Clinical Investigations

Inflammatory cytokine response in patients with septic shock

secondary to generalized peritonitis
Florence C. Riché, MD; Bernard P. Cholley, MD, PhD; Yves H. Panis, MD, PhD; Marie-
Josèphe C. Laisné, MD; Claudette G. Briard, MD; Anne-Marie Graulet, MS; Jean L. Gué
ris, MD; Patrice D. Valleur, MD
From the Departments of Anesthesiology (Drs. Riché, Cholley, Laisné, and Briard),
Surgery (Drs. Panis and Valleur), and Radio-Immunology (Drs. Graulet and Guéris), Hôpital
Lariboisière, Paris, France.
Supported, in part, by Assistance Publique-Hôpitaux de Paris.

Objectives: The aims of this study were
the following: a) to assess the
proinflammatory cytokine (tumor
necrosis factor [TNF]- α , interleukin
[IL]-1, and IL-6) response in patients with
septic shock secondary to generalized
peritonitis; and b) to evaluate the
influence of bacteremic status, type of
peritonitis (acute perforation or
postoperative), and peritoneal microbial
status (mono- or polymicrobial) on
cytokine expression and mortality.
Design: Prospective study.
Setting: Surgical intensive care unit of a
university hospital.
Patients: Fifty-two consecutive patients
with septic shock caused by generalized
peritonitis.
Interventions: Routine blood tests, blood
cultures, and cytokine assays were
performed during the first 3 days after
onset of shock.
Measurements and Main Results: Serum
TNF- α and IL-6 concentrations were
measured by using a radioimmunoassay,
and IL-1 concentrations were measured
by using ELISA. Median serum
concentrations on day 1 were: TNF-α, 90
pg/mL; IL-1, 7 pg/mL; and IL-6, 5000
pg/mL. TNF-α and IL-6 concentrations
decreased significantly between the first
and third days of septic shock (p = .0001),
whereas IL-1 concentrations remained
low. The decrease in IL-6 tended to be
more pronounced in the survivors group
(p = .057). Median TNF- α serum
concentrations were higher in bacteremic
compared with nonbacteremic patients
(151 vs. 73 pg/mL, p = .003). TNF-α, IL-1,
and IL-6 serum concentrations and
mortality were not different between
acute perforation vs. postoperative
peritonitis and mono- versus
polymicrobial peritonitis.
Conclusions: The systemic release of
TNF- α and IL-6 during septic shock
caused by generalized peritonitis was
maximal on day 1 and decreased rapidly
during the next days. No systemic release
of IL-1 was observed. IL-6 serum
concentrations remained higher in
patients who subsequently died. Among
the different features of peritonitis
studied, only bacteremia influenced the
systemic cytokine response (higher TNF-
α).
KEY WORDS: tumor necrosis factor- α ;
interleukin-1; interleukin-6; cytokine;
septic shock; peritonitis; Infection;
Cytokine serum concentrations
observed during infectious processes are
determined, in part, by the primary site of
infection. Experimentally, the magnitude of
the rise in tumor necrosis factor (TNF)-α
serum concentration secondary to peritonitis
is a hundred times less important than the
rise observed after an intravenous septic
challenge (1-3). This differential pattern of
response underlines the importance of the
body "compartment" infected in determining
the TNF-α serum profile. Clinical studies
investigating the cytokine response to septic
shock generally have enrolled patients with
various causes of infection (4-7). Those
focusing on intra-abdominal infections (not
only peritonitis) reported an elevation in
TNF- α and interleukin (IL)-6 serum
concentrations regardless of the presence of
shock and found a correlation between
cytokine serum concentrations (TNF-α and
IL-6) and patient mortality (8-11). However,
several studies found that a high TNF-α
concentration was associated with increased
survival (12-14).
Little information from
homogeneous cohorts of patients is available
regarding the cytokine response to septic
shock secondary to peritonitis. In addition,
the influence of microbiological
characteristics of the peritonitis on cytokine
response is unknown. Therefore, the aims of
this prospective study were to assess the
cytokine response (TNF-α, IL-1, and IL-6)
to septic shock secondary to peritonitis; and
to evaluate the influence of bacteremic
status, type of peritonitis (acute perforation
or postoperative), and peritoneal microbial
status (mono- or polymicrobial) on cytokine
expression and survival.
PATIENTS AND METHODS
surgery; bacteremia
The study protocol was approved by
the hospital committee for human
investigation. Because cytokine
measurements were performed on "leftover"
routine blood samples, the need for
informed consent was waived.
