| Background and Aims: The albumin-bilirubin(ALBI) grade is a recently reported, simpler, more objective and evidence-based alternative to the Child-Pugh(CTP) score for hepatocellular carcinoma(HCC). Serum prealbumin(PA) concentration might be a more sensitive indicator than albumin(ALB) in assessing liver reserve function as the shorter half-life time( 1.9 days) of PA. We plan to develop a modified ALBI grade(m ALBI) in which PA substitute for ALB in ALBI grade. The aims of this study were to examine roles and advantages of m ALBI predicted the prognoses of patients with HCC.Methods: A retrospective cohort study was conducted. Cases were 583 HCC patients who underwent curative resection and 489 HCC patients who underwent transarterial chemoembolization / embolization(TACE / TAE). Univariable and multivariable Cox regression analyses were performed to predict prognoses of HCC patients. The Kaplan-Meier(K-M) method was used to plot overall survival and disease-free survival curves,and the log-rank test was used to compare the differences. Akaike information criterion(AIC) assessed model fit,-2log likelihood estimated homogeneity, Harrell’s concordance index( C-index) evaluated predictive power, and linear trend χ2 test assessed heterogeneity were used to determined which is the best between the seven Liver Function Classification Systems.Results: There were 841 men(84.8%) and 151 women(15.2%) with male-to-female ratio of 5.6: 1. The median age was 55± 11.5 years. There were 518 patients dead during follow-up. The median survival time was 55.0 months and the mortality rate was 52.2%. Patients were divided into CTP A grade(CTP 5 or 6 scores) 909 patients and CTP B grade(CTP 7- 9 scores) 83 patients by Child-Turcotte-Pugh score(CTP). And the 1-year, 3-year, 5-year survival rates were 82.6% and 55.4%, 61.6% and 34.9%, 51.4% and 21.6%, respectively. Patients were divided into ALBI 1 grade 673 patients, ALBI 2 grade 315 patients, ALBI 3 grade 4 patients by ALBI grade. And the 1-year survival rates were 86.8%, 67.0%, 50.0%, respectively. The 3-year survival rates were 67.2%, 43.5%, 0.0%, respectively. The 5-year survival rates were 57.4%, 31.6%, 0.0%, respectively. Patients were also divided into m ALBI 1 grade 240 patients, ALBI 2 grade 623 patients, ALBI 3 grade 96 patients by m ALBI grade. And the 1-year survival rates were 91.7%, 82.5%, 44.8%, respectively. The 3-year survival rates were 79.2%, 58.3%, 36.0%, respectively. The 5-year survival rates were 68.2%, 47.3%, 22.9%, respectively.By comparing the betweell-group heterogeneity and predictive power of m ALBI, ALBI, CTP, MELD, MELD-Na, i MELD and MESO, the m ALBI grade were the best, ALBI followed, and CTP was poor. The values linear trend χ2 test and C-index were 61.288 and 0.612, 60.336 and 0.570, 27.030 and 0.541, 0.646 and 0.500, 11.349 and 0.548, 0.124 and 0.504, 6.653 and 0.531, respectively. The m ALBI grade was still the best in the comparison of the homogeneity within group and model fit, however, the CTP and ALBI grades were worse than others. The values-2log likelihood and AIC were 6304.196 and 6305.84, 6716.546 and 6748.7, 6754.206 and 6755.87, 6440.203 and 6441.98, 6423.935 and 6425.69, 6440.349 and 6442.12, 6431.923 and 6433.69, respectively. Regardless of liver function CTP gade A, or ALBI gade 1, the m ALBI gade always showed the best predictive performance prognoses of patients who had underwent initial hepatectomy, TACE or TAE treatment.Conclusions: The ALBI and CTP grades could well evaluate the liver function and prognosis of patient with HCC. However, regardless of liver function was CTP gade A or ALBI gade 1, the m ALBI could better evaluate the liver function and prognoses of patients who had underwent initial hepatectomy, TACE or TAE treatment. It may be superior to existing other Liver Function Classification Systems. However, it needs high-quality, multi-center prospective study further to demonstrate. |
PDF Full Text Request