- Email: [email protected]
Computed tomography findings in invasive pulmonary aspergillosis in non-neutropenic transplant recipients and neutropenic patients, and their prognostic value
Journal of Infection (2011) 63, 447e456
www.elsevierhealth.com/journals/jinf
Computed tomography findings in invasive pulmonary aspergillosis in non-neutropenic transplant recipients and neutropenic patients, and their prognostic value* Seong Yeon Park a,c,d, Chaehun Lim b,d, Sang-Oh Lee a, Sang-Ho Choi a, Yang Soo Kim a, Jun Hee Woo a, Jae-Woo Song b, Mi Young Kim b, Eun Jin Chae b, Kyung-Hyun Do b, Koun-Sik Song b, Joon Beom Seo b,**,e, Sung-Han Kim a,*,e a
Departments of Infectious Diseases, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap-dong, Songpa-gu, Seoul 138-736, Republic of Korea b Department of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea c Department of Internal Medicine, Dongguk University International Hospital, Dongguk University College of Medicine, Seoul, Republic of Korea Accepted 10 August 2011 Available online 16 August 2011
KEYWORDS Invasive pulmonary aspergillosis; Non-neutropenic transplant recipients; Neutropenic patients
Summary Objectives: We evaluated CT findings and their prognostic value in nonneutropenic transplant recipients with invasive pulmonary aspergillosis (IPA) compared with neutropenic patients with IPA. Methods: All adult patients during a 27-month who met the criteria for proven or probable IPA according to the 2008 EORTC/MSG criteria were retrospectively enrolled. Initial CT findings were reviewed by two radiologists blinded to the patients’ demographics and clinical outcomes. Results: A total of 50 non-neutropenic transplant recipients and 60 neutropenic patients were enrolled. Consolidation-or-mass, halo signs, and angio-invasive form were observed less often in non-neutropenic transplant recipients than in neutropenic patients: (56%, 26%, and 32%) versus (78%, 55%, and 60%, p Z 0.01, p Z 0.002, and p Z 0.003, respectively). Multivariate analysis revealed that macronodules (HR 0.31, p Z 0.001), multiple infarct-shaped consolidations
*
Trial Registration: ClinicalTrials.gov Identifier: NCT001178177. * Corresponding author. Tel.: þ82 2 3010 3305; fax: þ82 2 3010 6970. ** Corresponding author. Tel.: þ82 2 3010 4383; fax: þ82 2 3010 8127. E-mail addresses: [email protected] (J.B. Seo), [email protected] (S.-H. Kim). d S. Y. Park and C. Lim contributed equally to the study. e S-H Kim was responsible for the clinical aspects of the study and J B Seo for the radiologic aspects. 0163-4453/$36 ª 2011 The British Infection Association. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jinf.2011.08.007
448
S.Y. Park et al. (HR 4.26, p < 0.001), renal replacement therapy (HR 5.62, p < 0.001) and persistence of a positive serum galactomannan (HR 7.14, p < 0.001) were independently associated with 90-day mortality. Conclusions: Our data indicate that CT findings in non-neutropenic transplant recipients with IPA are similar to those in neutropenic patients with IPA except that consolidation-or-mass, halo sings, and angio-invasive form are less frequent, and certain CT findings such as macronodules and multiple infarct-shaped consolidations have prognostic implications in IPA. ª 2011 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
Introduction Invasive pulmonary aspergillosis (IPA) is an opportunistic infection that occurs mainly in patients with neutropenia and in hematopoietic stem cell and solid organ transplant recipients.1,2 The survival of these patients depends on early diagnosis and prompt initiation of therapeutic measures.3,4 However, early diagnosis of IPA based on culture and/or microscopy of histopathologic specimens is generally difficult because obtaining tissue biopsies to permit definite diagnosis is not without risk in critically ill patients due to coagulation abnormalities and thrombocytopenia.5 Thus, early diagnosis of IPA relies heavily on imaging modes such as chest computed tomography (CT).1 These are used to differentiate fungal from bacterial and viral pneumonia, because early antifungal therapy improves survival.4 The basis of the lung injury caused by Aspergillus in patients with neutropenia as the predominant mechanism of immunosuppression may differ from that in patients who have numerically adequate but functionally-impaired phagocytes.6,7 Although the clinical and radiological features of IPA in patients with neutropenia have been relatively well studied,4,8e11 little is known about the characteristics of IPA in non-neutropenic transplant recipients.12,13 We therefore evaluated CT findings and their prognostic value in non-neutropenic transplant recipients versus neutropenic patients with IPA.
Methods Study population and data collection This study was performed at the Asan Medical Center, a 2700-bed tertiary care teaching hospital in Seoul, South Korea, from January 2008 to March 2010. It was approved by the Institutional Review Board at Asan Medical Center (approval number 2010-0460). The study patients were classified as either nonneutropenic transplant recipients or neutropenic patients. Non-neutropenic transplant recipients were defined as those patients with IPA who underwent solid organ transplantation (SOT) or hematopoietic stem cell transplantation (HSCT),6,14,15 and there were no evidence of neutropenia at the time of diagnosis of IPA. Neutropenic patients were defined as those patients with IPA who received chemotherapy for acute leukemia, lymphoma, or a plastic anemia while having an absolute neutrophil count <500 cells/mm3 within 30 days prior to diagnosis of IPA.1 HSCT recipients with IPA who did not achieve primary
engraftment, or who, after engraftment, suffered from neutropenia related to recurrence of underlying diseases, were classified as neutropenic patients.15 Corticosteroid use was defined as use of a mean minimum dose of 0.3 mg/kg/day of prednisolone equivalent for >3 weeks.1 We excluded from our analysis patients with mixed pulmonary infections (i.e., fungal other than Aspergillus spp., bacterial, and viral co-infections). During the first 6 months after SOT, all liver transplant recipients (about 300 cases per year) received oral itraconazole or low dose liposomal amphotericin (1 mg/kg/day), as fungal prophylaxis. However, kidney (about 210 cases per year) and heart transplant patients (about 30 cases per year) did not receive any routine antifungal prophylaxis due to their lower incidence of IPA.2 Only one lung transplant was performed in our hospital during the study period. HSCT recipients (about 120 allogeneic HSCT and 40 autologous HSCT cases per year), received micafungin as fungal prophylaxis from day þ1 to ANC 3000 cells/mm3 after engraftment. However, neutropenic patients who were given cytotoxic chemotherapy did not receive any routine antifungal prophylaxis. Serum galactomannan (GM) antigen levels were measured as described previously.13,16,17 The decision to perform fiberoptic bronchoscopy with bronchoalveolar lavage (BAL) was at the discretion of each attending physician. Fiberoptic bronchoscopy was performed using four 50 mL aliquots of sterile saline solution. An optical density ratio of 0.5 or greater in serum or BAL samples was considered positive for GM.13,16,17 Positive GM levels in patients receiving piperacillin-tazobactam, amoxicillin-clavulanate or ampicillin-sulbactam before serum or BAL GM assay were excluded because these antibiotics can cause false-positive results in GM assays.16,17 Serum GM antigen levels in liver transplant recipients, HSCT recipients and neutropenic patients were measured once or twice in a week by ELISA. Serum GM was not regularly measured in other patients, although the assay was performed whenever IPA was suspected. Median follow-up duration of GM antigen level was 46 days (interquartile range [IQR] 16.5e88.0 days). GM antigen level was checked median 7 times (IQR, 3e11 times). We defined the persistence of a positive serum GM as serial GM monitoring did not revealed negative conversion within 3 months after the diagnosis of IPA.18,19
Case definitions of IPA Patients were assigned a proven or probable diagnosis of IPA according to the consensus definition of the EORTC/MSG and our previous report.1,13,16 Proven IPA was defined by
CT findings in invasive pulmonary aspergillosis histologic evidence of tissue invasion, including septated, acutely branching filamentous fungi and positive culture. Probable IPA was defined as the presence of a host factor, together with clinical indications such as dense, wellcircumscribed lesions with or without a halo sign, aircrescent sign, or cavity on CT; and mycologic evidence of fungal infection (by culture or cytologic analysis of BAL fluid for Aspergillus species or GM assay of serum or BAL). Possible IPA was defined as the presence of a host factor and clinical criteria without mycologic evidence for IPA. Cases that were classified as possible IPA were not included in this analysis.
Radiologic definitions and interpretation categories Chest CT scans with or without contrast enhancement were performed whenever IPA was suspected. Baseline chest CT findings were reviewed by two independent
449 radiologists unaware of patients’ characteristics and clinical outcomes. Where there was disagreement the final decision was made by consensus after discussion. The radiologists used a standard glossary of CT imaging definitions to categorize pulmonary lesions (Supplementary Table).20 Radiological patterns of IPA were classified as either angio-invasive or airway-invasive form, based on previous studies.21,22 A radiologic pattern was considered angioinvasive form when the predominant CT imaging fulfilled two of the three following findings: 1. a halo sign, 2. an infarct-shaped consolidation, 3. an internal low attenuation, cavity, or air-crescent sign (Fig. 1). It was considered airway-invasive form when the predominant CT fulfilled two of the three following findings following findings: 1. small airway lesions, 2. a peribronchial consolidation, 3. a bronchiectasis (Fig. 2). It was classified as a combined form was considered when CT finding met
Figure 1 Typical diagnostic findings of angio-invasive form. (A) CT scan with lung window setting shows subsegmental, triangular consolidation with the base abutting the pleura and the apex directed toward the hilum (arrows), characterized by infract-shaped consolidation. And ground glass attenuation is observed around consolidation, suggesting of CT halo sign (arrowheads). (B) Internal low attenuated lesion is noted within consolidation in left lower lobe. It is necrotic portion of consolidation and precursor of cavity. (C) Cavity is defined as a gas-filled space, seen as a lucency or low attenuation area within consolidation (double arrow). (D) Aircrescent sign is a collection of air in a crescentic shape that separates the wall of a cavity from an inner consolidation (black arrowheads).
