Approach to management of

Upper Gastrointestinal
Bleeding
By
Dr. Williams E.A
 Definition and terminology

•Upper gastrointestinal hemorrhage(UGIB):

Bleeding upon the ligament of treitz

•Hematemesis : vomiting of fresh blood

shown active/ massive bleeding
• Coffee ground: (blood+gastric acid) dark or
brownish vomit which shows presence of
denatured blood in the vomit.
• Melena: dark tarry & foul-smelling stool.
Hb+acid= acid hematin
• Hematochezia: means fresh or blood clot in the
stool.
 Approximately 50% of admissions for GI
bleeding are for UGI causes, 40% for lower GI
and 10% for obscure bleeding (from small
intestine)
 UGI bleeding accounts for > 500,000
admissions/year or about 170
patients/100,000 population/year
 Generally, peptic ulcer is the most common
cause of UGI bleeding
 Mortality rate from severe UGI bleeding has
remained at 5% -10% in spite of advances in
medical treatment.
 3 yrs admissions of GI cases in FMCY,
upper GI bleeding was responsible for
26.2% of the cases.
 GI bleeding accounted for 9.3% of all GI
mortality.
 High mortality is principally due to an increase
in the proportion of elderly patients with UGI
bleeding who also have co-morbidities and
increase in number of liver cirrhosis patients
with variceal bleeding
 In 80% of patients with UGI bleeding, the
bleeding is self-limiting even without specific
therapy. However, mortality is increased
(30%-40%) in those who continue to bleed or
rebleed.
Causes of Upper gastrointestinal bleeding
 Duodenal ulcer
 Gastritis and/or gastric erosions (usually
NSAID or alcohol-induced)
 Oesophageal varices
 Gastric ulcer
 Gastric tumours
 Oesophageal tumours
 Mallory-Weiss tear (or syndrome)
 Erosive gastro-oesophagealreflux disease
 Duodenitis
 Hereditary telengiectasia(Osler-Weber-
Rendu syndrome)
 Coagulopathy

 Aorto-duodenal fistula

 Portal hypertensive gastropathy

 Dieulafoy lesions (calibre-persistent arteries that

arise to the surface of the gastric mucosa, erode

through it and bleed; occur mainly in the elderly)

 Incidence(%)
 Duodenal ulcer 24.3
 Gastric erosions 23.4
 Gastric ulcer 21.3
 Esophageal varices 10.3----------20% (in cirrhosis)
 Malorry-Weiss tear 7.2
 Esophagitis 6.3
 Duodenitis 5.8
 Neoplasm 2.9
 Marginal( stomal) ulcer 1.8
 Esophageal ulcer 1.7
 Miscellaneous 6.8
Silverstein FE, Gilbert DA, Tadeseo FJ, et al,
The national ASGE Survey on upper gastrointestinal bleeding
Gastrointestinal Endoscopy, 1981
Clinical features of UGIB

 Melaena: commonest mode of presentation]

 Coffee grounds vomiting.

 Haematemesis.

 Haematochezia: occurs in 11% of patients

with rapid bleeding from an UGI source
Case 1
 24yr old male with massive hematemesis
(>1000ml), significant alcohol hx. Has pulse rate
of 100bpm, systolic Bp > 100, stable mental
status with splenomegaly and no ascites.
Normal liver function parameters, abdominal
scan showed normal size liver with coarse
echotexture and splenomegaly. Hb on admission
was 8g/dl and received 4 units of blood
transfusion.
 Case 2
 A 49yr old businessman with significant alcohol
intake admitted on account massive
hematemesis, passage of melena, hematokezia,
with associated altered sensorium, Hypovolemic
shock, oliguria, splenomegaly and abd
distension with ascites. Hb on admission was
4g/dl and received 13 units of blood transfusion
on admission. Liver function test was in-keeping
with chronic liver disease.
Management approach to Gastrointes
tinal Bleeding

 Initial Assessment

 Initial Resuscitation

 Critical care and monitoring

 Definite diagnosis and management

INITIAL ASSESSMENT
Assessment of Upper GIT Bleeding

1. Quick History
 Recurrent epigastric pain: peptic ulcer
disease.
 Current NSAID ingestion: Gastritis, gastric
erosion.
 Haematemesis following an alcoholic binge
and retching: Mallory-Weiss tear.
 Significant alcohol consumption:
oesophageal varices from cirrhosis & portal
hypertension or Mallory-Weiss syndrome
 Previous history of bleeding disorder (e.g.
epistaxis): coagulopathy
 Previous history of surgery (e.g. aortic
aneurysm repair): aorto-duodenal fistula.
2. Establish the Severity & Risk stratification of
the patient by Scoring to categorize the
patients;

- Active or ongoing/ massive/ continue/ or

intermittent bleeding

- High risk of re-bleed/poor prognosis

 Estimate the blood loss (≥ 3 pints is regarded
as severe)
 Check for co-morbidity, e.g. diabetes
mellitus, cancer, chronic obstructive
pulmonary disease, etc.
 Age (> 55 years higher risk & worse
outcome)
3. Haemodynamic status
 Check for evidence of shock

