| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Diagnostic Imaging, The University of Texas M. D. Anderson
Cancer Center, Unit 371, 1515 Holcombe Blvd., Houston, TX 77030.
2 Department of Infectious Diseases, Infection Control, and Employee Health, The
University of Texas M. D. Anderson Cancer Center, Houston, TX.
Received October 9, 2007;
accepted after revision December 26, 2007.
Address correspondence to E. M. Marom
(emarom{at}di.mdacc.tmc.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Pulmonary fusariosis has radiographic manifestations that are suggestive of an angioinvasive mold. Nodules or masses were the most common findings at CT, seen in 82% of patients compared with only 45% on chest radiography. The halo sign was not seen. Chest radiographs showed nonspecific findings in 30% of patients, and findings were normal at presentation in 25%. All of the patients had underlying hematologic malignancies. Thirteen of the 20 patients studied (65%) died within 1 month of diagnosis of pulmonary fusariosis. Because early initiation of intense antifungal therapy offers the best chance for survival in pulmonary fusariosis, early CT and appropriate microbiologic investigation should be obtained in severely immunocompromised patients.
Keywords: Fusarium species • immunocompromised • infection • pulmonary fusariosis
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We divided the patients included in the study into two groups: documented pulmonary fusariosis (proven or probable according to international consensus criteria [9]) and possible pulmonary fusariosis. The latter group included patients with signs or symptoms of pulmonary infection, radiologic signs of pulmonary infection, and Fusarium species isolated from a site other than the respiratory tract (e.g., blood or skin).
Images
Chest radiographs and chest CT scans were reviewed by three experienced
chest radiologists with 10, 10, and 5 years of experience, respectively, to
identify radiographic abnormalities associated with fusariosis. Differences in
interpretation were resolved by discussion to achieve consensus. When multiple
studies had been performed, the study closest to the onset of pulmonary
fusariosis was reviewed.
Consolidations; ground-glass opacities (GGO), defined as hazy opacities
that do not obscure the associated pulmonary vessels; nodules (< 3 cm); and
masses (
3 cm) along with their lobar locations were recorded. The largest
airway order (main bronchus, lobar, segmental, or subsegmental bronchi) that
abutted the pulmonary abnormality was recorded as well. Nodules and masses
were defined as solid, masslike consolidation (when containing air
bronchograms), or GGO, and their diameter was recorded. Studies were also
assessed for the presence of a halo sign (GGO surrounding a soft-tissue mass
or nodule), cavitations, tree-in-bud opacities, pleural effusions, and
lymphadenopathy.
All chest radiographs obtained between presentation and 12 months after presentation or death, whichever came first, were reviewed by an experienced chest radiologist. Evolution of the pulmonary abnormalities was documented.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All of the patients had underlying hematologic malignancies: acute leukemia in 14 (acute lymphocytic leukemia and acute myelogenous leukemia, seven each), myelo-dysplastic syndrome in three, and lymphoid malignancies in three (non-Hodgkin's lymphoma, Hodgkin's lymphoma, and chronic lymphocytic leukemia, one each). Nine patients were stem cell transplant recipients. Seventeen patients (85%) were receiving steroids and 19 (95%) were neutropenic at the time of diagnosis of fusariosis. Pulmonary signs or symptoms were present in 19 (95%) patients: shortness of breath (n = 14), cough (n = 5), chest pain (n = 5), hemoptysis (n = 1), lung crackles or rhonchi (n = 5). Fifteen patients were febrile. Thirteen patients (65%) died within 1 month of diagnosis of pulmonary fusariosis.
Imaging
CT scans—Chest CT scans were available in 11 patients and
were obtained within 0 to 41 days of the mycologic diagnosis (mean, 14 days;
median, 9 days). IV contrast agents were used in six patients. The slice
thickness was 10 mm in one patient, 7 mm in two patients, 5 mm in two, 3.75 mm
in four, 2.5 mm in one, and 1.25 mm in one.
The CT findings are summarized in Table 1. Pulmonary nodules were seen in nine patients (82%) and ranged from 0.3 to 2.7 cm. Three patients had a solitary nodule in an upper lobe, and six had more than 10 nodules. When nodules were multiple, they usually had no lobar predominance (n = 4), although lower lobe predominance (n = 1) or upper lobe predominance (n = 1) was seen. All nine patients had peripheral nodules involving the subsegmental bronchi, but three patients also had some nodules involving the segmental bronchi.
|
A lung mass was seen in six patients (55%), ranging from 3.0 to 6.7 cm (mean, 4.6 cm) (Figs. 1A and 1B). The mass was solitary in five patients whereas one patient had two masses. The masses were seen in the right upper lobe (n = 2), middle lobe (n = 1), right lower lobe (n = 2), and left upper lobe (n = 2). Masses usually involved the segmental-subsegmental airways; in one patient, the mass extended to the lobar airway. Four masses showed air bronchograms within them. Two masses were solid, and one of these was cavitary with an air-crescent sign (Figs. 2A and 2B).
|
|
|
|
Consolidation was seen in four patients (segmental in two; lobar or subsegmental in one each) and had a peripheral, lower lung predominance. When lobar or segmental, the consolidation involved two lobes, mainly in the lower lungs. GGOs were seen in three patients (multilobar in two; segmental in one). The halo sign and tree-in-bud opacities were not seen in any patient.
No patient had a normal chest CT scan. Four patients had small pleural effusions, one had a small pericardial effusion, and two had concomitant mild (< 1.5 cm) mediastinal or hilar adenopathy with the pulmonary infection.
