S. Ganguly et al., THE PREVALENCE AND SPECTRUM OF COLONIC LESIONS IN PATIENTS WITH CIRRHOTIC AND NONCIRRHOTIC PORTAL-HYPERTENSION, Hepatology, 21(5), 1995, pp. 1226-1231
Portal hypertension diffusely affects the gastrointestinal tract. The
frequency and profile of distinct colonic mucosal lesions (portal colo
pathy) and rectal varices (RV; veins >4 cm above the anal verge) is no
t well studied. Fifty consecutive patients with portal hypertension (2
5 with cirrhosis, 10 with noncirrhotic portal fibrosis [NCPF], and 15
with extrahepatic portal vein obstruction [EHPVO]) were assessed clini
cally and by upper and lower gastrointestinal (GI) endoscopy. Colorect
al lesions were seen in 35 (70%) patients, significantly more often in
bleeders than in nonbleeders. Rectal varices were detected in 22 (44%
) patients; larger and more often seen in EHPVO (80%) than in cirrhosi
s (28%) and NCPF (308) (P < .01) patients. Portal colopathy was seen i
n 26 (52%) patients, with nearly similar frequency in cirrhotics, NCPF
, and EHPVO patients. Previous sclerotherapy or presence of gastric va
rices had little influence on the development of these lesions. An ass
ociation (P < .01) was, however, seen between the presence of colopath
y and portal gastropathy. Overt bleeding was seen in 8% and 4% of pati
ents with RV and colopathy, respectively. In conclusion, our results d
emonstrate that colorectal lesions are present in about two thirds of
patients with portal hypertension. Patients with portal hypertension a
nd lower GI bleeding should be colonoscoped. Patients with extrahepati
c portal vein obstruction may in turn benefit from baseline sigmoidosc
opic examination to define the presence and size of rectal varices.