Occult hypovolaemia leading to poor organ perfusion is assumed to be a major factor in determining postoperative morbidity after bowel surgery. Routine haemodynamic monitoring such as heart rate (HR), arterial pressure and central venous pressure (CVP) may remain relatively ‘normal’ despite reduced blood flow to certain organs such as the gut and, therefore, are insensitive indicators of hypovolaemia . It has been shown that intraoperative hypoperfusion of the gut is associated with increased morbidity and length of hospital stay .
Impressive improvements in patient outcome have been reported when therapy has been targeted at optimizing oxygen delivery to tissues and avoiding hypovolaemia [2-5]. Such goal-directed therapy has been used in cardiac surgery where improved gut perfusion, as reflected by gastric tonometry, was achieved by increased fluid administration alone . Furthermore, other studies investigating intravascular fluid optimization have demonstrated improved patient outcomes in orthopaedic and general surgery [5,7,8]. In a survey of moderate-risk-elective surgery patients, gastrointestinal complications occurred in 22% of patients . Gastric hypoperfusion and arterial base deficit were among the strongest intraoperative predictors of these complications.
Hypovolaemia following major surgery [9,10] has been implicated in the development of poor intestinal perfusion and increased mucosal permeability. Furthermore, the incidence of leaking anastomosis in the colon correlates directly with low intestinal tissue oxygen pressure . Supplemental oxygen administration increases intestinal  and subcutaneous  oxygen pressures. However, even supplemental oxygen fails to improve tissue oxygen pressure in hypoperfused tissues .
In this study, we assessed whether increasing amounts of intravenous (i.v.) crystalloids during laparotomy to avoid occult hypovolaemia result in higher tissue oxygen pressure in the small and large bowel as well as in the subcutaneous tissue.
Materials and methods
This study was performed in accordance with National Institute of Health guidelines for the use of experimental animals. The local Animal Studies Committee approved the protocol. In all, 27 domestic pigs (weight: 20-28 kg) were fasted overnight but had free access to water.
The pigs were sedated with intramuscular telazol (2 mg kg−1), ketamine (1 mg kg−1) and xylazine (1 mg kg−1). Anaesthesia was induced by inhalation of isoflurane and maintained with isoflurane (1.1-1.4%). All pigs were orally intubated and ventilated with oxygen in nitrogen (fraction of inspired oxygen (FiO2) = 0.3). The animals were ventilated with a volume-controlled ventilator and tidal volume was kept at 10-13 mL kg−1, with the respiratory rate adjusted (11-14 breaths min−1) to maintain end-tidal carbon dioxide (etCO2) tension at 40 ± 3 mmHg.
Through a left groin cut down, indwelling catheters were inserted into the femoral artery and inferior vena cava. A balloon-tipped catheter was inserted into the pulmonary artery through the left femoral vein. Location of the catheter tip was determined by observing the characteristic pressure trace on the monitor as it was advanced through the right side of the heart into the pulmonary artery. With the pig in the supine position, a midline laparotomy was performed. A urinary catheter was directly inserted in the bladder.
Intramural tissue oxygen probes were inserted through 20-G cannulae from the serosal side into the jejunal and colonic wall. The probes were inserted into the tissue plane between the serosa and mucosa. Resistance at the end of the insertion canal indicated the correct position of the probes. If mucosal penetration occurred as indicated by loss of resistance the probe could immediately be repositioned before final fixation. Great care was taken to minimize handling of the intestine and to return the bowel to a neutral position; intestinal retractors were not used.
Subcutaneous tissue oxygen pressure was measured with a silicone tonometer. The silicone tubing was inserted subcutaneously via a 16-G needle in the left upper foreleg. After inserting a tissue oxygen sensor and a thermistor the tubing was flushed with deoxygenated saline.
Body temperature was taken from the thermistor of the pulmonary artery catheter and maintained at 37.0 ± 0.5°C using a warming mattress and a patient air warming system (Bair Hugger; Arizant Healthcare Inc., Eden Prairie, MN, USA). The abdomen was closed after completion of surgical preparation, and the pigs were allowed to recover for 30 min before measurements began.
The animals were preoperatively assigned to one of three fluid treatments. The respective treatments were started immediately after induction of general anaesthesia and maintained throughout the entire study. Fluids were administered with an infusion pump.
The fluid treatments were:
- Low fluid volume: nine animals received an initial bolus of 3 mL kg−1 of lactated Ringer's solution and subsequently a continuous infusion of 3 mL kg−1 h−1 of lactated Ringer's solution throughout the study.
