Dexmedetomidine enhances recovery after gastrointestinal cancer surgery by protecting the endothelial glycocalyx: A randomized, double-blind, placebo-controlled study

Abstract

BACKGROUND

The vascular endothelial glycocalyx (VEG) plays a critical role in maintaining vascular barrier integrity, regulating inflammation, and ensuring microcirculatory homeostasis. Surgical stress and systemic inflammation can disrupt the glycocalyx, leading to endothelial dysfunction, impaired microcirculation, and adverse postoperative outcomes. dexmedetomidine (DEX), an α2-adrenergic agonist with anti-inflammatory and organ-protective properties, has been suggested in preclinical and clinical studies to mitigate glycocalyx degradation, yet evidence in gastrointestinal cancer surgery remains limited.

AIM

To determine whether perioperative DEX attenuates surgical inflammation-induced VEG degradation and preserves endothelial barrier function in patients undergoing gastrointestinal cancer resection.

METHODS

This was a prospective, single-center, randomized, double-blind, placebo-controlled trial conducted at the First Affiliated Hospital of University of Science and Technology of China. A total of 110 patients undergoing elective gastric or colorectal tumor resection were randomly assigned (1:1) to receive intraoperative DEX or saline placebo. Anesthesia and analgesia were standardized across groups. The primary outcome was plasma syndecan-1 concentration, a marker of endothelial glycocalyx injury, measured at four perioperative timepoints (T0-T3). Secondary outcomes included inflammatory biomarkers [interleukin-6 (IL-6), tumor necrosis factor-alpha, C-reactive protein, heparan sulfate], microcirculatory parameters [perfused vessel density (PVD), flow index, P(v-a)CO₂, lactate], and clinical endpoints [extubation time, opioid use, Visual Analog Scale (VAS) scores, Quality of Recovery-15 Questionnaire (QoR-15), length of stay, and 30-day complications]. Postoperative complications were defined by Clavien-Dindo criteria and adjudicated by blinded investigators. The trial was registered prospectively (ChiCTR2500109633) and powered to detect a clinically meaningful difference in syndecan-1 levels.

RESULTS

A total of 110 patients were randomized equally to the DEX or control group, with well-balanced baseline characteristics. Compared with controls, DEX significantly reduced postoperative infections (7% vs 16%) and intensive care unit admissions (7% vs 13%), shortened extubation time (13.1 ± 3.0 minutes vs 18.4 ± 4.0 minutes; P < 0.001), and decreased opioid use (23.1 ± 5.0 mg vs 27.3 ± 6.0 mg; P = 0.004) and VAS pain scores (P = 0.002). At abdominal closure, DEX attenuated endothelial glycocalyx injury, as evidenced by lower plasma syndecan-1 (44.72 ± 7.10 ng/mL vs 48.73 ± 6.26 ng/mL; P = 0.002) and heparan sulfate levels (P = 0.001). IL-6 was significantly reduced at 24 hours (110.77 ± 29.72 pg/mL vs 138.86 ± 35.95 pg/mL; P < 0.0001) and positively correlated with syndecan-1 (r = 0.71). Microcirculatory function improved with DEX, including higher PVD (21.40 ± 3.50 mm/mm² vs 19.94 ± 2.93 mm/mm²; P = 0.019), increased flow index, lower P(v-a)CO₂ (P < 0.001), and reduced lactate (P = 0.003). DEX also improved recovery outcomes, with higher QoR-15 scores (P = 0.001), shorter hospital stays (6.49 ± 1.29 days vs 7.29 ± 1.59 days; P = 0.005), and fewer overall 30-day complications (12.7% vs 30.9%; P = 0.036). Receiver operating characteristic analysis identified syndecan-1 > 45 ng/mL at abdominal closure as a potential predictor of postoperative complications (area under the curve = 0.68, 95%CI: 0.59-0.76), and multivariable analysis showed a near-significant association (OR = 2.88, P = 0.057). Subgroup analyses demonstrated consistent anti-inflammatory and endothelial-protective effects of DEX across age and surgical approach strata.

CONCLUSION

Perioperative administration of DEX confers significant endothelial-protective effects by mitigating glycocalyx degradation, suppressing systemic inflammation, and promoting enhanced postoperative recovery. These findings support its clinical utility as a valuable adjunctive therapy in the perioperative management of patients undergoing oncologic gastrointestinal surgery.

Key Words: Dexmedetomidine; Vascular endothelial glycocalyx; Syndecan-1; Surgical inflammation; Gastrointestinal neoplasms

Core Tip: This randomized controlled trial demonstrates that perioperative dexmedetomidine (DEX) infusion attenuates vascular endothelial glycocalyx degradation (as evidenced by reduced syndecan-1 and heparan sulfate levels) and suppresses systemic inflammation in patients undergoing gastrointestinal tumor resection. These mechanistic benefits were associated with improved microcirculatory perfusion, reduced postoperative complications, and enhanced recovery, including shorter hospital stay. The study identifies syndecan-1 as a potential biomarker for perioperative risk stratification and supports the integration of DEX as an endothelial-protective adjunct within Enhanced Recovery after Surgery protocols for oncologic surgery.

INTRODUCTION

The vascular endothelial glycocalyx (VEG) is a fragile, carbohydrate-rich layer composed of membrane-bound proteoglycans and glycosaminoglycans that lines the vascular endothelium. Functioning as the primary interface between the bloodstream and the vessel wall, this structure is instrumental in regulating vascular permeability, Mechan transduction, and inflammatory cell recruitment, thereby maintaining microcirculatory homeostasis[1,2]. However, its delicate architecture renders it highly vulnerable to degradation in response to surgical trauma, especially during major abdominal procedures.

