| Objective:Alcoholic hepatitis(AH)refers to a group of clinicopathological syndromes caused by a large number of necrosis of liver cells in a short period of time.It can occur with or without cirrhosis.Severe alcoholic hepatitis(SAH)refers to acute liver dysfunction such as jaundice and coagulation dysfunction in patients with AH.The short-term mortality is high.Alcoholic liver disease related acute-on-chronic liver failure(ALD-ACLF)may occur when there are incentives such as infection or alcohol consumption.It is characterized by the combination of one or more organ failure.The short-term mortality of SAH patients with ACLF is significantly increased.However,due to the different causes and incentives caused by lifestyle and epidemiology in different regions,the clinical manifestations and course of disease of ACLF patients are very different,and the applicable population and diagnostic criteria formulated by societies in different regions are also different.Among them,the European Association for the Study of the Liver-Chronic liver failure(EASL-CLIF),the North American Consortium for the Study of End-Stage Liver Disease(NACSELD),and the large-scale multi-center prospective study led by the First Affiliated Hospital of Zhejiang University Medical College in China proposed the definitions of EASL-CLIF ACLF,NACSEL ACLF and COSSH-ACLF(Chinese Group on the Study of Severe Hepatitis BACLF)according to the characteristics of the disease in the region.The performance of different diagnostic criteria in the prognosis evaluation of this special group of SAH patients is still unclear,and the guiding role of diagnosis and treatment of the disease remains to be further explored.This study is an ambispective cohort study,a total of 309 patients with SAH were collected retrospectively and prospectively,and compared the characteristics of patients who met the three ACLF diagnostic criteria and the performance of the three diagnostic criteria in the prognosis evaluation of SAH patients.Methods:The clinical data of 309 SAH patients hospitalized in the Department of Hepatobiliary and Pancreatic Medicine of the First Hospital of Jilin University from March 2012 to December 2022 were collected and followed up.Baseline data,incidence and types of organ failure,and survival data on admission or on the 7th,28 th,and 90 th days after diagnosis of ACLF were collected.The performance of the three diagnostic criteria in the prognosis evaluation of SAH patients was compared and analyzed.Results:1.General characteristics of AH patients:Among the 309 SAH patients,293 cases were male(94.8%)and 16 cases were female(5.1%).The mean age was 50.1±8.8 years.There were 183(59.2%).ALD-ACLF patients who met the COSSH-ACLF diagnostic criteria,297(96.1%)cses with cirrhosis,106(34.3%)cases who met the EASL-CLIF ACLF criteria,and 16(5.2%)patients who met the NACSELD ACLF criteria.264 cases were followed up and 8 cases received liver transplantation.The 7-day,28-day and 90-day transplant-free survival rates of ALD-ACLF patients who met the COSSH-ACLF criteria were 96.2%、82.2%、64.2%,respectively.The 7-day,28-day and 90-day survival rates of ALD-ACLF patients who met the EASL-ACLF diagnostic criteria were 94.3%、73.4%、51.9%,respectively.The 7-day,28-day and 90-day survival rates of ALD-ACLF patients who met the NACSELD ACLF criteria were 68.8%、43.8%、18.8%,respectively.45 cases were lost to follow-up.2.Comparison of clinical characteristics between ALD-ACLF patients diagnosed according to COSSH-ACLF criteria and non-ALD-ACLF patients: Among the 309 SAH patients,183 cases(59.2%)were diagnosed as ALD-ACLF according to COSSH-ACLF criteria.Compared with the non-ALD-ACLF group,the incidence of hepatic encephalopathy,spontaneous bacterial peritonitis,sepsis,and infection in the ALD-ACLF group was higher than that in the non-ALD-ACLF group(P<0.05).WBC、ANC、NLR、PLT、AST、ALT、GGT、TBIL、DBIL、IBIL、BUN、SCr、PT、INR、TC in ALD-ACLF group were higher than those in non-ALD-ACLF group(P<0.05).3.Comparison of clinical characteristics between ALD-ACLF patients diagnosed according to EASL-CLIF ACLF