Accurate and reasonable method for estimation of graft volume in living donor liver transplantation.

Ding Yuan,Kefei Chen,Bo Li,Lvnan Yan,Yonggang Wei
DOI: https://doi.org/10.1097/TP.0b013e3181875959
2008-01-01
Transplantation
Abstract:The graft volume (GV) must be balanced between donor and recipient postoperative safety in living donor liver transplantation (LDLT). Therefore, accurate preoperative estimation of GV is crucial to confirm donor safety and avoid small-for-size syndrome in recipients. However, estimated GV with computed tomographic (CT) volumetry is usually associated with a large error compared with the actual graft volume (AGV) (1, 2). The aim of the study was to create a new formula for estimating GV and compare this with previous methods on a cohort of living liver donors. From March 2005 to December 2007, 88 donors undergoing right hepatectomy for LDLT were reviewed at the liver transplantation center of the West China Hospital. Donor selection criteria and operative techniques were performed according to previously described reports (3). The graft was harvested without the middle hepatic vein by using an ultrasonic dissector without inflow vascular occlusion. GV was preoperatively estimated by CT volumetry (CTGV). AGV was measured by water replacement method on the back table. The estimation formula was created from the initial 50 donors. CTGV was significantly higher than AGV (791.4±147.9 mL vs. 580.9±99.1 mL, P=0.000). The CT adjusted formula was created as follows: CTGV-GV (mL)= 0.548×CTGV (mL)+146.9 (R=0.819, P=0.000). The formula related body surface area (BSA) was derived from the stepwise regression analysis as follows: BSA-GV (mL)=356.9×BSA (m2)+2.74×age− 104.97 (R=0.536, P=0.000). The mean percentage of right GV to the standard liver volume (SLV) was 50.2%±7.6%, thus the formula-related SLV was adjusted as follows: SLV-GV (mL)=0.502×SLV (mL) (4, 5). We further validated these formulas from subsequent 38 donors (Table 1). There was only significant difference between AGV (600.7±109.7 mL) and CTGV by paired sample t test (P= 0.000). The error ratio of CTGV was significantly higher than the other error ratios of GVs (P=0.000). However, there was no significant difference among the error ratios of CTGV-GV, BSA-GV, or SLV-GV (P>0.05). According to the agreement analysis, Lin's concordance correlation coefficient (6) further indicated the best agreement between CTGV-GV and AGV.TABLE 1: Evaluation of different methods in subsequent 38 donorsCurrently, the method of graft size estimation using CT volumetry is popular and useful, but estimated GV is associated with a margin of error compared with AGV (1, 5, 7, 8). Dehydration status of University of Wisconsin solution, the inclusion of intrahepatic blood volume, the mismatching of cutting line, as well as the performance of the scanner and the technical expertise of the radiologist, can contribute to this discrepancy (4, 7, 9). In fact, the formula-related BSA or SLV estimated more accurately GV than CT volumetry, but the two formulas only depended on pure anthropometric variables. Although CT volumetry was disappointing for estimating GV (error ratio=32.8%±18.4%), there was a linear relationship between CTGV and AGV (R=0.819). Thus, we designed this study to identify accurately estimating GV. Our results indicated CTGV-GV was not only more accurate than CTGV and the CT adjusted formula but also was in better agreement than the formula related BSA or SLV. Because of a prominent variation in the size and shape of liver for some individuals, great variation between AGV and estimated GV using anthropometric variables derived formulas may appear in a few donors. Because our CT adjusted formula takes preoperative morphology of the liver into consideration, the individual variation is pre-excluded using this formula. The CT-adjusted formula is an optimal method for estimating GV in LDLT. Ding Yuan Kefei Chen Bo Li Lvnan Yan Yonggang Wei Department of Liver Surgery and Liver Transplantation Center of West China Hospital Sichuan University, Chengdu Sichuan, China
What problem does this paper attempt to address?