TO THE EDITORS:
Here we present a living donor liver transplant with simultaneously performed distal splenorenal shunting and splenic artery ligation. The recipient was an 18-year-old, 83.5-kg male with autoimmune hepatitis and end-stage liver disease who had a Model for End-Stage Liver Disease score of 23 (total bilirubin level = 5.8 mg/dL, creatinine level = 0.6 mg/dL, international normalized ratio = 2.51). The patient's brother, a healthy 24-year-old male weighing 88.6 kg, was the donor. A donor computed tomography angiogram demonstrated a nonreplaced arterial vasculature, a normal biliary anatomy, and patent hepatic veins. The portal vein had an anomaly in which the right posterior portal vein branch bifurcated from the main portal vein, and the right anterior portal vein branch bifurcated separately from the left portal vein. The total donor liver volume was 1780.5 mL, with 1274.4 mL coming from the right lobe and 506.1 mL coming from the left lobe (Fig. 1).
Because the remnant liver volume (28.4%) was less than our standard for right lobe procurement and the donor had a portal vein variant that generally indicates donation of the left lobe, we proposed an inflow modification (as detailed by Hill et al.) and the use of the smaller left lobe graft. Opting initially for splenic artery ligation, we would combine this maneuver with a portal vein shunt to further guard against small-for-size syndrome; the actual graft weight was noted to be 457 g on the back table (graft-weight/recipient ratio = 0.55). As such, we dissected out the splenic vein and the left renal vein below the transverse mesentery to perform a distal splenorenal shunt. Notably, the left adrenal vein was found to be sizeable when we isolated the renal vein, and it was used as a conduit. Thus, after dividing the gonadal vein to ensure greater renal vein mobility, we anastomosed the adrenal vein in an end-to-side fashion to the splenic vein as described by Aydin et al.
After the recipient liver was removed, the donor graft was implanted with the standard piggyback technique. We then ligated the splenic artery to optimize this undersized graft. Balancing the possibly reduced portal inflow against relative splanchnic hypertension, we opted to formally ligate the proximal splenic vein and convert the end-to-side spleno-adrenal shunt into a functional distal splenorenal shunt and establish a gradient of 2 mm Hg (Fig. 1).
Recovery was prolonged by a bile leak from the liver's cut surface and early graft dysfunction consistent with small-for-size syndrome, which was completely resolved within 2 months. At the time of this writing, the patient was symptom-free with normal hepatic function on standard immunosuppression.
The use of the left lobe and the acceptance of donors with a predicted graft weight/recipient ratio < 0.8 increases access to living donor liver transplantation, although theoretically there is an augmented risk of graft failure. The use of portal flow modulation has been postulated to increase the utilization of small grafts in living donor liver transplantation.[4, 5]
Guided by the principle that a selective shunt could lower the portal vein pressure without diverting the flow, we preferred to create a functional distal splenorenal shunt. Admittedly, in the comparison of shunts, there has been limited inquiry into the hemodynamics involved in living donor liver transplantation. Contrastingly, possibly because of the technical ease, splenic artery modulation in living donor liver transplantation has been more frequently explored and chronicled. Recently, Umeda et al. demonstrated that recipients with a graft weight/recipient ratio < 0.8 who underwent splenic artery embolization or ligation had decreased portal flow volume in comparison with nonmodulated cohorts. Moreover, splenic artery embolization has been shown to dramatically counter the effects of small-for-size syndrome in the postoperative period.[7, 8] Therefore, given our very real concerns about small-for-size syndrome in the recipient, we opted to ligate the splenic artery in combination with the creation of a functional distal splenorenal shunt.
We present here a novel dichotomous strategy of inflow modulation and portal vein flow diversion that has yet to be described. Appropriately, we submit that the utility of such an intervention may be questioned because of our inability to abrogate small-for-size syndrome in this patient. Contrastingly, the counterargument presupposes that without the aforementioned portal venous flow alteration, the resultant graft dysfunction may have been worse. Perhaps, in hindsight, intraoperative flow studies may have been of use, but it is unclear what clinical role a static assessment of the portal vein should play in the setting of intravenous pressors, large volume shifts, and a hyperdynamic cardiac state. Moreover, changes in the graft resistance interposed between the systemic and splanchnic venous circulations may occur long after the patient leaves the operating room.[9, 10]
A large, multicenter, prospective study of flow modulation in living donor liver transplantation is underway at our center as part of the Adult-to-Adult Living Donor Liver Transplantation Cohort Study to address these issues. This experience may provide yet another impetus for exploring the use of smaller grafts and the technical methods that make this more permissible.
Jay A. Graham, M.D.Benjamin Samstein, M.D.Jean C. Emond, M.D.Tomoaki Kato, M.D.
Department of Surgery Columbia University College of Physicians and Surgeons New York, NY