Rokcall score versus forrest classification in endoscopic management of bleeding peptic ulcer

doi:10.4236/health.2010.26096

Paper Menu >>

Journal Menu >>

Vol.2, No.6, 634-638 (2010) Health

doi:10.4236/health.2010.26096

Rokcall score versus forrest classification in

endoscopic management of bleeding peptic ulcer

Heba Sayed Assal1*, Ashraf Elsherbiny1, Hanan M. M. Badawy2, Ehab Hassan Nashaat3,

Hesham al Shabrawi4

1Internal Medicine, Internal Medicine Department, National Research Center, Cairo-Egypt;

*Corresponding Author: heba_assal@yahoo.com

2Internal Medicine, Department, Faculty of medicine, Ain-shams University, Cairo-Egypt

3Internal Medicine, Internal Medicine Department, Faculty of medicine, Ain-shams University, Cairo-Egypt

4Internal Medicine Department, Ahmed Maher Teaching hospital, Cairo-Egypt

Received 27 November 2009; revised 22 February 2010; accepted 24 February 2010.

ABSTRACT

Acute upper gastrointestinal bleeding (UGIB)

remains an important emergency situation. In

the last two decades, major developments took

place influencing incidence, etiology and out-

come of patients with acute UGIB. Peptic ulcer

bleeding is the most significant complication of

ulcer disease being responsible for 50% of all

cases mortality. Aim of the study: To compare

between endoscopic clip application versus

argon plasma coagulation in management of

bleeding peptic ulcer (BPU). Patients and Meth-

ods: Sixty patients suffering from acute UGIB

were randomly divided into two groups: group I

included 30 patients treated with endoscopic

clip application and Group II included 30 pa-

tients subjected to endoscopic APC. All patients

were classified according to Forrest classifica-

tion and the clinical Rockall score. Results:

There were significant differences between the

two groups as regard Forrest classification (P <

0.05) there were insignificant difference be-

tween the two groups as regard rockall score,

site of the ulcer and re bleeding (P > 0.05). Re

bleeding was significant with higher Rockall

score in group I (P < 0.05) but it was insignifi-

cant in group II (P > 0.05). Conclusion: Endo-

scopic application of hemoclips have a less re

bleeding rate than Argon plasma coagulation for

treatment of bleeding peptic ulcer, although this

was statistically insignificant, meanwhile APC is

still less cost and easy. Clinical and endoscopic

assessment (through Rokcall score and Forrest

classification) could help in making best choice

for endoscopic management.

Keywords: Endoscope; Bleeding Peptic Ulcer;

Argon Plasma Coagulation; Clip Applications

1. INTRODUCTION

Since the late 1980s, endoscopic haemostatic therapy has

been widely accepted as the first-line therapy for up-

per-gastrointestinal bleeding. Most clinical trials demon-

strated a reduction in both recurrent bleeding and the

need for surgical intervention when endoscopic hemo-

stasis was used [1]. Endoscopic therapy can be broadly

categorized into injection therapy, thermal coagulation,

and mechanical hemostasis. When analyzed separately,

injection therapy, thermal-contact devices, and me-

chanical treatment all decrease the frequency of recur-

rent bleeding and rate of surgical intervention [2]. Argon

plasma coagulation (APC) is a non contact type of co-

agulation that is easier to target to bleeding sites. A

high-frequency current is transmitted by the ionized,

electrically conductive argon gas. The argon gas flows

onto the target surface, even if approached tangentially.

APC has been used successfully to obtain hemostasis

during open surgery. The use of APC in digestive tract

endoscopy was first described in 1994. It is being ap-

plied more and more widely in the treatment of different

GI pathologic disorders, hemorrhagic lesions in particu-

lar [3]. The only mechanical therapies widely available

are endoscopically placed clips and band ligation de-

vices. Endoscopic clips usually are placed over a bleed-

ing site (e.g. visible vessel) and left in place [4]. This

consisted of a stainless steel clip (of size approximately

6 mm long and 1.2 mm wide at the prongs) with a metal

deployment device (that could be used to insert the clip

into the endoscopic camera, and deployed outside the

camera) enclosed in a plastic sheath. These clips were

H. S. Assal et al. / HEALTH 2 (2010) 634-638

635

initially reloadable [5]. Numerous prognostic scores

have been devised to aid the gastroenterologist in the

management of upper gastro-intestinal bleeding, strati-

fying individual patients by risk of re-bleeding and death.

These scores range from the simple, endoscopy-based

analysis of ulcer appearance described by Forrest et al.