450
S.Y. Park et al.
Figure 2 Typical diagnostic findings of airway-invasive form. (A, B) CT scan with lung window setting shows small, clustered or centrilobular micronodules and branching-linear opacities (white arrows) with mild bronchiectasis (arrowheads). (C) Peribronchial opacity or consolidation is classified as the manifestation of airway-invasive form (arrow).
both angio-invasive and airway-invasive form and it was designated undetermined form when the CT image did not fulfill any of the above criteria. To simply the patterns, we re-assigned these forms into two types, angioinvasive and non-angio-invasive types. The angio-invasive type was defined as the angio-invasive or combined form, and the non-angio-invasive type as the airwayinvasive or undetermined form.
Statistical analysis All statistical analyses were performed with SPSS version 14.0 (SPSS, Chicago, IL). Categorical variables were compared using the c2 or Fisher’s exact test, and continuous variables were compared using the ManneWhitney U test. Univariate and multivariate analyses for factors associated with time to death were performed using Cox proportional hazards models. All variables with p-value <0.10 by univariate analysis were included in a multivariate analysis. A p-value <0.05 were considered statistically significant.
Results Patient characteristics During the study period, a total of 112 patients comprising 52 transplant recipients (42 SOT and 10 HSCT) and 60 neutropenic patients were diagnosed with IPA. Of these, two SOT recipients who revealed neutropenia at the diagnosis of IPA were excluded from the analysis. Finally, the 110 patients with IPA (10 histologically proven and 100 probable cases) were analyzed. The characteristics of the overall study patients are shown in Table 1. Of the 40 SOT recipients, 9 (23%) received immunosuppression-pulse therapy because of chronic or acute rejection, and 21 (53%) patients had graft dysfunction at the time of IPA. Among l0 HSCT recipients (8 allogeneic and 2 autologous HSCT), 7 (70%) had graft versus host disease (GVHD). Of the 60 neutropenic patients with IPA, 15 (25%) received HSCT but did not achieve primary engraftment or suffered from neutropenia related to recurrence of underlying diseases after engraftment.
CT findings in invasive pulmonary aspergillosis
451
Table 1 Baseline clinical characteristics of non-neutropenic transplant recipients and neutropenic patients with invasive pulmonary aspergillosis.
Age, median years (IQR) Male gender Case classification Proven Probable Disseminated aspergillosis Underlying diseases
ANC, median count (IQR) Time from onset of neutropenia to diagnosis of IPA, median days (IQR) Time from transplantation to diagnosis of IPA, median days (IQR) Use of corticosteroidb Use of immunosuppressantc RRT before IPA Mechanical ventilation Serum GM assays Serum GM, positive Median serum GM value (IQR) BAL GM assays BAL GM positive Median BAL GM value (IQR) Positive Aspergillus culture Prophylactic or empirical antifungal therapy before diagnosis of IPA Initial amphotericin-B Initial voriconazole 1-month mortality 3-month mortality
Transplant group (n Z 50)
Neutropenic group (n Z 60)
52 (42e60) 39 (58)
51 (42e62) 34 (57)
4 (8) 46 (92) 6 (12) Liver transplant 33(65) HSCT 10a (20) Kidney transplant 4 (8) Heart transplant 3(8) 4347 (2415e7635) Not applicable
6 (10) 54 (90) 4 (7) Acute leukemia 50 (83) Aplastic anemia 6 (10) Lymphoma 4 (7)
54 (26e156)
Not applicable
31 46 27 27
27 (45) 12 (20) 8 (13) 13 (22)
(61) (90) (53) (53)
0 (0e90) 28 (13e48)
p-value 0.43 0.88 0.69
0.35
<0.001
0.10 <0.001 <0.001 0.001
37/49 (76) 0.76 (0.61e1.59)
47/59 (80) 0.73 (0.65e2.88)
0.65 0.36
11/19 (60) 4.66 (0.78e10.00) 16/41 (39) 40 (80)
4/7 (57) 0.67 (0.55e1.34) 8/33 (24) 53 (88)
0.97 0.07 0.20 0.15
15 33 13 22
26 33 13 25
0.15 0.24 0.64 0.88
(30) (66) (26) (43)
(43) (55) (22) (42)
Data are no. (%) of patients, unless otherwise indicated. IQR, interquartile range; HSCT, hematopoietic stem cell transplantation; ANC, absolute neutrophil count; IPA, invasive pulmonary aspergillosis; RRT, renal replacement therapy; GM, galactomannan; BAL, bronchoalveolar lavage. a Of l0 HSCT recipients, 8 received allogeneic and 2 autologous HSCT. b Corticosteroid users are defined as use of corticosteroids at a mean minimum dose of 0.3 mg/kg/day of prednisolone equivalent for >3 weeks. c Treatment with other recognized T cell immunosuppressants, such as cyclosporine, TNF-a blockers, specific monoclonal antibodies (such as alemtuzumab), or nucleoside analogs during the previous 90 days.
Radiological features of IPA in non-neutropenic transplant recipients and neutropenic patients
were more likely to exhibit infarct-shaped consolidations compared with HSCT recipients (50% vs 10%, p Z 0.03).
The most common chest CT finding was consolidation-ormass (75; 68%). Compared with neutropenic patients, nonneutropenic transplant recipients were less likely to exhibit consolidation-or-mass and halo signs (Table 2). Angioinvasive form was also less frequent in non-neutropenic transplant recipients than in neutropenic patients (32% vs 60%, p Z 0.003), and the undetermined form was more common in the non-neutropenic transplant recipients (58% vs 20%, p < 0.001). Multivariate analysis revealed that the halo sign was independently associated with neutropenic patients (OR 2.9; 95% CI 1.28e6.67; p Z 0.02). When we performed subgroup analysis including the nonneutropenic transplant recipients (n Z 50), SOT recipients
Outcomes and prognostic factors Mortality rates among the non-neutropenic transplant recipients and neutropenic patients were 43% and 42%, respectively (p Z 0.88). We analyzed the risk factors for 90-day mortality in the overall patient population (Table 3). Multivariate analysis revealed macronodules (HR 0.31; 95% CI 0.16e0.61; p Z 0.001), multiple infarctshaped consolidations (HR 4.26; 95% CI 2.05e8.87; p < 0.001), renal replacement therapy (HR 5.62; 95% CI 2.80e11.25; p < 0.001) and the persistence of a positive serum GM (HR 7.14; 95% CI 3.33e14.28; p < 0.001) were independently associated with 90-day mortality. The survival
452
S.Y. Park et al.
Table 2 Computed tomography findings in invasive pulmonary aspergillosis in non-neutropenic transplant recipients and neutropenic patients. Image findingsa
Total number (n Z 110)
Transplant group (n Z 50)
Neutropenic group (n Z 60)
Consolidation-or-massb Single Multiple Macronodule (1 cm in diameter)c Single Multiple Consolidation, infarct-shaped Single Multiple Halo sign Ground glass opacity Micronodules (<1 cm in diameter) Small airway lesions other than micronodulesd Bronchiectasis or wall thickening Pleural effusion Pattern of necrotizing pneumonia Internal low attenuation signe Air-crescent sign Cavityf Types Angio-invasive type Angio-invasive formg Combined formh Non-angio-invasive type Airway-invasive formi Undetermined formj
75 44 31 72 33 38 56 32 24 46 31 35 10
(68) (40) (28) (65) (30) (35) (51) (29) (22) (42) (28) (32) (9)
28 14 14 29 16 13 21 10 11 13 16 12 3
(56) (28) (28) (58) (32) (26) (42) (20) (22) (26) (31) (24) (6)
47 30 17 43 17 25 35 22 13 33 15 23 7
(78) (50) (28) (72) (28) (42) (58) (37) (22) (55) (25) (38) (12)
0.002 0.46 0.11 0.30
15 39 33 13 7 24
(14) (35) (30) (12) (6) (22)
5 21 14 3 2 13
(12) (41) (28) (6) (4) (26)
10 18 19 10 5 11
(17) (30) (32) (17) (8) (18)
0.31 0.19 0.68 0.08 0.34 0.36
59 52 7 51 10 41
(54) (47) (6) (46) (9) (37)
19 16 3 31 2 29
(38) (32) (6) (62) (4) (58)
40 36 4 20 8 12
(67) (60) (7) (39) (13) (20)
0.003 0.003 >0.99 0.003 0.11 <0.001
p-value 0.01
0.13
0.09
Data are no. (%) of patients, unless otherwise indicated. a Patients may have 1 type of lesion. b Includes infract-shaped consolidations. c Includes macronodules with or without halo sign. d Includes mosaic attenuation, air trapping, tree-in-bud, and peribronchial opacity. e Internal low attenuation area in consolidation-or macronodule. f Includes cavitary lesions with or without air-crescent signs. g Angio-invasive form was designated when the predominant CT image fulfilled two of the three following findings (1. a halo sign, 2. an infarct-shaped consolidation, 3. an internal low attenuation, cavity, or air-crescent sign). h Combined type was designated when CT findings included both angio-invasive and airway-invasive forms. i Airway-invasive form was designated when the predominant CT findings revealed two of the three following findings (1. small airway lesions, 2. a peribronchial consolidation, 3. a bronchiectasis). j Undetermined type was designated when the pattern did not conform to the above criteria.
curves, stratified according to macronodules or multiple infarct-shaped consolidations, are presented in Fig. 3. The halo sign (HR 1.55; 95% CI 0.89e2.75; p Z 0.13) was not associated with 90-day mortality. In subgroup analysis of the 60 neutropenic patients, the halo sign (HR 1.52; 95% CI 0.67e3.45; p Z 0.31) was also not associated with 90-day mortality.