- Medium fluid volume: nine animals received an initial bolus of 10 mL kg−1 of lactated Ringer's solution followed by a continuous infusion of 7 mL kg−1 h−1 throughout the study.
- High fluid volume: nine animals received an initial bolus of 10 mL kg−1 and subsequently a continuous infusion of 20 mL kg−1 h−1 of lactated Ringer's solution. Additional boluses of 100 mL lactated Ringer's solution were administered, only in the high volume group, when the pulse pressure variation (dPP) was greater than 13% (target controlled).
The dPP was calculated every 5 min from the arterial pressure reading. According to Michard and Teboul , dPP was calculated as the difference of maximum and minimum values divided by their respective means and expressed as a percentage. A variation in dPP of over 13% has been shown to be likely to increase stroke volume with additional i.v. fluid [15,16]. Consequently, a dPP under 13% was considered an adequate intravascular volume.
After surgery the animals were allowed to stabilize for at least 30 min. Subsequently, all pigs were subjected in a random crossover design to an FiO2 of 0.3 and oxygen challenge with an FiO2 of 1.0. The designated FiO2 concentration in each pig was maintained for 30 min to establish steady-state conditions in the tissues before measurements began. The conditions were considered stable when tissue oxygen pressure did not change more than 3 mmHg within 10 min. Then subcutaneous and intestinal tissue oxygen tensions were measured for 30 min with the same oxygen concentration (treatment period). After that the alternative oxygen concentration was started. Again, 30 min were allowed to elapse to establish steady-state conditions. Measurements were then repeated for 30 min. etCO2 tension was kept at 40 mmHg throughout the study. The total duration of the study and hence of the assigned volume treatment was about 4 h. We considered this time sufficient to produce significant effects of the respective i.v. fluid treatment.
Intramural intestinal tissue oxygen tension was measured with a polarographic tissue oxygen sensor (Licox CC 1.2; Gesellschaft fuer Medizinische Sondentechnik, Kiel, Germany). The oxygen sensor is a flexible microcatheter probe used for long-term monitoring of partial pressure of oxygen in tissue and body fluid. The sensor is an electrochemically reversible polarographic cell embedded into the tip of a completely sealed microcatheter tube. Temperature sensitivity of the sensor is 0.25% °C−1 but a thermistor (Licox 8.1; Gesellschaft fuer Medizinische Sondentechnik, Kiel, Germany) placed in close proximity provided accurate temperature compensation. The device was calibrated before insertion. Calibration remained stable (within 8% in room air) in vivo for at least 8 h. Oxygen sensors were calibrated in room air (ambient PO2, 154 mmHg). For calibration purposes, a calibration card was inserted into the Licox device. All PO2 values measured before insertion were within 8% of 154 mmHg.
Subcutaneous oxygen tension was measured with a tonometer consisting of a 12-cm length of silicone tubing (2.0 mm outer diameter, 1.4 mm inner diameter) (Dow Corning, Midland, MI, USA). Silicone is freely permeable to oxygen and after equilibration the PO2 of saline within it reflects the mean oxygen tension of the surrounding tissue. Within this silicone tubing, a polarographic tissue oxygen sensor (Licox CC 1.2) as well as a thermistor (Licox 8.1) were inserted to measure subcutaneous oxygen tension and temperature.
HR was recorded from the electrocardiogram. Systemic haemodynamics including CVP and pulmonary artery pressures were continuously displayed on a multimodular monitor recorded every 10 min. A thermodilution method was used to determine cardiac output at the end of each different oxygen condition (mean of three consecutives measurements). Cardiac output and systemic vascular resistance were indexed to body weight. Body temperature was recorded from the thermistor in the pulmonary artery catheter. An arterial blood gas sample was drawn after measuring cardiac output during each oxygen condition and immediately analysed. Urine was collected and the amount measured hourly before it was discharged.
Tissue oxygen pressure data for each inspired oxygen concentration were recorded for 30 min and stored on line to a portable computer. The mean value was calculated over this 30-min period using commercially available software (AcqKnowledge 3.9 for Mac)
Data are presented as mean±SD. All data were tested for normal distribution with a Kolmogorov-Smirnov test. One-way analysis of variance (ANOVA) followed by the Tukey-Kramer post test for multiple comparisons were used to describe differences between the treatment groups. Absolute values were used for all calculations. A P < 0.05 was considered statistically significant.