From January 1992 to September
1996, we studied 52 consecutive patients
with septic shock caused by generalized
peritonitis. Among them, 32 patients were
part of a previous study evaluating the
prognostic value of TNF- α serum
concentrations during septic shock
secondary to various abdominal infections
(12). Patients were entered into the present
study when they had both surgically proven
generalized peritonitis and septic shock as
defined by Bone et al. (15). The criteria for
sepsis included fever (temperature ≥38.3°
C) or hypothermia (temperature <36° C),
tachycardia (>90 beats/min), tachypnea (>20
breaths/min), white blood cell
count >12,000 or <4,000 109/L, and clinical
suspicion of infection. Shock was defined as
hypotension (systolic blood pressure <90
mmHg or a decrease in systolic blood
pressure ≥ 40 mmHg) in the absence of
other causes for hypotension,
unresponsiveness to intravenous fluids, or
the need for vasopressor agents and
perfusion abnormalities. Perfusion
abnormalities leading to organ dysfunction
may include hypoxemia (PaO2 <75 torr
while breathing room air or PaO2/FIO2
<250), oliguria (urine output <30 mL/hr or
0.5 mL/kg/hr), unexplained metabolic
acidosis (or high plasma lactate), or a recent
change in mental status. When the primary
site of infection was not a generalized
peritonitis (i.e., localized abscess,
cholangitis), patients were not considered
for the study.
 All patients were evaluated using the
Simplified Acute Physiologic Score (SAPS)
II (16) on day 1. From the first day of septic
shock to day 3, blood was withdrawn daily
through an indwelling arterial catheter for
routine laboratory tests (white blood cell and
platelet counts, blood gases, serum
electrolytes, liver and kidney function tests).
Systematic blood cultures were performed at
the time of admission and every 8 hrs (as it
is part of the standard care for such patients
in our institution) and during each episode
of shivering, peak hyperthermia (≥38.3°
C), or hypothermia (<36 ° C). Leftover
blood samples for routine tests were used for
TNF-α, IL-1, and IL-6 measurements. IL-1
and IL-6 serum concentration
determinations were obtained in only 26 and
41 patients, respectively, because these
assays were not available at the beginning of
the study. Blood samples were centrifuged
at 3000 rpm for 10 min, and the serum was
stored at -20°C until tested. TNF-α and
IL-6 concentrations were measured by using
a radioimmunoassay, and IL-1
concentrations were measured by using
ELISA. The sensitivity thresholds of these
assays were 5 pg/mL for TNF-α and IL-1
and 10 pg/mL for IL-6.
Statistical Analysis
Serum cytokine profiles during the
first 3 days were analyzed by using
Friedman repeated-measures analysis of
variance on ranks. Differences in profiles
between survivors and nonsurvivors were
tested by using two-way analysis of variance
on the log-transformed values of cytokine
serum concentrations. The Mann-Whitney
rank sum test was used to compare values on
day 1 between subgroups of patients
(bacteremic versus nonbacteremic, acute
perforation versus postoperative,
monoversus polymicrobial peritonitis).
RESULTS
Clinical and Biological Data
Fifty-two consecutive patients (21
women and 31 men) with septic shock
caused by generalized peritonitis were
studied. Their mean age was 67 ± 16 years
(range, 19-92 years). Onset of septic shock
occurred in the immediate perioperative
period, ranging from 12 hrs preoperatively
to 24 hrs postoperatively. All patients were
hypotensive despite adequate fluid
resuscitation and required a norepinephrine
infusion to maintain the mean arterial blood
pressure >60 mmHg. The mean duration of
norepinephrine infusion was 6 ± 5 days. In
addition, 31 of 52 patients received low-
dose dopamine (3-5 μg/kg/min) for 6 ± 8
days. Continuous arteriovenous
hemofiltration was required in 11 patients.
The mean peak lactate concentration was 6.1
± 5.2 mmol/L (range, 1.7-29.5 mmol/L).
The mean (± SD) SAPS II score
was 52 ± 18 (range, 23-103). The median
duration of hospital stay was 24 days (range,
1-214 days). Selected clinical and biological
findings are listed in Table 1.
