ENDOSCOPIC BAND LIGATION IN PATIENTS WITH VARICEAL BLEEDING

This was a prospec ve observa onal study conducted in the Department of Surgical Gastroenterology, College of Medical Sciences and Teaching Hospital (COMS-TH), Bharatpur, Nepal over a period of 2 years. Pa ents a ending the COMS emergency/surgical OPD with history of UGI bleed with features consistent with portal hypertension during this period were enrolled. The endoscopic variceal band liga on was done based on similar principles applied to Barren and liga on in the treatment of internal haemorrhoids.


INTRODUCTION
Esophageal varices (EV) affect about 50% of pa ents with 1 liver cirrhosis and forma on rate is about 5% per year.The overall incidence of bleeding from EV ranges from 25 to 30% at two years from the diagnosis.Mortality rate from the ini al EV bleeding can s ll rise up to 30% in pa ents with 2 advanced liver disease and large varices.Treatment op ons, such as vasoac ve drugs, endoscopic therapy, balloon tamponade, surgical and radiological shunt (TIPS) can be helpful to stop EV bleeding.
Endoscopic therapy is widely used for treatment of variceal bleeding.Sclerotherapy has been proven effec ve in control of acute bleeding and is also effec ve in preven ng 3 variceal rebleeding.However, endoscopic sclerotherapy requires a skilled endoscopists and is quite frequently associated with adverse events.Banding liga on has been compared with sclerotherapy in several RCT's for the longterm preven on of variceal bleeding and found superior to sclerotherapy.Banding liga on has fewer complica ons and fewer procedures are needed to eradicate oesophageal varices compared to sclerotherapy, and should therefore 4 most likely be preferred.
Upper gastrointes nal bleed secondary to oesophageal varices due to portal hypertension is one of the deadly presenta on in our emergency and OPDs.In this study we tried to study the pa ern of liver disease manifes ng with portal hypertension with oesophageal variceal bleeding in the Nepalese popula on, risk factors for variceal bleeding, endoscopic band liga on of the varices and its outcome.

METHODOLOGY
This was a prospec ve observa onal study conducted in the department of Surgical Gastroenterology, College of Medical Sciences and Teaching Hospital (COMS-TH), Bharatpur, st Nepal over a period of 2 years from 1 August 2014 to July st 31 2016.The ethical clearance was taken from ethical review commi ee of COMS-TH.Pa ents a ending the COMS emergency/surgical OPD with history of UGI bleed with features consistent with portal hypertension during this period were enrolled.Pa ents with endoscopic diagnosis of oesophageal varices secondary to portal hypertension were included in the study but those pa ents with age less than 18 years, with cirrhosis harbouring Hepatocellular carcinoma (HCC), with esophageal varices with prior endoscopic or surgical management of esophageal varices and with evidence of fundal varices were excluded.Wri en consent was taken before performing endoscopic banding.
In the emergency ac vely bleeding cases were resuscitated and stabilized by giving I.V. fluids, blood transfusion and inj.Octreo de.All the pa ents were admi ed in the hospital.Detailed history was taken and thorough clinical examina on was done.The pa ent's demographics were recorded in the history.Esophagogastro-duodenoscopy was done to establish the cause of UGI bleed.If esophageal varices were present they were graded into four grades (Paquet's classifica on).Pa ents with ac ve variceal bleed were subjected for endoscopic variceal band liga on while performing the diagnos c endoscopy.
The endoscopic variceal band liga on was done based on similar principles applied to Barren and liga on in the treatment of internal haemorrhoids, the bands used were elas c rubber bands ('O' rings).The band liga on procedure was conducted under topical anaesthesia and I.V.Seda on (Inj.Diazepam).
All the procedures were carried out using PENTAX EG-2770K Fibre Op c Endoscope and the Mul Band Ligator-Barrel device (SHAILI ENDOSCOPY, GUJARAT, INDIA).
Treatment by band liga on was started at the distal part of esophagus usually at the gastro esophageal junc on.The liga on was done on an average of 3-4 lesions and maximum of '6' bands were applied in a single si ng.
In ac vely bleeding pa ents the blood was flushed with water, sucked and the elas c rubber 'O' ring was directly placed on the varix at the point of bleeding site.If bleeding was not controlled than banding was done over the mucosa above and below the point of bleeding.
Pa ents who had emergency procedures were admi ed in the surgical ward with the following advices: 1 Pa ents who had elec ve endoscopic bandliga on were kept for two hours for observa on in the recovery room and were discharged with cap omeprazole 20 mg BD for 2-4 weeks depending on the interval between sessions and syrup Sucralfate 2tsf tds for 1 week.(To help in the healing process of esophageal ulcers produced due to liga on).
The follow up was done three weekly a er the index procedure ll all the varices were eradicated and then three monthly for next six months.Endoscopic sessions were done in between the follow-up if pa ents presented with features of ac ve UGI bleeding from esophageal varices.In the follow up following findings were noted endoscopically.
[ Ulcers of the oesophagus at the site of liga on.

