A scoring system to predict the risk of invasive fungal infection in children of general surgical department
DING Xiang-yu, JIA Chen-hong, ZHANG Gu-ying, LI Qian, LIU Yan-hui
2020, 15(6):
331-337,353.
Asbtract
(
205 )
PDF (1133KB)
(
69
)
References |
Related Articles |
Metrics
Objective To investigate the risk factors of invasive fungal infection in pediatric patients of general surgical department and to establish a risk scoring system. Methods Risk factors of 82 hospitalized children diagnosed with IFI in the general surgery department of a children's hospital from January 2010 to December 2018 and 82 non IFI hospitalized children with similar basic diseases in the same ward were investigated. Possible risk factors were analyzed by univariate logistic regression analysis and non-conditional multivariate logistic regression analysis to select the risk factors of IFI. According to the results of regression analysis, a scoring system was established, the regression coefficient was integer assigned, the scoring formula was established, the z-value was calculated, and the ROC curve was drawn to evaluate the effectiveness of the scoring system, and the z-critical value was determined according to the cut-off value. The scoring system was used to verify 227 children admitted to the same ward from January to December 2019 with similar underlying diseases in the scoring system group, and to compare and verify the positive rate of diagnosis according to the diagnostic criteria or clinical diagnosis. Results Single factor analysis showed that the length of stay in the observation group was longer than that in the control group, and the proportion of indicators such as abdominal drainage tube, parenteral nutrition, central venous catheter, mechanical ventilation application>3 days, broad-spectrum antibiotics application>2 weeks, combination of two or more antibiotics, hypoproteinemia and glucocorticoid application>7 days were all higher than those in the control group, with statistical significance(P>0.05). Multiple-factor logistic regression analysis showed the length of stay, the combination of two or more antimicrobial agents, glucocorticoid application>7 days were independent risk factors for infant IFI in general surgery department. According to the regression coefficients of 2, 26 and 15, and the scoring system was established, the AUC of ROC curve was 0.880 (95% CI:0.825-0.935, P<0.001), and the cut-off value was 0.700. In other words, when the Z critical value was 63.5, the sensitivity was 0.778, and the specificity was 0.922. There was no statistical difference between the positive rate predicted by the scoring system in the verification group and the positive rate predicted by the diagnostic criteria or clinical diagnosis (P=0.074), Kappa=0.769, indicating a good consistency between the two methods. The sensitivity and specificity of the scoring system were 0.750 and 0.981, which were basically consistent with those of the cut-off value. Conclusion The effectiveness of the scoring system was good and its correctness had been verified, which had a good application value for early clinical identification of high-risk children with general surgery IFI.
