The study of
UPPER GASTROINTESTINAL BLEEDING
Dr. Rajesh Kashyap (M.D.)

Part 6 .......

DISCUSSION

Upper GI bleed is one of the most common gastrointestinal emergency. Acute upper gastrointestinal (UGI) bleeding occurs frequently prompting hospital admission or complicating another illness⁵³. Despite advances in critical care, monitoring and hemodynamic support, mortality rates of 10% to 20% continue to be reported (54,55). While spontaneous cessation of bleeding occurs in as many as 85% of cases⁵⁶, early intervention is required in those in whom bleeding does not stop spontaneously. Endoscopic diagnosis and therapy is an integral part of the management of these patients⁵⁷.

Dagradi et al studied 500 patients and found variceal bleed in 16%, Mallory Weiss tear in 3%, peptic ulcer in 11% and acute mucosal lesions in 34%⁵⁸. ASGE studied 2097 cases in 1981 and found cause of UGIB variceal in 15%. Mallory Weiss tear in 8%, peptic ulcer in 45% and acute erosive mucosal disease in 30%³¹. In de Bombal ⁵⁹ study variceal bleed in 13%, Mallory Weiss in 2%, peptic ulcer in 36%. Acute gastric mucosal disease in 7%. Peptic ulcer remains the commonest cause of upper GI bleeding in nearly all series but there is considerable variation in proportion of patients with variceal bleeding which tends to be low in UK series and high in those from USA^19,20.

In the present hospital based study the incidence of peptic ulcer was found 67% (DU-48%, GU 19%) this incidence is higher than those of other studies but closely follows the study of C. Roll hanser⁶⁰. The reason for increased incidence of peptic ulcer in our study is not exactly known but may be due to the diet habits. Mallory Weiss syndrome was found in 10.8% cases of UGIB in our study and it closely resembles the ASGE study of 1981³¹. In our study of 111 patients studied over one year showed the incidence of erosive mucosal disease as 13%, which is quite low than that of other studies but closely relates to the study conducted by de-Bombal⁵⁹. The reason for low incidence of erosive mucosal disease in our study may be less consumption of NSAIDS by our patients, less alcohol intake by our patients and we may not be subjecting all the critically ill patients for upper GI endoscopy to rule out stress bleeding. In one series the gastro intestinal tumors were the cause in 5% cases in patients with UGIB⁶¹. In our study the tumors of gastrointestinal tracts accounts for 3.6% cases of UGIB which closely matches with the above study.

The incidence of variceal bleed in our study was 10.8% and it is low as compared to other studies^58,59,31. It may be due to less number of cirrhotic patients or the patient may not be able to reach our hospital which is tertiary level, may be collapsing at secondary level only. There is possibility of less consumption of alcohol also.

In our study of 111 patients there was history of drug intake in 43 (38.7%) patients and no history of drug intake in 68 (61.3%) patients the history of drug intake is not statistically significant. Also history of alcohol intake is not statistically significant. The history of symptoms ulcer like is present in 78 (70.3%) patients, history of portal hypertension present in 9 (8.1%), history of M.W. tear like is present in 11 (9.9%) patients which is highly significant statistically.

In our study of 111 patients clinical impression of DU was kept in 56 (50.5%) patients and finally 48 (43.9%) patients came out to be DU which is statistically significant. In GU clinical impression and final diagnosis is not statistically significant for portal hypertension. Clinical impression is kept in 11 patients and finally 12 patients came out to be of portal hypertension out of study group of 111 patients which is statistically highly significant, for MW clinical impression a final diagnosis is highly significant.

Sugowa et al in 1990 found that upper GI bleed was managed conservatively in most of the patients (89.6%) the emergency operation was performed in 7.8% of cases to control the bleeding⁶².

In our hospital based study of 111 patients PPI/Triple therapy was given to 99 (83.8%) patients sclerotherapy was given to five patients of portal hypertension. Bands were applied to two patients of portal hypertension and somatostatin/octeriotide to two patients of portal hypertension. In our series of 111 patients, stigmata of recent haemorrhage was present in 11 patients out of which ten were given local temponade in the form of injection therapy. The blood transfusion of more than 4 units are required in 20 patients.

In another study conducted in 1997 it was found that need for surgery is only 6.1% in the general medical community in USA. Almost three quarters of the patients who required surgery had no prior attempt at theraputic endoscopy⁶⁰.

In our study of 111 patients of upper GI bleed most of the patients managed conservatively 90.9% which matches the above study closely⁶². The patient in our study requiring surgery was 2.7% which is low than thatof other study but reason could be that in our institution surgery for portal hypertension is not being done or most of the patient with severe and massive bleed may not reach this institution due to topography and weather conditions.

In an overview of upper GI bleed by E.van de mierop in 1996 found that mortality for upper GI bleed is 7%⁶³ an another study revealed drop in mortality rate by almost half from high 15-20yrs ago. 5.9 per lakh in 1955 to 2.7 per lakh in 1977 for peptic ulcer. Total 4.5 mortality rate for all causes was .95-.96%⁴. In Silvershy series in 1981 found that mortality for esophageal varies was 30%, DU 7%, gastric ulcer 8%, Mallory Weiss 4% and acute mucosal lesion 10%³¹.

Larson in 1986 found that death rate in esophageal varices was 36%, DU 8%, gastric ulcer 16%, MW tear 17% and acute gastric mucosal lesion 27%⁵⁵. Katschinski in 1989 found mortality in esophageal varices 17%, in duodenal ulcer 9%, in gastric ulcer 12%, MW tear 2% and acute gastric mucosal lesion 5%⁶⁴.

In our study group the death rate for esophageal varices patient was 50% which is higher than that of other study may be due to non-availability of surgical back up for these patients or the patient reach in the hospital at terminal stage.

The mortality for duodenal ulcer patient in our series was 2.08% which is low than other series, this could be due to availability of PPI and triple therapy. No death was reported in patients of GU and MW tear. In our series while two patients of gastric tumor died due to excessive haemorrhage.

CONCLUSION

1. Peptic ulcer disease is the most common cause of upper GI bleed in our study. Duodenal ulcer patients were 48 (43.9%) and gastric ulcer patients 19 (17.1%)] followed by erosive mucosal disease, variceal bleed and Mallory Weiss syndrome.
2. In peptic ulcer group most of the patients were in the age group of 30-50 years; eleven patients had stigmata of recent hemorrhage on upper GI endoscopy, ten were given injection therapy, two patients required surgery and one patient died.
3. The incidence of variceal bleed, erosive mucosal disease and Mallory Weiss syndrome was almost equal. In variceal bleeding most patients falling into 30-60 years of age, two patients were given EVL, five sclerotherapy, two somatostatin/octeriotide infusion. In this group four patients died during hospital stay.

All rights reserved with the author, including the rights of reproduction in whole or part, in any form. No part of the publication, images, photographs, design or any other content may be reproduced or transmitted in any form or by any means, electronic or mechanical including photocopying, recording, electronic replication, or any information storage and retrieval system, without permission in writing from the copyright holder. However the study may be quoted and referred to in any other research, medical or allied work or/and project without any restriction.

Copyright © reserved with Rajesh Kashyap, Year- 2000