| Background and ObjiectivesInvasive pulmonary aspergillosis (IPA) is an opportunistic infection that occurs mainly among patients with hematologic malignancies, solid tumors, critically illness, HIV/AIDS, allogeneic stem cell transplantation, and solid organ transplantation, especially among patients with prolonged neutropenia. However, IPA has also merged as an important cause of morbidity and mortality in non-neutropenic patients without underlying diseases. Few data are available on IPA in such patients. IPA that occurs among non-neutropenic patients with underlying diseases is common, but the epidemiology, clinical characteristics, outcomes and prognosis is still unclear, which differ greatly from IPA in neutropenic patients. Therefore, we conducted a retrospective cohort study of non-neutropenic patients who had been diagnosed with IPA at our institution over the last10years to investigate epidemiology, clinical characteristics, outcomes and prognosis and survey the difference between non-neutropenic patients with and without underlying diseases.MethodsWe retrospectively review all cases of Invasive pulmonary aspergillosis (IPA) in non-neutropenic patients seen in the NanFang hospital from Jannuary1.2000to September1,2010. The clinical data of proven and probable IPA cases were collected using our hospital computerized medical records database, and the follow-up data were collected by telephone and letters.1. we conducted a retrospective study of non-neutropenic patients who had been diagnosed with IPA at our institution over the last10years to investigate epidemiology, clinical characteristics, radiological and endoscopic features and survey the difference between non-neutropenic patients with and without underlying diseases.2. we conducted a retrospective cohort study of non-neutropenic patients who had been diagnosed with IPA at our institution over the last10years to investigate outcomes, prognosis and prognostic factors using a Cox proportional-hazards model, furthermore, establish a PI prognostic model.3. we conducted a case-control study to investigate the risk factors of voriconazole-induced mental disorders.4. Statistical analyses were executed with SPSS13.0(SPSS Inc.Chicago, IL, USA). Continuous variables are described as mean±standard deviation (SD) and median (lower quartile-upper quartile). Qualitative variables were compared using the Chi-square test or Fisher’s exact test when appropriate. For quantitative data, t tests wre used. Survival was estimated by using Kaplan-Meier method. Overall survival was compared between groups using Log-Rank test. The prognostic sigificance of variates was studied in univariate analysis by the Log-Rank test. All variates identified by univariate analysis(p<0.05) were entered into a Cox proportional-hazards model for multivariate analysis, and hazard ratios (HR) and95%CL were reported. In the case-control study, non-conditional logistic regression model is used in multivariate analysis. All tests used were two-tailed and statistical significance was defined as a p value<0.05. Results1.52patients were included in this study, in which thirty-three cases (63.5%) were histologically proven, and19(36.5%) were probable. The distribution of underlying conditions among all patients:Chronic pulmonary disease (n=22,42.3%of the total), mostly including COPD, was the main underlying condition. Seventeen patients (32.7%of the total) underwent organ failure. Chronic cardiovascular disease (n=8,15.4%), pulmonary tuberculosis (n=7,13.5%), chronic liver disease (n=7,13.5%) were also common underlying condition in non-neutropenic patients with IPA. leukemia, lung cancer each presented in4cases(7.7%). only2(3.8%) patients were solid-organ transplant recipients. Ten patients (19.2%) had no underlying disease.2. All patients presented with at least one of the following symptoms:cough (n=44,84.6%), fever (n=23,44.2%), hemoptysis (n=12,23.1%), dyspnea (n=21,40.4%), chest tightness (n=8,15.4%), weight loss (n=8,15.4%), chest pain (n=2,3.8%).19patients (36.5%) only had moist crackly on auscultation.5(9.6%) had wheezes, both in9cases(17.3%), neither in19cases (36.5%). Seventeen patients (36.7%) were infected with IPA in spring.16(30.8%) were infected in autumn,12(23.1%) in winter,7(13.5%) in summer. There is no seasonal difference found in the distribution of cases in four seasons (χ2=6.359, P=0.095).3. Chest CT was performed in36(69.2%) patients within3days before or after diagnosis. The most frequent Chest CT abnormalities were infiltrates in19(52.8%) patients. Nodules in12cases(33.3%), segmental areas of consolidation in7(19.4%), pleural effusion in5(13.9%). Cavity or cavities were seen in4(11.1%) patients. Only3(8.3%) of36patients showed the air crescent sign, and none showed the halo sign. The most frequent locations of the chest CT abnormalities were the right upper lobe (n=22,61.1%) and the left upper lobe (n=21,58.3%).19(52.8%) patients were involved in the the left lower lobe. The right lower and middle lobe were involved in 17(47.2%) and16(44.4%) patients respectively. However, there is no significant difference in frenquency of involvement in each lobe (χ2=2.156, P=0.707). Chest radiographs, performed in46(88.5%) patients within3days before or after diagnosis, mainly