Gastrointestinal oncologic resections, albeit curative, trigger a substantial systemic inflammatory response characterized by a sharp increase in pro-inflammatory cytokines, notably interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNF-α). These cytokines promote the release of matrix metalloproteinases (e.g., MMP-9), which enzymatically cleave core glycocalyx components such as syndecan-1 and heparan sulfate, ultimately compromising endothelial barrier function[3,4].

In cancer patients, this pathophysiological process is further amplified by tumor-derived exosomes carrying bioactive molecules like IL-6 and transforming growth factor-β. These nanovesicles facilitate paracrine signaling that exacerbates endothelial activation and glycocalyx degradation, creating a vicious cycle of vascular inflammation and injury[5-7]. The resulting endothelial dysfunction clinically manifests as increased vascular permeability, tissue edema, and a higher risk of postoperative complications, including pulmonary dysfunction, gastrointestinal paralysis, and renal impairment[8].

Emerging preclinical evidence indicates that certain anesthetic adjuvants, particularly the α₂-adrenergic agonist dexmedetomidine (DEX), may confer endothelial protection through multimodal mechanisms. Investigations using models of systemic inflammation have demonstrated that DEX can attenuate cytokine storms, enhance microvascular perfusion, and reduce glycocalyx shedding[9,10]. However, its efficacy in preserving VEG integrity specifically in the context of oncologic surgery—where surgical trauma acts synergistically with tumor-mediated inflammatory pathways—remains inadequately established in clinical settings.

To address this knowledge gap, we conducted a randomized controlled trial to systematically evaluate whether perioperative DEX administration: (1) Preserves glycocalyx structural integrity; (2) Attenuates surgery-induced cytokine release; and (3) Translates to improved microcirculatory function and clinical outcomes in patients undergoing gastrointestinal tumor resection.

MATERIALS AND METHODS

This prospective, single-center, randomized, double-blind, placebo-controlled trial was conducted at The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, adhering to CONSORT guidelines and Good Clinical Practice principles. The study protocol was approved by the Institutional Ethics Committee of Anhui Provincial Cancer Hospital (approval No. 2024014). Written informed consent was obtained from all participants prior to enrollment. The trial was prospectively registered (ChiCTR2500109633, https://www.chictr.org.cn/showproj.html?proj=221761) with a fixed sample size and no adaptive re-estimation clause. Investigators and statisticians remained blinded. Any mid-course recalculation would have required unblinding or revealing pooled variances, risking inflation of type-I error (“operational bias”). CONSORT and International Conference on Harmonization-E9 discourage unplanned sample-size changes because they compromise the pre-specified α-control; doing so would necessitate α-spending or Bayesian re-analysis, beyond our resources/timeframe. Our data and safety monitoring committee charter allowed an interim look only for safety signals (mortality/serious adverse events), not for efficacy or biomarker variance; those stopping boundaries were never crossed.

No important changes to trial methods, eligibility criteria, or outcome definitions were made after trial commencement.

Participant selection and randomization

This prospective trial enrolled 110 patients aged 18-75 years, with an American Society of Anesthesiologists (ASA) physical status of II or III, who were scheduled for elective open or laparoscopic resection of gastric or colorectal neoplasms with an anticipated duration exceeding 2 hours.

Key exclusion criteria were as follows: (1) Bradycardia (resting heart rate < 50 beats per minute); (2) Hemoglobin level < 80 g/L or albumin < 25 g/L; (3) Chronic kidney disease (estimated glomerular filtration rate < 30 mL/minute/1.73 m²); (4) Hepatic dysfunction (Child-Pugh class C); (5) Active infection or ongoing immunosuppressive therapy; or (6) Coagulopathy, chronic opioid use, or requirement for emergency surgery.

Eligible participants were randomized in a 1:1 ratio using a computer-generated block randomization sequence (block size of 4) prepared by an independent biostatistician. Allocation concealment was strictly maintained through the use of sealed, opaque, sequentially numbered envelopes.

Intervention assignment was managed by an independent anesthesiologist not involved in subsequent patient care or outcome assessment. Patients allocated to the DEX group (n = 55) received an intravenous loading dose of 0.5 μg/kg over 10 minutes, followed by a continuous intraoperative infusion of 0.5 μg/kg/hour and a postoperative infusion of 0.05 μg/kg/hour via a patient-controlled intravenous analgesia pump. The control group (n = 55) received volume-matched normal saline administered identically. All study drugs were prepared in identical opaque syringes to ensure blinding.

To safeguard the blinding, all clinical staff, outcome assessors, and patients were masked to treatment allocation. Similarly, the data analysis statisticians were kept unaware of group codes until completion of the primary analysis.

The CONSORT flowchart is presented in Figure 1.

Figure 1 CONSORT flowchart. DEX: Dexmedetomidine.

Intraoperative management protocol

General anesthesia was induced using etomidate (0.2-0.3 mg/kg), propofol (2-2.5 mg/kg), sufentanil (0.3-0.5 μg/kg), and rocuronium (0.6 mg/kg). Following induction, patients in the DEX group received a loading dose of DEX (0.5 μg/kg administered over 10 minutes) followed by a continuous intraoperative infusion (0.2-0.5 μg/kg/hour) titrated to maintain a Narcotrend index between 40-60. The control group received a volume-matched saline placebo following an identical administration protocol. Anesthesia was maintained with ciprofol (2-3 mg/kg/hour), sevoflurane (1-1.3 minimum alveolar concentration), and remifentanil (0.1-0.2 μg/kg/minute). Neuromuscular blockade was maintained with atracurium, with doses adjusted to sustain a train-of-four count ≤ 1. The study drug and placebo were identical in appearance, packaging, and administration to preserve blinding.

