Analysis of influencing factors for treatment outcomes and establishment of prediction model for the treatment of carbapenem-resistant Klebsiella pneumoniae infection
Objective To analyze clinical characteristics of Carbapenem-resistant Klebsiella pneumoniae(CRKP)infections,identify factors influencing treatment outcomes,and construct a predictive model for assessing the treatment outcomes,providing insights to improve clinical treatment outcomes.Methods A retrospectively analysis was conducted on 207 CRKP infection cases admitted to the First Affiliated Hospital of Zhengzhou University from January 2022 to December 2022.Clinical features were examined,and single-factor and multifactorial logistic regression analyses were used to explore factors affecting the treatment effects.A predictive model was developed,and the receiver-operating characteristic(ROC)curve was utilized to assess its clinical applicability.Results The study included 147 males(71.0％)and 60 females(29.0％).The intensive care unit(ICU)was the primary source of cases(85 cases,41.1％).Sputum(54 cases,26.1％),bronchoalveolar lavage fluid(53 cases,25.6％),and blood samples(36 cases,17.4％)were the primary sources of specimens.With combined comorbidities such as hypoproteinemia,hypertension,and cerebrovascular disease(＞35.0％)were prevalent among CRKP patients,most of whom had underdone invasive procedures prior to infection.Post-treatment,45.89％(95 cases)experienced poor outcomes,while 54.11％(112 cases)had favorable responses.Multifactorial logistic regression identified male gender[OR(95％CI):0.278(0.120-0.648)],ICU admission[OR(95％CI):36.902(10.687-127.421)],and hospitalization duration[OR(95％CI):0.969(0.954-0.985)],albumin levels[OR(95％CI):0.849(0.790-0.912)],and hormone use[OR(95％CI):7.270(1.762-30.002)]as independent predictors of treatment outcomes(all P＜0.05).The predictive model formula is P=1/(1+e^-y),where Y=1.781-1.279 ×sex+3.608×ICU admission-0.031×hospital stay-0.164×albumin value+1.984×hormone use,and P represents the probability of a poor treatment effect,and Y is the prediction index.The model showed a sensitivity of 0.811 and specificity of 0.804.Conclusions CRKP infection predominantly originate from the ICU,with hypoalbuminemia,hypertension,and cerebrovascular disease being significant risk factors.Clinically,it is of imperative to focus on preventive measures and care for males and ICU patients,reduce hospital stay,and limit hormone use to curb CRKP infection and improve patient prognoses.
Carbapenem-resistant to Klebsiella pneumoniaeInfluencing factorsTreatment outcomesPredictive modeling
NETL
NSTL
万方数据
