CLINICAL PRACTICE GUIDELINES ON ACUTE
NON-VARICEAL UPPER GASTROINTESTINAI BLEEDING
|
MINISTRY OF HEALTH, MALAYSIA 2003 Members of the Expert Panel Panel Members: |
1. INTRODUCTION
Upper gastrointestinal bleeding
(UGIB) is a common medical emergency associated with significant morbidity
and mortality. Acute upper gastrointestinal haemorrhage accounts for about
2,500 hospital admissions each year in the United Kingdom (1). The annual
incidence varies from 47 to 116 (approximately 100) per 100,000 of the population
and is higher in socioeconomically deprived areas (2). The incidence is approximately
72 per 100,000 in Malaysia (3, 4). Bleeding peptic ulcer remains the most
common cause of acute non-variceal upper gastrointestinal bleeding. 80% of
such bleeds stop spontaneously. However, 20% of patients may have persistent
or recurrent bleeding. Much of the morbidity and mortality of upper gastrointestinal
bleeding occurs in patients with recurrent bleeding or significant co-morbid
illnesses.
Hospital mortality has not improved over the past 50 years and remains at about 10%. This may in part be due to the fact that older patients, who have advanced cardiovascular, respiratory, or cerebrovascular disease that puts them at increased risk of death, now comprise a much higher proportion of cases. Therapeutic endoscopy is considered a safe and effective form of treatment today (5). Analysis of clinical and endoscopic factors permits accurate risk assessment, rational treatment planning and improved outcome.
1.1 Grading of recommendations (6)
Grade A
Evidence from large, randomised clinical trials or meta-analyses
Grade B
High quality study of non-randomized cohorts who did not receive therapy or
high quality case series
Grade C
Opinions from experts based on arguments from physiology bench research or
first principles
1.2 Definitions (6)
Haematemesis
Vomiting fresh, red blood
Melaena
Passage of black, tarry stools
Haematochezia
Passage of red blood per rectum
Rebleeding
Fresh haematemesis and/or melaena associated with the development of shock
(pulse >100 bpm.; Systolic BP <100 mmHg), a fall in CVP>5 mmHg, or
a reduction in Haemoglobin >2gm% over 24hrs.
2. AETIOLOGY
The commonest cause of Non-variceal UGIB is peptic ulcer disease. A history
of proved ulcer or ulcer-like dyspepsia is absent in about 20% of cases. In
these patients consumption of aspirin or non-steroidal antiinflammatory drugs
(NSAIDS) is common. Infection with Helicobacter pylori (H. pylori) is less
prevalent in bleeding ulcers than in uncomplicated ulcers (7, 8).
2.1 Peptic ulcers
Peptic ulcer bleeding occurs predominantly from duodenal ulcer or gastric
ulcers. It occurs as a result of erosion of blood vessels and the severity
of the bleed is dependent on the size of the vessel affected. Simple oozing
is caused by damage to small submucosal vessels less than 0.1mm in diameter.
More severe arterial bleeding indicates a large vessel between 0.1 and 2mm
in diameter in the base of the ulcer has been eroded by the inflammatory process.
Large ulcers arising from the posterior part of the duodenal cap can erode
the gastroduodenal artery and provoke brisk bleeding.
2.2 Erosions
Acute erosive gastritis can cause persistent haemorrhage as a result of diffuse
loss of mucosal epithelium and small ulcers. This condition is often associated
with the use of non-steroidal antiinflammatory drugs, steroids and intake
of alcohol. Haemorrhagic gastritis which probably occurs as a result of impaired
mucosal blood flow is often caused by stressful stimuli including shock, hepatic
failure and head injury.
2.3 Oesophagitis
Oesophagitis usually only causes minor acute bleeding. Occasionally a significant
vessel may be involved with consequent massive arterial haemorrhage.
2.4 Mallory-Weiss tear
This is an acute tear at the gastro-oesophageal junction as a result of severe
vomiting or retching, often after excessive alcohol intake. Mallory-Weiss
tear occurs mostly in the gastric mucosa, but may extend into the oesophagus
resulting in profuse vomiting of bright red blood which usually settles spontaneously.
Endoscopic haemostasis may sometimes be required. Occasionally repeated vomiting
may result in a full thickness tear (Boerrhaave's syndrome) which is associated
with sudden onset of severe pain in the upper abdomen or chest.
2.5 Malignancy
Carcinoma and lymphoma of the stomach commonly bleed at an advanced ulcerated
stage, and occasionally present with acute haemorrhage. The prognosis is usually
dictatated by the stage of the disease.
2.6 Miscellaneous
There are several other causes which may present as upper non-variceal gastrointestinal
haemorrhage and these are listed in Table 1.
Table 1: Aetiology of Non-variceal upper gastrointestinal bleeding (8)
| Oesophagus |
| Mallory-Weissi tear, Reflux oesophagitis, Oesophaneal ulcer, Barret's ulcer, Cameron ulcer within hiatus hernia*, Oesophageal neoplasm |
| Stomach |
| Gastric ulcer, Gastric erosions, Haemorrhagic gastritis, Gastric carcinoma, Gastric lymphoma, Leiomyoma, Gastric polyp, Hereditary haemorrhagic telangiectasia, Dieulafoy lesion*, Gastric Antral Vascular Ectasia(GAVE)*, Angiodysplasia* |
| Duodenum |
| Duodenal ulcer, Duodenal erosions, Vascular malformation, Aorta-duodenal fistula, Polyps (including Peutz-Jeghers syndrome and other polyposis syndromes), Carcinoma of ampulla, Carcinoma of pancrease, Haemobilia* |
| Small bowel |
| Stomach ulcer, Diverticulum (including Meckel's diverticulum), Vascular malformation, Tumor |
*Important causes of obscure UGIB
3. EPIDEMIOLOGY
UGIB is a common reason for emergency admission to hospitals. A recent large
prospective study from the United Kingdom reported an overall incidence of
103 per 100,000 adults per year, with an overall mortality of 14% but only
0.6% for those below 60 years of age without comorbidity, Most deaths were
in elderly patients with considerable comorbidity (1). A retrospective study
from USA also showed a similar incidence of 102 per 100,000 adults (2). Figures
available from a small prospective study from Singapore more than a decade
ago showed an overall mortality of 10% (9).
The incidence of UGIB is twice as high in men as in women. The incidence increases markedly with age. Consequently, many patients presenting with UGIB have an active comorbid condition, a consistent risk factor for increased mortality.
A recent local multicentre prospective study has provided new information on the epidemiology of UCIB in Malaysia (3). Recruiting 1,830 patients from four government hospitals in East Malaysia over a period of two years, the study reported an incidence of UGIB of 72 per 100,000 population and this peaked around the 4th to 6th decade. Mortality rate from UGIB was 10.2% but increased substantially with age and did not differ between the sexes. Inpatients that were admitted for other diagnosis but developed UGIB had the highest mortality; at almost 5 times higher than those with emergency admissions or transfers from other hospitals for UCIB. 64% of those admitted in this series had peptic ulcer disease as a cause of bleeding. The next most frequent cause is mucosal erosive diseases at 16.5%. Variceal bleed accounted for 6.4% and malignancies 3.6%. Almost 9% of patients had no discernable cause for UGIB after endoscopy. These results are comparable to other reported series. Interestingly, 1 in 7 patients with variceal bleed also had concomitant peptic ulcer disease. Peptic ulcer disease has been documented as the main aetiological factor in UGIB in Malaysia. There are comparatively fewer cases of peptic ulcer disease in the Malay-majority east coast states of Terengganu and Kelantan, thought to be related to a low H. pylori infection rate (10-12). The Chinese patients predominate in the UGIB cases in urban centres such as lpoh and Kuala Lumpur (13, 14). In Kota Kinabalu, comparing the distribution of the UGIB cases with that of the racial distribution in the state, the Chinese and the Kadazandusuns, one of the major indigenous groups, have a significantly higher rate of UGIB than expected and, reflecting the situation in West Malaysia, the Malay and Bajau ethnic groups have the lowest rate and relative risk of UGIB (4).
