Dr Preethy John Chandy
Oral cavity & bowel as sources of infection in
critically ill patients
Microbial invasion of body: skin v/s GIT
• Skin: multi layered keratinised surface
• GIT: single cell layer mucosa 0.1 mm thick.
microflora outnumbers that in skin
• No of bacteria in 1 gram of stool (10 to 100 billion)
> the no of people on Earth (6.5 billion)
• Real threat of microbial infection comes from GIT
Microbes in GIT
• Moist environment in mouth & GIT - ideal for
• 400 to 500 different species of bacteria and fungi in
the adult GIT.
Levels of Protective Mechanisms from microbes in
• 1.) Lumen – Acid - anti septic
• 2.)Bowel wall – Physical barrier of Mucosa
• 3.)Extraluminal - Reticuloendothelial system (2/3rd
of it is located intra abdominally result of
frequent microbial invasion across bowel wall.)
Effect of pH on the growth of microbes-
antiseptic action of gastric acid
• Gastric acid endogenous antiseptic agent that
eradicates microorganisms swallowed in saliva and
• Drug-induced inhibition of gastric acid production
recurrent Salmonella enteritis (Lancet 1990)
• Achlorhydria increased risk of bacterial
gastroenteritis (Gut 1987, Scandinavian Jn Gastroenterology
• Gastric acid our own built-in method of
disinfecting the food we eat.
The Acid Phobia
• Gastric acid : long-standing reputation of being a
corrosive agent that can eat through an unprotected
stomach wall and "burn a hole in your stomach.“
• This is more fantasy than factis a direct result of
the traditional notion that gastric acid is the
principal cause of PUD
• Corrosive for inorganic compounds like metals and
enamel, but is not destructive for organic matter.
• Evidence indicates that local infection with H.
pylori is responsible for most cases of PUD.
Predisposing Conditions for sepsis from GIT
1.)Translocation: Movement of organisms across the
bowel wall into the systemic circulation.
3 conditions that promote it:
a.) microbial overgrowth in the bowel lumen
b.) disruption of the mucosal barrier
c.) defective clearance by the lymphatic system.
2.) Reduced gastric acidity:
Reduced gastric acidity: loss of antiseptic action
bacterial overgrowth in stomach infections :
- Pneumonia from aspiration of infectious gastric
-Septicemia from bacterial translocation across the
• Reason to avoid the use of drugs that inhibit gastric
STRESS-RELATED MUCOSAL INJURYINJURY
Erosions in the gastric mucosa that occur in almost all
patients with acute, life threatening illness.
-Superficial Erosions: confined to mucosa
-Stress Ulcer: deeper into submucosa
• Mucosal lining: normally shed and replaced every 2
- 3 days.
• When nutrient blood flow is inadequate to support
the replacement process, the surface of the bowel
becomes denuded, superficial erosions.
Actions of gastric acid may aggravate
this , but the principal cause of stress-related
mucosal injury is impaired blood flow, not gastric
• Erosions can be demonstrated in 75% to 100% of
patients within 24 hours of admission to the ICU.
1.) Clinically silent : majority
2.) Microbial translocation
3.) Clinically apparent GI bleeding in as many as
25% ( Chest 2001)
4.) Clinically significant bleeding (i.e., significant drop in
blood pressure or requires transfusion) occurs in only 1%
- 5% . (NEJM 1994) Disruption of only small capillaries in
High Risk Conditions for clinically significant
• 1. Mechanical ventilation for longer than 48
• 2. Coagulopathy (i.e., platelets <50,000, INR >1.5,
or PTT >2 X control)
• 3. Hypotension
• 4. Severe sepsis
• 5. Multisystem trauma
• 6. Severe head injury
• 7. Burns involving >30% of body surface area
• 8. Renal failure or hepatic failure
• For the other conditions, at least two must be
present to consider the patient high-risk for
• These high-risk conditions serve as indicators for
prophylactic therapy to prevent GI bleeding from
• Preserving Gastric Blood Flow
No readily available methods for monitoring gastric blood
• Sublingual capnometry measures PCO2
on the underside of the tongue, is promising method (Chest
But experience is currently limited.
• The best strategy is to maintain systemic blood flow and
O2 transport using standard markers (e.g., blood lactate
levels) or invasive parameters (e.g., oxygen delivery,
oxygen uptake) if available.
• It exerts a trophic effect on the bowel mucosa that
helps to maintain the structural and functional
integrity of the bowel mucosa (Int Care Med 1999).
• Can be considered adequate prophylaxis for stress-
related gastric hemorrhage unless there is some
other condition that raises special concern for GI
bleeding, such as a coagulopathy, a prior history of
bleeding from gastritis or PUD, or active PUD.