Discussion Although the clinical and radiological features of IPA in patients with neutropenia have been relatively well studied,4,8e11 comparisons between the image findings in IPA in non-neutropenic transplant recipients and neutropenic patients are rare.13 To our knowledge, this is the largest study
that systematically compares CT findings of IPA in nonneutropenic transplant recipients and neutropenic patients. Our findings reveal that non-neutropenic transplant recipients with IPA were less likely to exhibit consolidationor-mass and angio-invasive forms of IPA such as a halo signs than neutropenic patients. Our findings also suggest that certain radiologic findings such as macronodules and multiple infarct-shaped consolidations have prognostic implications in patients with IPA. In the present study, the most common radiologic findings in patients with IPA were consolidation-or-mass (68%) and macronodule (65%). Therefore, these radiologic findings are helpful for the diagnosis of IPA. However, the previous study reported that the vast majority of patients (95%) with IPA presented with macronodule,9 whereas only 65% of patients with IPA revealed macronodule in the
CT findings in invasive pulmonary aspergillosis Table 3
453
Univariate analysis of factors associated with 90-day mortality in 110 patients with invasive pulmonary aspergillosis.
Variable
Death (n Z 46)
Survival (n Z 64)
Hazard Ratio (95% CI)
p-value
Age, >65 years Male gender Proven IPA Disseminated aspergillosis Neutropenia Use of corticosteroid Uncontrolled underlying disease RRT before IPAa Persistence of a positive serum galactomannana Consolidation-or-mass Macronodulea Multiple infarct-shaped consolidationsa Halo sign Ground glass opacity Micronodules Internal low attenuation sign Cavitary lesion Air-crescent sign Angio-invasive form Airway-invasive form Initial amphotericin-B Initial voriconazole
10 29 3 6 22 27 32 26 18/28 37 28 17 24 17 17 5 8 3 23 4 14 19
7 35 7 4 31 31 36 9 8/45 38 46 11 22 14 19 9 16 4 29 6 11 33
1.98 0.75 1.47 1.63 0.95 1.24 1.45 3.98 5.14 2.08 0.52 2.08 1.55 1.87 1.15 0.79 0.63 0.94 1.13 0.97 1.80 0.69
0.06 0.34 0.52 0.26 0.85 0.46 0.24 <0.001 <0.001 0.04 0.03 0.02 0.13 0.04 0.65 0.61 0.24 0.92 0.68 0.96 0.07 0.22
(21) (62) (6) (13) (47) (57) (68) (55) (64) (79) (55) (41) (51) (37) (36) (11) (17) (6) (50) (9) (30) (41)
(11) (55) (11) (6) (48) (48) (56) (14) (18) (59) (72) (21) (34) (22) (30) (14) (25) (6) (45) (9) (17) (52)
(0.98e3.40) (0.42e1.35) (0.46e4.74) (0.69e3.85) (0.53e1.68) (0.70e2.22) (0.78e2.68) (2.23e7.12) (2.36e11.19) (1.03e4.19) (0.29e0.93) (1.12e3.86) (0.89e2.75) (1.03e3.40) (0.63e2.08) (0.31e1.99) (0.30e1.35) (0.29e3.03) (0.63e2.01) (0.35e2.72) (0.96e3.37) (0.38e1.25)
Data are no. (%) of patients, unless otherwise indicated. CI, confidence interval; IPA, invasive pulmonary aspergillosis; RRT, renal replacement therapy; GM, galactomannan. a Independent risk factors for 90-day mortality (p < 0.05 in Cox proportional hazard model).
current study. The relatively low proportion of macronodule in our study may be due to the high proportion of transplant recipients (45%) in our population, since these patients can present with variable CT findings.6,13,15 Therefore, the presence of macronodule may not be useful to exclude a diagnosis of IPA especially in SOT recipients. However, the presence of macronodule and/or consolidation-or-mass retained a high sensitivity (98%). Thus, the absence of these two CT findings might be helpful to exclude a diagnosis of IPA. The patterns of IPA in patients with neutropenia are characterized by prominent fungal angioinvasion, intraalveolar hemorrhage, tissue infarction, and minimal inflammatory cell response.7,23 In patients with immunosuppression but not myelosuppression, the histologic pattern is different, and is characterized by central liquefaction necrosis, fewer hyphae aligned in a radical pattern around the lesion, and prominent neutrophil infiltration.7,23 In the present study, angio-invasive forms (32% vs 60%, p Z 0.003) and halo signs (26% vs 55%, p Z 0.002) were more common in the patients with neutropenia, which agrees with previous reports.13,19,20 However, although our preliminary data suggested that the airway-invasive form of IPA was more common in SOT recipients than neutropenic patients,13 we did not find a significant difference in airway-invasive IPA between the nonneutropenic transplant recipients and neutropenic patients in the present work. One possible explanation may be the use of an earlier version of the EORTC criteria for IPA24 in the previous study.13 In the present analysis we used the modified EORTC/MSG criteria,1 so that patients who did not fulfill the clinical criteria (one of the following 3 signs on CT: a well-
defined dense lesion, air-crescent sign, or cavity) would not have been enrolled. Historically, nodular lesions with halo or air-crescent signs were associated with IPA.8,9,25 These imaging findings are commonly used to make a presumptive diagnosis and initiate empirical treatment for IPA in high risk groups (eg. neutropenic fever in patients with hematologic malignancy). Although the revised EORTC/MSG criteria for host factors include non-neutropenic immunosuppressed patients with SOT, hereditary immunodeficiencies, and connective tissue disorders,1 the CT findings of IPA have not been fully validated in this group. We showed that the non-angio-invasive types of IPA such as the airwayinvasive or undeterminate forms were more common in non-neutropenic transplant recipients than in neutropenic patients. Therefore, because some patients, such as nonneutropenic transplant recipients, may not meet the radiologic criteria of EORTC/MSG,1 radiologic findings other than prespecified radiologic criteria such as EORTC/MSG would be included in the clinical criteria for IPA. Recently, Nucci et al. also proposed adding GM assay-supported cases with nonspecific radiologic findings to the EORTC/MSG criteria.26 We found that serum GM titers, macronodules, multiple infarct-shaped consolidations, and renal replacement therapy, were significantly associated with 90-day mortality. Previous reports noted a strong correlation between serum GM index and IPA survival outcomes.13,18,19 The need for renal replacement therapy is also a clearly established risk factor for mortality from IPA.2,27 However, macronodules and multiple infarct-shaped consolidations have
454
A
S.Y. Park et al.
Presence of macronodules Absence of macronodules 0.31 0.16–0.61
B Absence of multiple infarctshaped consolidations Presence of multiple infarctshaped consolidations (HR 4.26, 95% CI 2.05-8.87)
Figure 3 Estimated survival curves for patients with invasive pulmonary aspergillosis according to the presence of macronodules (A) and multiple infarct-shaped consolidations (B) based on the Cox proportional hazard model.
not been described as clinical predictors associated with outcomes of IPA. Macronodule are one of the main findings of early IPA, and they are such a common feature on initial CT that their absence argues against the likelihood of IPA.9,11 Thus, early initiation of effective antifungal treatment for IPA, based on the finding of macronodules, may be associated with a significantly improved response to treatment, and with improved survival. However, we could not discover any evidence of a relationship between halo
signs and higher survival in previous reports.4,9,28 This may be explained by the fact that most of the patients (86%) included in the previous studies had underlying hematological malignancies,9 whereas we included a relatively large number (45%) of non-neutropenic transplant recipients. Since, in our study, only 26% of the nonneutropenic transplant recipients with IPA presented with halo signs, the halo sign may not have prognostic implications associated with early detection of IPA specifically in non-neutropenic transplant recipients. Another interesting result of our study is that the presence of multiple infarct-shaped consolidations on baseline CT was a significant predictor of poor outcome. The presence of multiple infarct-shaped consolidations may point to late diagnosis of the IPA, since multiple nodules gradually coalesce into areas of consolidation29 and these consolidations may be complicated by infarctions caused by angioinvasion by Aspergillus. This study had several limitations. First, some may object that the transplant groups were composed of recipients of both SOT and HSCT. A recent review of autopsy-proven IPA revealed that a histopathology characterized by diffuse inflammation and limited numbers of Aspergillus hyphae was more common in HSCT recipients with GVHD who were treated with glucocorticoids than in neutropenic patients.14 Thus, the recipients of SOT and HSCT after primary engraftment included in the present study were both glucocorticoid-immunosuppressed patients and developed a form of IPA that was pathologically distinct from that in neutropenic patients.7,30 Second, our study shares the limitations of retrospective studies and suffers from a low autopsy rate, a trend noted worldwide.14 Only 10 proven IPA cases were included in our study and no autopsies were performed on the patients who died. Some of the neutropenic patients with IPA also had thrombocytopenia, and allograft dysfunction occurred in 53% of the transplant recipients, especially the liver transplant recipients, at the time of IPA diagnosis. The patients with hepatic allograft dysfunction had severe coagulopathy and were often markedly thrombocytopenic. Thus, these patients could not have tolerated invasive procedures such as transbronchial lung biopsy. Third, our study was limited by a relative insensitivity of Aspergillus species culture,31 making the diagnosis of IPA dependent on serum and BAL GM assays. This might be explained by the fact that prophylactic or empirical antifungal agents had already been administered to 85% of the patients (93/110).13,17 In addition, these antifungal use prior to the diagnosis of IPA may affect the CT findings. Indeed, the subgroup analysis including the patients (n Z 17) without antifungal agents prior to the diagnosis of IPA showed that the most common finding was macronodule (77%) while consolidation-or-mass and infarct-shaped consolidation were observed in only 35% of patients, respectively. The relative high proportion of patients who revealed macronodules in their CT findings is consistent with previous data which showed that 95% of neutropenic patients revealed macronodule in their CT findings.9 It remains to be settled on the influence of prophylactic or empirical antifungal agents on radiologic findings of IPA. Finally, lung transplant recipients were not included in the present study and this might attenuate differences in radiologic
CT findings in invasive pulmonary aspergillosis findings between non-neutropenic transplant recipients and neutropenic patients.32 In conclusion, our data suggest that chest CT findings in non-neutropenic transplant recipients with IPA are similar to those in neutropenic patients with IPA, except that consolidation-or-mass, halo signs, and angio-invasive form are less common in non-neutropenic transplant recipients. So, nonspecific radiological findings might not exclude a diagnosis of IPA radiological findings in non-neutropenic transplant recipients. Our findings also suggest that certain CT findings in patients with IPA, such as macronodules and multiple infarct-shaped consolidations, have prognostic implications. Thus, when we observed multiple infarctshaped consolidations in patients with IPA, aggressive antifungal treatment is needed to overcome the poor prognosis.
