First published online December 6, 2006
British Journal of Radiology (2007) 80, 516-523
© 2007 British Institute of Radiology
doi: 10.1259/bjr/39696316
Differences and similarities of cytomegalovirus and pneumocystis pneumonia in HIV-negative immunocompromised patients – thin section CT morphology in the early phase of the disease
M N Vogel, MD
1
H Brodoefel, MD
1
T Hierl, MD
2
R Beck, MD
3
W A Bethge, MD
4
C D Claussen, PROF
1 and
M S Horger, MD
1
1 Department of Diagnostic Radiology, Eberhard-Karls-University, Hoppe-Seyler-Str. 3, 72076 Tübingen, 2 Institute of Medical Microbiology and Hygiene, Eberhard-Karls-University, Elfriede-Aulhorn-Str. 6, 72076 Tuebingen, 3 Institute of Medical Virology and Epidemiology of Viral Diseases, Eberhard-Karls-University, Elfriede-Aulhorn-Str. 6, 72076 Tuebingen, 4 Department of Internal Medicine II, Eberhard-Karls-University, Otfried-Müller-Str. 10, 72076 Tübingen, Germany
Correspondence: Monika Vogel, MD, Department of Diagnostic Radiology, Eberhard-Karls-University, Hoppe-Seyler-Str. 3, 72076 Tübingen, Germany. E-mail: monika.vogel{at}med.uni-tuebingen.de
 |
Abstract
|
|---|
The purpose of this study was to assess CT morphology of pneumocystis pneumonia (PcP) and cytomegalovirus (CMV) pneumonia for specific characteristic features, similarities as well as differences, which might contribute to an early diagnosis and, therefore, influence patient management
58 patients were included, 31 with CMV pneumonia and 27 with PcP. All patients with CMV pneumonia had underlying haematological malignancies (n = 31) mainly treated by haematopoietic cell transplantation (HCT) (n = 26). Patients with PcP had haematological malignancies (n = 17) treated by HCT in 6, solid tumours (n = 5) and corticosteroid therapy (n = 5). Thin section CTs were analysed retrospectively by two radiologists. 18 CT morphological criteria were evaluated for presence or absence. Significance was calculated by 
2 test. Interobserver correlation was tested by kappa-statistic (K)
Only 5 of the 18 features were found to have significantly different frequencies in the two entities. Apical distribution (p<0.01), mosaic pattern (p<0.01) and homogeneous structure of ground-glass opacities (GGO) (p<0.05) were found more frequently in PcP (each K: 0.7–0.9), whereas small nodules or unsharp demarcation of GGO and consolidation were typical of CMV pneumonia (p<0.05). Peripheral sparing, consolidation and septal thickening inter alia were found equally in both groups
In conclusion analysis of craniocaudal distribution, demarcation and structure of infiltrates may be helpful in prioritizing differential diagnosis of CMV pneumonia or PcP. However, some features thought typical for one or the other entities appear with similar frequency in both groups in HIV-negative patients.
|
Introduction
|
|---|
Pneumonias are amongst the most frequent, life threatening infectious complications in immunocompromised patients. In a clinical constellation of a pneumonia not only the characteristic and usually well containable fungal pneumonia comes into question, but also pneumonias from cytomegalovirus (CMV) and Pneumocystis jirovecii (previously </emph>carinii</emph>) (PcP) [1]. In recipients of haematopoietic cell transplantations (HCT) the incidence of CMV is 20–35% and the mortality is up to 50% [2, 3]. The incidence of PcP in patients with solid tumours and collagen vascular diseases undergoing steroid therapy is estimated to be below 2%. In transplant receivers it lies between 5% and 10% and over 50% in AIDS patients, where it is the most common opportunistic infection. Depending on the underlying disease, mortality lies between 20% and 50% and is at its highest in cancer patients [2, 3].