[6], through pre-endoscopic clinical scores such as the

‘clinical’ Rockall scores [7], to combined clinical and

endoscopic evaluation, best exemplified by the classical

Rockall [8]. Such a scheme should aid in making clinical

decisions, as to both the need for urgent intervention and

the prediction of continued or recurrent bleeding in the

context of endoscopic therapy [9].

The purpose of this study was designed to compare

between endoscopic clip applications versus argon plasma

coagulation in management of bleeding peptic ulcer.

2. PATIENTS AND METHODS

This study was conducted on 60 patients (30 males with

mean age 50.8 ± 8.9 and 30 females with mean age 49.7

± 9.8) between January 2007 and August 2008, all pa-

tients admitted to Ain-Shams University Hospital, pre-

senting with hematemesis. After fluid resuscitation, the

patients underwent endoscopy of the upper gastrointes-

tinal tract within 12 hours of admission. Those with

duodenal, gastric, or stomal ulcers and stigmata of recent

hemorrhage were enrolled in the study. The patients

were selected according to Forrest classification between

groups IA (spurting bleeding) to IIB (non bleeding ulcer

with an adherent clot). A score was calculated to them

according to Rockall’s score.

Patients were excluded from the study if they had se-

vere terminal illness that made endoscopic examination

hazardous or undesirable; profuse hemorrhage accompa-

nied by persistent shock, during which the upper gastro-

intestinal tract was filled with fresh blood, limiting visi-

bility through the endoscope and necessitating emer-

gency surgery as a life-saving procedure; or bleeding

from a Mallory—Weiss tear, varices, erosions, tumors,

or an unknown source.

All patients gave informed consent and the study was

approved by the Institutional Ethical Committee.

All participants were subjected to:

Resuscitation including IV fluids, packed RBC trans-

fusion until becomes hemodynamically stable. Routine

laboratory investigations: complete blood count, liver

function and kidney function tests, prothrombin time,

partial thromboplastin time.

Upper GIT endoscopy was done. Patients with selec-

tion criteria of bleeding ulcer were divided into two

groups: Group I: Consisting of 30 patients subjected to

clip application using a metallic clips (Hemoclip), Group

II: Consisting of 30 patients in whom Argon Plasma

Coagulation (APC) was done using an argon plasma

coagulator unit.

Clip application device: clip application was done us-

ing a metallic clips (Hemoclip; Olympus America, rota-

tional clip fixing device HX’6UR’1 through flexible

endoscopes). The clip fixing device length is 23 mm and

maximum insertion portion diameter is 2.8 mm with

processing port. Clips are loaded onto the fixing device

and drawn into a sheath. At the target lesion, the clip is

advanced out of the sheath, oriented with the rotational

handle, and then deployed. The mechanism of hemosta-

sis is mechanical compression [10].

Argon Plasma Coagulation (APC) was done using an

argon plasma coagulator unit (TERNO ABC TOM 201,

Germany). Spray mode was used with 2 power/gas set-

tings (respectively, 40 and 70 W and 1.5 to 3 L/min).

Probe of 2.3 mm was used with endoscopes with corre-

sponding channel diameters (2.8 mm diameter accessory

channels). The maximum coagulation depth achieved by

APC is 3 to 4 mm, which minimizes the risk of perfora-

tion. Continuous suction was applied to remove smoke

and prevent over inflation of the GIT [3].

Follow up: After endoscopy, all patients were closely

monitored clinically for one weak looking for symptoms

and signs of bleeding. All patients received the same

proton pump inhibitor, and Blood transfusion was given

to maintain the hemoglobin level above 8 g/dL. Clinical

recurrent bleeding was defined as signs of bleeding:

vomiting of fresh blood, passage of melena with pulse

rate higher than 100beat/min, decrease in systolic blood

pressure exceeding 30 mmHg, after the early stabiliza-

tion of pulse, blood pressure, and or decrease in hemo-

globin concentration by at least 2 g/dL over a 24-hour

period.

In case of re bleeding endoscopy was repeated as an

emergency procedure and the same primary endoscopic

management was used. Indications for surgery ; where

failed endoscopic treatment on second endoscopy, re-

currence of bleeding after a second therapeutic endo-

scopy, or a total blood transfusion requirement of greater

than 8 units to maintain a hemoglobin level of 10 g/dL.

2.1. Statistical Analysis

Data were statistically described in terms of mean ±

standard deviation (± SD).

Chi–square test of significance was used in order to

compare proportions between two categorical variables.

