Partial versus total portacaval shunt in alcoholic cirrhosis. Results of a prospective,randomized clinical trial.

OBJECTIVE: Results of the first prospective randomized clinical trial comparing partial and total portacaval shunt for variceal hemorrhage are reported. SUMMARY BACKGROUND DATA: Total portacaval shunts produce subnormal portal pressures, completely diverting hepatic portal flow. Partial shunts maintain higher pressures and preserve hepatopedal flow. No randomized trials of these two approaches have been performed. METHODS: Alcoholic patients with cirrhosis (n = 30) and variceal hemorrhage treated at one institution were randomized to receive partial (8-mm diameter portacaval H grafts with collateral ablation, n = 14) or total shunts (16-mm diameter grafts, n = 16). Portography was performed after operation and then yearly. Investigators blinded to shunt type assessed encephalopathy; hospitalizations were reviewed. RESULTS: Child''s class, age, and operative urgency were similar for the two groups. Two patients (with total shunts) died within 30 days. Hepatopedal flow was maintained in 13 partial and 0 total shunt patients (p < 0.0001). Shunt gradients were 16 +/- 5 compared with 6 +/- 3 cm saline after partial and total shunts (p < 0.0001). There were no shunt thromboses or variceal hemorrhages. Encephalopathy-free survival was significantly greater after partial shunts (p = 0.013; life table analysis). Five total compared with zero partial shunt patients required hospitalization for coma (p = 0.02). Long-term survival was not different for the two groups of patients. CONCLUSIONS: Partial shunts control variceal hemorrhage while maintaining hepatopedal flow and elevated portal pressures. By minimizing encephalopathy rates, partial shunts provide improved quality of survival compared with total shunts.     

Postprandial augmentation of portal hepatic inflow after prosthetic H-graft portacaval shunt.

Encephalopathy after portasystemic shunting generally occurs after eating. After partial portal decompression, encephalopathy is uncommon, presumably because of associated portal hemodynamics. However, after partial shunting, the changes in portal venous hemodynamics that occur with eating are unknown. With this in mind, 11 nonencephalopathic adults were studied more than 1 year after 8-mm H-graft portacaval shunt (PCS). The studies utilized color flow duplex ultrasound to determine the changes in portal vein (PV) and inferior vena cava blood flow that occur with eating a standardized meal. Following H-graft PCS, there is increased blood flow in the inferior vena cava after eating, particularly cephalad to the H-graft PCS, implying increased flow through the prosthetic shunt. Eating also increases hepatopedal blood flow in the PV distal to the H-graft PCS. Postprandial increases in hepatopedal portal blood flow may play an important role in avoiding encephalopathy after H-graft PCS.     

A systematic appraisal of portacaval H-graft diameters. Clinical and hemodynamic perspectives.

Over a period of 10 years, the authors have systematically reduced portacaval H-graft diameters. Their objective was to achieve partial shunting of portal flow without reversal of hepatic flow. This report summarizes their clinical and hemodynamic observations in 68 surviving patients with cirrhosis (mostly alcoholic) and variceal hemorrhage who underwent portacaval H-grafts ranging from 20 to 8 mm diameters. When shunt diameters were reduced to 10 and 8 mm and combined with aggressive portal collateral ablation, portal pressures increased significantly over larger H-grafts. Only 3% of patients with 20-12 mm H-grafts had prograde portal flow after operation, compared with 46 and 82% after 10 and 8 mm H-grafts, respectively (p less than 0.001). The incidence of encephalopathy diminished from 39% in the 20-12 mm H-graft group to 19 and 9% after 10 and 8 mm grafts, respectively (p less than 0.04). None of the patients with 10 or 8 mm PTFE grafts rebled from varices in the follow-up period (4-61 months). It is concluded that partial shunting of portal flow is hemodynamically feasible. It can be achieved in most patients using 8 mm polytetrafluoroethylene (PTFE) portacaval H-grafts combined with portal collateral ablation. Preserving prograde portal flow by partial shunting correlates with reduced encephalopathy rates after operation. Despite maintaining a relatively hypertensive portal system, partial shunts effectively prevent variceal hemorrhage.     