- Pulse rate > 100/min &

- Systolic BP < 100 mmHg

 Category of Hypovolemic Shock

 Class I:Impending (< 10% of blood volume

loss)

no symptom, pulse > 90-100 , BP normal

 Class II: mild (10-20% of blood volume loss)

fainting, pallor, cool skin, BP drop, pulse>120

 Class III: moderate (20-30% of blood
volume loss): urine output -oliguria
 Class IV: severe ( >40% of blood volume
loss) may caused unconcious and cardiac
arrest
4. Quick Examination
 External stigmata of chronic liver disease
(CLD) & portal hypertension (e.g. ascites &
splenomegaly)
 Inspection of the mouth for telangiectasia
 Abdominal examination for epigastric
tenderness (and for aortic aneurysm)
 Digital rectal examination (DRE) for melaena
Scoring to categorize the patients
 Forrest classification
severe, moderate, mild
Lancet 1974
 Rockall Risk Scoring
Gut 1996
 New Scoring System by Blatchford
Lancet 2000
 Modified Rockall Score for both Non-variceal and Variceal
bleeding
AJG 2002
 AIMS65 scoring.
 Rockall Scoring
 Age
 Shock
 Co morbid disease ( cancer diseases)
 Endoscopic diagnosis
 Stigmata of recent hemorrhage
Pre-endoscope score 0-7
Post –op endoscope score 0-11
* this scoring system is good to predict for the
mortality rate much than rebleeding
0-3 : mortality rate = 0 – 11%
4-7 : mortality rate = 24- 27%
 > 8 : motality rate = > 40%

Rockall TA et al GUT 1996; 38: 316-21

New Scoring System by Blatchford
 Admission Hb

 BUN

 Pulse

 Systolic BP

 Fainting or melena as chief complaint

 Liver disease or cardiac disease

• to predict the need for clinical interventions

 AIMS65 scoring system

 Albumin level < 30g/l

 INR > 1.5

 Alteration in mental status

 Systolic blood pressure < 90 mmHg

 Age greater than or equal to 65yrs

Score 1 point each, and score more than 2 is high risk.
 High Risk ~Criteria
 Host Factors
- Age >60yr
- Co-morbid conditions e.g. renal failure, cirrhosis,
cardiovascular disease, COPD
- Hemodynamic instability; mod to severe shock
- Coagulopathy include drug-related
 Bleeding character ; Active continue red blood from
NG after irriagtion and red blood per rectum
 Patient course; massive blood transfution> 4-6 units
to maintain Hb in 24 hr , re-bleeding in 72 hr , return
to have hemodynamic instability

2004 Concensus for Clinical Practice Guideline for the Management of

Upper GI bleeding; สมาคมโรคทางเดินอาหารแห่งประเทศไทย
 Investigations

 Haemogram: Hb/PCV, platelet count

 Coagulation tests: prothrombin time, INR,

bleeding time, whole blood clotting time, etc.
 Blood grouping & cross-matching

 Liver function tests (serum bilirubin,

aminotransferases, alkaline phosphatase,
albumin)
 Serum electrolytes, urea & creatinine

 Upper GI endoscopy (emergency or elective

depending on the severity & the risk)
 Capsule endoscopy/enteroscopy(especially for
obscure bleeding)
 Angiography (done only if endoscopy has
failed
INITIAL RESUSCITATION
 Initial Resuscitation

 What to do for good resuscitation?

 When will we give blood transfusion ?
 Which medication will be used?
 Large- bore intravenous lines or central lines

 NG tube aspiration (by hand) to decompress

clot in stomach
 Volume expansion with colloid or crystalloid

 Transfusion of blood immediately if patient

has hemodynamically unstable

* Blood products are the most efficient

volume expanders.
 If patients have coagulopathy, they should be corrected.

- PTT prolong > 1.5 times

- Platelet < 50,000/ mm3

- FFP should be given after 6 unit of PRC and plt

should add after 10 unit of PRC

 Monitoring V/S, urine out put /hour

 Airway protection in those who have alterated

consciousness or endotracheal intubations may be required.

Medications to use in UGIB
 Recommendation for empiric
Acid- suppression therapy
Traditionally treated, even before the cause is
determined, with acid suppression therapy.
Medications are extremely safe, although the
efficacy of this practice has not been proven
conclusively.
Kupfer, et al Gastroenterol Clin of North Amer, 2000

I.V. Proton pump inhibitor is more effective than

I.V. H 2 blocker in increasing intragastric pH
Vasopressin should not be used due to its systemic side
effect
 High dose omeprazole significantly
reduces the frequency of further bleeding
and of surgery in patients with bleeding
ulcer.
dosage 80mg i.v. every 12 hrs. for 5 days
Saltzman JR, N Engl J Med, 1997
NEW GENERATION PPI
- Lanzoplazole
- Pantoprazole
- Rabeprazole
- Esomeprazole
 SOMATOSTATIN

Somatostatin / Octreotide infusion

- In massive UGIB with Hx of advance liver

disease is recommended

PROSTAGLANDIN ANALOQUE
- Cytoprotective agent
Somatostatin causes
- Splanchnic vasocostriction

- Reduces Azygos venous blood flow

- Reduces portal collateral circulation and

decreases portal pressure

- it can be discontinued without tapering.

