective of this cohort study was to determine whether preoperative cardiac output is independently associated with the development of AKI. Specifically, we tested the hypothesis that baseline cardiac output levels stratify patients into different risk categories for developing AKI after off-pump coronary artery bypass grafting.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eThreshold Determination and Exposure Definition\u003c/h2\u003e \u003cp\u003eTo identify a clinically meaningful threshold for cardiac output, we systematically evaluated odds ratios (ORs) for AKI across a range of cutoffs. ORs were calculated across thresholds incremented by 0.1 L/min or 0.1 L/min/m\u003csup\u003e2\u003c/sup\u003e for cardiac index. These were plotted to visualize the relationship between baseline cardiac output thresholds and the risk of AKI. Thresholds were selected based on both statistical significance in the OR-threshold plots and clinical usability. The primary exposure was defined using the selected threshold, with patients categorized into high- or low-cardiac-output groups accordingly. The main analysis utilized cardiac index (cardiac output normalized to body surface area) to account for body size, whereas sensitivity analyses employed absolute cardiac output values.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003ePropensity Score Matching\u003c/h2\u003e \u003cp\u003eTo mitigate confounding, we employed propensity score matching. Propensity scores were derived from logistic regression models incorporating preoperative variables known or suspected to be associated with both cardiac output and AKI. These included demographics, comorbidities, medication use, laboratory values, and randomization group from the parent trial. Nearest-neighbor matching without replacement was used, with a variable matching ratio (1:1 to 1:4) to optimize sample size while maintaining balance. A caliper width of 0.2 standard deviations of the logit of the propensity score was applied. Covariate balance was assessed using absolute standardized differences, with values\u0026thinsp;\u0026le;\u0026thinsp;0.1 indicating adequate balance.\u003c/p\u003e \u003cp\u003eThe incidence of AKI and other categorical outcomes was compared using conditional logistic regression. Linear mixed-effects models were used for continuous outcomes, with matching incorporated as a random intercept. ORs with 95% confidence intervals (CIs) were used to quantify associations between cardiac output and postoperative AKI.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eNonlinear Association Analysis\u003c/h2\u003e \u003cp\u003eTo assess the shape of the relationship between cardiac output and AKI, restricted cubic spline regression was performed using continuous cardiac index or cardiac output as predictors. Models were adjusted for the same covariates used in propensity score estimation. These analyses enabled the detection of nonlinear associations, complementing the threshold-based approach.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eIntraoperative Hemodynamic Trajectory Analysis\u003c/h2\u003e \u003cp\u003eTo determine whether preoperative differences in cardiac output were sustained intraoperatively, we analysed cardiac output trajectories in propensity score\u0026ndash;matched cohorts. Measurements before and after anesthesia induction were compared, along with intraoperative median and minimum cardiac index values. Minimum values were calculated by smoothing cardiac index measurements (10-second rolling medians) and extracting the median of the ten lowest values (Supplementary Fig.\u0026nbsp;1). Between-group differences were assessed using linear mixed-effects models to account for unequal sample sizes and repeated measures.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eMediation Analysis\u003c/h2\u003e \u003cp\u003eTo explore whether intraoperative variables mediated the relationship between baseline cardiac output and postoperative AKI, we conducted mediation analyses using a two-step approach.[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] First, potential mediators, including intraoperative hemodynamic variables, oxygenation parameters, vasopressor use, fluid balance, and cardiac output changes, were modelled using regression analyses adjusted for baseline covariates. Second, logistic regression was used to assess associations between baseline cardiac index, mediators, and AKI. Continuous mediators were standardized (z-scores). Mediation effects were quantified by estimating the average causal mediation effect, average direct effect, total effect, and proportion mediated, using bootstrap resampling (1,000 iterations). A 95% CI excluding zero was considered statistically significant.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eExternal Validation\u003c/h2\u003e \u003cp\u003eTo assess generalizability, we replicated the primary analysis in an external cohort drawn from the MIMIC-IV v3.1 database.[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e] This dataset included patients who underwent cardiac procedures with documented preoperative cardiac output and postoperative AKI outcomes. Analyses used the same exposure thresholds and outcome definitions as the primary cohort.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eSoftware and Statistical Significance\u003c/h2\u003e \u003cp\u003eAll analyses were performed using R version 3.5.3 (R Foundation for Statistical Computing, Vienna, Austria). Key R packages included: readxl, dplyr, tableone, mice, autoReg, ggplot2, Matching, mediation, survival, and rms. Visualizations were created in Python 3.8.10 using NumPy, Pandas, Matplotlib, Seaborn, and Statsmodels. A two-sided p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec25\" class=\"Section2\"\u003e \u003ch2\u003eParticipants\u003c/h2\u003e \u003cp\u003eThe Bottomline-CS randomized controlled trial enrolled a total of 1,960 patients scheduled for elective off-pump CABG between June 8, 2021, and December 27, 2023. For the current cohort study, 11 patients were excluded due to cancelled surgery (n\u0026thinsp;=\u0026thinsp;5), age younger than 60 years (n\u0026thinsp;=\u0026thinsp;3), and missing postoperative AKI data (n\u0026thinsp;=\u0026thinsp;3), resulting in an analytic cohort of 1,949 patients. A detailed flowchart illustrating patient selection criteria and analysis plan is provided in Fig.\u0026nbsp;1.\u003c/p\u003e \u003cdiv id=\"Sec26\" class=\"Section3\"\u003e \u003ch2\u003eCohort Characteristics\u003c/h2\u003e \u003cp\u003eThe study included 1,949 patients with a mean age of 69\u0026thinsp;\u0026plusmn;\u0026thinsp;5 years; 70% of participants were male, and the mean body mass index was 25\u0026thinsp;\u0026plusmn;\u0026thinsp;3 kg/m\u003csup\u003e2\u003c/sup\u003e. Preoperative hemodynamic measures revealed a mean cardiac output of 4.5\u0026thinsp;\u0026plusmn;\u0026thinsp;1.5 L/min and a cardiac index of 2.