C57BL/6 mice (25–30 g, male) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China; production license SCXK [Hu] 2022-0004). The mice were housed under specific-pathogen-free (SPF) conditions with a 12 h light/12 h dark cycle, and had ad libitum access to irradiated standard rodent chow and autoclaved tap water. All experimental procedures were approved by the Animal Ethics Committee of Zhejiang University (approval No. ZJU20240505) and were conducted in accordance with the international Guidelines for the Care and Use of Laboratory Animals.

Eighteen male C57BL/6 mice (25–30 g) were used in this study and randomly divided into three groups (n = 6 per group): the Control group, which received no procedural intervention; the Sham group, which underwent surgical exposure of the jugular vein and femoral artery without cannulation, to account for the effects of anesthesia and surgical trauma; in brief, mice in the Sham group received the same preoperative preparation as the ECC group, including anesthesia and surgical dissection/exposure of the jugular vein and femoral artery. No vascular cannulation was performed and the extracorporeal circuit was not connected or initiated and the ECC group, in which the exposed vessels were cannulated to conduct ECC as the experimental model under investigation.

The murine extracorporeal circulation (ECC) system was designed as a miniaturized, closed-loop veno-arterial circuit providing partial circulatory support, operating in parallel with an intact, beating native heart rather than replacing total cardiac output. This configuration was intended to model partial extracorporeal circulatory assistance rather than full cardiopulmonary bypass. The circuit is primarily composed of the following components, connected in series: The circuit sequence was: right external jugular vein (drainage) → pump → membrane oxygenator → blood storage room (mini-reservoir) → left femoral artery (return). A miniature peristaltic pump (YZ1515, Baoding Qili Constant Flow Pump Co., Ltd., China) served as the artificial heart, providing the motive force for blood circulation. The pump was driven by its corresponding drive unit (BT100-02, Baoding Qili Constant Flow Pump Co., Ltd., China), which allowed for precise digital control of the flow rate. According to the manufacturer’s specifications, the pump operates under a wide flow range (0.07–2,400 mL/min, ≤600 rpm), enabling accurate regulation of the low-flow conditions required for murine ECC. The pump head was fitted with a segment of pharmaceutical-grade pump tubing (Pharmed 13#, Baoding Qili Constant Flow Pump Co., Ltd., China), which featured a proprietary anticoagulant coating on its inner surface to enhance hemocompatibility. The outlet of the pump tubing was connected to a miniature membrane oxygenator (Micro-MO, Dongguan Kewei Medical Device Co., Ltd., China) for gas exchange, effectively functioning as an artificial lung. According to the manufacturer, the oxygenator consists of a polycarbonate housing with polypropylene hollow-fiber membranes and provides a total membrane surface area of approximately 0.05 m². The oxygenator demonstrated stable gas exchange performance within the experimental flow range applied in this study (2.5–3.5 mL/min/100 g body weight). The oxygenated blood was then passed through a standard disposable medical three-way stopcock and a positive-pressure sterile infusion set (Guangdong Baihe Medical Technology Co., Ltd., NMPA Registration No. 20223140499) for fluid administration and sample collection. For vascular access, a 26G intravenous catheter (Shanghai Kindly Medical Group, NMPA Approval No. 20193141674 Co., Ltd., China) was used for cannulation of the right external jugular vein as the drainage (venous) line. A component labeled as “blood storage room” was incorporated into the circuit upstream of the arterial return line. It was assembled using a sterile micro 3-way Y connector (22G) (WP32001, Instech Laboratories, Inc., United States) connected to a modified 1-mL Luer-Lok™ syringe barrel (BD 309628, Becton, Dickinson and Company (BD), United States), forming a closed mini-reservoir. The left femoral artery was similarly cannulated using a 26G intravenous catheter of the same specification (Shanghai Kindly Medical Group, NMPA Approval No. 20193141674 Co., Ltd., China), which served as the return (arterial) line. The circuit was interconnected using 22G PE/PVC tubing (btcoex-22, Beijing Mingtai Jiaxin Technology Co., Ltd., China) to complete the closed-loop path. All components were meticulously assembled and primed prior to the procedure to ensure a sterile, bubble-free, and closed system.