Hemodynamic management was guided by cardiac output monitoring, complemented by real-time microcirculatory assessment using sidestream dark-field (SDF) imaging. Intraoperative hypotension, defined as mean arterial pressure < 65 mmHg lasting ≥ 1 minute, was continuously recorded. Vasopressor administration followed a standardized protocol: Norepinephrine infusion was initiated for hypotension refractory to fluid optimization, while phenylephrine boluses were administered for transient pressure reductions. Both the incidence of hypotension and total vasopressor consumption were documented for subsequent analysis to control for potential confounding effects on glycocalyx integrity.

Standardized ventilation parameters were applied to all patients using volume-controlled ventilation with a tidal volume of 6-8 mL/kg predicted body weight. Respiratory rate was adjusted to maintain normocapnia (end-tidal CO₂ 35-40 mmHg) with positive end-expiratory pressure set at 5 cmH2O. Plateau airway pressures were maintained below 30 cmH2O, with particular attention to peak pressures during pneumoperitoneum. Recruitment maneuvers were performed as clinically indicated. Surgical positioning was standardized according to procedure type: Laparoscopic cases were performed in 15°-20° Trendelenburg position with pneumoperitoneum maintained at 12-14 mmHg, while open procedures were conducted in the supine position. Extreme or prolonged steep Trendelenburg positioning was avoided to minimize hemodynamic and microcirculatory disturbances.

Perioperative fluid management followed a standardized, goal-directed protocol based on stroke volume variation and cardiac output trends. Crystalloids (Ringer's lactate and Plasma-Lyte) served as primary maintenance fluids, while colloids (5% albumin) were administered only when clinically indicated for volume resuscitation. Total intraoperative fluid volumes, including crystalloids, colloids, and blood products, were recorded to avoid excessive fluid loading that could independently affect endothelial glycocalyx integrity.

Postoperative analgesia was provided through a patient-controlled intravenous analgesia system containing sufentanil (0.2 μg/kg) and ondansetron (0.4 mg/kg) in 150 mL saline, programmed for a 3 mL/h basal infusion with 3 mL bolus doses available every 30 minutes. The DEX group's solution additionally contained DEX at 0.05 μg/kg/hour.

Patient recruitment was conducted from March 2022 to August 2023. The final 30-day postoperative follow-up was completed in September 2023.

Primary outcome

The primary outcome was the perioperative plasma concentration of syndecan-1, a specific biomarker of VEG injury. Measurements were taken at four predefined time points: Before anesthesia induction (T0), at abdominal closure (T1), and at 24 (T2) and 72 (T3) hours postoperatively, using a commercial enzyme-linked immunosorbent assay kit (Abcam, ab235649).

Secondary outcomes

Biomarkers of endothelial and inflammatory response: Plasma levels of the following biomarkers were assessed: Syndecan-1, heparan sulfate (MBS2600475), TNF-α, IL-6, C-reactive protein (CRP), procalcitonin, interleukin-10 (IL-10), vascular endothelial cadherin (VE-cadherin), and angiopoietin-2 (Ang-2). Blood samples were collected at baseline (T0), immediately after surgery (T1), and 24 hours postoperatively (T2).

Microcirculation and perfusion parameters: Microvascular function was evaluated using sublingual SDF imaging to measure the microvascular flow index, perfused vessel density (PVD), and the proportion of perfused vessels at T0, T1, and T2. Arterial lactate and the central venous-arterial carbon dioxide difference [P(v-a)CO₂] were measured as complementary markers of global tissue perfusion.

Clinical recovery and postoperative complications: Recovery endpoints included time to extubation, cumulative opioid consumption (converted to intravenous morphine equivalents), postoperative pain scores on the Visual Analog Scale (VAS), quality of recovery at 72 hours on the Quality of Recovery-15 Questionnaire (QoR-15), time to first ambulation, and length of hospital stay. Postoperative complications within 30 days included infection (e.g., surgical site infection, pneumonia), ileus, anastomotic leakage, and acute kidney injury (AKI). Data on intensive care unit (ICU) admission, 30-day readmission, and mortality were extracted from electronic medical records.

Anesthesia depth and sedation: Intraoperative anesthesia depth was monitored continuously using the Bispectral index. Postoperative sedation levels were assessed using the Richmond Agitation-Sedation Scale.

Standardized definitions

Postoperative complications were adjudicated by a blinded investigator using predefined criteria. Complications of Clavien-Dindo grade II or higher were included in the analysis.

Infectious complications: Pneumonia required radiographic confirmation; surgical site infection required antibiotic treatment; urinary tract infection required a positive culture.

AKI: Defined and staged according to KDIGO criteria[11] based on serum creatinine changes.

ICU admission: Defined as an unplanned transfer due to hemodynamic instability, respiratory failure, or need for invasive monitoring/therapy beyond routine care.

Sample size

The sample size was determined based on the anticipated difference in syndecan-1 levels at the end of surgery (T1). Preliminary data and existing literature indicated that syndecan-1 levels in the control group at T1 would be approximately 65 ± 25 ng/mL. We considered a 30% relative reduction (approximately 19.5 ng/mL) in the DEX group to be clinically meaningful. With a two-sided α of 0.05 and 80% power, a total of 100 patients (50 per group) were required. To account for an estimated 10% dropout rate, 110 patients were enrolled.

Although the observed syndecan-1 levels at T1 were lower than projected (control group mean: 48.73 ng/mL), the standard deviation was also substantially smaller than expected (approximately 6-7 ng/mL vs the assumed 25 ng/mL). As a result, the study retained sufficient power to detect a clinically meaningful between-group difference, and no sample size recalculation was performed, in accordance with the pre-registered protocol and ethical approval. No interim analyses or stopping guidelines were predefined for this trial.

