Wound infections and colonic anastomotic leakage are common and serious surgical complications (1,2). Successful surgical wound healing and resistance to wound infections critically depend on adequate tissue oxygen tension (3,4). Similarly, gut tissue oxygen tension correlates inversely with the risk of colonic leakage (1).
During surgery, local vascular supply is disrupted as a result of vessel injury and thrombosis. This leads to a cascade of events, including platelet degranulation, and release of complement, kinins, and chemotactic factors. In response, neutrophils, lymphocytes and, later, macrophages and fibroblasts migrate to the site of injury. The ensuing combination of a decreased vascular supply, high cellularity, and an increased oxygen demand causes wounds to be hypoxic compared with normal tissue (5).
A simple and effective method for increasing perioperative tissue oxygen tension is administration of supplemental oxygen. Supplemental oxygen increases subcutaneous (3) and intestinal tissue oxygen tension (6). Previous studies have shown that tissue oxygen monitoring during a sudden increase of inspired oxygen (“oxygen challenge”) can be used to determine the adequacy of fluid resuscitation (7,8).
Additional crystalloid fluid administration increases subcutaneous tissue oxygen tension (9) and improves outcome for patients undergoing minor surgery (10,11). However, this conclusion is far from confirmed, as several clinical studies demonstrated better outcome with restricted perioperative fluid administration (12–14).
The effect of supplemental oxygen and supplemental fluid on the tissue of primary interest, the injured perianastomotic and anastomotic tissue, has not been investigated. The objective of the current study was to test the hypothesis that supplemental fluid (primary outcome) and supplemental oxygen (secondary outcome) increase tissue oxygen tension in healthy colon, and in perianastomotic and anastomotic colon tissue.
After approval from the Bern University Animal Studies Committee, we studied 16 healthy domestic pigs. Pigs were chosen as they are omnivores with an intestinal physiology closely resembling that of humans. The pigs were approximately 8 months old and weighed 32.1 ± 1.8 kg. All animals were fasted overnight and had free access to water.
A peripheral IV catheter was inserted in the ear for administration of fluids and medications. Anesthesia was induced with midazolam 10 mg, atropine 1 mg, xylazine (1 mg/kg), and ketamine (1 mg/kg), and maintained with inhaled isoflurane (0.8%–1.0%), 0.6 μg · kg−1 · h−1 fentanyl, and pancuronium 0.02 mg · kg−1 · h−1 IV. All pigs were endotracheally intubated, and their lungs were mechanically ventilated with 11–14 breaths/min, a tidal volume of approximately 8–10 mL/kg and a Fio2 of 30% during surgery.
An arterial catheter was inserted in the femoral artery for direct arterial blood pressure monitoring. A balloon-tipped pulmonary artery catheter was inserted via the external or internal jugular vein. A urinary catheter was placed as well. After catheter insertion, initial hemodynamic and blood gas baseline values were recorded.
For intestinal tissue oxygen tension monitoring and anastomosis surgery, a midline laparotomy was performed. The colon anastomosis was performed by an experienced surgeon. For intramural intestinal oxygen tension measurement, the surgeon inserted the Clark-type tissue oxygen tension sensors through a 20-gauge cannula into a section of the healthy and perianastomotic colon between the serosal and the mucosal tissue planes. This method has been used by several authors (6,15), and results in tissue oxygen tension measurements that are slightly higher than noninvasive serosal tissue oxygen tension measurements (16). For intra-anastomotic measurement, the probe was sewn between the wound edges during anastomosis closure, and was similarly located between serosal and mucosal surfaces. Care was taken to minimize handling of the intestines and to return the bowel to a neutral position. Intestinal retractors were not used. To avoid fluid loss by evaporation, the abdomen remained closed after surgery until the end of the study.
Tissue Oxygen Tension Measurement
Tissue oxygen tension was measured with a temperature-corrected Clark-type electrode (Licox, Gesellschaft für Medizinische Sondensysteme, Kiel, Germany). In vitro accuracy of the electrodes in a water bath at 37°C is ±3 mm Hg for the range from 0 to 100 mm Hg, and ±5% for the range 100–360 mm Hg (17). Thermistors are incorporated in the probes, as the polarographic electrodes are temperature-sensitive with a deviation of 0.25%/°C. Calibration remains stable (within 8% of baseline value) for at least 8 h. For calibration purposes, a calibration card containing factory calibration settings was inserted into the Licox device. All sensors were exposed to room air (ambient Po2 154 mm Hg) before measurements to ensure adequate function. All air-Po2 values measured before insertion were ±10% of 154 mm Hg. To exclude a significant drift of the Licox probe (>10%), probes were again exposed to room air after each investigation. No relevant drift was observed throughout the entire study. Correct positioning of the Licox probe was confirmed by macroscopic inspection and measurement values typical of colon tissue at Fio2 of 30%, e.g., approximately 50 ± 15 mm Hg (6). After each change of Fio2 level, at least 30 min were allowed for electrode-tissue equilibration. Stabilization of tissue oxygen tension measurements was defined as less than ±3 mm Hg variation for tissue oxygen measurements in 10 min.