Portacaval H-graft: relationships of shunt diameter, portal flow patterns and encephalopathy

Small-diameter protacaval H-grafts, 10, 12, or 14 mm, were constructed in 29 cirrhotic patients with previous or active variceal hemorrhage. When 10 mm grafts were used in combination with portal collateral outflow ligation, varying degrees of prograde portal flow were maintained in 50% of the patients. When shunt size was greater, prograde flow was lost in more than 90%. The incidence of spontaneous postoperative encephalopathy was 11% in patients with prograde flow, compared with 50% in those with retrograde flow (p = 0.05). It is concluded that maintaining prograde portal flow after portacaval shunt is essential in minimizing postoperative encephalopathy. Prograde portal flow may be achieved in 50% of patients using 10 mm PTFE portacaval H-grafts combined with portal collateral ligation.     

Effect of portasystemic shunt operations on hepatic portal perfusion

We recently developed a radiocolloid technique for quantifying the fraction of superior mesenteric venous blood that bypasses liver sinusoids through extra- and intrahepatic collateral vessels. In the present investigation we applied this method, which is performed in conjunction with visceral angiography, to the assessment of patients with portal hypertension before and after surgical construction of portasystemic shunts. The mean corrected shunt index was 0.89 in 27 preoperative patients, and 48 percent of the patients had no evidence of sinusoidal perfusion by superior mesenteric venous blood (shunt index greater than 0.95). Sinusoidal perfusion was absent in five patients with residual hepatic portal flow by angiography, indicating that they had a high degree of intrahepatic shunting. Hepatic portal perfusion was preserved in 80 percent of patients after distal splenorenal shunt, and the corrected shunt index was significantly smaller after this procedure than after portacaval and interposition shunts. Three patients with no sinusoidal perfusion by superior mesenteric blood preoperatively had restoration of portal flow after distal splenorenal shunt. Five patients undergoing portacaval and interposition shunts had no evidence of portal sinusoidal perfusion by the radiocolloid technique either before or after the operative procedure.     

Mesenteric venous hypertension: importance after portal systemic shunting?

Hepatic dysfunction after portacaval shunting (PCS) has been attributed to loss of portal perfusion to the liver. Proponents of selective systemic shunting state that reduced encephalopathy and hepatic dysfunction with this procedure result from the maintenance of portal perfusion to the liver through the hypertensive mesenteric venous circulation. We questioned the importance of maintaining the diminished portal flow to the cirrhotic liver because hepatofugal flow is known to develop in many of these patients. We sought to further define mechanisms that may contribute to the maintenance of critical flow to the liver in compensated hepatic cirrhosis. We demonstrated a primary relationship between mesenteric venous hypertension (MVH) and increased hepatic arterial blood flow after diversion of portal blood flow. Fifteen dogs had vena caval stenosis above an end-to-side PCS to establish MVH and deprive the liver of portal blood flow. Another 15 dogs had end-to-side PCS alone. A half hour after shunting, hepatic arterial blood flow had increased significantly in all dogs. Hemodynamic parameters remained stable throughout. Six weeks later, mesenteric pressure increased 98% +/- 3% with intracaval stenosis (from 9.6 +/- 0.1 to 19.0 +/- 0.3 cm H2O). Mesenteric pressure was unchanged with PCS alone (9.0 +/- 0.1 cm H2O). Increased hepatic arterial flow was significantly elevated in all dogs above pre-shunt values by 6 weeks postshunt. With MVH, however, further augmentation in hepatic arterial flow was noted in the chronic state (1.5 +/- 0.1 vs 0.9 +/- 0.1 ml/min/gm, p less than 0.05). There was significant correlation between MVH and increased hepatic arterial flow in the chronic state (r = 0.79, p = 0.05). Hepatic arterial flow 6 weeks after PCS with MVH was associated with lower blood ammonia and improved hepatocellular function compared with animals with PCS alone. These results support the hypothesis that MVH is important in maintaining blood supply--beyond providing driving force for sustained portal flow to the liver. This is an important consideration in the medical and surgical management of portal hypertension, a condition in which profound reduction in portal pressure may negatively affect compensatory hepatic arterial blood flow.     