Statistical analysis

All statistical analyses were performed using SPSS Statistics version 26.0 (IBM Corp., Armonk, NY, United States) and GraphPad Prism version 9.5.1 (GraphPad Software, San Diego, CA, United States). Continuous variables were tested for normality using the Shapiro-Wilk test. Normally distributed variables were expressed as mean ± SD and compared between groups using the independent-samples t-test, while non-normally distributed variables were expressed as median with interquartile range and compared using the Mann-Whitney U test. Categorical variables were analyzed using the χ² test or Fisher’s exact test. Longitudinal changes in biomarkers (e.g., syndecan-1, IL-6, TNF-α) were assessed using two-way repeated-measures ANOVA with a time × group interaction, followed by Bonferroni post hoc corrections for multiple comparisons. Pearson correlation analysis was employed to examine associations between syndecan-1 and IL-6 levels, as well as between syndecan-1 and clinical outcomes such as postoperative length of stay. Area under the curve (AUC) analyses were conducted to evaluate cumulative inflammatory and endothelial injury burden over time. Effect sizes for continuous outcomes were expressed as mean differences with 95%CIs, and ORs with 95%CIs were reported for binary outcomes. A post hoc multivariable Logistic regression model was developed to explore the independent effect of DEX on 30-day postoperative complications, adjusting for baseline variables including age, comorbidities, and surgical approach.

Receiver operating characteristic (ROC) analyses were performed for the overall cohort, and stratified by treatment arm (DEX vs control), to assess whether DEX influences the predictive utility of syndecan-1 for postoperative complications.

All statistical tests were two-tailed, and a P value < 0.05 was considered statistically significant.

RESULTS

Participant flow and baseline characteristics

A consecutive cohort of 110 patients undergoing elective gastrointestinal tumor resection was prospectively enrolled and randomly allocated in a 1:1 ratio to either the DEX group (n = 55) or the control group (n = 55), with complete protocol adherence and no attrition or major protocol violations throughout the study. As detailed in Table 1, baseline demographic, clinical, and oncological characteristics were well balanced between groups, including age, sex, ASA physical status, operative duration, estimated blood loss, comorbidities, tumor site and stage, surgical approach, and preoperative laboratory values (all P > 0.05). Furthermore, standardized intraoperative management was successfully implemented, as evidenced by the comparable distribution of laparoscopic surgical positioning with a consistent Trendelenburg tilt (15°-20°) and maintained intra-abdominal pressure (12-14 mmHg) across both cohorts. Importantly, other key intraoperative variables with potential to confound glycocalyx assessment—including the incidence and duration of deep hypotension, vasopressor requirements (norepinephrine and phenylephrine), and the total volume and composition of fluid administration (crystalloids, colloids, and blood products)—also demonstrated no significant differences. This comprehensive balance across baseline characteristics and critical perioperative parameters, including surgical positioning and pneumoperitoneum pressure, strengthens the validity of the subsequent findings, indicating that the observed intergroup differences in outcomes are unlikely to be attributable to confounding variations in patient positioning or intraoperative management.

Table 1 Baseline characteristics of study participants, n (%).

Characteristic	DEX group (n = 55)	Control group (n = 55)	P value
Age (years)	62.6 ± 9.1	61.9 ± 10.9	0.704
Sex (male)	36 (65.5)	39 (70.9)	0.682
ASA II/III	35 (63.6)	40 (72.7)	0.413
Surgical duration (minute)	188.3 ± 29.7	189.5 ± 28.5	0.829
Blood loss (mL)	330.9 ± 59.3	338.1 ± 55.0	0.703
Hypertension	22 (40.0)	21 (38.2)	0.849
Diabetes mellitus	12 (21.8)	14 (25.5)	0.67
CAD	6 (10.9)	5 (9.1)	0.758
Tumor site (gastric)	25 (45.5)	30 (54.5)	0.446
Tumor stage I-II	31 (56.4)	28 (50.9)	0.566
Laparoscopic surgery	38 (69.1)	35 (63.6)	0.545
Hemoglobin (g/L)	124.8 ± 14.7	126.2 ± 13.9	0.589
Albumin (g/L)	38.6 ± 3.2	38.1 ± 3.4	0.436
CRP (mg/L)	6.3 ± 2.7	6.7 ± 3.0	0.483
WBC count (× 109/L)	6.8 ± 1.4	6.9 ± 1.5	0.734
Intraoperative deep hypotension	6 (10.9)	8 (14.5)	0.580
Duration of hypotension [minute, median (IQR)]	3 (2-5)	4 (2-6)	0.420
Norepinephrine use	18 (32.7)	20 (36.4)	0.290
Dose (μg/kg·minute)	0.04 ± 0.01	0.05 ± 0.02
Phenylephrine use	12 (21.8)	14 (25.5)	0.650
Total intraoperative fluid volume [mL, median (IQR)]	2450 (2100-2800)	2500 (2200-2900)	0.62
Crystalloid volume (mL)	2100 ± 450	2150 ± 470	0.59
Colloid volume (mL)	350 ± 120	360 ± 110	0.72
Blood products	3 (5.5)	4 (7.3)	0.69
Trendelenburg tilt (°)	17.5 ± 1.2	17.8 ± 1.4	0.481
Intra-abdominal pressure (mmHg)	13.1 ± 0.8	13.3 ± 0.9	0.355

DEX: Dexmedetomidine; ASA: American Society of Anesthesiologists; CAD: Coronary artery disease; CRP: C-reactive protein; WBC: White blood cell; IQR: Interquartile range.

