The effect of intra-abdominal pressure on the generation of 8-iso prostaglandin F2{alpha} during laparoscopy in rabbits

Ângela M.Bentes de Souza, Chi Chiu Wang, Ching Yan Chu, Po Mui Lam and Michael Scott Rogers1

Department of Obstetrics & Gynaecology, The Chinese University of Hong Kong, Hong Kong SAR

1 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, 1/EF Prince of Wales Hospital, Shatin–New Territories, Hong Kong SAR. e-mail: msrogers{at}cuhk.edu.hk


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: Carbon dioxide pneumoperitoneum induces peritoneal oxidative stress. The aim of this study was to verify the effect of intra-abdominal pressure on oxidative stress in the peritoneum and on post-operative adhesion formation. METHODS: Forty-one rabbits underwent laparoscopic surgery: either gasless, or with CO2-pneumoperitoneum at pressures of 5, 10 or 15 mmHg. Serial parietal peritoneal biopsies were taken at various time-points: immediately after reaching the abdominal cavity, 30, 60, 90 and 120 min afterwards, and 15 min after abdominal desufflation. 8-iso prostaglandin F2{alpha} (8-iso PGF2{alpha}), a marker of oxidative stresss, was assayed by enzyme immunoassay and adhesion formation was scored by second-look laparoscopy on day 14. RESULTS: The gasless group showed no significant changes in 8-iso PGF2{alpha}. Conversely, significant changes occurred in CO2-pneumoperitoneum in a time- and pressure-dependent manner. Adhesions developed only in the CO2-pneumoperitoneum groups, and total adhesion score was correlated with the amount of CO2 insufflated and intra-abdominal pressure, but not with 8-iso PGF2{alpha}, which was correlated with intra-abdominal pressure. CONCLUSION: Intra-abdominal pressure increased 8-iso PGF2{alpha} in the parietal peritoneum in a graded fashion, whilst gasless laparoscopy had no impact. It also influenced the frequency and severity of adhesion formation, but no causal link was found between 8-iso PGF2{alpha} and post-operative adhesion formation.

Key words: adhesion formation/intra-abdominal pressure/ischaemia/isoprostanes/laparoscopy


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Laparoscopic surgery represents an attractive approach that offers many advantages when compared with open surgery. However, laparoscopy has peculiar effects as it is usually performed in a different environment, the pneumoperitoneum. Carbon dioxide (CO2) is the most common gas used to create the pneumoperitoneum during laparoscopy, and the standards set for its use as an insufflation agent require 99% purity with relative humidity of 0.0002%. This environment produces intricate effects in the abdominal cavity that primarily depend on the intra-abdominal pressure and the duration of application. Locally, CO2-pneumoperitoneum induces a drop in the abdominal cavity pH with transient impairment of the release of inflammatory mediators by macrophages in peritoneal fluid (West et al., 1997Go); it dries the tissue surfaces causing loss of integrity, cellular compromise, and exposure of the sub-mesothelial layer (Volz et al., 1999Go); and it provokes haemodynamic changes in the abdominal viscera (Schilling et al., 1997Go).

The haemodynamic changes result from mechanical distension of the abdominal cavity and from liberation of systemically active vasoconstrictor substances (Viinamaki and Punnonen, 1982Go). Intra-abdominal pressure affects the splanchnic macro- and microcirculation in a graded fashion. It has a direct influence on abdominal vessels and visceral organs, as well as on the abdominal wall surfaces. In the venous system, the increased intra-abdominal pressure compresses the splanchnic veins, reducing the blood flow by elevating vascular resistance. Compression of the portal vein, which represents the major outflow tract of visceral organs, leads to blood flow stasis in the splanchnic circulation. This stasis impairs intestinal perfusion at the mucosal and submucosal layers, leading to a reduction in tissue oxygen tension, anaerobic cell metabolism, acidosis, and production of reactive oxygen species (ROS) (Eleftheriadis et al., 1996Go). Overproduction of ROS related to CO2-pneumoperitoneum has been addressed in few experimental and clinical trials with conflicting results. Some studies have found no difference in the plasma oxidative status between the laparoscopic and laparotomy approaches (Odeberg et al., 1998Go; Ozmen et al., 2002Go), whereas others have demonstrated local (Bentes de Souza et al., 2003Go) and systemic (Glantzounis et al., 2001Go) oxidative stress induced by CO2-pneumoperitoneum. The clinical significance of this peritoneal oxidative stress remains unclear, but there is some evidence to suggest that oxidative stress plays a role in post-operative adhesion formation (Taskin et al., 1999Go).

Direct measurement of ROS generation in vivo has proven problematic. The conventional approach has been to measure products of free radical activity. 8-iso prostaglandin F2{alpha} (8-iso PGF2{alpha}) is a member of a complex family of compounds produced in situ in cell membranes by direct free-radical action on arachidonic acid. It can be quantified as a reliable pathophysiological marker of free radical activity for investigation of oxidative stress in human diseases. 8-iso PGF2{alpha} is found in relatively large concentrations in blood and urine (10-fold greater than the cyclo-oxygenase-derived prostaglandins), and is also ‘stored’ as part of the cell membrane (Morrow et al., 1992Go). In addition this compound is biologically active: 8-iso PGF2{alpha} is a potent vasoconstrictor (John and Valentin, 1997Go) and a mitogen in several cell types, including vascular smooth cells (Yura et al., 1999Go).

