Septic shock remains a life-threatening condition characterized by widespread systemic inflammation and vascular dilatation, leading to multi-organ dysfunction. The complexity of this severe condition contributes to a high mortality rate in critically ill patients despite advancements in clinical care (1). According to the third international consensus definition (Sepsis-3), septic shock is defined as persistent hypotension with mean arterial pressure of less than 65 mmHg, requiring vasopressors to keep it above 65mmHg and elevated levels of serum lactate of more than 2 mmol/L and suspicion of infection (2). Peritonitis originating from abdominal infections, such as like bowel perforation constitutes one of the causes of sepsis and septic shock (3). The sympathetic nervous system (SNS)- with renal nerves detected circularly around renal arteries- plays a crucial role in septic shock pathophysiology because sympathetic hyperactivation modifies vascular stability and organ failure (4).

Renal sympathetic denervation (RSD) was introduced a couple of decates ago as a therapeutic option in conditions such as resistant hypertension and other cardiovascular disease (5, 6, 7, 8). Despite the initially controversial results, recent studies confirm effectiveness and safety of the method in controlling resistant hypertension (9, 10). However, by interrupting renal sympathetic nerve activity this intervention have detrimental effects in septic shock, due to the irreversible inhibition of the SNS.

Nevertheless, such severe conditions as septic shock generate obstacles to conducting studies on humans. For this reason experimental animal models are selected as means to get insight to humans. Pigs are very similar to human animals in terms of anatomy, physiology and genetics, with their immune effectors to be comparable to human structures and functions (11). Still there are very few studies in the international literature dealing with the impact of renal sympathetic denervation on systemic, organic and inflammatory, response in septic shock environment, but none of them was conducted in experimental porcine animal models (1, 12, 13, 14).  

Therefore the present study was designed aiming to inspect whether renal sympathetic denervation affects systemic response into a hyperdynamic state of septic shock and, simultaneously, to determine its degree of influence either locally in kidneys or at a systemic level. This knowledge may contribute to clinical implications regarding application of RSD for treatment of arterial hypertension, influencing the indications and contraindications of the method.

Materials and Methods

Animals and preparation

The present study was approved by the Bioethical Committee of Aristotle University of Thessaloniki, the Ethics Committee of Companion Animal Clinic of School of Veterinary Medicine of Aristotle University of Thessaloniki (Registration num:EL-54-BIOexp-18) and Animal Care Committee of the Region of Central Macedonia of Greece (Protocol num: 54162(167)/09-02-2024)

The study included a total number of 14 Large White pigs, female and male, aged 3-4 months and body weight 25±5kg. All animals were transported to the Companion Animal Clinic of School of Veterinary Medicine of Aristotle University of Thessaloniki, 24 hours prior to the day of the experiment and were housed in the stable facility of the inpatient unit in specially designed stainless-steel cages for pigs, with free access to food and water and under controlled environmental conditions (temperature 18–22 °C, relative humidity 55–65%, and a 12-hour light/dark cycle).

Anesthesia and ventilation

All animals received pre-anaesthetic medication consisting of midazolam (0.1 mg/kg, Richmond Vet Pharma Inc, Buenos Aires, Argentinae), butorphanol (0.1 mg/kg, Vetviva Richter GmbH, Wels Austria), medetomidine (0.1 mg/kg, Eurovet Animal Health, Bladel, Nederlands), and ketamine (10 mg/kg, Vetviva Richter GmbH, Wels, Austria), administered intramuscularly into the psoas muscles. Then animals were subsequently transferred on the operating table of a fully equipped experimental operating theatre of the clinic.

Catheterization of the auricular vein was performed using an abbocath  (22G, Polymed Medicure Limited, New Delhi, India)to establish peripheral venous access. General anesthesia was induced with intravenous administration of propofol (2-3 mg/kg, CP-Pharma Handlesgesellschaft mbH, Burgdorf, Germany) followed by endotracheal intubation with a size 7–8 endotracheal tube (Jiangsu Hendhong Medical technology Co. Ltd, Yizheng, China). General anesthesia was maintained with administration of 100% oxygen and Isoflurane (1-2%, Piramal Critical Care B.V. Inc, Voorshcoten, Nederlands) until the end of the study. Hemodynamic support was provided by intravenous administration of Lactated Ringer’s solution (10 mL/kg/h, Bradex Inc, Krioneri, Greece). In addition, intravenous ampicillin (5 g, Intervet Hellas Inc., Athens, Greece) was administered as preoperative antimicrobial prophylaxis (15).