Clinical outcomes

The DEX group exhibited a lower incidence of postoperative infections (7% vs 16%), primarily driven by reduced pneumonia rates. No cases of anastomotic leakage were observed in either group. Ileus was rare and not significantly different between groups. ICU admissions were lower in the DEX group (7% vs 13%). The 30-day readmission and mortality were infrequent (1%-2%) and comparable across groups. DEX also shortened time to extubation (13.1 ± 3.0 0 minutes vs 18.4 ± 4.0 minutes; P < 0.001), reduced opioid consumption (23.1 ± 5.0 0 mg vs 27.3 ± 6.0 mg; P = 0.004), and lowered VAS pain scores (P = 0.002).

Glycocalyx injury and inflammatory cytokines

DEX demonstrated significant protective effects against perioperative glycocalyx degradation and inflammatory activation. At abdominal closure (T1), the DEX group exhibited significantly lower syndecan-1 concentrations (44.72 ± 7.10 ng/mL vs 48.73 ± 6.26 ng/mL; P = 0.002), representing an 8.2% attenuation of glycocalyx shedding. While lower than expected from prior estimates, the reduced inter-patient variability preserved the study’s statistical power. Similar trends were observed at T2 and T3 (P < 0.01).

Heparan sulfate levels were also significantly reduced at T1 in the DEX group (12.19 ± 2.20 ng/mL vs 13.76 ± 2.44 ng/mL; P = 0.001). IL-6 levels at 24 hours postoperatively (T2) were significantly suppressed in the DEX group (110.77 ± 29.72 pg/mL vs 138.86 ± 35.95 pg/mL; P < 0.0001), with a strong positive correlation between IL-6 and syndecan-1 levels (r = 0.71, 95%CI: 0.55-0.82; P < 0.0001). TNF-α and CRP levels were also significantly lower. IL-10 and procalcitonin showed non-significant downward trends, while Ang-2 levels were significantly reduced in DEX-treated patients (P = 0.038). Baseline biomarker levels were comparable between groups (Table 2).

Table 2 Biomarker levels and between-group comparisons over time.

Biomarker	Time point	DEX group (n = 55)	Control group (n = 55)	P value
Syndecan-1 (ng/mL)	T0	20.66 ± 4.43	20.85 ± 5.06	0.832
	T1	44.72 ± 7.10	48.73 ± 6.26	0.002
	T2	34.16 ± 4.81	37.93 ± 5.78	< 0.001
	T3	29.21 ± 3.88	31.06 ± 3.47	0.009
Heparan sulfate (ng/mL)	T0	8.61 ± 1.71	8.70 ± 1.89	0.787
	T1	12.19 ± 2.20	13.76 ± 2.44	0.001
	T2	9.97 ± 2.27	10.98 ± 2.31	0.023
	T3	9.12 ± 2.82	9.29 ± 2.26	0.728
IL-6 (pg/mL)	T0	85.92 ± 24.31	85.50 ± 29.06	0.934
	T1	117.68 ± 40.77	131.03 ± 41.68	0.092
	T2	110.77 ± 29.72	138.86 ± 35.95	< 0.0001
	T3	97.19 ± 38.11	105.00 ± 42.71	0.314
TNF-α (pg/mL)	T0	41.99 ± 9.72	40.30 ± 9.57	0.36
	T1	58.61 ± 11.35	61.52 ± 10.73	0.17
	T2	56.90 ± 10.63	63.75 ± 12.17	0.002
	T3	56.88 ± 11.32	60.36 ± 12.37	0.126

DEX: Dexmedetomidine; IL-6: Interleukin-6; TNF-α: Tumor necrosis factor-alpha.

Microcirculatory and hemodynamic outcomes

SDF imaging demonstrated that perioperative DEX administration significantly enhanced microcirculatory perfusion, as evidenced by improved PVD (21.40 ± 3.50 mm/mm²vs 19.94 ± 2.93 mm/mm²; P = 0.019) and mean flow index (2.83 ± 0.28 vs 2.66 ± 0.33; P = 0.005) at 24 hours postoperatively (T2) compared to controls. These microvascular improvements were accompanied by significantly lower central venous-to-arterial CO₂ difference (3.89 ± 0.84 mmHg vs 4.63 ± 1.06 mmHg; P < 0.001) and reduced serum lactate levels (1.33 ± 0.40 mmol/L vs 1.65 ± 0.54 mmol/L; P = 0.003), indicating enhanced tissue oxygenation and more efficient microcirculatory flow. Temporal analysis revealed that while no baseline differences existed, these beneficial effects emerged distinctly at T2 before attenuating by T3, suggesting a time-dependent optimization of microvascular function with DEX administration (Table 3). The concordance between improved microcirculatory parameters and favorable metabolic markers underscores the potential of DEX to preserve perioperative microvascular integrity and tissue perfusion.

Table 3 Microcirculatory and metabolic parameters between dexmedetomidine and control groups over time.

Parameter	Time point	DEX group (n = 55)	Control group (n = 55)	P value
Perfused vessel density (mm/mm2)	T2	21.40 ± 3.50	19.94 ± 2.93	0.019
Mean flow index	T2	2.83 ± 0.28	2.66 ± 0.33	0.005
P(v-a)CO2 (mmHg)	T0	3.07 ± 0.78	3.27 ± 0.65	0.632
	T1	4.59 ± 0.86	4.94 ± 0.85	0.132
	T2	3.89 ± 0.84	4.63 ± 1.06	< 0.001
	T3	4.00 ± 1.03	4.40 ± 0.87	0.128
Lactate (mmol/L)	T0	1.00 ± 0.32	1.09 ± 0.31	0.504
	T1	1.74 ± 0.37	1.82 ± 0.36	0.968
	T2	1.33 ± 0.40	1.65 ± 0.54	0.003
	T3	1.31 ± 0.58	1.51 ± 0.43	0.183

DEX: Dexmedetomidine.

