Idiopathic Ileocolonic Varices Coexisting with a Colon Polyp Treated Successfully by Endoscopy: A Case Report and Literature Review.

Colonic varices are usually associated with portal hypertension. Idiopathic colonic varices are extremely rare. A 68-year-old man with a positive fecal occult blood test result underwent colonoscopy. We detected idiopathic ileocolonic varices and a coexisting ascending colon polyp. While reviewing the literature, we found cases of biopsies and polypectomies resulting in significant bleeding. We herein report a case of idiopathic ileocolonic varices coexisting with a colon polyp treated successfully by endoscopy. The coexistence of colonic varices and a colorectal lesion that requires endoscopic treatment may lead to significant bleeding. During management, the development of a treatment strategy and obtaining informed consent are necessary.

Introduction

Colonic varices are rare and usually associated with portal hypertension. Idiopathic colonic varices are very rare. In most cases, the major complement is hematochezia. Colon polyps are typically treated by endoscopy, and bleeding is the most common serious complication occurring after endoscopic mucosal resection (1). The coexistence of colonic varices and colorectal lesions that require endoscopic procedures may cause significant bleeding. When reviewing case reports of previous idiopathic colonic varices cases (2-30), we found cases wherein significant bleeding was caused by a biopsy or polypectomy (20,31).

We herein report the first case of idiopathic ileocolonic varices coexisting with a colon polyp treated successfully by endoscopy.

Case Report

A 68-year-old man was referred to our hospital because he had positive fecal occult blood screening test results. He had a history of duodenal ulcer and operation for sinusitis in his 30s. During the examination of the duodenal ulcer, no esophageal varices were detected. He also had a history of hematochezia twice in his 50s. He had never been found to have any liver function abnormality or liver disease, portal hypertension, colon malignancy, pancreatitis, or congestive heart failure. He had no history of abdominal operation, blood transfusion or a family predisposition. He was a social drinker and never drank more than 60 g alcohol per day.

Total colonoscopy revealed the presence of ileocolonic varices in the terminal ileum and throughout the colon, extending from the ascending colon to the rectum (Fig. 1). In addition to ileocolonic varices, the patient had a 12-mm sessile polyp with a clear margin, slightly uneven surface, and faded color without reddish area in the ascending colon on the colonic varices. The Japan Narrow-band imaging Expert Team (JNET) classification (32) was type 2A (Fig. 2A-C).

Figure 1.

Total colonoscopy revealed the presence of ileocolonic varices at the terminal ileum and throughout the colon, extending from the ascending colon to the rectum, A: Terminal ileum, B: Ascending colon, C: Transverse colon, D: Descending colon, E: Sigmoid colon, F: Rectum.

Figure 2.

A: Conventional endoscopy in the ascending colon shows a 12-mm polyp, existing on the colonic varices, B: Narrow-band imaging, C: Chromoendoscopy, D: Endoscopic ultrasound with a small-caliber ultrasonic probe showed 3- to 4-mm colonic varices in the submucosa.

Endoscopic ultrasonography (EUS) using a small-caliber ultrasonic probe showed 3- to 4-mm colonic varices in the submucosa under the polyp (Fig. 2D). Considering the risk of bleeding after resection, we decided to further continue with the investigation of the portal hypertension and hemodynamics of the ileocolonic varices. The laboratory test results revealed a history of HBV infection, but no other data supported the suspicion of a present viral hepatitis, autoimmune disease, endocrine disease, or other liver disease. Mac-2 binding protein glycan isomer (M2BPGi) was within normal levels (Table 1).

Table 1.

Laboratory Data on Admission.

<Peripheral blood>

ALT

13

U/L

<Serology>

WBC

5,580

/μL

LDH

173

U/L

IgG

1,197

mg/dL

RBC

466×104

/μL

ALP

151

U/L

IgA

226

mg/dL

Hb

14.6

g/dL

γ-GTP

19

U/L

IgM

54

mg/dL

Ht

42.8

%

ChE

310

U/L

CRP

0.03

μg/dL

Plt

21.6×104

/μL

T-chol

248

mg/dL

Ferritin

49.8

μg/dL

<Coagulation>

TG

100

mg/dL

M2BPGi

0.39

COI

PT

111

%

UA

5.4

mg/dL

<Viral markers>

PT-INR

0.98

BUN

13.6

mg/dL

HBs-Ag

(-)

APTT

23

sec

Cr

1.1

mg/dL

HBs-Ab

(-)

FIB

258

mg/dL

NH3

42

μg/dL

HBc-Ab

(+)

