Rectal varices (RVs) are an important cause of lower gastrointestinal bleed (LGIB) in portal hypertension (PHT) and have been reported to occur in 44% to 89% of cases of cirrhosis. RVs are dilated sub-mucosal porto-systemic communications which extend from mid rectum to the ano-rectal junction and are considered distinct from internal hemorrhoids, which are submucosal arterio-venous communications of the anorectal vascular plexus. The suspicion of RVs as the cause of bleeding can be made with a high index of suspicion when lower GI bleed is seen in absence of hemorrhoids, and colonoscopy shows blood in rectum. Bleeding usually happens from endoscopically evident rectal varices (EERV) but sometimes bleed can occur from varices, which are endoscopically in evident (EIERV). Endoscopic ultrasound (EUS) has been shown to be more sensitive in diagnosis of EIERV. Endoscopic and EUS correlation of RVs has shown that RVs, classified as tortuous, nodular, and tumorous on endoscopic examination, have corresponding appearances on rectal EUS as single, multiple, and innumerable submucosal veins, respectively. The hemodynamic evaluation (HDE) of RVs by EUS is routinely done at some centers to assess parameters like the site, size, velocity, or direction of flow.

AIM:

To describe the clinical, radiological, endoscopic and endoscopic-ultrasound features of non-bleeding and bleeding rectal varices, among patients with portal hypertension.

OBJECTIVES:

Primary objectives:

Anatomical and liver -related risk-factors for presence of rectal varices (RV) among patients with portal hypertension Secondary objectives

1. Etiological spectrum of bleeding per-rectum, among patients with PHT.
2. Frequency of coexistence of hemorrhoids with RV.
3. Distribution and size of variceal channels in the rectum on endoscopy and endoscopic ultrasound (EUS).
4. Location and numbers of inflow and outflow perforator channels for RV on endoscopic ultrasound (EUS).

Patients and methodology

Study Population:

All patients with PHT (cirrhotic or non-cirrhotic), with active or prior anorectal bleeding (defined later) in the preceding 6-months will be evaluated for inclusion. We will also include patients with incidental detection of RV during endoscopy done for other indications. We will include both indoor and outdoor patients, attending the Department of Hepatology, ILBS, New Delhi. An informed consent will be obtained from the participants in the study.

Study Design Cross- sectional, observational and descriptive study

Study period September 2020 to February 2020. Baseline data

The following clinical, biochemical and radiological features will be recorded (detailed in the proforma):

1. Liver disease etiology and severity, including Fibroscan and HVPG, where done.
2. Upper endoscopy findings.
3. Details of prior bleeding episodes if any (upper and lower).
4. Details of current bleeding episodes.
5. Radiological features (cirrhosis, splanchnic venous thrombosis, flow direction in portal vein (PV) and its tributaries, inferior mesenteric vein (IMV) diameter, enlarged para-rectal collateral circulation, large porto-systemic shunts).

Endoscopy protocol:

All patients will undergo a sigmoidoscopy or colonoscopy procedure, as per the clinical indication. Patients found to have RV will undergo EUS examination in addition. EUS evaluation of patients with hemorrhoids alone will be done at the discretion of the examiner after informed consent from the patient.

Bowel preparation: Patient will be advised liquid-soft diet for 24 hours before the examination day. Split dose PEG preparation (2-4 L volume) will be administered- 2L in evening before and 2L on day of procedure till 4hours before the study scheduled time.

Patient position: Left lateral position. Endoscopes: Colonoscope or gastroscope. EUS probes: Radial and linear EUS scopes, paired with Olympus compact EUS processor EU-M2.

Operators: Procto-sigmoidoscopy: HS supervised by VB/ VB EUS: VB/ HS supervised by VB

Study Definitions:

Lower GI bleeding LGIB will be defined as bleeding from a source distal to the ileocecal valve (24).

Acute lower gastrointestinal bleeding Acute LGIB will be defined as bleeding of recent duration (<3 days) that may result in hemodynamic instability, anemia, and/or the need for blood transfusion (25).

Clinically significant lower gastrointestinal bleeding[VB1] Requirement for blood transfusion, a hemoglobin drop of > 3g/dL from baseline or need for hospital admission.

A bleeding episode is clinically significant when there is (BAVENO III[VB2] ):

1. Transfusion requirement of ≥ 2 units of blood within 24 hours of time zero, 2. Systolic blood pressure < 100 mmHg or a postural change of >20 mmHg, 3. Pulse rate >100/min at time zero Anorectal bleeding

1. Anorectal bleeding will be defined as red bleed per-rectum without or with stool expulsion. In the latter instance it can be mixed with stools, may smear stool, may dribble after defecation, or may smear wipes.
2. Anorectal bleeding may be painless or may be associated with pain.
3. Anorectal bleeding may or may not be associated with hemodynamic changes.

Monitoring and assessment

Primary:

1. Anatomical and liver -related risk-factors for presence of rectal varices (RV) among patients with portal hypertension

Secondary

1. Etiological spectrum of bleeding per-rectum, among patients with PHT.
2. Frequency of coexistence of hemorrhoids with RV.
3. Distribution and size of variceal channels in the rectum on endoscopy and endoscopic ultrasound (EUS).
4. Location and numbers of inflow and outflow perforator channels for RV on endoscopic ultrasound (EUS).

Statistical analysis The dependent variable in the study will be presence of RV, presence of large RV (>5mm), and RV bleeding. Independent variables will be presence/ absence of esophageal varices, gastric varices, ectopic varices, prior UGIE bleeding, prior endotherapy for esophago-gastric varices, ascites, gastro-lieno-renal shunt, other large porto-systemic shunt, PV and/or SV and/or SMV thrombosis or flow reversal, composite liver function scores (CTP and MELD), number, location, and size of rectal perforator channels, size of para-rectal collaterals, pulsatile/ phasic flow in perforator(s) and/or RV, and presence of hemorrhoids. The continuous data will be represented by Mean ± SD or by Median (IQR) as appropriate. The categorical data will be represented as frequency (%). Student t-test or Mann Whitney test will be used for quantitative data and Chi square test will be used for qualitative data. Besides this an appropriate statistical analysis like uni-variate and multi-variate logistic regression will be used at the time of data analyss. The significance will be seen at 5 %.

Adverse Event:

1. Pain abdomen
2. Bloating sensation
3. Abdominal distension

Stopping Rule:No stopping rule

Ethical Issues in the study and plans to address these issues:No ethical issue related to my study

Inclusion Criteria:

1. Age >18-years.
2. Patients with PHT and active or prior anorectal bleeding in preceding 6-months.
3. Patients with RV (bleeding or non-bleeding) detected on LGI endoscopy.
4. Ability to understand study procedures, and to comply with them for the entire length of the study.

Exclusion Criteria:

1. Peri-anal infection, peri-anal warts, anal condyloma, or fistulizing disease.
2. Proctitis- infective, inflammatory bowel disease (IBD), radiation, or unspecified.
3. Pregnant or lactating females.
4. Presence of hemodynamic instability.
5. Presence of encephalopathy
6. Non-consenting adult patient or guardian.
7. Prior history of banding or surgery for hemorrhoids, or endotherapy for RV.
8. Prior shunt occlusion procedures like BRTO or PARTO, or shunt creation procedures like TIPS.

All patients with PHT (cirrhotic or non-cirrhotic), with active or prior anorectal bleeding (defined later) in the preceding 6-months will be evaluated for inclusion. We will also include patients with incidental detection of RV during endoscopy done for other indications. We will include both indoor and outdoor patients, attending the Department of Hepatology, ILBS, New Delhi. An informed consent will be obtained from the participants in the study.