Transcript GI bleeds: what’s old, what’s new? Grand Rounds, Department of Emergency Medicine

GI bleeds: what’s old,
what’s new?
Grand Rounds, Department of
Emergency Medicine
April 15, 2004
Christine Hall MSc MD FRCPC
Outline

Upper GI bleeds overall summary
–
–
–
–

PUD
Drinkers and non drinkers
Variance
NSAIDs
Upper GI bleeds specific physiology
– NSAIDs

Upper GI bleed Rx issues
– PPI’s
– EGD
– Discharge criteria



Variceal bleeds
Lower GI bleeds
Patient disposition
Introduction
GI bleeding:
 2% of all US hospital admissions
 5% of admissions originating in the ED
 Majority of bleeds are UGI

UGI bleeding incidence

103/100,00 adult per year in UK
– incidence 23/100,00 in age under 30 to
485/100,00 in over 75
men more than double women except in
elderly
 14% occurred in inpatients already in
hospital for some other reason
 27% of cases (37% female, 19% male)
patients were aged over 80.

Rockall TA, Logan RF, Devlin HB, Northfield TC.
Gastrointest Endosc. 2002 Jan;55(1):1-5.
UGI Bleeding mortality

Overall mortality 14%
– 11% in emergency admissions
– 33% in inpatients

In the emergency admissions:
– 65% of deaths in pts under 80 associated with
malignancy or organ failure at presentation

Mortality for patients under 60 in the absence of
malignancy or organ failure at presentation was
0.8%.
Rockall TA, Logan RF, Devlin HB, Northfield TC.
Gastrointest Endosc. 2002 Jan;55(1):1-5.
UGI bleeds in drinkers





From August 1, 1990 through September 9,
1994. Prospective.
Pts referred to GI (selection bias)
subnormal hematocrit on or within 12 h of
admission or a fall of at least 5 points from
previous
Upper GI endoscopy was performed in all
patients within 48 h of admission
Alcohol use was quantitated as chronic (80 g or
more per day for at least 1 month), binge,
occasional, or none.
UGI bleeds in drinkers
727 pts: 212 (29%) chronic alcohol users
 Peptic ulcer disease was most common
cause of bleeding (60%).
 Gastropathy etiological in only 32 patients
(4%)

– portal hypertension in 22 patients
– NSAID induced gastropathy in 5

Alcohol induced gastropathy in 5 pts
– bleeding was mild and self-limited.
Wilcox CM, Alexander LN, Straub RF, Clark WS
Am J Gastroenterol. 1996 Jul;91(7):1343-7
UGI bleeds in drinkers

Causes of bleeding were compared between
drinkers and nondrinkers
– drinkers were more likely to bleed from varices
(p = 0.024)
– portal hypertension-related causes (p < 0.01),

Peptic ulcer was more common in nondrinkers
compared with chronic ETOH
– (67 vs 53%; p < 0.01)

Esophagitis (p = 0.95) and Mallory-Weiss tear (p
= 0.15) were NS between the two groups.
Wilcox CM, Alexander LN, Straub RF, Clark WS
Am J Gastroenterol. 1996 Jul;91(7):1343-7
UGI bleed variance

Qiu, YJ. Zhonghua Hu Li Za Zhi. 1987
Jun;22(6):269-70.

Relation between hemorrhage and
the waxing or waning of the moon
– Unfortunately in Chinese so not available for
today…
Upper GI bleeds: pathophysiology

Esophagitis
– Candidal
– Non candidal
– bisphosphonates
Esophageal tears
 Gastritis
 That’s all we’ll say about the benign
causes

UGI bleed diagnosis: to OG or not
to OG?
Not great evidence for confirming UGI
bleed by putting in NG or OG
 Practice discouraged by most GI’s
 Can confound UGI bleed re: trauma of
insertion
 Relative contraindication in variceal bleed
although this is witchcraft…more about
that later…

