Effects of remote ischemic preconditioning on postoperative gastrointestinal function in patients undergoing off-pump coronary artery bypass grafting: a randomized controlled trial Authors: Wei Yang, Qinghai Liu, Xiaohua Wang, Wei Han, Hongli Li, Tianlong Wang Published: 2026 ABSTRACT AbstractBackgroundPostoperative gastrointestinal function is one of the important factors influencing the prognosis of off-pump coronary artery bypass grafting (OPCAB) patients. Remote ischemic preconditioning (RIPC) is a noninvasive intervention that may offer multiorgan protection. This study aimed to evaluate the impact of preoperative RIPC on postoperative gastrointestinal function and clinical outcomes in OPCAB patients.MethodsThis prospective, assessor- and statistician-blinded, sham-controlled, randomized clinical trial included 106 patients scheduled to undergo elective OPCAB. They were randomly assigned to the RIPC (n = 53) or sham-RIPC (n = 53) group. The RIPC group received bilateral upper limb ischemia/reperfusion cycles (200 mmHg, 5-min ischemia/5-min reperfusion, repeated for 5 cycles) twice daily for 3 consecutive preoperative days. The primary outcome was the incidence of postoperative gastrointestinal intolerance (POGI), defined as an I-FEED score of 3–5 within 7 days postoperatively. The secondary outcomes included markers of intestinal injury (intestinal fatty acid-binding protein (I-FABP)), systemic inflammation (interleukin-6 (IL-6), C-reactive protein (CRP)), and S-100 protein beta subunit (S-100β). The exploratory outcomes comprised gut microbiota analysis and other clinical parameters.ResultsThe incidence of POGI was lower in the RIPC than in the sham-RIPC group (35.4% vs. 56.5%, P = 0.040). The postoperative serum levels of I-FABP, IL-6, CRP, and S-100β were also lower in the RIPC group. Moreover, the relative abundance of Gemella and Dorea decreased in the RIPC group postoperatively. Notably, the abundance of Gemella showed a positive correlation with the postoperative levels of I-FABP and IL-6, whereas that of Dorea exhibited a positive correlation with the postoperative I-FABP levels.ConclusionsThe preoperative application of RIPC was associated with a lower risk of POGI, with no observed safety concerns. This effect of RIPC may be related to attenuated intestinal mucosal injury, reduced systemic inflammation, and potential modulation of the gut microbiota.Trial registrationThis study was registered on May 30, 2024 (ChiCTR2400085044).Supplementary InformationThe online version contains supplementary material available at 10.1186/s12871-026-03980-z.Keywords: Remote ischemic preconditioning, Off-pump coronary artery bypass grafting, Postoperative gastrointestinal intolerance, Postoperative gastrointestinal function Background Off-pump coronary artery bypass grafting (OPCAB) is currently performed in approximately 20% of all coronary artery bypass grafting (CABG) procedures worldwide. As the number of high-risk patients unsuitable for on-pump CABG increases, the demand for OPCAB gradually rises [1, 2]. As an irreplaceable classic surgical approach for the treatment of coronary atherosclerotic heart disease, OPCAB avoids respiratory and renal complications associated with cardiopulmonary bypass [1]. However, postoperative gastrointestinal dysfunction remains one of the most severe complications following OPCAB [3]. Intraoperative hemodynamic fluctuations caused by cardiac displacement and stabilizer compression are important contributors to intestinal injury, as they lead to splanchnic venous congestion and insufficient intestinal arterial perfusion [4, 5]. Complications during OPCAB, such as mesenteric ischemia, cardiac arrest, and hemorrhagic shock, can exacerbate intestinal injury [6]. Pain stimulates the sympathetic nervous system, while analgesic drugs activate intestinal opioid receptors, further inhibiting gastrointestinal motility [7]. When the intestinal epithelial and endothelial barriers are disrupted, bacteria and endotoxins translocate to the systemic circulation. The resulting interactions among inflammatory cell activation, immune cell activation, intestinal secretory cells, and the gut microbiota may create a self-perpetuating vicious cycle [8, 9]. Therefore, clinical studies in recent years have shown that OPCAB leads to poor early postoperative recovery of intestinal function [10], severe gastric mucosal hypoxia [11], and significant intestinal hypoperfusion [4]. Based on the I-FEED score, in the early postoperative period following OPCAB, postoperative gastrointestinal intolerance (POGI) may progress to the more severe condition [12], which is significantly associated with adverse clinical outcomes and a very high all-cause mortality rate [3]. Furthermore, the level of intestinal fatty acid-binding protein (I-FABP), a specific marker of intestinal epithelial cell injury, is significantly elevated after OPCAB, further suggesting the intestinal injury during OPCAB [4]. Therefore, interventions targeting POGI in OPCAB patients have become a significant clinical issue that warrants further investigation. Remote ischemic preconditioning (RIPC) is a non-pharmacological intervention that elicits endogenous protective mechanisms in the body by subjecting limbs to brief, non-invasive ischemia-reperfusion, thereby exerting protective effects on remote vital organs [13]. Characterized by its effectiveness, non-invasiveness, low cost, and multi-organ protection with minimal risk, RIPC achieves its protective effects through neural pathways, humoral factor release, and systemic anti-inflammatory and anti-apoptotic response induction [14]. Current research has indicated the protective effects of RIPC on multiple organs, including the heart [15], liver [16], lungs [17], brain [18], kidneys [19], and gastrointestinal tract [20]. As mentioned above, cardiac manipulation during OPCAB can impair abdominal organ perfusion, and the gut is highly vulnerable [4]. RIPC may activate endogenous protective mechanisms and has been shown to alleviate intestinal mucosal injury and preserve barrier function [14, 20]. However, evidence on RIPC for postoperative gastrointestinal function in OPCAB patients remains limited, especially regarding POGI and its effects on gut microbiota. This prospective, randomized, assessor- and statistician-blinded, sham-controlled clinical trial aimed to systematically evaluate the impact of RIPC on POGI and clinical outcomes in patients with OPCAB, and to explore whether RIPC could serve as a potential gastrointestinal protection strategy for OPCAB and its possible mechanisms. Methods Ethics This study was conducted as a prospective, assessor- and statistician-blinded, randomized, sham-controlled clinical trial. Written informed consent was obtained from all participants or their authorized representatives. The Ethics Committee of Xuanwu Hospital, Capital Medical University, reviewed and approved the study protocol (Approval No: Lin Yan Shen [2024] 110-001). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Furthermore, the study was registered with the Chinese Clinical Trial Registry (ChiCTR2400085044) (http://www.chictr.org.cn). Study design and participants This study included patients scheduled to undergo OPCAB surgery under general anesthesia at Xuanwu Hospital, Capital Medical University, China. The inclusion criteria were as follows: age 18–80 years, American Society of Anesthesiologists physical status classifications II-IV, New York Heart Association functional classes II-III, and body mass index between 22 and 30 kg·m2. The exclusion criteria were as follows: age > 80 years; the need for concurrent other major cardiovascular or carotid artery surgeries; presence of active systemic infection, uncontrolled immune system disease or peripheral vascular disease involving the upper limbs; combined severe hepatic, renal, pulmonary and brain dysfunction; and preoperative use of mechanical assist devices, such as an intra-aortic balloon pump (IABP), or extracorporeal membrane oxygenation. The discontinuation criteria included inability to tolerate the RIPC or sham-RIPC procedure and objective signs of tissue or neurovascular injury attributable to the intervention; changes in surgical plan (temporary cancellation of surgery or conversion to on-pump procedure) or requirement for postoperative reoperation; intraoperative special events, such as severe allergic reactions, major hemorrhage, or other significant hemodynamic instability; postoperative inability or refusal by the subject to undergo scale assessments; voluntary request to withdraw; or loss to follow-up. Randomization, intervention, and blinding Eligible patients were randomly assigned in a 1:1 ratio to either the RIPC or sham-RIPC group using a computer-generated block randomization sequence. The randomization sequence was concealed using sequentially numbered, opaque, sealed envelopes, kept by independent research personnel not involved in trial conduct. After screening and providing written informed consent, eligible subjects were randomized. Treatment allocation remained concealed until the intervention was administered. Due to the need for pressure setting of the RIPC device, trained medical staff were informed of the intervention measures. Due to the distinct inflation pressures (200 mmHg vs. 30 mmHg) applied twice daily for three consecutive days, complete blinding of participants was not feasible. However, outcome assessors, statisticians, and all clinical care teams (surgeons, anesthesiologists, ICU staff) were blinded to group assignment throughout the trial. The medical staff involved in RIPC procedures were excluded from all subsequent treatment and statistical processes after patient participation. Assignment data were kept separate from the clinical database until completion of the final analysis to preserve masking. An independent statistician performed the final analysis after database lock and unblinding. Outcome assessors did not participate in patient care or other trial procedures to maintain masking. Assessors were trained to use the published scoring systems according to a standardized protocol to minimize assessment bias and the risk of unmasking. Emergency unblinding was permitted only if required for the clinical management of a serious adverse event potentially related to the study intervention. In such cases, the treatment allocation was revealed by the principal investigator via a secure, password-protected system. All unblinding events were documented. Estimation of sample size Sample size was calculated for the primary outcome (POGI incidence). Based on pilot data, we assumed POGI rates of 53% in the RIPC group and 80% in the sham group. With a two-sided alpha of 0.05, power of 80%, and a 10% dropout rate, a total of 106 patients (53 per group) was required. Detailed calculations are provided in Supplementary 2. Protocol and intervention design In the RIPC group, a dual-chamber automatic cycle device (Product Name: Ischemic Preconditioning Training Device, Registration Certificate Number: Su Xie Zhu Zhun 20162090668, Registered Company: Beijing Renqiao Cardiovascular and Cerebrovascular Disease Prevention and Treatment Research Jiangsu Co., Ltd., Machine Model: IPC-906D) with a standard upper-arm cuff was used. The intervention was performed twice daily (8:00–10:00 and 14:00–16:00) for 3 consecutive preoperative days. Each session consisted of five cycles: 5-minute cuff inflation to 200 mmHg (confirmed by loss of bilateral radial pulses), followed by 5-minute complete deflation (confirmed by return of bilateral radial pulses). For the sham