The objective of this study was to investigate, in anaesthetized adult rabbits, the temporal changes of 8-iso PGF2{alpha} in the peritoneum during laparoscopy using different intra-abdominal pressures. We hypothesized that peritoneal oxidative stress results from the elevation of intra-abdominal pressure, and may be associated with post-operative adhesion formation.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Animals
Forty-one female New Zealand white rabbits weighing between 3.0 and 4.0 kg were used. The animals were housed in the facilities of the Laboratory Animal Services Centre (LASEC) of the Chinese University of Hong Kong. They were kept at temperatures between 20 and 25°C, with relative humidity between 40 and 70%, and a 12 h light–dark cycle. The rabbits had free access to water and standard rodent laboratory food before and after the surgical procedures. The Animal Research Ethics Committee of The Chinese University of Hong Kong approved the study.

Anaesthesia
Anaesthesia was induced with an i.m. injection of ketamine (50 mg/kg, Kelatin®; Alfasan, The Netherlands) and xylazine (6 mg/kg, XYL-M 2%®; Alfasan). A 3.5 mm endotracheal cannula was inserted into the trachea under direct visualization of vocal cords using a 2.9 mm hysteroscope attached to a video-monitor system. The anaesthesia was maintained with 2.0% isofluorane (Forane®; Abbott, UK) and 2 l/min of oxygen–room air mixture (80:20), with spontaneous ventilation. Pulse rate and oxygen saturation (SaO2) were monitored continuously throughout the experiment using an oxymeter sensor (Masimo, USA) attached to an ear vessel. After stabilization of anaesthesia the animal was placed in the supine position with the limbs secured to the table; the abdomen was shaved and disinfected with iodine solution (Betadine®; Mundipharma, Switzerland); and the animal was wrapped with bubble-plastic to prevent loss of body heat. The surgery was performed under strict aseptic conditions and no perioperative antibiotics were administered.

Surgical technique
Animals were randomly allocated to one of the experimental groups: group A (n = 10) gasless laparoscopy; group B (n = 10) 5 mmHg intra-abdominal pressure; group C (n = 10) 10 mmHg intra-abdominal pressure; and group D (n = 10) 15 mmHg intra-abdominal pressure.

Pneumoperitoneum was achieved using a Veress needle placed caudal to the sternum through a stab wound. In groups B, C and D the intra-abdominal pressure was initially kept at 3 mmHg until the first biopsy (T0) had been collected. Thereafter the pressure was raised according to the designated group. This was performed to minimize any influence of intra-abdominal pressure on the baseline values. Intra-abdominal pressure was maintained with a laparoflator (Striker® Endoscopy, USA) using non-pre-warmed carbon dioxide as the insufflation agent. Three puncture sites were used: a 10 mm trochar placed caudal to the sternum, and two 5 mm trochars; one in the right and one in the left flanks. A continuous flow was used to maintain a 100% intra-abdominal concentration of CO2. This was achieved by placing a 16 gauge catheter through the abdominal wall. The escape flow from the abdomen varied with the insufflation pressure. To minimize the differences in amount of CO2 used between the groups, a three way tap connected to the 16 gauge catheter was used to adjust and maintain a flow rate of ~1 l/min through the animal: the amount of CO2 insufflated by the laparoflator was checked every 5 min. In the gasless laparoscopy group (control), ports were placed by open technique and the abdominal wall was suspended using s.c. sutures. No pneumoperitoneum was used in this group.

The surgical procedures each lasted for 135 min, and six 1x1 cm squares of peritoneum were excised throughout the experiment (Figure 1B and C). Time-points and sites were defined as follows: T0, immediately after reaching the abdominal cavity: right lower quadrant; T1, 30 min: left lower quadrant; T2, 60 min: right flank; T3, 90 min: left flank; T4, 120 min: right hypochondrium; and T5, 15 min after deflation of pneumoperitoneum: left hypochondrium.



View larger version (72K):
[in this window]
[in a new window]
 
Figure 1. (A) Changes in pulse rate, CO2 amount and SaO2 throughout the study period. Pulse rate increased in the 10 and 15 mmHg groups at T1–T5 (P < 0.05). CO2 amount differed among the groups from T2 to T5 (P < 0.05), mainly due to the 15 mmHg group. Values are mean ± SD. *P < 0.05. (B) The rabbit was placed in supine position under general anaesthesia. Six 1x1 cm squares of peritoneum were excised from the abdominal wall as follows: T0, immediately after reaching the abdominal cavity: right lower quadrant; T1, 30 min: left lower quadrant; T2, 60 min: right flank; T3, 90 min: left flank; T4, 120 min: right hypochondrium; and T5, 15 min after deflation of pneumoperitoneum: left hypochondrium. (C) Intra-abdominal view of the peritoneal excision sites.

 
After T4 (120 min) the pneumoperitoneum was totally deflated for 15 min to allow reperfusion of the peritoneum. A 3 mmHg pneumoperitoneum was re-established for collection of the final peritoneal biopsy (T5). The peritoneum samples were snap-frozen in liquid nitrogen immediately after collection and stored at –80°C until being used for 8-iso PGF2{alpha} assay.