Table 1. Causes of generalized peritonitis
All patients had secondary peritonitis,
including 37 acute perforations and 15
postoperative peritonitis caused by
anastomotic leaks. Acute perforation
peritonitis was caused by colonic perforation
in 19 cases (6 diverticulitis, 5 colonic
carcinoma, 2 colonic necrosis, 3
colonoscopy, and 3 fecaloma), duodenal
ulcer perforation (n = 7), small bowel
perforation (n = 2), biliary peritonitis (n = 5),
and miscellaneous causes (n = 4). All acute
perforations peritonitis was "community-
acquired," as opposed to postoperative
peritonitis, which was "hospital-acquired."
The mortality rate was 15/37 (40%) for the
acute perforation and 7/15 (46%) for the
postoperative peritonitis (p = NS).
Bacteriologic Data
Peritoneal liquid was polymicrobial (n
= 37), monomicrobial (n = 11), or sterile (n
= 4). Bacteriologic data from the peritoneum
are shown in Table 2. There was no
difference in mortality between patients with
poly-versus monomicrobial peritonitis.
Table 2. Bacterial isolates from the
peritoneal liquid of 52 patients with
generalized peritonitis
52) 90 pg/mL (range, 6-4663 pg/mL); IL-1
(n = 26) 7 pg/mL (range, 0-66 pg/mL); IL-6
(n = 40) 5000 pg/mL (range, 44-5100
pg/mL). The serum cytokine concentration
courses during the first 3 days of septic
shock are shown in Figure 1. TNF-α and
IL-6 concentrations decreased significantly
between the first and third days of septic
shock, whereas IL-1 remained unchanged.
There were no difference between survivors
and nonsurvivors with respect to TNF- α
and IL-1 concentrations. In contrast, IL-6
serum concentrations tended to remain
higher in the nonsurvivors group (p = .057).

Of the 52 patients, 12 (23%) were

bacteremic: 5 of 22 in the nonsurvivors
group vs. 7 of 30 in the survivors group (p =
NS). Microorganisms identified in blood
cultures included Escherichia coli (n = 4),
streptococcus (n = 4), anaerobes (n = 4),
Pseudomonas (n = 2), Acinetobacter
baumanni (n = 1), and Candida (n = 1); they
were also present in the peritoneal cavity in
50% of cases. Among the 12 bacteremic
patients, 4 had polymicrobial blood cultures.
Cytokine Assay Figure 1. Box plots showing the profiles of
cytokine serum concentrations during the
The average time delay between the first 3 days after the onset of septic shock
onset of hypotension and the initiation of caused by peritonitis. Top, tumor necrosis
data collection was 5 hrs (range, 10 mins-12
factor- α (TNF- α ) (n = 52, p < .001);
hrs). The median cytokine serum
middle, interleukin-1 (IL-1) (n = 26, p = NS);
concentrations on day 1 were: TNF-α (n =
bottom, interleukin-6 (IL-6) (n = 40, p
< .001). 25th and 75th percentiles are
represented by a bar centered about the
median; 10th and 90th percentiles are error
bars; and data points beyond 10th or 90th
percentiles are dots.
Median TNF-α serum concentrations
were significantly (p = .003) higher in
bacteremic patients (n = 12) compared with
nonbacteremic patients (n = 40): 151 pg/mL
(range, 59-4663 pg/mL) versus 73 pg/mL
(range, 6-275 pg/mL), respectively (Fig. 2).
No difference was observed for IL-1 and IL-
6 serum concentrations in bacteremic versus
nonbacteremic patients. There was no
difference for TNF-α, IL-1, and IL-6 serum
concentrations in acute perforation versus
postoperative peritonitis and in
monomicrobial versus polymicrobial
peritonitis.
Figure 2. Box plot showing tumor necrosis
factor-α (TNF-α) serum concentrations in
bacteremic (n = 12) and nonbacteremic (n =
40) patients on day 1. p = .003.
DISCUSSION
The aim of this study was to assess the
inflammatory cytokine response in a
homogeneous group of patients with septic
shock caused by generalized peritonitis.
There was a systemic release of IL-6 and
TNF- α in these patients, whereas IL-1
barely increased. The serum concentrations
of TNF-α and IL-6 decreased significantly
between the first and third days of septic
shock, whereas IL-1 remained unchanged.
IL-6 concentrations tended to remain higher
in nonsurvivors than in survivors during the
first 3 days after the onset of shock, whereas
there was no difference for TNF-α and IL-1
concentrations. Higher TNF- α serum
concentrations were measured in bacteremic
versus nonbacteremic patients. No
difference in serum cytokine concentrations
and hospital mortality was observed
between monoversus polymicrobial and
acute perforation versus postoperative
peritonitis.
We observed large variability in serum
cytokine concentrations among patients.
