To determine if the venous drainage of the spleen into the portal circulation is essential for its ability to protect against encapsulated bacterial challenge. Three groups of dogs were randomly assigned to undergo either sham laparotomy, splenectomy, or splenectomy with autotransplantation of the intact spleen into the pelvis and formation of vascular anastomoses to the iliac vessels. Two weeks postoperatively, the dogs received a sublethal intravenous injection of Streptococcus pneumoniae type 25. Bacterial clearances and inflammatory damage to the liver. Bacterial clearance of the autotransplanted spleen should be no different from that of a sham-operation spleen and significantly different from that of a splenectomized animal. In addition, immunologic function of the autotransplanted spleen should not differ from that of a sham-operation spleen in the degree of inflammatory damage to the liver. No differences in bacterial clearance function were found between the animals that had undergone sham laparotomy or splenic autotransplantation. However, bacterial clearance in the splenectomized animals was significantly impaired. Histologic examination of the liver 2 weeks after the bacterial challenge revealed high-grade inflammatory damage to the livers of splenectomized dogs, intermediate liver damage in dogs that underwent autotransplantation, and essentially no damage in dogs that underwent sham laparotomy. Autotransplanted spleens were essentially nonreactive, lacking actively proliferating germinal centers, whereas splenic tissue from sham-operation animals showed reactivity. Although bacterial clearance function is unchanged in autotransplanted spleens, this method still does not fully protect the liver from inflammatory damage. © 1995 American Medical Association. All rights reserved.