Empyemas develop in three distinct stages, exudative, fibropurulent, and organizing, with important therapeutic implications, although it may be difficult to differentiate them radiographically (1). Contrast-enhanced CT has played a role in detecting chest wall changes in patients with empyema and change of parietal pleura, which may help to suggest occult pleural space infections, and has influenced therapeutic decisions (2). Catheter drainage has been reported as a primary method of empyema management, with cure rate of 80–90%(3,4). Fibrinolytic therapy has been tried in patients with empyema at the organizing stage (3,4). Park et al. (5) reported that it was not effective in pleural fluid with a honeycomb appearance on sonography or in parietal pleura with a thickness of >5 mm on CT scans. Decortication is still regarded as the treatment of choice for chronic empyema due to the propensity of pleural peel to cause lung entrapment with impairment of pulmonary function (1).
Without adequate treatment, the end result of empyema is pleural fibrosis. The ingrowth of fibroblasts along the pleural peel may occur over a more prolonged period of time and may result in thick-walled loculated pleural fluid collection. There is a particular tendency for this to occur in patients with partially treated tuberculosis (6).
The parietal pleural peel and extrapleural space in patients with chronic empyema have variable thickness and sometimes have layers with variable attenuation at CT scan (2).
There has been no report about what these layers at CT scan represent histopathologically. The purpose of our study was to describe CT findings of parietal pleura and extrapleural space in patients with chronic tuberculous empyema and to compare them with histopathologic findings in patients who underwent decortication.
MATERIALS AND METHODS
We performed a retrospective study of 13 patients with chronic tuberculous empyema whose chest CT scans and pathologic slides were available for review; the subjects were selected from 36 consecutive patients who underwent decortication or pleuropneumonectomy between July 1995 and November 1998 at our institution. Patients whose CT scans were obtained >1 month before decortication, whose etiology was confirmed as other than tuberculosis or uncertain, and whose pathologic slides did not show full thickness from the layer of necrotic debris on the pleural surface to the layer of extrapleural fat were excluded. All patients were men from 18 to 67 years old (mean 42 years old). Presenting symptoms were cough and dyspnea (n = 6), pleuritic chest pain or discomfort (n = 4), and hemoptysis (n = 2). One patient was asymptomatic. All patients had a history of antituberculous medication for pleurisy or empyema. Nine patients had previously undergone thoracentesis or thoracostomy. Decortication was performed at an average of 14.1 days (range 4 days to 1 month) after CT examination.
Contrast-enhanced CT scans were obtained with 10 mm collimation and 13 mm intervals (pitch = 1.3) by using Somatom plus S (Seimens, Erlangen, Germany; n = 2), Somatom plus 4 scanner (Siemens; n = 10), or HiSpeed scanner (GE Medical Systems, Milwaukee, WI, U.S.A.) (n = 1). Contrast material (60% iodine content) was injected at 2.5 ml/s for 100 ml with scan delay of 35 s.
The CT scans were reviewed by two chest radiologists. CT scans were assessed specifically for changes in the parietal pleura and the extrapleural space. The designations of the parietal pleura and extrapleural space used in this study are as follows: Anatomically, the parietal pleura is composed of four major layers: a single layer of mesothelial cells, a thin supportive fibroelastic layer, a layer of loose connective tissue, and the endothoracic fascia, a second fibroelastic layer that tethers the pleura to the overlying rib. Despite the inclusion of this layer as an anatomic component of the parietal pleura, the loose areolar connective tissue layer has been referred to as subpleural and extrapleural, a designation used herein (2). The combined mesothelial and submesothelial supportive fibroelastic layers are referred to herein as the parietal pleura.
The parietal pleura were evaluated for contrast enhancement, thickness, presence of a layer within the enhanced parietal pleura, and calcification. The extrapleural fat adjacent to the ribs was evaluated for visibility, thickness, and attenuation. The presence of extrapleural fat was considered thickened if it was >3 mm thick (2). That attenuation of the extrapleural fat was estimated to be the same as that of the chest wall fat or intermediate between that of fat and musculature.
The characteristics of attenuation within the extrapleural space were also evaluated.
The cause of chronic tuberculous empyema was diagnosed by means of histologic findings or clinical history. Tuberculosis was confirmed in 13 patients by histologic findings that showed chronic granulomatous inflammation with or without caseation necrosis (n = 10) and clinical history of tuberculosis (n = 3). Acid-fast bacilli were proven in the decorticated specimen in two patients.