The animals in the low fluid volume group received a total amount of 3.8 ± 0.1 mL kg−1 h−1 of lactated Ringer's solution during the study, including the initial bolus of 3 mL kg−1, and the medium volume group received 8.9 ± 1.1 mL kg−1 h−1 The animals in the high fluid treatment group received a total of 28 ± 2.8 mL kg−1 h−1 including both the initial bolus and the additional fluid boluses as triggered by a dPP of over 13%. Thus, the total amount of lactated Ringer's solution given was 360, 912 and 2650 mL for the low, medium and high volume groups, respectively. The bowel at the insertion sites of the tissue oxygen probes appeared normal at the end of the studies. We did not observe any haematomas at the probe insertion sites.
In the low and medium volume groups, mean arterial pressure (MAP) was 58 ± 7 and 62±8 mmHg, respectively. The high volume treatment resulted in a significantly increased MAP (85 ± 10 mmHg, P < 0.01). Cardiac index increased with the amount of volume administered from 92 ± 14 mL kg−1 min−1 in the low volume group to 105 ± 12 mL kg−1 min−1 in the medium volume group and to 118 ± 15 mL kg−1 min−1 in the high volume group. CVP and pulmonary capillary wedge pressure were significantly increased in the high volume group while HR tended to be lower during high volume treatment (Table 1).
In the jejunum, wall tissue oxygen pressure in the low, medium and high volume groups were similar (46 ± 15, 37 ± 7 and 37± 19 mmHg, respectively) with FiO2 of 0.3. Increasing FiO2 to 1.0 resulted in a marked increase in jejunal tissue oxygen pressure in all pigs to approximately 75 mmHg. However, tissue oxygen pressure in the jejunal wall between the three volume regimens did not differ (Fig. 1). Similarly, in the colon wall, tissue oxygen tension in the low, medium and high volume groups were 49 ± 11, 55 ± 12 and 53 ± 19 mmHg, respectively, with FiO2 of 0.3. These values were essentially doubled in all three groups with FiO2 of 1.0. Again, the increase was similar in all groups regardless of the volume treatment (Fig. 2).
Subcutaneous tissue oxygen tension in the low and medium volume group was with 30% oxygen 58 ± 14 and 53± ±13 mmHg, respectively, and thus significantly lower than in the high volume group with 84 ± 27 mmHg (P < 0.05). Increasing the inspired oxygen to 100% was followed by a significant increase in the subcutaneous tissue oxygenation in all groups. Subcutaneous tissue oxygen tension increased to 122 ± 27 and 102 ± 48 mmHg in the low and medium volume group, respectively, compared with 178 ± 56 mmHg (P < 0.05) in the high volume group (Fig. 3).
All three groups displayed similar pH, PaCO2 and core body temperatures (Table 1). Arterial oxygen pressure was approximately 110 mmHg with 30% inspired oxygen and 450 mmHg with 100% oxygen; no differences between the volume treatments were found (Table 1). Haemoglobin seemed to be lower in the high volume group but differences between the groups were not significant.
The main result of this study is that the intestinal wall oxygen pressure was not affected by any of the three different fluid replacement regimens. As oxygenation in the tissue is tightly linked to blood flow, our results suggest that blood perfusion of the intestinal wall did not improve with high volume fluid administration. On the other hand, a low volume fluid regimen, which resulted in decreased CVP, lower cardiac output and reduced urine output, apparently did not harm the intestine as judged by the tissue oxygen tension. We have previously shown that the intestinal mucosa has the ability to adjust blood flow even during low output conditions such as severe haemorrhage . We found that in anaesthetized pigs exposed to haemorrhage, small intestinal mucosal blood flow was maintained even when MAP was decreased to 40 mmHg. Such profound regulation of intestinal blood flow may explain the results found in the current study, which is in agreement with recent studies suggesting low fluid regimens may result in lower incidence of complications after bowel surgery [18, 19]. It can be speculated that beneficial effects of low volume therapy such as less oedema formation in the bowel wall may be beneficial for tissue healing and for outcome as long as the tissue oxygenation is sufficient
Subcutaneous tissue oxygen pressure increased with high fluid volume administration, which is in agreement with a recent study by Arkilic and colleagues in patients undergoing colon surgery . This suggests that systemic flow is diverted to ‘less important’ tissues such as subcutaneous tissues when oxygen requirements of ‘more important’ organs such as the gut are met. This assumption is also supported by the fact that hypovolaemia decreases tissue perfusion and oxygenation in subcutaneous tissue long before it affects oxygenation in central organs [14,22].