Additional measurements included hemodynamic and respiratory values. Core temperature was measured with a thermometer probe inserted into the distal esophagus. All values, including tissue oxygen tension values, were recorded manually at 5 min intervals throughout the study. Arterial blood gas and thermodilution cardiac output were measured at the end of each study period. The intestinal tissue was checked for macroscopic signs of edema (tissue looking bloated or swollen) by the surgeon during initial anastomosis surgery and at the end of the experiment.
Preoperatively, all animals were assigned to high or low crystalloid fluid administration using a computer-generated randomization list. The treatment was initiated after start of anastomosis surgery.
- Low crystalloid group (n = 8): 3 mL · kg−1 · h−1 crystalloid (lactated Ringer’s solution).
- High crystalloid group (n = 8): 10 mL/kg initial crystalloid bolus, 18 mL · kg−1 · h−1 crystalloid (lactated Ringer’s solution).
The pigs were ventilated with 30% oxygen for 3 h. Subsequently, the pigs were ventilated with 100% oxygen for 1 h. For analysis, all continuous measurements obtained over the last 30 min of each oxygen condition were averaged in each animal. Tissue oxygen tension measurements obtained during change of Fio2 indicated tissue oxygen tension response to the oxygen challenge (7,8).
The number of pigs required for this study was calculated as follows: Previous studies showed that a difference in subcutaneous tissue oxygen tension of >15 mm Hg is clinically important (4). The standard deviation of previous tissue oxygen tension studies in animals (6) and humans typically ranges between 10 and 20 mm Hg. Assuming a difference of at least 20 mm Hg between the fluid treatment groups and a standard deviation of 14 mm Hg, eight animals in each group provide an 80% power to detect a significant difference between the two fluid groups at an [alpha]-level of 0.05.
Data were tested for normal distribution using Q–Q-plot and Shapiro-Wilk-test. Normally distributed data are presented as mean ± sd. Non-normally distributed data are presented as median (interquartile range). Baseline data were compared between the groups with unpaired Student’s t-test. For comparison of Fio2 conditions, paired t-test and Wilcoxon’s signed rank test were used for normally distributed data and non-normally distributed data respectively. ANOVA for repeated measurements was used to test for differences between the treatment groups. As anastomosis tissue oxygen tension was not normally distributed, a natural log transformation was used in the repeated measurements ANOVA. SPSS 11.0 (Chicago, IL) was used for statistical calculations. P < 0.05 was considered statistically significant.
All animals (n = 16) survived until the end of the treatment period (4 h). There were no differences for hemodynamic or blood gas variables during baseline between the groups.
Hemodynamic Variables and Blood Gas Results During Treatment (Table 1)
Animals in the high group received 2998 ± 147 mL crystalloid during the study. Animals in the low group received 457 ± 27 mL crystalloid during the study. Consequently, urine output was significantly higher in the high fluid group (P < 0.001). Arterial blood pressure, pulmonary arterial pressure, and cardiac output did not differ between the groups. However, in the last study hour, there was a trend towards a decrease in cardiac output in the low fluid group, which did not reach statistical significance. Heart rate was slower and pulmonary capillary wedge pressure was higher in the high fluid group (both: P = 0.02). The hemoglobin concentration was lower in the high fluid group during the treatment period (P < 0.01). As expected, arterial oxygen increased significantly under ventilation with 100% oxygen (in both groups: P < 0.01). Pco2 remained stable throughout the study (Table 1).
Tissue Oxygen Tension Results (Table 2, Fig. 1)
Tissue temperature remained stable throughout the experiment. There was no difference in tissue temperature between the fluid groups among the three probe sites. In both groups, ventilation with 100% oxygen doubled tissue oxygen tension in healthy colon (P < 0.01) and perianastomotic colon (P < 0.01). Tissue oxygen tension measurements directly in the anastomosis showed a large interquartile range during ventilation with 30% and 100% oxygen. Ventilation with 100% oxygen increased intra-anastomotic tissue oxygen tension (low fluid: P = 0.04; high fluid: P = 0.01). Tissue oxygen tension increased at all probes sites by >20% within 5 min (Table 2, Fig. 1), fulfilling the requirements of a positive oxygen challenge (7).