• Cytoprotection: uses an agent that provides local
protection to the gastric mucosa.
• Gastric acid reduction:
• Aluminum salt of sucrose sulfate forms a
protective covering on the gastric mucosa and helps
to preserve the structural and functional integrity of
the mucosa (N Engl J Med 1990).
• Part of this effect may be due to local stimulation of
prostaglandin production, which helps to preserve
gastric blood flow.
• The pH of gastric secretions is not altered
• Least expensive.
• INTERACTIONS. binds to a number of drugs in the bowel
lumen and reduce their absorption.
• To avoid potential interactions in the bowel, these drugs
should be given at least 2 hours before sucralfate.
• The aluminum binds phosphate in the bowel, but
hypophosphatemia is only rarely reported in association
with sucralfate therapy (Nutr Clin Prc 1991). It is not advised for
patients with persistent or severe hypophosphatemia.
Histamine type-2 Receptor Antangonists.
• Continuous infusion ot H2 blockers is the most
effective method of maintaining gastric acid
inhibition (Crit Care Med 1988); however, intermittent
dosing is currently the favored regimen for stress
• Famotidine is longer lasting than ranitidine [i.e., a
single 20 mg i.v dose of famotidine inhibits gastric
acid-for 10-12 hours , while a single 50 mg i.v dose
of ranitidine inhibits gastric acid for 6-8 hrs ]
• i.v doses of are largely excreted unchanged in
urine, and accumulation of these drugs in renal
insufficiency can produce a neurotoxic condition
characterized by confusion, agitation, and even
seizures (Mosby drug consult 2005).
• Dose should be reduced in renal insufficiency.
• Benefit v/s risk:
Reduces incidence of clinically significant bleeding
Has been associated with increased risk of infection
(Lancet 1990, JAMA 2004) esp pneumonia.
Sucralfate v/s H2 Receptor Antagonists
The study involving 1,200 ventilator-dependent
patients in 16 ICUs (NEJM 1998) randomized to
receive sucralfate (1 gram q6H) or ranitidine (50 mg
• Clinically significant bleeding occurred more
frequently in the sucralfate group ( absolute
difference: 2.1%) while hospital-acquired
pneumonia occurred more frequently in the
Ranitidine group ( absolute difference : 2.9% not
• But a combined analysis of 8 other studies shows a
significantly greater incidence of pneumonia in the
Fewer bleeding episodes v/s Fewer pneumonias ?
• Can't be based on survival benefit because the
mortality in both groups is the same.
• Relative incidence of GI bleeding v/s pneumonia in
ICU patients: pneumonia occurs much more
frequently fewer patients would have to be
treated with sucralfate to see a benefit (i.e, fewer
pneumonias) as compared to ranitidine.
• A recent survey showed that H2 blockers are used
much more frequently than sucralfate for stress ulcer
prophylaxis in the ICU (Crit Care Med 2004).
• However, the major reason for this was drug
availability rather than clinical efficacy.
Proton Pump Inhibitors
• PPIs-prodrugs- bind irreversibly to the membrane
pump for hydrogen ion secretion and produce
complete inhibition of gastric acid secretion.
• More effective in reducing gastric acidity than H2
blockers, and unlike H2 blockers, they do not
produce tolerance with prolonged use (Crit Care med
• Intragastric administration of PPIs can be
problematic because of inactivation by gastric acid.
• So morning dose before first feed when gastric acid
secretion in minimum is advocated.
• Enteric coated granules of omeprazole &
lansoprazole mixed in 8.4% NaHCO3 (to
neutralise gastric acid) via NGT has been tried (Am J
Gastro 1999), but is time consuming and with variable
bioavailability (Crit Care Med 2002).
• Pantoprazole i.v 40 mg od
• Esomeprazole i.v 20 mg od
• There is school of thought :Low frequency of
bleeding from gastric erosions and the effectiveness
of other prophylactic measures, the use of PPIs for
stress ulcer prophylaxis seems unnecessary.
• Furthermore, the potency of PPIs in raising gastric
pH will create even greater risks from bacterial
overgrowth in the bowel than the H2 blockers.
Occult Blood Testing.
• Is not necessary for evaluating the efficacy of stress
ulcer prophylaxis. NGT aspirates almost always
contain occult blood in the presence of gastric
erosions (Ann of Surg 1994), and as few progress to
clinically significant bleeding, the presence of occult
blood in NGT aspirates has no predictive value for
assessing the risk of significant bleeding.