group, the same equipment and procedure were applied, except that the cuff was inflated to only 30 mmHg, which produced mild limb constriction without inducing ischemic stimulation. Anesthesia Patients were randomly allocated to groups using computer-generated numbers on preoperative day 3. Medical staff knowledgeable about the group allocation administered RIPC or sham-RIPC treatments twice daily for 3 consecutive days. Intravenous access was established upon arrival in the operating room, after which an initial bolus of 8 ml·kg⁻¹ of lactated Ringer`s solution was administered, followed by an intraoperative maintenance infusion at 4 ml·kg⁻¹·h⁻¹. Anesthesia was induced with intravenous sufentanil (1–2 µg·kg⁻¹), etomidate (1.5-2.5 mg·kg⁻¹), and rocuronium (0.6-1 mg·kg⁻¹). For mechanical ventilation, 6–7 ml·kg⁻¹ tidal volume, 12 breaths/min respiratory rate of, 6 cmH2O PEEP, 1:2 I/E ratio, 60% FiO2, and 35 ~ 37 mmHg PETCO2 were set. Anesthesia was maintained with 0.2–0.3 MAC sevoflurane, propofol (1-1.5 mg·kg⁻¹·h⁻¹), sufentanil (0.5–1 µg·kg⁻¹·h⁻¹), dexmedetomidine (0.5–1 µg·kg⁻¹·h⁻¹), and rocuronium (0.5 mg·kg⁻¹·h⁻¹). The bispectral index was maintained between 40 and 60, and the intraoperative body temperature at approximately 36.0 °C. Invasive arterial and central venous pressures were continuously monitored, and vasoactive agents were titrated accordingly to maintain hemodynamic stability. All patients received standardized anesthetic management according to a unified protocol. Postoperatively, the patients were transferred to the intensive care unit (ICU) under continuous haemodynamics monitoring, and extubation was performed once standard criteria were met. The same team of experienced cardiac surgeons performed all the OPCAB procedures in accordance with a standardized surgical protocol. Outcomes Primary outcome The incidence of POGI was defined as an I-FEED score between 3 and 5 within 7 days postoperatively. Postoperative gastrointestinal function was assessed daily by trained nurses using the I-FEED scoring system [21]. Secondary outcome Blood sample collection and processing Fasting peripheral venous blood was collected from the patients in the morning before RIPC or sham-RIPC and again on postoperative day 3. The serum levels of CRP, I-FABP, IL-6, and S-100β were measured before RIPC/sham-RIPC and on postoperative day 3. To separate the serum, blood samples were centrifuged at 3,000 rpm for 15 min within 2 h of collection. The serum was aliquoted and stored at − 80 °C in an ultra-low temperature freezer, avoiding repeated freeze-thaw cycles until all samples were processed in batches. Details of CRP, I-FABP, IL-6, and S-100β detection are presented in Supplement 4. Exploratory clinical outcomes Intestinal microbiota was characterized via 16 S rRNA analysis of fecal samples. Fecal samples were collected from each patient at two time points: before the start of RIPC or sham-RIPC (T0) and at the time of the first postoperative bowel movement, which occurred between postoperative days 3 and 5 (T1). After DNA extraction and 16 S rRNA sequencing, the relative abundance of bacterial genera was compared between the two groups. To explore baseline differences, we compared the genus-level composition between the RIPC and sham-RIPC groups using the pre-intervention samples. To assess the effect of the intervention, we compared the genera between the two groups using the postoperative samples. All microbiome comparisons were performed using the two-group comparison module of the Microbiome Consortium (Wekemo Bioincloud), and the top 15 features with the smallest P values were displayed. In addition, Spearman’s correlation analysis was conducted between the relative abundance of bacterial genera and clinical parameters (e.g., I-FABP, IL-6, CRP, S-100β), and the results were visualized as a heatmap showing the top 30 and top 10 correlated genera, respectively. Detailed procedures for sample collection, CTAB-based DNA extraction, PCR amplification, and sequencing are provided in Supplement 5. The incidence of POGD was defined as I-FEED score of 6 or more within 7 days postoperatively. The time to recovery of gastrointestinal function (first flatus, first stool, and first successful oral feeding) was recorded. The incidence of postoperative delirium (POD) within 7 days postoperatively was also assessed. The incidence of the composite endpoint (POGI + POD) was defined as the presence of both POGI (I-FEED score 3–5) and POD (positive 3D-CAM or CAM-ICU score) within 7 days postoperatively. POD was diagnosed using culturally adapted and validated Chinese versions of established tools. For non-ventilated patients, we employed the Chinese version of the 3-minute Diagnostic Interview for CAM (3D-CAM). For mechanically ventilated patients in the ICU, we used the Chinese version of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) [22]. Furthermore, the incidence of severe pain and severe sleep disorders were assessed, and intraoperative and postoperative outcomes such as the duration of endotracheal intubation, ICU stay, postoperative hospitalization, and total hospitalization were also recorded. We used a 10-point scale to assess the severity of pain and sleep quality during the first 7 days postoperatively, with scores of 7–10 defined as severe pain or severe sleep disturbance [23] (eTable-1, Supplement 1). Safety outcome Safety was monitored throughout the study. Device-related safety: The principal investigator immediately discontinued the intervention and performed emergency unblinding upon the appearance of objective signs of severe intolerance or local injury in the subject (e.g., persistent loss of pulse after cuff deflation, significant edema, erythema, or skin compromise) occurring during or after RIPC/sham-RIPC [24]. Systematic