Journal of Surgical Research 164, 28–37 (2010)
doi:10.1016/j.jss.2010.05.046
ASSOCIATION FOR ACADEMIC SURGERY
The Humoral Response After Laparoscopic Versus Open Colorectal
Surgery: A Meta-Analysis
Tarik Sammour, M.B.Ch.B.,*,‡,1 Arman Kahokehr, M.B.Ch.B.,*,‡ Sophie Chan, B.H.B.,‡ Roger J. Booth, Ph.D.,†
and Andrew G. Hill, M.D., F.R.A.C.S.*
*Department of Surgery, South Auckland Clinical School, University of Auckland, Auckland, New Zealand; †School of Medical Sciences,
University of Auckland, Auckland, New Zealand; and ‡Faculty of Medical and Health Sciences, University of Auckland, Auckland,
New Zealand
Submitted for publication December 24, 2009
Background. The local and systemic humoral
response after colorectal surgery is thought to affect
postoperative recovery. It is commonly claimed that
laparoscopic surgery elicits a diminished inflamma-
tory response than equivalent open surgery. Despite
these claims, the evidence is conflicting. Therefore,
we aimed to systematically review the results from
randomized controlled clinical trials comparing the
humoral response associated with laparoscopic versus
open colorectal surgery.
Materials and Methods. A high-sensitivity search
was conducted independently by two of the authors
with no language restriction. Studies were identified
from the Cochrane Central Register of Controlled Tri-
als (CENTRAL/CCTR), Cochrane Library, Medline
(January 1966 to January 2009), PubMed (1950 to
January 2009), and Embase (1947 to January 2009).
Relevant meeting abstracts and reference lists were
manually searched. Data analysis was performed
using Review Manager ver. 5.0.
Results. Thirteen randomized controlled trials
were included. Meta-analysis demonstrated a signifi-
cantly higher serum IL-6 on d 1 after open colorectal
resection for neoplasia (n [ 97) compared with laparo-
scopic resection (n [ 76, P [ 0.0008) without signifi-
cant heterogeneity. Data for plasma IL-6 were
heterogeneous, with no apparent difference between
groups. No other significant differences were identi-
fied, and there were not enough data on local perito-
neal humoral factors to allow meta-analysis.
1
To whom correspondence and reprint requests should be
addressed at Department of Surgery, South Auckland Clinical School,
Faculty of Medical and Health Sciences, University of Auckland,
Auckland, NZ. E-mail: tsammour@middlemore.co.nz.
0022-4804/$36.00 28
Ó 2010 Elsevier Inc. All rights reserved.
Conclusion. Open colorectal resection for neoplasia
is associated with higher postoperative serum levels of
IL-6 on d 1 than equivalent laparoscopic surgery. The
aetiology and clinical significance of this finding is un-
certain, and further studies are required to elucidate
any differences in the local humoral response which
may be more clinically relevant in surgery for this indi-
cation. Ó 2010 Elsevier Inc. All rights reserved.
Key Words: laparoscopy; colon, rectum; pneumoperi-
toneum; cytokine, immunology.
INTRODUCTION
Major abdominal surgery evokes an intense local and
systemic inflammatory reaction that has important
immunological consequences. Peritoneal mesothelial
cells and resident leukocytes, activated by injury,
secrete various mediators (Table 1) creating a local mi-
lieu that is directly responsible for the inflammatory
and repair processes that contribute to the injury re-
sponse [1–4]. The importance of this phenomenon lies
in its association with several clinical outcomes. The
exaggerated production of pro-inflammatory cytokines
such as interleukin 6 (IL-6) and tumor necrosis factor
(TNF-a) in the acute phase manifests systemically as
hemodynamic instability and metabolic derangement
[5]. This, in conjunction with the endocrine responses
exhibited by patients undergoing major surgery [6],
results in the classic pattern of reduced metabolism
for approximately 24 h postoperatively, followed by
a catabolic phase of at least 2 wk duration [5] associated
with muscle protein degradation, lipolysis, and distur-
bances in glucose metabolism [7, 8]. Even after these
physiological parameters have seemingly recovered,
postoperative patient fatigue can persist, lasting for
 SAMMOUR ET AL.: A SYSTEMATIC REVIEW AND META-ANALYSIS
TABLE 1
Classification of Humoral Mediators with Examples
Mediator
Chemokines
 IL-8, LTB4, MCP-1
Cytokines
 TNFa, IL-1b, IL-4, IL-6, IL-10, IL-13
Growth factors
 TGFa, TGF-b, PDGF, VEGF, EGF
Coagulation cascade
 tissue factor, tPA, uPA, PAI
Nitric oxide
Matrix metalloproteinases
 TIMP 1-4
IL ¼ interleukin; LT-B4 ¼ leukotriene B4; MCP-1 ¼ monocyte che-
moattractant protein 1; TNF-a ¼ tumor necrosis factor a; TGF-a ¼ tu-
mor growth factor a; TGF-b ¼ tumor growth factor b; PDGF ¼ platelet
derived growth factor; VEGF ¼ vascular endothelial growth factor;
EGF ¼ epidermal growth factor; tPA ¼ tissue plasminogen activator;
uPA ¼ urokinase plasminogen activator; PAI ¼ plasminogen activa-
tor inhibitor 1.
up to 3 mo after major uncomplicated gastrointestinal
operations [9, 10]. Research studies into fatigue
suggest a complex bio-psycho-social etiology [11], but
there is recent evidence that postsurgical inflammatory
responses may directly influence its development [12].
The local humoral environment after surgery also has
direct and specific consequences. For example, after
peritoneal injury, levels of local growth factors increase
substantially, and fibrinolytic activity is markedly
reduced [13–17], with the imbalance resulting
in increased interstitial collagen deposition and
peritoneal adhesion formation [18–21]. Also, there is
evidence that cytokine and chemokine up-regulation
has a direct inhibitory effect on the muscularis externa
of the bowel, contributing to postoperative ileus [22–26].
While these observations suggest a significant role
for the postoperative inflammatory response in recov-
ery after surgery, perhaps the most important conse-
quence is the detrimental impact that this response
may have on oncological outcome [27]. During and after
surgery for neoplasia, cells exfoliated from the primary
lesion (or leaked out of lymphatics) can bind to the en-
dothelium and mesothelium facilitated by interactions
between ligands induced by epidermal growth factor,
interleukin 1b, and other humoral factors [28, 29].
Peritoneal disruption during surgery dramatically
increases levels of these local mediators, leading to
a tumor promoting effect that is not restricted to the
inflicted site but rather has a generalized character,
promoting tumor growth in non-traumatised, as well
as traumatised peritoneum [29, 30].
Thus, the amelioration of the postoperative inflamma-
tory response is, and has been, the subject of intense
29
investigation. It is commonly claimed that laparoscopic
surgery is associated with a reduced systemic inflamma-
tory response compared with equivalent open surgery
[31–34]. It has also been suggested that oncological
outcomes may be improved by laparoscopic techniques
as a result of this diminished inflammatory response
[35, 36]. Conclusions have tended to be cautious,
however, as many of the studies are non-randomized,
[26] and focus on systemic levels of a heterogeneous
group of humoral factors, with relatively little attention
paid to the local peritoneal response which, as noted
above, may be the more important determinant of out-
come [37, 38]. Furthermore, while the immunological
benefits may be expected to be pronounced in
procedures where the access incision constitutes
a proportionally large burden of injury, such as in
laparoscopic cholecystectomy; in other procedures,
where the intra-abdominal insult is the more major, the
evidence remains unclear [39–43]. Colorectal surgery
serves as a case in point. Despite a large number of
trials comparing the postoperative humoral response
between laparoscopic and open surgery for this
indication, results are conflicting and opinion remains
divided [26].
The aim of this paper is, therefore, to systematically
review the results from randomized controlled clinical
trials comparing the local and systemic humoral
response in laparoscopic and open colorectal surgery.
MATERIALS AND METHODS
The principal study question is whether or not patients undergoing
elective laparoscopic colorectal surgery exhibit a reduced inflamma-
tory response as measured by local and systemic levels of humoral
markers. The comparison group is formed by patients undergoing
equivalent open colorectal surgery.
Systematic Literature Search
A high-sensitivity, low precision search was conducted with the
search terms outlined in Table 2, and in-line with the validated
methods of the QUORUM statement [44]. The search was run inde-
pendently by two of the authors (TS, AK), with no restrictions on lan-
guage. Relevant primary studies were identified from the Cochrane
Central Register of Controlled Trials (CENTRAL/CCTR), the Co-
chrane Library, Medline including in-process and non-indexed cita-
tions (from January 1966 to January 2009), PubMed (from 1950 to
January 2009), and Embase (from 1947 to January 2009). Relevant
scientific meeting abstracts and reference lists of included papers
were manually searched to identify further relevant publications.
Study Selection
A total of 1616 search results were entered into a unified database,
after which 111 duplicate results were removed. The titles and
abstracts of all the studies were screened. Published and unpublished
randomized controlled clinical trials that evaluated laparoscopic
versus open surgery for any colorectal indication were included.
Exclusion criteria were: animal studies, trials for non-colorectal indi-
cations, non-randomized trials (no restriction on randomisation
30 JOURNAL OF SURGICAL RESEARCH: VOL. 164, NO. 1, NOVEMBER 2010

