The gastrointestinal tract as a barrier in sepsis
Brian J Rowlands*, Chee Voon Soong+ and Keith R Gardiner*
*Section of Surgery, Queen's Medical Centre, University Hospital, Nottingham, UK and
^Department of Surgery, Queen's University of Belfast, Belfast, UK
The gastrointestinal tract is an organ of digestion and absorption which is
metabolically active and has specific nutrient requirements. In health, it has an
additional function as a major barrier, protecting the body from harmful
intraluminal pathogens and large antigenic molecules. In disease states, such as
sepsis when the mucosal barrier is compromised, micro-organisms and their toxic
products gain access to the portal and systemic circulations producing deleterious
effects. Under these circumstances, systemic inflammatory response syndrome
(SIRS) and multiple organ dysfunction syndrome (MODS) develop leading to
deterioration and death of the patient in the intensive care unit. Therapeutic
strategies for such patients in the intensive care unit aim to support general
immune function and maintain the structure and function of the gastrointestinal
tract. For these therapies to be successful, the underlying septic or necrotic focus
must be ablated using appropriate surgical or other invasive techniques.
Correspondence to:
Prof. Brian J Rowlands,
Section of Surgery,
Queen's Medical Centre.
University Hospital.
Nottingham NG7 2UH, UK
Sepsis, systemic inflammatory response syndrome (SIRS) and multiple
organ dysfunction syndrome (MODS) complicate the investigation and
management of several surgical diseases1. These include major trauma,
obstructive jaundice, inflammatory bowel disease, acute pancreatitis, intraabdominal sepsis and major vascular surgery. These clinical conditions and
their septic complications are characterised by a state of 'hypermetabolism', which leads to a rapid consumption of endogenous stores of
protein and energy, immunological dysfunction and deterioration of organ
function2. These changes which affect the liver, kidney, gastrointestinal
tract, heart and lungs are orchestrated by a series of neuroendocrine events
and the release of cytokines, activators and mediators. The 'gut-liver axis'
appears to have a central role in these responses.
In recent years, the gastrointestinal tract has assumed more importance
in the management of the septic patient in the intensive care unit.
Previously, the gastrointestinal tract was regarded as an organ that
contributed little to the pathophysiology of sepsis but it is now recognised
that the small intestine and colon make important contributions to the
maintenance of hypermetabolism in sepsis, SIRS and MODS 3 . This is due
to changes in gastrointestinal structure and function that promote loss
of intestinal barrier function and associated changes in hepatic Kupffer
British Medical Bulletin 1999;55 (No. 1)- 196-211
OThe British Council 1999
Gut barrier in sepsis
cell function. When taken in conjunction with abnormal colonisation by
luminal micro-organisms, bacterial translocation and absorption of toxins
due to increased intestinal permeability, there is a constant trigger to the
widespread activation of pro-inflammatory cells and the release of other
mediators of the metabolic response to sepsis. This produces profound
systemic changes in metabolic homeostasis and immunological dysfunction. These problems appear to be enhanced by malnutrition4.
Pathophysiology of sepsis
Sepsis and related conditions present a gradation of severity of illness.
Minimal derangement of normal physiology and rapid restoration of
metabolic homeostasis with therapy is at one end of the spectrum. At the
other extreme, is massive disruption of normal organ function, which
leads to progressive deterioration and death despite treatment. Central to
our understanding of these events is an appreciation that initially there is
a localised response to injury. If the local host defence mechanisms are
inadequate or overwhelmed, this may lead subsequently to systemic
manifestations. Localised infections due to bacteria, viruses, fungi and
parasites stimulate the release of various mediators, e.g. cytokines,
prostaglandins, thromboxanes, platelet activating factors and the
complement system. These mediators help to combat infection by
activating neutrophils with consequent degranulation and release of
oxygen radicals, which increase local blood flow and vascular
permeability allowing the influx of phagocytic cells. They also activate
white blood cells and induce chemotaxis. If the severity of the infection is
sufficient that these mediators spill over into the systemic circulation, a
septic cascade is initiated which leads to septic shock, SIRS, and MODS5.
Superoxide radicals now damage host cells. Endotoxin, tumour necrosis
factor, the interleukins, transforming growth factor beta and prostaglandin E2 all contribute to the initiation and maintenance of this cascade,
which may be beneficial or detrimental, depending on the clinical setting.
Sepsis is not synonymous with overwhelming infection and in many
patients who fit the criteria for SIRS, no micro-organism can be
demonstrated or cultured6. Our understanding of the development of
organ failure and death has expanded greatly since originally described7
and we now recognise three stages in the development of SIRS8. The
balance between the anti-inflammatory systemic response and the proinflammatory systemic response is important for metabolic homeostasis,
and the disruption of this equilibrium gives rise to a number of syndromes
(Fig. 1). Clinical manifestations of these syndromes are cardiovascular
compromise (shock), suppression of immunity, apoptosis and organ
dysfunction9-10. The balance of cytokines, their natural antagonists and
British Medical Bulletin 1999;55 (No. 1)
197
Intensive care medicine
Major Surgery
Acute pancreatitis
SEPSIS
Obstructive jaundice
SIRS
Intra-abdominal sepsis
Cardiovascular compromise (shock)
„ Suppression of immunity
CARS
Inflammatory bowel disease
Apoptosis
Organ dysfunction
Ischaemia-reperfusion
MODS
Homeostasis
Multisystem trauma
DEATH
Burns
Rg. 1 Common
surgical conditions
that lead to well
recognised syndromes,
disruption of metabolic homeostasis and
death.