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.8 L/min/m\u003csup\u003e2\u003c/sup\u003e. The mean arterial pressure before surgery was 90\u0026thinsp;\u0026plusmn;\u0026thinsp;12 mmHg. AKI occurred in 11% (213/1949) of patients. Comprehensive preoperative baseline characteristics, including demographics, medical history, medication use, cognitive assessments, cardiac measurements, hemodynamic data, and laboratory results, are presented in Supplementary Table\u0026nbsp;1.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec27\" class=\"Section3\"\u003e \u003ch2\u003ePreoperative Baseline Cardiac Output Threshold Determination\u003c/h2\u003e \u003cp\u003eThe relationship between preoperative baseline cardiac output thresholds measured at the patient ward 1\u0026ndash;2 days prior to surgery and the unadjusted OR of postoperative AKI is presented in Fig.\u0026nbsp;2. As the threshold for cardiac output increased, the OR for AKI correspondingly increased. Based on this analysis and clinical relevance, we selected a cardiac index threshold of 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e for the primary analysis. Additionally, thresholds based on a cardiac index of 3.5 L/min/m\u003csup\u003e2\u003c/sup\u003e and absolute cardiac output of 5.0 L/min and 6.0 L/min were chosen for sensitivity analyses to evaluate the robustness of the association.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003ePropensity Score-Matched Analysis of AKI Risk in Patients with High vs. Low Baseline Cardiac Output\u003c/h2\u003e \u003cp\u003eIn the primary analysis, patients were divided into two groups based on a preoperative cardiac index threshold of 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e: those with a cardiac index of less than 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e and those with a cardiac index of 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e or greater. Preoperative baseline characteristics for these two groups are detailed in Supplementary Table\u0026nbsp;2. Propensity score matching was performed to balance baseline covariates between groups, using variables including age, sex, height, body mass index, smoking status, alcohol consumption, history of myocardial infarction, arrhythmia, diabetes, hypertension, carotid artery disease, age-adjusted Charlson comorbidity index, medications (i.e., beta-blockers, angiotensin-converting enzyme inhibitors, calcium channel blockers, angiotensin receptor blockers, antidiabetic treatment, and diuretics), mean arterial pressure, C-reactive protein, estimated glomerular filtration rate, and randomization group assignment in the Bottomline-CS trial (Table\u0026nbsp;1).\u003c/p\u003e \u003cp\u003eThe outcomes of propensity score matching are summarized in Figs.\u0026nbsp;3A and 3B and Supplementary Table\u0026nbsp;3. Post-matching absolute standardized differences for all covariates were below 0.1, indicating adequate balance (Fig.\u0026nbsp;3B). After matching, the mean cardiac index was 2.2\u0026thinsp;\u0026plusmn;\u0026thinsp;0.8 L/min/m\u003csup\u003e2\u003c/sup\u003e in the lower cardiac index group (n\u0026thinsp;=\u0026thinsp;942) and 3.6\u0026thinsp;\u0026plusmn;\u0026thinsp;0.5 L/min/m\u003csup\u003e2\u003c/sup\u003e in the higher cardiac index group (n\u0026thinsp;=\u0026thinsp;314). The incidence of AKI was significantly higher among patients with a higher baseline cardiac index compared to those with a lower level (14.3% [45/314] vs. 9.0% [85/942]; OR, 1.69; 95% CI, 1.15\u0026ndash;2.48).\u003c/p\u003e \u003cp\u003eAdditional sensitivity analyses using alternative thresholds further supported this finding. Specifically, analyses based on cardiac index thresholds of 3.5 L/min/m\u003csup\u003e2\u003c/sup\u003e (Fig.\u0026nbsp;3C, Supplementary Table\u0026nbsp;4), and absolute cardiac output thresholds of 5.0 L/min (Fig.\u0026nbsp;3D, Supplementary Table\u0026nbsp;5) and 6.0 L/min (Fig.\u0026nbsp;3E, Supplementary Table\u0026nbsp;6), consistently demonstrated that patients with higher baseline cardiac outputs were at significantly greater risk of developing postoperative AKI (Fig.\u0026nbsp;3A and Supplementary Table\u0026nbsp;7).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eNonlinear Association Between Baseline Cardiac Output and Risk of Postoperative AKI\u003c/h2\u003e \u003cp\u003eRestricted cubic spline analyses revealed a nonlinear, J-shaped relationship between baseline cardiac output and the risk of postoperative AKI. As shown in Fig.\u0026nbsp;4A, the odds of AKI began to increase when the baseline cardiac index exceeded approximately 2.0 L/min/m\u0026sup2;, with a steeper rise beyond 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e. A similar pattern was observed for absolute cardiac output (Fig.\u0026nbsp;4B), where the odds ratio started to rise above a cardiac output of roughly 4.0 L/min and became significantly elevated beyond 5.0 L/min. This pattern supports the presence of a threshold effect and reinforces the use of 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e and 5.0 L/min as clinically meaningful cutoffs for cardiac index and cardiac output, respectively, in risk stratification.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePreoperative and Intraoperative Hemodynamic Patterns in Patients with Distinct Baseline Cardiac Output Levels\u003c/h3\u003e\n\u003cp\u003eUsing propensity score-matched patients (Supplementary Table\u0026nbsp;5), we compared hemodynamic parameters between groups stratified by a baseline cardiac output threshold of 5.0 L/min. As expected, baseline cardiac output clearly separated the two groups with no overlap (Fig.\u0026nbsp;5A). Systemic vascular resistance also differed significantly between groups, though values partially overlapped (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001, Fig.\u0026nbsp;5B). In contrast, mean arterial pressures were comparable between groups (P\u0026thinsp;=\u0026thinsp;0.083, Fig.\u0026nbsp;5C), as anticipated, given that mean arterial pressure was one of the covariates included in the propensity score model.\u003c/p\u003e \u003cp\u003eTo further explore the relationship between cardiac output and systemic vascular resistance, we generated a scatterplot using the same matched patients (Fig.\u0026nbsp;5D). An iso-pressure reference line corresponding to a mean arterial pressure of 90 mmHg was overlaid.[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] The distribution of data points along this line demonstrated that for a given mean arterial pressure, the underlying cardiac output and systemic vascular resistance varied widely among different patients, reflecting the physiologic heterogeneity and compensatory balance between flow and resistance. Cardiac output (rather than cardiac index) was used in these analyses (Figs.