CT Features Of Invasive Pulmonary Fungal Disease And Quantitative CT Combined With Laboratory Indicators For Evaluation Of Prognosis Of Pneumocystis Jirovecii Pneumonia

PartⅠCT features of invasive pulmonary fungal disease in different immune statesObjective:To compare the CT performance characteristics of pulmonary aspergillosis,pulmonary cryptococcosis and pulmonary candidiasis in different immune states.Methods:Retrospective analysis of the clinical and imaging data of 117 cases of invasive pulmonary fungal disease(IPFD)from February 2014 to August 2022 in our Hospital,with 62 cases of pulmonary aspergillosis,34 cases of pulmonary cryptococcosis and 21cases of pulmonary candidiasis confirmed or clinically diagnosed.All patients underwent two or more chest CT examinations before diagnosis.The characteristics of the patients’chest CT were analysed,including the type of lesions,distribution of lesions,number and size of lesions and accompanying signs.The patients were divided into 68 cases in the immunosuppressed group and 49 cases in the non-immunosuppressed group according to whether their immune status was suppressed or not,and the incidence and CT characteristics of the three types of IPFD under different immune status were compared.The count data between groups were compared byχ~2test(continuous correctionχ~2test or Fisher’s exact probability method if the conditions were not met),and Bonferroni method was used for the comparison between two groups.Results:40 of the 62 cases of pulmonary aspergillosis were in the immunosuppressed group and 22 in the non-immunosuppressed group,12 and 22 of the 34 cases of pulmonary cryptococcosis and 16 and 5 of the 21 cases of pulmonary candidiasis,respectively,and the incidence of the three invasive pulmonary fungal diseases differed by immune status(P<0.05),and the incidence of pulmonary aspergillosis and pulmonary candidiasis in the immunosuppressed group was higher than that of pulmonary cryptococcosis(all P<0.0167).Comparison of CT features between the three invasive pulmonary fungal diseases in the immunosuppressed group showed:pulmonary aspergillosis showed more of a nodule/mass-type than pulmonary candidiasis(P=0.000),compared with pulmonary cryptococcosis and pulmonary candidiasis were mostly scattered in the whole lung(P=0.004,0.007),and the incidence of halo signs was higher than that of pulmonary cryptococcosis and pulmonary candidiasis(P=0.012,0.002);pulmonary cryptococcosis tends to have an outer band distribution compared to pulmonary aspergillosis and pulmonary candidiasis(P=0.001,0.006),the incidence of lobulation sign was higher than that of pulmonary aspergillosis(P=0.010);the lesion type of pulmonary candidiasis was more favorable toward the consolidation type than the pulmonary aspergillosis(P=0.000);the nodule/mass-type pulmonary aspergillosis and pulmonary cryptococcosis were different in the number of lesions(P=0.010).Comparison of CT features between the three invasive pulmonary fungal diseases in the non-immunosuppressed group showed:pulmonary aspergillosis showed more nodules/mass-type than pulmonary candidiasis(P=0.016),more inclined to the inner middle band than pulmonary cryptococcosis(P=0.015),and the incidence of tree bud sign and air crescent sign was higher than pulmonary cryptococcosis(P=0.009,0.004);pulmonary cryptococcosis showed more nodule/mass-type than pulmonary candidiasis(P=0.001),more prone than pulmonary aspergillosis was distributed in the outer band distribution(P=0.006),moreover,the incidence of halo sign and lobulation sign was higher than that of pulmonary aspergillosis(P=0.009,P=0.016);pulmonary candidiasis showed more consolidation than pulmonary cryptococcosis(P=0.006),and multiple lobe distribution than pulmonary aspergillosis and pulmonary cryptococcosis(P=0.009,0.004);pulmonary aspergillosis and pulmonary cryptococcosis were different in lesion size(P=0.005).CT features of pulmonary aspergillosis in different immunosuppressed states showed:the immunosuppressed group showed more of a nodule/mass-type(P=0.029),was multilobar(P=0.000),had a random scattered distribution(P=0.000)and had a greater incidence of the halo sign than the non-immunosuppressed group(P=0.000).The non-immunosuppressed group tended to have an inner middle band distribution(P=0.001),and the tree bud sign was more frequent than in the immunosuppressed group(P=0.001).The nodule/mass-type pulmonary aspergillosis was different in the number of lesions between the two groups(P=0.000).There was no significant difference in CT features between pulmonary cryptococcosis and pulmonary candidiasis in different immune states.Conclusions:1.Pulmonary aspergillosis and pulmonary candidiasis are more common in immunosuppressed populations,while pulmonary cryptococcosis occurs higher in non-immunosuppressed populations;2.Pulmonary aspergillosis in the immunosuppressed state with multiple nodule/mass-type as the main CT signs,while the random scattered distribution of multiple lung lobes,accompanied by the halo sign is more typical;while in the non-immunosuppressed state,the distribution of single lung lobes in the