PUD Causation
10%
30%
60%
Other
H. Pylori
NSAID
H. Pylori
Serology never becomes negative once
positive
 Even if H. Pylori is eradicated
 Complicates repeat breath tests or
biopsies
 PPIs are antimicrobial for h.pylori despite
poor initial performance in development as
an antibiotic

Current Rx for H. Pylori Eradication
NSAID induced ulcers
NSAIDS: what’s the problem?
Toxic effects of NSAIDs thought to be due to
Cox-1 inhibition
 launch of the cyclooxygenase-2 (COX-2)
selective NSAIDs was based on 2 hypotheses:

– the major adverse effects limiting the usefulness of
nonselective NSAIDs are gastrointestinal in nature
– COX-2 selective NSAIDs are associated with fewer
gastrointestinal adverse effects than nonselective
NSAIDs.
The CLASS Trial

To determine whether celecoxib, a COX-2-specific
inhibitor, is associated with a lower incidence of
significant upper GI toxic effects and other adverse
effects compared with conventional NSAIDs.

double-blind, randomized controlled trial conducted from
September 1998 to March 2000

Three hundred eighty-six clinical sites in the United
States and Canada
The CLASS trial
8059 patients : >/=18 years old, OA or RA,
7968 received at least 1 dose of study drug.
 4573 patients (57%) received treatment for 6
months
 Random assign:

– celecoxib, 400 mg twice per day (2 and 4 times the
maximum RA and OA dosages, respectively
– n = 3987
– ibuprofen, 800 mg 3 times per day (n = 1985)
– diclofenac, 75 mg twice per day (n = 1996).

ASA for CV prophylaxis (</=325 mg/d) was
permitted.
CLASS Trial

Incidence of prospectively defined symptomatic upper GI ulcers and
ulcer complications (bleeding, perforation, and obstruction) and
other adverse effects during the 6-month treatment period.

Overall annual incidence rates of UGI ulcer complications alone:
0.76% vs 1.45% with NSAIDs (p=0.09)
Overall annual incidence rate of UGI ulcer complications plus
symptomatic ulcers 2.08% vs 3.54% , p=0.02


Pts not on ASA:
– UGI complication alone 0.44% vs 1.27%, p=0.04
– UGI complication plus symptoms: 1.40% vs 2.91%, p =.02

For patients onASA:
– UGI complications alone 2.01% vs 2.12%,l p=0.92
– UGI complications plus symptomatic ulcers 4.70% vs 6.00% (P =.49).
CLASS conclusions
Fewer celecoxib-treated patients than NSAID-treated
patients experienced chronic GI blood loss, GI
intolerance, hepatotoxicity, or renal toxicity.
 No difference was noted in the incidence of
cardiovascular events between celecoxib and NSAIDs,
irrespective of aspirin use.
 CONCLUSIONS: Celecoxib, at big doses, was associated
with a lower incidence of symptomatic ulcers and ulcer
complications combined, as well as “other clinically
important toxic effects”, compared with NSAIDs at
standard dosages (800 mg tid).
 The decrease in upper GI toxicity was strongest among
patients not taking aspirin concomitantly.

JAMA. 2000 Sep 13;284(10):1247-55.
The VIGOR trial





clinical upper gastrointestinal events occur in 2 to 4
percent of patients who are taking nonselective NSAIDs
Is rofecoxib associated with a lower incidence of
clinically important upper gastrointestinal events than
Naproxen among patients with rheumatoid arthritis
randomly assigned 8076 RA patients who were at least
50 years of age (or at least 40 years of age and
receiving long-term glucocorticoid therapy)
received either 50 mg of rofecoxib daily or 500 mg of
naproxen twice daily
primary end point was confirmed clinical upper
gastrointestinal events (gastroduodenal perforation or
obstruction, upper gastrointestinal bleeding, and
symptomatic gastroduodenal ulcers)
VIGOR Trial


Rofecoxib and naproxen had similar efficacy against
rheumatoid arthritis.
median follow-up of 9 months
– 2.1 per 100 pt years confirmed gastrointestinal events with
refecoxib
– 4.5 per 100 pt-years with naproxen
– relative risk, 0.5; CI 0.3 to 0.6; P<0.001