TABLE 2 authors [59]. This had not been explicitly stated in the text of the
manuscript.
Search Terms Used in Different Databases

Database (number Validity Assessment

of results) Search terms
Assessment was performed by one author (TS) who was blinded to
CENTRAL (15 results) (laparoscopy/or pneumoperitoneum, the journal title, article title, and authors of the publications. Method-
Medline (72 results) artificial/or surgical procedures, ological quality was determined using the Jadad score [60]. This score
Cochrane library minimally invasive/)and(adipokines/or depends on three items: random allocation, masking of patients, and
(120 results) angiogenic proteins/or cytokines/or dropouts and withdrawals. The scale ranges from 0 to 5 points, with 2
PubMed (1256 results) endothelial growth factors/or or less indicating low quality and 3 or more indicating high quality.
Embase (153 results) endothelins/or epidermal growth factor/
or fibroblast growth factors/or Data Abstraction
interferons/or kinins/or transforming
growth factors/or tumor necrosis Data on humoral factor levels were gathered and the results of each
factors/) trial summarized on an intention-to-treat basis in prospectively de-
signed 2 3 2 tables. The data were categorized by humoral factor mea-
sured and time interval at which this was done. For the purposes of
meta-analysis, standard deviation of the outcome data was used if
method), and studies that did not measure plasma, serum, or perito-
provided by the authors, or calculated from P values, confidence inter-
neal levels of chemokines, cytokines, or growth factors as defined by
vals, or data ranges if they were not. The corresponding author for
Table 1. Studies that only measured other markers of inflammation
each publication was contacted if information was missing or unclear
such as serum/plasma C-reactive protein, erythrocyte sedimentation
to obtain as much raw data as possible.
rate, leukocyte count, immunoglobulin levels, and human leukocyte
antigen expression were not included. A QUORUM diagram is pro-
vided in Fig. 1. Statistical Analysis
Application of these criteria yielded a total of 41 potentially relevant
publications that were retrieved for more detailed evaluation. These Analysis of combined data was performed using Review Manager
were scrutinized for inclusion independently by two of the authors ver. 5.0 (Copenhagen: The Nordic Cochrane Centre, The Cochrane
(TS, AK), with disagreement resolved by consensus, and consultation Collaboration, 2008). Results of the meta-analysis were assessed by
with the senior author (AGH) if consensus could not be reached. Four- graphical presentations of standardized mean difference with 95%
teen randomized controlled trials comparing the humoral response in confidence intervals on forest plots using the random effects model
laparoscopic and open colorectal surgery were selected for inclusion in for more conservative estimates [61]; P < 0.050 was considered statis-
the review [45–58]. One of these trials was subsequently excluded [58] tically significant. Statistical heterogeneity was evaluated using I2
when it was realized that a subset of the data presented had been ob- statistics, with values up to 25%, up to 50%, and above 50% indicating
tained from a non-randomized trial published previously by the same low, moderate, and high levels of heterogeneity. A c2 test for

FIG. 1. QUORUM diagram demonstrating study selection. RCT ¼ randomized controlled trial; n ¼ number of papers. (Color version of fig-
ure is available online.)
 SAMMOUR ET AL.: A SYSTEMATIC REVIEW AND META-ANALYSIS 31

TABLE 3
Summary Characteristics and Quality Assessment of Included Studies According to Jadad et al. [60]

No. No. in Jadad

Study Indication in lap open Randomization Blinded? Score Comparable groups?