inner middle band,while the number of nodular masses are mainly single,accompanied by tree bud sign is more typical.The incidence of air crescent sign has no significant difference under different immune states of pulmonary aspergillosis,but it is different from the characteristic performance of other two types of IPFD under non-immunosuppressive state;3.Pulmonary cryptococcosis in the non-immunosuppressive state with nodule/mass-type as the the main CT signs,regardless of the immune status,and the lesions tend to have an outer band distribution,with the accompanying signs being more typical of lobulation sign.The incidence of halo sign has no significant difference under different immune states of pulmonary cryptococcosis,but it is different from the characteristic performance of other two types of IPFD under non-immunosuppressive state;4.Pulmonary candidiasis is more manifested as consolidation type under any immune state.Part ⅡQuantitative CT combined with laboratory indicators for evaluation of prognosis of Pneumocystis jirovecii pneumoniaObjective: To investigate the relationship between quantitative chest CT analysis of Pneumocystis jirovecii pneumonia(PJP)and laboratory indicators,with the aim of assessing the severity of the disease and the value of prognosis.Methods: Retrospective analysis of clinical and chest CT imaging data of 58 patients with PJP from February 2014 to August 2022 in our hospital,of which 35 were HIV-infected PJP patients(HIV-PJP)and 23 were non-HIV-infected PJP patients(non-HIV-PJP),and the above two types of PJP were separately classified according to the outcome during hospitalization The patients were divided into survival and death groups.The 3D Slicer software was applied to quantitative analysis of chest CT according to the threshold segmentation method to obtain the corresponding indexes,including the percentage of ground glass opacity volume(GGOV%),the percentage of solid opacity volume(SOV%),the percentage of total lung lesion volume(Le V%),and the ratio of solid opacity volume to ground glass opacity volume(SOV/GGOV).The Fisher’s exact probability method,independent samples t-test and Mann-Whitney U test were used to compare the difference types of general clinical data,laboratory tests and CT quantitative indicators between HIV-PJP and non-HIV-PJP in the survival and death groups,respectively.Correlations between laboratory tests and CT quantitative indicators were analyzed between patients with both types of PJP using Spearman or Pearson tests.The predictive efficacy of in-hospital mortality in patients with PJP was assessed by plotting receiver operating characteristic curve(ROC).Results: There were no significant differences between the group comparisons of clinical general data between the two types of PJP patients.The LDH was higher than the survival group,and the CD4+T lymphocyte count was lower than the survival group(all P<0.05);on the CT quantification,GGOV%,SOV%,Le V% and SOV/GGOV were all higher than the survival group(all P<0.05).In the laboratory examination of non-HIV-PJP patients,the NEU and LDH in the death group were higher than those in the survival group(all P<0.05),and the other differences were not statistically significant;In terms of CT quantitative indicators,there was no significant difference in GGOV%,SOV%,Le V%,SOV/GGOV between the two groups.The GGOV%,SOV%,Le V%,SOV/GGOV and LDH of HIV-PJP patients were positively correlated(r=0.366,0.494,0.555,0.388,all P<0.05);However,GGOV%,Le V%,CD4+T lymphocyte count were negatively correlated(r=-0.343,-0.410,all P<0.05).Le V% of non-HIV-PJP patients was positively correlated with NEU(r=0.425,P<0.05);At the same time,GGOV% and Le V% were positively correlated with LDH(r=0.445,0.442,P<0.05).The area under the ROC curve of CT quantitative indicators GGOV%,SOV%,Le V%,SOV/GGOV and laboratory indicators LDH,CD4+T lymphocyte count predicting the death of HIV-PJP patients was0.812,0.868,0.936,0.756,0.906,0.863,respectively.The area under the ROC curve of laboratory indicators LDH and NEU predicting the death of non-HIV-PJP patients was0.762,0.746,respectively.Conclusions :1.LDH and each CT quantitative indicators were higher in the HIV-PJP death group than in the survival group,while CD4+ T lymphocyte counts were lower than in the survival group;LDH and NEU were higher in the non-HIV-PJP death group than in the survival group,while each CT quantitative indicators did not differ significantly between the two groups;2.Quantitative CT analysis is valuable for assessing the severity of PJP.each CT quantitative indicators of HIV-PJP was positively correlated with LDH,while GOV% and Le V% were negatively correlated with CD4+T lymphocyte count;GOV% and Le V% of non-HIV-PJP were positively correlated with LDH,while Le V% was also positively correlated with NEU;3.Quantitative CT indicators can provide an objective data model for HIV-PJP prognosis but have limited value in predicting death in non-HIV-PJP.LDH and CD4+T lymphocyte count are important predictors of in-hospital mortality in HIV-PJP;LDH and NEU are important predictors of in-hospital mortality in non-HIV-PJP.