The respective rates of complicated confirmed events
(perforation, obstruction, and severe upper
gastrointestinal bleeding)
– 0.6 per 100 patient-years refecoxib
– 1.4 per 100 patient-years naproxen
– relative risk, 0.4; CI 0.2 to 0.8; P=0.005
VIGOR trial

The incidence of myocardial infarction
– lower among patients in the naproxen group
than among those in the rofecoxib group
– 0.1 percent vs. 0.4 percent
– relative risk, 0.2; CI 0.1 to 0.7
– the overall mortality rate and the rate of
death from cardiovascular causes were similar
in the two groups.
Bombardier C, Laine L, Reicin A, Shapiro D, Burgos-Vargas R,
Davis B, Day R, Ferraz MB, Hawkey CJ, Hochberg MC,
Kvien TK, Schnitzer TJ
N Engl J Med. 2000 Nov 23;343(21):1520-8
So….
increased incidence of total and nongastrointestinal
serious adverse events with the COX-2 selective NSAIDs
remains a major concern
 increased morbidity may be a manifestation of the COX-2
selectivity or the supramaximal doses of these drugs
used in the trials
 proof that the increased harm was not caused by the
COX-2 selectivity of the drugs depends on a randomized
controlled trial of COX-2 selective NSAIDs at usual doses
vs. nonselective NSAIDs
 Thus far, no such trial
 What about Cox-2’s plus PPII’s?...no trial

Rx of UGI bleed

Shots in the dark?
Upper GI bleeds
Coagulation and fibrinolysis are dependent
on intragastric pH
 When pH is below 6, platelet aggregation
is abolished
 When pH is below 4, fibrin clots are
digested by gastric pepsin
 In UGI bleed, need to raise gastric pH and
keep it there for 72 hours

PUD H2 blockers
H2 receptor antagonists inhibit gastric acid
production
 Little if any evidence of improved outcome
in active GI bleed
 Insufficient acid suppression
 Rapid drug tolerance

Proton Pump Inhibitors






More profound and sustained acid suppression
Binds irreversibly to H+K+ATPase (proton
pump) inhibiting actively secreting parietal cells
Predictable: t1/2 1 hr but effect 24 hrs
No tolerance developed
Excellent safety profile, no dose reduction in
renal failure/cirrhosis
IV Omeprazole released in 1991
PPI’s in PUD

IV Proton pump inhibitors
– Lau et al NEJM 2000 Aug 3: 343 (5) 310-16
– After endoscopic Rx, bleed recurs in 15-20%
– Assess whether use of high dose PPI
decreased recurrence within 30 days
– Dbl blind rand to Omeprazole bolus and
infusion vs placebo
– All got Omeprazole orally x 2 weeks post
infusion
PPI’s in PUD (Lau, ct’d)
– 240 patients, recur in 6.7% of omeprazole vs
22.5% of placebo
– Most recurrent bleed occurs in first three days
– “Only evidence for IV infusion is in acute
bleed with ulcer with high risk stigmata (i,.e
visible vessel)”
– But, EGD was done within 1 hour 24/7
Waiting for EGD
PPI’s while waiting for EGD







R. Enns & C. Andrews, UBC, Can J Gastro 2003 &
Aliment Pharmacol Ther 2003.
To evaluate cost effectiveness of IV PPI while awaiting
EGD compared with EGD alone
Probabilities of rebleeding and surgery taken into
account
In a hypothetical cohort of 1000 patients, IV PPI
utilization while awaiting EGD = mean savings of
$20,700 Canadian
Using PPI pre EGD averted 37 rebleeds
Use of PPI while awaiting EGD is cost effective
Finidngs do not apply to low risk stigmata, variceal
bleeds, lower GI bleeds
PUD: RUGBE






