Objective:To investigate the related factors and prognosis of invasive Klebsiella pneumoniae liver abscess syndrome (IKLAS). Methods:The in-patients diagnosed with Klebsiella pneumoniae liver abscess in the Department of Infectious Diseases, Huashan Hospital, Fudan University from January 2015 to February 2021 were retrospectively enrolled. The patients were divided into IKLAS group and non-IKLAS group according to whether they had IKLAS or not. The clinical data between the two groups were compared, including the prevalence of diabetes mellitus, the details of liver abscess, clinical symptoms such as fever and abdominal pain, as well as laboratory tests such as glycosylated hemoglobin and hemoglobin. Statistical analysis was performed using chi-square test or independent sample t test. Multivariate logistic regression analysis was used to analyze the factors influencing the occurrence of IKLAS. Results:A total of 75 patients with Klebsiella pneumoniae liver abscess were enrolled, including 55 patients (73.33%) in the IKLAS group and 20 patients (26.67%) in the non-IKLAS group. Fifty-two point seven three percent (29/55) of the patients had diabetes mellitus and 12.73%(7/55) of the patients had abdominal pain in the IKLAS group, which were 20.00%(4/20) and 45.00%(9/20) in the non-IKLAS group, respectively, and the differences were both statistically significant ( χ2=6.38 and 7.28, respectively, both P<0.05). Most of liver abscesses were single (50/75, 66.67%), and more likely to occur in the right liver (50/75, 66.67%). The maximum diameter of liver abscess in the IKLAS group was (4.58±2.04) cm, which was smaller than that in the non-IKLAS group ((6.49±3.11) cm), and the difference was statistically significant ( t=2.82, P=0.011). Compared with those in the non-IKLAS group, patients in the IKLAS group had higher glycosylated hemoglobin (8.69%±2.64% vs 6.18%±1.31%) and hemoglobin ((112.25±22.04) g/L vs (100.05±18.59) g/L), and the differences were both statistically significant ( t=-4.25 and -2.21, respectively, both P<0.05). The proportion of patients using antibiotics combined with abscess drainage in the IKLAS group was 38.18%(21/55), and that in the non-IKLAS group was 85.00%(17/20). The difference between the two groups was statistically significant ( χ2=12.86, P<0.001). A total of 16 patients (21 eyes) were diagnosed as endogenous Klebsiella pneumoniae endophthalmitis (EKPE), and all of them were IKLAS patients, and 14 patients underwent monocular/binocular eyeball injection and/or vitrectomy and silicone oil filling. The visual acuity of 13 patients decreased significantly. Multivariate logistic regression analysis showed that complicated with diabetes mellitus was an independent risk factor for IKLAS (odds ratio ( OR)=5.02, 95% confidence interval (95% CI) 1.01 to 25.03, P=0.049). The large diameter of liver abscess was a protective factor for IKLAS ( OR=0.64, 95% CI 0.47 to 0.86, P=0.003). Conclusions:The patients with IKLAS have less abdominal pain, and most of them complicate with diabetes mellitus. Diabetes mellitus is an independent risk factor for the occurrence of IKLAS, while the large diameter of liver abscess is a protective factor. EKPE is associated with poor visual prognosis.