| Objective:To assess the value of serum (1,3)-β-D-glucan (G), galactomannan antigen (GM) and fungal culture for the diagnosis of invasive fungal infections (IFI) in the intensive care unit (ICU) patients.Method:We retrospectively analyzed 243 inpatients of ICU in Qilu Hospital of Shandong University, from January 2010 to December 2010. The inpatients must meet these follow conditions:first, the length of stay in ICU≥72h, the second, inpatients aged≥18 years. Demographic data, clinical indicators and auxiliary examination results were collected in all patients. According to the invasive fungus infection diagnosis standards (draft) of patients with hematologic disease/cancer, the patients involved in the study were divided into IFI group (proven, probable, possible) and excluded IFI group. According to pathogen, patients in IFI group were divided into three subgroups-Aspergillus, Candida and other fungal infection groups. (1) Calculate the sensitivity, specificity, positive predictive value, negative predictive value of G test for IFI group, invasive candidiasis (IC) subgroup and invasive aspergillosis (IA) subgroup respectively. Calculate the sensitivity, specificity, positive predictive value, negative predictive value of G test for IA group. Compare the diagnostic value of G test and GM test in the patients of IA. (2) Analyze the relationship between BG levels and the prognosis of patients and the effect of anti-fungal treatment. (3)Collect fungal culture results from all specimens of 243 cases. Analyze the compliance in Candida culture and the G test, and the value of Candida culture in the diagnosis of IC. Analyze the compliance in Aspergillus culture and the GM test, G test, and the value of Aspergillus culture in the diagnosis of IA. (4) Compare patients'age, sex, underlying diseases, the length of stay in ICU, APACHEⅡscore, SOFA score, septic shock rate, mechanical ventilation rate, central venous catheter rate, MODS rate and other related clinical information in IFI group and excluded IFI group to analyze predisposing factors of IFI.Results:(1) 243 patients collected were divided into 4 groups:4 cases (1.6%) were proven,36 cases (14.8%) were probable,22 cases (9.1%) were possible, and the rest 186 cases (75.0%) were out of fungal infection. (2) Taking G≥20pg/ml as G test positive, the sensitivity, specificity, positive predictive value and negative predictive value of G test for IFI group were:79.5%,78.6%,67.3% and 87.5%. The sensitivity, specificity, positive predictive value and negative predictive value of G test for IC subgroup were:82.1%,78.9%,57.9% and 91.8%. The sensitivity, specificity, positive predictive value and negative predictive value of G test for IA subgroup were:75.0%, 80.6%,48.0% and 93.1%.Taking GM index≥0.5 as index cutoff of the GM test, The sensitivity, specificity, positive predictive value and negative predictive value of G test for IA subgroup were: in the 11 proven and probable IA patients in whom both G test and GM test were carried out, the sensitivity of G test (81.8%) was higher than that of GM test (72.7%). (3)The BG levels significantly decreased in 33 patients who had received effective anti-fungal treatment (86.06±52.38 vs 35.24±22.32 P<0.001). But in 16 patients who had received Ineffective anti-fungal treatment, the BG levels were not significantly different (88.26±54.27 vs 86.50±52.16 P=0.568). Collect the highest values of serum BG in all IFI patients. We found that the mortality rate in group 50～100pg/ml and group 100～200pg/ml were higher than that of group <50 pg/ml (11/11 vs 5/20 P=0.030,14/5 vs 5/20 P<0.001), but in 50～100pg/ml group and 100～200pg/ml group there's no significant difference in mortality (11/11 vs 14/5 P= 0.121).(3) The positive rate of G test in patients with Candida culture positive was 15.4% (12/78). The positive rate of G test in cases with Aspergillus culture positive was 72.7%(8/11). In the IPA patients whose G tests were positive,40.0%(6/15) of the sputum culture for Aspergillus had positive results. The positive rate of GM test in cases with Aspergillus culture positive was 63.6%(7/11).In the IPA patients whose GM test were positive,42.9%(3/7) of the sputum culture for Aspergillus had positive results.(4) Comparing the chronic underlying diseases in IFI group and excluded IFI group, we concluded that the incidences of diabetes (17/62vs 28/181 P=0.037), COPD (13/62 vs 8/181 P=0.035), chronic liver or kidney disease (14/62 vs 19/181 P= 0.017), autoimmune disease (9/57 vs 2/56 P=0.046)and malignant tumor(17/62 vs 18/181 P=0.001) were higher in the IFI group than those in the excluded IFI group. However, the incidences of high blood pressure and coronary heart disease had no significant differences. The proportion who received long-term glucocorticoid treatment in the IFI group (10/62) was higher than that in the excluded IFI group (12/ 181), the difference was statistically significant (P=0.024). The APACHEⅡscore and SOFA score were significantly higher in the IFI group than in the excluded IFI group (14.33±5.61 vs 12.52±3.85 P=0.040), (6.24±3.83 vs 4.83±2.26 P= 0.015).The incidence of septic shock in the IFI group was higher than in the excluded IFI group (14/62 vs 11/181 P<0.001); the time of ICU stay in IFI group (39.24±29.65) was longer than in the excluded IFI group (26.98±18.50), the difference was statistically significant (P=0.008); the proportion of mechanical ventilation in the IFI group was higher than that of excluded IFI group (53/62 vs 113/181 P=0.001); the proportion of venous catheter in the IFI group was higher than that of excluded IFI group (50/62 vs 120/181 P=0.033);the incidence of MODS was higher in IFI group(7/62 vs 7/181 P=0.030).The age and sex had no difference between IFI group and cxcluded IFI group.Conclusion:(1) G test is a good diagnosis method for ICU patients with suspected IFI. Dynamic monitoring of serum BG benefits critically ill patients in the early diagnosis of IFI.(2) Persistent high level of BG in serum often prompts a serious fungal infection or an ineffective anti-fungal treatment. Therefore, monitoring serum BG level dynamically can assess the prognosis of patients and guide the adjustment of antifungal treatment.(3) Sputum culture positive for Candida are more considered as colonization in the respiratory tract, and rarely indicate the invasive Candida infections, so antifungal therapy should not be used. But the Candida (either from the same species or different species) in many parts (including respiratory tract, urinary tract and other open cavities) cultured positively can be prompted to IC, if necessary, early antifungal therapy should be done. Positive culture of Aspergillus in sputum should not be treated as the colonization. It may prompt more to the diagnosis of invasive aspergillosis.(4) Patients with underlying diseases such as diabetes, COPD, chronic liver failure or kidney failure, autoimmune diseases, long-term use of glucocorticoids, septic shock, prolonged mechanical ventilation, high APACHE II score and SOFA score and the long-term stay in ICU are more likely to suffer invasive fungal infections. |
PDF Full Text Request