assessment of postoperative systemic complications from surgery to discharge included postoperative new-onset ventricular arrhythmias, postoperative atrial fibrillation, postoperative myocardial infarction, postoperative heart failure, postoperative cerebral infarction, postoperative pneumonia, and postoperative acute kidney injury/acute renal failure. All-cause mortality was recorded at 7 days and 28 days via in-hospital follow-ups or telephone follow-up. Statistical analysis Data were analyzed using R software (version 4.4.3). Continuous variables were summarized as mean (SD) or median (interquartile range [IQR]), as appropriate, unless otherwise specified. Continuous data group comparisons were performed using the independent samples t-test or Mann-Whitney U test, as appropriate. Categorical data were expressed as numbers (percentages) and compared using the chi-squared test, with continuity correction or Fisher’s exact test applied where appropriate. A two-sided P value < 0.05 was considered statistically significant. For all secondary outcomes (IL-6, CRP, I-FABP, and S-100β), within-group changes from baseline to postoperative day 3 were evaluated using paired t-tests. For between-group comparisons at baseline, independent samples t-tests were employed, whereas postoperative comparisons were performed via analysis of covariance (ANCOVA) with adjustment for baseline values. Adjusted mean differences with 95% confidence intervals (CIs) were calculated from the regression coefficients, and F statistics were used to test for group effects. To correct for multiple comparisons across the four secondary biomarker outcomes, we controlled the false discovery rate (FDR) at 0.05 using the Benjamini-Hochberg method. Tests with P < 0.05 were considered significant. To evaluate microbial community differences between the RIPC and sham-RIPC groups at T1, a two-group comparison was performed using the two-group comparison module in Wekemo Bioincloud [25]. The analysis included mean abundance estimation, differential testing (Welch’s t-test), and 95% CI calculation with multiple testing corrections applied using the Benjamini-Hochberg method. Box plots were used for visualization to demonstrate statistical significance between groups. Furthermore, the correlation heatmaps were generated using the correlation heatmap module in Wekemo Bioincloud, mainly employing the pheatmap package of the R project (Raivo Kolde, Tartu, Republic of Estonia). This analysis determined whether clinical factors were significantly correlated with microbial communities by calculating Spearman’s correlation coefficients between clinical factors and microbial species, which were then visualized using heatmaps. Quality control To ensure excellent study quality, the following measures were implemented: Minimizing Selection Bias: Patient enrollment was carried out with strict adherence to the predefined inclusion and exclusion criteria. A rigorous randomization procedure was performed. Standardization of Procedures: All surgical procedures were performed by the same lead surgeon and management team. Anesthesia was administered by a dedicated, consistent team of anesthesiologists. Perioperative management, sample collection, and postoperative follow-up were strictly conducted in accordance with the study protocol. Blinded Outcome Assessment: All clinical scoring and follow-up evaluations were conducted by designated personnel who were not involved in the perioperative care of the patients and were blinded to the group allocation. Data Integrity and Completeness: The trial process and key data were regularly audited. Medical records were meticulously reviewed to ensure completeness, and every effort was made to reduce follow-up loss. Results Patient population and baseline demographic characteristics This study was conducted from June 2024 to June 2025. A CONSORT flow diagram is presented in Fig. 1. A total of 190 patients were assessed for eligibility. Of these, 84 were excluded before randomization: 44 did not meet the inclusion criteria and 40 declined to participate. The remaining 106 patients were randomly assigned in a 1:1 ratio to the RIPC group (n = 53) or the sham-RIPC group (n = 53). In the RIPC group, 2 patients had surgery temporarily canceled, 1 withdrew during the RIPC procedure, and 1 required reoperation for postoperative bleeding; thus, 49 patients received the allocated intervention and completed surgery. In the sham-RIPC group, 2 patients requested discharge before surgery and 3 withdrew during the sham procedure; thus, 48 patients received the allocated intervention and completed surgery. During follow-up, 1 patient in the RIPC group and 2 patients in the sham-RIPC group refused the daily I-FEED assessment and were lost to follow-up. Therefore, the modified intention-to-treat (mITT) population (defined as all randomized patients who received the study intervention, completed surgery, and underwent postoperative I-FEED evaluation) included 48 patients in the RIPC group and 46 patients in the sham-RIPC group (total n = 94). No significant differences were observed between the groups in terms of baseline characteristics (Table 1). Moreover, there were no significant differences in anesthesia duration, surgical duration, or total graft anastomosis time. The doses of anesthetic agents (sufentanil, propofol, dexmedetomidine, norepinephrine, phenylephrine, dopamine, nitroglycerin, weight-adjusted dose for all vasoactive drugs, and Vasoactive-Inotropic Score (VIS)) were also comparable between the groups. Similarly, surgery-related baseline indicators, including the number of bypass grafts, intraoperative BE value < − 4, blood loss, colloid and crystalloid infusion volumes, and urine output, exhibited no significant differences (Table 2). CONSORT flow diagram