вход по аккаунту



код для вставкиСкачать
Norbert Runkel
Department of Surgery. Benjamin Franklin
Medical Center. Free University of Berlin.
DigSurg 1996:13:269-272
Bacterial Translocation in
Acute Pancreatitis
K e y W o rds
A b s tra c t
Bacterial translocation
Acute pancreatitis
Understanding of the mechanisms for the development of sepsis during acute
pancreatitis has increased greatly over the last 5 years. It has become clear that
the gut serves as a source of bacteria in a variety of animal models of acute
pancreatitis. Bacterial translocation from the intestinal lumen to extraintestinal sites including pancreatic necrosis may occur via the lymphatic, hematoge­
nous or direct transperitoneal route. Increased mucosal permeability, reduced
bowel motility, and an impaired host defense have been identified as possible
causes for translocation. Standard prophylactic management for the pre­
vention of sepsis in acute pancreatitis includes the use of systemic broad-spec­
trum antibiotics with optimal penetration into the pancreas, such as imipenem. Hou'ever, selective gut decontamination clearly reduces the incidence of
pancreatic infection in animal models and, probably, in humans, too.
In tro d u ctio n
The mortality rate of acute necrotizing pancreatitis
remains high and is reported to be between 10 and more
than 50% [1-3]. Infection of pancreatic or peripancreatic
necrosis has emerged as the most serious complication of
acute pancreatitis and accounts for more than 80% of
deaths [4], As the prevention of sepsis has become a criti­
cal issue in the management, research into the mecha­
nisms for the development of sepsis has expanded over
the last 5 years. In the 1980s it became clear that the inci­
dence of pancreatic infection is related to the severity of
the disease and the extent of pancreatic necrosis [5]. The
microorganisms causing pancreatic infection were shown
to be common enteric bacteria [6]. We know now that sec­
ondary pancreatic infection most likely occurs several
days after the onset of the disease and is usually preceded
by an early septic-shock-like course without evidence of a
septic focus (systemic inflammatory response syndrome).
\I\V J
If-Mail kargcr(o »
Fax +4! 61 306 12 34
© 1996 S. Karger AG. Basel
0253-4886/96/0135-0269$ 10.00/0
This knowledge led to the concept of the gut as a potential
septic focus and a source of bacteria in acute pancreatitis.
This concept of gut-derived sepsis is supported by some
new and exciting data from experimental research.
B a c te ria l T ra n slo ca tio n fro m th e G u t
Detailed studies on the development of sepsis in acute
pancreatitis began in the late 1980s when pancreatic
infection was regarded as occurring infrequently in ani­
mals. Prior observations found rates of less than 5% in the
choline-deficient ethionine-supplcmented (CDE) diet
model in mice [7], of 50% in bile duct ligation pancreatitis
[8] and of 100% in the closed-duodenal-loop model [9].
We were the first to study bacterial dissemination and sys­
temic pancreatic infection in opossum [10. 11] and rat
[12] pancreatitis. Biliary pancreatitis in the opossum, a
North American carnivore, significantly increased the
PD Dr. N. Runkel
Abteilung für Allgemein-. Thorax- und Gcfasschirurgie
Universitätsklinikum Benjamin Franklin
Freie Universität Berlin. Hindenburgdamm 30
D-122(H) Berlin (Germany)
Downloaded by:
Kings's College London - 10/25/2017 7:57:11 PM
Experimental model
Common channel ligation
CDE diet
Pancreatic head ligation
Ductal glycodeoxycholate perfusion
+ intravenous prostaglandin Ei
Intraductal glycodeoxycholate
+intravenous caerulein
Runkel ct al., 1990 [10]
Runkel et al.. 1995 [11]
Runkel ct al., 1991 [12]
Tarpilact al., 199.3 [16]
Kazantsev ct al.. 1994 [17]
Isaji et al.. 1992 [31]
Gianottiet al.. 1995 [24]
Med ich et al.. 1993 [14]
Gianotti et al.. 1993 [15]
Gianotti et al.. 1993 [15]
Widdison et al.. 1994 [18]
Foitzik et al., 1994 [13]
Intraductal taurocholate
bacterial contamination rate of intestinal lymph nodes
(50 vs. 23% in controls), blood (70 vs. 10%) and the pan­
creas (40 vs. 0%). Biliary pancreatitis in rats also pro­
moted bacterial translocation into the mesenteric lymph
nodes (100 vs. 21% in controls) and to systemic sites (33
vs. 0%). This pattern of spread and the identification of
enteric microorganisms clearly defined the origin of these
bacteria, namely the intestinal tract. Others have since
confirmed our observation in various animal models (ta­
ble 1). Bacterial translocation describes this process of
bacterial migration from the gut lumen to extraintestinal
sites and is recognized as a potential cause of sepsis in
various experimental conditions. There should be little
doubt today that pancreatic infection in acute experimen­
tal pancreatitis is caused by bacterial translocation.
