LOW PLATELET COUNT AS AN EARLY INDEX FOR OESOPHAGEAL VARICES IN PATIENTS WITH PORTAL HYPERTENSION

Objective: To determine if low platelet count can be used as an index for detection of esophageal varices in patientswith portal hypertension.Material and Methods: This cross sectional analytic study was conducted in Bilal Medical Trust Hospital, Pir Baba andKuwait Teaching Hospital, Peshawar, Pakistan from January 2015 to December 2015. After approval from the hospitalethical and research committee the study was conducted and it included patients recruited through non probabilityconsecutive sampling, presenting to OPD. A detailed history and physical examination was done after which the liverwas thoroughly assessed by a detailed ultrasound of the abdomen. The liver size, echo pattern of established signs ofcirrhosis , uneven hepatic margins, increased parenchymal reflectivity, coarseness, increased echo graphic contrastbetween right lobe of liver and right kidney, hypertrophied caudate lobe, and attenuated hepatic veins were assessedto diagnose cirrhosis. Hepatic focal lesions and portal vein patency and diameter along with splenic size, Portal vein,hepatic artery, and splenic artery flow and patency were noted. Full blood count, ALT, hepatitis Band C serology wasdone at the time of admission. A screening endoscopy was undertaken next to assess the presence or absence of thevarices. The varices after detection was graded according to the Paquet grading system.Results: Among the total 123 patients examined 60% were females and 40% were males. Minimum age was 30 andmaximum age was 70. Average age was 51. Among different age groups 4 (3.2%) belonged to age group 30 to 40years, 33(26.2%) belonged to age group 40 to 50 years, 47 (38.2%) aged 50 to 60 years and 40(32.5%) belonged to60 to 70 years age group. The serology done for hepatitis revealed hepatitis C in all of the cases as an etiology 123(100%). The platelet count was measured next and the data further stratified accordingly. About 8(6.5%) had a plateletcount between 30-50,000. Around 18(14.6%) had a platelet count in the range of 50-70,000. Of all patients maximum33(26.8%) had the platelet count in the range of 70-90,000. Also another 31(25.2%) had the same counts in the range of90-110,000. In about 20(16.2%) the platelet counts were in the range of 110-130,000. Only 13(10.5%) had platelet countsin the range of 130-150,000. Next the data was analyzed for the esophageal varices. Again the data was subdividedin to four groups according to the presence and grades of varices. A total of 32(26%) had no varices on endoscopy.About 31(37%) had Grade 1 varices, whereas 14(12%) had Grade 2 varices. An alarming 31(38%) had Grade 3 variceson endoscopy. After running correlation the p value for the variables varices and platelet count was 0.098 (significantat 10%) and as such showed no clear and consistent relationship between the two variables. However on further extrapolatingthe study it was found that gender (females) showed a partial relationship (Table 1) between falling plateletcounts and higher grades of varices (p value 0.02).Conclusion: Low platelet count alone cannot be used as an early index for the presence of esophageal varices inpatients with portal hypertension.