SIRS
:- systemic inflammatory response
syndrome
CARS :- compensatory anti-inflammatory
response syndrome
MODS :- multiple organ dysfunction
syndrome
endogenous antibodies to endotoxins is important in determining
outcome in patients with sepsis syndrome11.
The gastrointestinal tract in health
The gastrointestinal tract is regarded primarily as an organ of digestion
and absorption but it is a metabolically active organ that requires specific
nutrients. It has a major barrier function, protecting the body from
harmful intraluminal pathogens and large antigenic molecules12. In
addition, it plays a pivotal role in the metabolism of glutamine13. The gut
mucosal barrier comprises both immunological and non-immunological
protective components, the former being divided into local and systemic
components and the latter comprises mechanical and chemical barriers as
well as intraluminal bacteria (Table 1). The maintenance of normal
epithelial cell structure prevents transepithelial migration of particles from
the gut lumen, and the preservation of tight junctions between the cells
prevents movement through the paracellular channels14. Acid secretion in
the stomach, alkali secretions in the small bowel and mucous production
throughout the gastrointestinal tract provides additional protection. The
lumen of the gut is colonised by aerobic and anaerobic micro-organisms,
198
British Medical Bulletin 1999,55 (No. 1)
Gut barrier in sepsis
•fable 1 Components of the gut mucosal barrier
Immunological
Non immunological
Local
•
•
•
•
•
•
Mechanical
•
Healthy enterocyte
•
Tight junction
Cell turnover
Normal motility
Gut associated lymphoid tissue (GALT)
Intra-epithelial lymphocytes
Submucosal aggregates
Peyer"s patches
Mesenteric lymph nodes
Secretory IgA
Systemic
•
Circulatory lymphocytes
•
Hepatic Kupffer cells
Bacteriological
>
Aerobic micro-organisms
•
Anaerobic micro-organisms
(Zhemical
Gastric acidity
Salivary lysozyme
Lactoferrin
Mucous secretion
Bile salts
and there is a progressive increase in their numbers from the stomach,
where gastric acid produces an almost sterile environment, to the colon,
which harbours 108 aerobes and 1011 anaerobes. Under normal circumstances, these micro-organisms remain within the lumen of the bowel
where they have important functions in metabolic and nutritional
homeostasis. In disease states, when the mucosal barrier is compromised,
these micro-organisms and their toxic products may 'escape' from the
lumen and reach the systemic circulation to produce deleterious effects,
despite the presence of other defences, e.g. the gut associated lymphoid
tissue (GALT), mesenteric lymph nodes and hepatic Kupffer cells (Fig. 2).
The gastrointestinal tract in disease
The demonstration of intestinal atrophy manifest by changes in weight,
structure and mucosal content of DNA and protein have been assumed
to indicate impaired intestinal barrier function, but this has not been
confirmed in animal studies of protein malnutrition15. Recent clinical
evidence demonstrates a strong association between compromise of gut
barrier function and malnutrition, which suggests a mechanism that
facilitates gut-derived infection and sepsis16. Dysfunction of the mucosal
barrier is thought to result from an imbalance of aggressive and
defensive factors on the gastrointestinal mucosa17. Genetic and environmental factors may modify the response of the gastrointestinal mucosa
British Medical Bulletin 1999;55 (No. 1)
199
Intensive care medicine
Hepatic Kupffer cells
Luminal secretions
Bowel
motility
Fig. 2 Components of
the 'gut-liver axis' that
provide a barrier to
invasion of the
systemic circulation by
bacteria and toxins.
Gut associated
lymphoid
tissue
Luminal
bacteria
Mesenteric
blood flow
to pro-inflammatory factors. Increased gastric acid production, colonisation with Helicobacter pylori, ingestion of non-steroidal anti-inflammatory analgesics and the use of broad-spectrum antibiotics, predispose
the gastrointestinal mucosa to ulceration. Maintenance of mucosal blood
flow, oxidative fuel supply and a normal intestinal bacterial flora protect
against barrier dysfunction. Failure of the intestinal mucosal barrier results
in permeation of microbial and dietary antigens across the intestinal wall.
The transmural migration of enteric bacteria or their products
(endotoxins) to extra-intestinal sites has been termed translocation, and
may occur by the transcellular or paracellular routes. Enhanced uptake of
macromolecules and bacteria has been demonstrated where the intestinal
mucosa is damaged by inflammation, infection, neoplasia or trauma. In
addition, systemic endotoxaemia, bacterial translocation, and increased
intestinal permeability to macromolecules have been demonstrated in
patients with rheumatic diseases, haemorrhagic shock, burns, burn sepsis,
major trauma, following chemotherapy or radiotherapy, and in
experimental endotoxaemia in the absence of macroscopic intestinal
disease. However, translocations of endotoxin or bacteria have not been
found commonly in patients after major trauma or burns18.