455
7.
8.
9.
10.
Funding 11.
No author received financial support.
Conflict of interest
12.
There are no potential conflicts of interest for any authors.
13.
Acknowledgments
14.
We thank Ji-Eun Hyun for administrative support. We are grateful to Julian Gross for assistance in language editing. 15.
Appendix. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jinf.2011.08. 007.
References 1. De Pauw B, Walsh TJ, Donnelly JP, Stevens DA, Edwards JE, Calandra T, et al. Revised definitions of invasive fungal disease from the European organization for research and treatment of cancer/invasive fungal infections cooperative group and the national institute of allergy and infectious diseases mycoses study group (EORTC/MSG) consensus group. Clin Infect Dis 2008;46(12):1813e21. 2. Singh N, Paterson DL. Aspergillus infections in transplant recipients. Clin Microbiol Rev 2005;18(1):44e69. 3. Denning DW. Therapeutic outcome in invasive aspergillosis. Clin Infect Dis 1996;23(3):608e15. 4. Caillot D, Casasnovas O, Bernard A, Couaillier JF, Durand C, Cuisenier B, et al. Improved management of invasive pulmonary aspergillosis in neutropenic patients using early thoracic computed tomographic scan and surgery. J Clin Oncol 1997; 15(1):139e47. 5. Wahba H, Truong MT, Lei X, Kontoyiannis DP, Marom EM. Reversed halo sign in invasive pulmonary fungal infections. Clin Infect Dis 2008;46(11):1733e7. 6. Milito MA, Kontoyiannis DP, Lewis RE, Liu P, Mawlawi OR, Truong MT, et al. Influence of host immunosuppression on ct
16.
17.
18.
19.
20.
21.
22.
23.
findings in invasive pulmonary aspergillosis. Med Mycol 2010; 48(6):817e23. Berenguer J, Allende MC, Lee JW, Garrett K, Lyman C, Ali NM, et al. Pathogenesis of pulmonary aspergillosis. Granulocytopenia versus cyclosporine and methylprednisolone-induced immunosuppression. Am J Respir Crit Care Med 1995;152(3): 1079e86. Caillot D, Couaillier JF, Bernard A, Casasnovas O, Denning DW, Mannone L, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic computed tomography scans in patients with neutropenia. J Clin Oncol 2001;19(1):253e9. Greene RE, Schlamm HT, Oestmann JW, Stark P, Durand C, Lortholary O, et al. Imaging findings in acute invasive pulmonary aspergillosis: clinical significance of the halo sign. Clin Infect Dis 2007;44(3):373e9. Gerson SL, Talbot GH, Hurwitz S, Strom BL, Lusk EJ, Cassileth PA. Prolonged granulocytopenia: the major risk factor for invasive pulmonary aspergillosis in patients with acute leukemia. Ann Intern Med 1984;100(3):345e51. Orr DP, Myerowitz RL, Dubois PJ. Patho-radiologic correlation of invasive pulmonary aspergillosis in the compromised host. Cancer 1978;41(5):2028e39. Park YS, Seo JB, Lee YK, Do KH, Lee JS, Song JW, et al. Radiological and clinical findings of pulmonary aspergillosis following solid organ transplant. Clin Radiol 2008;63(6):673e80. Park SY, Kim SH, Choi SH, Sung H, Kim MN, Woo JH, et al. Clinical and radiological features of invasive pulmonary aspergillosis in transplant recipients and neutropenic patients. Transpl Infect Dis 2010;12(4):309e15. Chamilos G, Luna M, Lewis RE, Bodey GP, Chemaly R, Tarrand JJ, et al. Invasive fungal infections in patients with hematologic malignancies in a tertiary care cancer center: an autopsy study over a 15-year period (1989e2003). Haematologica 2006;91(7):986e9. Kojima R, Tateishi U, Kami M, Murashige N, Nannya Y, Kusumi E, et al. Chest computed tomography of late invasive aspergillosis after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005;11(7): 506e11. Park SY, Lee SO, Choi SH, Sung H, Kim MN, Choi CM, et al. Aspergillus galactomannan antigen assay in bronchoalveolar lavage fluid for diagnosis of invasive pulmonary aspergillosis. J Infect 2010;61(6):492e8. Park SY, Lee SO, Choi SH, Jeong JY, Sung H, Kim MN, et al. Serum and bronchoalveolar lavage fluid galactomannan assays in patients with pulmonary aspergilloma. Clin Infect Dis 2011; 52(7):e149e52. Woods G, Miceli MH, Grazziutti ML, Zhao W, Barlogie B, Anaissie E. Serum aspergillus galactomannan antigen values strongly correlated with outcome of invasive aspergillosis. A study of 56 patients with hematologic cancer. Cancer 2007; 110:830e4. Maertens J, Buve ´ K, Theunissen K, Meersseman W, Verbeken E, Verhoef G, et al. Galactomaannan serves as surrogate endpoint for outcome of pulmonary invasive aspergillosis in neutropenic hematology patients. Cancer 2009;115:355e62. Hansell DM, Bankier AA, MacMahon H, McLoud TC, Muller NL, Remy J. Fleischner society: glossary of terms for thoracic imaging. Radiology 2008;246(3):697e722. Logan PM, Primack SL, Miller RR, Muller NL. Invasive aspergillosis of the airways: Radiographic, CT, and pathologic findings. Radiology 1994;193(2):383e8. Logan PM, Muller NL. High-resolution computed tomography and pathologic findings in pulmonary aspergillosis: a pictorial essay. Can Assoc Radiol J 1996;47(6):444e52. Stergiopoulou T, Meletiadis J, Roilides E, Kleiner DE, Schaufele R, Roden M, et al. Host-dependent patterns of tissue
456
24.
25.
26.
27.
S.Y. Park et al. injury in invasive pulmonary aspergillosis. Am J Clin Pathol 2007;127(3):349e55. Ascioglu S, Rex JH, de Pauw B, Bennett JE, Bille J, Crokaert F, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis 2002;34(1):7e14. Neofytos D. Chest computed tomography versus serm galactomannan enzyme immunoassay for the diagnosis of probable invasive pulmonary aspergillosis: to be decided. Clin Infect Dis 2010;51(11):1281e3. Nucci M, Nouer SA, Grazziutti M, Kumar NS, Barlogie B, Anaissie E. Probable invasive aspergillosis without prespecified radiologic findings: Proposal for inclusion of a new category of aspergillosis and implications for studying novel therapies. Clin Infect Dis 2010;51(11):1273e80. Baddley JW, Andes DR, Marr KA, Kontoyiannis DP, Alexander BD, Kauffman CA, et al. Factors associated with mortality in transplant patients with invasive aspergillosis. Clin Infect Dis 2010;50(12):1559e67.
