So Yeong Jeong1, Jeongjin Lee2, Kyoung Won Kim3, Jin Kyoo Jang1, Heon-Ju Kwon4, Gi Won Song5, Sung Gyu Lee5. 1. Department of Radiology and the Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-Ro 43-Gil, Songpa-Gu, Seoul 138-736, South Korea. 2. School of Computer Science and Engineering, Soongsil University, Seoul, South Korea. 3. Department of Radiology and the Research Institute of Radiology, University of Ulsan College of Medicine, Asan Medical Center, 88 Olympic-Ro 43-Gil, Songpa-Gu, Seoul 138-736, South Korea. Electronic address: kimkw@amc.seoul.kr. 4. Department of Radiology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, South Korea. 5. Division of Liver Transplantation and Hepatobiliary Surgery, Departments of Surgery, University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea.
Abstract
RATIONALE AND OBJECTIVES: To determine the accuracy of semiautomated CT volumetry using portal vein (PV) segmentation to estimate volume of the right posterior section (RPS) graft compared to intraoperative measured weight (W) in live liver donors. MATERIALS AND METHODS: Among 23 donors who donated RPS grafts for liver transplantation in our institution from April 2003 to August 2016, 17 donors with CT scans within 3 months of liver procurement and PV anatomy of type I-III were included. RPS volumes were retrospectively evaluated by semiautomated CT volumetry (RPSVCTV) and by measurement of standard liver volume (SLV) and PV area ratio (RPSVSLV). RPS volumes were compared to W for correlation coefficients, (absolute) difference, and (absolute) percentage deviation. Linear fitting was performed to identify the method that yielded the greatest correlation with W. RESULTS: Mean values of RPSVCTV, RPSVSLV, and W were 503.4 ± 97.8 mL (346.6-686.0), 516.54 ± 146.20 (274.06-776.32), and 518.8 ± 122.4 (370.0-789.0), respectively. RPSVCTV was strongly correlated with W (r = 0.9414; p < 0.0001), whereas RPSVSLV was only moderately did (r = 0.5899; p = 0.0127). RPSVCTV showed a significantly smaller absolute difference (35.20 ± 30.82 vs. 104.79 ± 60.27, p = 0.004) and absolute percentage deviation (6.61 ± 4.90 vs. 19.92 ± 10.37, p < 0.0001) from W. Equation correlating RPSVCTV and W was W = -74.7191 + 1.1791 RPSVCTV (R2 = 0.8862; p < 0.001). CONCLUSION: RPSVCTV yields smaller absolute difference than RPSVSLV for estimating intraoperative measured weight of RPS in live liver donors. Semiautomated CT volumetry using PV segmentation is feasible for the estimation of the volume of the RPS of the liver, and RPSVCTV was strongly correlated with W (r = 0.9414; p < 0.0001).
RATIONALE AND OBJECTIVES: To determine the accuracy of semiautomated CT volumetry using portal vein (PV) segmentation to estimate volume of the right posterior section (RPS) graft compared to intraoperative measured weight (W) in live liver donors. MATERIALS AND METHODS: Among 23 donors who donated RPS grafts for liver transplantation in our institution from April 2003 to August 2016, 17 donors with CT scans within 3 months of liver procurement and PV anatomy of type I-III were included. RPS volumes were retrospectively evaluated by semiautomated CT volumetry (RPSVCTV) and by measurement of standard liver volume (SLV) and PV area ratio (RPSVSLV). RPS volumes were compared to W for correlation coefficients, (absolute) difference, and (absolute) percentage deviation. Linear fitting was performed to identify the method that yielded the greatest correlation with W. RESULTS: Mean values of RPSVCTV, RPSVSLV, and W were 503.4 ± 97.8 mL (346.6-686.0), 516.54 ± 146.20 (274.06-776.32), and 518.8 ± 122.4 (370.0-789.0), respectively. RPSVCTV was strongly correlated with W (r = 0.9414; p < 0.0001), whereas RPSVSLV was only moderately did (r = 0.5899; p = 0.0127). RPSVCTV showed a significantly smaller absolute difference (35.20 ± 30.82 vs. 104.79 ± 60.27, p = 0.004) and absolute percentage deviation (6.61 ± 4.90 vs. 19.92 ± 10.37, p < 0.0001) from W. Equation correlating RPSVCTV and W was W = -74.7191 + 1.1791 RPSVCTV (R2 = 0.8862; p < 0.001). CONCLUSION: RPSVCTV yields smaller absolute difference than RPSVSLV for estimating intraoperative measured weight of RPS in live liver donors. Semiautomated CT volumetry using PV segmentation is feasible for the estimation of the volume of the RPS of the liver, and RPSVCTV was strongly correlated with W (r = 0.9414; p < 0.0001).