\u0026nbsp;5A\u0026ndash;5D) to directly correspond with the following physiologic equation in which cardiac output, not cardiac index, has been conventionally used:​\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:Mean\\:Arterial\\:Pressure=\\frac{Cardiac\\:Output\\times\\:Systemic\\:Vascular\\:Resistance}{80}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWe then evaluated intraoperative cardiac index trajectories in matched groups stratified by a baseline cardiac index threshold of 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e (Supplementary Table\u0026nbsp;3 and Fig.\u0026nbsp;5E). The cardiac index increased prior to anesthesia induction, likely reflecting anxiety-induced sympathetic activation, and decreased significantly following induction. Throughout the intraoperative period, cardiac index remained consistently higher in patients with a baseline index\u0026thinsp;\u0026ge;\u0026thinsp;3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e. Specifically, the intraoperative median cardiac index was 2.6\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6 L/min/m\u0026sup2; in the low index group and 2.7\u0026thinsp;\u0026plusmn;\u0026thinsp;0.7 L/min/m\u003csup\u003e2\u003c/sup\u003e in the high index group (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Minimum intraoperative cardiac index values were 1.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.4 L/min/m\u003csup\u003e2\u003c/sup\u003e and 1.6\u0026thinsp;\u0026plusmn;\u0026thinsp;0.4 L/min/m\u003csup\u003e2\u003c/sup\u003e (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05), respectively. These findings suggest that preoperative differences in cardiac index persist throughout surgery, despite a remarkable decrease in these differences.\u003c/p\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003eMediation analyses\u003c/h2\u003e \u003cp\u003eTo explore potential mechanisms underlying the association between preoperative cardiac output and postoperative AKI, we conducted mediation analyses focusing on intraoperative variables. These analyses consistently demonstrated a significant total effect of preoperative cardiac output on postoperative AKI yet revealed minimal or negligible indirect effects through intraoperative mediators (Supplementary Tables\u0026nbsp;8\u0026ndash;10). Specifically, variables such as intraoperative hemodynamics, oxygenation metrics, vasopressor use, fluid balance, and decreases in cardiac output did not meaningfully mediate the relationship. This pattern persisted across all modelled pathways and sensitivity analyses. Collectively, these findings suggest that the observed association between higher baseline cardiac output and increased risk of AKI is primarily attributable to direct effects, rather than being mediated by intraoperative physiological disturbances.\u003c/p\u003e \u003cdiv id=\"Sec32\" class=\"Section3\"\u003e \u003ch2\u003eExternal Validation in Patients Undergoing Cardiac Procedures\u003c/h2\u003e \u003cp\u003eTo evaluate the generalizability of our findings, we conducted an exploratory analysis using data from the MIMIC-IV v3.1 dataset. We identified 748 adult patients who underwent various cardiac procedures and had pre-procedure cardiac output measurements available (Supplementary Table\u0026nbsp;11). These procedures encompassed a broad range of cardiac interventions. Thermodilution via pulmonary artery catheter was the predominant method used to measure cardiac output (79.1%), followed by non-invasive bioreactance-based technology (18.2%), arterial pressure waveform analysis (1.5%), and transpulmonary thermodilution with pulse contour analysis (1.2%) (Supplementary Table\u0026nbsp;12). The heterogeneity of procedures and measurement modalities in the external cohort strengthens generalizability, offering more compelling validation than would be achieved in a uniform population.\u003c/p\u003e \u003cp\u003eUsing a preoperative cardiac index threshold of 3.0 L/min/m\u003csup\u003e2\u003c/sup\u003e, consistent with the primary analysis, 582 patients were classified as having a low cardiac index and 166 as having a high cardiac index. The incidence of postoperative AKI was 18.8% (109/582) in the low cardiac index group and 27.7% (46/166) in the high cardiac index group. Unadjusted analysis showed that patients with a high preoperative cardiac index had significantly increased odds of developing AKI (OR, 1.66; 95% CI, 1.11\u0026ndash;2.47; p\u0026thinsp;=\u0026thinsp;0.012) (Supplementary Table\u0026nbsp;13).\u003c/p\u003e \u003cp\u003eDue to substantial missingness, i.e., more than 50%, in key covariates used for propensity score modeling, we did not perform matched analyses in this cohort (Supplementary Table\u0026nbsp;14). Nevertheless, the direction and magnitude of the association were consistent with our findings based on the Bottomline-CS cohort. Importantly, patients in the low cardiac index group were at a higher risk of postoperative complications: they were older (mean age, 69 vs. 60 years), more likely to be male (45% vs. 34%), and had higher rates of myocardial infarction (34% vs. 19%), diabetes (40% vs. 30%), hypertension (79% vs. 57%), and greater comorbidity burden, as indicated by the age-adjusted Charlson Comorbidity Index (median 6 vs. 5) (Supplementary Table\u0026nbsp;15). Therefore, the findings from the MIMIC-IV cohort, although unadjusted, lend credibility to the observed association, as patients with lower baseline cardiac index, despite their higher comorbidity burden, demonstrated a lower incidence of AKI.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this large, prospective cohort study of patients undergoing elective off-pump coronary artery bypass grafting, we demonstrated that higher preoperative cardiac output is independently associated with a significantly greater risk of postoperative AKI. This association persisted across multiple analytic strategies, including propensity score matching and sensitivity analyses using alternative thresholds. Importantly, this unexpected finding was supported by an independent validation cohort derived from the MIMIC-IV v3.1 database. These results offer a novel and clinically important contribution to our understanding of perioperative hemodynamics and renal risk, and they challenge the conventional assumption that greater cardiac output necessarily implies improved renal perfusion and protection against AKI.\u003c/p\u003e \u003cp\u003eTo our knowledge, this is the first study to systematically investigate and demonstrate an inverse association between preoperative cardiac output and postoperative AKI in patients undergoing elective cardiac surgery. While reduced cardiac output has traditionally been viewed as a risk factor for renal hypoperfusion and injury, our findings reveal a paradoxical relationship: patients with higher resting cardiac output, measured under tightly standardized preoperative conditions, had a significantly higher risk of developing AKI following surgery. This counterintuitive observation necessitates a re-evaluation of long-standing physiological assumptions and suggests a new potential marker for perioperative risk stratification.