28. Horger M, Hebart H, Einsele H, Lengerke C, Claussen CD, Vonthein R, et al. Initial CT manifestations of invasive pulmonary aspergillosis in 45 non-hiv immunocompromised patients: association with patient outcome? Eur J Radiol 2005;55(3):437e44. nchez F, Franquet T, 29. Hidalgo A, Parody R, Martino R, Sa Gime ´nez A, et al. Correlation between high-resolution computed tomography and galactomannan antigenemia in adult hematologic patients at risk for invasive aspergillosis. Eur J Radiol 2009;71(1):55e60. 30. Lewis RE, Kontoyiannis DP. Invasive aspergillosis in glucocorticoidtreated patients. Med Mycol 2009;47(Suppl 1):S271e81. 31. Cornillet A, Camus C, Nimubona S, Gandemer V, Tattevin P, Belleguic C, et al. Comparison of epidemiological, clinical, and biological features of invasive aspergillosis in neutropenic and nonneutropenic patients: a 6-year survey. Clin Infect Dis 2006;43(5):577e84. 32. Husain S, Mooney ML, Danziger-Isakov L, Mattner F, Singh N, Avery R, et al. A 2010 working formulation for the standardization of definitions of infections in cardiothoracoc transplant recipients. J Heart Lung Transplant 2011;30(4):361e74.
www.elsevierhealth.com/journals/jinf
Computed tomography findings in invasive pulmonary aspergillosis in non-neutropenic transplant recipients and neutropenic patients, and their prognostic value* Seong Yeon Park a,c,d, Chaehun Lim b,d, Sang-Oh Lee a, Sang-Ho Choi a, Yang Soo Kim a, Jun Hee Woo a, Jae-Woo Song b, Mi Young Kim b, Eun Jin Chae b, Kyung-Hyun Do b, Koun-Sik Song b, Joon Beom Seo b,**,e, Sung-Han Kim a,*,e a
Departments of Infectious Diseases, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap-dong, Songpa-gu, Seoul 138-736, Republic of Korea b Department of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea c Department of Internal Medicine, Dongguk University International Hospital, Dongguk University College of Medicine, Seoul, Republic of Korea Accepted 10 August 2011 Available online 16 August 2011
KEYWORDS Invasive pulmonary aspergillosis; Non-neutropenic transplant recipients; Neutropenic patients
Summary Objectives: We evaluated CT findings and their prognostic value in nonneutropenic transplant recipients with invasive pulmonary aspergillosis (IPA) compared with neutropenic patients with IPA. Methods: All adult patients during a 27-month who met the criteria for proven or probable IPA according to the 2008 EORTC/MSG criteria were retrospectively enrolled. Initial CT findings were reviewed by two radiologists blinded to the patients’ demographics and clinical outcomes. Results: A total of 50 non-neutropenic transplant recipients and 60 neutropenic patients were enrolled. Consolidation-or-mass, halo signs, and angio-invasive form were observed less often in non-neutropenic transplant recipients than in neutropenic patients: (56%, 26%, and 32%) versus (78%, 55%, and 60%, p Z 0.01, p Z 0.002, and p Z 0.003, respectively). Multivariate analysis revealed that macronodules (HR 0.31, p Z 0.001), multiple infarct-shaped consolidations
*
Trial Registration: ClinicalTrials.gov Identifier: NCT001178177. * Corresponding author. Tel.: þ82 2 3010 3305; fax: þ82 2 3010 6970. ** Corresponding author. Tel.: þ82 2 3010 4383; fax: þ82 2 3010 8127. E-mail addresses: [email protected] (J.B. Seo), [email protected] (S.-H. Kim). d S. Y. Park and C. Lim contributed equally to the study. e S-H Kim was responsible for the clinical aspects of the study and J B Seo for the radiologic aspects. 0163-4453/$36 ª 2011 The British Infection Association. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jinf.2011.08.007
448
S.Y. Park et al. (HR 4.26, p < 0.001), renal replacement therapy (HR 5.62, p < 0.001) and persistence of a positive serum galactomannan (HR 7.14, p < 0.001) were independently associated with 90-day mortality. Conclusions: Our data indicate that CT findings in non-neutropenic transplant recipients with IPA are similar to those in neutropenic patients with IPA except that consolidation-or-mass, halo sings, and angio-invasive form are less frequent, and certain CT findings such as macronodules and multiple infarct-shaped consolidations have prognostic implications in IPA. ª 2011 The British Infection Association. Published by Elsevier Ltd. All rights reserved.
Introduction Invasive pulmonary aspergillosis (IPA) is an opportunistic infection that occurs mainly in patients with neutropenia and in hematopoietic stem cell and solid organ transplant recipients.1,2 The survival of these patients depends on early diagnosis and prompt initiation of therapeutic measures.3,4 However, early diagnosis of IPA based on culture and/or microscopy of histopathologic specimens is generally difficult because obtaining tissue biopsies to permit definite diagnosis is not without risk in critically ill patients due to coagulation abnormalities and thrombocytopenia.5 Thus, early diagnosis of IPA relies heavily on imaging modes such as chest computed tomography (CT).1 These are used to differentiate fungal from bacterial and viral pneumonia, because early antifungal therapy improves survival.4 The basis of the lung injury caused by Aspergillus in patients with neutropenia as the predominant mechanism of immunosuppression may differ from that in patients who have numerically adequate but functionally-impaired phagocytes.6,7 Although the clinical and radiological features of IPA in patients with neutropenia have been relatively well studied,4,8e11 little is known about the characteristics of IPA in non-neutropenic transplant recipients.12,13 We therefore evaluated CT findings and their prognostic value in non-neutropenic transplant recipients versus neutropenic patients with IPA.
Methods Study population and data collection This study was performed at the Asan Medical Center, a 2700-bed tertiary care teaching hospital in Seoul, South Korea, from January 2008 to March 2010. It was approved by the Institutional Review Board at Asan Medical Center (approval number 2010-0460). The study patients were classified as either nonneutropenic transplant recipients or neutropenic patients. Non-neutropenic transplant recipients were defined as those patients with IPA who underwent solid organ transplantation (SOT) or hematopoietic stem cell transplantation (HSCT),6,14,15 and there were no evidence of neutropenia at the time of diagnosis of IPA. Neutropenic patients were defined as those patients with IPA who received chemotherapy for acute leukemia, lymphoma, or a plastic anemia while having an absolute neutrophil count <500 cells/mm3 within 30 days prior to diagnosis of IPA.1 HSCT recipients with IPA who did not achieve primary
engraftment, or who, after engraftment, suffered from neutropenia related to recurrence of underlying diseases, were classified as neutropenic patients.15 Corticosteroid use was defined as use of a mean minimum dose of 0.3 mg/kg/day of prednisolone equivalent for >3 weeks.1 We excluded from our analysis patients with mixed pulmonary infections (i.e., fungal other than Aspergillus spp., bacterial, and viral co-infections). During the first 6 months after SOT, all liver transplant recipients (about 300 cases per year) received oral itraconazole or low dose liposomal amphotericin (1 mg/kg/day), as fungal prophylaxis. However, kidney (about 210 cases per year) and heart transplant patients (about 30 cases per year) did not receive any routine antifungal prophylaxis due to their lower incidence of IPA.2 Only one lung transplant was performed in our hospital during the study period. HSCT recipients (about 120 allogeneic HSCT and 40 autologous HSCT cases per year), received micafungin as fungal prophylaxis from day þ1 to ANC 3000 cells/mm3 after engraftment. However, neutropenic patients who were given cytotoxic chemotherapy did not receive any routine antifungal prophylaxis. Serum galactomannan (GM) antigen levels were measured as described previously.13,16,17 The decision to perform fiberoptic bronchoscopy with bronchoalveolar lavage (BAL) was at the discretion of each attending physician. Fiberoptic bronchoscopy was performed using four 50 mL aliquots of sterile saline solution. An optical density ratio of 0.5 or greater in serum or BAL samples was considered positive for GM.13,16,17 Positive GM levels in patients receiving piperacillin-tazobactam, amoxicillin-clavulanate or ampicillin-sulbactam before serum or BAL GM assay were excluded because these antibiotics can cause false-positive results in GM assays.16,17 Serum GM antigen levels in liver transplant recipients, HSCT recipients and neutropenic patients were measured once or twice in a week by ELISA. Serum GM was not regularly measured in other patients, although the assay was performed whenever IPA was suspected. Median follow-up duration of GM antigen level was 46 days (interquartile range [IQR] 16.5e88.0 days). GM antigen level was checked median 7 times (IQR, 3e11 times). We defined the persistence of a positive serum GM as serial GM monitoring did not revealed negative conversion within 3 months after the diagnosis of IPA.18,19
Case definitions of IPA Patients were assigned a proven or probable diagnosis of IPA according to the consensus definition of the EORTC/MSG and our previous report.1,13,16 Proven IPA was defined by
CT findings in invasive pulmonary aspergillosis histologic evidence of tissue invasion, including septated, acutely branching filamentous fungi and positive culture. Probable IPA was defined as the presence of a host factor, together with clinical indications such as dense, wellcircumscribed lesions with or without a halo sign, aircrescent sign, or cavity on CT; and mycologic evidence of fungal infection (by culture or cytologic analysis of BAL fluid for Aspergillus species or GM assay of serum or BAL). Possible IPA was defined as the presence of a host factor and clinical criteria without mycologic evidence for IPA. Cases that were classified as possible IPA were not included in this analysis.
Radiologic definitions and interpretation categories Chest CT scans with or without contrast enhancement were performed whenever IPA was suspected. Baseline chest CT findings were reviewed by two independent
449 radiologists unaware of patients’ characteristics and clinical outcomes. Where there was disagreement the final decision was made by consensus after discussion. The radiologists used a standard glossary of CT imaging definitions to categorize pulmonary lesions (Supplementary Table).20 Radiological patterns of IPA were classified as either angio-invasive or airway-invasive form, based on previous studies.21,22 A radiologic pattern was considered angioinvasive form when the predominant CT imaging fulfilled two of the three following findings: 1. a halo sign, 2. an infarct-shaped consolidation, 3. an internal low attenuation, cavity, or air-crescent sign (Fig. 1). It was considered airway-invasive form when the predominant CT fulfilled two of the three following findings following findings: 1. small airway lesions, 2. a peribronchial consolidation, 3. a bronchiectasis (Fig. 2). It was classified as a combined form was considered when CT finding met
Figure 1 Typical diagnostic findings of angio-invasive form. (A) CT scan with lung window setting shows subsegmental, triangular consolidation with the base abutting the pleura and the apex directed toward the hilum (arrows), characterized by infract-shaped consolidation. And ground glass attenuation is observed around consolidation, suggesting of CT halo sign (arrowheads). (B) Internal low attenuated lesion is noted within consolidation in left lower lobe. It is necrotic portion of consolidation and precursor of cavity. (C) Cavity is defined as a gas-filled space, seen as a lucency or low attenuation area within consolidation (double arrow). (D) Aircrescent sign is a collection of air in a crescentic shape that separates the wall of a cavity from an inner consolidation (black arrowheads).