4. RISK FACTORS
4.1 Nonsteroidal anti-inflammatory drugs (NSAIDs)
Some 15-30% of patients exposed to NSAIDs develop gastroduodenal ulcers, but
the rate of serious gastrointestinal events such as bleeding, perforation
and obstruction is approximately ten times less (15). The occurrence of injury
may depend on clinical factors, type and dose of NSAID used. The use of NSAIDs
has also been found to be a risk factor for peptic ulcer rebleeding in some
studies. It may increase the risk of ulcer complications by a factor of four
(16). The risk of bleeding increases with older age, the use of conventional
NSAIDs, co-rnorbid conditions, concomitant ingestion of steroids and anticoagulants
and a prior history of gastrointestinal complications (17). The advent of
the selective COX-2 inhibitors holds promise in terms of reduced gastrointestinal
toxicity (18).
4.2 Aspirin
Exposure to aspirin carries a definite risk of gastroduodenal injury (19).
The risk of upper gastrointestinal bleeding is similar among users of non-coated
low-dose aspirin and coated low-dose aspirin. Any dose of aspirin has the
potential to cause gastrointestinal bleeding, the enteric-coated form carrying
the same risk as plain aspirin (20). The concomitant use of NSAIDs increases
the risk of complications. A history of UGIB is a significant risk factor
for recurrent bleeding in those taking low-dose aspirin or other NSAIDs (21).
4.3 Helicobacter pylori
H. pylori is the main cause of uncomplicated peptic ulcer disease. The benefit
of H. pylori eradication to decrease ulcer recurrence and bleeding after eradication
of H. pylori infection at the index episode of bleeding is well established
for those lesions that have H. pylori as the sole etiological factor (22-24).
The protective effect afforded by H. pylori cure in this setting provides the same level of protection as that of continuous antisecretory therapy (25).
The mode of H. pylori testing in the setting of UGIB may be of importance. In particular, biopsy-based urease tests may be false-negative (26).
The interaction between NSAIDs and H. pyiori remains controversial (27). Eradication treatment may be appropriate for patients who had sustained UCIB secondary to low-dose aspirin, while treatment with omeprazole appears the best strategy for the prevention of recurrent-bleeding from NSAID-induced ulcers (28). H. pylori contributes to an increased ulcer risk for patients starting NSAID treatment, whereas NSAIDs probably account for the majority of ulcer disease in chronic NSAID users. The eradication of H. pylon substantially reduces the risk of ulcers for patients who are about to start long-term NSAIDs (29).
5. ASSESSMENT, RESUSCITATION AND RISK STRATIFICATION
5.1. Clinical presentation
Acute upper gastrointestinal bleeding presents with either haemetemesis or
melaena or both. Haemetemesis with bright red vomitus indicates acute bleeding
while recent bleeding appears as "coffee grounds" vomitus due to
gastric acid breaking down the haemoglobin in red cells to haematin. Melaena
consists of black, tarry, loose or sticky and malodorous stool due to degradation
o( blood in the intestine (30, 31).
Table 2: Upper Gastrointestinal Bleeding: Clinical Situations
| Acute |
| 1. Haemetemesis with or without melaena 2. Malaena with or without haemetemesis 3. Rarely haematochezia indicating massive life threatening bleed |
| Chronic |
| 4. Iron deficiency anaemia with or without evidence of visible blood
loss 5. Blood loss detected bv positive Occult Blood Test |
5.2 Patient assessment
Close monitoring of blood pressure, pulse and gross evidence of ongoing bleeding
is mandatory. Agitation, pallor, hypotension and tachycardia may indicate
shock requiring immediate volume replacement. Shock occurs when blood loss
approaches 40% of the total blood volume. Postural hypotonsion of 10 mmHg
or higher usually indicates at least 20% reduction in blood volume. Initial
haematocrit obtained for a patient with acute bleeding poorly reflects the
degree of blood loss due to haemoconcenlration. The immediate goal is to resuscitate
the patient to ensure a stable haemodynamic status prior to endoscopy.
5.3 Clear airway
A drowsy or comatose patient is at high risk of aspiration if vomiting or
haemetemesis continues. The patient is kept flat on his/her side. A cuffed
endotracheal tube may be inserted to protect the airways if needed.
Table 3: Mental status may be impaired due to
| 1. Cerebral hypoperfusion due to severe acute blood loss 2. Encephalopathy due to concomitant chronic liver disease or renal failure 3. Alcohol or drug intoxication/overdose |
5.4. Resuscitation
An immediate assessment of haemodynamic status and red cell transfusion requirements
must be made. A confused, clammy and sweaty patient with cold peripheries
and a fast thready pulse suggest hypovolaemia. The blood pressure may be low.
Although no controlled trials have examined each of the elements of resuscitation,
the following are recommended (32) (Grade C).
Resuscitation must be commenced immediately with the insertion of at least two large bore intravenous cannulae, which should be inserted into large peripheral veins. Supplemental oxygen may help a confused, agitated elderly patient. Central venous pressure (CVP) monitoring is advisable in patients with profound shock or organ failure and in elderly patients with significant comorbidity. Given the lack of evidence for colloids, crystalloids are the choice for fluid resuscitation. Fluid resuscitation can be commenced with isotonic crystalloid solutions (0.9N saline, lactated Ringer's solution). Blood samples must be drawn for urgent full blood count, blood grouping and cross matching, coagulation screen, blood urea and electrolyte and liver function tests.
Table 4: Blood tests on admission to hospital (6, 7)
| Haemoglobin - May be normal during the acute stages until haemodilution
occurs Urea and electrolytes - Elevated blood urea suggests severe bleeding Cross match for transfusion - Two units of blood are sufficient unless bleeding is extreme. If the transfusion is not needed urgently, group the blood and save the serum Liver function tests Prothrombin time |
The evidence to guide red cell transfusion is limited (Grade C). Packed cells are the preferred form of blood transfusion. The aim of transfusion is to restore blood volume and pressure and to correct anaemia to maintain the oxygen carrying capacity. This means maintaining a haemoglobin level of approximately 10 g/dl. Fresh frozen plasma may be given if the prothrombin time is at least 1.5 times higher than the control value. Platelet transfusion is indicated if the platelet count is below 50,000/mm3.
Table 5: Indications for blood transfusion in patients with gastrointestinal
bleeding (7, 8)
| 1.Systolic BP < 110 mmHg 2. Postural hypotension 3. Pulse > 110/min 4. Haemoglobin < 8g/dl 5. Angina or cardiovascular disease with a Haemoglobin < l0g/dl |
5.5 Assessment of ongoing bleeding
Continuous haemetemesis or persistent hypovolaemia despite aggressive resuscitation
suggest bleeding is still active. Passage of melaena does not imply continuous
bleeding. However passage of "fresh" melaena, which is maroon coloured
or passage of bright red visible clots suggest active bleeding. The insertion
of a nasogastric tube may be helpful in demonstrating active bleeding. However,
it may be poorly tolerated. The caveat is when there is a bleeding ulcer with
the pylorus in spasm. Aspirate without evidence of blood or "coffee-grounds"
material is seen in about 15% of patients with UGIB (32).
5.6 Elucidating the source of bleeding
A good history from the patient or relatives and a quick examination help
suggest the aetiology of the gastrointestinal bleeding. Ask for history of
retching (Mallory Weiss tear), history of NSAIDS or aspirin, previous peptic
ulcer disease, dyspepsia, ethanol ingestion, traditional medication and a
history of hepatitis B or C. Examine for stigmata of chronic liver disease
(e.g. palmar erythema, spider naevi, etc), features of portal hypertension
(eg ascites, splenomegaly, caput medusae), cutaneous and buccal telangiectasia
(Osler-Weber-Rendu syndrome).