\u003c/p\u003e\n\u003ch3\u003eChallenging a Traditional Paradigm\u003c/h3\u003e\n\u003cp\u003eHistorically, the prevailing pathophysiologic paradigm has posited that diminished cardiac output leads to decreased renal blood flow and perfusion pressure, triggering ischemic injury and contributing to the development of AKI.[\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] This model is particularly prominent in the heart failure literature, where low-output states are commonly associated with renal impairment. Numerous studies on patients with acute and chronic decompensated heart failure have highlighted reduced forward flow as a contributor to worsening renal function and adverse outcomes.[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e, \u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] Based on this premise, higher cardiac output would be expected to confer a protective effect on renal function.\u003c/p\u003e \u003cp\u003eHowever, our findings directly contradict this assumption. In this cohort of well-characterized, hemodynamically stable patients undergoing elective, off-pump CABG, those with higher cardiac output exhibited increased, rather than decreased, rates of postoperative AKI. This association remained evident despite balanced baseline characteristics, including mean arterial pressure, across comparison groups. Cardiac output was measured 24 to 48 hours prior to surgery under standardized, resting conditions, with patients in a supine position, calm, awake, breathing room air, and free from sedation or anesthesia. This provided a reliable snapshot of baseline physiology, independent of procedural or pharmacological influences.\u003c/p\u003e \u003cp\u003eThese data suggest that higher cardiac output in this context may not reflect superior cardiovascular function or more effective perfusion, but rather a maladaptive physiologic state. Potential explanations include systemic vasodilation, which requires compensatory increases in cardiac output to maintain perfusion pressure, impaired renal autoregulatory responses, or neurohormonal activation reflective of underlying physiologic stress. The fact that blood pressure was similar between groups further supports the interpretation that it is not perfusion pressure alone, but rather the broader hemodynamic configuration, that may predispose patients to AKI.\u003c/p\u003e \u003cp\u003eImportantly, caution is warranted when interpreting these findings in the context of prior studies. Unlike earlier investigations that were primarily based on patients with heart failure, our study population consisted of elective surgical patients without decompensated cardiac function. The mechanisms linking cardiac output to renal outcomes in heart failure, where forward flow is pathologically reduced and venous congestion plays a dominant role, may not fully translate to the perioperative setting of preserved cardiac function and controlled hemodynamics. Nonetheless, the paradoxical association we observed invites a re-evaluation of how cardiac output is conceptualized in relation to renal risk across diverse clinical populations.\u003c/p\u003e\n\u003ch3\u003eHemodynamic and Mechanistic Insights\u003c/h3\u003e\n\u003cp\u003eTo better understand this relationship, we examined the distribution of systemic vascular resistance and cardiac output in relation to mean arterial pressure. Our scatterplots revealed that, for comparable mean arterial pressures, patients exhibited a broad range of cardiac output and systemic vascular resistance combinations (Fig.\u0026nbsp;5D). This finding reinforces the concept that mean arterial pressure alone is insufficient to characterize hemodynamic state or predict renal perfusion adequacy. Notably, patients with higher cardiac output tend to exhibit lower systemic vascular resistance, supporting the hypothesis that high-output states may reflect compensatory responses to peripheral vasodilation, which in turn may affect renal perfusion heterogeneity or glomerular filtration dynamics.\u003c/p\u003e \u003cp\u003eDespite exploring a range of potential mediating factors, including intraoperative hypotension, cardiac index trajectories, tissue desaturation indices, vasoactive drugs, and fluids, we were unable to identify a definitive explanation for the observed association. Importantly, the cardiac output-AKI relationship appeared to be preoperatively determined and was not explained by intraoperative events. These findings underscore the complexity of the interplay between flow, pressure, resistance, and renal autoregulation and suggest that resting preoperative cardiac output may serve as a physiologic marker of latent renal vulnerability.\u003c/p\u003e \u003cdiv id=\"Sec36\" class=\"Section2\"\u003e \u003ch2\u003eConsistency with Heart Failure Literature\u003c/h2\u003e \u003cp\u003eAlthough our study is the first to demonstrate this association in the context of cardiac surgery, our findings are consistent with several investigations in heart failure populations, which have questioned the traditional model linking reduced cardiac output with renal dysfunction. For example, in patients diagnosed with heart failure undergoing pulmonary artery catheterization, no positive correlation was found between cardiac index and renal function.[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] On the contrary, a weak but statistically significant inverse relationship was observed, such that a higher cardiac index was associated with a lower estimated glomerular filtration rate.[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] Additional studies have similarly reported a lack of correlation, or even paradoxical associations, between cardiac output and renal function in patients with advanced heart failure.[\u003cspan additionalcitationids=\"CR25 CR26 CR27 CR28\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eIn one study of patients undergoing right heart catheterization for decompensated heart failure, central venous pressure, but not cardiac output, correlated with renal function.[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] Others have found that passive venous congestion may play a more critical role than forward flow in determining renal outcomes.[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] Collectively, these data suggest that renal dysfunction may be more closely tied to pressure dynamics, venous outflow resistance, or neurohormonal dysregulation than to absolute cardiac output.\u003c/p\u003e \u003cp\u003eOur study extends these insights to a new clinical context: elective cardiac surgery. Unlike heart failure cohorts, our patients were not in decompensated states and underwent off-pump procedures with proactive, prospective hemodynamic monitoring. Thus, our results provide novel evidence that the non-linear or inverse relationship between cardiac output and AKI risk may also apply to stable, surgical populations.