Factors influencing intestinal mucosal barrier function
Clinical studies strongly suggest that intestinal barrier dysfunction and
increased permeability occur in patients with intestinal inflammation
200
British Medical Bulletin 1999,55 (No. 1)
Gut barrier in sepsis
and other diseases associated with increased mortality and morbidity19.
The significance of this is difficult to determine, but a working hypothesis
suggests that translocating micro-organisms and toxins activate a systemic
inflammatory cascade and promote organ dysfunction and failure.
Factors that predispose to the development of sepsis syndromes are
changes in the luminal micro-environment, perfusion and oxygen defects,
ischaemia reperfusion injury, malnutrition,and hepatic dysfunction.
The luminal micro-environment
In the small intestine, peristalsis, mucus secretions rich in IgA, and
resident flora help to minimise the growth of pathogenic species. The
peristaltic waves discourage adherence of the organism to the bowel wall
and prevent colonisation with overgrowth. The higher bacterial counts of
the colon are a reflection of this decreased motility in comparison to the
small bowel. The development of paralytic ileus and the formation of a
blind loop allow stagnation of intraluminal contents, creating favourable
conditions for microbial proliferation. The use of systemic and oral
antibiotics cause a reduction in the normal colonisation-resistant resident
flora and an increase in number of potentially harmful pathogenic species.
Perfusion and oxygenation defects
The mucosa at the tip of the villi is particularly prone to ischaemia due to
the counter current exchange mechanism of the vessels and occurs when
there is shunting of blood during low flow states. More subtle mucosal
damage detectable only microscopically may occur following mild
ischaemia. In transient hypoperfusion or hypoxia, the mucosal injury is
more pronounced during reperfusion when the oxygen supply is reestablished. Blood loss and shock have detrimental effects on the immune
and hepatic cells, reducing their capacity to clear bacteria and endotoxin.
Blood transfusion can activate polymorphonuclear leucocytes which then
sequestrate in various sites and has been imphcated in the development of
transfusion-related renal and pulmonary pathology. Autologous blood
may have the same effect leading to leucocyte activation and cytokine
release. The activation and accumulation of neutrophils leads to sludging
of capillaries and reduction of mucosal blood flow.
Complete disruption of the bowel wall, which may occur with
perforation or frank infarction, can induce profound hypotension and
shock due to massive fluid losses into the bowel wall and peritoneal
cavity. This may result in the intraluminal contents entering the systemic
circulation either directly or indirectly. If there is minimal contamination
British Medical Bulletin 1999,55 (No. 1)
201
Intensive care medicine
and haemodynamic disturbance, the initial invasion will be via the
portal system with systemic bacteraemia delayed because of hepatic
Kupffer cell activity. With severe intestinal wall damage, direct massive
peritoneal contamination and profound shock, bacteria and endotoxin
enter the systemic circulation directly from the peritoneal cavity
bypassing the protective barrier of the 'gut-liver axis'.
Ischaemia-reperfusion injury
The causes of intestinal ischaemia are diverse and may represent part of a
generalised hypoperfusion insult or decreased blood flow confined to part
of the splanchnic circulation. In sepsis, intestinal ischaemia may develop
as a result of poor extraction and utilisation of nutrients by the intestine
despite normal oxygen content and delivery20. Selective splanchnic flowdependent tissue hypoxia may develop in sepsis whereby oxygen delivery
fails to meet demands and blood is shunted away from the mucosa. A
reduction in blood flow to a segment of intestine may occur due to
obstruction or strangulation of the bowel, or due to atherosclerosis and
thromboembolic disease of the splanchnic vessels.
The ischaemic injury sustained and changes observed are dependent on
the duration and severity of the ischaemia and whether these changes are
reversible. Three patterns of intestinal ischaemia are apparent. These are
complete vascular occlusion, compensated partial ischaemia and partial
ischaemia followed by reperfusion. In complete vascular occlusion rapid
progression to transmural infarction ensues. In compensated partial
ischaemia, the injury may be so mild that no morphological or physiological change can be detected or there is a slight alteration in vascular
permeability. An increase in intestinal capillary permeability is usually
found with 1 h partial regional ischaemia but may occur after only 20
min. If significant partial ischaemia is reversed, the injury sustained by the
bowel may be exacerbated by reperfusion21. A decrease of intestinal
arterial pressure to 25-35 mmHg for 3 h may lead to a loss of villous
height and a reduction in mucosal thickness. This mucosal damage is
enhanced if the ischaemic episode is followed by 1 h of reperfusion. This
reperfusion injury is more severe than that produced by 4 h of ischaemia
alone.
