Splenic artery aneurysms (SAAs), occurring in 7% to 17% of patients with cirrhosis, often result in catastrophic rupture after liver transplantation. We had experienced 3 cases of ruptured SAAs after adult living donor liver transplantation (LDLT), and we then performed this study to find risk factors for coexisting SAAs in liver transplant candidates with cirrhosis and to propose ideal approaches for them. Preoperative and postoperative computed tomography angiograms and axial views were reviewed for 310 adult LDLT recipients who had cirrhosis from January 2004 to August 2005. The recorded variables were the preoperative diagnosis, the presence of SAA and its characteristics, the splenic artery (SA) diameter, and the presence and size of portosystemic collaterals. Devastating SAA rupture accompanied by hypovolemic shock occurred on postoperative days 6, 82, and 8, respectively, and it was treated emergently by embolization in cases 1 and 2 and by splenectomy in case 3. Cases 1 and 3 recovered well, but case 2 died of an unrelated cause with a long hospital stay. The incidence of SAA during the study period was 14.2% (44/310), and the size was 16.6 +/- 5.7 mm. Most SAAs were single (70.6%, 31/44) and were located in the distal one-third of the SA (82.4%, 36/44). Large portosystemic collaterals demonstrating longstanding severe portal hypertension were significantly correlated with the occurrence of SAAs. Nine patients with SAAs were preventively treated by proximal ligation (n = 4) intraoperatively and by embolization (n = 5) 1 day before or after LDLT. No patient showed severe postembolization syndrome. In conclusion, a careful preoperative evaluation of SAAs by high-resolution 3-dimensional computed tomography in liver transplant candidates, especially in those showing large portosystemic collaterals, is merited. Preventive treatment should be encouraged regardless of the size in order to avoid severe morbidity and mortality related to SAA rupture, and methods such as radiological and surgical interventions need to be individualized according to the location and number of SAAs.