FDP

<2.0

μg/mL

Na

139

mEq/L

HBe-Ag

(-)

D dimer

0.8

μg/mL

K

4.2

mEq/L

HBe-Ab

(+)

AT III

94

%

Cl

106

mEq/L

HBV-DNA (PCR)

(-)

<Biochemistry>

Fe

139

μg/dL

HCV-Ab

(-)

TP

6.5

g/dL

UIBC

179

μg/dL

CMV pp65 (C10/C11)

(-)

Alb

4.2

g/dL

Cu

89

μg/dL

EBV VCA-IgM Ab

<10

×

T-Bill

1.9

mg/dL

Zu

86

μg/dL

EBV VCA-IgG Ab

<10

×

D-Bill

0.2

mg/dL

BTR

8.46

EBNA

40

×

AST

19

U/L

AFP

2.3

ng/mL

WBC: white blood cell, RBC: red blood cell, Ht: hematocrit, Plt: platelet, PT: protorombin time, INR: international normalized ratio, APTT: active partial thromboplastin time, FIB: fibrinogen, FDP: fibrinogen degradation products, AT III: antithrombin III, TP: total protein, Alb: albumin, T-Bill: total bilirubin, D-bill: direct bilirubin, AST: aspartate aminotransferase, ALT: alanine aminotransferase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, GTP: glutamyltransferase, ChE: cholinesterase, T-chol: total cholesterol, TG: triglyceride, UA: uric acid, BUN: blood urea nitrogen, Cr: creatinine, NH3: ammonia, Na: natrium, K: kalium, Cl: chloride, Fe: ferrum, UIBC: unsaturated iron binding capacity, Cu: copper, Zu: zinc, BTR: ratio of branched-chain amino acid to tyrosine, AFP: alfa-fetoprotein, M2BPGi: mac-2 binding protein glycan isomer, HBs-Ag: hepatitis B surface antigen, HBs-Ab: hepatitis B surface antibody, HBc-Ab: hepatitis B core antibody, HBe-Ag: hepatitis B e antigen, HBe-Ab: hepatitis B e antibody, HBV-DNA: hepatitis B virus deoxyribonucleic acid, PCR: polymerase chain reaction, HCV-Ab: hepatitis C virus antibody, CMV pp65: cytomegalovirus pp65 antigenemia, EBV VCA-IgM Ab: Epstein-Barr virus-viral capsid antigen immunoglobulin M antibody, EBV VCA-IgG Ab: Epstein-Barr virus viral capsid antigen immunoglobulin G antibody, EBNA: epstein-barr virus nuclear antigen

Liver ultrasonography with Shear wave measurement (Hitachi, Tokyo, Japan), an accurate ultrasonographic modality to evaluate liver fibrosis, showed no evidence of liver cirrhosis, according to Vs 1.39 m/s (33). Contrast computed tomography (CT) of the abdomen and pelvis revealed rectal varices, but ileocolonic varices were not detected. Digital subtraction angiography of the superior mesenteric artery showed delayed venous pooling in the ascending colon and terminal ileum. The blood flow was from the superior mesenteric artery to the superior mesenteric vein (Fig. 3A, B). CT during superior mesenteric angiography revealed the development of colonic varices in the ascending colon (Fig. 3C, D). There was no portal vein thrombosis detected on angiography. Considering the risk of bleeding after endoscopic treatment, we thoroughly informed the patient by explaining the possibility of bleeding and potential need for additional treatment, including endoscopic and operative management.

Figure 3.

A, B: Digital subtraction angiography of the superior mesenteric artery showed delayed venous pooling in the ascending colon. C, D: Computed tomography during superior mesenteric angiography showed the development of colonic varices in the ascending colon.

After obtaining his informed consent, we decided to perform endoscopic mucosal resection (EMR) because the JNET classification was type 2A, suggesting a low-grade intramucosal neoplasia (32). In addition, the tumor size was less than 20 mm, and we considered complete resection to be possible by EMR. During EMR, we shortened the time of exposure to electricity as much as possible in order to reduce the risk of electrocautery damage. After resection, we performed clip closure to help prevent post-EMR bleeding, and no immediate or delayed post-EMR bleeding was noted. A histopathological analysis revealed high-grade tubular adenoma of the colon. A colonoscopy six months after EMR revealed no change in the ileocolonic varices and no other lesions requiring excision.

