The endocrine, metabolic, and inflammatory responses to injury and infection consist of a variety of physiological changes, collectively called the surgical stress response. This process is considered to be detrimental to restoring homeostasis, wound healing, and returning the organism to a normal state of health and activity. Although the endocrine and metabolic changes have been extensively studied [1,2], little attention has been paid to the pathophysiology of intestinal changes, such as gastrointestinal ischemia during surgery, which may result in a decrease of bowel function  and a breakdown of the intestinal barrier in the normal gut . Global indices of adequacy of tissue perfusion, such as blood pressure, heart rate, cardiac output, urine flow, and arterial pH, may not reflect regional perfusion [5,6]. In experimental models, the intramucosal pH (pHi) of the gut is a reliable index of local tissue perfusion . The gastric pHi can be measured easily and reproducibly using a silicone balloon tonometer placed in the lumen of the stomach  by using the Henderson-Hasselbalch equation . pHi can be calculated from the corrected intramucosal CO2 (PiCO2) and arterial bicarbonate concentration.
The stress response to major abdominal surgery  may be inhibited by epidural anesthesia. Furthermore, thoracic epidural anesthesia (TEA) has a beneficial effect on bowel function [10,11]. Postoperative epidural pain management results in reduced postoperative morbidity  and shorter hospital stay . Therefore, we hypothesized that using TEA perioperatively may result in the maintenance of normal visceral perfusion and intraabdominal metabolism.
The aim of this intraoperative study was to evaluate the impact of TEA on pHi in patients undergoing major abdominal surgery.
After institutional approval and informed consent, 30 patients (ASA physical status II or III) scheduled for major abdominal surgery were included in the study. Patients with congestive heart failure, coronary artery disease, coagulopathies, perforated viscus, esophageal or gastric pathology, and esophageal or gastric surgery were excluded, as were pregnant patients. All patients were premedicated with oral triazolam (0.25 mg). On arrival at the operating room, an infusion of 500 mL of lactated Ringer's solution was started. A radial artery catheter was inserted for blood pressure monitoring and blood sampling. Additionally, a gastric tonometer tube (TRIP[trade mark sign] NGS Catheter; Instrumentarium Corp., Helsinki, Finland) was placed into the stomach via the nasogastric route. An epidural catheter was inserted into the epidural space through an 18-gauge Tuohy needle using the paramedian approach. Catheters were inserted at the T5-6 to T9-10 interspace. Using a placebo-controlled design, patients were randomly allocated into two groups. In the TEA group (n = 15), 8 mL of bupivacaine 0.5% was administered, and in the control group (n = 15), 8 mL of isotonic sodium chloride solution was injected into the epidural space. The level of epidural anesthesia was evaluated by the detection of cold 30 min after the local anesthetic injection. General anesthesia was induced with 3 mg/kg propofol, 50 [micro sign]g/kg alfentanil, and 0.15 mg/kg alcuronium and was maintained with a propofol drip (3-7 mg [center dot] kg-1 [center dot] h-1) and 25 [micro sign]g/kg alfentanil according to patient requirements.
Before the administration of the local anesthetic or placebo and anesthesia induction, the following baseline measurements were performed: systolic, mean arterial blood pressure, diastolic arterial blood pressure, heart rate (HR), oxygen saturation (SaO2), central body temperature (urine catheter probe), and arterial and gastric secretion samples. Arterial blood gas analysis and measurement of intramucosal CO (2) (PiCO2) were performed using a standard blood gas analyzer (BG Electrolyts[trade mark sign]; Instrumentation Laboratory, Lexington, MA).
The pHi was calculated after an equilibration period of 45 min by inserting the values of PiCO2 and the arterial bicarbonate into a modified Henderson-Hasselbalch equation. A pHi of 7.32 was taken as the lower limit of normal [4,15,18]. Measurements were obtained before and 60, 120, 180, and 240 min after local anesthetic injection. Surgery began immediately after anesthesia induction; therefore, measurements from 60 to 240 min were taken intraoperatively. After the study, the epidural catheter was used for pain management in all patients for 5-7 days using a patient-controlled motor pump (CADD-PCA[trade mark sign]; Pharmacia Deltec Inc., St. Paul, MN).
The size of the groups was estimated with reference to our pilot study  and previous studies [4,15]. Data are expressed in mean values +/- SD and were compared by using unpaired Student's t-tests and chi squared test using Yates rectification. A P value <0.05 was considered significant.
The maximal pHi decrease during the observation period was larger (P < 0.001) in the control group (0.16 +/- 0.04) than in the TEA group (0.07 +/- 0.05). There were 10 patients in the control group and 2 patients in the TEA group who had evidence of gastric mucosal ischemia (pHi <7.32) at the end of surgery, which is statistically significant (P < 0.01). The pHi changes are shown in Figure 1, and PiCO2 data are plotted in Figure 2. After 180 and 240 min, both pHi and PiCO (2) reached statistically significant differences between groups.