We tested three fluid regimens. Results for the low volume group are in agreement with recently published outcome studies indicating that fluid restriction might be rather beneficial for outcome after uncomplicated abdominal surgery [18-20]. The fluid regimen is quite different between these studies, however. They all have, in common, perioperative fluid restriction which results in fewer complications. For the low fluid volume group, we chose a similar fluid regimen as used by Nisanevich and colleagues: a fixed low infusion rate of lactated Ringer's solution . The 7-9 mL kg−1 h−1 of Ringer's lactate solution given in the medium volume group reflects the common clinical practice in our hospital. The amount of fluids given in the high volume group may appear large, however, intraoperative fluid administration of up to 20 mL kg−1 h−1 for abdominal surgery is recommended in many textbooks [23-25] and by other studies [26,27]. Additional fluid in our high volume group was given when dPP was over 13%. Several recent publications suggest that cardiac index can be increased with fluid loading when dPP is beyond this limit [15,28]. Since it can be assumed that intestinal blood flow would profit from increased cardiac output, we wanted to increase cardiac output in order to optimize bowel perfusion and oxygenation.
This approach is supported by several clinical studies, such as that by Mythen and Webb  who reported decreased incidence of gastrointestinal hypoperfusion after perioperative plasma volume expansion in cardiac patients and Wakeling and colleagues  who showed earlier return of normal bowel function and a reduced length of hospital stay after goal-directed fluid therapy in patients undergoing major bowel surgery. On the other hand, this disagrees with the restricted fluid approach discussed above [18,19].
In the present study, haemodynamic stability was greater in the high fluid volume group, as judged by blood pressure (BP), cardiac index and urinary output, which is in accordance with data reported by Campbell and colleagues  on patients undergoing major abdominal surgery. However, cardiovascular parameters were not significantly different between the low and the medium volume groups; indicating that the low fluid group was able to hemodynamically compensate for the restricted fluid administration except for a low urine output. The urine output of less than 0.5 mL kg−1 h−1 in the low volume group most likely indicates some degree of hypovolaemia while in the medium volume group urine output was normal.
The study has some limitations. Firstly, the study was performed in an animal model due to the fact, that direct measurements of intestinal tissue oxygenation in patients are invasive, time consuming and require special skills. We chose the pig because of its anatomical and physiological similarity in respect to the cardiovascular system and the digestive tract. Furthermore, clinical conditions in an operating room were imitated in the laboratory (general anaesthesia, mechanical ventilation and monitoring). Secondly, due to the relatively small number of animals per group and the rather short follow up, some biologically relevant effects may have been missed. Randomization of the animals, however, should have prevented a significant bias in the baseline fluid status. Thirdly, we did not measure tissue oedema in the gut. High fluid loads may have caused increased interstitial fluid accumulation in the gut wall. No major bowel surgery was performed. Thus induction of capillary leak in the bowel wall possibly resulting in gut wall oedema was apparently not severe. Still, low, medium and high volume administration resulted in similar gut wall tissue oxygen pressure. Consequently, we assume no harm was done by interstitial fluid in our study. On the other hand, we found no obvious benefit of high volume fluid therapy and high oxygen delivery for tissue oxygen pressure in the gut wall. Fourthly, it must be considered that the tissue oxygen sensor was placed into the small and large bowel wall, respectively. Thus, we measured tissue oxygen pressure in the submucosa and the muscularis layer of the bowel wall. This could also explain the possible differences between our study and findings in the study by Mythen and Webb  in which the mucosal PCO2 gap reflecting gastric mucosal perfusion was decreased during high fluid administration. We have examined the small intestinal and colon wall tissue oxygen pressure but the gastric wall or the mucosal layer in any other part of the intestinal tract may respond differently. Finally, our animals were young and healthy. The reduced capacity to compensate for a defined insult as seen in many elderly surgical patients is difficult to reproduce in animal models and should be considered when the results of this study are interpreted.
In conclusion, high fluid volume administration resulted in an increased BP and cardiac output but intestinal wall oxygen pressure was not affected. Low fluid volume management did not appear to have any negative effects on oxygen tension in the jejunum and colon. These results suggest a very efficient autoregulation of intestinal blood flow in healthy subjects undergoing uncomplicated abdominal surgery.
This work was supported by the Research Fund of the Department of Anesthesia, University of Bern, and by NIH Grants GM 58273 and GM 061655 (Bethesda, MD, USA).