5.7 Risk assessment
When a patient presents with gastrointestinal bleeding, risk assessment and
resuscitation proceed simultaneously. Such assessment aids in rational decision
making regarding treatment options. At the initial assessment it is important
to define the factors that have prognostic importance (33) (Table 6). The
main factors predicting death include increasing age, comorbidity and endoscopic
findings. Mortality is low in patients below 40 years of age but increases
steeply thereafter. Patients with severe comorbidity, particularly renal failure,
liver failure and disseminated malignancy have a poor prognosis (Grade A).
Death in these patients is more often due to disease progression rather than
to the upper gastrointestinal bleeding. Patients who developed UGIB after
hospitalisation for other serious illnesses have a much worse prognosis than
those who are admitted because of bleeding, with a mortality of about 30%.
Endoscopic findings of active, spurting haemorrhage and a non-bleeding visible vessel within an ulcer are associated with a definite risk of rebleeding (Tables 7, 8). The absence of these stigmata, varices or upper gastrointestinal cancer indicates a low risk of rebleeding (Grade A).
Table 6: Risk factors for death after hospital admission for acute upper gastrointestinal bleeding (33)
| 1. Advanced age 2. Shock on admission (pulse rate >100 beats/min; systolic blood pressure < 100 mmHg) 3. Comorbidity (particularly hepatic or renal failure and disseminated malignancy) 4. Diagnosis (worst prognosis for advanced upper gastrointestinal malignancy) 5. Endoscopic findings (active, spurting haemorrhage from peptic ulcer; non-bleeding visible vessel) 6. Rebleeding (increases mortality 10 fold) |
Table 7: Forrest Classification for bleeding peptic ulcer (34)
| Ia: Spurting Bleeding IIb: Non spurting active bleeding IIa: Visible vessel (no active bleeding) IIb: Non bleeding ulcer with overlying clot (no visible vessel) IIc: Ulcer with hematin covered base III: Clean ulcer ground (no clot, no vessel) |
SRH = Stigmata of recent haemorrhage
Major SRH = Forrest la, 1b, 2a and 2b
Minor SRH = Forrest 2c and 3
5.7.1 Endoscopy for risk assessment
Early upper gastrointestinal endoscopy (within 12-24 hours) is the cornerstone
of management of UCIB. Early endoscopy has 3 major roles viz. diagnosis, treatment
and risk stratification. It is the most accurate method available for identifying
the source of bleeding. Once the Forrest grade of ulcer is identified (Table
7; Appendix 2), a risk assessment may be made and a decision made on whether
ongoing hospitalisation is needed. Recently, a number of studies have indicated
that systematic assessment of clinical and endoscopic risk factors (endoscopic
triage) may obviate hospitalisation in some patients and may help in determining
the appropriate length of stay in others (36-38). Those determined to be at
low-risk based on clinical and endoscopic criteria were discharged on the
day of presentation and received out-patient care (39, 40). The aforementioned
findings have led to the development of practice guidelines and clinical care
pathways for UGIB (41, 42) with some incorporating an initial phase of endoscopic
triage (43).
Table 8: Risk of rebleeding and mortality in patients with peptic ulcer
bleeding (35)
|
Endoscopic finding (SRH)*
|
Risk of Rebleeding (%)
|
Mortality(%)
|
| Active Bleeding Visible Vessel Adherent Clot Flat Spot Clean Base |
55
43 22 10 5 |
11
11 7 3 2 |
*SRH=Stigmata of recent haemorrhage
5.7.2 Use of risk stratification scoring systems
A number of scoring systems have been designed to ascertain risk factors for poor outcome in patients with UGIB (33, 36, 38). One such system (Rockall risk assessment score), derived from the data of a national audit, is based on age, presence of shock, co-morbidity, diagnosis and endoscopic stigmata of recent haemorrhage (33). In the Rockall risk assessment score, a series of independent risk factors were scored (Appendix 1). Patients who score 2 or less have a mortality of 0.1% and a rebleeding rate of 4.3%, but a score in excess of 8 is associated with a 41% mortality and rebleeding rate of 42.1%. The score was more reliable in predicting mortality than it was in predicting rebleeding (Grade A). Such risk assessment scores may be useful in triaging patients for either outpatient care or admission to a high dependency unit (37).
6. ENDOSCOPIC THERAPY
Endoscopic therapy has been shown to improve outcome in nonvariceal haemorrhage
(Grade A). In a recent meta-analysis of 30 randomized trials involving more
than 2000 patients, endoscopic therapy reduced rates of further bleeding (OR
0.38; 95% confidence interval:0.32 to 0.45), the need for urgent surgery (OR
0.36; 95% CI:0.28 to 0.45), and mortality (OR 0.55; 95% Cl:0.40 to 0.76) (44).
Early gastroscopy is very valuable as a therapeutic and prognostic instrument,
decreases rates of blood transfusions and significantly reduces hospital length
of stay (45) (Grade A).
After resuscitation, endoscopy is undertaken. In most cases this is done electively on the next available routine list but within 24 hours of admission. Only a minority of profusely bleeding patients need "out of hours" emergency endoscopy. On-call endoscopists must be experienced and be able to apply a range of endoscopic treatments.
Endoscopic therapy is indicated when there are major stigmata of recent haemorrhage (SRH). There is little doubt that Forrest la, 1b and 2a ulcers should have endoscopic haemostasis (5, 35, 46, 51) (Grade A). Patients with an adherent clot may also constitute a high-risk group. Up to one-third of blood clots covering an ulcer can be removed to reveal major stigmata of recent haemorrhage. Current opinion favours the displacement of the clot by irrigation or mechanical removal, followed by endoscopic haemostasis of any underlying visible vessel (47-50) (Grade A). Minor SRH i.e. Forrest 2c and 3 ulcers, may be managed conservatively and discharged early.
The various modalities of endoscopic haemostasis are outlined in Table 9.
Table 9: Endoscopic treatment for non-variceal upper gastrointestinal bleeding
(7, 8)
| Thermal |
| Heater probe Multipolar electrocoagulation (BICAP, Gold Probe) Argon plasma coagulation Laser |
| Injection |
| Adrenaline (1:10,000) Procoagulants (fibrin glue, human thrombin) Sclerosants (ethanolamine, 1% polidoconal) Alcohol (98%) |
| Mechanical |
| Clips Band Ligation Endoloops Staples Sutures |
| Combination therapy |
| Injection plus thermal therapy Injection plus mechanical therapy |
Methods rarely used are depicted in italics.
6.1 Injection Therapy
6.1.1 Adrenaline
In experimental animal studies, mucosal injection of 1:10,000 dilution adrenaline
causes prolonged vasoconstriction for up to 2 hours. Adrenaline also causes
platelet aggregation and a local tamponade effect on the vessel when injected
in large volumes. A total volume of 4-16 ml (1 :10,000) may be injected safely
(6) as most of the adrenaline will undergo first-pass metabolism in the liver.
There are few systemic complications other than transient tachycardia. Adrenaline
is the injection agent of choice, because it is non-tissue damaging. After
bleeding has been controlled, a clear view of the vessel is then possible.
Adrenaline injection has reduced hospital stay, transfusion requirement and operative intervention (41% to 15%). The rebleeding rate in this randomized trial was 15% for adrenaline therapy versus 41% for controls in actively bleeding ulcers (51) (Grade A). It remains the gold standard (5, 52). It is cheap, easily available and achieves control in actively bleeding ulcers. It is an essential component of combination therapies.
6.1.2 Sclerosants
1% polidocanol, alcohol and ethanolamine are sclerosants used in ulcer haemostasis.
In animal studies, 1% polidocanol causes haemostasis by inducing bowel wall
spasm and early oedema with subsequent inflammation and thrombosis of the
vessel.
Absolute alcohol stops bleeding by causing rapid dehydration and fixation of the tissue, thus obliterating the bleeding vessel. The amount of tissue damage is directly related to the volume of sclerosant injected. Alcohol, being more ulcerogenic, induces ulceration which lasts for a longer period. It may reduce rebleeding as well as emergency surgery rates (53) (Grade B).