Circuit-Derived mixed venous blood sample (approximately 100 μL each) for evaluating systemic oxygenation, ventilation, and metabolic status during ECC were collected at three specified time points: baseline (BL), post anesthetic stabilization, 30 min post ECC initiation (T30), and 60 min post ECC initiation (T60) (Figure 2). The samples were obtained using heparinized syringes connected to the venous line through a three-way stopcock, ensuring the immediate removal of any air bubbles upon collection. Analysis of all samples was conducted within 60 s of collection using a Radiometer ABL90 FLEX blood gas analyzer (Radiometer, Denmark), which underwent daily automatic calibration. The parameters measured included pH, bicarbonate (HCO₃ ⁻), base excess (BE), lactate, and potassium (K⁺), with all values reported at 37 °C. It should be emphasized that these measurements reflect mixed venous (pre-oxygenator) blood gas and electrolyte composition rather than true systemic arterial values, and are therefore interpreted as indicators of global metabolic status rather than arterial oxygenation performance.

To evaluate the dynamic changes in blood composition during ECC, additional blood samples were collected in heparin-anticoagulated tubes at the same time points as the blood gas analyzes: baseline, 30 min on ECC, and 60 min on ECC. Following blood gas measurement, the remaining sample in the heparinized syringe was gently inverted several times to prevent clotting. A 20 μL aliquot was then aspirated and analyzed using a URIT-290 Vet Plus automated hematology analyzer (URIT Medical Electronic, China) to determine the counts of white blood cells (WBC), red blood cells (RBC), and platelets (PLT) at each stage. Daily quality control procedures were conducted using manufacturer-matched calibrators to ensure the accuracy and reliability of all measurements.

The above procedure outlines the perfusion circuit, surgical procedures, and monitoring of physiological parameters during ECC in a murine model. When performed by a microsurgeon with adequate expertise, the results are consistently reproducible.

Plasma samples were collected from mice, and the levels of pro-inflammatory cytokines (IL-1β, IL-6, TNF-α, and IL-18) were quantified using commercially available enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturers’ instructions. Briefly, whole blood was collected into anticoagulant-treated tubes and centrifuged at 1,500–2,000 × g for 10 min at 4 °C to obtain plasma. The plasma samples were aliquoted and stored at −80 °C until analysis. ELISA kits were obtained as follows: IL-1β (Cat. No. 88-7013a, Invitrogen, Thermo Fisher Scientific, United States), IL-6 (Cat. No. DY406, R&D Systems, United States), TNF-α (Cat. No. 88-7324, Invitrogen, Thermo Fisher Scientific, United States), and IL-18 (Cat. No. SEK50073, Sino Biological Inc., China). All samples were analyzed strictly according to the manufacturers’ instructions.

Organ injury–related biomarkers in mouse plasma were measured using commercially available assay kits according to the manufacturers’ instructions, with plasma samples prepared as described above. Creatinine (Cr) was measured to assess renal injury (Cat. No. C011-2-1, Nanjing Jiancheng Bioengineering Institute, China); cardiac troponin and lactate dehydrogenase (LDH) were used to evaluate cardiac injury (Cat. No. EM30867S, Weiao Biotechnology, China; Cat. No. A020-4-1, Nanjing Jiancheng Bioengineering Institute, China); aspartate aminotransferase (AST/GOT) was measured as a marker of hepatic injury (Cat. No. C010-2-1, Nanjing Jiancheng Bioengineering Institute, China); and myeloperoxidase (MPO) activity was determined to assess pulmonary injury (Cat. No. A044-1-1, Nanjing Jiancheng Bioengineering Institute, China). All samples were analyzed strictly according to the manufacturers’ instructions.