450
S.Y. Park et al.
Figure 2 Typical diagnostic findings of airway-invasive form. (A, B) CT scan with lung window setting shows small, clustered or centrilobular micronodules and branching-linear opacities (white arrows) with mild bronchiectasis (arrowheads). (C) Peribronchial opacity or consolidation is classified as the manifestation of airway-invasive form (arrow).
both angio-invasive and airway-invasive form and it was designated undetermined form when the CT image did not fulfill any of the above criteria. To simply the patterns, we re-assigned these forms into two types, angioinvasive and non-angio-invasive types. The angio-invasive type was defined as the angio-invasive or combined form, and the non-angio-invasive type as the airwayinvasive or undetermined form.
Statistical analysis All statistical analyses were performed with SPSS version 14.0 (SPSS, Chicago, IL). Categorical variables were compared using the c2 or Fisher’s exact test, and continuous variables were compared using the ManneWhitney U test. Univariate and multivariate analyses for factors associated with time to death were performed using Cox proportional hazards models. All variables with p-value <0.10 by univariate analysis were included in a multivariate analysis. A p-value <0.05 were considered statistically significant.
Results Patient characteristics During the study period, a total of 112 patients comprising 52 transplant recipients (42 SOT and 10 HSCT) and 60 neutropenic patients were diagnosed with IPA. Of these, two SOT recipients who revealed neutropenia at the diagnosis of IPA were excluded from the analysis. Finally, the 110 patients with IPA (10 histologically proven and 100 probable cases) were analyzed. The characteristics of the overall study patients are shown in Table 1. Of the 40 SOT recipients, 9 (23%) received immunosuppression-pulse therapy because of chronic or acute rejection, and 21 (53%) patients had graft dysfunction at the time of IPA. Among l0 HSCT recipients (8 allogeneic and 2 autologous HSCT), 7 (70%) had graft versus host disease (GVHD). Of the 60 neutropenic patients with IPA, 15 (25%) received HSCT but did not achieve primary engraftment or suffered from neutropenia related to recurrence of underlying diseases after engraftment.
CT findings in invasive pulmonary aspergillosis
451
Table 1 Baseline clinical characteristics of non-neutropenic transplant recipients and neutropenic patients with invasive pulmonary aspergillosis.
Age, median years (IQR) Male gender Case classification Proven Probable Disseminated aspergillosis Underlying diseases
ANC, median count (IQR) Time from onset of neutropenia to diagnosis of IPA, median days (IQR) Time from transplantation to diagnosis of IPA, median days (IQR) Use of corticosteroidb Use of immunosuppressantc RRT before IPA Mechanical ventilation Serum GM assays Serum GM, positive Median serum GM value (IQR) BAL GM assays BAL GM positive Median BAL GM value (IQR) Positive Aspergillus culture Prophylactic or empirical antifungal therapy before diagnosis of IPA Initial amphotericin-B Initial voriconazole 1-month mortality 3-month mortality
Transplant group (n Z 50)
Neutropenic group (n Z 60)
52 (42e60) 39 (58)
51 (42e62) 34 (57)
4 (8) 46 (92) 6 (12) Liver transplant 33(65) HSCT 10a (20) Kidney transplant 4 (8) Heart transplant 3(8) 4347 (2415e7635) Not applicable
6 (10) 54 (90) 4 (7) Acute leukemia 50 (83) Aplastic anemia 6 (10) Lymphoma 4 (7)
54 (26e156)
Not applicable
31 46 27 27
27 (45) 12 (20) 8 (13) 13 (22)
(61) (90) (53) (53)
0 (0e90) 28 (13e48)
p-value 0.43 0.88 0.69
0.35
<0.001
0.10 <0.001 <0.001 0.001
37/49 (76) 0.76 (0.61e1.59)
47/59 (80) 0.73 (0.65e2.88)
0.65 0.36
11/19 (60) 4.66 (0.78e10.00) 16/41 (39) 40 (80)
4/7 (57) 0.67 (0.55e1.34) 8/33 (24) 53 (88)
0.97 0.07 0.20 0.15
15 33 13 22
26 33 13 25
0.15 0.24 0.64 0.88
(30) (66) (26) (43)
(43) (55) (22) (42)
Data are no. (%) of patients, unless otherwise indicated. IQR, interquartile range; HSCT, hematopoietic stem cell transplantation; ANC, absolute neutrophil count; IPA, invasive pulmonary aspergillosis; RRT, renal replacement therapy; GM, galactomannan; BAL, bronchoalveolar lavage. a Of l0 HSCT recipients, 8 received allogeneic and 2 autologous HSCT. b Corticosteroid users are defined as use of corticosteroids at a mean minimum dose of 0.3 mg/kg/day of prednisolone equivalent for >3 weeks. c Treatment with other recognized T cell immunosuppressants, such as cyclosporine, TNF-a blockers, specific monoclonal antibodies (such as alemtuzumab), or nucleoside analogs during the previous 90 days.
Radiological features of IPA in non-neutropenic transplant recipients and neutropenic patients
were more likely to exhibit infarct-shaped consolidations compared with HSCT recipients (50% vs 10%, p Z 0.03).
The most common chest CT finding was consolidation-ormass (75; 68%). Compared with neutropenic patients, nonneutropenic transplant recipients were less likely to exhibit consolidation-or-mass and halo signs (Table 2). Angioinvasive form was also less frequent in non-neutropenic transplant recipients than in neutropenic patients (32% vs 60%, p Z 0.003), and the undetermined form was more common in the non-neutropenic transplant recipients (58% vs 20%, p < 0.001). Multivariate analysis revealed that the halo sign was independently associated with neutropenic patients (OR 2.9; 95% CI 1.28e6.67; p Z 0.02). When we performed subgroup analysis including the nonneutropenic transplant recipients (n Z 50), SOT recipients
Outcomes and prognostic factors Mortality rates among the non-neutropenic transplant recipients and neutropenic patients were 43% and 42%, respectively (p Z 0.88). We analyzed the risk factors for 90-day mortality in the overall patient population (Table 3). Multivariate analysis revealed macronodules (HR 0.31; 95% CI 0.16e0.61; p Z 0.001), multiple infarctshaped consolidations (HR 4.26; 95% CI 2.05e8.87; p < 0.001), renal replacement therapy (HR 5.62; 95% CI 2.80e11.25; p < 0.001) and the persistence of a positive serum GM (HR 7.14; 95% CI 3.33e14.28; p < 0.001) were independently associated with 90-day mortality. The survival
452
S.Y. Park et al.
Table 2 Computed tomography findings in invasive pulmonary aspergillosis in non-neutropenic transplant recipients and neutropenic patients. Image findingsa
Total number (n Z 110)
Transplant group (n Z 50)
Neutropenic group (n Z 60)
Consolidation-or-massb Single Multiple Macronodule (1 cm in diameter)c Single Multiple Consolidation, infarct-shaped Single Multiple Halo sign Ground glass opacity Micronodules (<1 cm in diameter) Small airway lesions other than micronodulesd Bronchiectasis or wall thickening Pleural effusion Pattern of necrotizing pneumonia Internal low attenuation signe Air-crescent sign Cavityf Types Angio-invasive type Angio-invasive formg Combined formh Non-angio-invasive type Airway-invasive formi Undetermined formj
75 44 31 72 33 38 56 32 24 46 31 35 10
(68) (40) (28) (65) (30) (35) (51) (29) (22) (42) (28) (32) (9)
28 14 14 29 16 13 21 10 11 13 16 12 3
(56) (28) (28) (58) (32) (26) (42) (20) (22) (26) (31) (24) (6)
47 30 17 43 17 25 35 22 13 33 15 23 7
(78) (50) (28) (72) (28) (42) (58) (37) (22) (55) (25) (38) (12)
0.002 0.46 0.11 0.30
15 39 33 13 7 24
(14) (35) (30) (12) (6) (22)
5 21 14 3 2 13
(12) (41) (28) (6) (4) (26)
10 18 19 10 5 11
(17) (30) (32) (17) (8) (18)
0.31 0.19 0.68 0.08 0.34 0.36
59 52 7 51 10 41
(54) (47) (6) (46) (9) (37)
19 16 3 31 2 29
(38) (32) (6) (62) (4) (58)
40 36 4 20 8 12
(67) (60) (7) (39) (13) (20)
0.003 0.003 >0.99 0.003 0.11 <0.001
p-value 0.01
0.13
0.09
Data are no. (%) of patients, unless otherwise indicated. a Patients may have 1 type of lesion. b Includes infract-shaped consolidations. c Includes macronodules with or without halo sign. d Includes mosaic attenuation, air trapping, tree-in-bud, and peribronchial opacity. e Internal low attenuation area in consolidation-or macronodule. f Includes cavitary lesions with or without air-crescent signs. g Angio-invasive form was designated when the predominant CT image fulfilled two of the three following findings (1. a halo sign, 2. an infarct-shaped consolidation, 3. an internal low attenuation, cavity, or air-crescent sign). h Combined type was designated when CT findings included both angio-invasive and airway-invasive forms. i Airway-invasive form was designated when the predominant CT findings revealed two of the three following findings (1. small airway lesions, 2. a peribronchial consolidation, 3. a bronchiectasis). j Undetermined type was designated when the pattern did not conform to the above criteria.
curves, stratified according to macronodules or multiple infarct-shaped consolidations, are presented in Fig. 3. The halo sign (HR 1.55; 95% CI 0.89e2.75; p Z 0.13) was not associated with 90-day mortality. In subgroup analysis of the 60 neutropenic patients, the halo sign (HR 1.52; 95% CI 0.67e3.45; p Z 0.31) was also not associated with 90-day mortality.