In view of the risk of perforation, caution should be exercised when injecting large volumes of sclerosant. Fatal gastric necrosis has been reported (54).
The addition of a sclerosant to the vessel after initial adrenaline injection has not conferred any advantage over adrenaline injection alone (55,56) (Grade A).
6.1.3 Procoagulants (Thrombogenic agents)
Human thrombin and fibrin sealant are procoagulants that have been investigated
in ulcer haemostasis. Human thrombin after epinephrine injection has been
compared with epinephrine injection alone. Significant reductions in recurrent
bleeding, blood transfusion and deaths were observed in the combined treatment
group (57).
Fibrin glue is a formulation of fibrinogen and thrombin which when combined instantly forms a fibrin network. The two substances are injected via a double-lumen needle. The advantage of fibrin injection is that very little tissue damage occurs, therefore reducing the risk of tissue necrosis and perforation and allowing repeated injections. In an European multicentre trial, patients with actively bleeding ulcers or ulcers with non-bleeding visible vessels were randomized to receive a single injection of polidocanol, single fibrin sealant injection and daily fibrin injection until clean ulcers were seen. Fibrin sealant significantly reduced recurrent bleeding only if injected daily. A programme of daily scheduled endoscopy and repeated treatment has been advocated (58, 59). There is concern regarding viral transmission with the use of fibrin glue.
6.1.4 Technique of injection therapy
A therapeutic video-gastroscope (3.7 or 4.2mm working channel) with a disposable
23 or 25 gauge sclerotherapy needle are recommended. Between 4-16 ml of 1:10,000
adrenaline in 0.5ml aliquots is injected into and around the bleeding point
until the bleeding stops (6).
Dehydrated ethanol (98%, Abbott Laboratories) is injected, using a 1ml disposable plastic tuberculin syringe, with a total dose of no more than 1.5-2 ml. The ethanol is injected slowly, in amounts of 0.1 to 0.2ml per injection, at three or four sites surrounding the bleeding vessel and 1 or 2mm from the vessel (60). Polidocanol is less irritant and 10-15 ml have been used.
6.2 Thermal modalities
This can be divided into contact and non-contact methods. A distinct advantage
of contact over non-contact electrocoagulation is that mechanical pressure
can be applied to the bleeding point using the electrode to compress the bleeding
vessel prior to coagulation. The principle of coaptive coagulation is that
a combination of mechanical compression and heat treatment produces a stronger
sealing of the blood vessel compared to non-mechanical treatment.
6.2.1 Thermal contact methods
6.2.1.1 Monopolar electrocoagulation
In monopolar electrocoagulation the current flows through the patient and
exits via a ground plate. Due to an unpredictable depth of coagulation, monopolar
electrocoagulation is no longer recommended for endoscopic haemostasis (60).
6.2.1.2 Multipolar electrocoagulation
A multipolar electrocoagulation probe consists of 3 pairs of electrodes arranged
in a linear array at the tip and connected to a power generator. The flow
of the electrical current is limited between the electrodes on the probe thus
avoiding problems with grounding and aberrant current. The depth of injury
is shallower and more predictable compared to monopolar electrocoagulation.
Small 7Fr and large 10Fr probes (BICAP, Gold Probe) are available for use
with 2.8mm and 3.7mm channel endoscopes respectively. Optimal effect can be
obtained by using a large 3.7mm probe with a low power setting of 3-5 on the
generator and prolonged coagulation using 10-14 pulses of 2 seconds (60).
The efficacy of BICAP is similar to that of the heater probe (61) (Grade B).
6.2.1.3 Heater probe
The tip of the heater probe consists of a metal tip covered by Teflon which
is heated by a computer-controlled coil to a temperature of 250°C. Practically
this requires (i) forceful tamponade using a 3.2mm probe and (ii) sustained
coagulation with 4 consecutive pulses at 30J for at least 8 seconds (62}.
The heater probe is useful because it includes a water jet to wash away any
blood.
In general, heater probe and adrenaline injection are comparable in their efficacy (63) (Grade A). The rebleeding rates with the use of heater probe alone in comparison with laser or controls, and in comparison with laser or BICAP was not statistically significant (64).
6.3 Combination therapy
The addition of heater probe therapy to epinephrine injection in the subgroup
of patients with a spurter significantly reduced the need for surgery when
compared to epinephrine injection alone (65) (Grade B). Most studies, however,
have not demonstrated any added benefit in combining injection therapy with
thermal coagulation. The latter notwithstanding, the current trend favours
combination therapy using injection as well as thermal or mechanical therapy
(8).
6.4 Thermal non-contact methods
6.4.1 Argon plasma coagulation
Argon plasma coagulation (APC) is a special electrosurgical modality in which
high-frequency electric current is conducted 'contact-free' through ionized
and thus electrically conductive argon (argon plasma) into the tissue to be
treated. The aim of this technique is to create therapeutically effective
temperatures for thermal haemostasis and/or the ablation of pathologic tissue.
In haemostasis, APC is especially useful for diffuse bleeding arising from a large area, bleeding owing to coagulation disorders or tumour bleeding. It has been used successfully to treat gastric antral vascular ectasia (GAVE) (69), angiodysplasia and haemorrhagic telangiectasia. Reported complications (< 1%) include bowel wall emphysema, pneumomediastinum and perforation.
Treatment of bleeding ulcers with APC does not appear to confer any advantage over the heater probe for endoscopic haemostasis (70) (Grade B).
6.4.2 Laser
Several trials comparing the methods of monopolar, multipolar, and heater-probe
electrocoagulation with Nd:YAG and argon laser, as well as the injection modalities
of adrenaline, ethanol and polidocanol revealed that all methods were effective
in lowering the incidence of rebleeding and the need for emergency surgery
(64, 66-66). Complications of laser therapy include perforation, bleeding,
fistula formation and stenosis.
Laser therapy is currently not recommended (6) (Grade B).
6.5. Other endoscopic modalities
6.5.1 Mechanical methods
Endoscopic placement of metal clips has recently been advocated for haemostasis
(Grade B). One prospective randomised trial compared haemoclips with a thermal
modality. Acute rebleeding occurred in 1.8% of the haemoclip patients compared
with 21% of heater probe patients (p<0.05). The median number of blood
units transfused and hospital days was also significantly lower for haemoclips.
There was no difference in emergency surgery rate or 30-day mortality (71).
Haemoclips maybe particularly useful for actively bleeding large vessels but may be difficult to apply in awkwardly placed ulcers (e.g. high lesser curve or posterior duodenal ulcers) (6, 72).
7. Pharmacological therapy
In vitro, platelet aggregation and disaggregation, coagulation and fibrinolysis
are strongly dependent on intragastric pH. Platelet aggregation and blood
coagulation are optimal at pH 7.4 (73). Peptic digestion of the thrombus is
maximal in the pH range of 1-3.5 and pepsin 1 may continue to function up
to pH of 5 (74). Platelet aggregation is also severely impaired at low pH
in-vitro. As blood coagulation and platelet aggregation are abolished at pH
lower than 5.4, the perceived failure of traditional antisecretory drugs to
promote haemostasis in bleeding peptic ulcers may reflect inadequate pH control.
Acid suppressive therapy also decreases the increased fibrinolytic activity
noted in bleeding ulcers (75). This has provided the rationale use of more
potent acid reducing agents such as proton pump inhibitors in the management
of peptic ulcer bleeding.
7.1 H2-Receptor Antagonists
A meta analysis examining 27 randomised trials of cimetidine or ranitidine
in the treatment of UGIB involving more than 2,500 patients showed no significant
difference between H2 antagonist therapy (21%} and placebo (23%) (76). In
fact, only one of the 27 individual trials reported a significant decrease
in rebleeding with a H2 antagonist therapy. A large, prospective, randomized,
double-blind, placebo-controlled trial evaluating the use of famotidine in
acute bleeding peptic ulcer found that recurrent bleeding rates, need for
surgery and the number of deaths were no different between the two groups
(77). A recent meta-analysis concluded that there was no evidence to support
the use of H2-receptor antagonists in the treatment of bleeding duodenal ulcers
but there is evidence of a moderate benefit in gastric ulcers (78).