Table 1Demographic and baseline dataCharacteristicTotal( n = 94 )Sham-RIPC group( n = 46 )RIPC group( n = 48 )X2 / t valueP valueAge (SD), y66.0 (7.8)67.5 (6.6)64.5 (8.6)1.8720.065Height (SD), cm166.3 (7.7)166.2 (8.1)166.4 (7.4)0.1650.869Weight (SD), kg71.6 (3.3)71.0 (8.8)72.2 (9.6)0.6360.527BMI (SD)25.87 (3.32)25.78 (3.09)25.96 (3.56)0.2740.785Sex, n (%)1.0750.300 Female28 (29.8)16 (34.8)12 (25.0)Smoking history, n (%)48 (51.06)19 (41.3)29 (60.4)3.4340.064ASA score, n (%)0.270.787 Ⅱ37 (39.4)19 (41.3)18 (37.5) Ⅲ45 (47.9)21 (45.7)24 (50.0) Ⅳ12(12.8)6 (13.0)6 (12.5)Hypertension, n (%)71 (75.5)35 (76.1)36 (75.0)0.0150.903Diabetes mellitus, n (%)47(50.0)20 (43.5)27 (56.2)1.5330.216Coronary heart disease, n (%)§66 (70.2)29 (63.0)37 (77.1)2.2140.137Cerebral infarction, n (%)22 (23.4)11 (23.9)11 (22.9)0.0130.909Peripheral arterial disease (PAD), n (%)53 (56.4)26 (56.5)27 (56.2)0.0010.979Preoperative diagnosis, n (%)-0.118UA57 (60.6)27 (58.7)30 (62.5)CAD12 (12.8)8 (17.4)4 (8.3)NSTEMI22 (23.4)8 (17.4)14 (29.2)STEMI3 (3.2)3 (6.5)0 (0.0)Graft type, n (%)1.0710.585LITA only (A)3 (3.2)2 (4.3)1 (2.1)SVG only (V)18 (19.1)9 (19.6)9 (18.8)LITA + SVG (A + V)73 (77.7)35 (76.1)38 (79.2)The data are expressed as n (%) or mean (SD). Continuous variables (age, height, weight, BMI) were compared using the independent samples t-test. Categorical variables (sex, ASA score, hypertension, diabetes mellitus, coronary heart disease, cerebral infarction, smoking history) were analyzed using the chi-squared test. Preoperative diagnosis and graft type were compared using Fisher’s exact test§ Coronary heart disease refers to documented history of coronary heart disease prior to admission, not the current surgical indication for CABG (which was 100%)UA, unstable angina; CAD, coronary artery disease; NSTEMI, non-ST-segment elevation myocardial infarction; STEMI, ST-segment elevation myocardial infarctionLITA only (A), left internal thoracic artery only; SVG only (V), saphenous vein graft only; LITA + SVG (A + V), both arterial and venous conduits. No patient received gastroepiploic artery (GEA), bilateral internal thoracic artery (BITA), radial artery, or other arterial conduits Demographic and baseline data The data are expressed as n (%) or mean (SD). Continuous variables (age, height, weight, BMI) were compared using the independent samples t-test. Categorical variables (sex, ASA score, hypertension, diabetes mellitus, coronary heart disease, cerebral infarction, smoking history) were analyzed using the chi-squared test. Preoperative diagnosis and graft type were compared using Fisher’s exact test § Coronary heart disease refers to documented history of coronary heart disease prior to admission, not the current surgical indication for CABG (which was 100%) UA, unstable angina; CAD, coronary artery disease; NSTEMI, non-ST-segment elevation myocardial infarction; STEMI, ST-segment elevation myocardial infarction LITA only (A), left internal thoracic artery only; SVG only (V), saphenous vein graft only; LITA + SVG (A + V), both arterial and venous conduits. No patient received gastroepiploic artery (GEA), bilateral internal thoracic artery (BITA), radial artery, or other arterial conduits Table 2Intraoperative dataCharacteristicSham-RIPC group( n = 46 )RIPC group( n = 48 )X2 /Z / t valueP valueAnesthesia time, (IQR), min290 (259–325)281 (259–314)0.1820.859Operation time, (IQR), min220 (207–260)222 (201–246)0.3440.734Total graft anastomosis time, (SD), min†80.7 (27.0)81.9 (24.6)0.2130.831Number of Bypass, n(%)0.9590.327 ≤ 28 (17.4)5 (10.4) ≥ 338 (82.6)43 (89.6)Bleeding, median (IQR), mL400 (300–600)500 (300–600)1.0060.316Colloid, median (IQR), mL500 (500–1000)1000 (500–1000)0.6720.504Crystal, median (IQR), mL2700 (2100–3200)2500 (2200–3030)0.6020.549urine volume, median (IQR), mL1150 (1000–1600)1200 (950–1800)0.6330.529Intraoperative BE<-4, n (%)6 (13.0)9 (18.7)0.5700.450Intraoperative Medication UseSufentanil, (IQR), µg480 (420–538)460 (400–545)0.0870.934Propofol, (IQR), mg582 (418–728)580 (460–742)0.5670.573Dexmedetomidine, (IQR), µg1120 (920–1320)1200 (920–1410)0.4430.661Norepinephrine, (IQR), µg580 (343–920)480 (354–690)1.2520.212Phenylephrine, (IQR), µg200 (62–350)200 (50–400)0.5020.619Dopamine, (IQR), mg26 (20–36)23 (17–34)0.9720.333Nitroglycerin, (IQR), mg2 (1–3)1 (0–3)1.3670.173Norepinephrine, (IQR), µg/kg8.8 (5.1–13.4)6.9(4.7–10.2)1.4940.135Phenylephrine, (IQR), µg/kg3.1 (0.9–5.1)2.6 (0.7–5.2)0.4160.677Dopamine, (IQR), µg/kg352.7 (259.3-534.9)320.3 (232.5–505.0)1.1160.265Nitroglycerin, (IQR), µg/kg27.3 (15.7–39.2)20.7 (4.1–34.5)1.4910.136VIS, (IQR)8.5 (5.7–10.6)7.0 (4.7–9.4)1.5320.126The data are expressed as mean (SD), median (IQR), or n (%)† Total time required to complete all distal anastomoses (sum of individual graft anastomosis times)The number of bypass grafts and intraoperative metabolic acidosis were analyzed using chi-squared tests. Anesthesia time, operation time, intraoperative drug doses (sufentanil, propofol, dexmedetomidine, norepinephrine, phenylephrine, dopamine, nitroglycerin), bleeding, colloid and crystalloid infusion volumes, and urine output were analyzed using nonparametric Mann-Whitney U testsVIS (Vasoactive-Inotropic Score) = dopamine (µg/kg/min) + 100 × norepinephrine (µg/kg/min) + 100 × phenylephrine (µg/kg/min) Intraoperative data The data are expressed as mean (SD), median (IQR), or n (%) † Total time required to complete all distal anastomoses (sum of individual graft anastomosis times) The number of bypass grafts and intraoperative metabolic acidosis were analyzed using chi-squared tests. Anesthesia time, operation time, intraoperative drug doses (sufentanil, propofol, dexmedetomidine, norepinephrine, phenylephrine, dopamine, nitroglycerin), bleeding, colloid and crystalloid infusion volumes, and urine output were analyzed using