Ligation of huge spontaneous porto-systemic collaterals to avoid portal inflow steal in adult living donor liver transplantation: A case-report

IntroductionIn adult living donor liver transplantation (LDLT), maintenance of adequate portal inflow is essential for the graft regeneration. Portal inflow steal (PFS) may occur due to presence of huge spontaneous porto-systemic collaterals. A surgical procedure to increase the portal inflow is rarely necessary in adult LDLT.PresentationA 52 years male patient with end-stage liver disease due to chronic hepatitis C virus infection. Preoperative portography showed marked attenuated portal vein and its two main branches, patent tortuous splenic vein, multiple splenic hilar collaterals, and large lieno-renal collateral. He received a right hemi-liver graft from his nephew. Exploration revealed markedly cirrhotic liver, moderate splenomegaly with multiple collaterals and large lieno-renal collateral. Upon dissection of the hepato-duodenal ligament, a well-developed portal vein could be identified with a small mural thrombus.The recipient portal vein stump was anastomosed, in end to end fashion, to the graft portal vein. Doppler US showed reduced portal vein flow, so ligation of the huge lieno-renal collateral that allows steal of the portal inflow. After ligation of the lieno-renal collateral, improvement of the portal vein flow was observed in Doppler US.DiscussionThere is no accepted algorithm for managing spontaneous lieno-renal shunts before, during, or after liver transplantation, and evidence for efficacy of treatments remains limited. We report a case of surgical interruption of spontaneous huge porto-systemic collateral to prevent PFS during adult LDLT.ConclusionComplete interruption of large collateral vessels might be needed as a part of adult LDLT procedure to avoid devastating postoperative PFS.     

Angiotensin II increases portosystemic collateral resistance in portal hypertension

The decreased vasoconstrictive response of the splanchnic vasculature in portal hypertension (PHT) to angiotensin II (ANGII) is newly established. This could explain the limited ability of PHT patients to compensate in hypovolemic shock. However, the effect of ANGII upon portosystemic collateral resistance (Rc) is not known. We hypothesized that ANGII could directly effect Rc and thus change portal venous pressure (PVP). Chronic PHT was induced in New Zealand white rabbits by partial portal vein ligation 3 weeks prior to study. Splanchnic blood flow and portosystemic shunt (PSS) were measured simultaneously in six normal rabbits and then in six PHT rabbits at baseline and during intravenous ANGII infusion at 1.0 microgram/kg/min. Flow and resistance were standardized to 100 g small intestine weight. Superior mesenteric artery flow (Qsma) in normal rabbits was 64.9 +/- 3.6 ml/min, increased to 134.6 +/- 13.5 ml/min in the PHT animals (P less than 0.001) and was reduced 30% (P less than 0.05 vs PHT baseline) with ANGII. Concomitantly, PVP in the PHT animals was twice normal, 7.0 +/- 0.32 vs 14.4 +/- 0.55 mm Hg (P less than 0.001) and rose slightly with ANGII. The high PSS in PHT (84 +/- 6.0%, P less than 0.001 vs normal) was not affected significantly by ANGII infusion. Rc in the PHT rabbits rose 50% from 0.06 +/- 0.001 to 0.09 +/- 0.01 mm Hg/ml/min (P less than 0.001) with ANGII. This is the first evidence for a vasoconstrictive response in portosystemic collaterals during ANGII infusion. This change in collateral resistance causes both PVP and PSS to remain pathologically elevated in PHT despite a fall in portal inflow, thus predisposing to repeat variceal bleeding.     