All surgery was carried out by the same surgeon, and biopsies were performed in a standard way. Briefly, the parietal peritoneum was grasped with an atraumatic forceps and a small cut was performed followed by blunt dissection (with scissor) of the peritoneum. Thereafter the tissue was excised and removed through 5 mm trochar. Mild bleeding occasionally occurred, but stopped spontaneously in all cases. None of the excisions led to emphysema. No electrical current was used during the experiments, and besides peritoneal biopsy, no other surgical procedures were performed.

At the end of the procedure, the incisions were closed with 4–0 polyglycolic acid sutures (Dexon II; Davis & Geck, UK) in double layers. The animals were allowed to recover from anaesthesia, and 0.04 mg/kg buprenorphine (Temgesic®; Schering-Plough, UK) was administered by s.c. injection 12 hourly (x4 doses) for post-surgical analgesia.

Second-look laparoscopy and assessment of post-operative adhesions
Fourteen days after the initial procedure, a second-look laparoscopy was performed to score adhesion formation. All surgeries were videotaped, and subsequently evaluated blindly by another surgeon. Following the procedure, the animals were euthanased using 150 mg/kg pentobarbital sodium i.v. injection (Dorminal 20%; Alfasan).

Adhesions were evaluated using a scoring system previously described (Molinas and Koninckx, 2000Go) that took into account: type (1 = filmy adhesions, transparent and avascular; 2 = dense and avascular; 3 = dense, capillaries present; 4 = dense, larger vessels present); tenacity (1 = fall apart; 2 = require traction; 3 = require sharp dissection); and extent of adhesions (1 = 1–25%; 2 = 26–50%; 3 = 51–75%; 4 = 76–100%).

All animals were used to calculate the frequency of adhesions per group. Regarding the adhesion score, each site was scored individually, and the animal score represented the sum of the six score sites divided by 6. Total adhesion score was the sum of type, tenacity and extent of lesions. All animals (with and without adhesions) were included in the score group.

In order to avoid any confusion in terminology, this study used the following terms to classify adhesion development (Diamond and Nezhat, 1993Go): type 1: de-novo adhesion formation; development of adhesions at sites that did not have adhesion initially; type 1A: no operative procedure performed at site of adhesion formation; type 1B: operative procedure at site of adhesion formation; type 2: adhesion reformation, redevelopment of adhesions at sites at which adhesiolysis was performed; type 2A: no operative procedure at site of adhesion reformation, besides adhesiolysis; type 2B: operative procedure performed at site of adhesion reformation (in addition to adhesiolysis).

Oxidative metabolite measurement
Total (free and esterified) 8-iso PGF2{alpha} was assayed by enzyme-linked immunosorbent assay as previously described (Bentes de Souza et al., 2003Go). Briefly, each tissue sample was homogenized with 100% ethanol containing 0.05% butylated hydroxytoluene (BHT) using a polytron type grind (Wheaton, USA), and then centrifuged at 3000 g for 10 min. The supernatant was decanted into a clean tube and 15% potassium hydroxide was added. An aliquot of 350 µl was used for phospholipid determination according to a method described by others (Mrsny et al., 1986Go). Samples were diluted with Ultra pure water pH 3, eluted through Sep-Pak C18 and Sep-Pak Silica cartridges (Waters Associates, USA). The elution was developed in an enzyme immunoassay kit (Cayman Chemical, USA) and analysed on a spectrophotometric microplate reader (Spectramax; Bio-Tek Instruments, USA) at 412 nm wavelength. Cross-reactivity for the enzyme immunoassay was 100% with 8-iso PGF2{alpha}, 20% with 8-iso PGF3{alpha}, and <1% with other prostaglandin metabolites. In this study, [3H]8-iso PGF was used as internal standard for method validation and calibration. 8-iso PGF2{alpha} concentration in the peritoneum was presented as pg/µg phospholipid.

Data analyses
Statistical analyses were performed using SPSS, version 10.1 for Windows. Within-group differences were evaluated by Wilcoxon signed ranks test, and differences between group A and other groups (B, C and D) were evaluated by {chi}2, Kruskal–Wallis and Mann–Whitney U-tests. Spearman’s correlation coefficients were reported where appropriate. P < 0.05 was considered statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
All animals tolerated the experiments well, except one from group D (15 mmHg) that died after 60 min of CO2-pneumoperitoneum. The death was attributable to abdominal wall emphysema caused by incorrect placement of a 5 mm trochar. This animal was excluded from the study. A replacement animal was then used (no. 41), but was unfortunately allocated to the 10 mmHg group (rather than the 15 mmHg group) by mistake. Statistical analyses were performed with the following no. of animals/group: gasless, n = 10; 5 mmHg, n = 10; 10 mmHg, n = 11; and 15 mmHg, n = 9.