Canadian Registry of patients with Upper GI
Bleeding Undergoing Endoscopy
Pt char and Rx from 1999 on
J. Romagnolo is local PI
Protective effect of PPI use and EGD in specific
patient subgroups
Meta-analysis supports high dose IV PPIs leads
to improvement when coupled with EGD
Optimum dose, timing, and subgroup analysis
needs further work
PPI’s the local challenge
CHR: Pharmanews
 Vol 28 Issue 3 Revised November 2003
 Kaplan Gf, Bates D, McDonald D,
Romagnulo J.
 “Inappropriate Prescribing Leading to
Rapidly Escalating Costs of IV PPI in the
Calgary Health Region”

PPI’s the local cost
Year
FMC
PLC
RGH
Total
1999
$56K
$35K
$9K
$100K
2000
$149K
$86K
$73K
$308K
2001
$207K
$96K
$121K
$425K
2002
$168K
$94K
$123K
$386K
PPI’s Local guidelines





Ideally withold PPI’s until EGD but case by case
adaptation…i.e. night
IV Omeprasole 80 mg bolus and 8mg/hr infusion
for max of 72 hours
Pts with HRS on EGD (arterial or venous active
bleed, visible vessel, adherent clot) should have
endo Rx and then 72 hr infusion
No HRS = stop infusion
Inpts who are NPO get bid dosing not infusion
PPI’s Local guidelines ct’d
Clinical equivalence between IV PPI and
oral PPI in GERD
 Clinical equivalence between IV PPI and
NG PPI
 40 mg PPI IV = $13.70
 40 mg PPI po = $0.45

Can UGI bleeds go home?
Selection of patients for early discharge or
outpatient care after acute UGI haemorrhage.
National Audit of Acute Upper GI Haemorrhage
 Attempted to identify patients who had
negligible risk of further bleeding or death and
for whom early discharge or even outpatient
management would be possible with no adverse
effect on standards of care.

UGI bleed and discharge

Used a validated risk scoring system based on:
–
–
–
–
–
–
age (score 0-2)
presence of shock (0-2)
comorbidity (0-3)
diagnosis (0-2)
endoscopic stigmata of recent haemorrhage (0-2)
maximum possible score was 11.
Rockall’s Scoring System
VARIABLE
0
1
2
3
Age
<60
60-79
> 80
----
Shock
HR<100
HR>100,
SBP >100
HR>100,
SBP<100
----
Comorbidity
None
----
CHF, IHD,
any major
dx
Renal failure
Liver failure
Mets
Dx post OGD
MW tear
All other dx
UGI malig,
---Blood in UGI
Major SRH
None or
dark spot
----
Adherent
clot, vis
vessel
----
UGI bleed and discharge



2531 patients who underwent endoscopy after an acute
admission
744 (29.4%) of the 2531 patients scored 2 or less on the
risk score
Of these:
– 32 or 4.3% rebled [95% Cl 3.0-6.0]
– 1 or 0.1% died [95% CI 0.006-0.75]



Conclusions: the risk score identifies patients at low risk
of rebleeding or death
Median hospital stay increased with risk score
Within low risk score categories of 5 or less:
– trend of increasing hospital stay as the time between admission
and endoscopy increased
Lancet. 1996 Apr 27;347(9009):1138-40
Rockall TA, Logan RF, Devlin HB, Northfield TC.
UGI bleed and discharge
Three year prospective validation of a PRE-EGD
risk stratification in patients with acute UGI
hemorrhage
 To assess the accuracy of a risk stratification
used at initial assessment (does not rely on
EGD)
 To identify groups with increased risk of
mortality and requirement for urgent treatment
intervention
 Prospective assessment of risk stratification in
consecutive patients with acute uppergastrointestinal haemorrhage.

UGI risk stratification
3-year period, 1349 consecutive patients,
risk stratified and followed up for 2 weeks
 Two-week, all-cause mortality, re-bleeding,
and need for urgent treatment
intervention.

UGI bleed and discharge without
EGD?