Independent from the underlying disease, the clinical presentation in both cases is fever, dyspnoea, unproductive coughing and uncharacteristic reduced transparency in conventional thorax images. In detecting and differentiating pneumonia in immunocompromised patients thin section CT of the thorax is the considered imaging modality [4]. However the differentiation of pneumonia by CMV or pneumocystis is usually difficult. Especially in the early phase of the disease, the morphological appearances of both entities on thin section CT are very similar to each other [5–9]. Furthermore, the analysis of a so-called typical pattern, especially in patients without AIDS, often seems to lead to an incorrect interpretation or, at least, to a false prioritization.
Thin section CT findings of CMV pneumonia in HIV-negative patients are described as a combination of small nodules, ground-glass opacities (GGO), nodular opacities and consolidation, as well as an association with interlobular septal thickening and pleural effusions [6, 10–15].
In PcP, the typical picture of a thin section CT in the early phase of the disease is described as extensive GGO, which is often "map like", diffused or focal with predilection in the central lung sections [16–18]. Furthermore, septal thickening and consolidation have also been observed [16]. Earlier observations in non-AIDS PcP patients consist mainly of perihilar, ground-glass style and reticular opacities [19]. The development of cysts, spontaneous pneumo-thorax, pneumomediastinum and the predominantly affected apical lung have all, on the whole, only been described in PcP patients with AIDS [16, 20, 21].
The aim of the presented study was to determine which patterns typical of CMV pneumonia or PcP in patients without AIDS were actually noticeable with varying frequency by means of an early thin section CT. In addition, we looked for significant morphological differences in both of these entities in order to expedite diagnosis for impact on patient management.
|
Methods
|
|---|
Thorax thin section CT was analysed retrospectively in 58 patients. The acquisition of the patients followed after positive proof of CMV or Pneumocystis jiroveci/carinii from broncho-alveolar lavage (BAL) probes from the Institute for Microbiology and Virology at our university by way of a computer search. The time span covered January 1997 to November 2005. Patients with either suspicion of or confirmed HIV seropositivity, or with pulmonary co-infection, as well as patients without any (close) thin section CT examination were excluded from the study. A further necessary inclusion criterion was a clinical course indicating pneumonia. Included in the study were 31 patients with CMV pneumonia and 27 patients with PcP. CMV pneumonia was diagnosed in 31 patients with haematological disease: 18 men and 13 women with an average age of 44 (min–max 19–77) years, of which 26 had previously received an allogenic HCT with an average time post transplant of 90 days (total 0–320 days). The remaining five were either in the conditioning phase (n = 3) or had received high dosage chemotherapy (n = 2).
PcP was diagnosed in 27 patients, 17 men and 10 women with an average age of 54 (min–max 20–81) years, 17 of whom had haematological illnesses (six had received allogenic HCT 88–893 days previously), five had solid tumours and five patients were undergoing long-term steroid therapy for a collagen vascular disease. None of the PcP patients had received specific inhalator or other antibiotic PcP prophylaxis prior to the onset of the symptoms. CT scans were performed in patients who gave their written informed consent to the examination. Anonymity and data protection were taken into account. Approval for publication of patients' data was obtained from our Institutional Research Ethics Board.
Characterization and cause of the immune incompetence of the 58 patients are listed in Tables 1
and 2.
In patients receiving more than one examination, the evaluated thorax CT was selected as close as possible to the date of diagnostic BAL. The average length of time from the onset of the clinical symptoms to the CT was 1 day (min–max 0–10 days) in all patients.
The BAL in all patients was carried out on average 1 day prior to or following the thin section CT (min–max: 0–6 days). CMV was identified by staining of immediate early antigen and its characteristic cytopathic effect.
The confirmation of Pneumocystis jiroveci followed through the use of polymerase chain reaction (PCR).