The M e c h a n ism o f B a c te ria l T ra n s lo c a tio n
Bacterial translocation to the pancreas is an early event
in experimental pancreatitis and reaches its maximum in
most models within 3 days after the insult. The preva­
lence of pancreatic infection increased from 33 in the first
24 h to 78% between 48 and 76 h in a combined glycodeoxycholatc/caerulein rat pancreatitis model used by
Foitzik et al. [ 13] in 1994. They could also clearly demon­
strate that the severity of pancreatitis correlated with the
frequency of translocation as well as the number of trans­
locating bacteria.
The pathways of bacterial spread are largely unknown.
But the pattern of colonization of organs or of the distri­
bution of fluorescent beads [14] gave some indirect evi­
dence for a possible lymphatic, hematogenous or trans­
Dig Surg 1996:13:269-272
peritoneal route. It is important to note that in different
models, different pathways of bacterial translocation are
preferred. In our biliary' models, the primary route was
the lymphatic spread to the mesenteric lymph nodes, and
the hematogenous dissemination was a secondary phe­
nomenon [11, 12], The caerulein model, however, pro­
motes translocation transperitoneally [14] or via the
bloodstream [15], Intraductal injection of taurocholate in
rats or dogs is associated with the lymphatic pathway [16.
17], whereas the feline duct perfusion model again pro­
motes translocation via the hematogenous and direct way
[18], This was elegantly demonstrated when they enclosed
the colon in an impermeable plastic sack and thus pre­
vented translocation to the pancreas and, thus, pancreatic
This work from Reber’s group also provided data
regarding the site of translocation from the intestinal
tract. Separating the transverse colon from the pancreas
with an impermeable sack successfully interrupted transperitoneal bacterial translocation. Whether the colon is
the site of spread from the gut in other experimental mod­
els remains doubtful and unproven at this time.
The P a th o g e n e s is o f B a cte ria l T ra n s lo c a tio n
The last few years have also seen some progress regard­
ing the pathogenesis of bacterial translocation in acute
pancreatitis. Three principle causes of translocation are
discussed in the literature: disruption of the indigenous
intestinal microflora, loss of barrier function of the bowel
wall, and impairment of host defense. Each of these fac­
tors seems to play a role in acute pancreatitis.
R unkel
Downloaded by:
Kings's College London - 10/25/2017 7:57:11 PM
Table 1 . Bacterial translocation in
experimental acute pancreatitis
Loss o f Bowel Wall Barrier Function
The second principle cause of bacterial translocation is
a loss of mucosal barrier function resulting in a leakage of
bacteria and toxins out of the gut lumen. A number of
authors have recently demonstrated an increased intesti­
nal permeability for bacteria in acute pancreatitis using
radiolabelled Escherichia coli [15], plasmid-labelled E.
coli [17] or a clinical strain of E. coli [18]. The intestinal
barrier is also impaired for fluorescent beads [14]. The
mechanism of alteration of permeability is unknown.
Reduced gut motility may contribute to increased perme­
ability by prolonging the contact time between bowel con­
tent and mucosa and thus increasing the chance of per­
meation. Acute pancreatitis impairs intestinal microcir­
culation; however, it is speculative whether this results in
destruction of mucosal integrity. We have shown that
additional severe hypovolemic shock did not further pro­
mote bacterial translocation in rat biliary pancreatitis
[ 21].