Such variability has also been noted in other
studies and may be attributed, in part, to
factors related to the measurement technique,
the time of blood sampling within the time
course of the disease (17), and host
susceptibility (18). All measurements were
performed in duplicate, and both intra- and
interassay variability were <7%. Values for
TNF- α and IL-6 in nonseptic surgical
patients are normally very low (TNF-α <20
pg/mL, IL-6 <50 pg/mL) and remain
unaltered during a 5-day period after
abdominal surgery for cancer in the absence
of postoperative complications (19). IL-1
values in healthy volunteers are below the
threshold of detection (<5 pg/mL).
Host defense against infection of the
peritoneal cavity is responsible for a local
acute-phase response involving cytokine
secretion by macrophages and other
leukocytes (20-23). Early release of TNF-α
and IL-1 triggers the secondary production
of IL-6 and IL-8 (1, 24). These molecules
modulate the local inflammatory response
aimed at fighting peritoneal infection.
Several studies have pointed out that the
primary site of infection is a major
determinant of the systemic cytokine
response (25, 26). During peritonitis,
cytokine serum concentrations are thought
to reflect, in part, the intraperitoneal
production (11, 27). High concentrations of
TNF-α and IL-6 have been measured in
ascites during primary peritonitis (i.e.,
spontaneous bacterial peritonitis or after
peritoneal dialysis), whereas plasma
concentrations obtained simultaneously
were only slightly increased (28-30).
However, little information is available
regarding cytokine serum profiles in the
acute setting of patients with septic shock
secondary to acute perforation or
postoperative generalized peritonitis. In the
present study, we measured high and
transient increases in TNF- α and IL-6
serum concentrations on day 1, whereas IL-
1 was moderately increased. The
proinflammatory TNF-α and IL-6 quickly
decreased during the first 3 days after the
onset of septic shock, which suggests some
control mechanism down-regulating the
initial triggers of inflammation. IL-1 serum
concentrations remained constant during this
initial phase of shock. IL-6 concentrations
tended to decrease less in patients who
subsequently died than in survivors. This
was also noted by investigators studying
sepsis from various intra-abdominal origins
and seems to indicate a bad prognosis value
for patients with persistently high IL-6
concentrations (10, 11). No such difference
between survivors and nonsurvivors was
observed in the first 3 days for TNF-α and
IL-1 concentrations, which confirms
findings by Patel et al. (10).
TNF- α serum concentrations were
significantly higher in bacteremic versus
nonbacteremic patients. This is in
accordance with experimental observations
in which the rise in TNF- α serum
concentrations were 50-100 times greater
after intravenous bacterial challenges
compared with intraperitoneal inoculation
(1-3). It therefore seems crucial to take into
account which compartment is involved
with the microbiological insult before
interpreting serum cytokine profiles. This is
illustrated by the fact that TNF-α antibody
is more efficient in reducing mortality when
animals are challenged intravenously rather
than intraperitoneally (3).
We did not observe any difference in
the serum cytokine concentrations between
acute perforation and postoperative
peritonitis. In our cohort, postoperative
peritonitis was not associated with a worse
prognosis. This last feature is still debated,
and controversy exists regarding the
pejorative character of postoperative
peritonitis (31-34). Similarly, cytokine
expression and mortality rates were not
different between patients with monoversus
polymicrobial peritonitis. However, other
authors have suggested that the
polymicrobial character of intraabdominal
infection could increase the severity of
peritonitis via a synergistic effect of
microbial association and thus worsen
outcome (35-37).
Although excessive blood
concentrations of TNF- α may be
detrimental, intraperitoneal TNF- α is
probably very useful in the process of
defense against infection (38, 39). This is
corroborated by the fact that the
intraperitoneal administration of TNF- α
antibodies increased mortality in
experimental peritonitis, and this was
reversed by intraperitoneal TNF- α
administration (1-3). Mechanisms by which
TNF- α could have beneficial effects in
peritonitis include the induction of
peritoneal inflammation, increase in
capillary permeability, local recruitment of
neutrophils, stimulation of bacterial
phagocytosis, and formation of fibrin
adhesions (2, 24, 40).
To summarize, the findings of the
present study indicate that the onset of septic
shock secondary to generalized peritoneal
infection was accompanied by a systemic
release of TNF- α , IL-1, and IL-6. This
increase was transient for TNF-α and IL-6
in patients who survived. Further studies are
required to establish the pejorative
prognostic value of persistent elevated IL-6
concentrations during generalized peritonitis.