STATISTICAL ANALYSIS
Data were entered directly into SPSS version 20 and were analysed.In descrip ve sta s cs for the qualita ve variables frequency and percentage was shown,for quan ta ve variable means and standard devia on was calculated.In the inferen al sta s cs to find the associa on between variables chi-square test was used.Stas cal significance was considered if p-value was less than 0.05.

RESULTS
During this period of two years though 63 pa ents with UGI bleed with oesophageal varices were evaluated in the study, however only 50 were analysed.Thirteen pa ents were excluded because five of them had history of prior endoscopic procedure, three had gastric varices, one pa ent had Hepatocellular carcinoma and four pa ents did not come for regular follow up.In this study oesophageal varices secondary to portal hypertension was more commonly seen in the age 40 years to 65 years as shown in figure 1.The mean age of the study popula on was 52.26+/-12.80years and male predominance was seen (M:F=42:8).Among the causes of portal hypertension alcoholic chronic liver disease (CLD) (74%) was the most common cause of portal hypertension with variceal bleed followed by Hepa s B virus CLD (12%), extra-hepa c portal hypertension (10%) and cryptogenic CLD.Coming to Child's Pugh score most of the pa ent were of Child's B (54%) followed by C(24%) and A(22%).Grade III varices were seen in 36%, grade II in 32%, grade IV in 24% and grade I in 8% cases.Acute bleeding of varices was present in nine (18%) cases.On an average 3.8 (minimum two and maximum five) sessions were required to obliterate the varices.Control of acute bleed was achieved in 77.78% cases.Variceal eradica on was achieved in 96% cases, recurrence of variceal bleed was seen in 22% cases and recurrence of varices a er eradica on was seen in 14% cases.Complica ons following banding were seen in 20% cases with dysphagia seen in 8% cases.Mortality was seen in two (4%) cases.The cause of mortality in both cases was acute bleed with decompensated liver failure in pa ent with Child's Pugh grade C with grade IV varices.From the table 1, 2, 3 and 4 we can see that variceal rebleeding and variceal recurrence were more with higher grades of varices and higher grades of Child's Pugh score and these findings were stas cally significant ( p<0.05).

DISCUSSION
Among several complica ons of portal hypertension, including ascites, hepa c encephalopathy, and renal insufficiency, variceal hemorrhage is the most drama c with a high mortality rate.Endoscopic therapies for varices aim to Figure 1 Male predominance was seen in the current study with males comprising 84% of the study popula on.This finding may be due to the fact that alcoholic cirrhosis was one of the major cause of portal hypertension leading to variceal bleed in our study.with acute bleed at index presenta on.
A meta-analysis comparing the use of sclerotherapy and band liga on published in 2006 involving 1309 pa ents has shown the effcacy of endoscopic SCL for ini al hemostasis to be on average 95% (76%-100%), and that of EVL to be on average 12 97% (86%-100%).In the current study EVL controlled acute bleed in 77.78% of pa ents which is slightly lower than shown in the meta-analysis and this can be explained by the small sample size in our study.
Various studies have shown size of varices as a risk factor for variceal bleed and recurrence.In the present study, the frequency of grade II varices in 32% cases, Grade III varices in 36% cases and grade IV varices in 24% cases were observed.These findings were more or less similar to the  1997) the eradica on of varices, re-bleeding of varices and recurrence of varices by endoscopic band liga on was 87%-96%, 6%-27% and 29-48% respec vely.In the current study the eradica on of varices, re-bleeding of varices and recurrence of varices by endoscopic band liga on was 96%, 22% and 14% [14][15][16] respec vely.
The probability of variceal bleeding is significantly greater in pa ents with larger esophageal varices, with more severe red wale marks and with more severe liver dysfunc on  1997) reported complica ons ranging from 5%-45% and mortality ranging from 6%-22% following endoscopic band liga on in the management of esophageal varices.In the current study complica ons was seen in 20% cases and mortality in 4% cases and these findings are more 4,14-16 or less comparable.