Delgado [45] Colon Ca 39 58 Computer generated N 2 Excluded conversions, more

left sided cases in open group
Dunker [53] IBD, FAP 16 14 Did not state N 1 Excluded transfused patients
Hasegawa [46] Colon and recto-sigmoid Ca. 24 26 Did not state N 1 Excluded conversions
Hewitt [47] Colon Ca 8 8 Did not state N 1 Excluded conversions
Leung [48] Recto-sigmoid Ca above 5 cm 17 17 Computer generated N 2 Y
Ordemann [49] Colon and recto-sigmoid Ca 20 20 Did not state N 1 Y
above 12 cm
Schwenk [50] Colon and rectal Ca 30 30 Did not state N 1 Y
Solomon [54] Full thickness rectal prolapse 20 19 Did not state Single 2 Y
Stage [51] Colon Ca 15 14 Random numbers N 3 Excluded conversions
Svendsen [55] Colon Ca, adenoma, diverticular, 23 30 Did not state N 1 Excluded conversions
sigmoid volvulus
Wu [52] Colon Ca 12 14 Computer generated N 3 More advanced stage
disease in lap
Wu [57] Colon Ca 12 14 Computer generated N 3 More advanced stage
disease in lap
Ytting [56] Colon Ca, adenoma, diverticular, 26 34 Did not state N 1 Y
sigmoid volvulus
No. ¼ number of patients; lap ¼ laparoscopic; Ca ¼ cancer, IBD ¼ inflammatory bowel disease; FAP ¼ familial adenomatous polyposis; cm ¼
centimetres.

heterogeneity was performed, in which P < 0.100 was regarded as sig-
nificant. Funnel plots were used to screen for publication bias. An
a priori sensitivity analysis was performed to assess the effect of se-
rum versus plasma cytokine measurements on data heterogeneity.
RESULTS
Thirteen studies were identified in total, published
over a 9 y period between 1997 and 2006. All of these
studies were in English. Nine of these studies described
surgery for colorectal neoplasia exclusively [45–52, 57],
one study described inflammatory bowel disease and
familial adenomatous polyposis [53], one study rectal
prolapse [54], and two studies resections for heteroge-
neous indications (including colonic neoplasia and be-
nign pathology) [55, 56]. Three of these studies [51,
52, 57] were rated as high quality according to the
Jadad score, and the rest as low quality. A summary
of study characteristics is presented in Table 3, and
a summary of results in Table 4.
Systemic Response
All thirteen studies included a measure of the hu-
moral response in the systemic circulation. Seven stud-
ies measured plasma levels of humoral markers, and
six studies used serum levels (Table 4).
surgery [46, 47, 53, 54, 56], and five studies
demonstrated higher levels in the open group [45, 48-
50, 52]. Only one study reported higher levels of plasma
IL-6 in laparoscopic resections for colonic cancer [51].
While measures of IL-6 levels were performed at a variety
of time-points, and on patients with diverse pathologies,
eight studies including only colorectal resections for neo-
plasia were congruous in that they measured systemic
levels of IL-6 on d 1 [45–52]. Therefore, a meta-analysis
of the data from these studies and at this time point
was performed, and the results of this are presented in
Fig. 2, with funnel plot analysis in Fig. 3. Patients in
the open group (n ¼ 187) had a significantly higher levels
of IL-6 on d 1 compared with patients in the laparoscopic
group (n ¼ 165, P ¼ 0.040). However, there was signifi-
cant heterogeneity in the results (I2 ¼ 80%; c2 ¼ 35.11,
P < 0.0001). Plasma and serum data were analyzed sep-
arately as per an a priori sensitivity analysis for sources
of heterogeneity. This revealed the majority of heteroge-
neity to be caused by studies measuring plasma levels of
IL-6. As can be seen in Fig. 4, serum IL-6 was signifi-
cantly higher in the open group (n ¼ 97) than in the lap-
aroscopic group (n ¼ 76, P ¼ 0.0008) with no significant
heterogeneity in the results (P ¼ 0.280). Data for plasma
were still heterogeneous, with no apparent difference be-
tween groups.

Interleukin-6 Tumor Necrosis Factor-a

IL-6 was the most commonly measured humoral fac- Three studies measured systemic levels of TNF-a.
tor. Five studies found no significant difference between Leung et al. found no significant difference between
systemic IL-6 levels in laparoscopic and open colorectal groups [48], and Ordemann et al. found significantly
32 JOURNAL OF SURGICAL RESEARCH: VOL. 164, NO. 1, NOVEMBER 2010

TABLE 4
Summary of Study Results. Significant Differences in Green, no Difference in Red

Study Sample Humoral factor Summary of results

Delgado [45] Serum IL-6 IL-6 sig higher in open group at 4 h, 12 h, 24 h.

