Effects of Carbon Dioxide Pneumoperitoneum on the Inflammatory Response and Bacterial Translocation in Intraabdominal Infection

Abstract

Background:

This study evaluated the effect of CO₂ pneumoperitoneum on intraabdominal infection and bacterial translocation in intraabdominal infection.

Materials and Methods:

Escherichia coli and Bacteroides fragilis were injected separately into the abdominal cavities of 30 New Zealand white rabbits to establish two animal models of intraabdominal infection. Each model was divided into a laparotomy group, a pneumoperitoneum group, and a control group. Before and 1, 2, 4, and 7 days after surgery, blood and peritoneal fluids were obtained to determine bacterial culture and serum interleukin-6, tumor necrosis factor-α, and C-reactive protein levels by enzyme-linked immunosorbent assay. The total number of white blood cells (WBCs) was measured. Seven days after surgery, the animals were sacrificed and dissected, and liver, kidney, and spleen tissues were obtained for bacterial culture.

Results:

In the two bacterial models, incidence rates of bacteremia were higher in the laparotomy and pneumoperitoneum groups than in the control group. However, there were no significant differences between the laparotomy and pneumoperitoneum groups. Visceral bacterial translocation was detected in each group with no significant difference among the three groups. The change of inflammatory factors in the E. coli group and the B. fragilis group was nearly the same: the inflammatory factor levels and WBC counts in the laparotomy group were significantly higher than in the pneumoperitoneum group. The inflammatory factor levels and WBC counts in the pneumoperitoneum group increased slowly and were restored to normal quickly.

Conclusions:

In the intraabdominal infection animal model of the pneumoperitoneum group, the inflammatory response was weaker and the immune function was less affected and restored to normal more quickly than in the laparotomy group. The incidence rate of visceral bacterial translocation was not higher than that in the laparotomy group.

Introduction

Because laparoscopy has become more widely used in the clinical setting, minimally invasive abdominal surgery has been used increasingly in the treatment of a variety of diseases due to infections such as acute appendicitis. The suitability of the application of laparoscopy and carbon dioxide (CO₂) pneumoperitoneum in the treatment of peritonitis, however, remains controversial. Clinical observation has revealed that patients with peritonitis disease who undergo laparoscopic surgery develop postoperative abdominal abscess, wound infection, and other complications.¹ It has been reported that an intraabdominal pressure increase can induce intestinal ischemia and bacterial translocation in an experimental sepsis model.² It has also been proposed that CO₂ pneumoperitoneum causes an increase of intraabdominal pressure, which not only affects heart and lung function but also can lead to bacterial translocation.³ It was also observed in animal studies, however, that CO₂ pneumoperitoneum caused an inflammatory response.⁴ To some extent, these controversial issues limit the application of CO₂ pneumoperitoneum laparoscopy in the clinical setting.

Most previous animal experiments, however, use a single Escherichia coli infection model or feces-mixed infection model,^2,5 which has some limitations for the study of bacterial infection and changes in immune function, especially for anaerobic bacteria. In this study, two types of peritonitis animal models using E. coli and Bacteroides fragilis were used to study the effects of CO₂ pneumoperitoneum on bacterial translocation and inflammatory response.

Materials and Methods

Establishment of the animal model

In total, 60 purebred male New Zealand white rabbits weighing 2.5±0.5 kg were provided by the Shandong University Experimental Animal Center. The rabbits were divided randomly into the E. coli infection group and the B. fragilis infections group. Each group was divided into the pneumoperitoneum group, laparotomy group, and control group with 10 in each. The E. coli infection group was injected intraperitoneally at 12 hours before surgery with 4 mL of suspension at a concentration of 10¹¹ colony-forming units/mL. The B. fragilis group was injected intraperitoneally at 12 hours before surgery with 4 mL of B. fragilis suspension at a concentration of 10¹¹ colony-forming units/mL.