• Guaiac and Hemoccult tests give false+ and false-
when pH is less than 4 (J Clin Gastro 1984).
• The Gastroccult test (Smith Kline Laboratories) is
not influenced by pH (J Clin Gastro 1984) and is the
more appropriate test.
DECONTAMINATION OF THE ALIMENTARY TRACT
1.) Oral Decontamination
2.) Selective digestive decontamination
The aspiration of mouth secretions into the upper
airways inciting event in most cases of hospital-
• Aspiration of one micro liter of saliva will
introduce about one million microbes into the
airways. Fortunately, normal mouth flora are
harmless saprophytes (e.g lactobacillus and a-
• Critically ill patients are not as fortunate
• In hospitalized patients colonization of oral cavity
occurs with pathogenic organisms eg aerobic gram-
negative bacilli like P.aeruginosa (Int Care Med 1995).
• The change in microflora is not environmentally
driven, but is directly related to the severity of
illness in each patient.
• Highlights the importance of host-specific factors
in the microbial colonization of body surfaces.
• Colonization requires microbes to adhere to the
underlying surface. Epithelial cells on body surfaces
have specialized receptor proteins that can bind to
adhesion proteins ( adhesins) on bacterial surface.
• Healthy subjects, express receptors that bind
harmless organisms (e.g., lactobacillus)
• But seriously ill patients, express receptors thatbind
pathogenic organisms prelude to hospital-
• Bacterial adherence hold promise as an exciting
field of study prevent colonization & infection.
Oral Decontamination Regimen
• Successful regime in ICU
Nonabsorbable antibiotics applied locally in the
• Preparation: mixture of 2% gentamicin, 2%
colistin, and 2% vancomycin as a paste.
• Regimen: Apply paste to the buccal mucosa with a
gloved finger q6h until the patient is extubated.
• Eradicates most aerobic bacteria and Candida
species from the mouth in about one week.
• This was a study of ventilator-dependent patients
• Decontamination reduced the incidence of
pneumonia ( 27% 10%) : 60% reduction
• and mortality rate ( 38% 29%) : 23% reduction.
• Prolonged use of this locally applied antibiotic
regimen has not resulted in the emergence of
antibiotic resistant organisms.
• Success of oral decontamination in reducing the
incidence of nosocomial pneumonia has prompted
the Centers for Disease Control (CDC) to include a
recommendation for oral decontamination in their
updated guidelines on preventing pneumonia in
Condition in the ICU that might benefit from
decontamination of the alimentary tract.
Selective digestive decontamination (SDD)
• More extensive version of oral decontamination covering entire
• Is selective as it does not eliminate normal inhabitants of the bowel
which are important in preventing colonization with opportunistic
• A successful SDD regimen (Int Care Med 1984):
• Oral cavity: A paste of 2% polymyxin, 2% tobramycin, and
2% amphotericin is applied to the inside of the mouth with a
gloved finger q6H.
• GI tract: A 10 mL solution containing 100 mg polymyxin E, 80 mg
tobramycin, and 500 mg amphotericin is given via a NGT q6H.
• Systemic: i.v cefuroxime, 1.5 grams q8H for the first 4 days of
• It will eradicate most gram-negative aerobic bacteria
and yeasts after 1 week.
• The i.v antibiotic provides systemic protection until
the bowel regimen is fully effective at 1 week.
• Some SDD regimens do not include an i.v
antibiotic, but they are less successful
• The oral and GI components of SDD are continued
until the patient is well enough to be discharged
from the ICU.
• This study shows the influence of SDD on the
incidence of ICU-acquired infections.
• All 3 infections (pneumonia, urinary tract infections,
and septicemia from vascular catheters) were
significantly less frequent in SDD group.
• Similar results were reported in 10 other clinical
trials of SDD, which showed a combined 40%
relative reduction in the frequency of acquired
infections in the ICU (Br Med J 1998).
• The Never-Ending Debate
• Despite over 20 years of experience with SDD and
numerous reports of efficacy there is a continuing debate
over its merits.
1.)Impact on Mortality : most of the early studies showed no
• reduction in mortality despite the decreased rate of
• However a recent large-scale study with 1,000 ICU patients
using an iv antibiotic for the first few days showed a relative
35% reduction in mortality (Lancet 2003).
• Many of the early SDD regimens did not include an i.v
antibiotic may explain the improved results of the most
• 2.)Fear of antibiotic resistance: there is no evidence
to support this (Curr Opin Crit Care 2002).
• The debate overlooks one simple fact: the goal of
SDD is to reduce hospital-acquired infections and it
achieves this goal consistently.
• Therefore, SDD must be considered an effective
method of infection control in the ICU.