Dunker [53] Plasma IL-6 IL-6 no sig difference at d 1, d 7.
Hasegawa [46] Plasma IL-6 IL-6 no sig difference at d 1, d 7.
Hewitt [47] SerumPDF IL-6 Serum IL-6 no sig difference at 0 h, 4 h, 8 h, d 1, d 2.
PDF IL-6 no sig difference at 0 h.
Leung [48] Serum IL-6 IL-6 sig higher peak in open.
IL-1b IL-1b sig higher peak in open.
TNF-a TNF-a no sig difference at 2 h, 8 h, d 1, d 2, d 3, d 7, d 28.
Ordemann [49] Plasma IL-6, IL-6 sig higher in open group at 1 h, 4 h, d 1.
TNF-a TNF-a sig higher in open group at 1 h, 4 h, d 1, d 2.
Schwenk [50] Plasma IL-6 IL-6 sig higher peak and AUC in open.
IL-10 IL-1RA no sig difference in peak or AUC.
IL-1RA IL-10 no sig difference in peak or AUC.
Solomon [54] Serum IL-6 IL-6 no sig difference at 4 h, d 1, d 2.
Stage [51] Plasma IL-6 IL-6 higher increase in lap group (d 1, d 3, d 10).
Svendsen [55] Plasma VEGF VEGF no sig difference 1 h, 2 h, 6 h, 24 h, 48 h, d 8, d 30.
Wu [52] SerumPDF IL-6 Serum IL-6 higher in open group at 2 h.
IL-8 Serum IL-8 higher in open group at 2 h.
TNF-a Serum TNF-a undetectable 2 h, d 1, d 4.
PDF IL-8 higher in lap group on 24 h collection d 4.
PDF IL-6 no sig difference on 24 h collection d 1, d 4.
PDF TNF-a undetectable on 24 h collection d 1, d 4.
Wu [57] SerumPDF VEGF Serum VEGF no sig difference 2 h, d 1, d 4.
PDF VEGF higher in lap group on 24 h collection d 4.
Ytting [56] Plasma IL-6 Plasma IL-6 no sig difference at 1 h, 2 h, 6 h, d 1, d 2, d 8, d 30.
IL ¼ interleukin; TNF-a ¼ tumor necrosis factor-a; VEGF ¼ vascular endothelial growth factor; h ¼ hours after surgery; d ¼ days after sur-
gery; lap ¼ laparoscopic; PDF ¼ peritoneal drain fluid; AUC ¼ area under the curve; sig ¼ significant.

higher levels in the open group at every time interval
where this was measured (Table 4) [49]. The analysis
performed by Wu et al. was not sensitive enough to de-
tect TNF-a levels [52]. Meta-analysis of this data was
not performed as this would have only included two
studies.
Interleukin-1
IL-1b exhibited significantly higher peak serum
levels after open surgery compared with laparoscopic
surgery for recto-sigmoid carcinoma in one study [48].
This was the only trial that measured levels of this
cytokine directly. However, Schwenk et al. measured
plasma levels of IL-1 receptor antagonist (IL-1RA) after
resection for colorectal cancer [50]. They argued that
since IL-1RA can block the pro-inflammatory effects of
IL-1b, it is a true IL-1b antagonist. No difference was
demonstrable between the two groups.
Other Humoral Factors
IL-8 and IL-10 levels were each only reported by one
study. Serum IL-8 was significantly higher after open