All patients received unenhanced thorax CTs. These were either obtained with a single-detector row CT scanner (Somatom Plus 4) or a multidetector row scanner Volume Zoom (both Siemens Medical Engineering, Forchheim, Germany). Scanning parameters for spiral CT of the chest with a single-detector row CT scanner were 120 kV (peak), 120 mAs, 5 mm collimation and a pitch of 1.5. Axial scans through the thorax were obtained during full inspiration. Additional thin section CT scans were obtained with 1.0 mm collimation and a 10 mm slice interval. On the Volume Zoom scanner, a collimation of 4x1.0 mm and a slice width of 1.25 mm were chosen. Table speed rotation was 6 mm and rotation time was 0.75 s, with a pitch of 1.5. We used an increment of 1.2 mm. The tube voltage was 120 kV (peak) and the tube current time product was 100 mAs. Images were reconstructed with a high spatial frequency algorithm (high-resolution CT) B70s kernel. All scans were viewed at standard mediastinal windows (level, 35 HU; width 450 HU) and lung windows (level, –700 HU; width, 1500 HU).
The 58 CT examinations were assessed retrospectively by two independent radiologists. Neither was previously involved in the diagnosis of the examined patients and both were blinded to the results of the BAL analysis. They were, however, informed that they were dealing with patients with pneumonia, either through CMV or Pneumocystis jiroveci and information regarding previous examinations as well as regarding symptoms and underlying disease was at hand.
Eighteen criteria in relation to distribution pattern, thickness and structure of the infiltrate were defined as follows:
GGO was defined as an area of hazy increased attenuation without obscuration of underlying vascular markings. It was termed homogeneous when its density had a smooth aspect and remained the same within an area at least as large as a secondary pulmonary lobule or heterogeneous if it was slightly denser in the central portion of the secondary pulmonary lobule or around the bronchioli as a hybrid between GGO and small nodules. Consolidation was defined as an area of increased attenuation with obscuration of underlying vascular markings. Consolidation and GGO were recorded as focal when their size lay under a lobule segment and when scattered and – as opposed to mosaic pattern – neither connected to one another nor following anatomic boundaries. In the case of being predominantly close to the bronchi, the distribution was described as being peribronchial.
Diffuse was defined as the widespread distribution throughout the lung parenchyma. In the case of bilateral and symmetric distribution of at least 50% of GGO or consolidation, the pattern was considered as bilateral and symmetric. The distribution was considered as peripheral when the affected lung section was generally no further than 1–4 cm from the visceral pleura. When >50% of the affected area was found to be in the caudal or cranial lung section, then it was considered to be of a basal or apical distribution, respectively. A perihilar distribution with peripheral sparing was defined as the affected area being more than 4 cm away from the visceral pleura. A mosaic pattern was defined as more or less sharply demarcated regions of different density within infiltrated lung parenchyma or as regions of nearly unattained secondary lobules in between infiltrates. Depending on demarcation of the infiltrate through anatomical structure, they were considered as being either with sharp (i.e. well defined) or unsharp (i.e. ill defined) demarcation. Nodules were defined as being round opacities at least moderately well marginated with a diameter of maximum 3 cm. Small nodules were defined as round opacities being no greater than 1 cm in maximum diameter. Septal thickening was defined as an abnormal widening of interlobular septa. Lymph node enlargement and pleural effusion were only registered when a noticeable increase was shown in comparison to that of the pre-examinations.
Patterns were only registered when they appeared in at least 50% of the affected lung section and had newly occurred after the comparison of pre-examination CTs. Pre-examination CTs of the thorax were available for 49 of the 58 patients and the time interval to the actual examination was no longer than 190 days (average 45 days).
Significance of the differing frequencies of the various patterns was calculated for each individual pattern by means of a 2 test. The null hypothesis was no differing variations in frequencies of each individual pattern. The result was treated as significant in the case of a p-value <0.05. An adjustment of multiple tests after Bonferroni–Holm was carried out [22]. To calculate the correlation of the two observers' findings, the Kappa value (K) was calculated [23]. The frequency of the occurrence of the individual pattern, which is listed in the text and in Tables 3 and 4 and the data used for 2-calculation, relate only to cases in which both observers, independent from another, arrived at identical results. The additional information of the Kappa value relates to the results of the initial blinded evaluation and should give an indication for objectivity of the various patterns. Consensus reading was not performed.