Surgical methods and specimen collection

First, 3% sodium pentobarbital at a concentration of 60 mg/kg was injected into the marginal ear vein for anesthesia. The abdomen was cut at the midline, and the abdominal wall was incised 8 cm along the white line of the abdomen. The abdomen was closed after exposure for 30 minutes. Stryker Endoscopy Insufflator (San Jose, CA) was used in the CO₂ pneumoperitoneum group. The pneumoperitoneum needle was inserted into the abdominal cavity to maintain 10 mm Hg pneumoperitoneum pressure for 30 minutes. The control group underwent intraperitoneal bacterial injection without other treatment.

Blood and ascites were collected before intraperitoneal injection of bacteria and 1, 2, 4, and 7 days after surgery. Blood (3.5 mL) from the rabbit ear artery was collected; 1 mL was used for bacterial culture, 0.5 mL was added to an EDTA anticoagulation blood collection tube for blood cell count, and the remaining blood was refrigerated at −70°C after centrifugation for the detection of cytokine preparedness. The animal's abdomen was then disinfected, and ascites was obtained by B ultrasound-guided abdominal puncture using sterile syringes.

The animals were sacrificed and dissected 7 days after surgery, and liver, spleen, and kidney tissues were put into an autoclave centrifuge tube for the bacterial culture study. Each specimen was divided into two parts for aerobic and anaerobic bacteria culture studies.

Bacterial culture

Each blood or ascites specimen was inoculated on two blood agar slants. One blood agar was incubated in a 37°C incubator for aerobic culture for 48–72 hours, and the other blood agar was incubated in an anaerobic bag (Merieux, Lyon, France) at 37°C for 48–72 hours. Next, the bacteria were identified. In the clean bench, the copper mesh was placed in sterile containers, and the animal organ tissue blocks were placed on the copper grids and ground up. Some filtrate was inoculated in the blood agar and cultured in the 37°C incubator. The remaining blood agar was loaded into the anaerobic bags. The airbags were sealed and incubated for anaerobic culture.

The morphology of the bacteria cell was observed under the microscope. Combing with the colony characteristics, the types of bacteria were preliminarily determined. The bacteria were determined ultimately by identification of biochemical reactions. The blood agar was determined to be negative if there was no bacterial growth after 7 days.

Blood cell count

The white blood cell (WBC) count of the blood collected was determined using an automatic hematology analyzer (model XT1800I; Sysmex, Kobe, Japan).

Detection of inflammatory cytokines

Serum interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) levels were determined with an enzyme-linked immunoabsorbent assay kit from RapidBio (West Hills, CA). C-reactive protein (CRP) levels were determined with an enzyme-linked immunoabsorbent assay test kit from TPI (Johnson City, TN). All kits were double-antibody sandwich assays.

Statistical analysis

Random experiments were performed, and the paired t test was used to detect the level of inflammatory cytokines in each group. The positive rate of the bacterial culture, abdominal abscess incidence, and organ bacterial translocation incidence between groups were detected using exact probabilities in fourfold tables. All statistical work was completed using SPSS version 10.0 statistical software (SPSS, Inc., Chicago, IL).

Results

Postoperative bacterial infection

Intraabdominal abscess

There were more intraabdominal abscesses in the laparotomy and pneumoperitoneum groups, with no significant difference between groups. In the laparotomy group, the incidence rate of mesenteric abscess was 65%, and the incidence rate of abdominal wall abscess was 35%. The incidence rates in the pneumoperitoneum groups were 55% and 5%, respectively.

Blood culture results

Postoperative blood culture results are shown in Table 1. On the first and second days after surgery, the rates of positive blood culture in the E. coli laparotomy group and the pneumoperitoneum group were significantly higher than in the control group (P<.05), but with no significant differences between the laparotomy and pneumoperitoneum groups. The blood culture–positive rate of the B. fragilis infection group was similar to that in the E. coli group.

Table 1.