Experimental protocol and Surgical Technique

All laboratory animals were randomly allocated into two groups:

• Group R (Renal denervation): Pigs undergoing bilateral sympathetic denervation of the renal vessels (n = 9)
• Group C (Control): Pigs undergoing a sham procedure without renal denervation (n = 5)

Abdominal surgical field was prepared by using 10% povidone–iodine solution (Betadine, Lavipharm Inc, Athens, Greece) and applying sterile draping for sterilization. A midline laparotomy incision was performed in order to enter into the peritoneal cavity through the musculoaponeurotic layers of the abdominal wall.

Bladder catheterization was performed with placement of a suprapubic urinary catheter (14Fr 2way foley, Shangai International Holding Corp. GmbH, Hamburg, Germany). Subsequently, bilateral incision of the lateral peritoneal reflections was carried out to mobilize the colon away from the surgical field and to allow access to the retroperitoneal space, the kidneys, and the renal vessels.

In Group R, circumferential dissection and complete skeletonization of the renal arteries and their branches were performed bilaterally along their entire course, from their origin at the abdominal aorta to the renal hilum. Perivascular tissue was excised to ensure interruption of bilateral renal sympathetic neural signaling.

In Group C, the procedure was completed after access to the retroperitoneal space and exposure of the renal vessels, without performing sympathetic denervation.

In all animals, the colon was subsequently opened, and fecal material was released into the peritoneal cavity in order to induce fecal peritonitis and septic shock according to the criteria. Finally, the abdominal wall was closed with continuous sutures (PDS Plus suture 1, Ethicon J&J Med Tech, Edinburg, Scotland), and the skin was closed in anatomical layers (Prolene suture 2/0, Ethicon J&J Med Tech, Edinburg, Scotland)

During the first phase of the experimental protocol (surgical intervention), animals received intravenous paracetamol (20 mg/kg, Fresenius Kabi Hellas Inc, Athens, Greece)  and intramuscular fentanyl  (0.05 mg/kg, Kalcex Inc, Riga, Latvia) every 2 hours. During the second phase (septic shock), additional analgesia was administered, including intramuscular morphine (3 mg/kg, Molteni Pharmaceutics Inc, Florence,Italy) and intravenous tramadol (2 mg/kg, Molteni Pharmaceutics Inc, Florence,Italy) as this condition was considered painful and of extremely high severity (15, 16, 17, 18, 19, 20).

Autopsy and Histopathological Analysis

Scheduled euthanasia was performed 6 hours after the induction of fecal peritonitis and septic shock, in accordance with the fundamental principles of laboratory animal euthanasia. Specifically, while remaining under general anesthesia, the pigs received an intravenous overdose of propofol (CP-Pharma Handlesgesellschaft mbH, Burgdorf, Germany).

The previous abdominal incision was reopened, and the thoracic wall was also opened for collection of tissue samples from the liver, lungs, and heart. The kidneys were excised en bloc together with the renal vessels, and tissue samples from the lungs, liver, and heart were collected and fixed in 10% neutral buffered formalin for histological examination. Histological sections from each organ were prepared using standard paraffin embedding techniques. Tissue sections of 4 μm thickness were stained with hematoxylin and eosin (H&E) and evaluated microscopically.

Each tissue section was examined under light microscopy for the presence or absence of pathological lesions. Kidney samples were assessed for acute tubular injury, glomerular damage, interstitial capillary congestion and microhemorrhages. Lung samples were evaluated for alveolar injury, edema, leukocytic infiltration, capillary congestion, atelectasis, and microhemorrhages. Heart samples were examined for edema by measuring the interfibrilar distance, congestion- increase in capillaries, microscopic bleeding and leukocytic infiltration. Liver samples were assessed for capillary congestion, degenerative hepatocellular changes, periportal infiltration and Kupffer cell hyperplasia.

Hemodynamics, monitoring and urine output

During both the surgical procedure and the septic shock phase, hemodynamic parameters were continuously monitored. A sphygmomanometer was used to measure systolic (SBP) and diastolic (DBP) arterial blood pressure, as well as heart rate (HR), with the cuff applied to the thigh of each animal. Measurements were recorded prior to exploratory laparotomy in both groups. In Group R, values were additionally recorded after renal vessel denervation, whereas in Group C measurements were taken after access to the retroperitoneal space. Finally, measurements were obtained after induction of fecal peritonitis and septic shock in both groups. At the same time points, urine output (URINE) was assessed via Foley urinary catheters.

Laboratory Parameters

Each pig underwent a total of three blood samplings: the first prior to laparotomy, the second following completion of the surgical procedure, and the third 6 hours after induction of fecal peritonitis and septic shock just before euthanasia. Approximately 7 mL of blood was collected from a central vein (femoral or jugular) using a 10 mL syringe and transferred into blood collection tubes, which were subsequently sent for laboratory analysis. The measured parameters included White Blood Cell count (WBC) and Platelet count (PLT), coagulation parameters, specifically Prothrombin time (PT), activated Partial Thromboplastin time (aPTT), and International Normalized Ratio (INR). Also biochemical analysis was performed including measurement of Blood Urea Nitrogen (BUN), Creatinine (Creat), Potassium (K), Sodium (Na), hepatic enzymes Aspartate Aminotransferase (AST/SGOT) and Alanine Aminotransferase (ALT/SGPT), and C-reactive protein (CRP).