There was no difference for tissue oxygen tension between the fluid treatment groups in healthy colon (P = 0.32), perianastomotic colon (P = 0.70), and intra-anastomotic colon (P = 0.24).
In both groups, tissue oxygen tension 2 cm proximal of the anastomosis was significantly lower than tissue oxygen tension in healthy colon (low fluid: P = 0.02, high fluid: P = 0.03). There was no difference for the healthy-perianastomotic or the healthy-intra-anastomotic tissue oxygen tension gradient between the groups (P = 0.69, 0.77, respectively).
This study demonstrates that supplemental crystalloid fluid therapy does not increase intestinal tissue oxygen tension in healthy, perianastomotic, or intra-anastomotic colon tissue. In contrast, supplemental oxygen markedly increased tissue oxygen tension in healthy, perianastomotic, or intra-anastomotic colon tissue.
Wound infection, anastomotic leakage, and dehiscence are complications of colon surgery associated with considerable morbidity and mortality. Adequate tissue oxygen tension can help prevent these complications (4). Supplemental inspired oxygen can increase subcutaneous tissue oxygen tension and colonic tissue oxygen tension in surgical patients (3,18) and animal models (1,6). Our study extends this observation by showing that supplemental oxygen also improves tissue oxygen tension in injured, perianastomotic, and intra-anastomotic tissue.
As expected from previous study results (6), supplemental oxygen doubled tissue oxygen tension in the healthy gut. In the injured, perianastomotic tissue approximately 2 cm proximal of the anastomosis, tissue oxygen tension increased comparably. In the anastomosis, tissue oxygen tension values were highly variable. Nevertheless, anastomotic tissue oxygen tension increased significantly during ventilation with 100% oxygen. Yet, despite a significant increase during ventilation with 100% oxygen, the intra-anastomotic tissue oxygen tension remained very low. From our experience, we conclude that intra-anastomotic tissue oxygen tension probe placement has to be considered problematic. The large variation in tissue oxygen tension values was likely the result of suture technique and of the varying proximity of suture and the oxygen sensitive probe area. The results of the perianastomotic tissue oxygen tension probe have a narrower range, and may therefore be more reliable as a possible predictor for anastomotic dehiscence.
As tissue oxygen tension in both groups increased by 20% within <5 min at all probe sites after the increase to 100% oxygen (Fig. 1), the requirements for a positive “tissue oxygen challenge” were fulfilled (7). This result indicates adequacy of the overall systemic fluid status and suggests that, even in the low volume group, fluid status was still sufficient.
Over the past few years, several clinical studies in patients undergoing major surgery showed that a restricted fluid regimen reduced postoperative complications (12–14). However, in a study with patients undergoing laparoscopic cholecystectomy, a liberal fluid treatment had beneficial effects for postoperative outcome (11). Similarly, Holte and Kehlet (10) concluded, in a review of 17 trials, that additional preoperative fluid decreased postoperative drowsiness and dizziness. Supplemental fluid has also been shown to increase subcutaneous tissue oxygen tension (9). Obviously, the controversy of liberal versus restricted perioperative fluid therapy continues. In this study, two different crystalloid fluid regimes were tested: a low fluid regime with a fixed rate of 3 mL · kg−1 · h−1 lactated Ringer’s solution corresponding to the restricted patient regimes (12,14) and a high fluid regime with an initial bolus of 10 mL/kg and a rate of 18 mL · kg−1 · h−1 as recommended by textbooks (19) and described in other studies (11,20). In spite of the large differences of crystalloid fluid administered in this study, we found no difference in colonic tissue oxygen tension between the fluid groups.
Surprisingly, the large amounts of crystalloid fluid administered also had no major effect on cardiac output. Only in the last hour of our study could a trend towards a decrease in cardiac output in the low fluid group be detected, which was nonetheless not statistically significant. It is likely that the two fluid regimens we chose were on the lower and upper end of “normovolemia,” especially in young, healthy pigs, which are able to compensate well on both ends of the scale. However, our intent was to measure the impact of restricted versus liberal fluid management on intestinal tissue oxygen tension. We did not titrate fluid to achieve any specific cardiac output or filling pressures. We do not know if administration of less fluid, or more fluid, would have altered tissue oxygenation or more profoundly affected the physiological response (e.g., edema, cardiac output).
Cardiac output is just one of several factors contributing to local tissue oxygen tension. It has been shown that aggressive fluid treatment can lead to increased tissue oxygen tension with no apparent hemodynamic changes (9). Likewise, a decreased tissue oxygen tension due to hypoperfusion is not reliably reflected by a decreased cardiac output (5,21).