Bacterial Translocation in Acute
Impairment o f Host Defense
Although there is no doubt that acute pancreatitis
influences the immune system, few studies have ad­
dressed this problem. The proportion of circulating levels
of CD4-positive lymphocytes was reduced in patients
with acute pancreatitis and correlated inversely with
translocating endotoxin [22]. Multiple skin tests revealed
a sustained anergic state in patients with pancreatic sepsis
[23]. Pancreatitis caused the failure of local pancreatic
and systemic clearance of bacteria in the caerulein model
and pancreatic head ligation model in rats [24], Widdison
et al. [25] observed an impairment of local clearence of E.
coli from the pancreas in Reber’s cat model, which could
be countered by levamisol, indicating a loss of phagocytic
function. Impaired bacterial killing ability may be an
important critical factor in the development of sepsis: by
failing to clear regional lymph nodes, a systemic inflam­
matory response may be induced.
The P re v e n tio n o f B a cte ria l T ra n slo c a tio n
The experimental data from the literature support the
concept of gut-derived sepsis in acute pancreatitis, al­
though the extent and route of bacterial spread may be
different in each model. Bacterial translocation may be
prevented by restoring gut integrity (enteral nutrition,
motility, biliopancreatic secretion) and more specifically
by inhibiting intestinal overgrowth with pathogens.
Whether an enterostomy has an effect on the indigenous
microfiora is unknown. The construction of a cecostomy
reduced the frequency of endotoxin detection from 22.6
to 5.6% and the mortality of rats with acute pancreatitis
from 23 to 17% [26]. Selective gut decontamination
(SDD) with oral antibiotics was shown to be highly effec­
tive in experimental studies [27-29] and clinical trials
[30], SDD reduced the prevalence of pancreatic infection
from 71 to 43% in the combined glycodeoxycholate/caerulein model in rats [27], This effect, however, was not as
good as that of imipenem alone (14%) or a combination of
SDD and imipenem (7%). The authors failed to observe a
significant improvement of early survival by any of the
antibiotic regimens. In contrast, Gianotti et al. [28] clearly
demonstrated the efficiency of SDD in the CDE model in
mice. The combination of polymyxin B, amikacin and
amphotericin B completely abolished intestinal growth
and translocation of aerobes and fungi, resulting in an
increase in survival from 20 to 90%. Similarly, bacterial
translocation was prevented and mortality of rats reduced
by intestinal lavage with saline containing rifaximin prior
Dig Surg 1996; 13:269-272
Downloaded by:
Kings's College London - 10/25/2017 7:57:11 PM
Disruption o f the Indigenous Intestinal Microflora
We have quantified the intestinal microfiora in rat bil­
iary pancreatitis. The most striking feature was an imbal­
ance in gram-negative and gram-positive aerobic bacteria
which progressed with time [12]. This shift in the intesti­
nal microfiora toward the pathogenic gram-negative aero­
bes was first noticed in the cecum as early as 2 days after
the induction of pancreatitis. A further 2 days later, the
indigenous flora was also significantly disrupted in the
ileum, jejunum and duodenum. In search of the causes for
the profund changes, we studied intestinal motility in rats
and discovered a dramatic prolongation of the gut transit
time. Fluorescently labelled dextrans pass two thirds of
the small bowel within 1 h in normal animals but need
about 4 h for the same distance in pancreatitic animals.
The causal connection between reduced gut motility, dis­
rupted intestinal microfiora and bacterial translocation
was proven in a subsequent experiment in rats using mor­
phine as a pharmacologic depressant of gut motility [19].
The concept was supported by Muncy et al. [20] who
examined the time course of these events in rat biliary
pancreatitis. Bowel motility was immediately impaired
after the onset of pancreatitis but normalized within 2-3
days. The disruption of intestinal microfiora began early
and reached its peak at 1-2 days. The enteric microfiora
was again restored by day 4. The prevalence of bacterial
translocation reached its maximum later at days 2-3 and
slowly decreased thereafter.
to induction of pancreatitis by intraductal enterokinase/
trypsin injection [29]. The first prospective randomized
multicenter trial on SDD in acute pancreatitis was suc­
cessfully completed in the Netherlands by Luitgen et al.