\u003c/p\u003e \u003cdiv id=\"Sec37\" class=\"Section3\"\u003e \u003ch2\u003eValidation in an External Cohort\u003c/h2\u003e \u003cp\u003eThe external validation of our findings using the MIMIC-IV database adds credibility and reproducibility to our results. In the diverse cohort of 748 patients undergoing cardiac procedures with available cardiac output data, we observed a similar trend: patients with a pre-procedure cardiac index of \u0026ge;\u0026thinsp;3.0 L/min/m\u0026sup2; had a significantly higher incidence of post-procedure AKI compared to those with a lower cardiac index. Although propensity score matching was not feasible due to missing data, the direction and magnitude of the effect were consistent with our primary analysis. Moreover, the patients in the lower cardiac index group of the MIMIC cohort tended to be older and carried a greater comorbidity burden, including a higher prevalence of myocardial infarction, diabetes, hypertension, and elevated Charlson comorbidity scores. Despite these higher baseline risks, they experienced lower rates of AKI, suggesting that higher cardiac output may reflect a more complex or maladaptive physiology not fully captured by conventional risk stratifications.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec38\" class=\"Section2\"\u003e \u003ch2\u003eClinical Implications\u003c/h2\u003e \u003cp\u003eThese findings carry important clinical implications. First, they challenge the traditional paradigm that greater cardiac output is invariably beneficial for renal perfusion. Second, they suggest that preoperative cardiac output, when measured under resting conditions, may serve as a valuable tool for risk stratification in AKI following cardiac surgery. Third, our results underscore the limitations of using mean arterial pressure alone to assess renal perfusion risk, highlighting the need for a more integrated hemodynamic assessment.\u003c/p\u003e \u003cp\u003eGiven that AKI is a major contributor to perioperative morbidity, prolonged hospitalization, and long-term adverse outcomes, improving prediction and prevention strategies remains a clinical priority. Incorporating cardiac output into risk assessment models could enhance predictive accuracy and inform preoperative optimization strategies. Moreover, our findings prompt a re-evaluation of whether interventions aimed at increasing cardiac output, such as the use of inotropic agents or fluid loading, are universally appropriate in the perioperative period, especially in patients who already exhibit high-output states.\u003c/p\u003e \u003cdiv id=\"Sec39\" class=\"Section3\"\u003e \u003ch2\u003eStrengths and Limitations\u003c/h2\u003e \u003cp\u003eThis study has several notable strengths. It utilized high-quality, prospectively collected hemodynamic data acquired under standardized, controlled conditions. Cardiac output was measured with patients in a supine position, breathing room air, awake, and calm, minimizing physiologic variability and reducing the risk of confounding. Propensity score matching allowed for rigorous adjustment of baseline characteristics, and the consistency of findings across multiple thresholds and analytic strategies enhanced the internal validity of the observed association. Furthermore, external validation using an independent cohort supports the broader applicability of our results.\u003c/p\u003e \u003cp\u003eNonetheless, several limitations should be acknowledged. First, as an observational cohort study, the potential for residual confounding cannot be completely ruled out, despite robust statistical adjustments. Second, while cardiac output was measured preoperatively in a consistent manner, intraoperative hemodynamic monitoring employed different methodologies, and we were unable to establish causality or delineate precise mechanistic pathways. Third, we were unable to validate our findings in an external cohort with an identical surgical population and covariate set, largely due to the rarity of routine preoperative cardiac output measurement in non-research settings. This limitation highlights a broader challenge in integrating physiologic data into large-scale clinical datasets, underscoring the need for further research in comparable populations.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e\n\u003ch3\u003eFuture Directions\u003c/h3\u003e\n\u003cp\u003eFurther research is needed to clarify the mechanisms linking higher preoperative cardiac output to increased risk of AKI. Prospective studies that incorporate simultaneous assessments of renal perfusion, oxygenation, and biomarkers of tubular injury could help elucidate the underlying pathophysiology and identify potential mediators. In addition, randomized trials evaluating hemodynamic management strategies in patients with elevated cardiac output may help determine whether targeted modulation of flow or vascular resistance can mitigate the risk of AKI. Broader implementation of non-invasive cardiac output monitoring during preoperative assessment could also enable more accurate, physiology-based risk stratification and inform individualized perioperative care.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, this study provides the first evidence that higher preoperative cardiac output is associated with an increased risk of AKI after elective off-pump CABG. This counterintuitive finding challenges conventional assumptions and introduces a novel physiologic marker of renal risk. By integrating high-fidelity hemodynamic measurements, robust statistical analyses, and external validation, our work advances the understanding of AKI pathophysiology and offers a foundation for future research into personalised perioperative hemodynamic management. These findings inform a more nuanced approach to assessing and mitigating AKI risk in cardiac surgical patients.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was funded by the Tianjin Science and Technology Project (No. 20JCZDJC00810).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data, code, and other materials:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePatient-level data will be available collaboratively with approval by the corresponding authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis secondary analysis was approved by the Institutional Review Board of Tianjin Chest Hospital (approval date: February 1, 2024), and informed consent was waived due to the use of de-identified data.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003efor\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;publish:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable (secondary analysis of the Bottomline-CS trial; original registration: NCT04896736).