Stratification within the high-risk group predicted a significant
increased risk of 2-week, all-cause mortality (P < 0.001)
– Hi risk 11.8%
– Intermediate risk 3%
– Low risk 0%

High risk strata significantly higher re-bleeding (P < 0.001)
– Hi risk 44.1%
– Intermediate 2.3%
– Low risk 0%

High risk strata significantly higher need for urgent treatment
intervention (P < 0.001)
– Hi risk 71%
– Intermediate risk 40.6%
– Low risk 2.6%
Cameron EA, Pratap JN, Sims TJ, Inman S, Boyd D,
Ward M, Middleton SJ. Eur J Gastroenterol Hepatol.
2002 May;14(5):497-501.
So, who admits?



Physician specialty and variations in the cost of treating patients
with acute upper gastrointestinal bleeding.
Pts stratified into three groups based on a validated risk score
Length of stay and hospital costs compared
–

primarily cared for by internists, surgeons, and gastroenterologists
Median length of stay (2 days) if admitted to GI
– Significantly shorter than admitted under other physicians (P < 0.05)

Median hospitalization cost ($2856) if admitted to GI
– Significantly lower than other services (P < 0.01)

No differences in the time to endoscopy among services.
Quirk DM, Barry MJ, Aserkoff B, Podolsky DK
Gastroenterology. 1997 Nov;113(5):1443-8
Now for some risky business…
Variceal bleeding
Esophageal and/or gastric varices account for
about 20% of all GI bleeds
 Most dramatic and life threatening presentation
of portal hypertension
 25-50% of bleeds in cirrhotic pts is non variceal

– Dave P et al, Clin Gastroenterol 5:113, 1983
After 1 bleeding episode, risk of a second is 70%
 Accounts for 50% of all deaths from cirrhosis

Variceal bleeding
Risk of mortality from each bleeding
episode is 20-84%
 Risk depends on etiology of portal
hypertension, hepatic reserve of the pt.,
duration of hemorrhage prior to
presentation

IV Octreotide
Octreotide or Somatostatin is a potent
nonselective vasoconstrictor
 Should be routine in known or strongly
suspected variceal bleed
 Can be used alone or in conjunction with
sclerotherapy/banding
 No role in non variceal UGI bleeds

Balloon tamponade
Originated in the 1930s, first filled with water
not air
 1950 Sengstaken and Blakemore invented
double lumen balloon
 Modified by Linton, Nachlasm, Boyce and Edlich
 Currently have:

– Blakemore: 3 lumens – gastric, esoph and gastric
suction
– Linton: as above with esoph aspiration port as well
Balloon tamponade
Roberts still recommends NG insertion in
all GI bleed..no real evidence for it…nice
in flight or if ongoing vomiting
 There is no evidence to support passage
of an NG tube increases variceal bleed

– Meeroff et al: Mangement of massive UGI
bleed Hosp Practice, 1984

Management of varices is substantially
different from other causes
Balloon tamponade
Used when sclerotherapy not available
 Used when vasoconstrictor therapy is not
adequate
 Can control bleeding in up to 80% of cases
 In the absence of endoscopy indications:

– pt with known portal hypertension
– prior variceal bleed
– not responding to vasoconstrictor Rx
Balloon Tamponade of Varices
Use of GEBT has serious/lethal
complications
 Overall complication rate higher than most
other ED procedures
 Clear indications and complete
understanding of placement is essential


“best placed by the individual most
familiar with technique” (which is
usually no-one including GI guys)
GEBT
Patient at risk re uncooperative, or hepatic
encephalopathy
 If patient is not awake, alert and fully
cooperative should be intubated

– Roberts, Procedures in emergency medicine
Patients at extremely high risk for regurg
and aspiration
 Threshold for intubation should be lower
than usual

GEBT Procedure
Unfamiliar to most docs
 Gastric balloon holds 500 to 700 mL, test
both balloons for patency
 Patient at 45 degrees or in left lateral
 Anesthetize nose and post pharynx if
necessary
 Deflate balloons and lock off, lubricate
tube

