Effect of band ligation on portal hypertensive gastropathy and development of fundal varices.

BACKGROUND: Use of endoscopic therapies for esophageal varices has resulted in increased prevalence of fundal varices and severe portal hypertensive gastropathy. This study was meant to compare the effect of band ligation and sclerotherapy on development of fundal varices and portal hypertensive gastropathy. METHODS: Patients with esophageal varices presenting in the endoscopy unit of Shiakh Zayed Hospital, with at least one previous endoscopy were included. Patient's past record was reviewed for findings and type of treatment given for varices during first endoscopy, number of endoscopies till date, number of esophageal varices band ligation (EVBL) or sclerotherapy sessions. All patients underwent upper GI endoscopy and findings were recorded. Type of treatment patient rendered during first endoscopy either EVBL or sclerotherapy was correlated to the presence of fundal varices and severity of portal hypertensive gastropathy observed on present endoscopy, using Chi square test (chi2). RESULTS: Eighty one patients were included. Mean age of patients was 48.70 +/- 12.63. Esophageal varices band ligation was carried out during first endoscopy in 49 (60.5%) patients and sclerotherapy in 31 (38.2%) patients. On fresh endoscopy, fundal varices were seen in 25 (30.8%) patients. Severe portal hypertensive gastropathy was found in 26 (32.1%) and mild in 54 (66.7%) patients. Severity of portal hypertensive gastropathy and presence of fundal varices in recent endoscopy was significantly more in patients with EVBL in first endoscopy. CONCLUSION: Band ligation of esophageal varices is associated with more frequent development of fundal varices and worsening of portal hypertensive gastropathy compared to sclerotherapy.

Predictors of fatal outcome in fulminant hepatic failure.

OBJECTIVE: To determine prognostic indicators for morbidity and mortality in patients with fulminant hepatic failure. DESIGN: Observational study. PLACE AND DURATION OF STUDY: Department of Gastroenterology and Hepatology, Shaikh Zayed Postgraduate Medical Institute, Lahore, from April 2003 to June 2005. PATIENTS AND METHODS: Forty five patients with confirmed diagnosis of fulminant hepatic failure were included in this study. They were investigated for acute viral serology, complete blood count, liver function tests, renal function tests, serum electrolytes, arterial blood gases, initially daily and then on alternate days. Patients were followed in high dependency/ICU at the Shaikh Zayed Postgraduate Medical Institute. Data were analyzed by logistic regression analysis to identify predictors of mortality and morbidity. RESULTS: Of the 45 subjects, 31 were males and 14 females. Twenty four patients were positive for acute hepatitis E. Twenty patients recovered fully, whereas 25 died with mortality rate of 55.6 %. Variables i.e. presence of non-A non-B viral hepatitis, serum creatinine > 2.0 mg/dl, serum phosphate > 3.5 mg/dl and sepsis were found to have significant association with mortality on linear correlation. Only serum creatinine more than 2.0 mg/dl, serum phosphate > 3.5 mg/dl and development of sepsis were found to predict the outcome after multivariate analysis. CONCLUSION: Patients with serum creatinine > 2.0 mg/dl, serum phosphate > 3.5 mg/dl and sepsis showed adverse prognosis for survival.

Bedside diagnosis of spontaneous bacterial peritonitis using reagent strips.

OBJECTIVE: To determine the efficacy of reagent strip for bedside diagnosis of spontaneous bacterial peritonitis (SBP). DESIGN: Cross-sectional analytical study. PLACE AND DURATION OF STUDY: Shaikh Zayed Postgraduate Medical Institute from November 2003 to August 2004. PATIENTS AND METHODS: Patients with cirrhosis and ascites underwent diagnostic paracentesis. Fluid was checked for leukocyte esterase released by PMN by using Combur 10 urine strip and graded for color change from 0-3. Fluid was also analyzed by cytology for PMN count. Results of both methods were compared to determine sensitivity, specificity and accuracy of strip for diagnosis of SBP. RESULTS: Of 214 paracentesis performed, SBP was diagnosed in 38 patients whereas 176 were negative for infection. Strip test was 97.7% sensitive and 89.4% specific with positive predictive value of 90%, negative predictive value of 97.7% and accuracy of 96.2%, when reagent strip grade 3 was considered as positive for diagnosis. CONCLUSION: Reagent strip is a quick bedside test, highly sensitive and specific for the diagnosis of SBP, based on polymorphonuclear count in ascitic fluid, to initiate early treatment, thus improving patient's outcome.

Non-endoscopic prediction of presence of esophageal varices in cirrhosis.

OBJECTIVE: To identify hematological, biochemical and ultrasonographic predictors of esophageal varices in patients of cirrhosis. DESIGN: Cross-sectional, analytical study. PLACE AND DURATION OF STUDY: Department of Gastroenterology, Shaikh Zayed Postgraduate Medical Institute, Lahore, from September 2003 to March 2004. PATIENTS AND METHODS: One hundred and one patients with established cirrhosis and no history of variceal bleed underwent physical examination, hematological, biochemical tests and abdominal ultrasound examination. Esophagogastroduodenoscopy (EGD) was carried out in all patients. Presence of varices on EGD was correlated with hematological, biochemical and ultrasonographic variables by regression analysis. RESULTS: Esophageal varices were seen in 65 patients while 36 patients had no varices. High grade varices were seen in 15 patients and 50 patients had low grade varices. Serum albumin less than 2.95 g/dl, platelet count less than 88 x 103/muL and portal vein diameter more than 11 mm were associated with presence of varices. High grade varices were predicted by serum albumin < 2.95 g/dl and portal vein diameter more than 11 mm. CONCLUSION: Patients with serum albumin < 2.95 g/dl, platelet count < 88 x 103/muL and portal vein diameter > 11 mm are more likely to have high grade varices. These patients are candidates for surveillance endoscopy.