Positive Cases of Blood Culture

	Postsurgery
Infection, group	Day 1	Day 2	Day 4	Day 7
E. coli
Laparotomy	10	8	3	1
Pneumoperitoneum	9	7	2	0
Control	3	1	0	0
B. fragilis
Laparotomy	9	6	2	0
Pneumoperitoneum	8	4	1	0
Control	2	1	0	0

Ascites culture results

The aerobic and anaerobic bacteria were cultured in the ascites, and the positive rate gradually decreased. The major aerobic bacteria found were E. coli (G–) and Staphylococcus aureus (G+), and mainly anaerobic bacteria were B. fragilis (G–), Peptostreptococcus (G+) anaerobic Clostridium (G+), Bacteroides (G–). The E. coli infection group ascites culture showed primarily E. coli, and the B. fragilis infection group ascites culture showed primarily B. fragilis and E. coli.

Bacterial translocation

Postoperative bacterial translocation results are shown in Table 2. Seven days after surgery, 8 cases in the laparotomy group, 9 cases in the pneumoperitoneum group, and 2 cases in the control group appeared to show bacterial translocation. There was no significant difference (P=.065) between the laparotomy group and the control group. There was a significant difference (P=.031) between the pneumoperitoneum group and the control group. There was no significant difference (P=1.000) between the laparotomy group and the pneumoperitoneum group. Tissue from three organs (liver, spleen, and kidney) showed bacterial translocation, particularly in the liver and kidney.

Table 2.

Positive Cases of Bacterial Translocation

Infection, group	Liver	Spleen	Kidney	Total
E. coli
Laparotomy	2	1	1	4
Pneumoperitoneum	2	0	2	4
Control	0	0	1	1
B. fragilis
Laparotomy	2	1	1	4
Pneumoperitoneum	3	0	2	5
Control	1	0	0	1

The aerobic bacteria with bacterial translocation were primarily E. coli. The anaerobic bacteria with bacterial translocation were mainly B. fragilis. In the laparotomy group of the E. coli model, postoperative organ bacteria with bacterial translocation were primarily aerobic bacteria (E. coli, S. aureus).

The pneumoperitoneum group of the E. coli infection model showed postoperative bacteria with bacterial translocation that were primarily anaerobic bacteria (B. fragilis, Peptostreptococcus). In the B. fragilis infection groups, the primary bacteria with bacterial translocation were anaerobic bacteria (B. fragilis).

Immune response

Changes in WBC

The postoperative WBC count changes are shown in Table 3. On the first and second days after laparotomy, the WBC count was significantly higher in the pneumoperitoneum group (P<.05), and on the fourth day, the WBC count gradually returned to normal levels.

Table 3.

Number of Blood Leukocytes

		Postsurgery
Infection, group	Before injection	Day 1	Day 2	Day 4	Day 7
E. coli
Laparotomy (Group 1)	12.05±4.53	16.03±2.3	13.96±1.77	12.51±1.68	12.00±2.46
Pneumoperitoneum (Group 2)	11.53±3.14	12.52±1.63	12.00±1.27	11.51±1.21	11.03±1.36
Control (Group 3)	11.22±4.79	12.51±1.95	12.22±0.99	12.02±1.32	11.25±1.51
P
Group 1 versus Group 3	.697	.002	.017	.481	.424
Group 2 versus Group 3	.866	.996	.677	.381	.729
Group 1 versus Group 2	.772	<.001	.012	.148	.291
B. fragilis
Laparotomy group	11.81±1.02	17.81±2.87	15.31±0.65	13.19±2.35	12.19±2.27
Pneumoperitoneum group	11.24±1.69	13.09±2.72	12.51±1.61	11.83±1.84	13.09±2.03
Control group	10.90±1.70	13.53±1.90	13.01±2.05	12.49±2.25	11.52±2.74
P
Group 4 versus Group 6	.167	<.001	.006	.508	559
Group 5 versus Group 6	.659	.678	.548	.475	.162
Group 4 versus Group 5	.375	<.001	<.001	.166	.360

Data are mean±standard deviation values (×10⁹/L).

