Infusion hepatic angiography was used, together with conventional angiogra-phy, for diagnosis in 68 patients with liver metastases. The combination of both techniques led to a diagnostic accuracy of 97%. Metastases were missed in only two patients, both of whom had underlying liver or biliary disease. In a comparison of the two techniques, the infusion study was found diagnostically essential in five patients (7%) and afforded improved diagnosis in 49 others (72%). In 10 patients (15%), it gave. equivalent information; and in four patients (6%) less information than the conventional technique. Infusion hepatic angiog-raphy is a useful complementary technique in anatomic liver diagnosis, especially in its ability to improve upon the diagnostic accuracy of the capillary phase of hepatic angiography. Canccr 38:2278-2286, 1976. NFUSION HEPATIC ANGIOGRAPHY WAS INTRO-I duced by Wise and Siber" and modified by Kaude et al.4 and Wirtanen." In contrast to conventional hepatic angiography, which involves a 3 to 5-second injection of contrast medium , the infusion study uses a protracted (10 to 15-second) injection, a lower injection rate, and larger total dose of contrast medium. Reportedly , infusion hepatic angiography gives less detail in the arterial phase of the hepatogram,3 but provides more diagnostic information durinig the parenchymal phase of the hepatogram, increasing its density and improving the visualization of focal hepatic lesions. 3,4911 O f 22 patients with liver metastases, Kaude et al. found the infusion technique diagnostically superior to conventional hepatic angiography in 14 patients and equivalent in eight patients. However, it has also been reported that the increased hepatic density achieved by infusion angiography can occasionally obscure hypervascular metastatic nodules. We have used both the conventional and infusion hepatic angiography in 68 patients with liver metastases. This report presents our experience and compares the diagnostic effectiveness of the two techniques. Of 68 patients, 36 were female and 32 male, aged from 20 to 81 years (mean 58 years). T h e diagnosis of liver metastases was established by surgery, biopsy or both. Sites and types of primary tumors are listed in Table 1. Studies were done by the transfemoral route in 65 patients and the transaxillary route in three. Preformed polyethylene (ID 1.2 mm, 011 2.2 mm) or Torcon (Cook, Inc.) (ID 1.09 mm, O D 2.3 mm) catheters were used. Selective ce-liac angiography was done first and in case of an accessory liver supply from the superior mesen-teric artery, it was followed by superior mesen-teric angiography. For catheterizing the hepatic artery, occasionally a selective single-curve cath-eter, but mostly a superselective double-curve catheter, was used.' It was placed in the common hepatic artery in 37 patients, the proper hepatic artery in 26 patients, and a major accessory hepatic artery-a branch of the superior mesenteric-in five patients. In all patients, conventional and infusion hepatic angiographic studies were done wit...