When ischaemia is relieved by reperfusion or resuscitation, oxygenderived free radicals (ODFR) are generated by the xanthine oxidase
pathway and cause direct injury to the lipid membranes of cells and
hyaluronic acid of their basement membrane22. Xanthine dehydrogenase
occurs naturally and is found in abundance in the intestine. The
dehydrogenase form catalyses the conversion of xanthine and water to
uric acid with the reduction of the nicotinamide adenine dinucleotide
202
British Medical Bulletin 1999,55 (No. 1)
Gut barrier in sepsis
molecule. During ischaemia, the xanthine dehydrogenase is converted
rapidly by a calcium dependent proteolytic enzyme to xanthine oxidase
which generates uric acid and superoxide anion from hypoxanthine and
oxygen. The accumulation of hypoxanthine during ischaemia favours the
generation of ODER in response to the improvement in oxygenation
during reperfusion. Evidence for the role of ODER in producing injury is
supported by the reduction in villous and crypt epithelial cell necrosis by
allopurinol (a xanthine oxidase inhibitor) and superoxide dismutase (a
free radical scavenger) following reperfusion of ischaemic bowel.
The production of ODER together with calcium will activate the plasma
membrane phospholipase A^ which releases arachidonic acid and
activates the complement cascade23. Arachidonic acid is then converted to
a leucotriene (LT) or thromboxane (TX) depending on whether it follows
the lipo-oxygenase or cyclo-oxygenase pathways. Some of these
arachidonic acid metabolites especially LTB4 and TXA^ and breakdown
products of complement, e.g. C5a are potent chemo-attractants and
chemoactivators which can stimulate neutrophils and up-regulate
adhesion molecules on the endothelial membrane. A consequence of this
is the margination and adhesion of neutrophils along the wall of the
vessel. This leads to the release of more ODER and other proteolytic
enzymes which will cause further damage to the epithelium. Sludging and
plugging of the capillaries may then occur giving rise to the 'no re-flow'
phenomenon which will compound the ischaemic insult24. The role played
by neutrophils in reperfusion injury is supported by the prevention of
ischaemia induced intestinal mucosa injury by pretreatment of animals
with monoclonal antibody against neutrophil adhesion molecules25.
Neutropenia and inhibition of neutrophil adherence attenuates the
increase in microvascular permeability26. The initial injury is related to the
direct toxicity of ODFR, independent of neutrophil activation, but later
damage is dependent on neutrophil activation.
Ischaemia reperfusion injury is not confined to the bowel, and other
remote organs and systems may sustain damage as a result of reduction in
blood flow followed by revascularisation and successful resuscitation. The
impairment sustained by remote organs is dependent on the mass of
revascularised ischaemic tissue. The production of ODFR and thenbyproducts is initially confined to the ischaemic region but subsequently
spillage into the systemic circulation occurs. An increase in pulmonary
permeability and hepatocellular injury have been observed when
ischaemic intestine is reperfused27. These injuries to the lungs and liver are
associated with sequestration of neutrophils but neutropenic animals are
protected. A plasma chemo-activator and chemo-attractant produced
following reperfusion of an ischaemic organ can stimulate ODFR
production by activating neutrophils and increasing microvascular
permeability28.
British Medical Bulletin 1999;55 (No. 1)
203
Intensive care medicine
An increase in plasma concentration of tumour necrosis factor (TNF)
has been found with intestinal ischaemia with further increases of 5-10fold following reperfusion. Increased portal vein endotoxin concentrations precede systemically detectable TNF, suggesting that its release
may be due to hepatic stimulation by gut-derived endotoxin. TNF is a
potent pyrogen and intravenous administration can cause tachycardia,
hypotension and death. It activates neutrophils and enhances their
phagocytic ability. Remote organ injury may be abrogated by induction of
neutropenia and by the use of anti-oxidants such as superoxide dismutase
and catalase. Attenuation of the increased microvascular permeability in
the lungs is found by pretreating animals with anti-TNF antibody prior to
bowel ischaemia reperfusion injury. Administration of anti-TNF does not
prevent neutrophil sequestration in the lungs. Pretreatment with
interleukin-1 receptor antagonist can reduce the injury and myeloperoxidase activity as a measure of neutrophil infiltration in both lungs and
liver following ischaemia reperfusion of the bowel.
Malnutrition
The nutritional status of the patient has been shown to be important in
determining the outcome of patients following trauma and surgical
disease. Patients may present with pre-existing malnutrition due to the
inability to swallow, malabsorption, poverty, self-neglect and prolonged
starvation following surgery and in the intensive care unit. In patients
recovering from abdominal surgery enteral feeding improves muscle
function and reduces morbidity and mortality. Malnutrition and total
parenteral nutrition have been shown to cause atrophy of the
enterocytes and colonocytes. Although total parenteral nutrition (TPN)
may improve outcome in those with intestinal failure and in the severely
malnourished it may lead to atrophy of the intestinal mucosa and a
reduction in IgA antibody production. This is because the enterocytes
and colonocytes are better supported by enteral intraluminal infusion of
nutrients. Patients who develop sepsis, SIRS and MODS very rapidly
develop severe nutritional depletion due to an increase in their rate of
metabolism and consumption of their endogenous stores of protein and
energy, especially if no attempt is made to support their nutritional
status using parenteral or enteral feeding protocols.