CT-histopathologic correlation was retrospectively attained by an experienced pathologist and two chest radiologists. Five to seven pathologic slides with hematoxylin-eosin stain were reviewed in each patient. Two representative slides of decorticated specimens that were sectioned with full thickness (from necrotic debris on the pleural surface to extrapleural fat layer) and length of 3–5 cm were selected. The thickness of each layer of different histopathology on the pathologic slides was measured. Because the parietal pleural peel showed even thickness in most sections on CT scan, the layers with different histopathology on the slides and the layers showing different enhancement on CT were matched to each other by their thickness. The layer of extrapleural fat was a baseline for measuring thickness of parietal pleura both on pathologic slides and on CT scans. Then, histopathologic findings were analyzed.
Contrast-Enhanced CT Findings
Twelve of 13 patients showed enhancement of parietal pleura with variable thickness of 3–15 mm (mean 5.7 mm) (Fig. 1). Pleural calcification was noted in four patients, either partial (n = 3) (Fig. 2) or extensive (n = 1). Two patients showed a low attenuated layer of 2 mm in the middle of the enhanced parietal pleural peel (Figs. 2 and 3). The inner layer of the enhanced parietal pleura was 3 and 2 mm, and the outer layer was 2 and 10 mm, respectively. The inner surfaces of the enhanced parietal pleural peel were smooth in all patients except one. In one patient, peripheral rim-enhancing nodules were noted on the surface of the parietal pleural peel (Fig. 4).
Extrapleural space with intermediate attenuation was noted in six patients with 5–15 mm in thickness (Figs. 4 and 5). In one patient, multiple enhancing septa were noted within the extrapleural tissue with intermediate attenuation. Extrapleural space with fatty attenuation was noted in nine patients with 3 to 10 mm thickness (mean 6.2 mm). Linear soft tissue attenuation interrupting the extrapleural fat was noted in seven patients (Fig. 2).
Correlation of Contrast-Enhanced CT and Pathologic Findings
The thickened enhancing parietal pleura on CT corresponded histopathologically to layers of fibrous collagen tissue with varying degrees of inflammation (Fig. 1). The layer of low attenuation within the enhanced parietal pleura on CT corresponded histopathologically to a layer of histiocytes or caseation necrosis between the layers of collagenous fibrous tissue with inflammatory cells (Figs. 2 and 3). The peripheral rim-enhancing nodules on the surface of the enhanced parietal pleura represented tuberculous granulomas with central caseation necrosis histopathologically (Fig. 4).
Extrapleural space with intermediate attenuation on CT corresponded histopathologically to the layer of proliferation of vessels, inflammatory cells, and granulomas (Fig. 4). The multiple enhancing septa within the extrapleural tissue with intermediate attenuation revealed histopathologically multiple granulomas with caseation necrosis surrounding fibrous septa (Fig. 5). Linear soft tissue attenuation interrupting the extrapleural fat corresponded histopathologically to the layers of fibrous tissue with proliferation of vessels and inflammatory cells shown in six patients (Fig. 2). No muscular fibers were found.
The CT detection and anatomic correlation of the layers allow confident isolation of the deeper mesothelial and submesothelial fibroelastic layers as that portion of the pleura thickened by fibrin deposition and the ingrowth of capillaries and fibroblasts during formation of a pleural peel in patients with empyema (2). External to the parietal pleural peel, the subcostal extrapleural space shows fat attenuation, intermediate attenuation, or soft tissue attenuation interrupting the extrapleural fat. If active infection remains at the organizing stage of empyema or to relieve the entrapped lung, decortication may be required for treatment (3). Delay in diagnosis of empyema, inadequate chest tube drainage, or inappropriate antibiotic therapy make the need for decortication more likely (3,7–10). The extrapleural fat layer serves as the cleavage plane during decortication. In this study, we correlated the parietal pleural peel and extrapleural space seen on CT to the pathologic specimens following decortication in patients with chronic tuberculous empyema.
The parietal pleural peel in chronic empyema was enhanced with variable thickness (3 to 15 mm) on CT scan. This layer was seen histopathologically as fibrous collagen tissue with few inflammatory cells located internal to extrapleural fatty tissue.
In the organizing stage of empyema, there is an ingrowth of fibroblasts along the fibrin sheets lining the visceral and parietal pleura. With time, this membrane may remain thin or become thicker and denser and its cellularity becomes progressively less (1).
The layer of low attenuation within the parietal pleural peel was noted in two patients on CT. These layers corresponded pathologically to the layers of histiocyte or caseation necrosis between collagenous fibrous layers with varying degrees of inflammation. We could not explain the exact pathogenesis or time sequence of these layers because the patients had no previous CT scans for comparison. Nevertheless, these findings might indicate recurrent, but different episodes of, infections. The layers of histiocytes or of caseation necrosis might be entrapped between the fibrous collagen layer during the other episodes of infection, showing different degrees of inflammatory cells. Because the ingrowth of capillaries and inflammatory process occur from the chest wall side rather than cavity side, the outer layer of the enhanced parietal pleural peel is likely later infection, which showed more inflammatory cells and capillaries histopathologically.