GEBT procedure
Pass through mouth is preferred
(especially in intubated) but nose can be
used
 Pass tube to at least 50 cm mark or max
depth allowed by tube length
 Suction to esoph and gastric ports to
minimize aspiration risk even further prior
to inflation
 Confirm radiographically prior to inflation

GEBT procedure
Once markers seen: can begin to inflate
gastric balloon
 Should monitor pressure in balloon as
inflated
 Use AIR not water
 Repeat radiograph once inflated
 Clamp air ports on tube and gently pull
back on tube to snug up to GE junction

Indications to inflate the
esophageal balloon
If continuous bloody aspirate from gastric
suction
 Absolutely must be done with manometer
guidance
 Pressure at lowest level that stops
bleeding and should not exceed 45 mm
Hg
 If ongoing bleeding with both balloons up,
likely gastric and NOT esophageal source

GEBT traction
Use if inflation of balloons has been done
and active gastric bleeding persists
 External traction on the tube
 Foam rubber block often included in the
kit
 Various traction mechanisms

GEBT Complications
Occur in 8-16% of patients
 Mortality considered in about 3%
 Variceal hemorrhage itself carries 40-80%
mortality
 One study reported GEBT directly casued
death in 22% of pts in which it was used

GEBT complications
Major occur in 8-16%
 Mortality around 3%

– Common
 Aspiration pneumonia
 Asphyxiation (balloon migration, cut the tubes)
 Esophageal necrosis
– Uncommon





Esophageal perforation (overinflation or duration tube)
Duodenal rupture
Tracheobronchial rupture
Periesophageal abcess
mediastinitis
GEBT Complications

Minor
– Common





Gastroesophageal ulceration
Regurgitation
Chest discomfort
Back pain
Pressure necrosis re: traction
– Uncommon




Epistaxis
Pharyngeal erosions
Pressure necrosis of tongue
hiccups
Problems at the bottom end
Lower GI bleeds





defined as blood loss that originates from a source distal
to the ligament of Treitz
results in hemodynamic instability or symptomatic
anemia
approximately 10% to 15% of patients presenting with
acute severe hematochezia have an upper
gastrointestinal source of bleeding
most common causes of lower gastrointestinal bleeding
are diverticulosis, hemorrhoids, ischemic colitis, and
angiodysplasia
lower gastrointestinal bleeding ceases spontaneously in
most cases
LGI bleed, does colonoscopy affect
outcome?
management of lower-GI hemorrhage remains
controversial largely because outcomes data are
lacking
 Hypothesized that clinical factors, such as
comorbidity, hemodynamic instability, and timing
of colonoscopy, are associated with hospital
lengths of stay
 Retrospective review of LGI hospitalized, 1993 to
2000 (selection bias)
 regression model was constructed to explore
associations between time to discharge (i.e.,
length of stay) and clinical parameters.

LGI, colonoscopy continued




565 hospitalizations, mean length of stay was 6.7 days
Colonoscopy was performed during 415 hospitalizations
1/3 discharged within 48 hours after colonoscopy
Regression model, decreased likelihood of discharge if:
– hemodynamic instability
– higher comorbidity
– performance of a tagged red blood cell nuclear scan
– surgery for hemostasis

Colonoscopy at any timepoint associated with an increased
likelihood of being discharged
– hazard ratio 1.5: 95% CI [1.2, 1.8].