For comparing between two means, t-test of significance

was done and one way analysis of variance ANOVA was

used when comparing between more than two means.

When data were not normally distributed, nonparametric

Mann-Whitney test was used for comparing between

H. S. Assal et al. / HEALTH 2 (2010) 634-638

636

two means. A probability value (p value) less than 0.05

was considered statistically significant.

All statistical calculations were done using computer

programs Microsoft Excel version 7 (Microsoft Corpo-

ration, NY, USA) and SPSS (Statistical Package for the

Social Science; SPSS Inc., Chicago, IL, USA), version

16, statistical program.

2.2. Results

The characteristic stigmata of bleeding ulcers according

to their appearance at endoscopy were listed in Table 1,

there was insignificant difference between the 2 groups

regarding the presence of ulcer with visible or with ooz-

ing vessel (P > 0.05).

Table 2 shows significant difference between the 2

groups regarding Forrest’s classification (P < 0.05).

Table 3 shows insignificant difference between the 2

groups regarding the Rockall’S Score (P > 0.05).

Table 4 shows the characteristic stigmata of bleeding

ulcers according to their site at endoscopy, there was no

significant difference between the 2 groups regarding

presence of gastric or duodenal ulcer (P > 0.05).

There was a highly significant difference in re bleed-

ing incidence in different Forrest’s classes in group I (P

< 0.01), while there was insignificant difference in re

bleeding incidence in different Forrest’s classes in group

II (P > 0.05) as shown in Table 5.

Also there was a significant occurrence of re bleeding

with higher Rockall’s Score in group I. (P < 0.05), while

there was insignificant occurrence of re bleeding with

higher Rockall’s Score in group II (P > 0.05) as shown

in Table 6.

Table 7 shows insignificant relation between < 5 or

_5 Rockall’s score and the occurrence of re bleeding in

the whole patient population (P > 0.05). There was also

insignificant difference in the 2 groups regarding occur-

rence of re bleeding in relation to the site of ulcer, gas-

tric ulcer (GU) or duodenal ulcer (DU) (P > 0.05).

Table 1. Characteristic stigmata of bleeding ulcers according to

their appearance at endoscopy.

Ulcer with visible vessel Ulcer with oozing vessel

Negative Positive Total Negative PositiveTotal

N 26 4 30 16 14 30

Group I

% 86.67 13.33 100.00 53.33 46.67100.00

N 24 6 30 22 8 30

Group II

% 80.00 20.00 100.00 73.33 26.67100.00

Chi-

square X2 0.240 1.292

P-value > 0.05 (N.S) > 0.05 (N.S)

N.S: non significant; S: significant

Table 2. Forrest’s classification for the patient groups.

Forrest’s Classification

IA IIA IB IIB Total

N 0 4 14 12 30

Group I

% 0.0013.33 46.67 40.00100.00

N 10 6 8 6 30

Group II

% 33.3320.00 26.67 20.00100.00

X2 8.981

Chi-square

P-value< 0.05 (S)

Table 3. Rockall’s Score for the patient groups.

Rockall’s Score T-test

Group

Range Mean SD t P-value

Group I3.0008.0004.933 1.668

Group II3.0009.0005.333 2.059

–0.585 > 0.05(N.S)

Table 4. Characteristic stigmata of bleeding ulcers according to

their site at endoscopy.

GASTRIC OR DUODENAL

DU GU Total

N 14 16 30

Group I

% 46.67 53.33 100.00

N 20 10 30

Group II

% 66.67 33.33 100.00

X2 1.222

Chi-square

P-value> 0.05 (N.S)

Table 5. Incidence of rebleeding in relation to Forrest’s classi-

fication.

Forrest’s Classification Chi-square

Group Rebleeding

IA IIAIB IIB Total X2P-value

N 0012 12 24

Negative

%0.000.00 40.0040.00 80.00

N0 4 2 0 6

Group

Positive

%0.0013.336.67 0.00 20. 0

N6 6 4 4 20

9.643 < 0.01

(H.S)

Negative

% 20.0020.0013.3313.33 66.67

N4 0 4 2 10

Group

Positive

%13.33 0.00 13.33 6.67 33.33

2.100 > 0.05

(N.S)

H. S. Assal et al. / HEALTH 2 (2010) 634-638

637

Table 6. Incidence of rebleeding in relation to Rockall’s Score.

Rebleeding

Negative Positive T-test

Rockall’s

Score

Mean SD MeanSD t P-value

Group I 4.417 1.240 7.0001.732 –3.014 < 0.05 (S)

Group II 5.900 2.234 4.2001.095 1.587 > 0.05 (N.S)

Table 7. Correlation between high risk Rockall’s score and

rebleeding.