[30], They entered 102 patients with severe acute pancre­
atitis into the study (Imrie score >2; computed tomogra­
phy Balthazar grade D or E) between 1990 and 1993.
SDD consisted of an oral/rectal administration of colistin
sulfate, amphotericin, and norfloxacin plus short-term
systemic cefotaxim. SDD resulted in a reduction of pan­
creatic infections (18 vs. 38% of controls; p = 0.03), fre­
quency of laparotomies per patient (0.9 vs. 3.1; p < 0.05)
and mortality (22 vs. 35%; adjusted for Imrie score and
Balthazar grade: p < 0.048).
In summary, the concept of bacterial translocation as the
initial step in the development of pancreatic infection and
sepsis explains many findings in clinical acute pancreatitis.
This concept is strengthened by a series of recent animal
experiments. They have turned attention to the prevention
of sepsis by restoring intestinal integrity or by inhibiting
bacterial overgrowth. SDD has proven to be effective in
animals and in the first controlled clinical trial.
R e fe re n ces
2 Bradley EL 111: Operative management of
acute pancreatitis: Ventral open packing. Hepatogastroenterology 1991:38:134-138.
3 Allardyce DB: Incidence of necrotizing pancre­
atitis and factors related to mortality. Am J
Surg 1987;154:295-299.
4 Remitter IG. Savage WT III. Pantoja JI, Ren­
ner VJ: Death due to acute pancreatitis: A ret­
rospective analysis of 405 autopsy cases. Dig
DisSci 1985:30:1005-1018.
5 Beger HG. Bittner R. Block S. Biichler M: Bac­
terial contamination of pancreatic necrosis: A
prospective clinical study. Gastroenterology
6 Lumsden A, Bradly EL III: Secondary pan­
creatic infections. Surg Gynecol Obstet 1990:
7 Ratlner DW, Compton CC. Gu ZY. Wilkinson
R. Warshaw AL: Bacterial infection is not nec­
essary for lethal necrotizing pancreatitis in
mice, int J Pancreatol 1989;5:99-105.
8 Horn J, Heymer B. Merkle N. Meyer H. Diebolder H: Die akute Obstruktion als pathogene­
tischer Faktor der akuten Pankreatitis: Eine
tierexperimcntelle Untersuchung. Z Gastroen­
terol 1980:18:98-106.
9 Dickson AP. Foulis AK. Imrie CW: Histology
and bacteriology of closed duodenal loop mod­
els of experimental acute pancreatitis in the rat.
Digestion 1986;34:15-21.
10 Runkel N, Rodriguez L. LaRocco MT. Moody
FG: Mechanisms of pancreatic infection in
acute pancreatitis in opossums. Curr Surg
11 Runkel N, Rodriguez LF, Moody F: Mecha­
nisms of sepsis in acute pancreatitis in opos­
sums. Am J Surg 1995:169:227-232.
12 Runkel N. Moody FG. Smith GS. Rodriguez
LF, LaRocco MT. Miller TA: The role of the
gut in the development of sepsis in acute pan­
creatitis. J Surg Res 1991;51:18-23.
Dig Surg 1996;13:269-272
13 Foitzik T. Mithofcr K. Ferraro MJ. Femândezdel Castillo C. Lewandrowski KB. Rattner
DW. Warshaw AL: Time course of bacterial
infection of the pancreas and its relat ion to dis­
ease severity in a rodent model of acute necro­
tizing pancreatitis. Ann Surg 1994:220:193198.
14 Medich DS, Lee TK. Melhcm MF. Rowe ML
Schraut WH. Lee KKW: Pathogenesis of pan­
creatic sepsis. Am J Surg 1993;165:46-52.
15 Gianotti L. Munda R, Alexander JW, Tchervenkov JI. Babcock GF: Bacterial transloca­
tion: A potential source for infection in acute
pancreatitis. Pancreas 1993;8:551-558.