Similar changes in peripheral SaO2 were observed in all four groups. However, significant rise in the mean pulse rate from baseline values was observed between the groups at T1–T5 time-points (Kruskal–Wallis test, P < 0.05). Post-hoc test showed that it was due to the 10 and 15 mmHg groups (Mann–Whitney U-test, P < 0.05). Despite attempts to standardize the amount of CO2 between groups by maintaining a flow rate of 1 l/min, they differed significantly at T2–T5 time-points (Kruskal–Wallis test, P < 0.05). The difference was due to higher amount of CO2 in the 15 mmHg group (Mann–Whitney U-test, P < 0.05). Figure 1A summarizes the changes of SaO2, pulse rate and amount of CO2 throughout the experiments.

Comparison of peritoneal levels of 8-iso PGF2{alpha} between and within the groups
A significant difference in 8-iso PGF2{alpha} baseline values (T0) was detected between the four study groups (Kruskal–Wallis Test, P < 0.05). It should be noted that there was little difference between the means of the gasless, 10 and 15 mmHg groups and, therefore, the difference was due to the 5 mmHg group (Mann–Whitney U-test, P < 0.05). In order to overcome the dissimilarity in the baseline values, within-group changes were examined ({Delta}T = Tn – T0).

The levels of 8-iso PGF2{alpha} in group A (gasless) did not change significantly throughout the 135 min study period. Conversely, significant rises in 8-iso PGF2{alpha} occurred in the CO2-pneumoperitoneum groups: in group B (5 mmHg), a rise was only observed after release of the pneumoperitoneum (T1 to T4, P > 0.05; T5, P < 0.05); in group C (10 mmHg) at 30 min (T1, P < 0.01) and 60 min (T2, P < 0.05), but did not reach significance at the other time-points (T3 to T5, P > 0.05); whereas in group D (15 mmHg), all time-points (T1, T2, T4 and T5, P < 0.05; T3, P < 0.01) showed significant increase of 8-iso PGF2{alpha} in the peritoneal biopsies compared with baseline values (T0). Table I shows the concentration of 8-iso PGF2{alpha} in the peritoneum among the four study groups.


View this table:
[in this window]
[in a new window]
 
Table I. Concentrations of 8-iso PGF2{alpha} (pg/µg phospholipids) in peritoneal tissue taken from abdominal wall at different time-points, and under different intra-abdominal pressures
 
When the 8-iso PGF2{alpha} trends were examined, the 10 and 15 mmHg groups showed similar rises up to 60 min; thereafter further increase occurred in the 15 mmHg group, reaching a peak level at 90 min. In the 5 mmHg group, 8-iso PGF2{alpha} decreased in the initial 30 min of pneumoperitoneum; thereafter a progressive increase occurred reaching a peak level in the final biopsies taken 15 min after abdominal deflation. It is noted that abdominal deflation contributed to further generation of 8-iso PGF2{alpha} in both the 5 and 10 mmHg groups, but the effect on the 15 mmHg group was less pronounced. Figure 2 illustrates the trends of the 8-iso PGF2{alpha} changes.



View larger version (24K):
[in this window]
[in a new window]
 
Figure 2. Changes of 8-iso PGF2{alpha} concentrations (pg/µg phospholipids) in peritoneal tissue among the four experimental groups. Values represent mean ± 0.5 SD. *P < 0.05 compared with gasless group. Intra-abdominal pressure influenced the levels of 8-iso PGF2{alpha} in a time- and pressure-dependent manner. In the high pressure groups (10 and 15 mmHg) the rise occurred during abdominal insufflation, whereas in the low pressure group (5 mmHg) it happened only after pneumoperitoneum release.

 
The net gain of 8-iso PGF2{alpha} was used to verify the differences between groups. The reasons for this were that (i) the change of 8-iso PGF2{alpha} in the peritoneum was not linear, and (ii) there was a significant difference at T0 between the four groups, therefore direct comparison of 8-iso PGF2{alpha} values could not be performed. Net gain of 8-iso PGF2{alpha} in each subject was estimated using the area under the curve created by the differences in 8-iso PGF2{alpha} levels over the whole experiment. Statistical analyses showed a significant difference in net gain of 8-iso PGF2{alpha} (Kruskal–Wallis test, P < 0.01, mean ± SD) among the four experimental groups (gasless = –1.78 ± 101, 5 mmHg = 2.89 ± 28.94, 10 mmHg = 33.7 ± 42.62, and 15 mmHg = 55.28 ± 23.24 pg/µg phospholipid). The difference was predominantly due to the 15 mmHg when compared with 5 mmHg (Mann–Whitney U-test, P < 0.05). Figure 3 illustrates these differences.



View larger version (12K):
[in this window]
[in a new window]
 
Figure 3. Effect of intra-abdominal pressure on net gain of 8-iso PGF2{alpha}. Error bars represent mean ± SD. Significant difference was observed between 5 and 15 mmHg groups.

 
Post-operative adhesion formation
No rabbits had peritoneal adhesions prior to the initial procedure, and none of the groups developed de-novo adhesion formation away from the biopsy sites (type 1A). The frequency of adhesions (type 1B) varied significantly between the groups ({chi}2-test, P < 0.05, Figure 4). The difference was predominantly due to the 10 and 15 mmHg groups when compared with the gasless group (P < 0.05), but not to the 5 mmHg group (P > 0.05). In the gasless group, no post-operative de novo adhesion formation (type 1B) was observed at any of the peritoneal biopsy sites. In the CO2-pneumoperitoneum groups (5, 10 and 15 mmHg) most of the animals (20 out of 30) did not develop adhesions (type 1B). In those in which it occurred (10 out of 30), adhesions were more frequent (79%) on the left side of the abdominal wall. No significant difference in total adhesion score was found among the four groups (Kruskal–Wallis test, P = 0.094, mean ± SD; gasless = 0, 5 mmHg = 0.1 ± 0.21, 10 mmHg = 0.5 ± 0.75, 15 mmHg = 0.6 ± 0.95).