1. Mythen MG, Webb AR. Intra-operative gut mucosal hypoperfusion is associated with increased post-operative complications and cost. Intensive Care Med
2. Wilson J, Woods I, Fawcett J et al
. Reducing the risk of major elective surgery: randomised controlled trial of preoperative optimisation of oxygen delivery. BMJ
3. Shoemaker WC, Appel PL, Kram HB, Waxman K, Lee TS. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest
4. Boyd O, Grounds M, Bennett E. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA
5. Wakeling HG, McFall MR, Jenkins CS et al
. Intraoperative oesophageal Doppler guided fluid management shortens postoperative hospital stay after major bowel surgery. Br J Anaesth
6. Mythen MG, Webb AR. Perioperative plasma volume expansion reduces the incidence of gut mucosal hypoperfusion during cardiac surgery. Arch Surg
7. Conway DH, Mayall R, Abdul-Latif MS, Gilligan S, Tackaberry C. Randomised controlled trial investigating the influence of intravenous fluid titration using oesophageal Doppler monitoring during bowel surgery. Anaesthesia
8. Gan TJ, Soppitt A, Maroof M et al
. Goal-directed intraoperative fluid administration reduces length of hospital stay after major surgery. Anesthesiology
9. Roumen RM, van der Vliet JA, Wevers RA, Goris RJ. Intestinal permeability is increased after major vascular surgery. J Vasc Surg
10. Ohri SK, Somasundaram S, Koak Y et al
. The effect of intestinal hypoperfusion on intestinal absorption and permeability during cardiopulmonary bypass. Gastroenterology
11. Shandall A, Lowndes R, Young HL. Colonic anastomotic healing and oxygen tension. Br J Surg
12. Ratnaraj J, Kabon B, Talcott MR, Sessler DI, Kurz A. Supplemental oxygen and carbon dioxide each increase subcutaneous and intestinal intramural oxygenation. Anesth Analg
13. Greif R, Akca O, Horn EP, Kurz A, Sessler DI. Supplemental perioperative oxygen to reduce the incidence of surgical-wound infection. Outcomes Research Group. N Engl J Med
14. Gottrup F, Firmin R, Rabkin J, Halliday BJ, Hunt TK. Directly measured tissue oxygen tension and arterial oxygen tension assess tissue perfusion. Crit Care Med
15. Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest
16. Michard F, Boussat S, Chemla D et al
. Relation between respiratory changes in arterial pulse pressure and fluid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care Med
17. Krejci V, Hiltebrand L, Banic A, Erni D, Wheatley AM, Sigurdsson GH. Continuous measurements of microcirculatory blood flow in gastrointestinal organs during acute hemorrhage. Br J Anaesth
18. Brandstrup B, Tonnesen H, Beier-Holgersen R et al
. Effects of intravenous fluid restriction on postoperative complications: comparison of two perioperative fluid regimens: a randomized assessor-blinded multicenter trial. Ann Surg
19. Joshi GP. Intraoperative fluid restriction improves outcome after major elective gastrointestinal surgery. Anesth Analg
20. Arkilic CF, Taguchi A, Sharma N et al
. Supplemental perioperative fluid administration increases tissue oxygen pressure. Surgery
21. Nisanevich V, Felsenstein I, Almogy G, Weissman C, Einav S, Matot I. Effect of intraoperative fluid management on outcome after intraabdominal surgery. Anesthesiology
22. Jonsson K, Jensen JA, Goodson 3rd WH, West JM, Hunt TK. Assessment of perfusion in postoperative patients using tissue oxygen measurements. Br J Surg
23. Hwang G, Marota JA. Anesthesia for abdominal surgery. In: Hurford WE, Bailin MT, Davison JK, Haspel KL, Rosow C, eds. Clinical Anesthesia Procedures of the Massachusetts General Hospital
. Philadelphia, USA: Lippincott-Raven, 1997: 330-346.
24. Sendak M. Monitoring and management of perioperative fluid and electrolyte therapy. In: Rogers MC, Longnecker DE, Tinker JH, eds. Principles and Practice of Anesthesiology
, 1st edn. New York, USA: Mosby-Year Book, 1993: 863-966.
25. Tonnesen AS. Crystalloids and colloids. In: Miller RD, ed. Anesthesia
, 3rd edn. New York, USA: Churchill Livingstone, 1990: 1439-1465.
26. Campbell IT, Baxter JN, Tweedie IE, Taylor GT, Keens SJ. IV fluids during surgery. Br J Anaesth
27. Holte K, Klarskov B, Christensen DS et al
. Liberal versus restrictive fluid administration to improve recovery after laparoscopic cholecystectomy: a randomized, double-blind study. Ann Surg
28. Michard F, Teboul JL. Using heart-lung interactions to assess fluid responsiveness during mechanical ventilation. Crit Care