FIG. 2. Forest plot of plasma and serum IL-6 levels on d 1. SD ¼ standard deviation; CI ¼ confidence interval; Std ¼ standardized, % ¼
percentage. (Color version of figure is available online.)
 SAMMOUR ET AL.: A SYSTEMATIC REVIEW AND META-ANALYSIS
FIG. 3. Funnel plot of plasma and serum IL-6 levels on d 1. SE ¼
standard error; SMD ¼ standardized mean difference. (Color version
of figure is available online.)
surgery for colonic cancer [52], while no difference in
plasma IL-10 was demonstrable after surgery for colo-
rectal cancer [50]. Two studies measured levels of
plasma and serum vascular endothelial growth factor
(VEGF), with no difference demonstrable between
groups in either study [55, 57]. There was no data on
any of the other humoral factors mentioned in Table 1.
Local Response
Only two randomized trials (results reported in three
publications) measured peritoneal levels of humoral
FIG. 4. Forest plot of (A) plasma IL-6 levels on d 1, and (B) serum IL-6 levels on d 1. SD ¼ standard deviation; CI ¼ confidence interval; Std ¼
standardized; % ¼ percentage. (Color version of figure is available online.)
33
factors, and both of these were in patients with colonic
resection for neoplasia [47, 52, 57] Hewitt et al. did not
detect a significant difference in peritoneal IL-6 in fluid
aspirated from the peritoneal cavity at the end of
surgery [47]. Wu et al. performed a 24h collection of
peritoneal fluid at d 1 and 4 after surgery, and while
there was no difference in levels of IL-6 or TNF-a, sig-
nificantly higher levels of IL-8 were detected in the lap-
aroscopic group on d 4 [52]. In a follow-up analysis on
the same samples, the same authors presented results
showing higher levels of peritoneal VEGF in the laparo-
scopic group on d 4 [57].
DISCUSSION
This systematic review included 13 randomized con-
trolled trials comparing the local and systemic humoral
response after laparoscopic versus open colorectal
surgery. The identified papers were heterogeneous in
surgical indication, humoral factor measured, mea-
surement timing, and method of sample analysis. Nev-
ertheless, meta-analysis of the data on colorectal
resection for neoplasia demonstrated a statistically sig-
nificant higher serum level of IL-6 on d 1 after open sur-
gery. No other trends were identified, and there were
not enough data for other humoral factors and at other
time points to allow further meta-analysis.
The finding that serum IL-6 is significantly higher
after open colorectal surgery for neoplasia signifies a
potentially greater systemic stress response. IL-6 is
34 JOURNAL OF SURGICAL RESEARCH: VOL. 164, NO. 1, NOVEMBER 2010
considered to be a major mediator of the acute phase pro-
tein response following injury [2] and, in comparison
with other cytokines, the concentration of IL-6 is most
consistently increased in the circulation of injured pa-
tients [43]. The main producers of IL-6 are endothelial
cells of the peritoneal capillaries [62]. Production occurs
within the first hour of surgery and significantly in-
creases after 4 h, with a significant rise in serum
detected after 6 h [63]. Postoperatively, an early
exaggerated IL-6 response has been shown in one study
to precede the clinical onset of major complications by
12–48 h [2]. Although this is likely to be effect rather
than cause, in experimental animal studies systemic
IL-6 administration has been shown to have a direct det-
rimental effect on the healing of colonic anastomoses [64].
In this study, the difference in IL-6 levels was only de-
monstrable in analysis of serum data, with no signifi-
cant difference in plasma data despite an equivalent
combined number of patients in each data-set. Centrifu-
gation of samples after blood is collected in plain tubes
and allowed to clot yields serum, whereas spinning
non-clotted samples collected in anticoagulant tubes
provides plasma. It is unclear which technique more ac-
curately reflects the in vivo state, but it has been previ-
ously reported (at least for VEGF) that plasma rather
than serum assays may be more appropriate [36]. This
is because humoral factors stored in white blood cells
(WBCs) and platelets are released when blood clots,
which may result in overestimation [65].
The reason for the difference in the systemic humoral
response between equivalent laparoscopic and open
operations for colorectal neoplasia is unclear. One an-
swer is seemingly obvious: a smaller access incision
leads to a reduced burden of injury. This is true, but
simplistic, because in colorectal surgery the intra-
abdominal wound is considerably more extensive than
the access incision, dwarfing its systemic impact by
comparison [52, 53, 66, 67]. In the study by Dunker
et al, patients were grouped into two groups based on
the size of the access incision regardless of operation
(>8 cm and <8 cm), and there was no demonstrable
difference in plasma IL-6 concentrations on d 1 and
d 7 postoperatively between these two groups [53].
Differences between the two techniques are not limited
to the size of the access incision, and there are other factors
at play that may influence the systemic response. Laparo-
scopic colorectal procedures tend to be associated with
a longer operating time, a variable which has been directly
and positively correlated with systemic IL-6 levels postop-
eratively [68]. Anaesthetic and analgesia protocols are
also confounding variables, with spinal afferent blockade
[69] and opiate usage [47] both known to blunt the sys-
temic humoral response. Both of these modalities are gen-
erally utilized to a greater extent in open surgery due the
higher postoperative analgesia requirement [70, 71].
While the above would seem to favor a diminished
humoral response after open surgery (which is counter
to what the current meta-analysis results suggest), the
postoperative course may have the opposite effect. Laparo-
scopic surgery is generally associated with faster recovery,
and earlier resumption of mobilization and gastrointesti-
nal function [72]. How much of this is attributable to
surgical technique rather than differing postoperative
care, and patient/care-giver attitudes towards the size
of the incision, is a matter of debate [72–74]. The only
blinded study performed demonstrated almost identical
functional recovery of various organ functions after
equivalent laparoscopic and open colorectal procedures
[74]. This randomized controlled trial was performed in
the setting of a multi-modal enhanced recovery after sur-
gery (ERAS) program, which was credited with bridging
the gap between laparoscopic and open technique as far