View larger version (16K):
[in this window]
[in a new window]
 
Figure 4. Effect of intra-abdominal pressure on the frequency and severity of adhesion formation. Adhesions were scored at the sites of peritoneal excisions 14 days after the initial procedure. Total adhesion score of each animal was calculated by summing up the total score of the six sites and dividing by six. All animals were used to calculate the total adhesion score of the group. Comparison of the frequency of adhesions showed significant differences between gasless and 10 and 15 mmHg (*P < 0.05). However, no significant difference in total adhesion was found between the four experimental groups. Values are mean ± SD.

 
When the relationship between peritoneal oxidative levels, intra-abdominal pressure, total adhesion score, and amount of CO2 was examined, net gain of 8-iso PGF2{alpha} was correlated with intra-abdominal pressure (r = 0.452, P = 0.004), but not with total amount of CO2 insufflated (r = 0.207, P = 0.20) or with total adhesion score (r = 0.232, P = 0.15). Total adhesion score was correlated with total amount of CO2 insufflated (r = 0.326, P = 0.04) and intra-abdominal pressure (r = 0.399, P = 0.011). Total amount of CO2 insufflated was also correlated with intra-abdominal pressure (r = 0.763, P = 0.001). All animals from the four groups were used to calculate these correlations.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Oxidative stress has been implicated in the pathogenesis of many disorders including adhesion formation. Different compounds with antioxidant properties have been used to decrease post-operative adhesion formation with encouraging results, confirming a role for ROS in the healing process (Hemadeh et al., 1993Go; Galili et al., 1998Go). Some sources of oxidative stress during surgery have been identified, such as laparotomy with mild intestinal handling (Anup et al., 1999Go), severe ischaemic conditions (Tsimoyiannis et al., 1989Go) and activation of inflammatory cells (Shimanuki et al., 1986Go). We previously demonstrated that conditions dictated by the surgical route also influence the production of ROS: peritoneal levels of 8-iso PGF2{alpha} rose significantly during laparoscopy and seemed to be due to effects of CO2-pneumoperitoneum, whereas no changes in peritoneal levels of 8-iso PGF2{alpha} were observed during open surgery (Bentes de Souza et al., 2003Go).

The pathophysiology involved in the overproduction of ROS during laparoscopy seems to be related to elevated intra-abdominal pressure which induces changes in the splanchnic macro- and microcirculation with subsequent tissue ischaemia. Previous studies have investigated the changes in the splanchnic microcirculation using different models and intra-abdominal pressures. In humans, an intra-abdominal pressure of 15 mmHg decreased the blood flow in all abdominal organs, most notably in the parietal peritoneum and intra-peritoneal organs, whereas no changes in the parietal peritoneal blood flow were observed with pressures of 10 mmHg (Schilling et al., 1997Go). Research animals tend to be much smaller than humans, and are therefore more sensitive to lower (raised) intra-abdominal pressures. A study using pigs showed that 10 mmHg intra-abdominal pressure resulted in a 42% decrease in the rectus sheath blood flow (Diebel et al., 1992bGo), but did not significantly change the intestinal mucosal blood flow (Diebel et al., 1992aGo). Another study using a different model (rats) demonstrated that intra-abdominal pressures of 10 and 15 mmHg caused a significant stepwise decrease in jejunal mucosa perfusion index (29 and 78%, respectively) resulting in a severe impairment of mucosal microcirculation (Samel et al., 2002Go).

The present study made use of intra-abdominal pressures of 5, 10 and 15 mmHg to determine the threshold pressure that induced oxidative stress in the peritoneum of rabbits. The results showed a significant difference in net gain of 8-iso PGF2{alpha} among the four experimental groups, with higher intra-abdominal pressures yielding higher levels of 8-iso PGF2{alpha} (P < 0.01). When the temporal changes of 8-iso PGF2{alpha} levels were examined, the gasless (A) group showed no significant changes throughout the 135 min study period. Conversely, significant changes occurred in CO2-pneumoperitoneum (B, C and D) groups at different time-points. In the high pressure groups (C and D), increase in 8-iso PFG2{alpha} occurred during the abdominal insufflation, whereas in the low pressure group (B) it happened only after pneumoperitoneum release. Furthermore, at high pressures these changes in 8-iso PGF2{alpha} were not linear. The reasons for these findings are not clear, but we speculate that acute reduction of peritoneal blood flow plays a key role in the process. Raising intra-abdominal pressure leads to ischaemia in the peritoneum in a graded fashion (Diebel et al., 1992bGo). Ischaemia causes a series of disarrangements in the cell metabolism predisposing to the formation of ROS (Ferrari et al., 1998Go). One of these changes is inhibition of aerobic ATP synthesis, which leaves the mitochondrial carriers (ubiquinone) in a more reduced state (Freeman and Crapo, 1982Go). This condition results in a higher increase in electron leakage from the respiratory chain that, in turn, reacts with the residual molecular oxygen trapped within the inner mitochondrial membrane, leading to the formation of superoxide radicals. Re-introduction of oxygen with reperfusion will re-energize the mitochondria, but electron egress through cytochrome oxidase will be reduced because of the lack of ADP. As a result, the percentage of electron leakage will further increase, enabling more reactions with molecular oxygen. The electron leakage with subsequent formation of superoxide radicals during ischaemia may explain the rise in 8-iso PGF2{alpha} observed in our study during abdominal insufflation. Regarding the non-linear increase in 8-iso PGF2{alpha} at high pneumoperitoneum pressures (groups C and D) we can only speculate that acute ischaemia caused an overproduction of ROS which overwhelmed the tissue antioxidant system. However, after the initial ‘unexpected assault’, the tissue antioxidant defences were gradually rebuilt and started scavenging the high levels of ROS. It may even be possible that with longer duration of pneumoperitoneum, the levels of 8-iso PGF2{alpha} would have progressively decreased if no further insult was applied. Future studies monitoring the genes and proteins involved in the tissue antioxidant defense system over longer observational periods will help to elucidate the true situation.