Changes in inflammatory cytokines

The changes of the IL-6 levels in each group are shown in Table 4. The IL-6 levels in the laparotomy group on the first and second days after surgery were significantly higher than levels in the pneumoperitoneum group (P<.001). TNF-α level changes in each group are shown in Table 5. The TNF-α level of the laparotomy group during the first to seventh day after surgery was significantly higher than that in the pneumoperitoneum group (P<.05). The CRP level changes among groups are shown in Table 6. The CRP levels of the laparotomy group during the first to seventh days after surgery were significantly higher than in the pneumoperitoneum and control groups (P<.05).

Table 4.

Blood Interleukin-6 Level

		Postsurgery
Infection, group	Before injection	Day 1	Day 2	Day 4	Day 7
E. coli
Laparotomy (Group 1)	23.94±5.27	225.95±63.75	153.56±42.14	83.49±32.55	33.19±7.00
Pneumoperitoneum (Group 2)	25.01±5.21	126.21±35.48	56.75±19.12	28.49±9.65	24.88±7.54
Control (Group 3)	27.28±6.01	44.41±10.00	24.94±6.05	23.89±5.29	19.65±2.42
P
Group 1 versus Group 3	.204	<.001	<.001	<.001	<.001
Group 2 versus Group 3	.379	<.001	<.001	.207	.051
Group 1 versus Group 2	.654	<.001	<.001	<.001	.018
B. fragilis
Laparotomy (Group 4)	25.28±6.41	202.03±40.09	113.40±22.626	81.66±22.37	28.78±7.86
Pneumoperitoneum (Group 5)	28.01±5.40	113.31±25.31	47.90±19.27	24.94±6.39	23.35±5.42
Control (Group 6)	28.39±7.23	40.44±7.38	25.10±6.48	22.59±4.74	19.61±3.79
P
Group 4 versus Group 6	.323	<.001	<.001	<.001	.006
Group 5 versus Group 6	.895	<.001	.002	.364	.093
Group 4 versus Group 5	.318	<.001	<.001	<.001	.089

Data are mean±standard deviation values (in pg/mL).

Table 5.

Blood Tumor Necrosis Factor-α Level

		Postsurgery
Infection, group	Before surgery	Day 1	Day 2	Day 4	Day 7
E. coli
Laparotomy (Group 1)	40.57±8.68	101.03±27.19	137.75±13.88	106.39±12.98	85.54±10.06
Pneumoperitoneum (Group 2)	44.57±11.35	66.09±9.85	50.64±9.54	55.81±12.77	43.40±9.78
Control (Group 3)	51.38±6.95	57.79±9.21	52.70±9.45	46.42±11.85	51.46±12.33
P
Group 1 versus Group 3	.007	<.001	<.001	<.001	<.001
Group 2 versus Group 3	.126	.067	.635	.106	.124
Group 1 versus Group 2	.388	<.001	<.001	<.001	<.001
B. fragilis
Laparotomy (Group 4)	45.57±9.33	84.89±11.94	117.57±20.79	84.53±12.10	73.15±13.53
Pneumoperitoneum (Group 5)	50.54±7.90	70.61±13.38	62.85±12.94	53.54±11.97	49.49±12.29
Control (Group 6)	42.52±9.55	52.71±9.88	47.39±13.10	48.62±11.77	45.60±12.57
P
Group 4 versus Group 6	.478	<.001	<.001	<.001	<.001
Group 5 versus Group 6	.056	.003	.016	.366	.494
Group 4 versus Group 6	.215	.022	<.001	<.001	<.001

Data are mean±standard deviation values (in pg/mL).

Table 6.