Hepatic dysfunction
Hyperbilirubinaemia is a common accompaniment to sepsis. It usually
represents a direct toxic effect on liver parenchymal cells causing
204
British Medical Bulletin 1999;5S (No. 1)
Gut barrier in sepsis
inflammation in the portal triads and intrahepatic cholestasis. It may be
exacerbated by hypovolaemia, the use of intravenous nutrition, drugs and
excessive haemolysis. The combination of fever, rigors and obstructive
jaundice should always be investigated quickly as ascending biliary sepsis
can lead to rapid deterioration of the patient with onset of SIRS and
MODS, especially acute renal failure. Extrahepatic biliary obstruction due
to stone or stricture may cause ascending cholangitis. Acute acalculous
cholecystitis may lead to gall bladder necrosis and biliary peritonitis.
Unrecognised and untreated, these conditions have a high mortality so
they should be investigated and treated urgently using surgical,
endoscopic and radiological techniques. In the absence of extrahepatic
pathology which is treatable, sepsis and hyperbilirubinaemia usually
indicate severe parenchymal liver disease and/or intrahepatic cholestasis
which carries a bad prognosis.
Treatment of intestinal mucosai barrier dysfunction
Support of the gut mucosai barrier includes adequate mucosai perfusion,
optimal oxygen delivery, prevention of gastrointestinal haemorrhage
and maintenance of luminal micro-ecology. These general measures
together with more specific measures support enterocyte and colonocyte
structure and function.
Fluid resuscitation
Hypotension should be reversed with the appropriate intravenous fluids.
Initial management should be with crystalloids and depending on
response, colloids may be given. Synthetic colloids increase the circulating
volume by a greater degree per volume infused than crystalloids but carry
the risk of anaphylaxis and coagulopathy. It may be necessary to
supplement volume replacement with blood and fresh frozen plasma. In
some cases, hypotension is resistant to fluid alone, possibly due to the
generation of inflammatory mediators, e.g. TNF and myocardial
depressant factor. Inotropic support with dopamine may be required.
Dopamine at low doses (3-5 ug/kg/min) increases renal blood flow via its
dopaminergic receptors but the effect on splanchnic perfusion remains
controversial29. It may reduce mucosai perfusion and accelerate the onset
of intestinal ischaemia in haemorrhagic shock and in patients with
congestive cardiac failure. Higher doses produce vasoconstriction by ctadrenergic stimulation off-setting the vasodilatory effect. If dopamine is
inadequate noradrenaline or dobutamine should be considered. However,
inotropes are no substitute for adequate fluid replacement therapy.
British Medical Bulletin 1999;55 (No. 1)
205
Intensive care medicine
Tissue oxygenation
Oxygen delivery and degree of shunting due to ventilation/perfusion
mismatch in the lungs can be measured from blood gas analysis and the
inspired oxygen concentration. An accurate indication of tissue
perfusion can be obtained using a silicone tonometer. The principles of
the tonometer are based on the assumption that the partial pressure of
carbon dioxide within the bowel is similar to that of the lumen, and the
concentration of standard bicarbonate within the tissues is similar to
that of arterial blood. The tonometer measures the intramucosal pH of
the bowel which correlates well with diminished perfusion once oxygen
delivery falls below a critical level30. The fall in gastric intramucosal pH
following cardiac surgery is associated with a higher incidence of
morbidity and mortality. Tonometric measurements of gastric pH are an
accepted method of monitoring systemic oxygenation and outcome in
ICU patients31. Sustained acidosis beyond 2 h is highly predictive of
mortality and major complications in aortic surgery. Intramucosal
acidosis may be an early warning of an impending complication and
therapy guided by intramucosal pH measurements may significantly
improve outcome in critically ill patients32.
Selective digestive decontamination
Gastrointestinal haemorrhage used to be a major problem in septic
patients requiring ICU management, but improvements in overall
patient management have diminished the problem to an extent that
prophylaxis with antacids, H2 receptor antagonists or sucralfate are
used sparingly. The use of selective digestive decontamination (SDD) of
the gastrointestinal tract has its advocates. Meta-analysis of the
randomised controlled trials of SDD carried out over 15 years showed
that it was an effective technique for reducing infection related
morbidity and mortality in ICU patients33. The data showed that to
prevent one respiratory tract infection you would need to treat 5
patients and to prevent one death you would need to treat 23 patients.
Although SDD continues to be evaluated in specific groups of patients
admitted to ICU, it has not been universally adopted due to doubts
about efficacy in a general ICU population and concerns about the
development of multiresistant organisms with widespread usage.
Gut barrier dysfunction improves when the intestinal pool of Gramnegative bacteria/endotoxin is reduced by the administration of nonselective or selective antibiotics, lactulose to promote non-pathogenic
lactobacilli, or adsorbents to bind intraluminal endotoxin. The
administration of anti-inflammatory drugs (steroids), systemically or
206
British Medical Bulletin 1999;55 (No. 1)
Gut barrier in sepsis
topically, may enhance the healing of mucosal inflammatory lesions.
Intravenous injection of taurolidine, a drug with antiseptic, antibiotic and
anti-endotoxin activity, significantly reduces systemic endotoxaemia in
experimental colitis whereas systemic broad-spectrum antibiotic (metronidazole and cefuroxime) therapy was ineffective34. Administration of
anti-TNF antibody (cTN3) significantly reduced systemic endotoxaemia,
plasma IL-6 concentration, acute phase protein response and weight loss.