Recovery and clinical endpoints

DEX administration was associated with significant improvements in recovery, including a 0.8-day reduction in length of hospital stay (P = 0.005), improved QoR-15 scores at 72 hours, and reduced overall 30-day complication rates (12.7% vs 30.9%; P = 0.036). The most affected complications were pneumonia and AKI, both of which were less frequent in the DEX group (Table 4). No differences were observed in readmission or mortality.

Table 4 Comparison of outcomes between dexmedetomidine and control groups, n (%).

Outcome measure	DEX group (n = 55)	Control group (n = 55)	Mean difference or OR (95%CI)	Absolute/relative effect	P value
Intraoperative fentanyl (μg)	156.9 ± 31.6	175.5 ± 42.1	-18.6 (-32.6 to -4.5)	-	0
Length of hospital stay (days)	6.49 ± 1.29	7.29 ± 1.59	-0.8 (-1.35 to -0.25)	-	0.005
QoR-15 score at 72 hours	124.3 ± 9.2	118.7 ± 8.4	5.6 (2.3-9.0)	-	0.001
Postoperative infection	4 (7.3)	9 (16.4)	OR = 0.40 (0.11-1.37)	ARR 9.1%; RRR 55.5%; NNT approximately 11	0.135
Pneumonia	2 (3.6)	6 (10.9)	OR = 0.30 (0.06-1.47)	ARR 7.3%; RRR 67.0%; NNT approximately 14	0.142
Acute kidney injury	1 (1.8)	4 (7.3)	OR = 0.23 (0.02-2.12)	ARR 5.5%; RRR 75.3%; NNT approximately 18	0.195
Ileus	1 (1.8)	2 (3.6)	OR = 0.49 (0.04-5.53)	ARR 1.8%; RRR 50.0%; NNT approximately 56	0.564
Anastomotic leakage	0 (0)	0 (0)	-	-	-
ICU admission	4 (7.3)	7 (12.7)	OR = 0.54 (0.14-2.04)	ARR 5.4%; RRR 42.5%; NNT approximately 19	0.358
30-day readmission	1 (1.8)	1 (1.8)	OR = 1.00 (0.06-16.06)	-	> 0.05
30-day mortality	0 (0)	1 (1.8)	-	ARR 1.8%; RRR 100%; NNT approximately 56	> 0.05
Overall 30-day complications	7 (12.7)	17 (30.9)	18.2% absolute risk reduction (3.1%-33.2%)	ARR 18.2%; RRR 58.9%; NNT approximately 6	0.036

Overall 30-day complications include pneumonia, acute kidney injury, ileus, surgical site infection, and other clinically significant postoperative adverse events. DEX: Dexmedetomidine; QoR-15: Quality of Recovery-15 Questionnaire; ARR: Absolute risk reduction; RRR: Relative risk reduction; NNT: Number needed to treat; ICU: Intensive care unit.

The incidence of postoperative pneumonia, diagnosed according to CDC NHSN criteria based on clinical symptoms, elevated inflammatory markers, and new radiographic infiltrates, was lower in the DEX group [2/55 (3.6%)] than in the control group [6/55 (10.9%)]. All cases of AKI were stage 1 according to KDIGO staging, with no patients progressing to stage 2 or 3; the incidence was 1/55 (1.8%) in the DEX group and 4/55 (7.3%) in the control group. For key binary outcomes, absolute and relative effect sizes were calculated. DEX administration was associated with an absolute risk reduction of 18.2% (95%CI: 3.1%-33.2%) for overall 30-day complications, corresponding to a number needed to treat of 6.

Multivariable Logistic regression

Multivariable Logistic regression identified syndecan-1 > 45 ng/mL at T1 as an independent predictor of postoperative complications, with an OR of 2.88 (95%CI: 0.97-8.59, P = 0.057). Although the association approached statistical significance, the wide confidence interval reflects uncertainty likely due to sample size. Other covariates including age, surgical duration, and preoperative CRP levels were not independently associated with adverse outcomes (P > 0.05). The multivariable Logistic regression model yielded an Akaike information criterion of 120.74 and a Nagelkerke pseudo-R² of 0.064, indicating modest explanatory power. While syndecan-1 > 45 ng/mL demonstrated a near-significant association with postoperative complications, the overall model explained approximately 6.4% of the variance in adverse outcomes (Table 5).

Table 5 Multivariable Logistic regression results.

Variable	Coef.	SE	z	P > |z|	OR (95%CI)
Const	-0.481	2.431	-0.198	0.843	0.618 (0.005-72.488)
Syndecan_high	1.059	0.557	1.902	0.057	2.884 (0.968-8.593)
Age	-0.009	0.029	-0.305	0.76	0.991 (0.936-1.050)
Surgery_duration	-0.005	0.008	-0.605	0.545	0.995 (0.979-1.011)
CRP	-0.006	0.081	-0.075	0.94	0.994 (0.848-1.165)

CRP: C-reactive protein.

Correlation analysis of syndecan-1 with inflammatory response and clinical outcome

A significant time × group interaction was observed for IL-6 and syndecan-1 levels (both P < 0.001), indicating that DEX modified the postoperative inflammatory and endothelial injury trajectories. Cumulative AUC analysis revealed 25% and 19% reductions in IL-6 and syndecan-1 exposure, respectively, in the DEX group. At 24 hours postoperatively (T2), syndecan-1 positively correlated with IL-6 (DEX: r = 0.71; control: r = 0.59, Figure 2A), contrary to the initial summary statement, no statistically significant correlations were observed at individual timepoints. Specifically, the Pearson correlation coefficients were as follows: T0: R = -0.10, P = 0.476; T1: R = -0.02, P = 0.882; T2: R = -0.04, P = 0.796; T3: R = -0.11,P = 0.444. Moreover, syndecan-1 levels at T1 correlated with prolonged hospital stay in DEX patients (β = 0.101, P < 0.0001, Figure 2B), reinforcing its prognostic value as a marker of vascular injury and recovery delay. Here are the time-series plots for IL-6 and syndecan-1, clearly showing their temporal trajectories across T0 to T3 in both DEX and control groups (Figure 3).