Seven of 13 patients with empyema in our study demonstrated the extrapleural fat thickening of >3 mm in thickness on CT. Thickening of the extrapleural fat is frequently noted in patients with chronic empyema (2,11). Intermediate attenuation in the extrapleural space on CT showed granulomas or collagen tissue with proliferation of vessels, inflammatory cells, and granuloma histopathologically. On CT, granulomas with central caseation necrosis showed low attenuation center with peripheral enhancement in two patients, which had multiple septum-like lesions in extrapleural tissue with intermediate attenuation or nodules with peripheral enhancement on the surface of the pleural peel. Intermediate attenuation in the extrapleural space with or without enhancing septa might indicate an ongoing inflammatory process. Thickening and increased attenuation of extrapleural fat constituted another finding that was suggestive of pleural inflammation or infection and had been reported in patients with empyema, malignancy, and asbestos exposure (2,11,12). Approximately 75% of patients with the late stage of fibrinopurulent and organizing stage of empyema showed thickened increased attenuation of the extrapleural fat. These findings may show persistent thickening of the extrapleural tissues but with return to homogeneous fat attenuation in patients with resolved empyema at follow-up on CT scan (2).
Waite et al. (2) observed that streaky soft tissue areas of attenuation interrupted the subcostal fat, and they determined that these areas represented the subcostal muscle without pathologic correlation. However, we found no muscle fibers in pathologic specimens. In our study, the soft tissue area of attenuation was a fibrous band with proliferation of vessels and inflammatory cells, which was identified by the fact of attachment to the outermost portion of extrapleural fat.
The retrospective nature of this study did not allow for a true one-to-one correlation between the CT findings and pathologic specimens. Nevertheless, with the layer of extrapleural fat as a baseline, both on CT and on pathologic slides, we measured the thickness of each layer with different histology on pathologic slides following correlation of the layers showing different enhancement by their thickness on CT images. The parietal pleural peel had even thickness in most of the sections on CT; therefore, the layers of different histology matched by their thickness with CT images should provide no chance to correlate incorrectly. Furthermore, no other tissues except the layer of fibrous tissue external to extrapleural fatty tissue on pathologic slides, which were matched to the soft tissue areas of attenuation, interrupted extrapleural fat on CT.
In summary, the enhancing parietal pleura was collagenous fibrous tissue with varying degrees of inflammation. The low attenuation layer within the enhancing parietal pleura was a layer of histiocytes or caseation necrosis, suggesting different episodes of infection. The intermediate attenuation area in the extrapleural space was an ongoing inflammatory process showing granulomas and collagen tissue with proliferation of vessels, inflammatory cells, and granulomas. The linear soft tissue attenuation interrupting the extrapleural fat was a fibrotic band rather than the subcostalis muscle.
1. Armstrong P, Wilson AG, Dee P, et al., eds. Imaging of Diseases of the Chest.
2nd ed. St. Louis: Mosby-Year Book, 1995:641–703.
2. Waite RJ, Carbonneau RJ, Balikian JP, et al. Parietal pleural changes in empyema: appearances at CT. Radiology 1990; 173:145–50.
3. Kaplan DR. Treatment of empyema thoracis [Editorial]. Thorax 1994; 49:845–6.
4. Neff CC, vanSonnenberg E, Lawson DW, et al. CT follow-up of empyemas; pleural peels resolve after percutaneous catheter drainage. Radiology 1990; 176:195–7.
5. Park CS, Chung WM, Lim MK, et al. Transcatheter instillation of urokinase into loculated pleural effusion: analysis of treatment effect. AJR 1996; 167:649–52.
6. Patten W, Watson T, Gaensler E. Pulmonary function, before and at intervals after surgical decortication of the lung. Surg Gynecol Obstet 1952; 95:477.
7. McLoud TC, Flower CDR. Imaging the pleura: sonography, CT, and MR imaging. AJR 1991; 145:1145–53.
8. Muller NL. Imaging of the pleura. Radiology 1993; 186:297–309.
9. Alfageme I, Munoz F, Pena N, et al. Empyema of the thorax in adults: etiology, microbiologic findings, and management. Chest 1993; 103:839–43.
10. Schmitt WGH, Hubener KH, Rucker HC. Pleural calcification with persistent effusion. Radiology 1983; 149:633–8.
11. Im JG, Webb WR, Rosen A, et al. Costal pleura: appearances at high-resolution CT. Radiology 1989; 171:125–31.
12. Takasugi JE, Godwin JD, Teefey SA. The extrapleural fat in empyema: CT appearance. Br J Radiol 1991; 64:580–3.
Index Terms: Empyema; Pleura; Tuberculosis; Computed tomography
© 2001 Lippincott Williams & Wilkins, Inc.