001)).Of the patients who had lower urine-NGAL on day1(n=14), none had persistent AKI and 3 had transient AKI. Day1 urine-NGAL had a high probability to predict persistent AKI as well as mortality. At a cut off of 221ng/ml, urine-NGAL had a sensitivity and specificity of 65% in predicting severe pancreatitis (AUC=0.755,p=0.013). NGAL levels were significantly more than the controls

both in serum and urine on both days. Conclusion: NGAL(both serum and urinary) predicts AKI in acute pancreatitis. Day1 urine-NGAL can be used to predict AKI, both its occurrence and persistence and can be used to monitor renal failure in patients with SAP. Key Word(s): 1. pancreatitis; 2. NGAL; 3. AKI; 4. severity; Presenting Author: RAGESHBABU THANDASSERY Additional Authors: USHA DUTTA, SREEKANTH APPASANI, RAGHAVENDRA Small Molecule Compound Library PRASADA, THAKURDEEN YADAV, KARTAR SINGH, NIKHIL CHAUDHARY, AJAY BAHL, RAKESH KOCHHAR Corresponding Author: RAGESHBABU THANDASSERY Affiliations: EX 527 in vivo Department of Gastroenterology, PGIMER; Department of Cardiology, PGIMER Objective: Cardiovascular failure occurs in

one third of patients with severe acute pancreatitis (SAP). There is paucity of information on underlying mechanisms contributing to cardiovascular failure, the spectrum of cardiovascular dysfunction and its impact on outcome. AimTo study the occurrence of electrocardiography from changes (ECG), cardiac dysfunction (CD) in SAP, characterize the type of CD (systolic, diastolic or combined) and describe its impact on final outcome.

Methods: 72 patients with SAP and hypotension were prospectively studied for the occurrence of CD, nature of CD and its influence on hospital course and mortality. All patients had 12 lead ECG recorded on day 1, day 3 and day 7 in addition to the continuous ECG monitoring during ICU stay. Cardiac enzymes (Troponin I and Creatine phospho kinase MB) were measured at day 1 and day 3. All the patients underwent trans-thoracic echocardiography examination on day1 of hospitalization. Results: Of 72 patients (mean age of 39.1±12.9, 44 males); 58 (80.6%) had transient and 14 (19.4%) persistent hypotension. 47 (65.3%) patients had CD and of them 28 (59.6%) had diastolic dysfunction (DD), 8 (17%) had systolic dysfunction and 11(23.4%) had combined CD. Abnormal ECG findings were noted in 58 (80.6%) patients and were mostly ST segment and T wave changes. 10 (13.9%) patients had percutaneous drain placement, 12 (16.7%) underwent surgery and 14 (19.4%) succumbed to illness. Univariate analysis showed that patients with diastolic dysfunction had significantly higher mortality (hazard ratio- 3.6, 1.0, 12.5 and p value 0.032). Multivariate analysis showed APACHE II (odds ratio 20.60, CI=3.31-128.26, p=0.001) (odds ratio 7.2, CI=6.1-8.1, p<0.001) as independent predictors of mortality.