In the present study, the healthy colon and perianastomotic tissue oxygen probes were directly inserted into the tissue through a 20-gauge cannulae from the serosal side between serosal and mucosal tissue. In this respect, our technique differed from previous studies that evaluated intestinal oxygen tension with Clark-type multiwire electrodes placed noninvasively on the serosal or mucosal surfaces (22,23). The drawback of these methods is that measurements can be easily confounded by air, fecal contamination, or poor contact between the electrode and the tissue surface during longer measurement periods. Our baseline values for intestinal oxygen partial pressure remained stable throughout the study, and were comparable to those reported previously (9). Although we did not observe edema or hematoma at the insertion sites before probe removal, some microscopic tissue damage due to insertion is inevitable. This damage deliberately mimics a minor surrogate wound, much as subcutaneous tissue oxygen tension measurement in the arm deliberately mimics a surgical incision. It is also important to note that, although the polarographic probe is located subserosally in the wall of the colon, tissue oxygen tension values are nevertheless an average of all three colonic tissue planes due to probe size and measurement technique.
A limitation of the study is that tissue edema in the gut was not evaluated. A liberal approach to fluid therapy has been associated with an increased edema in anastomotic tissue (24) and prolonged recovery of bowel functions (14). However, at all probe sites, both groups had comparable tissue oxygen tensions, which increased similarly under ventilation with 100% oxygen. No macroscopic signs of edema were observed by the surgeon at the end of the experiment. Thus, we considered a significant third space loss in the gut in the high fluid group with resulting severe tissue edema unlikely.
In the present study, pigs were ventilated with 100% oxygen for 1 h. Undoubtedly, ventilation with 100% may lead to atelectasis (25,26). However, the extent of atelectasis formation and surfactant disintegration, and its impact on gas exchange and pulmonary function, remains disputed (27). In our study hemodynamic and respiratory variables remained stable during 1 h of ventilation with 100% oxygen. We assume the deleterious effect of oxygen to be negligible in healthy study animals during this relatively short exposure.
An obvious limitation of our study is that we evaluated pigs rather than humans. However, pig gut approximates the human intestinal system and, consistent with this theory, subcutaneous oxygen tension with 30% oxygen was comparable to values observed in humans, as was the effect of supplemental oxygen (15). Another limitation is the relatively short measurement period of 4 h, as anastomotic leakage usually develops several hours to days after surgery. However, the intraoperative and immediate postoperative phases are likely the decisive periods during which a low tissue oxygen tension has the biggest impact on the development of bacterial infection and wound healing (28).
In conclusion, the administration of supplemental crystalloid fluid did not increase tissue oxygen tension in healthy, perianastomotic, or intra-anastomotic colon tissue. In contrast, supplemental oxygen increased tissue oxygen tension in healthy and in injured perianastomotic, and intra-anastomotic colon tissue.
1. Shandall A, Lowndes R, Young HL. Colonic anastomotic healing and oxygen tension. Br J Surg 1985;72:606–9
2. Moran B, Heald R. Anastomotic leakage after colorectal anastomosis. Semin Surg Oncol 2000;18:244–8
3. 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 2000;342:161–7
4. Hopf HW, Hunt TK, West JM, Blomquist P, Goodson WH III, Jensen JA, Jonsson K, Paty PB, Rabkin JM, Upton RA, von Smitten K, Whitney JD. Wound tissue oxygen tension predicts the risk of wound infection in surgical patients. Arch Surg 1997;132:997–1004
5. Chang N, Goodson WH III, Gottrup F, Hunt TK. Direct measurement of wound and tissue oxygen tension in postoperative patients. Ann Surg 1983;197:470–8
6. Ratnaraj J, Kabon B, Talcott MR, Sessler DI, Kurz A. Supplemental oxygen and carbon dioxide each increase subcutaneous and intestinal intramural oxygenation. Anesth Analg 2004;99:207–11