postoperative clinical outcome [74]. Furthermore, a recent
study by Wichmann et al. demonstrated that ERAS tech-
niques lead to better preserved cell mediated immune
function [75]. It is notable that in the two studies included
in this meta-analysis that were performed in an ERAS set-
ting, no difference in the systemic response was demon-
strable between the laparoscopic and open groups [55, 56].
Another point worthy of discussion is that levels of lo-
cal peritoneal factors are thought to be of greater clini-
cal importance than systemic levels [18, 37]. There is
a much higher concentration of humoral factors in
peritoneal fluid than in plasma after colorectal
surgery suggesting that cytokine production occurs in
a compartmentalized fashion within the abdominal
cavity and at the site of dissection [37, 52]. While
some of these factors are thought to be incompletely
absorbed into the portal circulation, and end up in the
systemic system after being degraded by the liver and
then diluted in the plasma, there is an apparent
independence of the peritoneal response from the
systemic one [76]. For example, the IL-6 level in the
peritoneal fluid peaks later and lasts longer than
the systemic level measured in plasma [52]. Other
observed contrasting physiological responses to the
same humoral factor after intra-abdominal stimulation
serve to further illustrate this point [76]. Only two stud-
ies identified in this review measured levels of local
peritoneal humoral factors (results reported in three
papers). While one study did not identify a difference
between open and laparoscopic surgery, it is of interest
that the other found higher levels of IL-8 and VEGF
after laparoscopic surgery.
Further studies to elucidate any differences in the local
humoral response between open and laparoscopic colo-
rectal surgery are required. This is particularly impor-
tant because carbon dioxide pneumoperitoneum has
known local immunological effects [38]. Local acidifica-
tion caused by the formation of carbonic acid in the
 SAMMOUR ET AL.: A SYSTEMATIC REVIEW AND META-ANALYSIS
peritoneal fluid results in microscopically visible damage
to the mesothelial ultra-structure and reduced numbers
of activated neutrophils, which is more pronounced than
when peritoneum is exposed to room air [77–82]. This
acidification is independent of systemic acidosis [83].
There is also evidence that CO2 directly blunts peritoneal
macrophage function in vivo [84]. Another recognized is-
sue with pneumoperitoneum is the cold temperature
(room temperature) and dryness (0% relative humidity)
of the gas typically used [85]. Large volumes are often re-
quired in prolonged procedures (up to 500 L) owing to the
imperfect seal of the laparoscopic ports, and peritoneal
carbon dioxide absorption [85]. The effect of this cold,
dry gas flow on the peritoneal environment is significant,
producing visible structural changes in the peritoneal
mesothelial surface layer and an increased local
inflammatory response [80, 84, 86, 87]. Warming the
insufflated CO2 has been shown to be associated with
a reduced peritoneal cytokine response after
laparoscopic cholecystectomy in humans [88].
This systematic review is limited by the quality of the in-
cluded studies. Only three out of the 13 studies were rated
as high quality based on the Jadad scale, and only one
study was blinded (single blinded). While this is unlikely
to have influenced measured humoral factor concentra-
tions, true group allocation may be biased or concealed
by selective patient inclusion, data collection, and report-
ing. Also, there is a fundamental difficulty in the interpre-
tation of any studies on immune function. While reduced
levels of humoral factors tend to be interpreted as repre-
senting lesser injury, they could potentially also signify
suppression of host immunity by a more major surgical in-
sult [38]. Unless the concentration change is correlated di-
rectly with clinical outcome, the distinction is almost
impossible to make. Addressing this fundamental issue
should be a priority of future studies in this area.
CONCLUSION
Open colorectal resection for neoplasia is associated
with higher postoperative serum levels of IL-6 on d 1
than equivalent laparoscopic surgery. The aetiology
and clinical significance of this finding is uncertain,
and further studies are required to elucidate any differ-
ences in the local humoral response, which may be more
clinically relevant in surgery for this indication.
ACKNOWLEDGMENTS
TS is supported by a Surgeon Scientist Scholarship administered by
the Royal Australasian College of Surgeons. AK is supported by the
Ruth Spencer Fellowship administered by the Auckland Medical
Research Foundation.
35
SUPPLEMENTARY DATA
Supplementary data associated with the article can
be found in the online version, at doi:10.1016/j.jss.
2010.05.046.
REFERENCES
1. Chuang D, Paddison J, Booth R, et al. Differential production of
cytokines following colorectal surgery. ANZ J Surg 2006;76:821.
2. Baigrie R, Lamont P, Kwiatkowski D. Systemic cytokine re-
sponse after major surgery. Br J Surg 1992;19:757.
3. Whawell SA, Scott-Coombes DM, Vipond MN, et al. Tumor ne-
crosis factor-mediated release of plasminogen activator inhibitor
1 by human peritoneal mesothelial cells. Br J Surg 1994;81:214.
4. Riese J, Denzel C, Zowe M, et al. Secretion of IL-6, monocyte che-
moattractant protein-1, macrophage inflammatory protein-1a,
and TNFa by cultured intact human peritoneum. Eur Surg
Res 1999;31:281.
5. Choileain N, Redmond H. Cell response to surgery. Arch Surg
2006;141:1132.
6. Kehlet H. Manipulation of the metabolic response in clinical
practice. World J Surg 2000;24:690.
7. Singer M, De Santis V, Vitale D, et al. Multiorgan failure is an
adaptive, endocrine-mediated, metabolic response to over-
whelming systemic inflammation. Lancet 2004;364:545.
8. Hill A, Hill G. Metabolic response to severe injury. Br J Surg
1998;85:884.
9. Rubin GJ, Hardy R, Hotopf M. A systematic review and meta-
analysis of the incidence and severity of postoperative fatigue.
J Psychosom Res 2004;57:317.
10. Zargar-Shoshtari K, Paddison JS, Booth RJ, et al. A prospective
study on the influence of a fast-track program on postoperative
fatigue and functional recovery after major colonic surgery.
J Surg Res 2009;154:330.
11. Salmon P, Hall G. A theory of postoperative fatigue: An interac-
tion of biological, psychological, and social processes. Pharmacol
Biochem Behav 1997;56:623.
12. Paddison JS, Booth RJ, Fuchs D, et al. Peritoneal inflammation
and fatigue experiences following colorectal surgery: A pilot
