HEMATOCRIT AS AN EARLY PREDICTOR OF SEVERE ACUTE PANCREATITIS: A PROSPECTIVE OBSERVATIONAL STUDY IN TERTIARY CARE CENTRE OF NEPAL

Abstract

Background: The predictor system for acute severe pancreatitis is complex and takes up to 48 hours to complete. This study aimed to investigate the role of admission hematocrit in predicting acute pancreatitis severity. Methods: This was an observational study, conducted at Chitwan Medical College where 112 patients who met the inclusion criteria and were diagnosed with acute pancreatitis were examined. The patients were followed for 48 hours by evaluating the admission hematocrit and were categorized to have mild to moderately severe pancreatitis or severe pancreatitis according to the modified Atlanta criteria. All analyzes were performed using SPSS statistical software, version 26. Descriptive and inferential statics, sensitivity and specificity were calculated. Results: The mean age was 44.72 ± 14.5 years and 70.5% were male. The most common etiology was ethanol (52.2%), followed by gallstone disease (26.5%). The prevalence of severe pancreatitis in patients in the high-risk group was 88.89%, and the incidence of severe pancreatitis in patients in the low-risk group was 10.59%. The detection sensitivity of hematocrit level >/= 44% for severe acute pancreatitis was 72.73%, the specificity of 96.2%, the positive predictive value of 88.89%, the negative predictive value of 89.41%, the accuracy of 89.3%, likelihood ratio for positive test result of 19.14 and likelihood ratio for negative test result of 0.283. Conclusions: Patients with high hematocrit on admission were associated with the occurrence of severe acute pancreatitis, so these patients should be considered as a high-risk group for development of severe acute pancreatitis and monitored meticulously.