 Octreotide (Somatostatin analoque)

50 microgram i.v.bolus then

50 microgram/ hr for infusion rate for 5 days

 Terlipressin: 1 to 2 milligram i.v. bolus then

1 milligram 4 -6hrly (slow peripheral injection),

Max. duration of 72 hours.
 Propranolol: nonselective beta-blocker, to
lower the portal system pressure.
 20mg to 40mg daily

 Help to reduce the risk/chance of rebleed.

CRITICAL CARE AND MONITORING
 Critical care and monitoring

 ICU is needed, when?

- Massive/ continue or on going bleeding

with or without coagulopathy
- High Rockall scoring patients ( high risk
of morbidity & mortality due to continue
or rebleeding
- Severe co-morbid disease
DEFINITE DIAGNOSIS AND MANAGEMENT
 Definite diagnosis and management

 Esophagogastroduodenoscope

( EGD for Dx and Rx)

 Technique of operative intervention
 Endoscogastroduodenoscope
( EGD for Dx and Rx)
Indication and Timing
- In high score patients ( > 3)
- Shock Category II, III
- Promptly as a double set up in active /massive
bleeding
- Under specialist to perform endoscopic
therapy for hemostasis or localized potential
angiographic or surgical therapy

* Initial diagnostic procedure of choice should

be performed in first 6- 24 hour after onset of
bleeding
Precaution and contraindication
 Absolute contraindication
- GI perforation
- Acute uncontrolled unstable angina
- Severe untreated coagulopathy
- uncontrolled respiratory decompensation
- unexperience endoscopist and patient
agitation and uncooperation
* Intraoperative endoscopy ( on ET-tube
and G/A ) in selected cases or shift the
intervention to surgery or conservative only
 Prediction of further ulcer bleeding

 The most important endoscopic predictor of persistent

or recurrent bleeding is active bleeding( arterial
spurting or oozing) at the time of endoscopy
 The rate of rebleeding is approximately
3 % in the low risk group
25% in the high risk group
 Number of blood transfusion units
> 5 units = 57% needing Surgery, mortality = 43%
Adverse Prognostic Factors in UGIB
Endoscopic criteria for endoscopic intervention
because of high rate of continue or re-bleeding

Stigmata of recent hemorrhage: Forrest classification I,II

 Active bleeding lesion, oozing
 Visible vessel, Adherent clot
Ulcer location
 Posterior duodenal bulb
 Higher lesser gastric curvature, High lying ulcer
Ulcer size and character
 Large and hard edge
 Endoscopic Intervention For PU bleeding

1) Thermal Techniques ; monopolar/bipolar/heater

probe/laser photo coagulation
2) Injection Methods; 3.6% hypertonic saline+1:20,000
adrenalin 9-12 cc or 1:10,000 10 cc via 23-25 gauge
needles. 0.5 cc each point
3) Topical Agents; cyanoacrylate tissue glues/
microcrystalline collagen hemostat : * good for
diffuse gastric mucosal lesions or adjunct to other
modalities
4) Mechanical methods; Hemoclips (1.5mm)/ balloon
tamponad

Sukawa, et al, Surg Clin of North Amer, 1992

 Post combind adrenalin injection
And Heat probe coagulation in acute
GU bleeding
Post injection + Heater probe coagulation
in active DU bleeding
Follow up EGD of DU bleeding 1 month
Dieulafoy’s lesion : Therapeutic Hemoclip via EGD
Endoscopic Intervention

For Esophageal varices :

1) Endoscopic band ligation.

2) Endoscopic sclerosing therapy: 1% Ethoxyscleral

solution 0.5-1cc /point

3) Combined

Ballon Tamponad for temporary control after fail

endoscopic intervetion control
( Senstaken Blakemore tube preparation)
INTRAVARICEAL
INJECTION
( underfluoroscope
and venogram)
PARAVARICEAL
INJECTION
ENDOSCOPIC
MUCOSAL
VARICEAL
BAND LIGATION
TIPS( Transcutaneous-jugular intrahepatic
portosystemic shunts)
 Non-operative shunt
 Use in stage of cirrhosis with liver failure
 In non-randomize trial : Less effective to stop
GI bleeding than operative shunt, but less
invasive.
 Technique need radiointervention
( described by Zemel G, Katzen B T, Becker G J, et al TIPS,
JAMA, 266:390,1991 )
Topic of interest…..

 Video capsule Endoscopy

 Intraoperative endoscopy

 Rare causes of upper gastrointestinal

hemorrhage from an obscure source;

small intestine distal to ligament of treiz

that EGD could not exam, new scope was
developed
Gastric Varices

Primary
 Lessor curvature : common- resolute after
Endoscopic intervention for RX of EV
 Greater curvature: Less common
Secondary
Isolated Gastric Varices: Fundus, due to
splenic vein thrombosis treated by
SPLENECTOMY
THANK YOU FOR LISTENING