Figure 2 Correlation analyses of syndecan-1. A: Correlation between syndecan-1 and interleukin-6 levels at T2 (24 hours postoperatively); B: Correlation between syndecan-1 at T1 and length of hospital stay in the dexmedetomidine group. IL-6: Interleukin-6; DEX: Dexmedetomidine.

Figure 3 Interleukin-6 and syndecan-1 levels over time. A: Interleukin-6; B: Syndecan-1. IL-6: Interleukin-6; DEX: Dexmedetomidine.

Syndecan-1 > 45 ng/mL at T1 predicts postoperative risk

ROC analysis identified syndecan-1 > 45 ng/mL at abdominal closure as an optimal cut-off for predicting postoperative complications, with an AUC of 0.68 (95%CI: 0.59-0.76), sensitivity of 79.2%, and specificity of 43.0% (Figure 4A). Subgroup analyses revealed comparable discriminatory performance between groups. In the DEX arm, the ROC analysis yielded an AUC of 0.71 (95%CI: 0.60-0.82; sensitivity 77.8%, specificity 48.0%), while in the control arm, the AUC was 0.66 (95%CI: 0.55-0.77; sensitivity 80.0%, specificity 42.9%; Figure 4B).

Figure 4 Receiver operating characteristic curves for syndecan-1 levels at abdominal closure (T1) in predicting postoperative complications. A: Overall cohort: A threshold of > 45 ng/mL yielded an area under the curve of 0.68 (95%CI: 0.59-0.76), with 79.2% sensitivity and 43.0% specificity; B: Subgroup analyses: In the dexmedetomidine group, the receiver operating characteristic curve analysis yielded an area under the curve (AUC) of 0.71 (95%CI: 0.60-0.82; sensitivity 77.8%, specificity 48.0%), while in the control group the AUC was 0.66 (95%CI: 0.55-0.77; sensitivity 80.0%, specificity 42.9%). AUC: Area under the curve; ROC: Receiver operating characteristic curve; DEX: Dexmedetomidine.

Subgroup analysis

The anti-inflammatory and endothelial-protective effects of DEX were preserved across subgroups. In patients aged > 65 years, the IL-6 reduction at T2 was even more pronounced (mean difference: -35.7 pg/mL; P = 0.001). In laparoscopic surgery patients, syndecan-1 levels were lower overall, but the relative benefit of DEX remained significant (P = 0.014 for interaction). No interaction was found by tumor type (gastric vs colorectal; Figure 5).

Figure 5 Subgroup analysis: Interleukin-6 reduction at T2 (dexmedetomidine vs control). IL-6: Interleukin-6.

DISCUSSION

EG disruption has emerged as a pivotal mechanism underlying postoperative organ dysfunction, systemic inflammation, and impaired recovery, particularly following major abdominal oncologic surgery. The synergistic effects of surgical trauma and malignancy-associated cytokine release create a profound inflammatory surge that potentiates glycocalyx degradation, resulting in microvascular dysfunction and increased capillary permeability[12]. While DEX, a selective α₂-adrenoceptor agonist with demonstrated endothelial-protective and anti-inflammatory properties, has shown efficacy in preclinical models, robust clinical evidence in surgical oncology populations remains scarce.

Our randomized controlled trial provides compelling evidence that DEX administration significantly preserves glycocalyx integrity, as quantified by an 8.2% reduction in syndecan-1 and an 11.4% decrease in heparan sulfate levels at abdominal closure. Importantly, syndecan-1 is a structural glycocalyx component, and even modest increases in its plasma concentration reflect substantial microvascular damage, as its release is non-linear and threshold-sensitive.

By contrast, IL-6 is a rapidly amplified soluble cytokine, and its large fluctuations over time reflect systemic inflammation rather than direct vascular injury. The strong correlation observed between syndecan-1 and IL-6 (r = 0.71, P < 0.0001) suggests that even small elevations in syndecan-1 are biologically meaningful and may contribute to downstream inflammatory cascades. Furthermore, syndecan-1 at T1 independently predicted prolonged length of stay (β = 0.101, P < 0.0001), reinforcing its clinical relevance as a biomarker of perioperative vascular injury and recovery burden.

These biochemical findings correlate with the pathophysiological model proposed by Becker et al[10], wherein syndecan-1 shedding directly reflects the magnitude of surgical stress response. The observed 20.2% attenuation of IL-6 and 10.7% reduction in TNF-α at 24 hours postoperatively demonstrate DEX's potent immunomodulatory effects, consistent with prior mechanistic studies in both surgical[9] and septic[12] settings.

The clinical translation of these molecular effects was particularly noteworthy. DEX-treated patients exhibited superior microcirculatory perfusion, improved tissue oxygenation, and reduced opioid requirements. These benefits culminated in a 0.8-day reduction in hospitalization and an 18.2% absolute risk reduction in 30-day complications, predominantly pneumonia and AKI - outcomes that compare favorably with established Enhanced Recovery after Surgery (ERAS) protocols[13].

A critical advancement of our study is the identification of syndecan-1 > 45 ng/mL at closure as an independent predictor of adverse outcomes, validating its role as a biomarker of endothelial injury across clinical contexts. This threshold aligns with critical care studies where syndecan-1 levels predicted multi-organ dysfunction[14], suggesting its potential utility for risk stratification in surgical populations.