Discussion Although the clinical and radiological features of IPA in patients with neutropenia have been relatively well studied,4,8e11 comparisons between the image findings in IPA in non-neutropenic transplant recipients and neutropenic patients are rare.13 To our knowledge, this is the largest study
that systematically compares CT findings of IPA in nonneutropenic transplant recipients and neutropenic patients. Our findings reveal that non-neutropenic transplant recipients with IPA were less likely to exhibit consolidationor-mass and angio-invasive forms of IPA such as a halo signs than neutropenic patients. Our findings also suggest that certain radiologic findings such as macronodules and multiple infarct-shaped consolidations have prognostic implications in patients with IPA. In the present study, the most common radiologic findings in patients with IPA were consolidation-or-mass (68%) and macronodule (65%). Therefore, these radiologic findings are helpful for the diagnosis of IPA. However, the previous study reported that the vast majority of patients (95%) with IPA presented with macronodule,9 whereas only 65% of patients with IPA revealed macronodule in the
CT findings in invasive pulmonary aspergillosis Table 3
453
Univariate analysis of factors associated with 90-day mortality in 110 patients with invasive pulmonary aspergillosis.
Variable
Death (n Z 46)
Survival (n Z 64)
Hazard Ratio (95% CI)
p-value
Age, >65 years Male gender Proven IPA Disseminated aspergillosis Neutropenia Use of corticosteroid Uncontrolled underlying disease RRT before IPAa Persistence of a positive serum galactomannana Consolidation-or-mass Macronodulea Multiple infarct-shaped consolidationsa Halo sign Ground glass opacity Micronodules Internal low attenuation sign Cavitary lesion Air-crescent sign Angio-invasive form Airway-invasive form Initial amphotericin-B Initial voriconazole
10 29 3 6 22 27 32 26 18/28 37 28 17 24 17 17 5 8 3 23 4 14 19
7 35 7 4 31 31 36 9 8/45 38 46 11 22 14 19 9 16 4 29 6 11 33
1.98 0.75 1.47 1.63 0.95 1.24 1.45 3.98 5.14 2.08 0.52 2.08 1.55 1.87 1.15 0.79 0.63 0.94 1.13 0.97 1.80 0.69
0.06 0.34 0.52 0.26 0.85 0.46 0.24 <0.001 <0.001 0.04 0.03 0.02 0.13 0.04 0.65 0.61 0.24 0.92 0.68 0.96 0.07 0.22
(21) (62) (6) (13) (47) (57) (68) (55) (64) (79) (55) (41) (51) (37) (36) (11) (17) (6) (50) (9) (30) (41)
(11) (55) (11) (6) (48) (48) (56) (14) (18) (59) (72) (21) (34) (22) (30) (14) (25) (6) (45) (9) (17) (52)
(0.98e3.40) (0.42e1.35) (0.46e4.74) (0.69e3.85) (0.53e1.68) (0.70e2.22) (0.78e2.68) (2.23e7.12) (2.36e11.19) (1.03e4.19) (0.29e0.93) (1.12e3.86) (0.89e2.75) (1.03e3.40) (0.63e2.08) (0.31e1.99) (0.30e1.35) (0.29e3.03) (0.63e2.01) (0.35e2.72) (0.96e3.37) (0.38e1.25)
Data are no. (%) of patients, unless otherwise indicated. CI, confidence interval; IPA, invasive pulmonary aspergillosis; RRT, renal replacement therapy; GM, galactomannan. a Independent risk factors for 90-day mortality (p < 0.05 in Cox proportional hazard model).
current study. The relatively low proportion of macronodule in our study may be due to the high proportion of transplant recipients (45%) in our population, since these patients can present with variable CT findings.6,13,15 Therefore, the presence of macronodule may not be useful to exclude a diagnosis of IPA especially in SOT recipients. However, the presence of macronodule and/or consolidation-or-mass retained a high sensitivity (98%). Thus, the absence of these two CT findings might be helpful to exclude a diagnosis of IPA. The patterns of IPA in patients with neutropenia are characterized by prominent fungal angioinvasion, intraalveolar hemorrhage, tissue infarction, and minimal inflammatory cell response.7,23 In patients with immunosuppression but not myelosuppression, the histologic pattern is different, and is characterized by central liquefaction necrosis, fewer hyphae aligned in a radical pattern around the lesion, and prominent neutrophil infiltration.7,23 In the present study, angio-invasive forms (32% vs 60%, p Z 0.003) and halo signs (26% vs 55%, p Z 0.002) were more common in the patients with neutropenia, which agrees with previous reports.13,19,20 However, although our preliminary data suggested that the airway-invasive form of IPA was more common in SOT recipients than neutropenic patients,13 we did not find a significant difference in airway-invasive IPA between the nonneutropenic transplant recipients and neutropenic patients in the present work. One possible explanation may be the use of an earlier version of the EORTC criteria for IPA24 in the previous study.13 In the present analysis we used the modified EORTC/MSG criteria,1 so that patients who did not fulfill the clinical criteria (one of the following 3 signs on CT: a well-
defined dense lesion, air-crescent sign, or cavity) would not have been enrolled. Historically, nodular lesions with halo or air-crescent signs were associated with IPA.8,9,25 These imaging findings are commonly used to make a presumptive diagnosis and initiate empirical treatment for IPA in high risk groups (eg. neutropenic fever in patients with hematologic malignancy). Although the revised EORTC/MSG criteria for host factors include non-neutropenic immunosuppressed patients with SOT, hereditary immunodeficiencies, and connective tissue disorders,1 the CT findings of IPA have not been fully validated in this group. We showed that the non-angio-invasive types of IPA such as the airwayinvasive or undeterminate forms were more common in non-neutropenic transplant recipients than in neutropenic patients. Therefore, because some patients, such as nonneutropenic transplant recipients, may not meet the radiologic criteria of EORTC/MSG,1 radiologic findings other than prespecified radiologic criteria such as EORTC/MSG would be included in the clinical criteria for IPA. Recently, Nucci et al. also proposed adding GM assay-supported cases with nonspecific radiologic findings to the EORTC/MSG criteria.26 We found that serum GM titers, macronodules, multiple infarct-shaped consolidations, and renal replacement therapy, were significantly associated with 90-day mortality. Previous reports noted a strong correlation between serum GM index and IPA survival outcomes.13,18,19 The need for renal replacement therapy is also a clearly established risk factor for mortality from IPA.2,27 However, macronodules and multiple infarct-shaped consolidations have
454
A
S.Y. Park et al.
Presence of macronodules Absence of macronodules 0.31 0.16–0.61
B Absence of multiple infarctshaped consolidations Presence of multiple infarctshaped consolidations (HR 4.26, 95% CI 2.05-8.87)
Figure 3 Estimated survival curves for patients with invasive pulmonary aspergillosis according to the presence of macronodules (A) and multiple infarct-shaped consolidations (B) based on the Cox proportional hazard model.
not been described as clinical predictors associated with outcomes of IPA. Macronodule are one of the main findings of early IPA, and they are such a common feature on initial CT that their absence argues against the likelihood of IPA.9,11 Thus, early initiation of effective antifungal treatment for IPA, based on the finding of macronodules, may be associated with a significantly improved response to treatment, and with improved survival. However, we could not discover any evidence of a relationship between halo
signs and higher survival in previous reports.4,9,28 This may be explained by the fact that most of the patients (86%) included in the previous studies had underlying hematological malignancies,9 whereas we included a relatively large number (45%) of non-neutropenic transplant recipients. Since, in our study, only 26% of the nonneutropenic transplant recipients with IPA presented with halo signs, the halo sign may not have prognostic implications associated with early detection of IPA specifically in non-neutropenic transplant recipients. Another interesting result of our study is that the presence of multiple infarct-shaped consolidations on baseline CT was a significant predictor of poor outcome. The presence of multiple infarct-shaped consolidations may point to late diagnosis of the IPA, since multiple nodules gradually coalesce into areas of consolidation29 and these consolidations may be complicated by infarctions caused by angioinvasion by Aspergillus. This study had several limitations. First, some may object that the transplant groups were composed of recipients of both SOT and HSCT. A recent review of autopsy-proven IPA revealed that a histopathology characterized by diffuse inflammation and limited numbers of Aspergillus hyphae was more common in HSCT recipients with GVHD who were treated with glucocorticoids than in neutropenic patients.14 Thus, the recipients of SOT and HSCT after primary engraftment included in the present study were both glucocorticoid-immunosuppressed patients and developed a form of IPA that was pathologically distinct from that in neutropenic patients.7,30 Second, our study shares the limitations of retrospective studies and suffers from a low autopsy rate, a trend noted worldwide.14 Only 10 proven IPA cases were included in our study and no autopsies were performed on the patients who died. Some of the neutropenic patients with IPA also had thrombocytopenia, and allograft dysfunction occurred in 53% of the transplant recipients, especially the liver transplant recipients, at the time of IPA diagnosis. The patients with hepatic allograft dysfunction had severe coagulopathy and were often markedly thrombocytopenic. Thus, these patients could not have tolerated invasive procedures such as transbronchial lung biopsy. Third, our study was limited by a relative insensitivity of Aspergillus species culture,31 making the diagnosis of IPA dependent on serum and BAL GM assays. This might be explained by the fact that prophylactic or empirical antifungal agents had already been administered to 85% of the patients (93/110).13,17 In addition, these antifungal use prior to the diagnosis of IPA may affect the CT findings. Indeed, the subgroup analysis including the patients (n Z 17) without antifungal agents prior to the diagnosis of IPA showed that the most common finding was macronodule (77%) while consolidation-or-mass and infarct-shaped consolidation were observed in only 35% of patients, respectively. The relative high proportion of patients who revealed macronodules in their CT findings is consistent with previous data which showed that 95% of neutropenic patients revealed macronodule in their CT findings.9 It remains to be settled on the influence of prophylactic or empirical antifungal agents on radiologic findings of IPA. Finally, lung transplant recipients were not included in the present study and this might attenuate differences in radiologic
CT findings in invasive pulmonary aspergillosis findings between non-neutropenic transplant recipients and neutropenic patients.32 In conclusion, our data suggest that chest CT findings in non-neutropenic transplant recipients with IPA are similar to those in neutropenic patients with IPA, except that consolidation-or-mass, halo signs, and angio-invasive form are less common in non-neutropenic transplant recipients. So, nonspecific radiological findings might not exclude a diagnosis of IPA radiological findings in non-neutropenic transplant recipients. Our findings also suggest that certain CT findings in patients with IPA, such as macronodules and multiple infarct-shaped consolidations, have prognostic implications. Thus, when we observed multiple infarctshaped consolidations in patients with IPA, aggressive antifungal treatment is needed to overcome the poor prognosis.