No surgical or anesthetic complications were seen in either group. No surgical gastrointestinal complications were recorded postoperatively. The block in the TEA group was established from T1 to T12 (median 10 segments, range 8-12 segments) using 8 mL of bupivacaine 0.5%. There were no significant differences between the TEA and control groups in gender (male to female 7:8 in both groups), age (59 +/- 16 vs 63 +/- 12 yr), weight (76 +/- 10 vs 71 +/- 14 kg), or height (170 +/- 9 vs 168 +/- 7 cm). Surgical procedures are summarized in Table 1. The duration of surgery was similar in both groups (229 +/- 51 vs 219 +/- 47 min). Mean arterial pressure was unchanged without intergroup differences during the study. A slight increase in HR was observed in the TEA group, reaching a level of significance after 120 min (Figure 3). There was no significant difference in SaO2 and decrease of central body temperature between the TEA and control groups (1.2 +/- 0.6 vs 1.5 +/- 0.9[degree sign]C). Blood gases were not different between groups (Table 2). A significant increase of PaO2 in both groups was caused by a fraction of inspired oxygen of 0.4-0.5 during mechanical ventilation.
The major findings of the present study, which focused on the intraoperative period, was that the control group showed a significant decrease in pHi as an effect of increased PiCO2. However, none of our study patients showed clinical signs of postoperative organ dysfunction. A potential influence of these findings by changes in hemodynamics, oxygenation, and central body temperature can be excluded, as there were no differences in these variables between the two groups. Only a small but significant increase in HR was recorded in the TEA group. This increased HR may result in an increased cardiac output and increased visceral perfusion. It is possible that cardiorespiratory variables greater than normal levels are necessary for perfusion of gastric mucosa during surgery . Previous studies aiming to reduce the incidence of postoperative complications have relied on supernormal cardiac output . The assumption is that an increased total oxygen consumption is synonymous with increased oxygen demand. We do not have data for total body oxygen transport or consumption, but adequacy of the global circulation does not guarantee adequacy of perfusion in vulnerable tissues, as various studies have shown [6,7]. Gastric pHi indirectly reflects the adequacy of oxygen delivery to the splanchnic bed. This area is particularly vulnerable to insufficient cardiac output . Doubts regarding the calculation of pHi have been raised when normocapnia is not maintained or when bicarbonate is administered . However, in this study, arterial pH and HCO3 showed no significant changes and no intergroup differences, whereas pHi and PiCO2 displayed an unstable course in the control group and stable conditions in the TEA group. Additionally, 10 of the 15 patients in the control group had evidence of gastric ischemia (pHi <7.32), whereas only 2 patients in the TEA group had a pHi <7.32. These findings lead to the conclusion that the pHi changes recorded in this study may have been influenced by the administration of bupivacaine into the epidural space.
Our results are in accordance with previous studies. Sutcliffe et al.  demonstrated that the injection of epidural bupivacaine 0.125% supplemented with diamorphine 0.2 mg/mL, followed by a bypass, resulted in a decreased incidence of low pHi postoperatively, compared with a group in which diamorphine 0.2 mg/mL alone was administered epidurally in the same way. Johansson et al.  demonstrated that the epidural administration of bupivacaine resulted in an immediate increase of signal level using a laser Doppler flowmeter as a measure of increased arterial perfusion of the intestinal wall. Additionally, an obvious visible increase in small bowel motility was observed in 10 of 15 patients. Similar results have been reported by Aitkenhead et al. , who observed a 22% increase in blood flow in the canine colon after subarachnoidal spinal nerve block.
However, previous studies have not discussed the question of whether epidural anesthesia blocks the reduction of visceral blood flow and small bowel motility caused by surgical manipulation. This effect on intestinal blood flow may be due to decreased mesenteric vascular resistance secondary to sympathetic block, resulting in an adequacy of intramucosal oxygenation, as the data from our study, which were collected during surgery, show.
Postoperative epidural bupivacaine application prevents short intestinal paralysis, as reported by Ahn et al. . Their deduction was that general sympatholysis is of major importance for this effect. Thorn et al. , who studied the effects of epidural morphine and bupivacaine in healthy volunteers, concluded that local anesthetic blockade of sympathetic efferents may increase gastrointestinal motility. In the present study, bowel motility was not recorded. However, our data suggest that the maintenance of pHi as a measure of unchanged intestinal metabolism may result in stable bowel function during surgery. Additionally, evidence of gastric mucosal ischemia is associated with increased postoperative complications and costs . Regardless of the initial stimulus (surgery, anesthesia, bacterial infection), the integrity of gut mucosa is compromised by ischemia. This leads to translocation of bacteria and endotoxin from the gut lumen . The subsequent stimulation of inflammatory pathways can eventually lead to the characteristic tissue destruction seen in multiorgan failure. Postoperative organ dysfunction involves the whole organism, with tissue destruction distant from the site of surgery secondary to uncontrolled activation of inflammatory pathway. Therefore, attention must be focused on prevention .
