| BackgroundFungemia is a highly lethal and medically burdensome disease.Early prediction of the 30-day mortality risk in patients with fungemia in intensive care unit(ICU)is of great significance for disease management.Nonetheless,there is currently a lack of prediction tools for 30-day mortality of ICU patients with fungemia.As echinocandin has replaced fluconazole as the first-line therapy for yeast-associated fungemia,the resistance rate to echinocandin has increased accordingly.Therefore,it is important to reassess the prognosis of ICU patients with fungemia who were initially treated with echinocandin or fluconazole.Objective1.Based on the Medical Information Mart for Intensive Care(MIMIC-Ⅲ)database,a 30-day mortality risk prediction model for ICU patients with fungemia was constructed,and the model was verified using the data of ICU patients with fungemia in a Grade-Ⅲ Class-A hospital in China,in the hope of providing 30-day mortality risk screening tools for ICU patients with fungemia.2.Based on the data of ICU patients with fungemia from MIMIC-Ⅲ database and from the Grade-Ⅲ Class-A hospital in China,this paper analyzed whether there was a difference in the 3 0-day prognosis between ICU patients with fungemia initially treated with echinocandin and with fluconazole,in order to provide clinical evidence for the management of antifungal drugs.Method1.Data of ICU patients with fungemia from both the MIMIC-Ⅲ database and the Grade-Ⅲ Class-A hospital in China were collected.The data extracted from the MIMIC-Ⅲ database functioned as the training dataset,which was used to construct the predictive model for 30-day mortality risk in ICU patients with fungemia;the data from the hospital functioned as the validation dataset,which was used to validate the model.The training dataset served to screen the potential risk factors by means of Least Absolute Shrinkage and Selection Operator(LASSO)regression,and statistically significant variables(P<0.05)were included via logistic regression analysis.The predictive model for 30-day mortality risk in ICU patients with fungemia was then built based on R software.Such indicators as C-index and calibration curve were utilized to evaluate the prediction ability of the model.The data of ICU patients with fungemia from the hospital were used as the validation dataset to validate the model.2.Through a retrospective cohort study of the data of ICU patients with fungemia from both the MIMIC-Ⅲ database and the Grade-Ⅲ Class-A hospital in China,initial fluconazole treatment(the case group)and initial echinocandin treatment(the control group)were propensity-score-matched(PSM)(match tolerance 0.02,match 1:1).so as to analyze whether there is a difference in 30-day mortality between initial fluconazole treatment and initial echinocandin treatment for ICU patients with fungemia.Result1.Predictive models were constructed by age,international normalized ratio(INR),renal failure,liver disease,respiratory rate(RR),glucocorticoid therapy,antifungal therapy,and platelets.The C-index value of the models was 0.838(95%CI:0.79096-0.88504).Attested by the external validation results,the model has satisfactory predictive ability.2.234 ICU patients with fungemia were included in the study,with 107(45.7%)in the case group and 127(54.3%)in the control group.After propensity score matching(1:1),there was no difference in 30-day mortality between the two groups(OR 1.043,95%CI 0.615-1.770,P=0.875).ConclusionThe 30-day mortality risk predictive model for ICU patients with fungemia constructed in this study has good predictive ability,and may hopefully provide a 30day mortality risk screening tool for ICU patients with fungemia.In addition,there was no difference in 30-day mortality between echinocandin and fluconazole as initial antifungal therapy for ICU patients with fungemia and the choice of initial antifungal therapy for fungemia in ICU can be made based on local fungal distribution and fungal resistance in ICU. |
