Profile of gastric varices among Sri Lankan cirrhotics

Abstract

INTRODUCTION: Gastric varices (GV) can result in life threatening bleeding. There have been no studies on the characteristics of GV among Sri Lankan cirrhotics. Aims: To perform a descriptive analysis of GV among a Sri Lankan cirrhotic population METHODS: Medical records of all upper gastrointestinal endoscopies performed on cirrhotics at presentation, in the University Endoscopy Unit, Colombo North Teaching Hospital, Ragama, from 2006 to 2011 were screened. Demographics, indication and findings of endoscopy were analysed. RESULTS: 628/641(98%] of the cirrhotic population screened had a complete data set for analysis. 70/628(11%) patients had GV[ma!e-61(87%), female-9(13%); mean age 55(SD = +/- 10.7, range 23-81 years]]. 48/70(69%] had oesophageal varices (Gastro Oesophageal Varices (GOVl-18/48(38%), GOV2- 30/48(62%] in addition to GV and 22/70(31%] had Isolated GV (JGV1 -10(45%], IGV2 -12(55%]]. 38/70(54%] had portal hypertensive gastropathy and 3/70(4%] had gastric antral vascular ectasia in addition to GV. 51/70 (73%] were detected on initial screening of cirrhotics, while 19/70(27%][6/22(27%] with IGV] were detected at presentation with upper gastrointestinal bleed (UGIB]. 288/628(46%] had oesophageal varices (EV). 240/288(83%] had only EV without GV. 163/240(68%) were detected on initial screening of cirrhotics, while 77/240(32%) were detected at presentation with UGIB. There was no statistical significance between proportion of patients detected to have isolated EV or IGV at presentation with UGIB (p=0.64; ?2=0.2) CONCLUSIONS: The profile of GV among our patients is comparable to what is previously described elsewhere. Findings suggest in cirrhotic patients presenting with UGIB searching for presence of GV, by retroflexion at endoscopy, is as important as identifying EV.