Statistical analyzes were performed using GraphPad Prism software (version 10.1.2; GraphPad Software, San Diego, CA, United States). All data are presented as mean ± standard error of the mean (SEM). Prior to parametric testing, data normality was assessed using the Shapiro–Wilk test, and homogeneity of variance was evaluated using Levene’s test (or Brown–Forsythe test, where appropriate). All datasets satisfied the assumptions required for analysis of variance. For comparisons among the three experimental groups (Control, Sham, and ECC) at the final time point (60 min), a one-way analysis of variance (ANOVA) was performed, followed by Tukey’s post hoc test for multiple comparisons. For comparisons across time within the ECC group (0, 30, and 60 min), a one-way repeated-measures ANOVA was applied. Sphericity was assessed using Mauchly’s test, and when violations were detected, the Greenhouse–Geisser correction was applied automatically. No data points were excluded from the final analyzes. All measurements were successfully obtained at all predefined time points, therefore, no missing data imputation was required.

To assess the ability of this system to replicate the hematologic changes induced by ECC, we first conducted a complete blood cell analysis. Following the initiation of ECC in mice, we noted rapid changes in peripheral leukocyte populations. During the initial 60 min, total white blood cell counts (WBC) increased in a time-dependent manner, primarily due to a concurrent rise in neutrophils, which exhibited significant pairwise differences between time points (Figure 3A). Conversely, lymphocyte numbers decreased during this period, while monocytes demonstrated a modest yet consistent increase (Figure 3B). Comparisons among endpoint groups confirmed these trends: relative to sham and non-instrumented controls, ECC mice displayed pronounced leukocytosis and neutrophilia, alongside reduced lymphocyte counts and elevated monocyte levels (Figures 3C,D).

Collectively, these data suggest that ECC triggers an early, innate-skewed hematologic response characterized by rapid neutrophil mobilization and relative lymphopenia, which becomes apparent within the first hour following circuit initiation. This significant leukocyte remodeling reflects the acute inflammatory cascade observed in clinical ECC scenarios, where neutrophil activation and lymphoid suppression contribute to postoperative complications, including systemic inflammatory response syndrome (SIRS) (Misokalou et al., 2025). The pronounced increase in neutrophils, alongside the decrease in lymphocytes and moderate elevation of monocytes, indicates that our murine ECC model accurately replicates the hallmark hematologic and inflammatory changes associated with clinical ECC.

Building on the characterization of ECC-induced inflammatory leukocyte dynamics, we subsequently examined the hemostatic changes occurring during early ECC. Within 60 min of initiating ECC, both plateletcrit (PCT) and platelet count (PLT) exhibited significant declines (Figure 4A), with endpoint analyzes confirming marked thrombocytopenia in the ECC group compared to sham and non-instrumented controls (Figure 4B). This swift platelet loss reflects the clinical context of ECC-associated thrombocytopenia, where artificial surfaces and shear stress promote platelet activation and sequestration. In contrast, mean platelet volume (MPV) and the platelet–large cell ratio (P-LCR) increased during early ECC (Figure 4C), suggesting a transition toward larger, more reactive platelets. Furthermore, terminal assessments revealed a significant increase in P-LCR among ECC animals (Figure 4D), indicating an increased proportion of larger circulating platelets at the endpoint.

This inverse relationship—thrombocytopenia associated with macrothrombocytosis—is consistent with a state of high platelet turnover, indicating ongoing platelet activation on the surface of the artificial circuit and compensatory hematopoietic activity (Wang et al., 2021). However, based on the current data, it is not possible to definitively distinguish whether the observed platelet phenotype primarily reflects peripheral platelet consumption, enhanced platelet clearance, or activation of compensatory thrombopoietic pathways. These alterations create a prothrombotic and proinflammatory environment that predisposes to end-organ injury, as activated platelets not only facilitate thrombus formation but also enhance the inflammatory cascade through cytokine release and leukocyte interaction (Tucker et al., 2021).

Accordingly, future studies will incorporate direct markers of thrombopoiesis and platelet production, including analysis of the immature platelet fraction (IPF), measurement of circulating thrombopoietin (TPO) levels, and, where feasible, bone marrow–based assessments of megakaryopoiesis. Integration of these complementary readouts will allow direct discrimination between peripheral platelet consumption and true compensatory thrombopoietic responses, thereby providing a more mechanistically grounded interpretation of platelet dysregulation during extracorporeal circulation.