View this table:
[in this window]
[in a new window]
|
Table 3. Distribution of the morphological CT criteria without significantly different frequency. Values relate to cases where both observers independently made a unanimous decision. Characteristics of GGO and consolidation relate equally to both types of infiltration and are listed according to their frequency
|
|
View this table:
[in this window]
[in a new window]
|
Table 4. Distribution of the CT morphological criteria with significant differentiating frequency. Values relate to a unanimous decision from both observers. The criteria are listed according to their significance
|
|
|
Results
|
|---|
13 of the 18 criteria showed no significant differences in frequency in relation to either CMV pneumonia or PcP (Table 3
). GGO was found in 26 patients (84%) with CMV pneumonia and in 24 patients (89%) with PcP (K: 0.7; 2: 0.31). However, as shown in Table 4, GGO does show differences in structure and distribution. In the case of CMV pneumonia, GGO often had an inhomogeneous aspect and were often associated with focal consolidation (Figure 1). In the case of PcP, GGO were typically homogeneous and either diffuse (Figure 2) or focally (Figure 3) distributed and often forming a mosaic pattern in the apical lung section. A septal thickening was found to occur at almost the same frequency in CMV pneumonia (n = 14; 45%) as in PcP (n = 11; 41%) (K: 0.9; 2: 0.12 (Figures 4–6)). In the case of CMV pneumonia, a consolidation was found in 11 patients (35%) and was mainly basal and, in the case of bed-ridden patients, was often found in the dorsal lung section. In this case, sparing of the dorsobasal lung periphery was not shown in the majority of cases (Figure 7). In the case of PcP, consolidations were detected in six patients (22%) and were generally focally distributed (Figure 8). The Kappa value was 0.6 and 2 1.24 for the criteria of consolidation. Perihilar distribution with sparing of the lung periphery occurred somewhat more frequently in CMV pneumonia (n = 10; 32%) compared with PcP (n = 4; 15%). A significant difference was, however, not present (K: 0.8; 2: 2.47; Figures 3 and 4). A peripheral distribution occurred with nearly the same frequency in both of the entities (K: 0.8; 2: 0.17; Figures 6 and 7). In CMV pneumonia as well as in PcP, the distribution of GGO or consolidation was focal (K: 0.8; 2: 0.89), diffuse (K: 0.7; 2: 2.25), symmetrical (K: 0.7; 2: 1.38) or, in some cases, peribronchial (K: 0.8; 2: 0.36). It was shown that a sharp demarcation (K: 0.6; 2: 2.29) also occurred at nearly the same frequency in both groups. Although a predominantly basal distribution (K: 0.8; 2: 4.6) occurred in more CMV pneumonia cases than in PcP, it however showed no statistical significance. The additional occurrence of pleural effusion (K: 0.8: 2: 2.95) and lymph node enlargements (K: 0.8; 2: 0.04) were furthermore nearly equally apparent on both groups.
View larger version (122K):
[in this window]
[in a new window]
|
Figure 1. CMV pneumonia in a 39-year-old man with HCT for acute myeloid leukaemia 42 days previously and acute fever beginning the previous day. Thin section CT shows unsharp demarcated regions of GGO with zones of consolidation and centrilobular nodules (arrow).
|
|
View larger version (140K):
[in this window]
[in a new window]
|
Figure 2. PcP in a 60-year-old man with chronic lymphocytic leukaemia following chemotherapy and presenting with acute fever beginning the previous day. Thin section CT shows homogeneous GGO, which forms a mosaic pattern especially in the apical lung section (arrow).
|
|
View larger version (141K):
[in this window]
[in a new window]
|
Figure 3. PcP in a 50-year-old woman after HCT for non-Hodgkin's Lymphoma 88 days previously and acute fever beginning the day of thin section CT despite antibiotic therapy. Patchy, homogeneous GGO is distributed predominantly in the periphery of the lobes.