455
7.
8.
9.
10.
Funding 11.
No author received financial support.
Conflict of interest
12.
There are no potential conflicts of interest for any authors.
13.
Acknowledgments
14.
We thank Ji-Eun Hyun for administrative support. We are grateful to Julian Gross for assistance in language editing. 15.
Appendix. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jinf.2011.08. 007.
References 1. De Pauw B, Walsh TJ, Donnelly JP, Stevens DA, Edwards JE, Calandra T, et al. Revised definitions of invasive fungal disease from the European organization for research and treatment of cancer/invasive fungal infections cooperative group and the national institute of allergy and infectious diseases mycoses study group (EORTC/MSG) consensus group. Clin Infect Dis 2008;46(12):1813e21. 2. Singh N, Paterson DL. Aspergillus infections in transplant recipients. Clin Microbiol Rev 2005;18(1):44e69. 3. Denning DW. Therapeutic outcome in invasive aspergillosis. Clin Infect Dis 1996;23(3):608e15. 4. Caillot D, Casasnovas O, Bernard A, Couaillier JF, Durand C, Cuisenier B, et al. Improved management of invasive pulmonary aspergillosis in neutropenic patients using early thoracic computed tomographic scan and surgery. J Clin Oncol 1997; 15(1):139e47. 5. Wahba H, Truong MT, Lei X, Kontoyiannis DP, Marom EM. Reversed halo sign in invasive pulmonary fungal infections. Clin Infect Dis 2008;46(11):1733e7. 6. Milito MA, Kontoyiannis DP, Lewis RE, Liu P, Mawlawi OR, Truong MT, et al. Influence of host immunosuppression on ct
16.
17.
18.
19.
20.
21.
22.
23.
findings in invasive pulmonary aspergillosis. Med Mycol 2010; 48(6):817e23. Berenguer J, Allende MC, Lee JW, Garrett K, Lyman C, Ali NM, et al. Pathogenesis of pulmonary aspergillosis. Granulocytopenia versus cyclosporine and methylprednisolone-induced immunosuppression. Am J Respir Crit Care Med 1995;152(3): 1079e86. Caillot D, Couaillier JF, Bernard A, Casasnovas O, Denning DW, Mannone L, et al. Increasing volume and changing characteristics of invasive pulmonary aspergillosis on sequential thoracic computed tomography scans in patients with neutropenia. J Clin Oncol 2001;19(1):253e9. Greene RE, Schlamm HT, Oestmann JW, Stark P, Durand C, Lortholary O, et al. Imaging findings in acute invasive pulmonary aspergillosis: clinical significance of the halo sign. Clin Infect Dis 2007;44(3):373e9. Gerson SL, Talbot GH, Hurwitz S, Strom BL, Lusk EJ, Cassileth PA. Prolonged granulocytopenia: the major risk factor for invasive pulmonary aspergillosis in patients with acute leukemia. Ann Intern Med 1984;100(3):345e51. Orr DP, Myerowitz RL, Dubois PJ. Patho-radiologic correlation of invasive pulmonary aspergillosis in the compromised host. Cancer 1978;41(5):2028e39. Park YS, Seo JB, Lee YK, Do KH, Lee JS, Song JW, et al. Radiological and clinical findings of pulmonary aspergillosis following solid organ transplant. Clin Radiol 2008;63(6):673e80. Park SY, Kim SH, Choi SH, Sung H, Kim MN, Woo JH, et al. Clinical and radiological features of invasive pulmonary aspergillosis in transplant recipients and neutropenic patients. Transpl Infect Dis 2010;12(4):309e15. Chamilos G, Luna M, Lewis RE, Bodey GP, Chemaly R, Tarrand JJ, et al. Invasive fungal infections in patients with hematologic malignancies in a tertiary care cancer center: an autopsy study over a 15-year period (1989e2003). Haematologica 2006;91(7):986e9. Kojima R, Tateishi U, Kami M, Murashige N, Nannya Y, Kusumi E, et al. Chest computed tomography of late invasive aspergillosis after allogeneic hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2005;11(7): 506e11. Park SY, Lee SO, Choi SH, Sung H, Kim MN, Choi CM, et al. Aspergillus galactomannan antigen assay in bronchoalveolar lavage fluid for diagnosis of invasive pulmonary aspergillosis. J Infect 2010;61(6):492e8. Park SY, Lee SO, Choi SH, Jeong JY, Sung H, Kim MN, et al. Serum and bronchoalveolar lavage fluid galactomannan assays in patients with pulmonary aspergilloma. Clin Infect Dis 2011; 52(7):e149e52. Woods G, Miceli MH, Grazziutti ML, Zhao W, Barlogie B, Anaissie E. Serum aspergillus galactomannan antigen values strongly correlated with outcome of invasive aspergillosis. A study of 56 patients with hematologic cancer. Cancer 2007; 110:830e4. Maertens J, Buve ´ K, Theunissen K, Meersseman W, Verbeken E, Verhoef G, et al. Galactomaannan serves as surrogate endpoint for outcome of pulmonary invasive aspergillosis in neutropenic hematology patients. Cancer 2009;115:355e62. Hansell DM, Bankier AA, MacMahon H, McLoud TC, Muller NL, Remy J. Fleischner society: glossary of terms for thoracic imaging. Radiology 2008;246(3):697e722. Logan PM, Primack SL, Miller RR, Muller NL. Invasive aspergillosis of the airways: Radiographic, CT, and pathologic findings. Radiology 1994;193(2):383e8. Logan PM, Muller NL. High-resolution computed tomography and pathologic findings in pulmonary aspergillosis: a pictorial essay. Can Assoc Radiol J 1996;47(6):444e52. Stergiopoulou T, Meletiadis J, Roilides E, Kleiner DE, Schaufele R, Roden M, et al. Host-dependent patterns of tissue
456
24.
25.
26.
27.
S.Y. Park et al. injury in invasive pulmonary aspergillosis. Am J Clin Pathol 2007;127(3):349e55. Ascioglu S, Rex JH, de Pauw B, Bennett JE, Bille J, Crokaert F, et al. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis 2002;34(1):7e14. Neofytos D. Chest computed tomography versus serm galactomannan enzyme immunoassay for the diagnosis of probable invasive pulmonary aspergillosis: to be decided. Clin Infect Dis 2010;51(11):1281e3. Nucci M, Nouer SA, Grazziutti M, Kumar NS, Barlogie B, Anaissie E. Probable invasive aspergillosis without prespecified radiologic findings: Proposal for inclusion of a new category of aspergillosis and implications for studying novel therapies. Clin Infect Dis 2010;51(11):1273e80. Baddley JW, Andes DR, Marr KA, Kontoyiannis DP, Alexander BD, Kauffman CA, et al. Factors associated with mortality in transplant patients with invasive aspergillosis. Clin Infect Dis 2010;50(12):1559e67.
28. Horger M, Hebart H, Einsele H, Lengerke C, Claussen CD, Vonthein R, et al. Initial CT manifestations of invasive pulmonary aspergillosis in 45 non-hiv immunocompromised patients: association with patient outcome? Eur J Radiol 2005;55(3):437e44. nchez F, Franquet T, 29. Hidalgo A, Parody R, Martino R, Sa Gime ´nez A, et al. Correlation between high-resolution computed tomography and galactomannan antigenemia in adult hematologic patients at risk for invasive aspergillosis. Eur J Radiol 2009;71(1):55e60. 30. Lewis RE, Kontoyiannis DP. Invasive aspergillosis in glucocorticoidtreated patients. Med Mycol 2009;47(Suppl 1):S271e81. 31. Cornillet A, Camus C, Nimubona S, Gandemer V, Tattevin P, Belleguic C, et al. Comparison of epidemiological, clinical, and biological features of invasive aspergillosis in neutropenic and nonneutropenic patients: a 6-year survey. Clin Infect Dis 2006;43(5):577e84. 32. Husain S, Mooney ML, Danziger-Isakov L, Mattner F, Singh N, Avery R, et al. A 2010 working formulation for the standardization of definitions of infections in cardiothoracoc transplant recipients. J Heart Lung Transplant 2011;30(4):361e74.