CONCLUSIONS
From the current study it can be concluded that alcoholic cirrhosis is the most common cause of portal hypertension and hence variceal bleeding in the Nepalese popula on.Pa ents with larger varices and high Child's pugh score were having more recurrent bleeding and more recurrence.Endoscopic banding of varices in the current study was comparable to other studies.

[
Scars at the site of liga on [ The number of residual varices, if any [ Number of variceal columns [ Reduc on in the grading of varices compared to varices before liga on [ Re-bleeding of the varices, if any.[ During the follow up the remaining varices were ligated or sclerosed (depending upon the index procedure) un l they reduced in diameter to grade one or they could not be ligated or injected.A er successful treatment of varices the pa ents were advised to come three monthly for next six months during which the recurrence of varices or any complica ons were noted.Belbase N et al Birat Journal of Health Sciences Maskey et al. (2011) in a study from Nepal showed male predominance in alcoholic cirrhosis.⁷In different studies comparing endoscopic band liga on with endoscopic sclerotherapy, S egmann et al. (1992) and 5-6 Shafqat et al. (1998) also found male predominance.Although portal hypertension could result from pre-hepa c abnormali es, post-hepa c abnormali es, intrahepa c non-cirrho c causes, cirrhosis is by far the most common cause of portal hypertension.In the current study we found alcoholic cirrhosis as a cause of portal hypertension in 74% study popula on.Study done by Karki et al. (2013) found cirrhosis as a cause of portal hypertension in 78.3% of the study popula on in Nepal.S egmann et al. (1992) and de la Pena et al. (1999) in their study found alcoholic cirrhosis to be the major cause (81.6% and 65.9% respec vely) of 5,8,9 oesophageal varices.Different studies have shown that acute bleeding of the varices at presenta on is an independent risk factor for 10 variceal recurrence.In the current study around 18% of the study popula on was bleeding at index presenta on.S egmann et al. (1992) from USA; Lo et al. (1995) from China and Luz et al. (2011) from Brazil found similar presenta on of pa ents with 15-27% of the study popula on presen ng 5,11,12

12 findings
of Luz et al. (2011).Child-Pugh score at the bedside, is widely used as a simple descrip ve or prognos c indicator.Mul variate analyses using Child-Pugh score as an en ty have shown that it has an independent prognos c value in the se ngs of ascites, ruptured oesophageal varices, alcoholic cirrhosis and 13 decompensated HCV-related cirrhosis.In the present study Child's score A in 22%, Child's B in 54% cases and Child's C in 24% cases were found.Similar distribu on of pa ents was seen in study done by de la Pena et al. (1999) repor ng 9 Child's A-23.9%;Child's B-50.0% and Child's C in 26.1%.Variceal bleeding is the most severe outcome of portal hypertension.Endoscopic sclerotherapy and band liga on are endoscopic treatment modali es for both ac ve variceal bleeding and secondary prophylaxis.Variceal recurrence and re-bleeding are the common drawbacks with endoscopic treatment.From the studies by Hou et al. (1995), Sarin et al. (1997) and Avgerinos et al. (

17 according
to the Child-Pugh classifica on.de la Pena et al. (1999) in his study has shown that Child-Pugh class C and variceal grade IV had an influence on eradica on failure, Child-Pugh class was a definite index of survival and recurrence of variceal bleeding, and variceal grade was a 9 prognos c factor for variceal bleeding.In the current study also we found that variceal recurrence and rebleed increased with increase in the child pugh's score and increase in the grade of varices.Laine et al. (1993); Hou et al. (1995); Sarin et al. (1997); and Avgerinos et al. ( Belbase N et al Birat Journal of Health Sciences

Table 1 :
Rela onship between grade of varices and recurrent bleeding : Histogram showing age distribu on in the study

Table 2 :
Rela onship between grade of varices and recurrence of varices

Table 3 :
Rela on between Child's pugh score and recurrent bleeding

Table 4 :
Rela on between Child's pugh score and recurrence of varices tension by oblitera on of the varix.The two principal methods available for esophageal varices are endoscopic sclerotherapy (SCL) and band liga on (EVL).This study was conducted to evaluate the pa ern of liver disease manifes ng with portal hypertension with oesophageal variceal bleeding in the Nepalese popula on, risk factors for variceal bleeding, endoscopic band liga on of the varices and its outcome.
Birat Journal of Health Sciences reduce variceal wall