Nonocclusive hepatic artery hypoperfusion syndrome (splenic steal syndrome) in liver transplant recipients

Abstract

There are numerous causes of reduced arterial inline flow to the liver transplant despite a patent hepatic artery. These include causes of increased peripheral resistance in the hepatic arterial bed, siphoning of the hepatic arterial flow by a dominant splenic artery (splenic steal syndrome), functional reduction of hepatic arterial flow in response to hyperdynamic portal inline flow, and small hepatic graft relative to normal portal inline flow (relative increase of portal flow). These causes are incompletely understood, and perhaps the most controversial of all is the splenic steal syndrome, which is possibly an underrecognized cause of graft ischemia in the United States. Splenic steal syndrome presents nonspecifically as graft dysfunction; if overlooked, it may lead to graft failure. Its incidence is reported to be 0.6 to 10.1% in liver transplant recipients, with some institutions performing prophylactic and/or posttransplant treatment procedures in up to a quarter of their transplant recipients. This wide disparity in the incidence of the diagnosis is probably because there are no objective diagnostic imaging criteria. This article presents a review of the literature that addresses the differential diagnostic considerations of hepatic artery hypoperfusion (splenic steal syndrome included) in the absence of an anatomical defect (hepatic artery stenosis, thrombosis, and/or kinks).

Keywords: buffer response; hypoperfusion; liver; splenic steal; transplant; vascular complications.

Figure 1

Middle-age woman status post–liver transplantation 3 months previously, presenting with abnormal liver function tests (dysfunctional graft). Liver biopsy was suggestive of ischemia. (A) Selective image from a celiac angiogram of the liver transplant recipient demonstrating delayed arterial flow (open arrow) in the hepatic artery with no peripheral filling of the hepatic arterial distribution at this stage of the angiogram. In contrast, the splenic artery is filling peripherally into the spleen (Sp). (B) Selective image from a celiac angiogram of the same liver transplant recipient after balloon-test occlusion (asterisk) of the proximal splenic artery. This is the same catheter positioning, injection rate and volume, and frame within the angiogram as (A). The distal splenic artery (open arrows) is filling distally via reconstitution via the left gastroepiploic (LGA-PA) and left gastric arteries (LGAs) and continues to supply the spleen (Sp). In contrast with (A), obtained before the temporary splenic artery occlusion, the peripheral hepatic arterial branches are now filling in normal fashion.

Figure 2

Schematic of the splanchnic circulation demonstrating the full circle between the celiac axis (CAx), the spleen, the splenic vein, the portal vein, and the liver. To complete the circle is the mediation of the portal venous inflow with the hepatic arterial inflow of the liver by the hepatic artery buffer response (HABR).

Figure 3

Schematic of the splanchnic circulation showing the hemodynamic balances between the splenic peripheral vascular resistance (splenic vasc. bed) and the hepatic peripheral vascular resistance (hepatic vasc. bed) on the arterial side of the splanchnic circulation. On the portovenous side of the splanchnic circulation, there is a balance between the graft size (and graft quality) and the portal venous flow itself. Mediation between the arterial (top half of schematic) and the venous (bottom half of schematic) circulation is by the hepatic arterial buffer response (HABR), which is a partly reciprocal relationship between the hepatic arterial and hepatic portovenous inflow (Fig. 4).

Figure 4

Simplified schematic of the hepatic arterial buffer response (HABR), which is a partly reciprocal relationship between the hepatic arterial and hepatic portovenous inflow. This relationship is incompletely understood in both native livers and after liver transplantation. The HABR is speculated to be altered in the relatively denervated transplant hepatic graft, but in what way and to what extent, if any, is unknown. At baseline (normal, top row), ~30% of the hepatic arterial inline flow shunts to the portal circulation. The middle rows shows the largely agreed-upon arterial and portal responses. The response specific to nonocclusive hepatic artery hypoperfusion syndrome or splenic steal syndrome is an increase in inline portal venous flow (hyperdynamic portal circulation), causing a hepatic arterial response with a concomitant reduction of inline arterial flow and a reduction of the hepatic artery-to-portal shunting (more than the normally expected 30% of hepatic arterial flow). The HABR may be mediated by adenosine washout. Adenosine, a known vasodilator, accumulates in the liver. When inline portal venous flow increases, it washes out the adenosine and its vasodilator effect, leading to vasoconstriction, an increase in peripheral arterial resistance, and diminished hepatic arterial flow.