Role of prostacyclin in the splanchnic hyperemia contributing to portal hypertension.

To determine the possible role of prostacyclin (PGI2) as a mediator of the splanchnic hyperemia seen with portal hypertension, the portal and mesenteric hemodynamics in normal and portal hypertensive rabbits were studied before and after cyclo-oxygenase blockade. Three weeks after partial portal vein ligation, splenic pulp pressure was elevated from 4.3 +/- 0.9 to 9.8 +/- 0.8 mmHg (p less than 0.01). Mesenteric blood flow increased from 77.0 +/- 4.7 ml.min-1.100 g-1 to 99.1 +/- 5.19 ml/min-1/100 g-1. Mesenteric vascular resistance fell from 0.82 +/- 0.6 mmHg/ml-1/min-1 to 0.49 +/- 0.07 mmHg/ml-1/min-1 (p less than 0.01). These hemodynamic changes were associated with a 27.3 +/- 0.2% rise in systemic arterial levels of PGI2 (p less than 0.01) and were substantially ameliorated by cyclo-oxygenase blockade with indomethacin. The effects of indomethacin blockade were reversed by exogenous PGI2. Moreover, in normotensive rabbits, infusion of PGI2 reproduced the splanchnic hyperemia and caused a very small but significant increase in portosystemic shunting. These findings support the previously proposed concept that splanchnic hyperemia may contribute to the maintenance of chronic portal hypertension. Furthermore, they suggest that this effect may be partially mediated by splanchnic PGI2 production.     

Distal splenorenal anastomosis in the treatment of digestive hemorrhage in portal hypertension

One hundred and ten distal splenorenal anastomoses (DSRA) were performed between 1976 and 1988 for gastrointestinal haemorrhages secondary to portal hypertension. Thirty five patients underwent DSRA (Warren's operation) followed by a mesenteric disconnection procedure. Eleven patients (10%) died (nine Child B and two Child A), including two after an emergency shunt. The mortality was influenced by age (p less than 0.01) and by the number of previous episodes of ictero-ascitic decompensation (p less than 0.02). A single anastomotic thrombosis was observed; the high flow rate of the anastomoses (mean: 1200 +/- 650 ml/min) explains the high patency rate (more than 90%). Eleven patients developed partial or total thrombosis of the portal trunk less than 6 weeks after the operation. The quality of hepatic perfusion was not significantly influenced by the mesenterico-splenic disconnection and this was omitted in 75 patients. The mesenterico-splenic collateral venous circulation was more developed in the absence of disconnection (p less than 0.05). The long-term actuarial survival was 83% at one year, 53% at five years, 47% at eight years and 28% at ten years. Survival was not influenced by the disconnection. Warren's operation is more difficult to perform than a porto-caval or mesenterico-caval shunt. DSRA appears to have three advantages: 1) a high flow rate as it is arterialised by the splenic artery, 2) hepatopetal portal perfusion maintained for several days, weeks or months, possibly reducing the risk of encephalopathy, 3) the absence of dissection of the hepatic pedicle preserves the possibilities of liver transplantation.     

Small-diameter prosthetic H-graft portacaval shunt: Definitive therapy for variceal bleeding

Partial portal decompression has become a popular option in the treatment of complicated portal hypertension. This study was undertaken to report long-term follow-up after partial portal decompression obtained utilizing 8 mm prosthetic H-graft portacaval shunts. A total of 110 consecutive patients underwent H-graft portacaval shunting through a protocol that detailed care and studies from 1988 to 1996. Prospective follow-up recorded efficacy of partial portal decompression, shunt patency, morbidity of shunting, and survival. Seventy males and 40 females, whose average age was 54 ±12.7 years (standard deviation), underwent shunting. Cirrhosis was due to alcohol abuse in 64%. Fourteen percent were in Child’s class A, 55% in Child’s class B, and 31% in Child’s class C. Shunts were undertaken as emergencies in 20%, urgently in 13%, and electively in 67%. Shunting decreased portal pressure in all patients (30 ±5.3 mm Hg to 19.9 ±5.5 mm Hg; P <0.001). Early and late thrombosis was 6.4% and 3.6%, respectively. Late rebleeding occurred in 5.4%. Perioperative (30-day) mortality was 11.8%, and was highest for patients in Child’s class C. Three-year survival was 53 %. Five-year survival was 41%. Partial portal decompression is achieved with H-graft portacaval shunting. Rebleeding, shunt occlusion, and encephalopathy are uncommon. In this series of unselected older patients with alcoholic cirrhosis, 5-year survival after H-graft portacaval shunting was greater than 40% with minimal intervention.     