Rebleeding

Rockall’s Score

Negative Positive Total

N 20 4 24

< 5

% 33.33 6.67 40.00

N 24 12 36

≥ 5

% 40.00 20. 00 60.00

X2 1.02

Chi-square

P-value > 0.05 (N.S)

3. DISCUSSION

Peptic ulcer bleeding is the most common cause of upper

gastrointestinal bleeding, responsible for about 50% of

all cases. Mortality is increasing with increasing age and

is significantly higher in patients who are already admit-

ted in hospital for co-morbidity [3].

Risk factors for peptic ulcer bleeding are non-steroid

anti-inflammatory drugs (NSAIDs) use and Helicobacter

Pylori (HP) infection [8].

In patients with ulcers presenting with ongoing bleed-

ing or high risk features (Forrest I, IIA, IIB); surgery

was frequently required in the past to solve the situation.

However, endoscopic therapy has been well documented

to treat these ulcers [11].

The timing of the initial endoscopy has been debated.

In general, red hematemesis indicates emergency upper

endoscopy while black hematemesis and/or melena

without haemodynamic instability can wait until normal

working hours. However, from a logistic point of view

early endoscopy has been advocated to ensure optimal

utilization of resources [11] .In this study there was no

significant difference in both groups regarding age,

shock, presence of co morbid illness or liver cell failure,

ulcer size, rockall score and site of ulcer; factors known

to affect prognosis in many previous studies.

Our study showed that the rate of re bleeding was

slightly higher in APC group despite of being statisti-

cally insignificant. Also there was no significant relation

between the rates of re bleeding and the size of the ulcer.

Few reports have concerned the indication for and effi-

cacy of each hemostatic therapy according to location,

depth and size of ulcer and bleeding activity of the ex-

posed vessel as if the ulcer is large or deep, the possibil-

ity of complications including further ulceration, recur-

rence of bleeding and perforation is high [12].

A great care is required in performing the procedure if

the bleeding ulcer is located on the posterior wall or

lesser curvature of the gastric body or on the posterior

wall of the duodenal bulb, the hemostatic rate is lower

than for other therapies because of the technical diffi-

culty of approaching the lesion [13].

In the present study although there was no statistical

significance difference in re bleeding incidence in both

groups there was highly significant difference in re

bleeding incidence in relation to different Forrest’s

classes in group I (P < 0.01), while there was insignifi-

cant difference in re bleeding incidence in different

Forrest’s classes in group II. Also, the rate of surgical

interference of both groups was 0%.

In recent years, the Rockall score has been used to se-

lect patients with a low risk of rebleeding for early dis-

charge. Almost all patients in this low risk group belong

to patients without any stigmata of recent hemorrhage

(SRH). However, patients with a SRH are a high-risk

group for further re-bleeding and also mortality. It is

therefore important to determine whether the Rockall

score could be useful in patients who have undergone

endoscopic therapy for UGIB to identify high-risk pa-

tients and thus improve their management and outcome

[14].

In the present study we assessed correlation between

high risk Rockall’s score (> 5) and occurrence of re

bleeding which re bleeding was 6.8 % in low risk Rock-

all’s score (< 5) while re bleeding was 20% in high risk

Rockall’s score (> 5). However, this was statistically non

significant, but incidence of re bleeding in relation to

high risk Rockall's score was significant in group I. This

did not go in agreement with others [12], who concluded

that the Rockall scoring system accurately identifies

patients at high risk of death but not of re bleeding [12].

In spite that our study partially goes with others, who

observe good correlation between the Rockall score and

both the probability of re bleeding and mortality in pa-

tients undergoing endoscopic therapy for peptic ulcer

hemorrhage [15,16].

In the present study the mortality rates between the

two groups were the same which was 0% in the two

groups despite of significantly higher need for surgery in

group II.

This goes with others, who concluded that there was

H. S. Assal et al. / HEALTH 2 (2010) 634-638

638

no difference in all-cause mortality irrespective of the

modalities of endoscopic treatment [13,16].

Sung et al. in a meta-analysis of 15 studies reported

that regardless of improvements in sustaining hemostasis

by clipping leading to less re bleeding and fewer inter-

ventions with surgery, mortality has not been reduced

and there is no indication of a reduction in the death rate

[17]. Nevertheless, it is a mystery that despite successful

control of hemorrhage in many studies using various

combinations of endoscopic and pharmacological thera-

pies the mortality rate remains unchanged.