Discussion

Varices are frequently associated with portal hypertension resulting from liver cirrhosis and other less-common causes. Colonic varices are rare, and it is reported that the incidence of colonic varices is 0.07% (2 of 2,912 cases) (34), whereas that of rectal varices is 3-7% in cirrhotic patients on endoscopy (35,36). Idiopathic ileocolonic varices with no underlying causes of portal hypertension are extremely uncommon. The most common cause of portal hypertension is liver cirrhosis. Other causes of portal hypertension, such as pre-hepatic abnormalities (e.g., portal or splenic vein thrombosis, malignancy, pancreatitis, postsurgical state), post-hepatic abnormalities (e.g., Budd-Chiari syndrome, congestive heart disease), or intrahepatic non-cirrhotic abnormalities (37,38), can induce colonic varices. Varices develop as venous collaterals when the portal pressure remains higher than the hepatic venous pressure, a condition called portal hypertension. Reopening of collapsed embryonic channels and reversal of the flow of existing adult veins is caused by portal hypertension, which can lead to the development of varices (39). The diagnosis of idiopathic colonic varices is made by detecting varices and excluding portal hypertension as a cause. In our case, laboratory testing revealed a history of hepatitis B virus (HBV) infection, but no other data suggested viral hepatitis, autoimmune disease, endocrine disease, or any other liver disease. Mac2 Binding Protein Glucosylation Isomer (M2BPGi), an accurate, reliable, and reproducible marker for the assessment of liver fibrosis (40), was within normal levels (Table 1). Shear wave measurement (Hitachi), an ultrasonographic modality for assessing liver cirrhosis (33), showed no evidence of liver cirrhosis. There was no portal or splenic vein thrombosis, according to contrast-enhanced CT and angiography of the abdomen and pelvis.

To our knowledge, there have been 29 reports on 33 patients with idiopathic colonic varices published since 1986 (Table 2) (2-30). On reviewing these cases, the average age was 44.8 years old, and 17 (77.3%) of the 22 patients were men (4,6-10,12,13,18,19,21,23,24,27,29). A family tendency was evaluated in 16 cases, and 6 cases had familial tendency (9,20,25). Occurrence site of varices was mentioned in 23 cases, and 6 cases (26.1%) were ileocolonic varices (6,16,21,22,27,28) and 12 (52.2%) were total colonic varices (2,3,5,17,19,20,23,24,29,30). The entire colon was involved in almost 80% of cases. A major complement was hematochezia in most cases (3,4,6-8,10,12,13,16-24,28,29), and some cases presented with diarrhea (20,24,26). Among the 19 cases of hematochezia, 7 (36.8%) required an operation as treatment (3,7,9,12,18,21,22). Endoscopic treatment of active bleeding from idiopathic ileocolonic varices has not been reported yet. There were only two asymptomatic cases (27,30), and both were detected by screening colonoscopy.

Table 2.

Characteristics of 33 Cases of Idiopathic Colonic Varices.

No.		Age		Sex		Location		Family tendency		Symptom		Treatment		Prognosis		Ref.
1		71		F		colonic		ND		asymptomatic		none		stable		[30]
2		76		M		colonic		ND		hematochezia		preserved		stable		[29]
3		37		M		ileocolonic		ND		hematochezia		preserved (ARB)		disappeared		[28]
4		54		M		ileocolonic		none		asymptomatic		none		stable		[27]
5		38		F		colonic		none		diarrhea		none		stable		[26]
6		ND		ND		ND		(+)		ND		ND		ND		[25]
7		ND		ND		ND		(+)		ND		ND		ND		[25]
8		ND		ND		ND		(+)		ND		ND		ND		[25]
9		30		M		colonic		none		hematochezia, diarrhea		preserved		stable		[24]
10		44		M		colonic		none		hematochezia		preserved		stable		[23]
11		20		F		ileocolonic		ND		hematochezia		operation		stable		[22]
12		21		M		ileocolonic		none		hematochezia		operation		stable		[21]
13		61		F		colonic		(+)		hematochezia, diarrhea		preserved		stable		[20]
14		43		M		colonic		none		hematochezia		preserved (propranolol)		decreased		[19]
15		24		M		rectum-descending		none		hematochezia		operation		stable		[18]
16		74		F		colonic		ND		hematochezia		preserved		stable		[17]
17		64		ND		ileocolonic		ND		hematochezia		operation		stable		[16]
18		ND		ND		ND		ND		ND		ND		ND		[15]
19		ND		ND		ND		ND		ND		ND		ND		[14]
20		30		M		hepatic flexure		ND		hematochezia		preserved		ND		[13]
21		37		M		sigmoid		ND		hematochezia		operation		stable		[12]
22		ND		ND		ND		ND		ND		ND		ND		[11]
23		27		M		ND		ND		hematochezia		preserved		stable		[10]
24		56		M		ND		(+)		ND		ND		ND		[9]
25		28		M		ND		(+)		ND		ND		ND		[9]
26		81		M		ND		ND		hematochezia		ND		ND		[8]
27		32		M		rectum-sigmoid		none		hematochezia		operation		stable		[7]
28		25		M		ileocolonic		ND		hematochezia		preserved		stable		[6]
29		ND		ND		colonic		ND		ND		ND		ND		[5]
30		58		M		rectum-sigmoid, caecum		none		hematochezia		preserved		stable		[4]
31		ND		ND		colonic		none		hematochezia		operation		stable		[3]
32		ND		ND		colonic		ND		ND		ND		ND		[2]
33		ND		ND		colonic		ND		ND		ND		ND		[2]