The emergency portacaval H graft in alcoholic cirrhotic patients: influence of shunt diameter on clinical outcome

Emergency partial portal decompression was achieved with 8 or 10 mm portacaval H graft shunts combined with aggressive collateral ligation in 18 patients in whom bleeding esophageal varices could not be controlled medically. They were compared with 11 similar risk patients undergoing larger diameter portacaval H graft shunts (12 to 14 mm) for the same indications. Variables studied included 90 day operative mortality, hepatic encephalopathy rates, corrected portal pressure, and variceal re-bleeding. Operative mortality was similar in both groups and correlated strongly with Child's class. However, the incidence of portasystemic encephalopathy in survivors was significantly lower after partial decompression than after total decompression. No patient in either group rebled from varices. We conclude from our series of high risk alcoholic cirrhotic patients, that although mortality after partial and total portal decompression is similar, the lower incidence of encephalopathy in survivors suggests that partial decompression has advantages over total decompression when emergency control of variceal bleeding is necessary.     

Beneficial Effect of Partial Portal Decompression Using the Inferior Mesenteric Vein for Intractable Gastroesophageal Variceal Bleeding in Patients With Liver Cirrhosis

BACKGROUND: Use of the inferior mesenteric vein (IMV) for partial portal decompression has not been recommended as a first-line option for intractable gastroesophageal variceal bleeding because of the thin diameter of the vein. Although these indications remain relevant, few reports have compared partial portal decompression using the IMV with other therapies. We propose that partial portal decompression using the IMV is a useful alternative treatment for intractable variceal bleeding. METHODS: We performed partial portal decompression using the IMV in eight patients with intractable variceal bleeding that had been uncontrolled using medical and endoscopic therapies. All patients were classified into Child's class B or C. The surgical data, morbidity, and mortality were assessed. RESULTS: Mean portal venous pressure significantly decreased from 26.9 +/- 2.0 mmHg before the surgery to 19.8 +/- 3.9 mmHg after the surgery. The operative mortality rate was 0%. The mean duration of hospital stay was 25.5 +/- 13.3 days. Although one patient experienced recurrent bleeding, shunt patency was well maintained in all patients during the follow-up period (mean 28.9 +/- 14.1 months). Six patients are still alive and well without ascites or hepatic encephalopathy. Two of the Child's class C patients who underwent emergency shunt died owing to hepatic decompensation. CONCLUSION: Partial portal decompression using the IMV can be a safe, effective way to treat intractable variceal bleeding in patients with liver cirrhosis. However, use of the shunt procedure may have the most survival benefits for cirrhotic patients with preserved liver function.     

Graft interposition splenocaval shunt for total or selective decompression of portal hypertension

A new operation for selective or total decompression of the portal venous system in cases of intrahepatic portal hypertension is described. It involves interposition of a large-caliber Dacron graft between the splenic vein and the inferior vena cava. The graft-interposition splenocaval shunt is performed readily and quickly, satisfying the variable hemodynamic needs of patients with portal hypertension. It can be either selective (S-SCS) or total (T-SCS) from the beginning, or a T-SCS may be converted subsequently to a S-SCS should surgically induced hepatic decompensation supervene. It is less demanding technically than distal splenorenal shunt (D-SRS). The S-SCS conserves portal venous perfusion of the liver, preserves hepatocellular function and architecture at the preoperative levels, avoids precipitation of postshunt portal-systemic encephalopathy, and decompresses gastric-esophageal varices with prevention of further variceal bleeding even better than D-SRS. One hundred percent graft patency has been obtained, and the surgical results have been superior to those following portacaval shunt in patients with large liver blood flow and relative benignity of the liver disease, be it cirrhosis or noncirrhotic portal fibrosis. In patients with advanced cirrhosis, variceal bleeding, and small liver blood flows, T-SCS would be indicated. Patients of this category obtained inferior surgical results and had operative deaths (16.7%) following S-SCS. The concept of the operation has merits and deserves further evaluation.     