The present study confirms that perioperative DEX attenuates endothelial glycocalyx degradation in patients undergoing gastrointestinal tumor resection. At abdominal closure, DEX-treated patients exhibited significantly lower levels of syndecan-1 and heparan sulfate, key markers of glycocalyx shedding. These findings are consistent with the mechanistic understanding that DEX inhibits glycocalyx degradation via suppression of systemic inflammation and oxidative stress. Our finding that perioperative DEX significantly reduced syndecan-1 (-8.2%) and heparan sulfate (-11.4%) aligns with animal data showing DEX protects the endothelial glycocalyx. For example, Kobayashi et al[15] found DEX reduced both syndecan1 and TNFα in rat hemorrhagic shock, improving survival. In cell-based research, DEX preserved glycocalyx integrity and VEcadherin in sepsis models. The 20% IL-6 and 11% TNF-α reductions we observed parallel clinical findings: Cho et al[16] (spinal surgery) and Taniguchi et al[17] (sepsis models) reported significant cytokine suppression with DEX. Improved microcirculation, demonstrated by higher PVD and lower lactate/P(v-a)CO₂ in DEX patients, mirrors cardiac-surgery findings: Mohamed et al[18] reported similar enhancements in sublingual microcirculation using DEX. Additionally, a recent laparoscopic cholecystectomy trial showed reduced inflammatory and glycocalyx injury markers as well as better pain control with DEX[19-22].

Preclinical studies provide additional insights into the potential mechanisms by which DEX may protect the endothelial glycocalyx. Experimental models have demonstrated that DEX attenuates systemic inflammation by downregulating proinflammatory cytokines such as TNF-α and IL-6, while concomitantly enhancing anti-inflammatory mediators including IL-10. Its sympatholytic properties, mediated via α2-adrenergic receptor activation, reduce catecholamine-induced endothelial stress and oxidative injury. Furthermore, DEX has been shown to preserve endothelial barrier integrity by stabilizing adherens junction proteins (e.g., VE-cadherin) and limiting glycocalyx shedding of syndecan-1 and heparan sulfate. Collectively, these mechanistic pathways provide a biologically plausible rationale supporting the observed reductions in glycocalyx injury markers and improved microcirculatory profiles in the present study.

This study demonstrated that perioperative administration of DEX significantly attenuated endothelial glycocalyx degradation, reduced systemic inflammatory responses, and improved postoperative recovery outcomes in patients undergoing gastrointestinal tumor resection. Although the actual syndecan-1 concentrations at T1 were lower than those estimated during study planning, the observed inter-patient variability was substantially reduced. This improved precision allowed the study to maintain sufficient statistical power to detect significant between-group differences without the need for mid-study sample size recalculation. Adhering to ethical and registration guidelines, no unblinded interim analysis or adaptive design was implemented.

The discrepancy likely reflects differences in surgical modality (with a higher proportion of laparoscopic cases), timing of sample collection (immediately at closure rather than 30 minutes post-closure), and perioperative fluid and temperature management, all of which may reduce endothelial stress. Nonetheless, the significant difference in syndecan-1 between groups and its correlation with IL-6 underscore its utility as a surrogate marker for vascular injury and inflammation. These findings support the role of DEX in protecting vascular integrity and mitigating inflammation in oncologic surgery, with implications for broader clinical adoption and future multicenter trials incorporating adaptive design features.

Despite the strengths of this study, several limitations should be acknowledged. First, as a single-center trial with a moderate sample size consisting predominantly of ASA II patients, the generalizability of our findings to more diverse surgical populations may be limited. Second, although we quantified circulating glycocalyx components, direct visualization of the endothelial glycocalyx—for instance, via intravital microscopy—was not performed, which could have provided additional mechanistic insight. Third, despite standardization of intraoperative ventilation parameters (tidal volume, positive end-expiratory pressure, plateau pressure), subtle interindividual variations during pneumoperitoneum may have influenced glycocalyx shedding. Future studies incorporating continuous respiratory mechanics monitoring and stricter lung-protective strategies may help clarify these effects. Fourth, although episodes of deep hypotension and vasopressor use were prospectively documented, these variables were not fully integrated into the current statistical model. Fifth, while perioperative fluid management was protocolized and balanced between groups, the study was not powered to discern subtle effects of fluid type (crystalloid vs colloid) or cumulative volume on glycocalyx integrity. Sixth, despite standardized patient positioning, minor variations in Trendelenburg angle or pneumoperitoneum pressure may have affected microcirculatory stress and glycocalyx injury. Future trials may benefit from continuous monitoring of intra-abdominal pressure and patient position to better control for these factors. Finally, it is important to recognize that glycocalyx integrity is influenced by a range of systemic factors—including renal, pulmonary, and cardiovascular function—as well as perioperative interventions such as ventilation, hemodynamic management, and fluid therapy. Thus, changes in biomarkers such as syndecan-1 and heparan sulfate likely reflect systemic glycocalyx dysfunction rather than injury specific to the gastrointestinal tract. Although we sought to minimize confounding through standardized management, the interpretation of our results should take this broader physiological context into account.

CONCLUSION

This study demonstrates that perioperative DEX is associated with reduced glycocalyx degradation, enhanced microcirculatory function, and accelerated recovery following gastrointestinal tumor surgery. Although the data show strong correlations between inflammation and glycocalyx injury, the study design precludes causal inference. Further investigation using time-series analysis, dose-response validation, and confounder-adjusted modeling is needed to determine whether DEX directly mitigates inflammation-induced endothelial damage. These findings nonetheless support the integration of DEX into ERAS pathways as an anti-inflammatory adjuvant and opioid-sparing agent. Syndecan-1 and related glycocalyx components may serve as dynamic biomarkers for perioperative risk stratification. Future research should focus on real-time glycocalyx monitoring and evaluate whether such interventions yield durable oncologic and functional benefits.