Statistical Analysis

Sample size calculation was performed using G*Power software version 3.1.9.7, based on data from the international literature. Data are presented as means ± SD. Statistical analysis was performed using R: A Language and Environment for Statistical Computing, version 1.8.4 (R Foundation for Statistical Computing, Vienna, Austria). Normality of data distribution was assessed using the Shapiro–Wilk test. To assess changes over time, linear mixed-effects models were applied with group, time, and their interaction as fixed effects, and subject-specific random intercepts to account for repeated measurements. Within-group changes were studied using estimated marginal means from the mixed-effects models. For qualitative variables, odds ratios (ORs) and expected frequencies (EFs) were calculated. A p value less than 0.05 was considered to be statistically significant.

Results

Table 1 summarizes the baseline demographic, clinical, and laboratory characteristics of the study population. A total of 14 pigs were included, with 9 in the R group and 5 in the C group. Overall, baseline demographic, clinical, and laboratory characteristics were comparable between the R and C groups, with no statistically significant differences observed across all variables (p > 0.05).

Table 1. Baseline demographic, clinical and laboratory characteristics in total population and groups of study.

SBP: systolic blood pressure, DBP: diastolic blood pressure, HR: heart rate, URINE: urine output, WBC: white blood cell, CRP: C-reactive protein, PT: prothrombin time, INR: international normalized ratio, aPTT: partial thromboplastine time, PLT: platelet count, BUN: blood urea nitrogen, CREA: creatinine, K: potassium, Na: sodium, SGOT: aspartate aminotransferase (AST), SGPT: alanine aminotransferase (ALT).  Values are mean ± SD. R group: renal denervation, C group: control.

Systemic hemodynamic measurements

Renal sympathetic denervation causes a strong, immediate, and sustained drop in both systolic (SBP) and diastolic (DBP) blood pressure in pigs of R group. Systolic blood pressure (SBP) decreased significantly immediately after surgery in denervated pigs (p < 0.001) and continued to drop at 6 hours post-operatively (p < 0.001) Compared to control group, the post-operative SBP decrease in R pigs was significant (p < 0.001). Similarly, DBP decreased significantly immediately after surgery in denervation group (p < 0.001) and further at shock time (p < 0.001). Compared to control group, DBP in R group was significantly lower immediately postoperatively (p < 0.001). However, after the induction of shock the two groups had no statistically significant difference (p = 0.848).

Renal denervation strongly affects heart rate (HR), with an immediate and sustained increase, both within the R group and relative to control animals at 6 hours postoperatively. Heart Rate (HR) increased significantly immediately after surgery in denervated pigs (p < 0.0001) and continued to rise 6 hours postoperatively (p < 0.001). Nevertheless, at shock time point the increase in R pigs was not significantly different compared to control group, (p = 0.078).

Moreover, denervation induces a dramatic and sustained decrease in urine output in R pigs, most pronounced at shock time point. Urine output in R group decreased significantly immediately after surgery (p < 0.001) and further at 6 hours post-operatively (p < 0.001) Compared to control animals, urine output was significantly lower in R group immediately after operation (p <0.0042) but as opposed to 6 hours time point when two groups did not show statistically significant difference (p = 0.183) (Figure 1).

Inflammatory biomarkers

White blood cell count (WBC) did not change significantly after surgery in the denervated pigs (p> 0.5) and there were no significant differences compared to control group at any time point (p > 0.5). On the other hand, renal denervation caused a clear and statistically significant increase in CRP in R group, both immediately (p = 0.0028) and 6 hours after the shock induction (p = 0.0098).  Although CRP increased in the R group, the change was not significantly larger than the control group at any time point (p > 0.26) (Figure 2).

Coagulation system markers

Renal denervation caused a delayed and strong increase in PT in the R group, but no immediate change was observed. Prothrombin Time (PT) did not change immediately after surgery in the operated pigs (p = 0.254), but increased markedly 6 hours post-operatively (p < 0.001). The denervation effect is evident compared to control group after shock-induction, meaning this increase is likely attributable to the renal sympathetic denervation rather than baseline fluctuations. The 6h postoperatively increase was significantly larger than in control animals (p = 0.0028), indicating a denervation-specific effect on PT.