The use of H2 antagonists in upper gastrointestinal bleeding is not reconnmended (Grade A).
7.2 Proton Pump Inhibitors (PPIs)
The use of intravenous boluses of omeprazole in comparison with placebo revealed
less endoscopic evidence of persistent bleeding in the omeprazole treated
patients but other end points, including mortality, were similar in both groups
(79). A single centre study revealed that high dose oral omeprazole resulted
in less rebleeding and lower transfusion requirements when compared to placebo
(80). Endoscopic therapy was not used in this trial. Various trials have compared
the use of high dose intravenous omeprazole with placebo following primary
endoscopic haemostasis (81-84). The most convincing study (84) revealed a
significant reduction in rebleeding within 30 days in the omeprazole group
(6.7%) in comparison with placebo (22.5%). Although the mortality rate and
the number requiring surgery were also lower in the omeprazole group, these
differences were not significant.
Intravenous pantoprazole has also been used in peptic ulcer bleeding. Pantoprazole infusion was compared with ranitidine in patients with Forrest la, lb, lla and llb after undergoing endoscopic haemostasis with adrenaline or adrenaline with polidocanol. There was a tendency for lower rebleeding rate in the pantoprazole group as opposed to ranitidine (85).
In a meta-analysis comparing proton pump inhibitors with H2 antagonists, it was observed that persistent or recurrent bleeding was less frequent with proton pump inhibitors (6.7%) than with H2 antagonists (13.4%) (OR 0.4; 95% Cl:0.27-0.59%). The need for surgery and mortality rates did not reach statistical significance but showed a favourable trend towards PPIs. When the analysis was stratified according to endoscopic therapy, only the subgroup of patients who were not treated endoscopically showed a significant reduction in persistent or recurrent bleeding (OR 0.24; 95% Cl:0.13 to 0.14) (86).
It is recommended that following endoscopic therapy in major peptic ulcer bleeding, high dose intravenous PPI (eg IV Omeprazole 80 mg stat followed by an infusion of 8 mg hourly for 72 hours) be commenced (Grade B).
8. Management of other causes of UGIB
8.1 Mallory-Weiss tears
Occassionally endoscopic therapy is required to arrest severe bleeding. Adrenaline
(87), thermal methods or mechanical clips (88} have been used (Grade C).
8.2 Vascular malformations (including telangiectasia and
GAVE)
Multiple sessions of Argon plasma coagulation (APC) or heater probe therapy
(89) may be required to achieve haemostasis (Grade B).
8.3. Dieulafoy lesion
Uncontrolled series report success with band ligation, injection and thermal
methods (90) (Grade G).
9. Aftercare
After the initial endoscopy and the institution of endoscopic therapeutic
measures where necessary, the key point in the aftercare is the recognition
of patients at high risk of rebleeding and death who would require careful
monitoring in an intensive care or high dependency setting. Predictors of
an increased risk of rebleeding and death (as well as failure of endoscopic
therapy) include (i) clinical factors such as shock at the time of presentation,
advanced age, co-existing illnesses, (ii) endoscopic features such as ulcer
location (posterior duodenal ulcer), size of the ulcer (>2cm), stigmata
of recent haemorrhage and the presence of blood at the time of endoscopy as
well as (iii) laboratory features such as haemoglobin (<10g/dl) and elevated
blood urea levels (91-95). While there is some controversy as to which of
these factors are more important in predicting rebleeding and death, an overall
picture emerges that having a severe initial bleed, being elderly and having
coexisting severe illnesses increases the risk of an adverse outcome. While
rebleeding is an important cause of death, mortality could occur in the absence
of rebleeding especially in patients with coexisting illnesses.
The role of second-look endoscopy is unclear. Published studies on the routine use of second-look endoscopy consist of inadequate numbers (96-98). The difference is probably marginal.
10. Management of rebleeding
Recurrent bleeding remains the single most important adverse prognostic factor.
Morbidity and mortality are higher in those with rebleeding and 95% of rebleeding
occurs within the first 72 hours of hospitalisation (99).
10.1 Rebleeding After Initial Endoscopic Control of Bleeding Ulcers
The major challenge in applying endoscopic therapy for bleeding peptic ulcers
is that haemostasis is not permanent and re-bleeding occurs in about 15-20%
of the cases.
Endoscopic treatment would avoid the surgical risk. However, delay in establishing haemostasis may result in hypotension and adversely affect the survival.
In patients with peptic ulcers and recurrent bleeding after initial endoscopic control of bleeding, endoscopic retreatment reduces the need for surgery without increasing the risk of death and is associated with fewer complications than is surgery (100) (Grade A).
Surgery if decided upon should be performed early rather than late to avoid an unfavorable outcome especially in the hypotensive elderly patient. In some patients, endoscopic appearances (eg a giant posterior duodenal ulcer) may suggest that surgery be the preferred option (6) (Grade C).
11. Role of surgery
The role of surgery has changed with wider use of endoscopic hemostasis in
bleeding ulcers, no longer aiming to cure the disease but primarily to stop
the hemorrhage. Mortality after urgent surgery correlates with the preoperative
Apache 2 score (101).
11.1 Indications for Surgery as the Primary Mode of Treatment
11.1.1. Massive bleeding
There is still no proven alternative to emergency operation for massive bleeding
uncontrolled by endoscopic procedures. This may be due to bleeding that is
unresponsive to endoscopic hemostasis or failure of endoscopic visualization
of the bleeder due to profuse hemorrhage. A continued attempt with endoscopic
treatment is futile and dangerous.
11.1.2. Ulcer inaccessible to endoscopic control
There are situations where the bleeding ulcer is inaccessible to endoscopic
control. This can occur in duodenums that are often deformed and narrowed.
Primary surgery is indicated in such circumstances. The rate oi primary-emergency
surgery varies depending on the case mix and the expertise of endoscopic management.
Thus the surgeon should be involved from the outset in the team caring for the patient. Early and close cooperation between endoscopists/gastroenterologists and surgeons is vital.
11.2 Type of surgery for bleedins peptic ulcer
There appears to be no difference between local (under-running/over-sewing
or excision of ulcer) and radical surgery (gastric resection or vagotomy)
with respect to mortality although rebleeding rate may be higher in the local
group (102).
While under-running or over-sewing for bleeding ulcers is advisable in a large proportion of cases, ulcer excision or even more radical surgery (e.g. gastric resection for large, chronic, penetrating gastric ulcers) may be performed in selected cases. There is only one trial of different surgical procedures for bleeding duodenal ulcers (103). The rebleeding rate was lowest in patients having a gastrectomy to include the ulcer either with Billroth I or Billroth II reconstruction when compared with more conservative surgery. However, the bile leak following gastrectomy was much higher and the overall mortality was similar in the two randomized groups. The same study suggested that when a bleeding duodenal ulcer is under-run, ligation of the gastroduodenal and right gastroepiploic arteries reduced the rebleeding rate to a similar level as gastrectomy (Grade B).
Currently it is not possible to make definite recommendations in the absence of any good prospective randomized trials. The magnitude of surgery should be tailored to the type of ulcer, severity of illness in the patient and experience of the surgeon.
12. Interventional Radiology
In the critical or unstable patient who is not amenable to immediate surgical
intervention radiological intervention appears increasingly as a very effective
option. In a recent retrospective evaluation of interventional embolization
therapy over an 8 year period, bleeding was stopped in 83% of cases. The rate
of complications was 14%. Sodium diatrizoate, metal coils, tissue adhesives
and Gelfoam particles were used (104).