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCheruku SR, Raphael J, Neyra JA, Fox AA. Acute Kidney Injury after Cardiac Surgery: Prediction, Prevention, and Management. Anesthesiology. 2023;139(6):880\u0026ndash;98.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarkouti K, Wijeysundera DN, Yau TM, Callum JL, Cheng DC, Crowther M, Dupuis J-Y, Fremes SE, Kent B, Laflamme C, et al. Acute Kidney Injury After Cardiac Surgery. Circulation. 2009;119(4):495\u0026ndash;502.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO\u0026rsquo;Neal JB, Shaw AD, Billings FT. Acute kidney injury following cardiac surgery: current understanding and future directions. Crit Care. 2016;20(1):187.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeacche M, Winkelmayer WC, Paul S, Lin J, Unic D, Rawn JD, Cohn LH, Byrne JG. Predicting survival in patients requiring renal replacement therapy after cardiac surgery. Ann Thorac Surg. 2006;81(4):1385\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCorredor C, Thomson R, Al-Subaie N. Long-Term Consequences of Acute Kidney Injury After Cardiac Surgery: A Systematic Review and Meta-Analysis. J Cardiothorac Vasc Anesth. 2016;30(1):69\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRonco C, Bellomo R, Kellum JA. Acute kidney injury. Lancet. 2019;394(10212):1949\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZarbock A, K\u0026uuml;llmar M, Ostermann M, Lucchese G, Baig K, Cennamo A, Rajani R, McCorkell S, Arndt C, Wulf H, et al. Prevention of Cardiac Surgery\u0026ndash;Associated Acute Kidney Injury by Implementing the KDIGO Guidelines in High-Risk Patients Identified by Biomarkers: The PrevAKI-Multicenter Randomized Controlled Trial. Anesth Analgesia. 2021;133(2):292\u0026ndash;302.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRonco C, Haapio M, House AA, Anavekar N, Bellomo R. Cardiorenal Syndrome. J Am Coll Cardiol. 2008;52(19):1527\u0026ndash;39.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang KV, Williams AW, Greene EL, Redfield MM. Acute decompensated heart failure and the cardiorenal syndrome. Crit Care Med 2008, 36(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun S, Ma F, Li Q, Bai M, Li Y, Yu Y, Huang C, Wang H, Ning X. Risk model for deaths and renal replacement therapy dependence in patients with acute kidney injury after cardiac surgery. Interact Cardiovasc Thorac Surg. 2017;25(4):548\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan J, Zhai W, Wu Z, Zhang Z, Wang T, Ren M, Liu Z, Sessler DI, Guo Z, Meng L, et al. Care guided by tissue oxygenation and haemodynamic monitoring in off-pump coronary artery bypass grafting (Bottomline-CS): assessor blind, single centre, randomised controlled trial. BMJ. 2025;388:e082104.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhwaja A. KDIGO clinical practice guidelines for acute kidney injury. Nephron Clin Pract. 2012;120(4):c179\u0026ndash;184.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalsh M, Devereaux PJ, Garg AX, Kurz A, Turan A, Rodseth RN, Cywinski J, Thabane L, Sessler DI. Relationship between intraoperative mean arterial pressure and clinical outcomes after noncardiac surgery: toward an empirical definition of hypotension. Anesthesiology. 2013;119(3):507\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaron RM, Kenny DA. The moderator\u0026ndash;mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Personal Soc Psychol. 1986;51(6):1173.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohnson AE, Bulgarelli L, Shen L, Gayles A, Shammout A, Horng S, Pollard TJ, Hao S, Moody B, Gow B. MIMIC-IV, a freely accessible electronic health record dataset. Sci data. 2023;10(1):1.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeng L. Heterogeneous impact of hypotension on organ perfusion and outcomes: a narrative review. Br J Anaesth. 2021;127(6):845\u0026ndash;61.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRay SC, Mason J, O'Connor PM. Ischemic Renal Injury: Can Renal Anatomy and Associated Vascular Congestion Explain Why the Medulla and Not the Cortex Is Where the Trouble Starts? Semin Nephrol. 2019;39(6):520\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagener G, Brentjens TE. Renal Disease: The Anesthesiologist's Perspective. Anesthesiol Clin North America. 2006;24(3):523\u0026ndash;47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eProwle JR, Molan MP, Hornsey E, Bellomo R. Measurement of renal blood flow by phase-contrast magnetic resonance imaging during septic acute kidney injury: A pilot investigation*. Crit Care Med. 2012;40(6):1768\u0026ndash;76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBobbio E, Bollano E, Polte CL, Ekelund J, R\u0026aring;degran G, Lundgren J, Hagg\u0026aring;rd C, Gjesdal G, Braun O, Bartfay S-E, et al. Association between central haemodynamics and renal function in advanced heart failure: a nationwide study from Sweden. ESC Heart Fail. 2022;9(4):2654\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarubelli V, Metra M, Lombardi C, Bettari L, Bugatti S, Lazzarini V, Dei Cas L. Renal dysfunction in acute heart failure: epidemiology, mechanisms and assessment. Heart Fail Rev. 2012;17(2):271\u0026ndash;82.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDamman K, Kalra PR, Hillege H. PATHOPHYSIOLOGICAL MECHANISMS CONTRIBUTING TO RENAL DYSFUNCTION IN CHRONIC HEART FAILURE. J Ren Care. 2010;36(s1):18\u0026ndash;26.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHanberg JS, Sury K, Wilson FP, Brisco MA, Ahmad T, Ter Maaten JM, Broughton JS, Assefa M, Tang WHW, Parikh CR, et al. Reduced Cardiac Index Is Not the Dominant Driver of Renal Dysfunction in Heart Failure. J Am Coll Cardiol. 2016;67(19):2199\u0026ndash;208.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNohria A, Hasselblad V, Stebbins A, Pauly DF, Fonarow GC, Shah M, Yancy CW, Califf RM, Stevenson LW, Hill JA. Cardiorenal interactions: insights from the ESCAPE trial. J Am Coll Cardiol. 2008;51(13):1268\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDamman K, Navis G, Smilde TD, Voors AA, van der Bij W, van Veldhuisen DJ, Hillege HL. Decreased cardiac output, venous congestion and the association with renal impairment in patients with cardiac dysfunction. Eur J Heart Fail. 2007;9(9):872\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDupont M, Mullens W, Finucan M, Taylor DO, Starling RC, Tang WH. Determinants of dynamic changes in serum creatinine in acute decompensated heart failure: the importance of blood pressure reduction during treatment. Eur J Heart Fail. 2013;15(4):433\u0026ndash;40.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuglin M, Rivero A, Matar F, Garcia M. Renal dysfunction in heart failure is due to congestion but not low output. Clin Cardiol. 2011;34(2):113\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMullens W, Abrahams Z, Skouri HN, Francis GS, Taylor DO, Starling RC, Paganini E, Tang WH. Elevated intra-abdominal pressure in acute decompensated heart failure: a potential contributor to worsening renal function? J Am Coll Cardiol. 