Building on the insights regarding ECC-induced hematologic and hemostatic alterations, we further examined the acid-base and metabolic profiles to elucidate the systemic metabolic stress experienced during early ECC. ECC induced a critical state of compensated metabolic stress, characterized by tissue hypoperfusion and systemic inflammation that promoted anaerobic metabolism, while physiological buffers preserved acid-base homeostasis (Jentzer et al., 2022). During the initial hour of ECC, blood lactate (cLac⁺) progressively increased (Figure 5C, left), indicating a significant oxygen debt at the tissue level, which is a hallmark of impaired tissue perfusion and mitochondrial dysfunction. Nevertheless, this metabolic burden was effectively mitigated by integrated compensatory mechanisms: arterial pH and plasma bicarbonate (cHCO₃ ^-(P)) remained stable across all time points (Figure 5A), with no significant differences observed among groups at the endpoint (Figure 5B). The consistently negative yet stable values of actual and standard base excess (ABE.c and SBE.c) throughout the procedure (Figure 5D) further underscored a state of fully compensated metabolic acidosis.

This dissociation between rising lactate and preserved pH illustrates a precarious equilibrium—early lactate accumulation signals underlying hypoperfusion and mitochondrial dysfunction, while the intact acid-base parameters demonstrate the remarkable capacity of systemic buffers (including bicarbonate and hemoglobin) to mitigate overt acidemia during the initial phase of ECC (Jentzer et al., 2022). The elevated endpoint lactate in the ECC group compared to controls (Figure 5C, right) indicates the presence of early metabolic stress during short-term ECC, preceding overt acid–base imbalance or hemodynamic instability. Collectively, these data demonstrate that our murine ECC model faithfully recapitulates the early metabolic signature of clinical ECC, characterized by compensated metabolic acidosis driven by tissue hypoperfusion.

Building on the characterization of inflammatory, hemostatic, and metabolic responses, we further examined electrolyte and hematologic dynamics to elucidate the systemic homeostatic breakdown during ECC—providing key evidence that validates the clinical relevance of our model. Serum potassium (cK⁺) in the ECC group demonstrated a significant time-dependent increase (Figure 6A), with endpoint levels markedly higher than those observed in the control and sham groups (Figure 6B), indicating progressive hyperkalemia. This finding has profound physiological implications: elevated extracellular potassium disrupts cardiac membrane potential, impeding action potential propagation and heightening the risk of arrhythmias—characteristics of clinical ECC-associated cardiac electrical instability. Such elevations in extracellular potassium can depolarize cardiomyocyte resting membrane potential, slow impulse conduction, and increase arrhythmogenic susceptibility, consistent with cardiac electrical instability observed in clinical extracorporeal support settings. In contrast, serum sodium (cNa⁺) remained stable across groups and time points (Figures 6A,B), arguing against generalized hemodilution as the primary driver and supporting a potassium-predominant disturbance.

Time-course analysis of chloride (cCl⁻) and ionized calcium (cCa²⁺) showed a mild upward trend in cCl⁻ and a slight downward trend in cCa²⁺ within the ECC group, neither reaching statistical significance (Figure 6C). At the endpoint, cCl⁻ remained modestly elevated without significance, whereas cCa²⁺ in the ECC group was significantly lower than in controls (P < 0.05) and closely aligned with values observed in the sham group (Figure 6D). Ionized calcium is essential for excitation–contraction coupling via troponin C activation and sarcomeric cross-bridge formation; thus, reduced cCa²⁺ may contribute to impaired myocardial contractility and heightened cardiac vulnerability during ECC (Jeong and Lim, 2020). Together, the coexistence of hyperkalemia and reduced ionized calcium represents a combined electrophysiological and contractile stressor that may predispose to arrhythmia and pump dysfunction during extracorporeal support.