nonparametric Mann-Whitney U tests VIS (Vasoactive-Inotropic Score) = dopamine (µg/kg/min) + 100 × norepinephrine (µg/kg/min) + 100 × phenylephrine (µg/kg/min) Primary outcome The incidence of POGI within 7 days postoperatively was lower in the RIPC than in the sham group (35.4% vs. 56.5%, χ² = 4.216, P = 0.040). Secondary outcome Analysis of inflammatory markers and I-FABP levels in the two groups revealed no significant differences between them before RIPC/sham-RIPC. However, on postoperative day 3, the RIPC group exhibited lower levels of the following markers than the sham group: I-FABP (29.07 [0.76] vs. 23.34 [0.75], F = 28.70, P < 0.001), CRP (99.05 [3.72] vs. 85.42 [3.64], F = 5.93, P = 0.02), IL-6 (151.69 [24.07] vs. 75.54 [23.81], F = 6.29, P = 0.01), and S-100β (127.51 [16.60] vs. 66.12 [16.25], F = 6.86, P = 0.01). To correct for multiple comparisons, the Benjamini-Hochberg method was applied to control the false discovery rate at 0.05 across the four outcomes. All four comparisons remained lower in the RIPC group than in the sham group on postoperative day 3 (Table 3). Table 3The secondary outcomesCharacteristicTime pointSham-RIPC group (n = 46)RIPC group (n = 48)Difference #95% CIStatistic FP valueBH-adjusted P valueI-FABP, ng/mlPre-operation12.34 (2.53)12.49 (2.23)-0.15-1.11 ~ 0.810.090.7600.869Postoperative day 329.07 (0.76)23.34 (0.75)5.733.64 ~ 7.8328.70< 0.001*< 0.001*Within-group P< 0.001*< 0.001*---- -CRP, mg/lPre-operation11.84 (3.31)10.74 (2.64)1.10-0.11 ~ 2.303.160.0800.106Postoperative day 399.05 (3.72)85.42 (3.64)13.633.34 ~ 23.925.930.020*0.040*Within-group P< 0.001*< 0.001*---- -IL-6, pg/mlPre-operation7.57 (12.66)3.94 (4.17)3.63-0.18 ~ 7.453.480.0710.106Postoperative day 3151.69 (24.07)75.54 (23.81)76.159.18 ~ 143.126.290.010*0.027*Within-group P< 0.001*< 0.001*---- -S-100β, pg/mlPre-operation49.84 (34.77)51.14 (56.31)-1.30-20.31, 17.720.020.890.890Postoperative day 3127.51 (16.60)66.12 (16.25)61.3915.87, 106.926.860.010*0.027*Within-group P< 0.001*0.0032*---- -Data are presented as follows: pre-operation values are expressed as mean (SD); postoperative day 3 values are expressed as baseline-adjusted least-squares mean (SE) estimated from ANCOVA models adjusted for baseline values. Linear regression was used for between-group comparisons at baseline. ANCOVA with adjustment for baseline values was used for comparisons on postoperative day 3; adjusted mean differences (sham-RIPC minus RIPC) and 95% CIs are shown. Within-group changes were evaluated using paired t-tests. To correct for multiple comparisons across the four secondary outcomes, the Benjamini-Hochberg method was applied to control the false discovery rate at 0.05; the last column shows the adjusted P values. # Mean difference (sham-RIPC minus RIPC). * P < 0.05 indicates statistical significance The secondary outcomes Data are presented as follows: pre-operation values are expressed as mean (SD); postoperative day 3 values are expressed as baseline-adjusted least-squares mean (SE) estimated from ANCOVA models adjusted for baseline values. Linear regression was used for between-group comparisons at baseline. ANCOVA with adjustment for baseline values was used for comparisons on postoperative day 3; adjusted mean differences (sham-RIPC minus RIPC) and 95% CIs are shown. Within-group changes were evaluated using paired t-tests. To correct for multiple comparisons across the four secondary outcomes, the Benjamini-Hochberg method was applied to control the false discovery rate at 0.05; the last column shows the adjusted P values. # Mean difference (sham-RIPC minus RIPC). * P < 0.05 indicates statistical significance Exploratory clinical outcomes In the analysis of the gut microbiota at the 16 S rRNA genus level, no significant difference was observed in gut microbiota composition between the groups before RIPC/sham-RIPC (eFigure-1, Supplement 3). However, compared with the sham-RIPC group, the relative abundance of Gemella (P = 0.001) and Dorea (P = 0.043) at the genus level in the postoperative first stool microbiome analysis was reduced in the RIPC group (Fig. 2). Postoperative I-FABP exhibited a strong positive correlation with Dorea (P < 0.001) and Gemella (P = 0.009). In addition, Gemella showed a positive correlation with IL-6 (P = 0.016) (Fig. 3). Composition of the gut microbiota at the genus level between the groups. The relative abundance of Gemella (P = 0.001) and Dorea (P = 0.043) was lower in the RIPC than in the sham-RIPC group in the postoperative period Structure of the correlation heatmap. A. The heatmap shows the correlations of bacterial genera (right) with systemic inflammatory, intestinal injury markers, and S-100β as well as the highest I-FEED scores (I-FEEDmax) . It also shows the top 30 bacterial genera exhibiting correlations. Color Coding: Red indicates a positive correlation, whereas blue denotes a negative correlation. Deeper colors represent stronger correlations. Significance Markers: *(P < 0.05), **(P < 0.01), and ***(P < 0.001). Bacterial Genera: Gemella showed a positive correlation with IL-6 (red *) and significant positive correlation with I-FABP (red **). Dorea exhibited a strong positive correlation with I-FABP (red, ***). B. The heatmap displays the top 10 bacterial genera exhibiting correlations, demonstrating a strong positive correlation between I-FABP and Dorea (P < 0.001) and a significant positive correlation between I-FABP and Gemella (P = 0.009). Gemella also showed a positive correlation with IL-6 (P = 0.016) The rates of POGI + POD and POD within 7 days postoperatively were lower in the RIPC than in the sham-RIPC group (12.5% vs. 30.4% [χ² = 4.511, P = 0.034] and 27.1% vs. 47.8% [χ² = 4.325, P = 0.038], respectively) (Table 4). No significant difference was observed in the time to first flatus, time