Blood C-Reactive Protein Level

		Postsurgery
Infection, group	Before injection	Day 1	Day 2	Day 4	Day 7
E. coli
Laparotomy (Group 1)	8.28±3.69	139.01±53.41	179.71±49.84	120.32±17.46	71.16±12.56
Pneumoperitoneum (Group 2)	6.26±3.44	95.15±11.16	112.27±8.36	67.20±6.19	48.19±6.98
Control (Group 3)	4.60±2.75	51.19±4.60	52.20±5.51	40.16±4.63	33.33±6.73
P
Group 1 versus Group 3	.022	<.001	<.001	<.001	<.001
Group 2 versus Group 3	.251	<.001	<.001	<.001	<.001
Group 1 versus Group 2	.220	.030	.001	<.001	<.001
B. fragilis
Laparotomy (Group 4)	7.19±2.66	124.33±9.30	157.36±12.74	112.38±7.71	62.33±5.74
Pneumoperitoneum (Group 5)	5.32±2.05	67.27±7.13	93.45±8.05	61.27±7.31	35.21±6.55
Control (Group 6)	9.23±2.05	58.23±7.36	67.30±15.95	43.15±10.66	37.20±6.29
P
Group 4 versus Group 6	.072	<.001	<.001	<.001	<.001
Group 5 versus Group 6	<.001	.012	<.001	<.001	.496
Group 4 versus Group 5	.097	<.001	<.001	<.001	<.001

Data are mean±standard deviation values (in ng/mL).

Discussion

Infection and bacterial translocation

The influence of the pneumoperitoneum on peritonitis remains controversial. Some researchers regard the probability of occurrence of CO₂ pneumoperitoneum as higher than that of bacteremia from the laparotomy.⁶ The results of this study showed that intraabdominal infections of the pneumoperitoneum group caused by either E. coli or B. fragilis appeared to be bacteremia. The laparotomy group also showed significant bacteremia, and its incidence is not significantly different from that in the pneumoperitoneum group. Therefore, although bacteremia occurred in the pneumoperitoneum group, the incidence was not higher than that in the laparotomy group.

Our study also showed both aerobic and anaerobic bacteria in blood cultures. In the E. coli model group, there are primarily aerobes in the blood cultures and ascites, with only a small amount of anaerobic bacteria compared with the B. fragilis group. Therefore, it can be inferred that a single injection of aerobic or anaerobic bacteria in the peritoneal cavity produces a mixed bacterial infection in ascites culture. This has been confirmed in the ascites cultures. In addition, we speculate that CO₂ pneumoperitoneum caused by a low oxygen environment will also promote the growth of anaerobic bacteria.

Antibiotics were not used in the animal experiments. There was no significant difference in the incidence of abdominal abscess in the pneumoperitoneum and laparotomy groups, indicating that pneumoperitoneum does not increase the risk of postoperative intraabdominal infections.

Ozmen et al.⁶ observed that pneumoperitoneum can lead to bacterial translocation of the liver, spleen, kidney, mesenteric lymph nodes, and other organs. This experiment also showed that both the laparotomy group and the CO₂ pneumoperitoneum group have organ bacterial translocation; however, there was no significant difference between these groups. The bacterial species in the organ bacterial culture and postoperative blood culture are consistent, which indicates that the bacteria may reach distant organs through the blood pathway. This study suggests that anaerobic bacteria play a very important role in infection in the CO₂ pneumoperitoneum group.

The mechanism of the pneumoperitoneum bacterial translocation is not clear at present. Some studies found that laparoscopic surgery can cause a peritoneal changes leading to hypoxia and acidosis.⁷ Bloechle et al.⁸ observed using electron microscopy that in CO₂ pneumoperitoneum, because of the increased abdominal pressure, the elongation, deformation, and increase of the cell gap of the peritoneal mesothelial cell may cause the increase in abdominal lymphatic duct openings and introduce bacteria into the blood circulation.

