Management of ascites with hydrothorax

Hydrothorax occurs in 5.3 percent of ascitic patients. Our experience with 22 cases forms the basis of this report. Of the 22 cases, 21 were spontaneous and 1 was due to transdiaphragmatic incision. Eighteen occurred on the right side. Usually fluid enters the chest through tiny defects in the diaphragm. These defects are often covered by pleuroperitoneum, but the high abdominal pressure raises a bleb on the superior surface of the diaphragm. Rupture produces hydrothorax. The ascites is often relieved with the onset of the hydrothorax. Blockage of the thoracic duct has produced chylous ascites. The thoracoabdominal communication is immediately confirmed by a scan of the chest and abdomen after intraperitoneal injection of technetium-99 colloid. Fluid is tapped from the chest immediately before intraperitoneal injection. The rate at which the technetium-99 enters the chest is related to the size of the defect in the diaphragm. A significant transfer should occur within 12 hours. Immediate transfer occurs with large defects. The ruptured blister on the diaphragm forms a one-way valve. Intrathoracic injection does not migrate into the peritoneal cavity. The valvular characteristics of the leak force ascitic fluid into the thorax because the differential pressure between the abdominal and pleural cavities is intensified by inspiration. If tension hydrothorax has occurred, urgent thoracocentesis and paracentesis may be required. A chest tube should not be introduced. The main principle of surgery is to supply a low resistance pathway for the return of fluid to the venous system and to eliminate the diaphragmatic defect by obliteration of the pleural space. A LeVeen peritoneovenous shunt is performed after emptying the abdomen of its fluid load. After completion of the shunt operation, the chest is emptied of fluid, and a sclerosing agent (tetracycline or nitrogen mustard) is injected into the pleural cavity. Closure of the defect is verified by technetium-99 labeled scans which also confirm shunt patency. With this regime, the defect closed or was rendered insignificant in 18 of 22 patients. One patient had a post-transdiaphragmatic surgical defect which was too extensive to be closed by the aforementioned procedures. One patient remained well but did not have closure of the defect, one patient with a ruptured hiatial hernia did not have closure, and one patient who had previous placement of a chest tube could not be closed. Therefore, 18 of 22 patients were su