to first stool, and time to first oral feeding (Table 4). Furthermore, the groups did not show differences in mechanical ventilation time, incidence of mechanical ventilation time > 48 h, reintubation rate, postoperative ICU stay, postoperative hospital stay, hospitalization time, postoperative drainage volume, postoperative fever rate, postoperative laxative usage rate, postoperative epinephrine usage rate, severe postoperative sleep disorders, severe postoperative pain, or postoperative IABP usage rate; the two groups showed no difference (eTable-1, Supplement 1). Table 4The exploratory clinical outcomesOutcomeSham-RIPC group(n = 46 )RIPC group(n = 48 )χ2 /Z ValueP valuePOGI + POD, n(%)14 (30.4)6 (12.5)4.5110.034*POD, n(%)22 (47.8)13 (27.1)4.3250.038*POGD, n(%)1 (2.2)1 (2.1)—1.000Time to first flatus (IQR), h49.5 (42.2–64.8)50.5 (40.5–65.0)0.3940.697Time to first stool (IQR), h91.0 (68.5-108.8)94.0 (72.0-115.0)0.5070.615Time to first oral feeding (IQR), h18.0 (16.2–38.8)21.0 (17.0-29.2)0.2840.779The data are expressed as median (IQR) or absolute numbers (percentage of the total [%]). POGD was assessed using Fisher’s exact test. * P values < 0.05 indicate significant differences The exploratory clinical outcomes The data are expressed as median (IQR) or absolute numbers (percentage of the total [%]). POGD was assessed using Fisher’s exact test. * P values < 0.05 indicate significant differences Safety outcome No procedure-related local or neurovascular injuries requiring unblinding or discontinuation occurred in either the RIPC or sham-RIPC group. There were no statistically significant differences in postoperative complication rates from surgery until discharge, or in mortality rates at 7 and 28 days after surgery (eTable-2, Supplement 1). Discussion Our study indicated that the preoperative systematic application of RIPC can reduce the risk of postoperative POGI in patients with OPCAB. One possible explanation is that RIPC may induce a delayed protective (“second window”) effect [14]. Our protocol of twice-daily RIPC for three consecutive preoperative days might have engaged this delayed phase, thereby providing sustained mitigation of intestinal ischemia-reperfusion injury during and after OPCAB. The I-FEED (Intake, Feeling nauseated, Emesis, physical Exam, Duration of symptoms) scoring system classifies postoperative gastrointestinal function as normal, postoperative gastrointestinal intolerance (POGI), or postoperative gastrointestinal dysfunction (POGD) [12]. This tool has been validated in various surgeries, including gynecological, colorectal, and lumbar spine procedures [20, 21, 26], and facilitates early recognition of gastrointestinal dysfunction [27]. POGI typically presents as nausea, mild non-bilious vomiting, and abdominal distension after postoperative day 2, and may progress to POGD, which has been associated with worse clinical outcomes [12]. In addition, postoperative levels of I-FABP, a widely used biomarker of intestinal mucosal injury, were lower in the RIPC group. I-FABP is rapidly released into the bloodstream during intestinal ischemia and has been associated with poor prognosis in cardiac surgery patients [4, 28, 29]. Previous studies have shown that RIPC reduces I-FABP and inflammatory cytokines in patients undergoing abdominal aortic aneurysm repair and laparoscopic colorectal resection, suggesting a potential enhancement of intestinal mucosal tolerance to ischemia-reperfusion injury [20, 30]. These findings are consistent with our observation that RIPC may reduce POGI, possibly by attenuating intestinal mucosal injury. Our study also detected that RIPC was associated with lower postoperative serum levels of IL-6 and CRP. Elevated IL-6 has been reported to correlate with delayed colonic motility and may play a role in the pathogenesis of multiorgan failure [4, 31]. Similar reductions in IL-6 and CRP after RIPC have been reported across various settings, including mitral valve replacement [32], renal injury [33], acute lung injury [34], and marathon running [35]. These observations are consistent with the possibility that RIPC may suppress systemic inflammation, which could in turn help protect the intestinal barrier and potentially reduce the risk of POGI. In our study, serum S-100β levels at postoperative day 3 were lower in the RIPC group. S-100β has been associated with glial cell activation. Reactive enteric glial cells have been reported to release excessive S-100β, which may contribute to enteric neuronal dysfunction and has been linked to gastrointestinal symptoms [36–39]. Clinical studies have reported that RIPC can reduce S-100β levels after cardiac surgery [40]. The lower S-100β levels in the RIPC group may reflect reduced glial-related neuroinflammation, although the exact source (gut, brain, or both) remains unclear. Given the lack of specificity of S-100β, this finding should be interpreted cautiously. Nevertheless, it suggests a possible link between RIPC and glial modulation, which might contribute to the reduced POGI observed in our study. Systemic inflammation after OPCAB may disrupt the blood-brain barrier, activate microglia, and induce neuroinflammation, all of which could potentially contribute to delirium. As an exploratory outcome, we compared the incidence of postoperative delirium (POD) between the two groups and found that RIPC was associated with a lower incidence of POD. This finding is consistent with several randomized controlled trials in cardiac surgery, severe traumatic brain injury, and acute ischemic stroke, which reported similar effects [40–42]. One plausible hypothesis is that RIPC might stabilize the intestinal mucosal barrier, thereby reducing systemic translocation of gut-derived factors that could otherwise reach the brain and promote POD. The concurrent reduction