Abstract:
Objective
To investigate the risk factors for infection in patients with hematological malignancies (HM) undergoing chemotherapy and neutropenia, and to construct and validate a column chart model for predicting the risk of infection.
Methods
A retrospective analysis was conducted on the clinical data of 120 patients admitted to Longhua District People’s Hospital in Shenzhen from January 2021 to January 2023 who underwent initial HM chemotherapy and had a neutropenia phase, all patients were divided into infected group (43 cases) and control group (77 cases) based on whether infection occurred. Independent risk factors for infection during the neutropenia phase after HM chemotherapy were analyzed through univariate and multivariate Logistic regression analysis, and a predictive model (R software) was constructed.The discrimination and accuracy of the predictive model on the risk of infection during the neutropenia phase after HM chemotherapy were evaluated by receiver operating characteristic (ROC) curves and calibration curves. Among the 120 patients with neutropenia after HM chemotherapy, 43 patients developed infection,with an infection rate of 35.83% (43/120). Total of 47 strains of pathogenic bacteria were detected in 43 specimens of infected patients, including 33 strains (70.21%) of Gram negative bacteria, 12 strains (25.53%)of Gram positive bacteria, and 2 strains (4.26%) of fungi. Multivariate analysis showed that the frequency of chemotherapy ≥ 3 times (OR = 2.561, 95%CI:0.019-5.031, P < 0.001), cutaneous mucosal lesion (GradeⅠ:OR = 1.547, 95%CI:1.215-1.978, P < 0.001; Grade Ⅱ:OR = 2.649, 95%CI:1.134-4.547, P < 0.001;Grade Ⅲ:OR = 3.423, 95%CI:1.753-6.686, P < 0.001), and the improved infection likelihood score (mIPS)in the neutropenia phase ≥ 13 points (OR = 4.447, 95%CI:1.830-8.842, P < 0.001) were all risk factors for infection of patients with HM. Patients with HM chemotherapy who have a neutropenia period ≥ 7 days (OR =5.571, 95%CI:1.842-9.421, P < 0.001) and a hospital stay ≥ 14 days (OR = 2.213, 95%CI:1.264-4.431, P <0.001) were more likely to develop infections during the neutropenia phase. The area under the curve of the column chart model predicting infection in patients with granulocyte deficiency after HM chemotherapy was 0.846 (95%CI:0.809-0.884), with the sensitivity of 87.15% and the specificity of 89.67%. The slope of the calibration curve for predicting infection was close to 1, and the goodness of fit test results showed that the difference between the predicted probability of infection risk in patients with granulocyte deficiency after HM chemotherapy and the actual probability was not statistically significant (χ2 = 0.169, P = 0.643).
Conclusions
Patients with granulocyte deficiency stage after HM chemotherapy have a higher risk of infection. The number of chemotherapy cycles ≥ 3, skin and mucosal damage, mIPS score ≥ 13, granulocyte deficiency ≥ 7 days, and hospital stay ≥ 14 days are all high-risk factors for infection in patients with granulocyte deficiency stage after HM chemotherapy.
Key words:
Hematological malignancy,
Chemotherapy,
Agranulocytosis stage,
Infected,
Risk factors,
Risk prediction model
Hongchu Huang, Meirong Huang, Lihong Wen. Risk factor analysis and risk prediction model of granulocytopenia infection after chemotherapy in patients with hematological malignancies[J]. Chinese Journal of Experimental and Clinical Infectious Diseases(Electronic Edition), 2024, 18(05): 285-292.