study. Psychoneuroendocrinology 2008;33:446.
13. van der Wal JB, Jeekel J. Biology of the peritoneum in normal
homeostasis and after surgical trauma. Colorectal Dis 2007;
9(Suppl 2):9.
14. Chegini N, Rossi MJ, Schultz GS, et al. Cellular distribution of
epidermal growth factor, transforming growth factor-a, and epi-
dermal growth factor receptor in fascia and peritoneum during
healing in the rat: An autoradiographic and immunohistochem-
ical study. Wound Repair Regen 1993;1:28.
15. Wilkosz S, Epstein J, de Giorgio-Miller A, et al. Remodelling of
adipose tissue during experimental omental adhesion formation.
Br J Surg 2008;95:387.
16. Merlo G, Fausone G, Castagna B. Fibrinolytic activity of meso-
thelial lining of the displaced peritoneum. Am J Med Sci 1983;
286:12.
17. Whitaker D, Papadimitriou JM, Walters M. The mesothelium:
Its fibrinolytic properties. J Pathol 1982;136:291.
18. Cheong YC, Laird SM, Shelton JB, et al. The correlation of adhe-
sions and peritoneal fluid cytokine concentrations: A pilot study.
Hum Reprod (Oxford, England) 2002;17:1039.
19. Holmdahl L, Kotseos K, Bergstrom M, et al. Overproduction of
transforming growth factor- b1 (TGF- b1) is associated with ad-
hesion formation and peritoneal fibrinolytic impairment. Sur-
gery 2001;129:626.
20. Jones P, Werb Z. Degradation of connective tissue matrices by
macrophages. II. Influence of matrix composition on proteolysis
36 JOURNAL OF SURGICAL RESEARCH: VOL. 164, NO. 1, NOVEMBER 2010
of glycoproteins, elastin, and collagen by macrophages in cul-
ture. J Exp Med 1980;152:1527.
21. van der Poll T, Levi M, Buller H. Fibrinolytic response to tu-
mor necrosis factor in healthy subjects. J Exp Med 1991;
174:729.
22. de Jonge W, van Den Wijngaard R, The F. Postoperative il-
eus is maintained by intestinal immune infiltrates that acti-
vate inhibitory neural pathways in mice. Gastroenterology
2003;125:1137.
23. The FO, Bennink RJ, Ankum WM, et al. Intestinal handling-
induced mast cell activation and inflammation in human postop-
erative ileus [see comment]. Gut 2008;57:33.
24. Frasko R, Maruna P, Gurlich R, et al. Transcutaneous electro-
gastrography in patients with ileus. Relations to interleukin-
1b, interleukin-6, procalcitonin, and C-reactive protein. Eur
Surg Res 2008;41:197.
25. Moore BA, Albers KM, Davis BM, et al. Altered inflammatory
gene expression underlies increased susceptibility to murine
postoperative ileus with advancing age. Am J Physiol Gastroint-
est Liver Physiol 2007;292:G1650.
26. Sido B, Teklote JR, Hartel M, et al. Inflammatory response after
abdominal surgery. Best Pract Res Clin Anaesthesiol 2004;18:439.
27. Varghese S, Burness M, Xu H, et al. Site-specific gene expression
profiles and novel molecular prognostic factors in patients with
lower gastrointestinal adenocarcinoma diffusely metastatic to
liver or peritoneum. Ann Surg Oncol 2008;14:3460.
28. van Rossen M, Hofland L, van den Tol M, et al. Effect of inflam-
matory cytokines and growth factors on tumor cell adhesion to
the peritoneum. J Pathol 2001;193:530.
29. van den Tol P, van Rossen E, van Eijck C, et al. Reduction of peri-
toneal trauma by using nonsurgical gauze leads to less implan-
tation metastasis of spilled tumor cells. Ann Surg 1998;227:242.
30. Coffey J, Smith M, Wang J, et al. Cancer surgery: Risks and op-
portunities. Bio Essays 2006;28:433.
31. Schafer M, Krahenb hl L, Buchler MW. Comparison of adhesion
formation in open and laparoscopic surgery. Dig Surg 1998;
15:148.
32. Novitsky YW, Litwin DEM, Callery MP. The net immunologic
advantage of laparoscopic surgery. Surg Endosc 2004;18:1411.
33. Novitsky YW, Callery MP. Laparoscopic surgery and the nu-
ances of systemic immunity. J Clin Gastroenterol 2002;34:109.
34. Vittimberga FJ Jr, Foley DP, Meyers WC, et al. Laparoscopic
surgery and the systemic immune response. Ann Surg 1998;
227:326.
35. Whitson BA, D’Cunha J, Maddaus MA. Minimally invasive can-
cer surgery improves patient survival rates through less perio-
perative immunosuppression. Med Hypotheses 2007;68:1328.
36. Belizon A, Balik E, Feingold DL, et al. Major abdominal surgery
increases plasma levels of vascular endothelial growth factor:
Open more so than minimally invasive methods. Ann Surg
2006;244:792.
37. Wiik H, Karttunen R, Haukipuro K, et al. Maximal local and
minimal systemic cytokine response to colorectal surgery: The
influence of perioperative filgrastim. Cytokine 2001;14:188.
38. Hegarty N, Dasgupta P. Immunological aspects of minimally in-
vasive oncologic surgery. Curr Opin Urol 2008;18:129.
39. Gupta A, Watson DI. Effect of laparoscopy on immune function.
Br J Surg 2001;88:1296.
40. Hill AG. Manipulation of the small intestine as a cause of the in-
creased inflammatory response after open compared with lapa-
roscopic surgery. Br J Surg 2006;93:195 [comment]. Br J Surg
2006;93:504; author reply 505.
41. Hill AG, Connolly AB. Minimal access colonic surgery: Is it truly
minimally invasive? ANZ J Surg 2006;76:282.
42. Jacobi CA, Wenger F, Opitz I, et al. Immunologic changes during
minimally invasive surgery. Digest Surg 2002;19:459.
43. Sylla P, Kirman I, Whelan RL. Immunological advantages of ad-
vanced laparoscopy. Surg Clin North Am 2005;85:1.
44. Moher D, Cook DJ, Eastwood S, et al. Improving the quality of
reports of meta-analyses of randomized controlled trials: The
QUOROM statement. Quality of reporting of meta-analyses.
Lancet 1999;354:1896.
45. Delgado S, Lacy AM, Filella X, et al. Acute phase response in lap-
aroscopic and open colectomy in colon cancer: Randomized
study. Diseases Colon Rectum 2001;44:638.
46. Hasegawa H, Kabeshima Y, Watanabe M, et al. Randomized
controlled trial of laparoscopic versus open colectomy for ad-
vanced colorectal cancer. Surg Endosc 2003;17:636.
47. Hewitt PM, Ip SM, Kwok SP, et al. Laparoscopic-assisted vs.
open surgery for colorectal cancer: Comparative study of
immune effects. Dis Colon Rectum 1998;41:901.
48. Leung KL, Lai PB, Ho RL, et al. Systemic cytokine response after
laparoscopic-assisted resection of rectosigmoid carcinoma: A
prospective randomized trial. Ann Surg 2000;231:506.
49. Ordemann J, Jacobi CA, Schwenk W, et al. Cellular and humoral
inflammatory response after laparoscopic and conventional colo-
rectal resections: Results of a prospective randomized trial. Surg
Endoscop 2001;15:600.
50. Schwenk W, Jacobi C, Mansmann U, et al. Inflammatory response
after laparoscopic and conventional colorectal resections - results
of a prospective randomized trial. Langenbecks Arch Surg 2000;
385:2.
51. Stage JG, Schulze S, Moller P, et al. Prospective randomized