in POGI and POD rates observed in our study is consistent with this hypothesis. However, further in-depth research is still warranted. In exploratory analyses, the relative abundance of Gemella and Dorea was lower in the RIPC group postoperatively, while other abundant commensals such as Faecalibacterium and Blautia showed no significant differences. This pattern suggests that RIPC may exert targeted rather than broad-spectrum modulation of the gut microbiome. Both Dorea and Gemella have been suggested to be associated with intestinal inflammation [43, 44]. In the present study, postoperative I-FABP levels showed positive correlations with Dorea and Gemella, and Gemella also correlated positively with IL-6. These correlational observations raise the possibility that these two genera may be involved in postoperative intestinal barrier disruption and systemic inflammation, and that their reduced abundance in the RIPC group might reflect a potential protective effect on the intestinal barrier. Whether these gut microbiota alterations directly mediate the clinical effects of RIPC requires further investigation using mediation analysis, metagenomics, and germ-free animal models. Nevertheless, these exploratory findings are consistent with the possibility that RIPC-associated changes in gut microbiota might contribute to the reduced POGI observed in our study. This study has several limitations. First, due to the single-center design and the relatively small sample size, several of the exploratory outcomes may have insufficient statistical power. Second, because of the different cuff pressures (200 mmHg vs. 30 mmHg), patients in the RIPC group likely became aware of their allocation, which could introduce performance bias and potentially affect the subjective components of the I-FEED score; additionally, the variable timing of stool collection may have introduced bias in the microbiota analysis. Third, follow-up was relatively short (7 days for most outcomes, 28 days for mortality), precluding assessment of long-term benefits. Fourth, standardizing perioperative care in cardiac surgery is challenging due to variations in comorbidities, genetic backgrounds, and concomitant medications. Despite these limitations, the observed beneficial effects of RIPC on inflammatory markers, intestinal injury biomarkers, and exploratory microbiota measures, together with the absence of major safety concerns, suggest that RIPC may be a promising strategy for reducing postoperative gastrointestinal intolerance in OPCAB patients. Accordingly, our team will next undertake a large-scale, multicenter trial to confirm these findings. Conclusion This study indicates that preoperative systematic application of RIPC is associated with a reduced incidence of POGI in patients undergoing OPCAB. The RIPC group exhibited lower levels of the intestinal mucosal injury marker I-FABP, reduced inflammatory cytokines IL-6 and CRP, and lower S-100β levels. Exploratory analyses also indicated that RIPC decreased the postoperative relative abundances of Gemella and Dorea. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements The authors thank Di Zhang for performing the RIPC and the sham-RIPC, Te Lv and Xinxin Chen for assisting in blood and fecal specimen collection and preparation, Xiaona Zhang for assisting in the ELISA test, and Jing Li for assisting in the microbiome test. Abbreviations American Society of Anesthesiologists Body mass index Bispectral index Coronary artery bypass grafting Cardiopulmonary bypass Creactive protein Intestinal fatty acidbinding protein Intake, Feeling nauseated, Emesis, physical Exam, Duration of symptoms Interleukin6 Intensive care unit Left internal thoracic artery New York Heart Association Offpump coronary artery bypass grafting Peripheral arterial disease Postoperative delirium Postoperative gastrointestinal dysfunction Postoperative gastrointestinal intolerance Remote ischemic preconditioning S100 protein beta subunit Standard deviation Saphenous vein graft Vasoactiveinotropic score Authors’ contributions WY; Contribution: performed the operation and wrote and drafted the manuscript.QL; Contribution: contributed to trial conduction, helped evaluate the patients’ responses and funding support.XW; Contribution: contributed to trial conduction and performed data acquisition.WH; Contribution: participated in the data analysis and interpretation and helped revise the manuscript.HL; Contribution: contributed to the design of the research, performed the surgery as the lead surgeon.TW; Contribution: contributed to the design of the research and funding support and agreed to be accountable for all aspects of this work. All authors read and approved the final manuscript version. Funding Post-subsidy funds for National Clinical Research Center, Ministry of Science and Technology of China. Project ID: (Tianlong Wang: 303-01-001-0272-03); Beijing Municipal Hospitals’ Scientific Research Cultivation Program. Project ID: (Qinghai Liu: P X 2023031) Data availability The raw data of this study are available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The local ethics committee of Xuanwu Hospital, Capital Medical University approved the study protocol (Approval No: Lin Yan Shen [2024] 110-001). The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. This study was registered with the Chinese Clinical Trial Registry (ChiCTR2400085044) (http://www.chictr.org.cn/). Written informed consent was obtained from all individual participants included in the study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Hongli Li, Email: lihongli3021@sohu.com. Tianlong Wang, Email: w_tl5595@hotmail.com. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement The raw data of this study are available from the corresponding author on reasonable request.