Purpose : To evaluate the indication for intracavitary Urokinase(UK) in the treatment of loculated pleural effusion. Materials and Methods : We analyzed CT and US in 31 patients who were treated with intracavitary UK inloculated pleural effusion. in each patient, a single chest catheter (10-12 F) was insected under imaging guidance. When the amount of drainage was less than 100 ml/day, UK was instilled through the catheter until less than 50 ml/day was drained. On follow-up chest radiographs of more than 1 month, we classified the results of treatment into 3 groups: (1) completely effective (lung expansion, over 80%); (2) partially effective (20-80%);(3) ineffective (bellow 20%) group. Sonographic pattern of pleural fluid was classified into anechoic, linear septated, and honeycomb appearances and the thickness of parietal pleura was measured on CT. Results : Sixteen patients were completely effective, nine were partially effective, and six were ineffective group. All patients with completely or partially effective outcome had anechoic and linear septated appearance on US and had less than 4 mm of parietal pleural thickness on CT. Of six ineffective patients, US showed linear septated in one patient and honeycomb appearance in five patients and the thickness of parietal pleura on CT was 3 mm in one patient, 4 mm in two patients, 5 mm in one patient, and 6 mm in two patients. Conclusion : UK instillation through percutaneous catheter was an effective method in the treatment of loculated pleural effusion. However, we found near complete reaccumulation of pleural fluid when honeycomb appearance of pleural fluid on US or more than 5 mm parietal pleural thickness on CT was observed, which might suggest that we should consider the other kinds of treatment method in those patients.