7. Waxman K, Annas C, Daughters K, Tominaga GT, Scannell G. A method to determine the adequacy of resuscitation using tissue oxygen monitoring. J Trauma 1994;36:852–6
8. Yu M, Morita SY, Daniel SR, Chapital A, Waxman K, Severino R. Transcutaneous pressure of oxygen: a noninvasive and early detector of peripheral shock and outcome. Shock 2006;26:450–6
9. Arkilic CF, Taguchi A, Sharma N, Ratnaraj J, Sessler DI, Read TE, Fleshman JW, Kurz A. Supplemental perioperative fluid administration increases tissue oxygen pressure. Surgery 2003;133:49–55
10. Holte K, Kehlet H. Compensatory fluid administration for preoperative dehydration—does it improve outcome? Acta Anaesthesiol Scand 2002;46:1089–93
11. Holte K, Klarskov B, Christensen DS, Lund C, Nielsen KG, Bie P, Kehlet H. Liberal versus restrictive fluid administration to improve recovery after laparoscopic cholecystectomy: a randomized, double-blind study. Ann Surg 2004;240:892–9
12. Brandstrup B, Tonnesen H, Beier-Holgersen R, Hjortso E, Ording H, Lindorff-Larsen K, Rasmussen MS, Lanng C, Wallin L, Iversen LH, Gramkow CS, Okholm M, Blemmer T, Svendsen PE, Rottensten HH, Thage B, Riis J, Jeppesen IS, Teilum D, Christensen AM, Graungaard B, Pott F. Effects of intravenous fluid restriction on postoperative complications: comparison of two perioperative fluid regimens: a randomized assessor-blinded multicenter trial. Ann Surg 2003;238:641–8
13. Lobo DN, Bostock KA, Neal KR, Perkins AC, Rowlands BJ, Allison SP. Effect of salt and water balance on recovery of gastrointestinal function after elective colonic resection: a randomised controlled trial. Lancet 2002;359:1812–8
14. Nisanevich V, Felsenstein I, Almogy G, Weissman C, Einav S, Matot I. Effect of intraoperative fluid management on outcome after intraabdominal surgery. Anesthesiology 2005;103:25–32
15. Fleischmann E, Herbst F, Kugener A, Kabon B, Niedermayr M, Sessler DI, Kurz A. Mild hypercapnia increases subcutaneous and colonic oxygen tension in patients given 80% inspired oxygen during abdominal surgery. Anesthesiology 2006;104:944–9
16. Sheridan WG, Lowndes RH, Young HL. Intraoperative tissue oximetry in the human gastrointestinal tract. Am J Surg 1990;159:314–9
17. Hoelper BM, Alessandri B, Heimann A, Behr R, Kempski O. Brain oxygen monitoring: in-vitro accuracy, long-term drift and response-time of Licox- and Neurotrend sensors. Acta Neurochir (Wien) 2005;147:767–74
18. Sheridan WG, Lowndes RH, Young HL. Tissue oxygen tension as a predictor of colonic anastomotic healing. Dis Colon Rectum 1987;30:867–71
19. Kaye AD. Fluid and electrolyte physiology. In: Miller’s anesthesia. Philadelphia: Churchill Livingstone, 2005:1763–99
20. Campbell IT, Baxter JN, Tweedie IE, Taylor GT, Keens SJ. IV fluids during surgery. Br J Anaesth 1990;65:726–9
21. Gosain A, Rabkin J, Reymond JP, Jensen JA, Hunt TK, Upton RA. Tissue oxygen tension and other indicators of blood loss or organ perfusion during graded hemorrhage. Surgery 1991;109:523–32
22. Germann R, Haisjackl M, Schwarz B, Salak N, Deusch E, Pajk W, Wolf HJ, Riedmann B, Hasibeder W. Dopamine and intestinal mucosal tissue oxygenation in a porcine model of haemorrhage. Br J Anaesth 1997;79:357–62
23. Sheridan WG, Lowndes RH, Williams GT, Young HL. Determination of a critical level of tissue oxygenation in acute intestinal ischaemia. Gut 1992;33:762–6
24. Chan ST, Kapadia CR, Johnson AW, Radcliffe AG, Dudley HA. Extracellular fluid volume expansion and third space sequestration at the site of small bowel anastomoses. Br J Surg 1983;70:36–9
25. Reber A, Englberg G, Wegenius G, Hedenstierna G. Lung aeration. The effect of pre-oxygenation and hyperoxygenation during total intravenous anaesthesia. Anaesthesia 1996;51:733–7
26. Rothen HU, Sporre B, Englberg G, Wegenius G, Hogman M, Hedenstierna G. Influence of gas composition on recurrence of atelectasis after a reexpansion maneuver during general anesthesia. Anesthesiology 1995;82:832–42
27. Akca O, Podolsky A, Eisenhuber E, Panzer O, Hetz H, Lampl K, Lackner FX, Wittmann K, Grabenwoeger F, Kurz A, Schultz AM, Negishi C, Sessler DI. Comparable postoperative pulmonary atelectasis in patients given 30% or 80% oxygen during and 2 hours after colon resection. Anesthesiology 1999;91:991–8
28. Miles AA, Miles EM, Burke J. The value und duration of defense reactions of the skin to the primary lodgement of bacteria. Br J Exp Pathol 1957;38:79–96