In addition, it has been reported that CO₂ pneumoperitoneum increases survival in an animal model of peritonitis.^5,9 Peritoneal acidosis plays an important role in CO₂ pneumoperitoneum.^10,11 It is observed that CO₂ suppresses macrophage superoxide anion production independent of extracellular pH and mitochondrial activity¹² and affects the cytokine release of macrophage subpopulations exclusively via alteration of extracellular pH.¹³

Inflammatory reaction

The change in immune response in the postoperative laparoscopic pneumoperitoneum has become an important topic of research in recent years.¹⁴ Infection and surgical trauma may cause the activation of the immune system. CRP, IL-6, and TNF-α are important mediators in the acute inflammatory response. The increase in levels of inflammatory cytokines activates the body's immune response, which helps the body's response to infection and trauma. The high levels of postoperative inflammatory factors, however, may cause damage to the body and may be closely associated with postoperative infectious complications.

Balague et al.¹⁵ observed in the mouse E. coli abdominal infection model that there was a significant decrease in levels of inflammatory cytokines in the pneumoperitoneum group compared with those in the laparotomy group, because the anaerobe plays a very important role in celiac bacterial infection. To clarify changes in inflammation response in aerobic bacterial infections and anaerobic infections, our study used both E. coli and B. fragilis infection models. The results show that both models reached the same conclusion: that the postoperative IL-6, TNF-α, and CRP levels in the pneumoperitoneum group rose slightly but soon returned to normal. Compared with the pneumoperitoneum and control groups, inflammatory cytokine levels in the laparotomy group were significantly increased and decreased slowly. It is therefore inferred that, compared with the pneumoperitoneum group, the inflammatory response in the laparotomy group was more dramatic and postoperative recovery was slow, whereas the pneumoperitoneum can better protect immune function and improve postoperative recovery.

Clary et al.¹⁶ observed that the total number of WBCs in the bacterial peritonitis after surgery increased. The numbers of single-core and multicore bacterial peritonitis after surgery increased, whereas the number of lymphocytes decreased and returned to normal after 48 hours. The change in WBCs, however, was not observed as occurring over a longer time. Our study showed that the total number of blood leukocytes in each group peaked on the first day after surgery. On the first and second days after surgery, the total number of leukocytes in the laparotomy group was significantly higher than that in the pneumoperitoneum and control groups and decreased slowly, indicating that the impact of trauma caused by laparotomy on leukocytes was greater than caused by the pneumoperitoneum. Our study revealed that compared with laparotomy, the pneumoperitoneum can better protect the immune function of WBCs.

In summary, in both intraabdominal infection models of E. coli and B. fragilis, the WBC count of the pneumoperitoneum group rose more slowly and recovered more quickly than in the laparotomy group. Compared with the laparotomy group, the pneumoperitoneum group showed a lesser inflammation response, and the influence on immune function is weaker with a faster postoperative recovery. CO₂ pneumoperitoneum can cause organ bacterial translocation, but it is not higher than that in the laparotomy group. Pneumoperitoneum compared with laparotomy, therefore, will not aggravate inflammation. The CO₂ pneumoperitoneum promotes the growth of anaerobic bacteria. After pneumoperitoneum and laparoscopic surgery, respectively, in patients with intraabdominal infections, the use of sensitive antibiotics to prevent anaerobic infections should be considered.

Footnotes

Acknowledgments

We gratefully acknowledge the valuable cooperation of Prof. Yabin Zhou (Microbiology Laboratory, Shandong University).

Disclosure Statement

No competing financial interests exist.

References

Esposito

, Calvo

, Castagnetti

, Alicchio

, Suarez

, Giurin

, Settimi

. Open versus laparoscopic appendectomy in the pediatric population: A literature review and analysis of complications. J Laparoendosc Adv Surg Tech A, 2012; 22:834–839.