Repair of the intestinal barrier may be hastened by multi-targeted therapy
aimed at reducing luminal aggressive factors, reducing the inflammatory
response (steroids, nitric oxide synthase inhibition, anticytokines) and
selective gut nutrition.
Nutritional support
The use of nutritional support techniques improves outcome of patients
with sepsis and SIRS35. Recently, there have been two major trends in the
use of nutritional support. First there is a major shift from intravenous
administration of nutrients to enteral feedings and second the quantity
of nutrients being administered is decreasing and the quality of nutrient
mix is improving. In sepsis, a number of studies have demonstrated that
enteral feeding has a major advantage over parenteral nutrition36-37. The
beneficial effects of nutrition are its support of generalized immune
function, enhancement of mucosal barrier function, reduction of
bacterial translocation, modification of hepatic Kupffer cell function,
cytokine release and acute phase protein production by the liver.
Additional benefits of enteral nutrition are the maintenance of the
structural and functional integrity of the gastrointestinal tract by
stimulation of 'gut' hormone release, 'gut' motility and mucous
production and maintenance of luminal milieu of nutrients, microorganisms and trophic factors that are important for normal digestion
and barrier function. The theoretical advantages of enteral nutrition are
sometimes undermined by an inability to deliver sufficient nutrients into
the gastrointestinal tract due to prolonged ileus, bloating, abdominal
distention or diarrhoea. The latter may have a small additional benefit
in reducing numbers of luminal bacteria and toxins which are
potentially deleterious. If diarrhoea persists, it may lead to dehydration
and electrolyte imbalance.
The role of novel substrates in maintaining gut integrity has been
extensively reviewed by considering their metabolism in health and
disease, the effects on the gut mucosa and the experimental and clinical
evidence supporting their use in clinical practice18. Evidence that these
substrates improve gut mucosal barrier function and increase survival
does not necessarily imply that restored barrier function and improved
British Medical Bulletin 1999;55 (No. 1)
207
Intensive care medicine
survival are causally linked. The substrates may exert their beneficial
effects by improving nitrogen balance and metabolism or by enhancing
immune function and clearance of translocated bacteria. The strongest
candidates as selective gut nutrients are glutamine for the enterocyte and
short chain fatty acids for the colonocyte. Recent evidence shows that
glutamine supplemented parenteral nutrition improves outcome in
critically ill intensive care unit patients and surgical patients38'39. The
results from studies of enteral supplementation of glutamine are
disappointing40, but a recent study shows encouraging results in severely
injured patients with a low frequency of pneumonia, sepsis and
bacteraemia41. Modest intakes of standard oral diet (0.6 g nitrogen/kg
daily) are sufficient to maintain gut integrity and immune function.
Short chain fatty acids and n-3 polyunsaturated have modest beneficial
effect in patients with intestinal inflammation42. Arginine has proven
immunological benefits43. Despite this, oral supplementation with
arginine has been shown to be detrimental in experimental colitis
causing increased colonic inflammation44 but beneficial to mucosal
barrier function and survival after experimental ischaemia/reperfusion
injury45. There is growing evidence that the L-arginine-nitric oxide
pathway is important in the development of colonic inflammation and
that this may lead to new therapeutic strategies in inflammatory bowel
diseases46-47. Arginine supplements for critically ill patients should be
used cautiously at present. There is insufficient evidence from clinical
studies to support the use of orthinine, branched chain amino acid or
nucleotide supplementation by themselves. Supplementation of enteral
diets with combinations of novel substrates is beneficial to patients in a
number of different clinical situations where gut barrier function is
compromised by malnutrition and hypermetabolism48'49. The use of diets
(containing fibre, fermented oats and lactobacillus) that support
probiotic bacteria (microbial interference treatment) has been advocated
as an important new development in the support of these patients50'51.
Newer therapies
A number of new therapies have been developed based on three
important theories about the sepsis syndrome namely: (i) oxygen debt
causes organ injury; (ii) endotoxin is a critical mediator; and (iii) the
host inflammatory response is harmful52. A number of clinical trials
employing new therapeutic strategies have failed to show convincing
evidence of an improved outcome for patients with sepsis or septic shock
and some agents have caused harm9'53. An explanation of these
therapeutic failures is that our concept of the pathophysiology of sepsis
is too simplistic or incorrect and fails to consider the influence of the
208
British Medical Bulletin-\999;55 (No. 1)
Gut barrier in sepsis
compensatory anti-inflammatory response. More studies of specific
diseases are clearly indicated.
