Gastric Microperfusion in Patients Undergoing Gastroesophageal Resections

Background:

Gastroesophageal resection because of gastroesophageal junction (GEJ) adenocarcinoma is a massive surgical intervention. Nevertheless, surgery is the only treatment with significant long-term survival. Currently, gastroesophageal cancer surgery is performed with upper laparotomy followed by thoracotomy at the Department of Surgical Gastroenterology, Rigshospitalet, Denmark. However, minimal invasive techniques (MIT), e.g. robotic assisted laparoscopy, is in the progress of being implemented in this field as they are hypothesized to be more beneficial for the patients, and in some aspects better than conventional laparoscopic surgery [1]. The benefits of MIT are believed to be due to several factors, a smaller surgical stress response [2] and less pain [3] resulting in earlier mobilization, to mention a few.

When reconstructing the gastrointestinal continuity, a gastric tube is prepared from the upper part of the remaining stomach. It is challenging to visualize, if the remaining stomach has adequate blood supply. Earlier studies have shown an up to 70 % reduction in blood flow, to the upper part of the remnant stomach [4-6], and ischemia is one of the most significant causes of anastomotic leakage [7, 8]. Anastomotic leakage occurs in up to 10 % of the patients and is fatal in up to 50 %. Furthermore, a recent study originated from the investigators department demonstrated, that patients experiencing anastomotic leakage after gastroesophageal cancer resection, have a significantly reduced long-term survival, even when early death and other postoperative complications were accounted for [9]. The overall 5-year survival rates in patients with and without anastomotic leakage were 20 and 35 %, respectively. Therefore, sufficient blood flow to the area of anastomosis is of paramount concern. Different techniques have been used in the attempt to assess the microcirculation during gastroesophageal resection [5, 10-13]. However, none of these techniques has proven to be practical, reliable, and time-efficient, and therefore not being routinely implemented in the clinical settings.

During open gastroesophageal resection, hypotension is common (systolic blood pressure less than 90 mm Hg is experienced in more than 30 % of the patients) and is probably accompanied by a reduced gastric microcirculatory flow. The hypotension may be due to several factors;

• Sympathic nerve system blockage due to thoracic epidural neuraxial anesthesia: Several studies have found reduced splanchnic flow after epidural anesthesia, which was associated with a decrease in systemic resistance and mean arterial pressure [14, 15]. The impact of the time-point for activation of epidural analgesia, on complications in the postoperative period is unknown. An unpublished study (Nielsen T: "Increased cardiac output after Whipple's procedure for pancreatic cancer") suggests that early activation during operation leads to significantly longer stay in the recovery ward after the surgery.
• The initial manipulation of viscera, results in a mesenteric traction reflex (also called eventration syndrome) with vasodilatation and secondary hypotension [16, 17]. The reason for mesenteric traction reflex is unknown, but is believed to be elicited by traction on the mesenterial blood vessels, resulting in a local release of prostacyclin (PGI2) from the endothelium. This reflex has mainly been shown to occur during pancreatic, gastric and aortic surgery, where manipulation of the mesenteric root is inevitable [18-21]. PGI2 is a potent vasodilatator, with inhibitory effect on platelet aggression and on vascular smooth cell proliferation [22-24].

Microcirculatory measurements:

Laser Speckle Contrast Imaging (LSCI) is a relatively new technique for microcirculatory monitoring. When laser light encounters a surface of an object, a random interference effect generates, called laser speckle contrast. The speckle pattern changes depending on the velocity of the object, and it is possible to get information regarding the object's movements, by the fluctuations in the speckle pattern. By measuring the concentration and velocity of blood cells, tissue perfusion may be assessed by LSCI technique. A real-time and non-touch measurement can be made on a large field (0.5 cm x 0.7 cm up to 15 cm x 20 cm). The camera is placed at a distance of 20-30 cm and measures the relative flow (flux) in the regions of interest. The measurement is at a depth of 1-2 mm by infrared light reflected from circulating erythrocytes in the micro-vessels.

Hypothesis and aim:

Robot-assisted laparoscopic surgery alters the hemodynamics and microcirculation in the splanchnic system by reduced release of vasoactive substances, such as PGI2, compared to open surgery. With this study, the investigators wish to evaluate hemodynamic fluctuations and gastric microcirculation, and correlate these to serum levels of PGI2 at different stages of surgery in robotic assisted laparoscopic surgery and open surgery, respectively.

Methods:

2 × 25 consecutive patients selected for robotic assisted laparoscopic or open surgery. Hemodynamic assessments are according to standardized methods. Blood samples will be drawn at: 1) after induction of anesthesia, 2) when the peritoneum is first entered, 3) after 15 min. of surgery, 4) after liberation of stomach, 5) after abdominal skin closure, 6) after extraction of the gastric conduit into the thorax, 7) after formation of the anastomosis, 8) after thoracic skin closure, 9) 18 hours after surgery. LSCI will be used to measure the microcirculation at two locations (3 cm from the pylorus and just below the site of the anastomosis (the body)) on the stomach five times during open surgery, and twice during the thoracic part of robotic assisted laparoscopic surgery.

Statistics:

Statistical power-calculation is carried out on an expected difference of the occurrence of mesenteric traction reflex between patients operated with open vs. MIT. By using sample size of 50 comparing percentages where group one is 55 % and group two is 12 % [17], a requirement of 18 patients in each group is calculated in order to obtain a statistical power greater than 0.80 with a α-level of <0.05.