A selective approach to preexisting portal vein thrombosis in patients undergoing liver transplantation.

Splanchnic venous inflow is considered mandatory to ensure graft survival after liver transplantation. Over a 68-month period, we performed 570 liver transplants in 495 patients. Portal vein thrombosis was present in 16 patients. At transplant, the extent of the occlusion included portal vein alone (n = 4), portal including confluence of the splenic and superior mesenteric veins (n = 8), portal, splenic, and distal superior mesenteric veins (n = 2), and the entire portal vein, splenic vein, and superior mesenteric vein (n = 2). The operative approach included thrombectomy alone (n = 5), anastomosis at the confluence of the splenic and superior mesenteric splenic veins (n = 8), and extra-anatomic venous reconstruction (n = 3). The mean operative blood loss was 22 +/- 22 units, and the mean operative time was 9.7 +/- 4.8 hours. The 1-year actuarial survival rate was 81%, with a mean follow-up of 12.5 months. In summary, with a selective approach and the use of innovative forms of splanchnic venous inflow, portal vein thrombosis is no longer a contraindication to liver transplantation.     

Significance of selective arteriographic patterns in the celiac axis and superior mesenteric artery in portal hypertension

Selective celiac and superior mesenteric arteriographies were performed in patients with portal hypertension. An arterioarterial (A-A) shunt between the superior mesenteric artery and the celiac axis via pancreatic arcades was found in fifteen of forty-three patients with associated massive splenomegaly. A mild A-A shunt disappeared after portacaval anastomosis alone, whereas a prominent A-A shunt was reduced but persisted. The persisting A-A shunt disappeared after splenectomy. These findings led us to suggest that the paucity of the blood flow in the common hepatic artery concomitant with increased splenic arterial flow to the massively enlarged spleen may result in a compensatory supply to the liver from the superior mesenteric artery via the shunt.     

Predictability and maintenance of portal flow patterns after small-diameter portacaval H-grafts in man.

Patients undergoing small-diameter (8, 10, 12, and 14 mm) portacaval H-grafts were followed up to 3.5 years. Eight- and 10-mm grafts maintained prograde portal perfusion in 50% of the patients. Follow-up studies performed from 6 to 36 months after surgery show late shunt patency to be 97%. Recurrent variceal hemorrhage has not occurred in any patients. Direction of portal flow after a shunt was related to the size of the portal vein and the size of the shunt. If the shunt diameter was less than 50% that of the portal vein measured on the preoperative angiogram, portal flow was prograde. Encephalopathy rates remained significantly lower in patients with prograde flow after small diameter (8 and 10 mm) portacaval H-graft (p = .0.1). If thrombosis and encephalopathy rates remain low, the small-diameter, polytetrafluoroethylene portacaval H-graft is an attractive alternative to standard portacaval and mesocaval shunts.     

Noninvasive assessment of portosystemic shunting in extrahepatic portal hypertension in rats

A simple reproducible animal model of extrahepatic portal hypertension (EHPHT) has been developed in weanling Wistar rats using a two-stage ligation of the portal vein. This model consistently produces substantial collaterals, both portosystemic (hepatofugal) and portoportal (hepatopetal). Using dynamic hepatic scintigraphy (DHS) with 99mTechnetium sulphurcolloid, hepatopetal collateral flow was measured as the mesenteric fraction (MF) of total hepatic blood flow and compared with measurement of hepatofugal collateral flow (portosystemic shunting) following intraportal injection of radiolabeled microspheres. Strong and significant correlation between the two assessments was found with reduction in MF denoting increased portosystemic shunting (PSS). The technique of DHS has been used successfully in adults to assess compromised portal venous flow and is a simple noninvasive test to aid diagnosis, assessment, and follow-up of children with EHPHT.     