Chest Radiographs
Chest radiographs were available in all 20 patients. The chest radiographs
were obtained at presentation within 2 to 31 days of the mycologic diagnosis
(mean, 9.9 days; median, 7.5 days). Abnormal chest radiography findings were a
solitary mass (n = 8), which ranged from 3.3 to 7.0 cm and was seen
in the right upper lobe (n = 4), left upper lobe (n = 2),
middle lobe (n = 1), and left lower lobe (n = 1);
heterogeneous opacities (n = 5), which were lobar (n = 1, in
the right upper lobe) or segmental (n = 4, with lower lobe
predominance); lobar consolidation (n = 4) in the left lower lobe
(n = 4) and right lower lobe (n = 2); and nodules
(n = 3), which ranged from 0.5 to 2.8 cm. Nodules were few (1-3 per
patient) and were seen peripherally in the right upper or right middle
lobe.
Nine (45%) of the 20 patients had a nodule or mass on their chest radiograph. In seven patients (35%), a nodule or mass was the only parenchymal abnormality seen on the chest radiograph. Among the nine patients who had a nodule or mass on CT, a nodule or mass was seen on the corresponding chest radiograph in only four cases (44%). The only parenchymal abnormality in six (30%) patients was airspace disease (consolidation and heterogeneous opacities). Chest radiographs were normal in five patients (25%) (Figs. 3A, 3B, 3C, 3D and 3E). Pleural effusions were seen on chest radiographs in eight patients. Adenopathy was not seen on any chest radiograph.
|
|
|
|
|
Of the patients who had a longer than 1-month survival after diagnosis of fusariosis, chest radiography follow-up was available in six patients. Three patients (50%) showed fluctuations in their airspace disease within weeks and died 2-6 months after fusariosis diagnosis. Three patients survived more than 1 year: in one, the chest radiograph showed resolution of airspace disease over 2 months, one patient's chest radiograph showed growing pulmonary nodules that eventually resolved within 4 months, and one patient's chest radiograph showed cavitation of an initial mass at 3 weeks with a gradual decrease in the soft-tissue component to a thin-walled cyst at 8-month chest radiography follow-up.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The Fusarium species are soil saprophytes and plant and bacterial pathogens. Only a few of the 50 different species have been reported as pathogenic in humans, mainly: Fusarium solani, Fusarium oxysporum, Fusarium moniliforme, Fusarium verticillioides, Fusarium dimerum, and Fusarium proliferatum [8]. Among these, F. solani has the highest virulence, causing half of the reported human fusariosis cases. The skin and respiratory tract are the primary portals of entry for Fusarium species infections.
Fusarium species infections in normal hosts are typically localized to skin, eye, or bone after trauma. In contrast, fusariosis in severely immunocompromised patients, such as those with hematologic malignancies, is a much more severe infection. There are four major clinical manifestations: refractory fever of unknown origin, sinopulmonary infection or pneumonia, disseminated infection, and a variety of focal single-organ infections [8]. The increased incidence is multifactorial and thought to reflect the use of new, highly immunosuppressive chemotherapeutic regimens; the broad use of antifungal agents as either or both prophylactic and empiric therapy, resulting in selection of more resistant fungi; and possibly environmental exposures [8]. Differentiating pulmonary fusariosis from the more common pulmonary aspergillosis is difficult clinically because both present with fever despite the use of broad-spectrum antibiotics and with sinopulmonary infections. Differentiating the two is also difficult histopathologically because both look similar, with angioinvasion of acute branching septate hyphae [8].
Because of the rarity of this opportunistic infection and the complex clinical scenarios when it occurs, the management of fusariosis is not well defined and is highly individualized [13]. Fusarium species is less susceptible to antifungal treatment as compared with Aspergillus species. Only high doses of liposomal amphotericin B and the newer triazoles, such as voriconazole and posaconazole, have been shown to be effective against Fusarium species. Fusarium species infection has a worse outcome than that of Aspergillus species and fewer therapeutic options. Any efforts to differentiate early Fusarium species infections from Aspergillus species infections would lead to an aggressive strategy with the combined antifungal therapy previously described and WBC transfusion if the patient is neutropenic.
Our study shows that the pulmonary imaging features of pulmonary fusariosis are similar to those of pulmonary aspergillosis, with most patients presenting with a mass or nodules on CT. The existing literature on fusariosis does not focus on radiographic characteristics and is limited to reviews and case reports in which chest radiographs were assessed by clinicians and not by radiologists [3]. These earlier studies found that the presence of nodules or cavitary lesions, thought to be characteristic of fungal infections, was uncommon in pulmonary fusariosis, with a prevalence of no more than 19% among patients with any pulmonary abnormality. Instead, the majority of patients (up to 81%) had nonspecific opacities [3]. In our study, radiologic findings consistent with pulmonary lesions caused by an angiotropic mold, such as nodules or masses, were seen on 45% of chest radiographs at presentation. However, because the majority either had nonspecific airspace disease or a normal-appearing chest radiograph, CT should be helpful early in the investigation of febrile neutropenic patients. Whether the use of early CT translates into improved survival, as it does with pulmonary aspergillosis [14], remains to be seen because Fusarium species is one of the most resistant fungi to antifungal agents [8].
Because our small study was retrospective, the CT scans obtained varied from patient to patient in the time from diagnosis. Although this noncomparative study found that CT scans were highly sensitive in showing findings suggestive of fungal infection, our study could not answer the question of how long it takes from the first clinical symptoms until the appearance of parenchymal abnormalities detectable on CT, and we could not assess the evolution of pulmonary lesions on CT owing to the high short-term mortality rate in our highly immunocompromised patients.
In conclusion, pulmonary fusariosis has radiographic manifestations that are suggestive of an angiotropic mold. Nodules or masses are the most common findings on CT; a halo sign is not seen. Chest radiographs can be negative or show nonspecific findings.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