2008;51(3):300\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMullens W, Abrahams Z, Francis GS, Sokos G, Taylor DO, Starling RC, Young JB, Tang WHW. Importance of venous congestion for worsening of renal function in advanced decompensated heart failure. J Am Coll Cardiol. 2009;53(7):589\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoppie TM, Serio AM, Vickers AJ, Vora K, Dalbagni G, Donat SM, Herr HW, Bochner BH. Age-adjusted Charlson comorbidity score is associated with treatment decisions and clinical outcomes for patients undergoing radical cystectomy for bladder cancer. Cancer. 2008;112(11):2384\u0026ndash;92.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eInker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, Crews DC, Doria A, Estrella MM, Froissart M, et al. New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race. N Engl J Med. 2021;385(19):1737\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHan J, Zhai W, Wu Z, Zhang Z, Wang T, Ren M, Liu Z, Sessler DI, Guo Z, Meng L. Care guided by tissue oxygenation and haemodynamic monitoring in off-pump coronary artery bypass grafting (Bottomline-CS): assessor blind, single centre, randomised controlled trial. BMJ. 2025;388:e082104.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cdiv class=\"gridtable\"\u003e\n\u003ctable id=\"Tab1\" border=\"1\"\u003e\u003ccaption\u003e\n\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n\u003cdiv class=\"CaptionContent\"\u003e\n\u003cdiv class=\"SimplePara\"\u003ePreoperative baseline characteristics of 1949 patients\u003c/div\u003e\n\u003c/div\u003e\n\u003c/caption\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eCharacteristics*\u003c/div\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eSummary\u003c/div\u003e\n\u003c/th\u003e\n\u003cth align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eMissingness, n\u003c/div\u003e\n\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eAge (year), mean (SD)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e69 (5)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eMale, n (%)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e1364 (70.0)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eHeight (cm), mean (SD)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e168 (8)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eBody mass index (kg/m\u003csup\u003e2\u003c/sup\u003e), mean (SD)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e25.3 (3.3)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eSmoking, n (%)\u0026dagger;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e1124 (57.7)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eDrinking, n (%)\u0026Dagger;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e498 (25.6)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eMyocardial infarction, n (%)\u0026sect;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e545 (28.0)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eArrhythmia, n (%)\u0026para;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e186 (9.5)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eDiabetes, n (%)#\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e843 (43.3)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eHypertension, n (%)**\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e1422 (73.0)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eCarotid artery disease, n (%)\u0026dagger;\u0026dagger;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e482 (24.7)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eAge adjusted CCI, median (IQR ) \u0026Dagger;\u0026Dagger;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e5 (4\u0026ndash;7)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eBeta blocker, n (%)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e1664 (85.4)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eACEI, n (%)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e63 (3.2)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eCalcium channel blocker, n (%)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e801 (41.1)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eAngiotensin receptor blocker, n (%)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e695 (35.7)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eAntidiabetic, n (%)\u0026sect;\u0026sect;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e899 (46.1)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eDiuretic, n (%)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e201 (10.3)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eMAP (mmHg), mean (SD)\u0026para;\u0026para;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e90 (12)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eC-reactive protein (mg/L), median (IQR)##\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e1.9 (0.8\u0026ndash;5.0)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e271\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eeGFR (mL/min/1.73 m\u003csup\u003e2\u003c/sup\u003e)), median (IQR)***\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e93 (81\u0026ndash;99)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003eGuided care group, n (%)\u0026dagger;\u0026dagger;\u0026dagger;\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"left\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e975 (50.0)\u003c/div\u003e\n\u003c/td\u003e\n\u003ctd align=\"char\" char=\".\"\u003e\n\u003cdiv class=\"SimplePara\"\u003e0\u003c/div\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eSD, standard deviation; CCI, Charlson comorbidity index; IQR, interquartile range; ACEI, angiotensin-converting enzyme inhibitors; MAP, mean arterial pressure; eGFR, estimated glomerular filtration rate\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e* Summary statistics are reported as n (%) for categorical variables and as mean (SD) or median (IQR) for continuous variables.\u003c/p\u003e\n\u003cp\u003e\u0026dagger; Smoking was defined as a self-reported history of tobacco use, as obtained during the preoperative interview.\u003c/p\u003e\n\u003cp\u003e\u0026Dagger; Drinking was defined as a self-reported history of alcohol consumption, excluding occasional or purely social drinking, as obtained during the preoperative interview.\u003c/p\u003e\n\u003cp\u003e\u0026sect; Myocardial infarction was defined as a prior event confirmed by clinical presentation, electrocardiographic changes, elevated cardiac biomarkers, or imaging evidence of myocardial infarction.\u003c/p\u003e\n\u003cp\u003e\u0026para; Arrhythmia was defined as an abnormal cardiac electrical rhythm that was symptomatic, required treatment, or prompted further diagnostic evaluation.\u003c/p\u003e\n\u003cp\u003e# Diabetes was defined as a documented clinical diagnosis established according to standard care practices.