|
|
View larger version (153K):
[in this window]
[in a new window]
|
Figure 4. CMV pneumonia in a 56-year-old man 17 days after HCT due to acute myeloid leukaemia presenting with breathing problems and fever. The thin section CT shows diffuse consolidation zones, peripheral sparing, septal thickening (arrow) and bilateral pleural effusions.
|
|
View larger version (146K):
[in this window]
[in a new window]
|
Figure 6. CMV pneumonia in a 51-year-old man following HCT for acute myeloid leukaemia 115 days previously and presenting with acute fever beginning the day before. Thin section CT shows unsharp demarcated focal zones with GGO and septal thickening mainly distributed in the lung periphery.
|
|
View larger version (140K):
[in this window]
[in a new window]
|
Figure 7. CMV pneumonia in a 49-year-old man, 106 days after HCT due to acute myeloid leukaemia. He suffered from constant fever for 2 days, despite antibiotic therapy. The thin section CT shows small areas of consolidation predominantly in the lower lobes and dorsal parts of the lung. Simultaneously, centrilobular nodules, peripheral sparing and unsharp demarcated GGO are present.
|
|
View larger version (174K):
[in this window]
[in a new window]
|
Figure 8. PcP in a 65-year-old woman undergoing steroid medication for therapy of a collagen vascular disease and presenting with dyspnoea and fever beginning the previous day despite antibiotic therapy. The thin section CT shows patchy areas of consolidation, a mosaic pattern (arrow) and peripheral sparing in the upper lobes.
|
|
The following five patterns – listed in order of statistical significance importance – were found with differing frequency in patients with CMV pneumonia and PcP (Table 4).
The largest significance was in the frequency of an apical distribution of GGO or consolidation in PcP patients, where 19 cases (70%) were found unanimously in both viewings. In patients with CMV pneumonia, an apical lung invasion only occurred in six cases (19%) (p<0.01; K: 0.8). No influence was found to be caused through the length of the symptoms in both groups to the cranocaudal distribution.
Furthermore the mosaic pattern was mainly observed in patients with PcP and was recorded as being present in 14 cases with PcP (54%) and in 3 cases with CMV pneumonia (10%) (p<0.01; K: 0.08) according to both observers. See also Figures 2 and 8.
The demarcation of GGO and consolidation was also different in both groups. These were defined as having unsharp demarcations (p<0.05) in 15 patients with CMV (48%) as opposed to six PcP patients (23%). The fact that the demarcation of an infiltrate is a less objective criterion is clearly demonstrated by viewing the Kappa values of K: 0.5 (Figures 1, 3 and 6).
Nodules appeared in this study as only small nodules with centrilobular localization.
Subsequently, the occurrence of small nodules appeared as being helpful in differentiating between the two entities. These were detected by both observers in 15 patients with CMV pneumonia (48%) and three patients with PcP (11%) (p<0.05; K: 0.8) (Figures 1 and 3).
The frequency of GGO in itself was not seen to vary significantly between the two groups. As already explained above, its structure in PcP patients was, however, more homogeneous (Figures 2, 5 and 7). GGO was recorded as being homogeneous by both viewers in 14 patients with PcP (54%) as opposed to only five patients with CMV pneumonia (16%) (p<0.05; K: 0.7). In the case of these five CMV patients, the symptoms had lasted longer than 3 days.
View larger version (98K):
[in this window]
[in a new window]
|
Figure 5. PcP in a 40-year-old woman with acute myeloid leukaemia undergoing chemotherapy and presenting with a continuing fever despite antibiotic therapy. The thin section CT shows septal thickening (short, slanted arrow) and GGO (long, straight arrow) in the upper lobes.
|
|
To summarize, it can be said from the differences shown that the appearance of small nodules and/or unsharp demarcated GGO or consolidations can be more often detected in CMV pneumonia patients, whereas apical predominance and/or mosaic patterns of GGO or consolidation as well as a homogeneous structure of GGO appears more often in PcP patients.