Radioisotopic splenoportography in patients with portal hypertension

Radio-isotopic splenoportography was performed by injecting 99mTcO4- into the spleens of 46 patients with portal hypertension and 14 patients with various disorders not having portal hypertension. No collateral circulation was demonstrated in the 14 patients without portal hypertension whereas some RI-images of portosystemic collaterals were found in 40 (87.0 per cent) of the 46 patients with portal hypertension. Collaterals were divided into an ascending group and a descending group, the appearance rate of ascending collaterals being 80.4 per cent and that of descending collaterals, 41.3 per cent. There were 3 image patterns in the ascending group, namely, an AZ-pattern in which the azygos vein was demonstrated; a SC-pattern in which the RI-bolus ascended along the esophagus to the neck and the subclavian vein; and an EG-pattern which showed stagnation of the RI-bolus in the esophagogastric region. There were 4 patterns in the descending group, namely; a pattern of gastro-renal caval shunt (GR-pattern); reverse flow patterns into the umbilical or paraumbilical veins (UV-pattern); into the superior mesenteric vein (SMV-pattern); and into the inferior mesenteric vein (IMV-pattern). The appearance of the EG-pattern was seen most frequently (74.4 per cent). The usefulness of this method for surveying the collateral circulation in portal hypertension, estimating the risk of esophageal variceal bleeding and evaluating its treatments, was suggested by the results of this study.     

Directing portal flow is essential for graft survival in auxiliary partial heterotopic liver transplantation in the dog.

BACKGROUND/PURPOSE: Auxiliary liver transplantation is an attractive alternative for orthotopic liver transplantation in patients with certain inborn errors of metabolism of the liver in which complete resection of the liver is unnecessary or even contraindicated. Because in these diseases portal hypertension is mostly absent, finding a balance in portal blood distribution between native liver and graft is complicated. The objective of this study was to investigate requirements for long-term (180 days) graft survival in auxiliary partial heterotopic liver transplantation (APHLT) in a dog model. METHODS: A metabolic defect was corrected in 26 dalmation dogs with a 60% beagle heterotopic auxiliary liver graft. Four groups of different portal inflow were studied. In the ligation group the portal vein to the host liver was ligated. In the split-flow group graft and host liver received separate portal inflow. In the banding group the distribution of the portal flow was regulated with an adjustable strapband and in the free-flow group the portal blood was allowed to flow randomly to host or graft liver. RESULTS: Metabolic correction increased in all groups after transplantation from 0.19 +/- 0.02 to 0.70 +/- 0.05 (P< .0001) but remained significantly better in the ligation and split-flow groups (graft survival, 135 +/- 27 and 144 +/- 31 days). In the banding group metabolic correction decreased significantly after 70 days, and although the grafts kept some function for 155 +/- 14 days, in 4 of 6 dogs portal thrombosis was found. In the free-flow group, competition for the portal blood led to reduced correction within 12 days and total loss of function in 96 +/- 14 days. Graft function also was assessed with technetium (Tc) 99m dimethyl-iminodiacetic acid uptake. A good linear association between HIDA uptake and metabolic correction was observed (r = 0.74; P < .0005). Grafts that contributed more than 15% to the total uptake of HIDA showed biochemical correction. This indicates a critical graft mass of about 15% to 20% of the hepatocyte volume to correct this metabolic defect. CONCLUSION: Auxiliary partial heterotopic liver transplantation can be a valuable alternative treatment for inborn errors of hepatic metabolism if the native liver and the graft receive separate portal blood inflow.