Follow up
Patients admitted for bleeding peptic ulcer should be discharged with oral
proton pump inhibitors. Those with gastric ulcers should be re-endoscoped
in 6 weeks to assess healing and rule out malignancy. Attention should be
paid to H. pylori eradication for all H. pylori positive ulcers. The latter
is also recommended for those on long-term aspirin. Those who need to continue
on NSAIDs should consider COX-2 inhibitors, or the least damaging NSAID with
a proton pump inhibitor (Section 4.3).
Summary
Upper gastrointestinal bleeding is one of the most common medical emergencies.
In any patient with UGIB, history-taking, physical examination and resuscitation
need to proceed simultaneously. Endoscopy has become the cornerstone of diagnosis,
risk stratification and treatment of peptic ulcer bleeding. Clinical assessment
and endoscopic recognition of stigmata of recent haemorrhage can allow the
identification of patients with a high risk of rebleeding. Upper endoscopy
to assess the risk of rebleeding in patients with non-variceal upper gastrointestinal
bleeding may be used for triage, allowing outpatient care of selected patients
and leading to significant cost savings.
Patients with active bleeding at the time of endoscopy and with non-bleeding visible vessels should receive endoscopic treatment. Endoscopic methods can be divided into thermal (multipolar coagulation, heater probe, argon plasma coagulalor) and nonthermal (injection therapy, mechanical clips); both types are effective. A combination of injection and thermal therapy with initial injection of adrenaline followed by coagulation of the identified vessel is recommended. Bleeding recurs in 15% of patients. In selecled patients with recurrent ulcer bleeding, endoscopic retreatment may be attempted. If this fails to provide haemostasis, surgery is recommended. High-dosage parenteral proton-pump inhibitors after endoscopic treatment of bleeding peptic ulcers are recommended to reduce the risk of rebleeding. In selected patients with failed endoscopic therapy who are unfit for surgery, interventional radiology may be considered.
All medical and general surgical units need to be familiar with the management of UGIB and close collaboration between medical and surgical teams is essential. In addition, trained endoscopy assistants and interventional radiologists are all important team members in managing patients with severe haemorrhage. In the Malaysian context, factors such as local endoscopic expertise and the availability of adequate therapeutic endoscopic and radiological facilities should dictate the clinical care pathway for the management of UGIB.
Algorithm for management of Non-Variceal Upper Gastrointestinal Bleeding
(6, 7, 8)
Press here to view the Algorithm
Press here to view the Rockall Risk Assessment Score Table
Press here to view
the Forrest Classification of Peptic ulcers
References
Introduction
1. Rockall TA, Logan RFA, Devlin HB, Northfield TC on behalf of steering committee
and members of the national audit of acute gastrointestinal haemorrhage. Incidence
of and mortality from acute upper gastrointestinal haemorrhage in the United
Kingdom. BMJ 1995;311 :222-6
2. Longstreth GF. Epidemiology of Hospitalization for Acute Upper Gastrointeslinal
Hemorrhage: A Population-Based Study. Am J Gastroenterol 1995;90(2):206-210
3. Cheng JLS, Gunn A, Menon J, Arokiasarny J, Ong P, Loong SY, Oommen C, Damodaran
A. Aetiology of Acute Upper Gastrointestinal Bleeding in East Malaysia. Med
) Mal. 2001:56 (supp A) D31
4. Cheng JLS, Menon J, Arokiasamy J, Ong P. Racial Differences in Acute Upper
Gastrointestinal Bleeding in East Malaysia. Med J Mal. 2001;56 (supp A) D29
5. NIH Consensus Conference. Therapeutic endoscopy and bleeding ulcers. JAMA
1989;262:369-72
6. Palmer KR. British Society of Gastroenterology Endoscopy Committee. Non-variceal
upper gastrointestinal haemorrhage: guidelines. Gut 20t)2:5l(Supplement IV):1-6.
Aetiology
7. Dallal HJ, Palmer KR. Upper gastrointestinal haemorrhage. BMJ 2002:Vol.323:1
115-1117
8. Ghosh S, Watts D, Kinnear M. Management of gastrointestinal haemorrhage.
Postgrad Med J 2002,78:4-14
Epidemiology
9. Kang JY, Chua CL, Guan R, et al. A six month study of upper gastrointestinal
haemorrhage at Singapore General Hospital.SingMedJ. 1983;24:124-7.
10. Mahendra SR. Peptic ulcer disease in Kelantan: an underdiagnosed condition.
Med J Mal 1991 ;46:183-6
11. Loke YK, Gunn A, Tan MH. Endoscopic Findings in Patients with Suspected
Upper
Gastrointestinal Haemorrhage - The East Coast Experience. J Gastroenterology
and
Hepatology 1996; 11 (supp 2) A62.
12. Rosemi S, Nazirn M, Raj SM. Acute non-variceal upper gastrointestinal
bleeding in Universiti Sains Hospital, Kelantan. J Gastroenterology and Hepatology
19963;11 (supp 2) A65
13. Tan SS, Chee KH, Singh GP, Chua A. Survey of endoscopic findings in patients
presenting to lpoh Hospital for gastroscopy. I Gastroenterology and Hepatology
1996; 11(supp 2) A65.
14. Lakhwani MN. Ismail AR. Rarras CD. Tan WJ. Upper gastrointestinal bleeding
in Kuala Lumpur Hospital, Malaysia. Medical Journal of Malaysia. 55(4):498-505.
2000 Dec.
Risk factors
NSAIDs
15. Laine L. Approaches to non-steroidal anti-inflammatory drug use in the
high-risk patient. Gastroenterology 2001; 120:594-606
16. Hemandez Diaz S et al. Association between non-steroidal anti-inflammatory
drugs and upper gastrointestinal tract bleeding /perforation:an overview of
epidemiologic studies published in the 1990s. Arch IntMed 2000:160:2093-9
17. Hernandez-Diaz S, Garcia-Rodriguez LA. Epidemiologic assessment of the
safety of conventional non-steroidal anti-inflammatory drugs. AmJ Med 2001;110(Suppl
1):20-27
18. Bombardier C, Laine L, ReincinA, et al. Comparison of upper gastrointestinal
toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis.
N Engl J Med 2000;343:1520-1528
Aspirin
19. Sorensen HT, Mellemkjaer L, Blot WJ et al. Risk of upper gastrointestinal
bleeding
associated with the use of low-dose aspirin. Am J Gastroenterol 2000;95:2218-2224
20. De Abajo FJ, Garci Rodriguez LA. Risk of upper gastrointestinal bleeding
and perforation associated with low-dose aspirin as plain and enteric-coated
formulations. BMC Clin Pharmacol 2001 ;1:1
21. Lanas A, Bajador E, Serrano P et al. Nitrovasodilators, low dose aspirin,
other non-steroidal anti-inflammatory drugs and the risk of upper gastrointestinal