|
Discussion
|
|---|
CMV and Pneumocystis jiroveci are two of the most common infectious pathogens for a life threatening pneumonia in the case of immunocompromised patients. CMV belongs to the family of the herpes viridae, and has a seroprevalence of 40–100% in the world's population. Symptomatic infections or reactivations are almost exclusively observed in immunoincompetent carriers [3]. One effective therapy for CMV pneumonia is the nephrotoxic foscarnet. Ganciclovir (also for therapy of CMV pneumonia) can cause dangerous bone marrow toxicity. Pneumocystis is on the other hand classified as a fungus [24]. Humans are normally subject to Pneumocystis jirovecii (previously carinii) which is usually acquired through environmental inhalation [25]. Although the sulphonamide therapy of PcP is relatively inexpensive, it is nevertheless contraindicated for patients with sulphonamide allergy. A limited contraindication in the case of patients with kidney insufficiency or heminephrectomy is present due to its nephrotoxicity. Not only with CMV pneumonia, but also with PcP, the diagnosis indicates, without exception, a clinical stay for the patient. The therapeutical antibiotic or virostatic medication needs to be intravenously administered and at a notably higher dosage than that of a specific prophylaxis. Therefore a safe and, if possible, confirmed diagnosis by multimodalities (imaging plus laboratory) is desirable. In some cases, the invasive diagnostical BAL can be contraindicated or results of microbiological and virological analysis may be unclear. Nevertheless, the decision regarding the correct therapy is needed as early as possible in order to prevent unnecessary toxic side effects, as well as to reduce the patient's stay as an inpatient. Microbiological and especially virological BAL analysis requires more time than a thin section CT, meaning that the CT has the initial purpose of pointing the way for further patient management. So a prioritization evaluation is strongly recommended after CT morphology. CT morphology of both entities often overlaps, especially in the early phases of the illness. Nevertheless, there are numerous CT morphological criteria which, in the case of either disease, are described as occurring frequently or suggesting either CMV pneumonia or PcP [6, 19, 26]. The result of a direct comparison of the two entities has, to our knowledge, not been published. The main aim of this study was to determine whether these criteria actually occur with varying frequency in the two entities. In the case of non-AIDS, immunocompromised patients with CMV pneumonia, small nodules, GGO and consolidation zones, as well as septal thickening, were described as being the most common patterns [5, 10–12, 15]. Our data correlate with these previous results. In addition, we were able to show that small nodules, an unsharp demarcation and a basal distribution of GGO and consolidation in the early phases of the disease occur more frequently with CMV pneumonia than in PcP.
Described as classical characteristics of CT morphology of a PcP are a central, perihilar distribution with peripheral sparing, a mosaic pattern and, in the case of some authors, an apical distribution and cystic changes [6, 8, 16, 17, 20, 27, 28]. Earlier descriptions in the case of non-AIDS affected patients are limited to perihilar GGO and reticular interstitial opacities [19]. According to our knowledge the predominant involvement of the apical lung region and the occurrence of consolidation on CT were only rarely described in the last patient group [29].
Our results show that even in patients without AIDS, an apical distribution of GGO and consolidation as well as a mosaic pattern often occur in PcP. In addition, it could be shown that this pattern appeared more often with PcP than with CMV pneumonia. Infections involving peripheral, perihilar or both lung regions occurred with similar frequency in both CMV pneumonia and PcP in our study. These results somewhat contradict previous descriptions, which show a perihilar distribution of infiltrates and peripheral sparing as patterns especially occurring in PcP patients [6, 8, 28].
A further result of our study is the regular occurrence of consolidation zones in non-AIDS PcP patients [16]. Furthermore, no significant difference could be found between CMV pneumonia and PcP in relation to the frequency of consolidation. The same fact pertains to the presence of any GGO. In the majority of cases of incurring diffused GGO in immunocompromised patients, an atypical pneumonia is the underlying cause [30]. The most common infectious agents that come into question in this setting are pneumocystis and CMV [1]. However, in patients undergoing chemotherapy or following HCT, differential diagnosis of further infectious and non-infectious lung diseases has to be considered [1, 31]. It is to be especially noted that drug toxicity, hypersensitive reactions, fluid overload or diffuse alveolar haemorrhage could affect the analysed morphology and should, therefore, be diagnostically taken into account [1, 31]. When patient selection was made, such causes were ruled out wherever possible in our study. However, an eventual weak point of this work is the absence of additional diagnosis confirmation by biopsy or autopsy despite the presence of BAL results in all cases.