The lack of any changes in 8-iso PGF2{alpha} during 5 mmHg pneumoperitoneum (T1 to T4) imply that this pressure is relatively harmless to the tissue (i.e. only inducing ROS within the antioxidant capacity of the tissue), and furthermore that CO2 itself is not a cause of oxidative stress. However, decompression of the pneumoperitoneum led to a significant rise in 8-iso PGF2{alpha} in the 5 mmHg and the 10 mmHg groups. This suggests that a second element, that of reperfusion, contributes to ROS overproduction in laparoscopy. Laparoscopic surgery under conditions of pneumoperitoneum must therefore be considered a hypoxia–reperfusion model, where pneumoperitoneum induces low perfusion and relative hypoxia in the abdominal organs, and its release re-establishes splanchnic blood flow and tissue oxygenation. In the present study, elevated intra-abdominal pressure is the most likely cause of hypoxia in the 10 and 15 mmHg groups. However, it is uncertain whether that is the case in the 5 mmHg group, since there is lack of experimental data demonstrating that such low intra-abdominal pressure impairs the splanchnic microcirculation. We speculate that CO2 may be the hypoxic factor in this group. Normally, the extracellular oxygen partial pressure (pO2) is 40 mmHg and the intracellular pO2 varies from 5 to 40 mmHg (average, 23 mmHg), which is adequate to fully support the intracellular metabolic processes (Guyton, 2000Go). During laparoscopy 99% CO2 (United States Pharmacopeia, 1984Go) is insufflated, which translates into a pCO2 close to 760 mmHg and a pO2 close to zero. Consequently, oxygen will tend to diffuse from the extracellular space into the abdominal cavity. This will result in a decrease in extracellular and intracellular pO2. If there is no disturbance in the microcirculation, oxygen supply to the cells is maintained by the blood flow. However, even with an intra-abdominal pressure of 5 mmHg, some disturbance to the microcirculation is inevitable, and the oxygen supply may be compromised. This mild hypoxia followed by reperfusion would explain the significant rise in 8-iso PGF2{alpha} after abdominal deflation in the 5 mmHg group.

The second outcome of this study was adhesion formation, which was scored 14 days after the initial procedure. The results showed that in the gasless group no adhesions formed at the peritoneal biopsy sites, whereas adhesion formation occurred in all three CO2-pneumoperitoneum groups. This suggests that peritoneal biopsy itself was not adhesiogenic, but conditions dictated by CO2-pneumoperitoneum interfered in the normal peritoneal healing process in some cases. Indeed, higher frequency of adhesions was observed in the groups with higher intra-abdominal pressure (P < 0.05) and a positive correlation was found between adhesion score and amount of insufflated CO2 (P < 0.05). It is of note that the experiments were performed with a non-humidified CO2, therefore differences in amount of CO2 also mean differences in desiccation, which may have contributed to adhesion formation. Our findings are in agreement with previous studies that reported intra-abdominal pressure and amount of insufflated gas were positively correlated with the frequency and severity of adhesion formation (Yesildaglar and Koninckx, 2000Go).

One question remains unanswered: what is the clinical significance of peritoneal oxidative stress on the peritoneal healing process? We initially hypothesized that overproduction of ROS due to elevated intra-abdominal pressure might be associated with post-operative adhesion formation. Our findings do not support this hypothesis, as no correlation was found between adhesion formation and net gain of 8-iso PGF2{alpha}. Two possibilities may explain this absence of correlation: (i) oxidative stress induced by elevated intra-abdominal pressure causes reversible damage in the peritoneum, and (ii) peritoneal biopsy is not adhesiogenic enough to explore this association. The former seems to be the most likely one as 75% of the animals did not develop adhesions, even in the 15 mmHg group, and in the 25% remaining, adhesions occurred in the cases exposed to a higher amount of insufflated CO2. The latter is still possible, and further experiments with a more appropriate experimental design need to be done to clarify this issue.