\u003c/p\u003e\n\u003cp\u003e** Hypertension was defined as a documented clinical diagnosis or recorded use of antihypertensive medication, in accordance with standard medical practice.\u003c/p\u003e\n\u003cp\u003e\u0026dagger;\u0026dagger; Carotid artery disease was defined as \u0026ge;50% stenosis in at least one carotid artery, as identified by preoperative carotid ultrasound screening.\u003c/p\u003e\n\u003cp\u003e\u0026Dagger;\u0026Dagger; The age-adjusted CCI quantifies comorbid conditions based on their number and severity, with additional points assigned for increasing age. The original CCI includes 17 weighted comorbidities, while the age-adjusted version incorporates age to improve risk stratification and predictive accuracy for clinical outcomes.[30]\u003c/p\u003e\n\u003cp\u003e\u0026sect;\u0026sect; Antidiabetic agents were defined as any pharmacologic treatments used for the management of established diabetes, including both orally administered medications and the use of insulin.\u003c/p\u003e\n\u003cp\u003e\u0026para;\u0026para; MAP was measured 24 to 48 hours before surgery with patients resting in a supine position, eyes closed, and breathing room air in a quiet ward environment.\u003c/p\u003e\n\u003cp\u003e## C-reactive protein values were missing for 271 patients (13.9%). Multiple imputations by chained equations (50 imputations) were performed using the R package \u0026lsquo;mice\u0026rsquo; with the Predictive Mean Matching method. The imputed datasets were combined using Rubin\u0026rsquo;s rules.\u003c/p\u003e\n\u003cp\u003e*** eGFR was calculated using the 2021 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation.[31]\u003c/p\u003e\n\u003cp\u003e\u0026dagger;\u0026dagger;\u0026dagger; Each patient was randomly assigned to either the guided care group or the usual care group as part of the Bottomline-CS randomized controlled trial, which was designed to evaluate the effectiveness of guided care interventions.[32]\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-anesthesiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bane","sideBox":"Learn more about [BMC Anesthesiology](http://bmcanesthesiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bane","title":"BMC Anesthesiology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"cardiac output, acute kidney injury, coronary artery bypass grafting, off-pump surgery, hemodynamics, risk stratification, hyperdynamic circulation, propensity score matching","lastPublishedDoi":"10.21203/rs.3.rs-8225284/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8225284/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe relationship between cardiac output and kidney function following cardiac surgery remains poorly defined. We aimed to evaluate the association between preoperative cardiac output and postoperative acute kidney injury (AKI) in patients undergoing off-pump coronary artery bypass grafting (CABG), a setting without exposure to cardiopulmonary bypass.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eThis cohort study included 1,949 patients aged\u0026thinsp;\u0026ge;\u0026thinsp;60 years from the Bottomline-CS trial who underwent elective off-pump CABG. Preoperative cardiac output was measured under standardized resting conditions, 1\u0026ndash;2 days prior to surgery. The primary outcome was AKI within the postoperative 7 days, defined according to KDIGO serum creatinine criteria. Propensity score-matched analyses were performed to compare the risk of AKI between patients with low and high baseline cardiac output. Restricted cubic spline models were used to examine the continuous relationship between cardiac output and the risk of AKI. The findings were validated in an external cohort to assess their robustness.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eOverall AKI incidence was 11% (213/1,949). In matched analyses, a preoperative cardiac index (CI)\u0026thinsp;\u0026ge;\u0026thinsp;3.0 L/min/m\u0026sup2; was associated with higher AKI than CI\u0026thinsp;\u0026lt;\u0026thinsp;3.0 (14.3% vs 9.0%; OR 1.69; 95% CI, 1.15\u0026ndash;2.48). A similar association was observed for cardiac output\u0026thinsp;\u0026ge;\u0026thinsp;5.0 L/min versus \u0026lt;\u0026thinsp;5.0 (13.3% vs 8.9%; OR 1.56; 95% CI, 1.09\u0026ndash;2.23). Spline analyses showed a J-shaped relationship, with rising risk above CI about 3.0 L/min/m\u0026sup2; and cardiac output about 5.0 L/min. High-output patients had lower systemic vascular resistance with similar mean arterial pressure. Mediation analyses found no explanatory effect of intraoperative hypotension, vasopressor use, oxygenation deficits, fluid balance, or intraoperative cardiac output change. Findings were supported in an external cohort.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eHigher, not lower, preoperative cardiac output is independently associated with increased risk of postoperative AKI in patients undergoing off-pump CABG.\u003c/p\u003e","manuscriptTitle":"Inverse Association Between Preoperative Cardiac Output and Postoperative Kidney Function in Off-Pump Coronary Artery Bypass Grafting","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-17 08:56:40","doi":"10.21203/rs.3.rs-8225284/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-23T10:11:02+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-20T21:33:03+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"122095509171747962860455633302859419196","date":"2026-02-01T21:15:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"269984514727159411922495115197820374818","date":"2026-01-05T05:25:24+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-22T16:07:42+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"297110178383187776653073851135813125126","date":"2025-12-18T05:32:49+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-12T11:03:13+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-12-10T05:26:39+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-09T13:40:27+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-09T13:39:35+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Anesthesiology","date":"2025-11-27T23:09:54+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-anesthesiology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bane","sideBox":"Learn more about [BMC Anesthesiology](http://bmcanesthesiol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bane","title":"BMC Anesthesiology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"01021288-2b5b-4b10-93c6-e1f04c3b40cc","owner":[],"postedDate":"December 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-23T20:38:12+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-17 08:56:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8225284","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8225284","identity":"rs-8225284","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}