Specificity and sensitivity of these patterns in all non-AIDS immunocompromised patients must be further evaluated in a future prospective or blinded retrospective study.
|
Conclusion
|
|---|
Of the 18 examined criteria, only five showed a significant variation in the frequency of CMV pneumonia and PcP. The occurrence of small, centrilobular nodules and unsharp demarcation of GGO and consolidation could help to make the diagnosis of CMV pneumonia clearer, whereas an apical distribution and the occurrence of a mosaic pattern points more to PcP.
The remaining 13 criteria with sparing of the periphery and early consolidation occur with similar frequency in both pneumonias and, therefore, are not thought to be helpful in differentiating the two diseases.
Received for publication March 17, 2006.
Revision received September 3, 2006.
Accepted for publication September 25, 2006.
|
References
|
|---|
- Miller WT, Shah RM. Isolated diffuse ground-glass opacity in thoracic CT: causes and clinical presentations. AJR Am J Roentgenol 2004;184:613–22.
- Nichols WG, Boeck MJ. Recent advances in the therapy and prevention of CMV infections. Clin Virol 2000;16:25–40.[CrossRef]
- Sepkowitz KA. Opportunistic infections in patients with and patients without acquired immunodeficiency syndrome. Infect Dis 2002;34:1089–107.
- Hidalgo A, Falco V, Mauleon S, Andreu J, Crespo M, Ribera E, et al. Accuracy of high-resolution CT in distinguishing between Pneumocystis carinii pneumonia and non-Pneumocystis carinii pneumonia in AIDS patients. Eur Radiol 2002;13:1179–84.[Medline]
- Raoof Sa, Raoof Su, Naidich DP. Imaging of unusual diffuse lung diseases. Cur Opin Pulmon Med 2004;10:383–9.[CrossRef]
- Webb WR, Müller NL, Naidich DP, editors. High-resolution CT of the lung. Philadelphia, USA: Lippincott-Raven, 2001
- McGuinness G, Gruden JF. Viral and Pneumocystis carinii infections of the lung in the immunocompromised host. J Thoracic Imaging 1999;14:25–36.[Medline]
- Fraser RS, Müller NL, Colman N, Pare PD. Diagnosis of diseases of the chest. Saunders, 1999. 909–15.
- Fraser RS, Müller NL, Colman N, Pare PD. Diagnosis of diseases of the chest. Saunders, 1999. 1006–8.
- Aafedt BC, Halvorsen R, Tylen U, Hertz M. Cytomegalovirus pneumonia: computed tomography findings. Can Assoc Radiol J 1990;41:276–80.[Medline]
- Franquet T, Lee KS, Müller NL. Thin-section CT findings in 32 immunocompromised patients with cytomegalovirus pneumonia who do not have AIDS. AJR Am J Roentgenol 2003;181:1059–63.[Abstract/Free Full Text]
- Gasparetto EL, Ono SE, Escuissato D, Marchiori E, Roldan L, Marques HL, et al. Cytomegalovirus pneumonia after bone marrow transplantation: high resolution CT findings. Br J Radiol 2004;77:724–7.[Abstract/Free Full Text]
- Kang EY, Staples CA, McGuinness G, Primack SL, Müller NL. Detection and differential diagnosis of pulmonary infections and tumours in patients with AIDS. AJR Am J Roentgenol 1996;166:15–9.[Abstract/Free Full Text]
- McGuinness G, Scholes JV, Gary SM, Leitman BS, McCauley DI, Naidich DP. Cytomegalovirus pneumonitis: spectrum of parenchymal CT findings with pathologic correlation in 21 AIDS patients. Thoracic Radiol 1994;192:451–9.