16 Tarpila E. Nystrôm PO. FranztSn L. Ihse 1: Bac­
terial translocation during acute pancreatitis in
rats. Eur J Surg 1993;159:109-113.
17 Kazantse GB. Hecht DW. Rao R. Fedorak IJ.
Gattuso P, Thompson K. Djuricin G, Prinz
RA: Plasmic labeling confirms bacterial trans­
location in pancreatitis. Am J Surg 1994:167:
18 Widdison AL, Karanjia ND. Reber HA:
Routes of spread of pathogens into the pan­
creas in a feline model of acute pancreatitis.
Gut 1994;35:1306-1310.
19 Runkel NS, Moody FG. Smith GS, Rodriguez
LF, Chen Y. La Rocco MT, Miller TA: Altera­
tions in rat intestinal transit by morphine pro­
mote bacterial translocation. Dig Dis Sci 1993:
20 Muncy DM, Halcv-Russell D. Stanislawska M,
Vakil A, Rodriguez L, Moody FG: Bacterial
translocation following acute pancreatitis in
rats. Gastroenterology 1993:I02:A490.
21 Runkel N. Smith GS. Rodriguez LF. LaRocco
MT. Moody FG. Miller TA: Influence of shock
on development of infection during acute pan­
creatitis in the rat. Dig Dis Sci 1992:37:1418—
22 Curley PJ. McMahon MJ. Lancaster F. Banks
RE, Barclay GR. Shcfta J, Boylston AW,
Whichcr JT: Reduction in circulating levels of
CD4-positive lymphocytes in acute pancreati­
tis: Relationship to endotoxin, interleukin 6
and disease severity. Br J Surg 1993:80:13121315.
23 Garcia-Sabrido JL. Valdecantos E, Bastida E,
Tcllado JM: The anergic state as a predictor of
pancreatic sepsis. Zcntralbl Chir 1989:114:
24 Gianotti L, Solomkin JS. Munda R. Alexander
JW: Failure of local and systemic bacterial
clearance in rats with acute pancreatitis. Pan­
creas 1995;10:78-84.
25 Widdison AL. Karanjia ND, Alvarez C, Reber
HA: Influence of levantisole on pancreatic in­
fection in acute pancreatitis. Am J Surg 1992:
26 Sulkowski U. Boin C. Spiegel HU, Bünte H:
Der Einfluss der Zäkostomie auf die Pathophy­
siologie und Prognose der akuten experimen­
tellen Pankreatitis. Langenbecks Arch Chir
27 Foitzik T. Femändez-del Castillo C, Ferraro
MJ. Mithöfer K. Ratlncr DW, Warshaw AL:
Pathogenesis and prevention of early pancrea­
tic infection in experimental acute necrotizing
pancreatitis. Ann Surg 1995:222:179-185.
28 Gianotti L, Munda R, Gennari R. Pyles T,
Alexander JW: Effect of different regimens of
gut decontamination on bacterial translocation
and mortality in experimental acute pancreati­
tis. Eur J Surg 1995; 161:85—92.
29 Marotta F. Tajiri H. Kimura H, ideo G: Topi­
cal treatment affecting bacterial translocation
prevents the outburst of full-blown experimen­
tal acute pancreatitis (AP). Pancreas 1993:8:
30 Luitgen EJT, Hop WCJ, Lange JF, Bruining
HA: Controlled clinical trial of selective decon­
tamination for the treatment of severe acute
pancreatitis. Ann Surg 1995;222:57-65.
31 Isaji S. Suzuki M. Frey CF, Ruebner B. Carlson
J: Role of bacterial infection in diet-induced
acute pancreatitis in mice. Int J Pancreatol
R unkel
Downloaded by:
Kings's College London - 10/25/2017 7:57:11 PM
1 Heger HG: Operative management of necrotiz­
ing pancreatitis - Necrosectomy and contin­
uous closed postoperative lavage of the less­
er sac. Hepatogastrocnterology 1991:38:129—
Без категории
Размер файла
759 Кб
Пожаловаться на содержимое документа