When we interpret the data provided by this study there are a few methodological concerns. Animals were spontaneously ventilated. Transperitoneal absorption of CO2 may cause hypercapnia and acidosis, especially when associated with elevated intra-abdominal pressure. Both hypercapnia and acidosis may produce vasoconstriction, thus increasing vascular resistance and hypoperfusion. In this study, no significant changes in pO2 were observed, but significant variations in pulse rate occurred at higher intra-abdominal pressures (10 and 15 mmHg). Since no estimation of pCO2 was recorded throughout the experiments, we cannot rule out hypercapnia being responsible for the pulse rate variations, or that hypercapnia may have influenced the rise of 8-iso PGF2{alpha} induced by elevated intra-abdominal pressure.

There is evidence from an in vitro study (Kang et al., 1998Go) that buprenorphine has antioxidant properties. In the present study, buprenorphine was administrated post-operatively and therefore did not affect the peritoneal levels of 8-iso PGF2{alpha}. Nevertheless, it may have influenced the frequency and severity of post-operative adhesion formation. Buprenorphine was used in all experimental groups and therefore should not influence inter-group comparisons, but it may have influenced the correlation between 8-iso PGF2{alpha} and post-operative adhesion formation.

In conclusion, this study suggests that intra-abdominal pressure plays a role in the production of ROS during laparoscopy. Elevation of intra-abdominal pressure increased 8-iso PGF2{alpha} in the parietal peritoneum in a graded fashion, whilst gasless laparoscopy had no impact. The temporal rise of oxidative stress was influenced by the levels of intra-abdominal pressure: 10 and 15 mmHg significantly increased 8-iso PGF2{alpha} after 30 min, whereas with 5 mmHg it occurred only after pneumoperitoneum release. Intra-abdominal pressure also had an effect on post-operative adhesion formation, a higher frequency of adhesions was observed in the groups with higher intra-abdominal pressures, and total adhesion score was correlated with the amount of CO2 insufflated and intra-abdominal pressure. However, no correlation was found between 8-iso PGF2{alpha} and post-operative adhesion formation: the clinical implications of oxidative stress due to CO2-pneumoperitoneum therefore remain unclear.


    Acknowledgements
 
We thank Dr Anthony James and Mr John Tse, from Laboratory Animal Services Centre (LASEC) of the Chinese University of Hong Kong, for their support with the animal surgical procedures; and Stryker® China Limited for the loan of endoscopic equipment. We also thank Masimo SET for the Radical RDS-1 and accessories.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Anup, R., Aparna, V., Pulimood, A. and Balasubramanian, K.A. (1999) Surgical stress and the small intestine: role of oxygen free radicals. Surgery, 125, 560–569.[CrossRef][ISI][Medline]

Bentes de Souza, A.M., Rogers, M.S., Wang, C.C., Yuen, P.M. and Ng, P.S. (2003) Comparison of peritoneal oxidative stress during laparoscopy and laparotomy. J. Am. Assoc. Gynecol. Laparosc., 10, 65–74.[ISI][Medline]

Diamond, M.P. and Nezhat, F. (1993) Adhesions after resection of ovarian endometriomas. Fertil. Steril., 59, 934–935.

Diebel, L.N., Dulchavsky, S.A. and Wilson, R.F. (1992a) Effect of increased intra-abdominal pressure on mesenteric arterial and intestinal mucosal blood flow. J. Trauma, 33, 45–49.[ISI][Medline]

Diebel, L.N., Saxe, J. and Dulchavsky, S.A. (1992b) Effect of intra-abdominal pressure on abdominal wall blood flow. Am. Surg., 58, 573–576.[ISI][Medline]

Eleftheriadis, E., Kotzampassi, K., Papanotas, K., Heliadis, N. and Sarris, K. (1996) Gut ischemia, oxidative stress, and bacterial translocation in elevated abdominal pressure in rats. World J. Surg., 20, 11–16.[CrossRef][ISI][Medline]

Ferrari, R., Agnoletti, L., Comini, L., Gaia, G., Bachetti, T., Cargnoni, A., Ceconi, C., Curello, S. and Visioli, O. (1998) Oxidative stress during myocardial ischaemia and heart failure. Eur. Heart. J., 19, B2–11.[ISI][Medline]

Freeman, B.A. and Crapo, J.D. (1982) Biology of disease: free radicals and tissue injury. Lab. Invest., 47, 412–426.[ISI][Medline]

Galili, Y., Ben-Abraham, R., Rabau, M., Klausner, J. and Kluger, Y. (1998) Reduction of surgery-induced peritoneal adhesions by methylene blue. Am. J. Surg., 175, 30–32.[CrossRef][ISI][Medline]

Glantzounis, G.K., Tselepis, A.D., Tambaki, A.P., Trikalinos, T.A., Manataki, A.D., Galaris, D.A., Tsimoyiannis, E.C. and Kappas, A.M. (2001) Laparoscopic surgery-induced changes in oxidative stress markers in human plasma. Surg. Endosc., 15, 1315–1319.[Medline]

Guyton, A.C. (2000) Transport of oxygen and carbon dioxide in the blood and body fluids. In Guyton, A.C. and Hall, J.E. (eds), Textbook of Medical Physiology, 10th edn. W.B.Saunders, Philadelphia, PA, pp. 463–473.