showed nonspecific patchy shadows(n=31,67.4%). Nodules were seen in11patients (23.9%), consolidation in5(10.9%), cavity or cavities in5(10.9%). Typical imaging, such as air crescent sign and halo sign, were not observed.4. Only24reports of bronchoscopy are available. The typical endoscopic manifestations, such as ulceration, nodule, pseudomembrane, plaque, or eschar were occasionally observed. However, airway secretion found in bronchoscopy vary greatly (χ2=11.111, P=0.011), in which Cheese-like secretions or necrosis was seen in10patients (41.7%), transparent secretions in9(37.5%), purulent in3(12.5%), bloody secretions in2(8.3%).5. Peripheral neutrophil cout was surveyed in all patients with a mean of7.51±6.27×109cells/L (range,0.74-22.4×109cells/L). Serum lactate dehydrogenase (LDH) levels were elevated in52patients, with a mean of228.3±152U/L (range,56-745U/L). Serum β-D-glucan levels were elevated in only14patients, with a mean of117.8±156.5pg/L (range,5-522.8pg/L). One or more kinds of bacteria or other fungi were isolated from sputum culture in17patients, which included Staphylococcus. aureus (n=4), Escherichia Coli (n=4), Streptococcus (n=3), Pseudomonas aeruginosa (n=3), Staphylococcus epidermidis (n=2), Stenotrophomonas spp (n=2), Candida albicans (n=2), Acinetobacter baumannii (n=2). Fecal bacteria (n=1), Others (n=1).11patients had cultured only one microorganism,6had two microorganisms.6. No significant difference was observed in the ratio of male to female patients, age, body weight between the two groups. In seasonal distribution, the significant difference was observed in non-neutropenic patients with underlying disease (p=0.026). However, no seasonal difference was found in the other group (p=0.622). The only significant difference in symptoms between the2groups was fever (>38℃)(p=0.015). The group without underlying disease presents less fever, and commonly no signs on auscultation by comparison with the other group. There was no significant difference in predilection sites, radiological findings (including thoracic CT and chest radiography) and endoscopic manifestations between the2groups, nor did the complication by infection of the other pathogens. The significant difference in laboratory findings was peripheral platelet count (p=0.01). serum LDH level (p=0.00), and serum albumin level (p=0.044).7. The overall crude mortality rate among52patients was38.1%. ninteen patients died, and3were lost during follow-up. There is no significant difference in survival curve between the two groups (p=0.056). The survival rate in non-neutropenic patients with underlying disease was55.1%(mortality rate44.9%), while that in cases without underlying disease was90%(mortality rate10%). The attribuable mortality of IPA in non-neutropenic patients was22.3%(10%divided by44.9%).8. Cox multivariate analysis showed organ failure (HR:8.739,95%CI:3.770-20.255; p=0.000) and high LDH levels in serum (HR:1.004,95%CI:1.001-1.007; p=0.004) to be independently associated with overall mortality.9. Prognostic formula:PI=0.004*LDH+2.168*organ failure. Death within the first, third, sixth months since diagnosis of IPA would be predicted with sensitivity of100%,100%,100%, specificity of87.8%,92.3%,94.7%, positive predictive value of68.8%,81.3%,87.5%and negative prediction value of100%,100%,100%, consistency of90.3%,94.2%,96.2%, kappa value of0.753.0.856,0.906respectively when the PI takes the optimal cutoff value of2.7680.10. Delirium is the most common psychotic manifestation of voriconazole-induced mental disorders (50%), followed by the visual hallucinations and/or auditory hallucinations (25%). The mental symptoms occurred in3±2days after receiving voriconazole and disappeared within2±1days after drug withdrawal. Univariate analysis and multivariate Logistic regression analysis eventually revealed that the risk factors for voriconazole-induced mental disorders were age (OR=1.136;95%CI=1.044～1.238; P=0.003) and daily maintenance dose per kilogram body weight (OR=1.546;95%CI=1.062～2.251; P=0.023)Conclusion1. Clinical, radiological and endoscopic features of IPA are frequently lacked in non-neutropenic patients, especially in those without underlying disease.2. Presentations of CT scan and X-ray are nonspecific in non-neutropenic patients, and the typical imaging, such as air crescent sign, was less frequently observed in non-neutropenic patients. IPA has no infected predilection in lung lobes among non-neutropenic patients.3. Seasonal difference of distribution is great in non-neutropenic patients with underlying disease, while the patients without underlying disease manifested no difference in four seasons.4. Invasive pulmonary aspergillosis in non-neutropenic patients carries a overall mortality of38.1%, a attributable mortality of22.3%.5. Two risk factors were independently associated with overall mortality in non-neutropenic patients with IPA:organ failure and high LDH levels in serum.6. Prognostic formula:PI=0.004*LDH+2.168*organ failure. The optimal cutoff value of PI is2.768. The PI index model is a better tool for prognosis in non-neutropenic patients with IPA.7. Age and daily maintenance dose per kilogram body weight were independent risk factors for mental disorders induced by voriconazole. |
PDF Full Text Request