4. CONCLUSIONS

Endoscopic application of hemoclips have a less re

bleeding rate than Argon plasma coagulation for treat-

ment of bleeding peptic ulcer, although this was statisti-

cally insignificant meanwhile APC is still less cost and

easy. Clinical and endoscopic assessment (through

Rockall score and Forrest classification) could help in

making best choice for endoscopic management.

REFERENCES

[1] Barkun, A., et al. (2003) Consensus for managing pa-

tients with non variceal upper gastrointestinal bleeding.

Annals of Internal Medicine, 139(10), 843-857.

[2] Park, C.H., et al. (2004) Optimal volume for epinephrene

for endoscpic prevention of recurrent peptic bleeding.

Gastrointestinal Endoscopy, 60(6), 875-880.

[3] Canard, J.M. and Vedrenne, B. (2001) Clinical applica-

tion of argon plasma coagulation in gastrointestinal en-

doscopy: has the time come to replace the laser? Endo-

scopy, 33(4), 353-357.

[4] Church, N.I., Dallal, H.J., Masson, J., et al. (2003) A

randomized trial comparing heater probe plus thrombin

with heater probe plus placebo for bleeding peptic ulcer.

Gastroenterology, 125(2), 396-403.

[5] Devereaux, C.E. and Binmoeller, K.F (1999) Endoclip:

closing the surgical gap. Gastrointestinal Endoscopy, 50

(3), 440-442.

[6] Forrest, J.A.H., Finlayson, N.D.C. and Shearman, D.J.C.

(1974) Endoscopy in gastrointestinal bleeding. Lancet,

2(7877), 394-397.

[7] Rockall, T.A., Logan, R.F., Devlin, H.B. and Northfield,

T.C. (1995) Incidence of and mortality from acute upper

gastrointestinal haemorrage in the United Kingdom.

Steering Committee Gastrointestinal Haemorrhage, Brit-

ish Medical Journal, 311(6999), 222-226.

[8] Leerdam, V. (2008) Epidemiology of acute upper

gastrointestinal bleeding. Best Practice & Research Clini-

cal Gastroenterology, 22(2), 209-224.

[9] Sung, J. (2006) Current manegement of peptic ulcer

bleeding. Nature Clinical Practice Gastroenterology &

Hepatology, 3, 24-32.

[10] Chuttani, R., Barkun, A., Carpenter, S., et al. (2006)

Endoscopic clip application devices. American Society

for Gastrointestinal Endoscopy, 63(6), 16-51.

[11] Aabakken, L. (2008) Current endoscopic and pharma-

cological therapy of peptic ulcer bleeding. Best Practice

& Research Clinical Gastroenterology, 22(2), 243-259.

[12] Saperas, E., Sebastian, V., Carolina, B., Joan, D., Jose, R.

and Juan, R.M. (2008) Applicability of the Rockall scor-

ing system and prediction of rebleeding and mortality af-

ter combined pharmacologic and endoscopic treatment of

high-risk bleeding peptic ulcers. Gastrointestinal Endo-

scopy, 67(5), AB255.

[13] Chung, I.K., Ham, J.S., Kim, H.S., et al. (1999) Com-

parison of the hemostatic efficacy of the endoscopic

hemoclip method with hypertonic saline-epinephrine in-

jection and a combination of the two for the management

of bleeding pepic ulcers. Gastrointestinal Endoscopy,

49(1), 13-18.

[14] Bessa, X., O’Callaghan, E., Ballesté, B., Nieto, M., et al.

(2006) Applicability of the Rockall score in patients un-

dergoing endoscopic therapy for upper gastrointestinal

bleeding. Digestive and Liver Disease, 38(1), 12-17.

[15] Church, N.I. and Palmer, K.R. (2001) Relevance of the

Rockall score in patients undergoing endoscopic therapy

for peptic ulcer hemorrhage. European Journal of Gas-

troenterology and Hepatology, 13(10), 1149-1152.

[16] Chuan, C., Ming-Szu, H., Te-Fa, C., Jih-Chang, C. and

Cheng-Ting, H. (2007) Risk scoring systems to predict

need for clinical intervention for patients with non

variceal upper gastrointestinal tract bleeding. American

Journal of Emergency Medicine, 25(7), 774-779.

[17] Sung, J.J., Tosi, K.K., Lai, L.H., et al. (2007) Endoscopic

clipping versus injection and thermocagulation in treat-

ment of non variceal upper gastrointestinal bleeding: A

meta-analysis. Gut, 56, 1364-1373.