bleeding. N Engl J Med 2000;343:834-9
Helicobacter pylori
22. Jaspersen D, et al. Helicobacter pylori eradication reduces the
rate of rebleeding in ulcer haemorrhage. Gastrointest Endosc 1995;41:5-7
23. Labenz J, Borsch G. Role of Helicobacter pylori eradication in
the prevention of peptic ulcer relapse. Digestion 1994;55:19-23
24. Graham DY, Heeps KS, Ramirez FC et al. Treatment of Helicobacter pylori
reduces the rate of rebleeding in peptic ulcer disease. Scand J Gastroenterol
1993:28:939-42
25. Capurso G, Annibale B, Osborn J et al.Occurrence and relapse of bleeding
from duodenal ulcer: respective roles of acid secretion and Helicobacter
pylori infection. Aliment Pharmacol Ther 2001 ;15:821-829
26. Colin R, Czemichow P, Baty V et al. Low sensitivity of invasive tests
for the detection of Helicobacter pylori infection in patients with
bleeding ulcer. Gasterenterol Clin Biol 2000;24:31-35
27. Hawkey CJ. Risk of ulcer bleeding in patients infected with Helicobacter
pylori taking non-steroidal anti-inflammatory drugs. Gut2000;46:310-311
28 Chan FK, Chung SC, Suen BY et al. Preventing recurrent upper gastrointestinal
bleeding in patients with Helicobacter pylori infection who are taking
low-dose aspirin or naproxen. N Engl J Med 2001;344:967-973
29. Chan FK. Helicobacter pylori, NSAIDs and gastrointestinal haemorrhage.
Eur J Gastroenterol Hepatol 2002;14:1-3
Assessment, Resuscitation and Risk assessment
30. Yamada T, Alpers DH et al. Textbook of Gastroenterology, JB Lippincott
Company. 1995:671 -685
31. Bouchier IAD, Allan RN et al. Gastroenterology: Clinical Science and Practice
1993:957-975
32. Terdiman JP. Update on upper gastrointestinal bleeding. Postgrad Med 1998;
103(6):43-64
33. Rockall TA, Logan RFA, Devlin HB, et al. Risk assessment after acute upper
gastrointestinal haemorrhage. Gut 1996:38:316-321
34. Forrest JA, Finlayson ND, Shearman DJ. Endoscopy in gastrointestinal bleeding.
Lancet 1974;ii:394-7
35. Laine L, Peterson WL: Bleeding peptic ulcer. N Engl J Med 1994;331:717-727
36. Longstreth GF, Feitelberg SP. Outpatient care of selected patients with
acute non-varieceal upper gastrointestinal haemorrhage. Lancet 1995;345(8942):
108-11
37. Rockall TA, Logan RF, Devlin HB et al. Selection of patients for early
discharge or outpatient care after acute upper gastrointestinal haemorrhage.
National Audit of Acute Upper Gastrointestinal Haemorrhage. Lancet 1996;347(9009):1138-40
38. Blatchford 0, Murray WR, Blatchford M. A risk score to predict need for
treatment for upper gastrointestinal haemorrhage. Lancet 2000;356:1318-21
39. Lee JG et al. Endoscopy-based triage significantly reduces hospitalisation
rates and costs of treating upper Gl bleeding:a randomised controlled trial.
Gastrointest Endosc 1999;50:755-61
40. Cipoletta L, Bianco MA, Rotondano G, Marmo R, Piscopo R. Out-patient management
for low-risk non-variceal upper gastrointestinal bleeding: a randomized controlled
trial. Gastrointest Endosc 2002:55:1-5
41 . Hay JA, Maldonado L, Weingarten SR et al. Prospective evaluation of a
clinical guideline recommending hospital length of stay in upper gastrointestinal
tract haemorrhage. JAMA 1997;278(24):2151-6
42. Longstreth GF, Feitelberg SP. Successful outpatient management of acute
upper
gastrointestinal haemorrhage: Use of practice guidelines in a large patient
series. Gastrointest Endosc 1998;47:219-22
43. Podila PV, Ben-Menachem T et al. Managing patients with acute, nonvariceal
gastrointestinal haemorrhage: development and effectiveness of a clinical
care pathway. Am J Gastroenterol 2001;96:208-219
Endoscopic therapy
44. Cook Dj, Guyatt GH, Salena BJ et al: Endoscopic therapy for acute nonvariecal
upper gastrointestinal hemorrhage: A meta-analysis. Gastroenterology 1992;102:139-148
45. Spiegel BMR, Vakil NB, Offman JJ. Endoscopy for acute, non-variceal upper
gastrointestinal tract haemorrhage: is sooner better? Arch IntMed 2001 ;161:1
393-404
46. Palmer KR, Choudari CP. Endoscopic intervention in bleeding peptic ulcer.
Gut. 1995;37(2): 161-4
47. Lin HJ, Wang K, Perng CL, Lee FV, Lee CH, Lee SO. Natural history of bleeding
peptic ulcers with a tightly adherent blood clot: a prospective observation.
Gastrointestinal Endoscopy.1996; 43(5):470-3.
48. Ulrich CD, Costout CJ, Balm RK. Prognostic importance of the densely adherent
clot in bleeding ulcer disease (PUD). A 4-year retrospective review of the
Mayo clinic experience. Gastrointest Endosc 1994:40:89
49. Laine L, Stein C, Sharma V. A prospective outcome study of patients with
clot in an ulcer and the effect of irrigation. Gastrointest Endosc 1996;43:107-10.
50. Lau JY, Chung SCS. Management of upper gastrointestinal haemorrhage. 'J
Gastroenterology and Hepatology2001;15(s3):G8-G12.
Injection therapy
Adrenaline
51. Chung SCS, Leung JW, Steele RJ et al. Endoscopic injection of adrenaline
for actively bleeding ulcers:a randomized trial. BMJ 1988;296:1631-33.
52. Kubba AK. 'Murphy W., Palmer KR. Endoscopic injection for bleeding peptic
ulcer: a comparison of adrenaline alone with-adrenaline plus human thrombin.
Gastroenterology. 1996; 111(3):623-8
Sclerosants
53. Asaki S. Efficacy of endoscopic pure ethanol injection method for gastrointestinal
bleeding. World J Surg 2000;24:294-298
54. Levy J, Khakoo S, Barton R et al. Fatal injection sclerotherapy of a bleeding
peptic ulcer. Lancet 1991;337:504
55. Chung SCS, Leung JWC, Leough HT et al. Adding a sclerosant to endoscopic
epinephrine injection in actively bleeding ulcers:a randomised trial. Gastrointest
Endosc 1993;39:611-15
56. Choudari CP, Palmer KR. Endoscopic injection therapy for bleeding peptic
ulcer: a comparison of adrenaline alone with adrenaline plus ethanolamine
oleate. Gut 1994 ;35:608-10
Thrombogenic agents
57. Kubba AK, Murphy W, Palmer KR. Endoscopic injection for bleeding peptic
ulcer: A
comparison of adrenaline alone with adrenaline plus human thrombin. Gastroenterology
1996; 111: 623-8.
58. Kutgeerts P, Rauws E, Wara P etal. Randomised trial of single and repeated
fibrin glue compared with injection of polidocanol in treatment of bleeding
peptic ulcer. Lancet 1997;350:692-6
59. Blackstone MO. Fibrin glue for bleeding peptic ulcers (Commentary). Lancet
1997;350:679-8.
60. Leung JW, Chung SCS. Practical managernent of nonvariceal upper gastrointestinal
bleeding In Tytgat GNJ, Classen M, ed. Practice of Therapeutic Endoscopy.
Churchill Livingstone.1994;1-16
Thermal methods
61 . Jensen DM, Machicado GA, Kovacs TOG et al. Controlled randomised study
of heater probe and BICAP for haemostasis of severe ulcer bleeding. Gastroenterology
1996;94:A 208
62. Jensen DM. Heater probe for endoscopic haemostasis of bleeding peptic
ulcer. Gastrointest Clin N Am 1991;1:319-39
63. Choudahari CP, Rajagopal C, Palmer KR. A comparison of endoscopic injection
therapy versus heater probe in major peptic ulcer haemorrhage. Gut 1992;33:1159-61
64. Hui WM, Ng MMT, Lok ASF et al. A randomized comparative study of laser
photocoagulation, heater probe and bipolar electrocoagulation in the treatment
of actively bleeding ulcers. Gastrointest Endosc 1991;7:299-304
65. Chung SCS, Lau JY, Sung JJ et al. Randomized comparison between adrenaline
injection alone and adrenaline injection plus heat probe treatment for actively
bleeding ulcers. BMJ 1997;314:1307-11
66. Loizou LA, Bown SG. Endoscopic treatment for bleeding peptic ulcers-randomized
comparison of adrenaline injection and adrenaline injection + Nd-YAG laser
photocoagulation. Gut 1991:32:1100-03
67. Mathewson K, Swain CP, Bland M et al. Randomized comparison of Nd-YAG
laser, heat probe, and no endoscopic therapy for bleeding peptic ulcers. Gastroenterology
1990:98:1239-44
68. Swain CP, Bown SG, Salmon PR et al. Controlled trial of neodymium YAG
laser
photocoagulation in bleeding peptic ulcers. Lancet 1986;i:1113-4
69. Probst A, Scheubel R, Wienbeck M. Treatment of watermelon stomach (GAVE
syndrome) by means of endoscopic argon plasma coagulation (APC) : long term
outcome. Zeitschrift fur Gastroenterologie 2001 ,39:447-52
70. Cipolletta L, Bianco MA, Totondano G, Piscopo R, Prisco A, Garofano ML.
Prospective comparison of argon plasma coagulator and heater probe in the
endoscopic treatment of major peptic ulcer bleeding. Gastrointest Endosc 1998;48:191-5
Other Endoscopic modalities
71. Cipolletta L, Bianco MA, Marmo R et al. Endoclips versus heater probe
in preventing early recurrent bleeding from peptic ulcer: a prospective and
randomised trial. Gastrointest Endosc 2001;53:147-151
72. Quadri A, Vakil N. Peptic ulcer bleeding: Clips, Heat and Outcome. AmJ
Gasteroenterol 2002;97:200-201
Pharmacultigical therapy
73. Green PJ et al. Effect of acid and pepsin on blood coagulation and platelet
aggregation : A possible contributor to prolonged gastrointeslinal haemorrhage.