criteria and non-ALD-ACLF patients :106 cases(34.3%)were diagnosed with ALD-ACLF according to EASL-CLIF ACLF criteria.Compared with the non-ALD-ACLF group,the incidence of hepatic encephalopathy,spontaneous bacterial peritonitis,sepsis,and co-infection in the ALD-ACLF group was higher than that in the nonALD-ACLF group(P<0.05).WBC、ANC、NLR、PLT、AST、ALT、TBIL、DBIL、IBIL、BUN、SCr、PT、INR in ALD-ACLF group were higher than those in non-ALDACLF group(P<0.05).ALC、ALB、Na、TG、CHE in ALD-ACLF group were lower than those in non-ALD-ACLF group(P<0.05).4.The clinical characteristics of ALD-ACLF patients and non-ALD-ACLF patients diagnosed according to the NACSELD ACLF criteria :16 patients(5.2%)with ALD-ACLF diagnosed according to the NACSELD ACLF criteria.Compared with the non-ALD-ACLF group,the incidence of hepatic encephalopathy,gastrointestinal bleeding,sepsis combined infection in the ALD-ACLF group was higher than that in the non-ALD-ACLF group(P <0.05).The WBC、ANC、NLR、BUN、SCr、PT、INR in the ALD-ACLF group were higher than those in the non-ALD-ACLF group(P<0.05).The GGT、ALP、TG in the ALD-ACLF group were lower than those in the non-ALD-ACLF group(P<0.05).5.There were 183 patients with COSSH-ACLF,including 101 patients(55.2%)with ACLF-1,59 patients(32.2%)with ACLF-2 and 23 patients(12.6%)with ACLF-3.The 7-day survival rates of patients with ACLF-1,2 and 3 were 99.0%,98.3% and 78.3%,respectively.At 28 days,they were 93.3%,83.5% and 32.2%,respectively.At 90 days,they were 75.8%,63.7% and 18.4%,respectively.There were 106 patients with EASL-CLIF ACLF,including 31 patients(29.2%)with ACLF-1,56 patients(52.8%)with ACLF-2 and 19 patients(17.9%)with ACLF-3.The 7-day survival rates of patients with ACLF-1,2 and 3 were 96.8%,98.2% and 78.9%,respectively.The 28-day survival rates were 89.3%,80.6% and 28.2%,respectively.The 90-day survival rates were 59.5%,62.0% and 11.3%,respectively.6.Organ failure in patients with SAH according to different diagnostic criteria: Among the 183 patients who met the COSSH-ACLF criteria,159 cases(86.8%)had liver failure,45cases(24.5%)had coagulation failure,44 cases(24.0%)had renal failure,27 cases(14.7%)had respiratory failure,26 cases(14.2%)had nerve failure,and 16 cases(8.7%)had circulatory failure.Among the 106 patients who met the criteria of EASL-CLIF ACLF,83cases(78.3%)had liver failure,38 cases(35.8%)had coagulation failure,39 cases(36.7%)had kidney failure,24 cases(22.6%)had respiratory failure,23 cases(21.6%)had nerve failure,and 14 cases(13.2%)had circulatory failure.Among the 16 patients who met the NACSELD ACLF criteria,2 cases(12.5%)had renal failure,10 cases(62.5%)had respiratory failure,16 cases(100.0%)had nerve failure,and 10 cases(62.5%)had circulatory failure.7.Prognostic analysis of SAH patients who met different ACLF diagnostic criteria:regardless of the diagnostic criteria,compared with the non-ALD-ACLF group,the MDF,MELD score,ABIC score and GAHS of the ALD-ACLF group were higher than those of the non-ALD-ACLF group,and the difference was highly statistically significant(P<0.001).The Kaplan-Meier curve was drawn according to the survival of patients.The mortality rate of patients in the ACLF group was high and the prognosis was poor.The difference was statistically significant(Log-rank test P<0.001).8.The predictive value of different diagnostic criteria for mortality:In predicting 7-day mortality.NACSELD ACLF criteria had higher overall accuracy(95.58% vs 67.69%,P<0.001 and 95.58% vs 42.86%,P<0.001),specificity(96.49% vs67.25%,P<0.001 and 96.49% vs 41.46%,P<0.001)and positive predictive value(37.50%vs 6.00%,P<0.001 and 37.50% vs 4.00%,P<0.001)than EASL-CLIF ACLF criteria and COSSH-ACLF criteria.But the sensitivity is low.EASL-CLIF ACLF criteria also had higher overall accuracy(66.69% vs 42.86%,P<0.001)and specificity(67.25% vs 41.46%,P<0.001)than COSSH-CLIF ACLF criteria.The prediction of 28-day mortality was similar to that of 7-day mortality.NACSELD ACLF criteria had higher overall accuracy(90.19% vs 72.08%,P<0.001 and 90.19% vs51.32%,P<0.001),specificity(97.45% vs 70.94%,P<0.001 and 97.45% vs 