Surgical treatment
None of the aforementioned therapies will be successful if the
underlying source of infection is not controlled. This may be easy if the
patient has an obvious wound infection, an abscess or soft tissue
infection or a necrotic limb. Frequently, the source of the sepsis or SIRS
is not apparent and intrathoracic or intra-abdominal sepsis may be
difficult to diagnose. Treatment may require thoracatomy or
laparotomy. Surgical exploration should be carried out after appropriate
resuscitation of the patient. The extent of the surgical procedure will
depend on the location and extent of the septic lesion. Adequate
drainage and debridement should be carried out. In abdominal sepsis it
may be necessary to consider the use of stomas, tubes or drains and, in
certain circumstances, planned relaparotomy or laparostomy may be a
treatment option53. The chances of survival are enhanced by the early
eradication of the source of sepsis or SIRS. The surgical approach is
dictated by the condition of the patient and the experience of the
surgeon. The survival rate for patients requiring laparotomy for
abdominal sepsis in an ICU setting is approximately 40% and multiple
operations are associated with diminishing returns and increasing
mortality.
References
1
2
3
4
5
6
7
8
9
10
McCrory DC, Rowlands BJ. Septic syndrome and multiple system organ failure. Curr Pract
Surg 1993; 5: 211-5
Cerra FB. Hypermetabolism; organ failure and metabolic support. Surgery 1987; 101: 1-14
Deitch EA. Multiple organ failure. Pathophysiology and potential future therapy. Ann Surg
1992; 216: 117-34
Reynolds JV, O'Farrelly C, Feighery C et al. Impaired gut barrier function in malnourished
patients. BrJ Surg 1996; 83: 1288-91
Davies MG, Hagen P-O. Systemic inflammatory response syndrome. Br ] Surg 1997; 84,
920-35
Rowlands BJ. Management of severe sepsis and intensive care. In: Patterson-Brown S (ed) A
companion to specialist surgical practice - emergency surgery and critical care. London:
Saunders, 1997; 289-308
Baue AE. Multiple progressive or sequential systems failure. Arch Surg 1975; 110: 779-81
Bone RC. Toward a theory regarding the pathogenesis of the systemic inflammatory response
system - what we do and do not know about cytokine regulation. Crit Care Med 1996; 24:
163-72
Bone RC. Sir Isaac Newton, sepsis, SIRS and CARS. Crit Care Med 1996; 24: 1125-8
Nieuwenhuijzen GAP, Deitch EA, Goris RJA. Infection, the gut, and the development of the
multiple organ dysfunction syndrome. Eur ] Surg 1996; 162: 259-73
British Medical Bulletin 1999;55 (No. 1)
209
Intensive care medicine
11 Goldie AS, Fearon KCH, Ross JA et al. Natural cytokine antagonists and endogenous antiendotoxin core antibodies in sepsis syndrome. JAMA 1995; 274: 172—7
12 Saadia R, Schein M, McFarlane C, Boffard KD. Gut barrier function and the surgeon. BrJSurg
1990; 77: 487-92.
13 Souba WW, Smith RJ, Wilmore DW. Glutamine metabolism by the intestinal tract. / Parent Ent
Nutr 1985; 9: 608-17
14 van Lcewuwcn PAM, Boermeester MA, Houdijk APJ et al. Clinical significance of translocation.
Gut 1994; 35 (Suppl 1): S28-34
15 Deitch EA. Bacterial translocation of the gut flora. / Trauma 1990; 30: S184-9
16 Welsh FKS, Farmery SM, MacLennan K et al. Gut barrier function in malnourished patients.
Gut 1998; 42: 396-401
17 Sartor RB. Role of intestinal microflora in initiation and perpetuation of inflammatory bowel
disease. Can J Gastroenterol 1990; 4: 271-7
18 Gardiner KR, Kirk SJ, Rowlands BJ. Novel substrates to maintain gut integrity. Nutr Res Rev
1995; 8: 43-66
19 Rowlands BJ, Gardiner KR, Nutritional modulation of gut inflammation. Proc Nutr Soc 1998;
57: 395-401
20 Fiddian-Green RG. The role of gut in shock and resuscitation. Clin Intensive Care 1992; 3
(suppl): 5-11
21 Parks DA, Granger DN. Contribution of ischaemia and reperfusion to mucosal lesion
formation. Am J Physiol 1986; 250: G749-53
22 Parks DA, Bulldey GB, Granger DN, Hamilton SR, McCord JM. Ischaemia injury in the cat
small intestine: role of superoxide radicals. Gastroenterology 1982; 82: 9-15
23 Wellbourn CRB, Goldman G, Paterson IS, Valeri CR, Shepro D, Hetchman HB.
Pathophysiology of ischaemia reperfusion injury: central role of neutrophil. BrJSurg 1991; 78:
651-5
24 Schmid-Schonbein GW. Capillary plugging by granulocytes and the no-reflow phenomenon in
the microcirculation. Fed Proc 1987; 46: 2397-401
25 Hernandez LA, Grisheam MB, Twohig B, Arfors K-E, Harlan JM, Granger DN. Role of
neutrophil in ischaemia reperfusion induced microvascular injury. Am J Physiol 1987; 253:
699-703
26 Schocnberg MH, Poch B, Younes M et al. Involvement of neutrophil in postischaemic damage
to the small intestine. Gut 1991: 32; 905-12
27 Simpson R, Alon R, Kobzik L, Valeri R, Shepro D, Hetchman HB. Neutrophil and nonneutrophil mediated injury in intestinal ischaemia reperfusion. Ann Surg 1993; 218: 444-54
28 Paterson IS, Smith FCT, Tsang GMK, Hamer JD, Shearman CP. Reperfusion plasma contains a
neutrophil activator. Ann Vase Surg 1993; 7; 68-75
29 Soong CV, Halliday MI, Hood JM et al. The effects of low dose dopamine on intramucosal
acidosis of the sigmoid colon in patients undergoing elective abdominal aortic aneurysm repair.