Hemadeh, O., Chilukuri, S., Bonet, V., Hussein, S. and Chaudry, I.H. (1993) Prevention of peritoneal adhesions by administration of sodium carboxymethyl cellulose and oral vitamin E. Surgery, 114, 907–910.[ISI][Medline]

John, G.W. and Valentin, J.P. (1997) Analysis of the pulmonary hypertensive effects of the isoprostanes derivative, 8-epi-PGF2{alpha}, in the rat. Br. J. Pharmacol., 122, 899–905.[Abstract]

Kang, M.-Y., Tsuchiya, M., Packer, L. and Manabe, M. (1998) In vitro study on antioxidant potential of various drugs used in the perioperative period. Acta Anaesthesiol. Scand., 42, 4–12.

Molinas, C.P. and Koninckx, P.R. (2000) Hypoxaemia induced by CO2 or helium pneumoperitoneum is a co-factor in adhesion formation in rabbits. Hum. Reprod., 15, 1758–1763.[Abstract/Free Full Text]

Morrow, J.D., Award, J.A., Boss, H.J., Blair, I.A. and Roberts, L.J. (1992) Non-cyclooxygenase-derived prostanoids (F2-isoprostanes) are formed in situ on phospholipids. Proc. Natl Acad. Sci. USA, 89, 10721–10725.[Abstract]

Mrsny, R.J., Volwerk, J.J. and Griffith, O.H. (1986) A simplified procedure for lipid phosphorus analysis shows that digestion rates vary with phospholipid struture. Chem. Phy. Lipids, 39, 185–191.[CrossRef][ISI]

Odeberg, S., Ljungqvist, O. and Sollevi, A. (1998) Pneumoperitoneum for laparoscopic cholecystectomy is not associated with compromised splanchnic circulation. Eur. J. Surg., 164, 843–848.[CrossRef][ISI][Medline]

Ozmen, M.M., Kessaf Aslar, K., Besler, H.T. and Cinel, I. (2002) Does splanchnic ischemia occur during laparoscopic cholecystectomy? Surg. Endosc., 16, 468–471.[CrossRef][Medline]

Samel, S.T., Neufang, T., Mueller, A., Leister, I., Becker, H. and Post, S. (2002) A new abdominal cavity chamber to study the impact of increased intra-abdominal pressure on microcirculation of gut mucosa by using video microscopy in rats. Crit. Care Med., 30, 1854–1858.[ISI][Medline]

Schilling, M.K., Redaelli, C., Krahenbuhl, L., Signer, C. and Buchler, M.W. (1997) Splanchnic microcirculatory changes during CO2 laparoscopy. J. Am. Coll. Surg., 184, 378–382.[ISI][Medline]

Shimanuki, T., Nakamura, R.M. and DiZerega, G.S. (1986) A kinetic analysis of peritoneal fluid cytology and arachidonic acid metabolism after abrasion and reabrasion of rabbit peritoneum. J. Surg. Res., 41, 245–251.[ISI][Medline]

Taskin, O., Sadik, S., Onoglu, A., Gokdeniz, R., Yilmaz, I., Burak, F. and Wheeler, J. (1999) Adhesion formation after microlaparoscopic and laparoscopic ovarian coagulation for polycystic ovary disease. J. Am. Assoc. Gynecol. Laparosc., 6, 159–163.[ISI][Medline]

Tsimoyiannis, E.C., Tsimoyiannis, J.C., Sarros, C.J., Akalestos, G.C., Moutesidou, K.J., Lekkas, E.T. and Kotoulas, O.B. (1989) The role of oxygen-derived free radicals in peritoneal adhesions formation induced by ileal ischaemia/reperfusion. Acta Chir. Scand., 155, 171–174.[ISI][Medline]

United States Pharmacopeia and National Formulary and Supplements (1984) Vol. 21-NF. US Government Printing Office, Washington, DC.

Viinamaki, O. and Punnonen, R. (1982) Vasopressin release during laparoscopy: role of increased intra-abdominal pressure. Lancet, 1, 175–176.[Medline]

Volz, J., Koster, S., Spacek, Z. and Paweletz, N. (1999) Characteristic alterations of the peritoneum after carbon dioxide pneumoperitoneum. Surg. Endosc., 13, 611–614.[CrossRef][ISI][Medline]

West, M.A., Hackam, D.J., Baker, J., Rodrigues, J.L., Bellingham, J. and Rotstein, O. (1997) Mechanism of decreased in vitro murine macrophage cytokine release after exposure to carbon dioxide: relevance to laparoscopic surgery. Ann. Surg., 226, 179–190.[CrossRef][ISI][Medline]

Yesildaglar, N. and Koninckx, P.R. (2000) Adhesion formation in intubated rabbits increases with high insufflation pressure during endoscopic surgery. Hum. Reprod., 15, 687–691.[Abstract/Free Full Text]

Yura, T., Fukunaga, M., Khan, R., Nassar, G.N., Badr, K.F. and Montero, A. (1999) Free-radical-generated F2-isoprostane stimulates cell proliferation and endothelin-1 expression on endothelial cells. Kidney Int., 56, 471–478.[CrossRef][ISI][Medline]

Submitted on January 29, 2003; resubmitted on April 3, 2003; accepted on June 20, 2003.