Overall, the electrolyte profile observed here—pronounced hyperkalemia with concomitant reduction in ionized calcium in the setting of stable sodium—supports a pattern of selective homeostatic disruption during ECC.

Given that hemodilution is a recognized potential confounder in ECC studies, particularly in small-animal models, we assessed whether dilutional effects occurred during ECC and whether they could influence the interpretation of hematologic, metabolic, and electrolyte data.

Dilution-sensitive hematologic parameters were analyzed at baseline (0 min), 30 min, and 60 min following ECC initiation, including red blood cell (RBC) count, hemoglobin (Hgb) concentration, red cell distribution width (RDW-CV and RDW-SD), mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH) (Figure 7). Throughout the 60-min ECC period, RBC counts and hemoglobin levels remained stable, with no significant temporal changes or differences compared with sham-operated or non-instrumented control groups (Figure 7C), indicating the absence of overt dilutional anemia. RDW-CV and RDW-SD values also remained unchanged (Figure 7D), suggesting preserved erythrocyte size homogeneity and arguing against substantial hemolysis, fluid overload, or pathological red blood cell turnover. Similarly, MCV and MCH showed no significant temporal or intergroup differences (Figures 7A,B), supporting the absence of meaningful erythrocyte swelling, shrinkage, or intracellular hemoglobin dilution. Collectively, these findings indicate that erythrocyte count, size distribution, and hemoglobin content remained stable during short-term ECC.

Taken together, these hematologic data are consistent with the experimental design features intended to limit dilutional effects, including the short ECC duration (60 min) and restricted blood sampling volumes. Under the present conditions, systemic hemodilution was effectively controlled and did not represent a dominant contributor to the observed metabolic or electrolyte alterations, such as lactate accumulation or electrolyte imbalance described above.

Nevertheless, hemodilution remains an inherent limitation of the current model. To further reduce blood loss and improve animal welfare in future studies, we plan to adopt advanced microsampling and analytical approaches, such as micro-volume blood gas analyzers and multiplex assays requiring minimal sample volumes. These strategies will allow precise physiological and biochemical assessment while substantially reducing per-sample blood volume, thereby enhancing data fidelity and animal welfare.

To further characterize the early systemic responses elicited by short-term ECC, we next examined circulating inflammatory mediators and organ injury–associated biochemical markers at the 60-min ECC time point (Figure 8).

As shown in (Figure 8A), ECC induced a robust early inflammatory response, reflected by significantly elevated plasma levels of pro-inflammatory cytokines, including interleukin-1β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and interleukin-18 (IL-18), compared with both sham-operated and non-instrumented control groups. These findings indicate rapid activation of innate immune signaling pathways following blood exposure to artificial circuit surfaces during ECC.

In parallel, we assessed a panel of plasma biomarkers associated with early organ stress or injury across multiple organ systems (Figure 8B). Renal stress was evidenced by a significant increase in plasma creatinine levels in the ECC group. Cardiac involvement was reflected by elevated circulating cardiac troponin and lactate dehydrogenase (LDH), indicating early myocardial cellular stress and membrane permeability alterations. Hepatic stress was suggested by increased plasma aspartate aminotransferase (AST/GOT) activity, while pulmonary involvement was indicated by a significant elevation in myeloperoxidase (MPO) activity, consistent with early neutrophil activation and pulmonary inflammatory signaling.

Importantly, these biochemical alterations were detected within a short, 60-min ECC exposure window, and therefore represent early inflammatory activation and organ injury–associated signals rather than established or irreversible organ dysfunction. Accordingly, the observed biomarker changes should be interpreted as early indicators of organ stress and inflammatory engagement, consistent with the acute phase of ECC exposure.

Collectively, these results demonstrate that short-term ECC in this murine model is sufficient to trigger rapid systemic inflammation accompanied by measurable biochemical signals associated with early organ stress (kidney, heart, liver, and lung). These findings provide biochemical evidence of early systemic and organ-specific responses to ECC, distinct from definitive multiorgan failure or terminal organ injury.