- Moon JH, Kim EA, Lee KS, Kim TS, Jung KJ, Song HJ. Cytomegalovirus pneumonia: high-resolution CT findings in ten non-AIDS immunocompromised patients. Kor J Radiol 2000;1:73–8.
- Boiselle PM, Crans CA Jr, Kaplan MA. The changing face of Pneumocystis carinii pneumonia in AIDS patients. AJR Am J Roentgenol 1999;172:1301–9.[Abstract/Free Full Text]
- Gruden JF, Huang L, Turner J, Webb WR, Merrifield C, Stansell JD, et al. High-resolution CT in the evaluation of clinically suspected Pneumocystis carinii pneumonia in AIDS patients with normal, equivocal or nonspecific radiographic findings. AJR Am J Roentgenol 1997;169:967–75.[Abstract/Free Full Text]
- Kuhlman JE. Pneumocystis infections: the radiologist's perspective. Radiology 1996;198:623–35.[Free Full Text]
- Bergin CJ, Wirth RL, Berry GJ, Castellino RA. Pneumocystis carinii pneumonia: CT and HRCT observations. J Comput Assist Tomogr 1990;14:756–9.[Medline]
- Crans CA Jr, Boiselle PM. Imaging features of Pneumocystis carinii pneumonia. Crit Rev Diagn Imaging 1999;40:251–84.[Medline]
- Rumbak MJ, Winer-Muram HT, Beals DH, Fry P. Tension pneumomediastinum complicating Pneumocystis carinii pneumonia in acquired immunodeficiency syndrome. Crit Care Med 1992;20:1492–4.[Medline]
- Adam J. Statistisches Know-how in der medizinischen Forschung. Berlin, Germany: Ullstein Mosby GmbH & Co, 1992;485–6.
- Dawson-Saunders BTR. Basic and clinical biostatistics. United States: McGraw-Hill, 2004. 118–9.
- Edman JC, Kovacs JH, Masur H, Santi DV, Elwood HJ, Sogin ML. Ribosomal RNA sequence shows Pneumocystis carinii to be a member of the fungi. Nature 1988;334:519–22.[CrossRef][Medline]
- Morris A, Beard CB, Huang L. Update on the epidemiology and transmission of Pneumocystis carinii. Microbes Infect 2002;4:95–103.[CrossRef][Medline]
- Fraser RS, Müller NL, Colman N, Pare PD. Diagnosis of diseases of the chest. Saunders, 1999. 707–13.
- Kuhlman JE, Kavuru M, Fishman EK, Siegelman SS. Pneumocystis carinii pneumonia: spectrum of parenchymal CT findings. Thoracic Radiol 1990;175:711–14.
- Schmidt G, Becker-Gaab C, Bogner JR, Sing A, Reiser M. Röntgendiagnostik pulmonaler Manifestationen bei HIV-Patienten unter hochaktiver antiretroviraler Therapie (HAART). Fortschr Röntgenstr 2005;177:
- Tschampa HJ, Sommer T, Tasci S. Pneumocystis-Pneumonie – atypische Manifestation bei einer immunsupprimierten Patientin mit rheumatoider Arthritis. Fortschr Röntgenstr 2004;176:1843–8.[CrossRef]
- Shah RM, Miller WTJ. Widespread ground-glass-opacity in consecutive cases: does lobar distribution assist diagnosis? AJR Am J Roentgenol 2003;180:965–8.[Abstract/Free Full Text]
- Franquet T, Müller NL, Lee KS, Giménez A, Flint JD. High-resolution CT and pathologic findings of noninfectuous pulmonary complications after hematopoetic stem cell transplantation. AJR Am J Roentgenol 2005;184:629–37.[Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
L. B. Gadkowski and J. E. Stout
Cavitary Pulmonary Disease
Clin. Microbiol. Rev.,
April 1, 2008;
21(2):
305 - 333.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