Gastroenterology 1974;74:3843
74. Pearson JT, Ward R, Allen A et al. Mucus degradation by pepsin-comparison
of mucolytic activity of pepsin 1 and pepsin 3. Implications in peptic ulcpration.
Gut 1986;27:243-8.
75. Vreeburg EM, Levi M, Rauws EAJ et al. Enhanced mucosal fibrinolytic activity
in gastroduodenal ulcer haemorrhage and the beneficial effect of acid suppression.
Aliment Pharmacol Ther 2002 Jun;1 6:1 1 37-42
76. Collins R, Langman M. Treatment with Histamine H2 antagonists in upper
gastrointestinal hemorrhage. Implications of randomized trials. N Engl JMed
1985;313:660.
77. Walt RP et al. Continuous intravenous famotidine for haemorrhage from
peptic ulcer. Lancet 1992;340:1058-62.
78. Levine J et al. Meta-analysis: The efficacy of H2-receptor antagonists
in bleeding peptic ulcer. Aliment Pharmacol Ther 2002 Jun;16:1 137-42
79. Daneshmend TK, Hawkey CJ, Langman MJS, Logan RFA, Long RG, Walt RP. Omeprazole
versus placebo for acute upper gastrointestinal bleeding: randomised double
blind controlled trial. BMJ 1992;304:143-7.
80. Khuroo MS, Yattoo GN, Javid G, et al. A comparison of omeprazole and placebo
for bleeding peptic ulcer. N Engl J Med 1997;336:1054-8.
81. Schaffaalitzky D, Muckadell OB, Havelund T, Harling H et al. Effect of
omeprazole on the outcome of endoscopically treated bleeding peptic ulcers:
randomized double blind placebo controlled multicenter study. Scan J Gastroenterol
1997;32:320-7.
82. Hasselgren G, Lind T, Lundell L et al. Continuous intravenous infusion
of omeprazole in elderly patients with peptic ulcer bleeding: results of a
placebo-controlled mullicenter trial. Scan J Gastroenterol 1997:32:328-33.
83. Lin HJ, Lo WC, Lee FY, Perng CL, Tseng GY. A prospective randomized comparative
trial showing that omeprazole prevents rebleeding in patients with bleeding
peptic ulcer after successful endoscopic therapy. Arch Intern Med 1998;158:54-8.
84. Lau JY. Sung JJ, Lee KK et al. Effect of intravenous omeprazole on recurrent
bleeding after endoscopic treatment of bleeding peptic ulcers. N Engl J Med
Aug 3 2000:343:310-16.
85. Fried R, Begliner C, Meier R, Stumpf J, Adier G, Schepp W, Klein M, Fischer
R. Comparison of intravenous pantoprazole with intravenous ranitidine in peptic
ulcer bleeding. Gut 1999 Suppl V;(45):P0104.
86. Gisbert JP, Gonzalez L, Calvet X, Roque M, Gabriel R, Pajares JM. Proton
pump inhibitors versus H2 antagonists: a meta-analysis of their efficacy in
treating bleeding peptic ulcer. Aliment Pharmacol Ther 2001,15(7):917-26.
87. Laine L .Multipolar electrocoagulation in the treatment of active upper
gastrointestinal tract haemorrhage-a prospective controlled trial. N Engl
J Med 1987;316:1613.
88. Yamasuchi Y et al. Endoscopic haemoclipping for upper gastrointestinal
bleeding due to Mallory-Weiss syndrome. Gastrointest Endosc.2001;53:427-30
89. Machicado GA, Jensen DM. Randall GM. Upper gastrointestinal angiomata-
diagnosis and treatment.. Gastrointest Clin NorthAm 1991;21:241-62
90. Savides TJ, Jensen DM. Therapeutic endoscopy for non-variceal gastrointestinal
bleeding, Gastrointest Clin North Am 2000,29:465-87
Aftercare
91. Wong SK, Yu L M, Lau J Y W et al. Prediction of therapeutic failure after
adrenaline injection plus heater probe treatment in patients with bleeding
peptic ulcer. Gut 2002:50:322-5
92. Pundzius J. Clinical and endoscopic signs for the prediction of recurrent
bleeding from gastroduodenal ulcers. Eur J Surg 1994;160:689-92
93. Thomopoulos et al. Factors associated with failure of endoscopic injection
haemostasis in bleeding peptic ulcers. Scand J Gastroenterol 2001;36:664-8
94. Saeed ZA et al. Prospective validation of the Baylor bleeding score for
predicting the likelihood of rebleeding after endoscopic haemostasis of peptic
ulcers. Gastrointest Endosc.1995;41:561-5
95. Hasselgren G et al. Risk for rebleeding and fatal outcome in elderly patients
with acute peptic ulcer bleeding. Eur J Gastroenterol Hepatol 1998;10:667-72
96. Messmann H, Schaller P, Andus T et al. Effect of programmed endoscopic
follow-up examinations on the rebleeding rate of gastric or duodenal peptic
ulcers treated by injection therapy:a prospective,randomized controlled trial.
Endoscopy 1998;30:583-9
97. Villanueva C, Batanzo J, Torras X et al. Value of a second-look endoscopy
after injection therapy for bleeding peptic ulcer : a prospective and randomized
trial. Gastrointest Endosc 1994;40:34-9
98. Saeed ZA. Second thoughts about second-look endoscopy for ulcer bleeding?
Endoscopy 1998:30:650-1
Management of Rebleeding
99. Baillie J. Bleeding peptic ulcer-risk factors for rebleeding and sequential
changes in
endoscopic findings. Gastrointest Endosc. 1994 Sep-;40:656-7
Role of surgery
100. Lau YW, Sung JY, Lam Y, Chan CW, Ng KW, Lee WH, Chan KL, Suen CY, Chung
CS. Endoscopic retreatment compared with surgery in patients with recurrent
bleeding after initial endoscopic control of bleeding ulcers. N.Engl.J.Med.1999;340:751-56.41
101. Scheim M, Gecalter G. Apache II score in massive upper gastrointestinal
haemorrhage from peptic ulcer; a prognostic value and potential clinical applications.
Br J Surg 1989,76:733-6
102. Poxon VA, Keighley MRB, Dykes PW, Hepinstall K, Jaderberg M. Comparison
of minimal and conventional surgery in patients with bleeding peptic ulcer:
a multicentre trial. Br J Surg 1991,78:1344
103. Millat B, Hay JM, Valleur P, Fingerhut A, Fagniez PL. French Associations
for Surgical Research: Emergency surgical treatment for bleeding duodenal
ulcer: oversewing plus vagotomy versus gastric resection, a controlled randomized
trial. World J Surg 1993.: 1 7:568
Interventional Radiology
104. Kramer SC, Gorich J, Rilinger N et al. Embolization for gastrointestinal
haemorrhages. Eur Radiol 2000;10:802-805b