45.53%,P<0.001)and positive predictive value(62.50% vs 26.88%,P=0.005 and 62.50% vs 8.47%,P<0.001)than EASL-CLIF ACLF criteria and COSSH-ACLF criteria.But the sensitivity and negative predictive value were low.EASL-CLIF ACLF diagnostic criteria also had higher overall accuracy(72.08% vs 51.32%,P<0.001)and specificity(70.94% vs 45.53%,P<0.001)than COSSH-ACLF criteria.In the 90-day mortality prediction,the sensitivity and negative predictive value of NACSELD ACLF criteria decreased significantly,and the overall accuracy was equal to that of EASL-CLIF ACLF criteria(78.52% vs 73.47%,P=0.187),which was still higher than that of COSSH-ACLF criteria(78.52% vs 58.59%,P<0.001).NACSELD ACLF criteria still had higher specificity(98.94% vs 75.66%,P<0.001 and 98.94% vs 49.74%,P<0.001)and positive predictive value(87.50% vs 49.45%,P=0.005 and 87.50% vs 37.09%,P<0.001)than EASL-CLIF ACLF criteria and COSSH-ACLF criteria.The sensitivity(20.90% vs67.16%,P<0.001 and 20.90% vs 83.58%,P<0.001)and negative predictive value(77.92%vs 86.67%,P=0.026 and 77.92% vs 89.52%,P=0.011)were still low.EASL-CLIF ACLF criteria also had higher overall accuracy(73.44% vs 58.59%,P<0.001)and specificity(75.66%vs 49.74%,P<0.001)than COSSH-ACLF criteria.9.Clinical characteristics and prognosis of patients who met the COSSH-ACLF criteria alone and those who met the EASL-CLIF ACLF criteria and did not meet the NACSELD ACLF criteria: Among the 309 SAH patients,77 cases(24.9 %)patients only met the COSSH-ACLF criteria.Among the 77 patients,10 cases(13.0%)patients had gastrointestinal bleeding,7 cases(9.1%)had hepatic encephalopathy,20 cases(26.0%)had spontaneous bacterial peritonitis,69 cases(89.6%)had ascites,and 1 cases(1.3%)had sepsis.44 cases(57.1%)were complicated with infection(bacteria or fungi).There were 76 cases(98.7%)of liver failure,7 cases(9.1%)of coagulation failure,5 cases(6.5%)of renal failure,3 cases(3.9%)of respiratory failure,3 cases(3.9%)of nerve failure and 2 cases(2.6%)of circulatory failure.Of the 68 patients,70 cases(90.9%)were ACLF-1,3 cases(3.9%)were ACLF-2,and 4 cases(5.2%)were ACLF-3.The 7-day,28-day and 90-day survival rates were 98.7 %,91.4 % and 81.6 %,respectively.There were 90 patients who met the EASL-CLIF ACLF criteria but did not meet the NACSELD ACLF criteria,including 24 cases(26.7%)of gastrointestinal bleeding,38 cases(42.2%)of hepatic encephalopathy,39 cases(43.3%)of spontaneous bacterial peritonitis,84 cases(93.3%)of ascites,9 cases(10.0%)of sepsis,and 73 cases(81.1%)of combined infection(bacteria or fungi).There were 73 cases(81.1%)of liver failure,32 cases(35.6 %)of coagulation failure,32 cases(35.6%)of renal failure,9 cases(10.0%)of respiratory failure,11 cases(12.2%)of nerve failure and 2 cases(2.2%)of circulatory failure.Among them,32 cases(35.6%)were ACLF-1,52 cases(57.8%)were ACLF-2,6 cases(6.7%)were ACLF-3.The 7-day,28-day and 90-day survival rates were 97.8%,80.2% and 59.8%,respectively.Conclusions:1.ALD-ACLF often occurs in patients with SAH,and the mortality rate is significantly increased.2.The number of patients with ACLF,the incidence of organ failure and mortality were significantly different according to different ACLF diagnostic criteria.3.The grade distribution of COSSH-ACLF patients and EASL-CLIF ACLF patients was different,and the mortality of grade 3 was significantly higher than that of grade 1-2.4.NACSELD ACLF criteria had higher overall accuracy,specificity and positive predictive value than EASL-CLIF ACLF and COSSH-ACLF criteria in predicting 7-day mortality.Similar results were found in the prediction analysis of 28-day mortality,but in the 90-day mortality prediction,the sensitivity and negative predictive value of NACSELD ACLF criteria decreased significantly,and the overall accuracy was equal to EASL-CLIF ACLF criteria and still higher than COSSH-ACLF criteria.5.The mortality of patients who met EASL-CLIF ACLF criteria but did not meet NACSELD ACLF criteria was still high.The mortality of SAH patients who only met COSSH-ACLF criteria was comparable to that of non-ACLF patients. |
PDF Full Text Request