Pharmacological and electrical stimulation of the enteric nervous system as a new therapeutic approach for postoperative ileus.
Each abdominal surgical intervention leads to impaired motility of the entire gastrointestinal (GI) tract lasting several days with symptoms like nausea, vomiting, intolerance to food and absence of defecation, which is referred to as postoperative ileus (POI). This condition is a major source of patient morbidity and prolonged hospital stay and represents a significant economic burden to health care systems.
Previous research has showed that POI results from activation of resident macrophages in the muscularis of the intestine induced by intestinal manipulation during abdominal surgery. This event leads to an inflammatory cascade which greatly impacts the muscular function. A novel approach which could prevent the initiation of this inflammatory cascade (i.e. inhibition of macrophage activation) could thus be able to reduce the incidence of POI.
In a murine model of POI, we and others previously showed that electrical stimulation of the cervical vagus nerve (VNS) has anti-inflammatory properties and reduces POI. Cervical VNS namely dampens the activation of resident macrophages via the binding of acetylcholine with α7 nicotinerge acetylcholine receptor leading to a reduced influx of inflammatory cells, a decreased pro-inflammatory gene expression in the muscularisexterna and a fastened recovery of the GI tract. These data suggest that cervical VNS could be a novel therapeutic approach to shorten POI.
Currently, cervical VNS is safely used in patients with epilepsy, rheumatoid arthritis and Crohn’s disease. In brief, a coiled electrode is surgically positioned around the cervical vagus nerve and connected to a pacemaker for daily and chronic stimulation. In our preclinical model however, one single period of electrical stimulation of the cervical vagus nerve already reduces GI inflammation and improves POI. Hence, adding an additional surgical procedure to isolate the cervical vagus nerve is not acceptable. The branches of the abdominal vagus nerve are however easily accessible at the hiatal diaphragm during abdominal surgery, and thus can be electrically stimulated without the need of an extra surgical procedure. In this thesis, the possibility of abdominal VNS was explored in detail in mouse and pig as alternative technique to stimulate the vagus nerve and thus treat POI and to subsequently translate this novel approach to clinical care.
First, we compared the composition of cervical and abdominal vagus nerve (Chapter 1). We observed that the cervical vagus contained a significant higher percentage of myelinated fibers compared to the abdominal vagus nerve. In addition, the cervical vagus nerve was composed of 3 different myelinated fiber types (small, medium and large), while the abdominal vagus nerve only consisted of small myelinated fibers. This insights into the composition and type of vagal nerve fibers are important, since they could determine the optimal stimulation parameters.
In Chapter 2, we therefore determined in a model of POI whether abdominal VNS has the same anti-inflammatory properties as cervical VNS using the same stimulation parameters for both approaches. This study showed that similar to cervical VNS, abdominal VNS improved intestinal transit and reduced manipulation-induced inflammation.
In Chapter 3, we investigated which was the best approach to electrically stimulate the vagus nerve. To this end, we tested several electrodes in pigs and determined which electrode caused the greatest vagal activation using different stimulation parameters (Chapter 2-3). This study showed that abdominal VNS is best achievable with pacing wires within a 15 minute period and does not cause cardiovascular effects. Based on these findings, we conducted a pilot study in patients undergoing abdominal surgery. These patients were treated peroperatively with low or high frequency stimulation of the abdominal vagus nerve to determine which frequency had the greatest anti-inflammatory effect in human. We showed that high, but not low frequency stimulation of the abdominal nervus vagus led to anti-inflammatory effects in the blood as compared to sham-stimulated patients. Furthermore, this technique was safe in humans, i.e. VNS did not cause cardiovascular side effects or leakage of the anastomosis (Chapter 2).
Recently, we and others also reported that the vagus nerve does not directly interact with the resident macrophages, but rather indirectly modulates them via enteric neurons. Therefore, we determined whether enteric neurons possess anti-inflammatory properties and contribute to the tolerogenic phenotype of muscularis macrophages. Using immunohistochemical stainings, we showed that muscularis macrophages are in close proximity to cholinergic enteric neurons both in murine and human ileum. Moreover, supernatant of enteric neuron cultures imprinted bone marrow derived macrophages with a tolerogenic (M2-like) phenotype similar to that of muscularis macrophages. Next, using live Ca2+ imaging of Cx3cr1GFP/wt muscularis macrophages in an isolated muscularis preparation, we showed that electrical or pharmacological (using the 5-HT4 receptor agonist prucalopride) stimulation of enteric neurons reduced ATP-induced activation of muscularis macrophages. To further explore the anti-inflammatory properties of enteric neurons, we evaluated the effect of VNS (acting via activation of enteric neurons) and treatment with prucalopride in a murine model of POI. Both treatments significantly reduced the surgery-induced inflammatory response and improved intestinal transit, a mechanism mediated by α7 and β2 nicotinic receptors located on the muscularis macrophages. Finally, we confirmed our findings in patients undergoing abdominal surgery. Preoperative administration of prucalopride decreased intestinal inflammation in the muscularis collected during surgery and significantly improved clinical recovery. Taken together, our data demonstrate that enteric neurons release mediators that install the tolerogenic phenotype and dampen the activation of muscularis macrophages. This anti-inflammatory property of enteric neurons should be further explored as new treatment for POI.