ND: Not detected, ARB: Angiotensin receptor blocker

Screening colonoscopy is an effective method of reducing the risk of death due to colon cancer (41), and its spread might be the reason for these recent asymptomatic cases. We detected colonic varices by conducting screening colonoscopy and established the diagnosis by EUS, digital subtraction angiography, and CT angiography. CT angiography might be the most reliable method for making a diagnosis, and digital subtraction angiography might be the only method for determining the blood flow in the varices. As we showed in our case, EUS or transrectal Doppler sonography have also been reported to be useful for making a diagnosis (18,19). Capsule endoscope may also be a reliable tool for investigating the location of varices (28). A barium enema can reportedly depict colonic varices as tumor-like serpiginous lesions (20). However, it is difficult to exclude polyps, cancer, or other colorectal lesions using a barium enema.

Colonic varices may be misinterpreted as polyps, cancer, air bubbles, or fecal material (2,20). There are cases of a biopsy or polypectomy resulting in a significant bleeding (20,31). In asymptomatic ileocolonic varices, as in our case, a careful examination using a variety of techniques is required in order to avoid significant complications of endoscopy.

The treatment of idiopathic colonic varices remains unclear. While most cases with hematochezia show no active bleeding during colonoscopy, others require emergency operation (16) or extensive colectomy (3,7,21,22). There have been no reports of endoscopic treatment for active bleeding caused by idiopathic ileocolonic varices. The short-term prognosis seems good, but the long-term prognosis is unclear. A family history of death from gastrointestinal hemorrhaging was reported in one case (20). Very few attempts have been made to treat colorectal lesions (e.g., polyps, carcinoma) accompanied by colonic varices. Depending on the bleeding status, an operation seems to be the safest method, but it is highly invasive. As mentioned above, there are cases of significant bleeding arising from a biopsy or a polypectomy (20,31). Such cases show that endoscopic treatment can cause significant complications related to colorectal varices. Generally, active lower gastrointestinal bleeding is first treated by endoscopic hemostasis therapy, including mechanical, thermal, and injection therapy or a combination thereof (42). The majority of bleeding events after a colonoscopy are managed successfully by endoscopy, whereas very few cases require operation (43). When colorectal lesions coexist with colonic varices and require treatment, an endoscopic approach should be considered from the point of invasiveness. In our case, an interventional approach to stop the bleeding was considered difficult because the blood flow of the colorectal varices was from the mesenteric artery to the mesenteric vein. Therefore, we planned endoscopic clipping or endoscopic injection sclerotherapy to manage the bleeding, and informed consent was obtained, including consent regarding the operative method. At the time of EMR, we shortened the electricity time as much as possible. We proceeded with clip closure in order to prevent post-EMR bleeding, and no immediate or delayed bleeding was noted.

Cold snare polypectomy has a lower risk of post-polypectomy bleeding than does hot snare polypectomy (44). This shows that a longer time of exposure to electricity might increase the risk of post-polypectomy bleeding. Prophylactic clipping is reportedly unnecessary for preventing post-polypectomy bleeding with polyps smaller than 2 cm in diameter (45). The polyp in our case was smaller than 2 cm, but we proceeded with clip closure to prevent post-EMR bleeding, as this case had coexistent colonic varices.

To our knowledge, this is the first report of idiopathic ileocolonic varices with a colon polyp treated successfully by endoscopy. The diagnosis of idiopathic colonic varices requires a careful examination using a variety of techniques. During management, the development of a treatment strategy and obtaining informed consent are required.

The authors state that they have no Conflict of Interest (COI).