Equivalent was the observed effect on INR, which did not change immediately after surgery in denervated animals (p = 0.253), but increased significantly 6 hours postoperatively (p < 0.001). The increase at shock time was significantly larger than in the control group (p = 0.0027), indicating an effect on coagulation system. As for the impact on Partial Thromboplastin Time (aPTT), this marker did not change immediately after surgery in R group (p = 0.37), but increased significantly at shock time point (p < 0.001). However, this time point increase was not significantly different from control group (p = 0.22), though the magnitude suggests a biologically meaningful effect. Lastly, Platelets count (PLT) did not change immediately after surgery in R group (p = 0.15), but decreased significantly 6 hours postoperatively (p = 0.0002), indicating a delayed thrombocytopenic response. However, the drop was not statistically different from control group either immediately postoperatively or at shock time (p > 0.11), likely due to small sample size or variability in controls, although the magnitude suggests a clear denervation-associated decrease (Figure 3).

Renal function and Electrolytes biomarkers

The effect of renal denervation on renal function in septic shock was based on Blood Urea Nitrogen (BUN) and Creatinine (CREA) measurements. BUN did not change immediately after surgery in the group of denervated pigs (p = 0.731), but increased significantly after the shock induction (p < 0.001). However, this reported increase was not significantly different from the control group (p > 0.12). The renal denervation effect on creatinine (CREA) appears delayed, showing up only at 6 hours postoperatively in the R group (p < 0.001). Although creatinine increased in the R group at the 6 hours after surgery time point, the change was not significantly larger than any change seen in control animals (p = 0.095).

Moreover, potassium (K) and sodium (Na) measurements took place in both groups in order to study the effect on electrolytes in septic shock. Potassium (K) did not change immediately after surgery in denervated animals (p = 0.245), but increased significantly at 6 hours time point of shock (p = 0.0004). However, the increase was not significantly different from the control group (p > 0.15). Accordingly, Sodium (Na) did not change immediately or 6 hours after surgery at shock time in the R group of renal denervation (p= 0.884 and p= 0.474, respectively). There were no significant differences compared to the control group at any time point (p > 0.45) (Figure 4).

Hepatic function biomarkers

Hepatic function was based on SGOT and SGPT measurements as depicted on Figure 5, respectively. Neither SGOT nor SGPT were affected by renal denervation in this dataset. SGOT (AST) did not change immediately or 6 hours after surgery in R group (p> 0.6). There were no significant differences compared to control group at any time point (p > 0.39). Likewise, SGPT (ALT) did not change immediately after denervation or at shock time in the R group (p= 0.731 and p= 0.254, respectively). Similarly, there were no significant differences compared to the control group at any time point (p > 0.24).

Analysis of alterations over time

Additionally, renal denervation effects on laboratory and clinical parameters were evaluated by comparing the change of values from preoperative to induction of shock between the R (denervation) and the C (control) groups (Table 2).

Table 2: Changes from baseline to septic shock.

SBP: systolic blood pressure, DBP: diastolic blood pressure, HR: heart rate, URINE: urine output, WBC: white blood cell, CRP: C-reactive protein, PT: prothrombin time, INR: international normalized ratio, aPTT: partial thromboplastine time, PLT: platelet count, BUN: blood urea nitrogen, CREA: creatinine, K: potassium, Na: sodium, SGOT: aspartate aminotransferase (AST), SGPT: alanine aminotransferase (ALT).  Values are mean ± SD. R group: renal denervation, C group: control.

Significant differences between groups were observed for several variables:

• International Normalized Ratio (INR) increased more in the R group compared to C (p= 0.015), indicating a greater post-operative prolongation of coagulation time.
• Prothrombin time (PT) also increased significantly in R versus C (p= 0.016).
• Systolic blood pressure (SBP) decreased more in the R group (p= 0.003).
• Urine output (URINE) was lower in R compared to C (p 0.049).

For other parameters, including WBC, PLT, CRP, BUN, CREA, PTT, HR, SGOT, SGPT, K, and Na, no statistically significant difference was observed between groups over the 6-hour post-operative period (all p > 0.05).

Microscopic findings and histopathological analysis

Table 3 represents the pathological lesions found microscopically in tissue sections after histopathological analysis. No lesion differed significantly between groups after correction for multiple statistical testing. However, directional trends were observed, with pulmonary injury in R group of denervation and reduced hepatic degenerative changes compared to controls. In kidney, glomerular damage was more frequent in C group (80%) rather than R group (33%), but no statistically significantly difference observed (p= 0.265). In liver, R group had much lower degenerative changes (22%) compared to control group (80%), although the difference was not statistically significant (p= 0.09). Histopathology of lung tissue revealed higher atelectasis in R group (89%) compared to C group (40%) and, also, higher leukocytic infiltration in R group (78%) rather than C group (20%), but none was statistically significantly different (p= 0.094 and p= 0.09, respectively). Additionally, denervated animals manifested more frequent alveolar damage but not statistically different than C group (p= 0.265). Myocardium examination had no meaningful group separation.   

Table 3: Histopathological analysis.