BrJSurg 1995; 82: 912-5
30 Gram CM, Fiddian-Green RG, Pitenger GL et al. Adequacy of tissue oxygenation in intact dog
intestine.; Appl Physiol 1984; 56: 1065-9
31 Guittierez G, Pahzas F, Doglio P et al. Gastric intramucosal pH as a therapeutic index of tissue
oxygenation in critically ill patients. Lancet 1992; 339: 195-9
32 Maynard ND, Bihari DJ, Dalton RN, Smithies MN, Mason RC. Increasing splanchnic blood
flow in the critically ill. Chest 1995; 108: 1648-54
33 D'Amico R, Pifferi S, Lionetti C, Torri V, Tinazzi A, Liberati A. Effectiveness of antibiotic
prophylaxis in critically ill adult patients: systemic review of randomised controlled trials. BMJ
1998; 316: 1275-85
34 Gardiner KR, Anderson NH, McCaigue MD, Erwin PJ, Halliday MI, Rowlands BJ. Enteral and
parenteral antiendotoxin treatment in experimental colitis. Hepatogastroenterology 1994; 41:
554-8
35 Wilmore DW. Catabohc illness: strategies for enhancing recovery. N Engl J Med 1991; 325:
695-702
36 Kudsk KA, Groce MA, Fabian TC et al. Enteral vs parenteral feeding: effects on septic
morbidity after blunt and penetrating abdominal trauma. Ann Surg 1992; 215: 503—13
210
British Medical Bulletin 1999;55 (No. 1)
Gut barrier in sepsis
37 Moore FA, Feliciano DV, Andrassy RJ et al. Early enteral feeding, compared with parenteral
reduces postoperative septic complications - the result of a meta analysis. Ann Surg 1992; 216:
172-83
38 Griffiths RD, Jones C, Allan-Palmer TE. Six month outcome of critically ill patients given
glutamine-supplemented parenteral nutrition. Nutrition 1997; 13: 295-302
39 Morlion BJ, Stehle P, Wachtler PC et al. Total parenteral nutrition with glutamine dipeptide
after major abdominal surgery: a randomised double blind, controlled study. Ann Surg 1998;
227: 302-8
40 Powell-Tuck J. Glutamme supplementation in artificial nutritional support. Lancet 1997; 350:
534
41 Houdijk APJ, Rijnsburger ER, Jansen J et al. Randomised trial of glutamine enriched enteral
nutrition on infectious morbidity in patients with multiple trauma. Lancet 1998; 352: 772-6
42 Ross E. The role of marine fish oils in the treatment of ulcerative colitis. Nutr Rev 1993; 51,
47-9
43 Barbul A. Arginine: biochemical, physiology and therapeutic implications. / Parent Ent Nutr
1986; 10: 227-38
44 Neilly PJD, Kirk SJ, Gardiner KR, Anderson NH, Rowlands BJ. Manipulation of L-argininenitric oxide pathway in experimental colitis. Br J Surg 1995; 82: 1188-91
45 Scruffier R, Raul F. Prophylactic administration of L-arginine improves the intestinal barrier
function after mesentenc ischaemia. Gut 1996; 39: 194-8
46 Kimura H, Hokan R, Miura S et al. Increased expression of an inducible isoform of nitric oxide
synthase and the formation of peroxynitrite in colonic mucosa of panents with active ulcerative
colitis. Gut 1998; 42: 180-7
47 Warner TO, McCartney SA. NOS inhibitors in colitis: a suitable case for treatment. Gut 1998:
42: 152-3
48 Bower RH, Cerra FB, Bershadsky B et al. Early enteral administration of formula (Impact)
supplemented with arginine, nucleotides and fish oil in intensive care unit patients: results of a
mukicenter prospective, randomised clinical trial. Crit Care Med 1995: 23: 436—49
49 Daly JM, Weintraub FN, Shou J, Rosato EF, Lucia M. Enteral nutrition during multimodality
therapy in upper gastrointestinal cancer patients. Ann Surg 1995; 221: 327-38
50 Bengmark S. Econonutrition and health maintenance - a new concept to prevent GI
inflammation, ulceration and sepsis. Clin Nutr 1996; 15: 1-10
51 Bengmark S. Ecological control of the gastrointestinal tract - the role of probioac flora. Gut
1996; 42: 2-7
52 Freeman BD, Natanson C. Clinical trials in sepsis and septic shock in 1994 and 1995. Curr
Opin Crit Care 1995; 1; 349-57
53 Anderson ID, Fearon KCH, Grant IS. Laparotomy for abdominal sepsis in the critically ill. Br
} Surg 1996; 83: 535-9
British Medical Bulletin 1999;55 (No. 1)
211