In conclusion, the major finding of our study is that TEA prevented the decrease of pHi during major abdominal surgery as an effect of stable visceral perfusion. TEA may be a helpful method for intra- and postoperative treatment of surgical stress.
1. Kehlet H. Modification of response to surgery by neural blockade: clinical implications. In: Cousins MJ, Brdenbaugh PO, eds. Neural blockade in clinical anesthesia and management of pain 2nd ed. Philadelphia: JB Lippincott, 1987;145-88.
2. Kehlet H. The stress response to anaesthesia and surgery: release mechanisms and modifying factors. Clin Anaesthesiol 1984;2:315-39.
3. Carpenter RL. Regional anesthesia improves postoperative outcome. What further studies do we need? In: A. Van Zundert, ed. Highlights in pain therapy and regional anesthesia. V. Permanyer Publications, 1996:325-8.
4. Mythen GM, Webb AR. Intraoperative gut mucosal hypoperfusion is associated with increased post-operative complications and cost. Intensive Care Med 1994;20:99-104.
5. Shoemaker WC, Czer LS. Evaluation of the biologic importance of various hemodynamic and oxygen transport variables; which variables should be monitored in post-operative shock? Crit Care Med 1979;7:424-30.
6. Fiddian-Green RG, Baker S. Predictive value of stomach wall pH for complications after cardiac operations: comparison with other monitoring. Crit Care Med 1987;15:153-6.
7. Hartmann M, Montgomery A, Jonsson K, et al. Tissue oxygenation in hemorrhage shock measured as transcutaneous oxygen tension, subcutaneous oxygen tension, and gastrointestinal intramucosal pH in pigs. Crit Care Med 1991;19:205-10.
8. Riddington D, Venkatesh B, Clotton-Brock T, et al. Measuring carbon dioxide tension in saline and alternative solutions: quantification of bias and precision in two blood gas analyzers. Crate Care Med 1994;22:96-100.
9. Seeling W, Kustermann J, Schneider E. Postoperative Katheder-periduralanalgesie nach abdominellen Eingriffen. Bupivacain versus Buprenorphin peridural. Reg Anaesth 1990;13:78-87.
10. Thorn T, Sundberg A, Wattwill M, et al. Effects of epidural bupivacaine and morphine on bowel function and pain after hysterectomy. Acta Anaesthesiol Scand 1989;33:181-5.
11. Thorn SE, Wattwill M, Naslund I. Postoperative epidural morphine but not epidural bupivacaine delays gastric emptying on the first day after cholecystectomy. Reg Anesth 1992;17:91-4.
12. Hasenbos M, van Egmond J, Gielen M, et al. Postoperative analgesia with high thoracic epidural versus intramuscular nicomorphin after thoracotomy: part III-the effect of per- and postoperative analgesia on morbidity. Acta Anaesthesiol Scand 1987;31:645-51.
13. Guinard SO, Mavocordatos P, Chiolero R, et al. A randomized comparison of intravenous versus lumbar and thoracic epidural fentanyl for analgesia after thoracotomy. Anesthesiology 1992;77:1108-15.
14. Kapral S, Gollmann G, Lehofer F. Gastric tonometry as a visceral perfusion monitoring during thoracic epidural anaesthesia. Acta Anaesthiol Scand Suppl 1996;109:178-80.
15. Johansson K, Ahn H, Lindhagen J, et al. Effect of epidural anaesthesia on intestinal blood flow. Br J Surg 1988;75:73-6.
16. Shoemaker WC, Appel PL, Kram HB, et al. Postoperative trial of supranormal values of survivors as therapeutic goal of high-risk surgical patients. Chest 1988;94:1176-86.
17. Benjamin E, Polokoff E, Oropello JM, et al. Sodium bicarbonate administration affects the diagnostic accuracy of gastrointestinal tonometry in mesenteric ischemia. Crit Care Med 1992;20:1181-3.
18. Sutcliffe NP, Mostafa SM, Gannon J, et al. The effects of epidural blockade on gastric intramucosal pH in the peri-operative period. Anaesthesia 1996;51:37-40.
19. Aitkenhead AR, Wishart HY, Brown DA. High spinal nerve block for large bowel anastomosis. Br J Anaesth 1978;50:177-83.
20. Ahn H, Bronge A, Johansson K, et al. Effects of continuous postoperative epidural analgesia on intestinal motility. Br J Surg 1988;75:1176-8.
21. Thorn SE, Wattwil M, Kallander A. Effects of epidural morphine and epidural bupivacaine on gastroduodenal motility during the fasted state and after food intake. Acta Anaesthesiol Scand 1994;38:57-62.
22. Deitch EA. The role of intestinal barrier failure and bacterial translocation in the development of systemic infection multiorgan failure. Arch Surg 1990;125:403-5.