OBJECTIVES: Choosing the best treatment option at the optimal point of time for patients with walled-off necrosis (WON) is crucial. We aimed to identify imaging parameters and clinical findings predicting the need of necrosectomy in patients with WON. METHODS: All patients with endoscopically diagnosed WON and pseudocyst were retrospectively identified. Post hoc analysis of pre-interventional contrast-enhanced computed tomography was performed for factors predicting the need of necrosectomy. RESULTS: Sixty-five patients were included in this study. Forty patients (61.5%) were diagnosed with pseudocyst and 25 patients (38.5%) with WON. Patients with WON mostly had acute pancreatitis with biliary cause compared with more chronic pancreatitis and toxic cause in pseudocyst group (P = 0.002 and P = 0.004, respectively). Logistic regression revealed diabetes as a risk factor for WON. Computed tomography scans revealed 4.62% (n = 3) patients as false positive and 24.6% (n = 16) as false negative findings for WON. Reduced perfusion and detection of solid findings were independent risk factors for WON. CONCLUSIONS: Computed tomography scans are of low diagnostic yield when needed to predict treatment of patients with pancreatic cysts. Reduced pancreatic perfusion and solid findings seem to be a risk factor for WON, whereas patients with diabetes seem to be at higher risk of developing WON.
OBJECTIVES: Choosing the best treatment option at the optimal point of time for patients with walled-off necrosis (WON) is crucial. We aimed to identify imaging parameters and clinical findings predicting the need of necrosectomy in patients with WON. METHODS: All patients with endoscopically diagnosed WON and pseudocyst were retrospectively identified. Post hoc analysis of pre-interventional contrast-enhanced computed tomography was performed for factors predicting the need of necrosectomy. RESULTS: Sixty-five patients were included in this study. Forty patients (61.5%) were diagnosed with pseudocyst and 25 patients (38.5%) with WON. Patients with WON mostly had acute pancreatitis with biliary cause compared with more chronic pancreatitis and toxic cause in pseudocyst group (P = 0.002 and P = 0.004, respectively). Logistic regression revealed diabetes as a risk factor for WON. Computed tomography scans revealed 4.62% (n = 3) patients as false positive and 24.6% (n = 16) as false negative findings for WON. Reduced perfusion and detection of solid findings were independent risk factors for WON. CONCLUSIONS: Computed tomography scans are of low diagnostic yield when needed to predict treatment of patients with pancreatic cysts. Reduced pancreatic perfusion and solid findings seem to be a risk factor for WON, whereas patients with diabetes seem to be at higher risk of developing WON.