Bacteremic patients had significantly higher
systemic release of TNF- α , a finding of
importance if trials for the therapeutic use of
TNF-α antibody are considered.
REFERENCES
1. Bagby G, Plessala K, Wilson L, et al:
Divergent efficacy of antibody to
tumor necrosis factor-alpha in
intravascular and peritonitis models of
sepsis. J Infect Dis 1991; 163: 83-88
2. Echtenacher B, Falk W, Männel D, et
al: Requirement of endogenous tumor
necrosis factor/cachectin for recovery
from experimental peritonitis. J
Immunol 1990; 145: 3762-3766
3. Zanetti G, Heumann D, Gérain J, et al:
Cytokine production after intravenous
or peritoneal gram-negative bacterial
challenge in mice. J Immunol 1992;
148:1890-1897
4. Cannon J, Tompkins R, Gelfand J, et
al: Circulating interleukin-1 and tumor
necrosis factor in septic shock and
experimental endotoxin fever. J Infect
Dis 1990; 161:79-84
5. Debets J, Kampmeijer R, van der
Linden M, et al: Plasma tumor
necrosis factor and mortality rate in
critically ill septic patients. Crit Care
Med 1989; 17:489-494
6. Marks J, Berman-Marks C, Luce J, et
al: Plasma tumor necrosis factor in
patients with septic shock: mortality
rate, incidence of adult respiratory
distress syndrome, and effects of
methylprednisolone administration.
Am Rev Respir Dis 1990; 141:94-97
7. Calandra T, Baumgartner J, Grau G, et
al: Prognostic values of tumor necrosis
factor/cachectin, interleukin-1,
interferon-alpha, and interferon-
gamma in the serum of patients with
septic shock. J Infect Dis 1990;
161:982-987
8. Damas P, Reuter A, Gysen P, et al:
Tumor necrosis factor and interleukin-
1 serum levels during severe sepsis in
humans. Crit Care Med 1989; 17:975-
978
9. Függer R, Zadrobilek E, Götzinger P,
et al: Perioperative TNF and IL-6
concentrations correlate with septic
state, organ function, and APACHE II
scores in intra-abdominal infection.
Eur J Surg 1993; 159:525-529
10. Patel R, Deen KI, Youngs J, et al:
Interleukin-6 is a prognostic indicator
of outcome in severe intra-abdominal
sepsis. Br J Surg 1994; 81:1306-1308
11. Holzheimer R, Schein M, Wittmann D:
Inflammatory response in peritoneal
exudate and plasma of patients
undergoing planned relaparotomy for
severe secondary peritonitis. Arch
Surg 1995; 130:1314-1319
12. Riché F, Panis Y, Laisn é M, et al:
High tumor necrosis factor serum level
is associated with abdominal septic
shock: A prospective study in 59
patients. Surgery 1996; 120:801-7
13. Hamilton G, Hofbauer S, Hamilton B:
Endotoxin, TNF-alpha, interleukin-6
and parameters of the cellular immune
system in patients with intraabdominal
 sepsis. Scand J Infect Dis 1992;
24:361-368
14. Rogy M, Oldenburg H, Coyle S, et al:
Correlation between acute physiology
and chronic health evaluation
(APACHE) III score and
immunological parameters in critically
ill patients with sepsis. Br J Surg 1996;
83:396-400
15. Bone R, Balk R, Cerra F, et al:
Definition for sepsis and organ failure
and guidelines for the use of
innovative therapies in sepsis: the
ACCP/SCCM Consensus conference
committee. Chest 1992; 101:1644-
1661
16. Legall J, Lemeshow S, Saulnier F:
New simplified acute physiology score
(SAPS II) based on a European/North
American multicenter study. JAMA
1993; 270:2957-2963
17. Martich G, Danner R, Ceska M, et al:
Detection of interleukin 8 and tumor
necrosis factor in normal humans after
intravenous endotoxin: the effect of
antiinflammatory agents. J Exp Med
1991; 173:1021-1024
18. Yoshimoto T, Nakanishi K, Hirose S,
et al: High serum IL-6 level reflects
susceptible status of the host to
endotoxin and IL-1/tumor necrosis
factor. J Immunol 1992; 148: 3596-
3603
19. Riché F, Dosquet C, Panis Y, et al:
Levels of portal and systemic blood
cytokines after colectomy in patients
with carcinoma or Crohn's disease. J
Am Coll Surg 1995; 180:718-724