Kesici

, Kesici

, Polat

, Agca

, Turkmen

, Ozcan

, Sari

. Effects of intra-abdominal pressure increase on intestinal ischemia and bacterial translocation in experimental sepsis model. Saudi Med J, 2011; 32:813–817.

Sukhotnik

, Bejar

, Srugo

, Krausz

, Bernshteyn

, Hirsh

, Mogilner

. Adverse effects of increased intra-abdominal pressure on small bowel structure and bacterial translocation in the rat. J Laparoendosc Adv Surg Tech A, 2006; 16:404–410.

Hsieh

, Tain

, Chen

, et al. Carbon dioxide pneumoperitoneum induces anti-inflammatory response and hepatic oxidative stress in young rats with bacterial peritonitis. Pediatr Surg Int, 2011; 27:289–294.

Metzelder

, Kuebler

, Shimotakahara

, Chang

, Vieten

, Ure

. CO₂ pneumoperitoneum increases survival in mice with polymicrobial peritonitis. Eur J Pediatr Surg, 2008; 18:171–175.

Ozmen

, Col

, Aksoy

, et al. Effect of CO₂ insufflation on bacteremia and bacterial translocation in an animal model of peritonitis. Surg Endosc, 1999; 13:801–803.

Brokelman

, Lensvelt

, Borel Rinkes

, et al. Peritoneal changes due to laparoscopic surgery. Surg Endosc, 2011; 25:1–9.

Bloechle

, Kluth

, Holstein

, et al. A pneumoperitoneum perpetuates severe damage to the ultrastructural integrity of parietal peritoneum in gastric perforation-induced peritonitis in rats. Surg Endosc, 1999; 13:683–688.

Chatzimavroudis

, Pavlidis

, Koutelidakis

, Giamarrelos-Bourboulis

, Atmatzidis

, Kontopoulou

, Marakis

, Atmatzidis

. CO₂ pneumoperitoneum prolongs survival in an animal model of peritonitis compared to laparotomy. J Surg Res, 2009; 152:69–75.

10.

Kuebler

, Vieten

, Shimotakahara

, Metzelder

, Jesch

, Ure

. Acidification during carbon dioxide pneumoperitoneum is restricted to the gas-exposed peritoneal surface: Effects of pressure, gas flow, and additional intraperitoneal fluids. J Laparoendosc Adv Surg Tech A, 2006; 16:654–658.

11.

Hanly

, Aurora

, Shih

, Fuentes

, Marohn

, De Maio

, Talamini

. Peritoneal acidosis mediates immunoprotection in laparoscopic surgery. Surgery, 2007; 142:357–364.

12.

Kuebler

, Kos

, Jesch

, Metzelder

, van der Zee

, Bax

, Vieten

, Ure

. Carbon dioxide suppresses macrophage superoxide anion production independent of extracellular pH and mitochondrial activity. J Pediatr Surg, 2007; 42:244–248.

13.

Kos

, Kuebler

, Jesch

, Vieten

, Bax

, van der Zee

, Busche

, Ure

. Carbon dioxide differentially affects the cytokine release of macrophage subpopulations exclusively via alteration of extracellular pH. Surg Endosc, 2006; 20:570–576.

14.

Montalto

, Bitto

, Irrera

, Polito

, Rinaldi

, Antonuccio

, Impellizzeri

, Altavilla

, Squadrito

, Romeo

. CO₂ pneumoperitoneum impact on early liver and lung cytokine expression in a rat model of abdominal sepsis. Surg Endosc, 2012; 26:984–989.

15.

Balague

, Targarona

, Pujol

, et al. Peritoneal response to a septic challenge. Comparison between open laparoscopy, pneumoperitoneum laparoscopy, and wall lift laparoscopy. Surg Endosc, 1999; 13:792–796.

16.

Clary

, Bruch

, Lau

, et al. Effects of pneumoperitoneum on hemodynamic and systemic immunologic response to peritonitis in pigs. J Surg Res, 2002; 108:32–38.