study of laparoscopic versus open colonic resection for adenocar-
cinoma. Br J Surg 1997;84:391.
52. Wu FP, Sietses C, von Blomberg BM, et al. Systemic and perito-
neal inflammatory response after laparoscopic or conventional
colon resection in cancer patients: A prospective, randomized
trial. Dis Colon Rectum 2003;46:147.
53. Dunker MS, Ten Hove T, Bemelman WA, et al. Interleukin-6, C-
reactive protein, and expression of human leukocyte antigen-DR
on peripheral blood mononuclear cells in patients after laparo-
scopic vs. conventional bowel resection: A randomized study.
Dis Colon Rectum 2003;46:1238.
54. Solomon MJ, Young CJ, Eyers AA, et al. Randomized clinical
trial of laparoscopic versus open abdominal rectopexy for rectal
prolapse. Br J Surg 2002;89:35.
55. Svendsen MN, Werther K, Christensen IJ, et al. Influence of
open versus laparoscopically assisted colectomy on soluble vas-
cular endothelial growth factor (sVEGF) and its soluble receptor
1 (sVEGFR1). Inflamm Res 2005;54:458.
56. Ytting H, Christensen IJ, Basse L, et al. Influence of major sur-
gery on the mannan-binding lectin pathway of innate immunity.
Clin Exp Immunol 2006;144:239.
57. Wu FP, Hoekman K, Sietses C, et al. Systemic and peritoneal an-
giogenic response after laparoscopic or conventional colon resec-
tion in cancer patients: A prospective, randomized trial. Dis
Colon Rectum 2004;47:1670.
58. Hildebrandt U, Kessler K, Plusczyk T, et al. Comparison of sur-
gical stress between laparoscopic and open colonic resections.
Surg Endosc 2003;17:242.
59. Hildebrandt U, Kessler K, Pistorius G, et al. Granulocyte elas-
tase and systemic cytokine response after laparoscopic-assisted
and open resections in Crohn’s disease. Dis Colon Rectum
1999;42:1480.
60. Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of re-
ports of randomized clinical trials: Is blinding necessary? Con-
trol Clin Trials 1996;17:1.
61. Demets DL. Methods for combining randomized clinical trials:
Strengths and limitations. Stat Med 1987;6:341.
62. Riese J, Niedobitek G, Lisner R, et al. Expression of interleukin-6 and
monocyte chemoattractant protein-1 by peritoneal sub-mesothelial
cells during abdominal operations. J Pathol 2004;202:34.
63. Riese J, Schoolmann S, Beyer A, et al. Production of IL-6 and
MCP-1 by the human peritoneum in vivo during major abdomi-
nal surgery. Shock 2000;14:91.
 SAMMOUR ET AL.: A SYSTEMATIC REVIEW AND META-ANALYSIS
64. Greca FH, Souza Filho ZA, Giovanini A, et al. Interleukin-6 (IL-
6) influence on colonic anastomosis healing in rats. Acta Cir Bras
2007;22:110.
65. Webb N, Bottomley M, Watson C. Vascular endothelial growth
factor (VEGF) is released from platelets during blood clotting:
Implications for measurement of circulating VEGF levels in clin-
ical disease. Clin Sci 1998;94:395.
66. Hill AG, Connolly AB. Minimal access colorectal surgery: Is it
truly minimally invasive? Dis Colon Rectum 2006;49:144.
67. Mehigan BJ, Hartley JE, Drew PJ, et al. Changes in T cell subsets,
interleukin-6 and C-reactive protein after laparoscopic and open
colorectal resection for malignancy. Surg Endosc 2001;15:1289.
68. Fukushima R, Kawamura YJ, Saito H, et al. Interleukin-6 and
stress hormone responses after uncomplicated gasless
laparoscopic-assisted and open sigmoid colectomy. Dis Colon
Rectum 1996;39:S29.
69. Kehlet H. Manipulation of the metabolic response in clinical
practice. World J Surg 2000;24:690.
70. Schwenk W, Bohm B, Muller JM. Postoperative pain and fatigue
after laparoscopic or conventional colorectal resections. A pro-
spective randomized trial. Surg Endosc 1998;12:1131.
71. Weeks JC, Nelson H, Gelber S, et al. Short-term quality-of-life
outcomes following laparoscopic-assisted colectomy vs open
colectomy for colon cancer: A randomized trial. JAMA 2002;
287:321.
72. Hewett PJ, Allardyce RA, Bagshaw PF, et al. Short-term out-
comes of the Australasian randomized clinical study comparing
laparoscopic and conventional open surgical treatments for co-
lon cancer: The ALCCaS trial. Ann Surg 2008;248:728.
73. Poole G, Ooi S, Scott S, et al. How much has the introduction of
laparoscopic surgery changed open surgery? N Z Med J 2003;
116:U518.
74. Basse L, Jakobsen DH, Bardram L, et al. Functional recovery af-
ter open versus laparoscopic colonic resection: A randomized,
blinded study. Ann Surg 2005;241:416.
75. Wichmann MW, Eben R, Angele MK, et al. Fast-track rehabilita-
tion in elective colorectal surgery patients: A prospective clinical
and immunological single-centre study. ANZ J Surg 2007;77:502.
37
76. Riese J, Schoolmann S, Denzel C, et al. Effect of abdominal infec-
tions on peritoneal and systemic production of interleukin 6 and
monocyte chemoattractant protein-1. Shock 2002;17:361.
77. Rosario MT, Ribeiro U Jr, Corbett CE, et al. Does CO2 pneumo-
peritoneum alter the ultra-structure of the mesothelium? J Surg
Res 2006;133:84.
78. Tahara K, Fujii K, Yamaguchi K, et al. Increased expression of
P-cadherin mRNA in the mouse peritoneum after carbon dioxide
insufflation. Surg Endosc 2001;15:946.
79. Suematsu T, Hirabayashi Y, Shiraishi N, et al. Morphology of
the murine peritoneum after pneumoperitoneum vs. laparot-
omy. Surg Endosc 2001;15:954.
80. Neuhaus SJ, Gupta A, Watson DI. Helium and other alternative
insufflation gases for laparoscopy. Surg Endosc 2001;15:553.
81. Moehrlen U, Ziegler U, Boneberg E, et al. Impact of carbon diox-
ide versus air pneumoperitoneum on peritoneal cell migration
and cell fate. Surg Endosc 2006;20:1607.
82. Hanly EJ, Aurora AA, Shih SP, et al. Peritoneal acidosis mediates
immunoprotection in laparoscopic surgery. Surgery 2007;142:357.
83. Hanly EJ, Aurora AR, Fuentes JM, et al. Abdominal insufflation
with CO2 causes peritoneal acidosis independent of systemic pH.
J Gastrointest Surg 2005;9:1245. discussion 1251.
84. Neuhaus SJ, Watson DI. Pneumoperitoneum and peritoneal
surface changes: A review. Surg Endosc 2004;18:1316.
85. Sammour T, Kahokehr A, Hill A. The effect of warmed, hu-
midified insufflation on postoperative pain in laparoscopy –
A meta-analysis of randomized controlled trials. Br J Surg
2008;95:950.
86. Koster S, Spacek Z, Paweletz N, et al. [A scanning microscopy
study of the peritoneum in mice after application of a CO2-
pneumoperitoneum]. Zentralbl Gynakol 1999;121:244.
87. Erikoglu M, Yol S, Avunduk MC, et al. Electron-microscopic
alterations of the peritoneum after both cold and heated carbon
dioxide pneumoperitoneum. J Surg Res 2005;125:73.
88. Puttick MI, Scott-Coombes DM, Dye J, et al. Comparison of
immunologic and physiologic effects of CO2 pneumoperito-
neum at room and body temperatures. Surg Endosc 1999;
13:572.