Mean lengths of stay for colonoscopy within 24 hours of
hospitalization was shorter
– 5.4 vs. 7.2 days; p<0.008
Early colonoscopy for acute lower GI bleeding
predicts shorter hospital stay: a retrospective
study of experience in a single center.
Schmulewitz N, Fisher DA, Rockey DC.
Predictors of severe LGI bleed





Harvard group
Goal to determine risk factors for severe acute LIB
252 consecutive patients admitted with acute LIB
Data were collected on 24 clinical factors available in the
first 4 hours of evaluation
outcome was severe bleeding, which was defined as:
– continued bleeding within the first 24 hours of hospitalization
documented by transfusion of > or = 2 units of blood and/or
hematocrit decrease of > or = 20%)
– recurrent bleeding after 24 hours of stability (additional
transfusions, further hematocrit decrease of > or = 20%, or
readmission for LIB within 1 week of discharge)

Severe LGI bleeding occurred in 123 pts (49%)
Predictors of Severe LGI bleed


Independent correlates of severe bleeding were as follows:
Heart rate, > 100/min
– OR 3.67 (1.78-7.57)

Systolic blood pressure, < 115 mm Hg
– OR 3.45 (1.54-7.72)

Syncope
– OR 2.82 (1.06,7.46)

Nontender abdominal examination
– OR 2.43 (1.22-4.85)

Bleeding per rectum during the first 4 hours of evaluation
– OR, 2.32 (1.28-4.20)

Aspirin use
– OR, 2.07 (1.12-3.82)

More than 2 active comorbid conditions
– OR, 1.93 (1.08-3.44)
Strate LL, Orav EJ, Syngal S.
Arch Intern Med. 2003 Apr 14;163(7):838-43.
So, who can be discharged from
the ED?

Lit search = 0 papers,
so we are adrift on
this one
Questions?
Can you differentiate based on
stool color?






Now, here was a fun trial….
Subjective reporting of the color of blood passed per
rectum has been used to predict the location of
gastrointestinal bleeding
validity of this clinical approach has never been
evaluated systematically
this study determined the spectrum of patient and
physician descriptors used to characterize the color of
blood passed per rectum
evaluated prospectively if an objective test of stool color
would correlate with or improve upon subjective
descriptions in predicting bleeding locations
I know you can hardly wait for the results...
Stool color continued (how’s that
breakfast going?)
objective test employed was a card containing
five numbered colors that typify the spectrum of
stool blood colors
 120 pts used 23 different descriptors or terms to
verbalize the color of blood they passed per
rectum
 in 22% of cases there was a seeming
discrepancy between their verbalized color and
the color they pointed to on the test card

Stool color results
Pts pointing to black card resulted in closer matching to
an upper bleeding source than physicians using
terminology such as melena or tarry stools, p <0.02
 Pts picking the brightest red colors resulted in closer
matching to a coloanorectal bleeding source than
physicians using the terms hematochezia or bright red
blood per rectum, p <0.02
 PPV of the black card for UGI bleed was very high both
when patients pointed to this color or when it was
determined from the available stool (0.95 and 0.98,
respectively)
 PPV of brightest red card for LGI bleed was 100% when
patients picked it than and 0.83 if stool was directly
compared.
 Anyone want to order cards?

Hematochezia vs melena and
outcome…






Consecutive patients in Atlanta
August 1, 1990 - September 31, 1994
Prospective
Vital signs and stool color were recorded on
admission to ED
Endoscopy was performed in all patients, usually
within 48 h of admission
Cause of bleeding determined by endoscopy,
surgery, or autopsy.
Hematochezia results…





727 pts
104 (14%) presenting with hematochezia (18 with bright
red blood and 86 with maroon blood).
In those with hematochezia most common causes of
bleeding were duodenal ulcer (44%) and gastric ulcer
(20%)
Hematochezia vs melena: patients with hematochezia
were older (55 vs 50 yr, p < 0.01) and more likely to
present with duodenal ulcer bleeding (43 vs 25%, p <
0.01)
No differences in vital signs, including prevalence of
shock, or admission Hb concentration were found b/t
hematochezia vs melena.
Conclusions

Transfusion requirements:
– 5.4 vs 4.0 units, p = 0.01

Need for surgery
– 11.7 vs 5.7%, p = 0.03

Mortality
– 13.6 vs 7.5%, p = 0.05

All significantly higher in patients with
hematochezia than in those with melena
Wilcox CM, Alexander LN, Cotsonis G
Am J Gastroenterol. 1997 Feb;92(2):231-5