20. McMasters KM, Cheadle WG:
Regulation of macrophage TNF alpha,
IL-1 beta, and Ia (I-A alpha) mRNA
expression during peritonitis is site
dependent. J Surg Res 1993; 54:426-
430
21. Fieren M, Van den Bemd G, Bonta I,
et al: Peritoneal macrophages from
patients on continuous ambulatory
peritoneal dialysis have an increased
capability to release tumor necrosis
factor during peritonitis. J Clin Lab
Immunol 1991; 31:1-9
22. Betjes M, Tuk C, Struijk D, et al:
Interleukin 8 production by human
peritoneal mesothelial cells in response
to tumor necrosis factor alpha, and
interleukin conditioned by
macrophages cocultured with
Staphylococcus epidermidis. J Infect
Dis 1993; 168:1202-1210
23. Holzheimer R, Muhrer K, L'Allemand
N, et al: Intraabdominal infections:
classification, mortality, scoring and
physiopathology. Infection 1991;
19:447-452
24. Zemel D, Koomen G, Hart A, et al:
Relationship of TNF-alpha,
interleukin-6, and prostaglandins to
peritoneal permeability for
macromolecules during longitudinal
follow-up of peritonitis in continuous
ambulatory peritoneal dialysis. J Lab
Clin Med 1993; 122: 686-696
25. Nieuwenhuijzen GA, Knapen MF,
Hendriks T, et al: Elimination of
various subpopulations of
macrophages and the development of
multiple-organ dysfunction syndrome
in mice. Arch Surg 1997; 132:533-539
26. Schein M, Wittmann D, Holzheimer R,
et al: Hypothesis:
compartimentalization of cytokines in
intraabdominal infection. Surgery
1996; 119:694-700
27. Gau-Jun T, Cheng-Deng K, Tzu-Chen
Y, et al: Perioperative plasma
concentrations of tumor necrosis factor
alpha and interleukin-6 in infected
patients. Crit Care Med 1996; 24:423-
428
28. Propst T, Propst A, Herold M, et al: 37. Dupont H, Montravers P, Mohler J, et
Spontaneous bacterial peritonitis is al: Disparate findings on the role of
associated with high levels of virulence factors of enterococcus
interleukin-6 and its secondary faecalis in mouse and rat models of
mediators in ascitic fluid. Eur J Clin peritonitis. Infect Immun 1998;
Invest 1993; 23:832-6 66:2570-2575
29. Zeni F, Tardy B, Vindimian M, et al: 38. Shapiro L, Gelfand J: Cytokines and
High levels of tumor necrosis factor- sepsis: pathology and therapy. New
alpha and interleukin-6 in the ascitic Horiz 1993; 1:13-22
fluid of cirrhotic patients with 39. Natanson C, Hoffman W, Suffredini A,
spontaneous bacterial peritonitis. Clin et al: Selected treatment strategies for
Infect Dis 1993; 17:218-223 septic shock based on proposed
30. Zemel D, Krediet R, Koomen G: mechanisms of pathogenesis. Ann
Interleukin-8 during peritonitis in Intern Med 1994; 120:771-783
patients treated with CAPD: an in vivo 40. Whawell S, Scott-Commbes M,
model of acute inflammation. Nephrol Vipond S, et al: Tumor necrosis factor-
Dial Transplant 1994; 9:169-174 mediated release of plasminogen
31. Bohnen J, Boulanger M, Meakins J, et activator inhibitor 1 by human
al: Prognosis in generalized peritonitis: peritoneal mesothelial cells. Br J Surg
Relation to cause and risk factors. 1994; 81:214-216
Arch Surg 1983; 118:285-290
32. Dellinger E, Wertz M, Meakins J:
Surgical infection stratification system
for intra-abdominal infection. Arch
Surg 1985; 120:21-29
33. Bohnen J, Mustard R, Oxholm S, et al:
APACHE II score and abdominal
sepsis: a prospective study. Arch Surg
1988; 123: 225-229
34. McLaughan G, Anderson I, Graut I, et
al: Outcome of patients with
abdominal sepsis treated in an
intensive care unit. Br J Surg 1995;
82:524-529
35. Onderdonk A, Bartlett J, Louie T, et al:
Microbial synergy in polymicrobial
surgical infections. Rev Infect Dis
1976; 7:151-170
36. Burnett R, Haverstock D, Dellinger P,
et al: Definition of the role of
enterococcus in intrabdominal
infection: analysis of a prospective
randomized trial. Surgery 1995;
118:717-721