Rosita M. Shah

Significance of ground-glass opacity on HRCT in long-term follow-up of patients with systemic sclerosis.

Despite little supportive evidence, ground-glass opacity at high-resolution computed tomography, without other signs of fibrosis, has been equated with potentially reversible disease, and in systemic sclerosis, frequently prompts aggressive anti-inflammatory treatment to prevent pulmonary fibrosis. Our study evaluates ground-glass opacity on sequential high-resolution computed tomography in 41 patients with systemic sclerosis over a mean follow-up period of 27 months (r6 to 60 mo). Ground-glass opacity was the most common imaging finding, present in 66%, and usually associated with other signs of interstitial disease, including nonfibrotic interstitial opacities in 27% and fibrotic interstitial opacities in 32%. Improvement was only documented in 2(5%) patients with ground glass and nonfibrotic interstitial opacities. In systemic sclerosis, ground-glass opacity is most commonly associated with irreversible disease. Disease progression or improvement could not be predicted by the presence of ground-glass opacity.

Bronchioloalveolar cell carcinoma: impact of histology on dominant CT pattern.

The authors set out to determine how histologic variability in bronchioloalveolar cell carcinoma impacts dominant radiographic patterns shown by computed tomography (CT). Thoracic CTs of all patients with pathologically confirmed bronchioloalveolar cell carcinoma diagnosed over a 36-month period were reviewed without knowledge of underlying histologic type. The dominant CT pattern was recorded as 1) air space consolidation; 2) focal nodule or mass; and 3) multicentric nodules or masses. Nodules and masses were further characterized according to borders, distribution, and associated findings, including spiculations and air bronchograms. Histology was independently reviewed. Twenty-seven patients, 16 women and 11 men, mean age 60 years, were diagnosed with bronchioloalveolar cell carcinoma. In 6 (22%) of the 27 cases, the histology was mucinous, with malignant goblet cells identified. Five (83%) of the six mucinous neoplasms manifested as air space consolidation and three (50%) of the six presented with multiple nodules, in which two had coexisting air space consolidation. Of the remaining 21 cases (78%) with nonmucinous histology, the primary malignant cells of origin included Clara cells (n = 8), tall columnar epithelial cells (n = 7) and alveolar type II pneumocytes (n = 6). Sclerosis was a dominant histologic feature in 14 (67%) of the 21 cases. Seventeen (81%) of the nonmucinous neoplasms presented as isolated nodules or masses and four (19%) presented as multiple nodules or masses. Of these four patients with multifocal disease and nonmucinous histology, multiple bronchioloalveolar adenomas accounted for multicentricity in two of the patients. Significant correlations included air space consolidation with mucinous histology (p = 0.001) and focal nodule or mass with nonmucinous histology (p = 0.001). At CT of bronchioloalveolar cell carcinoma, the patterns of air-space consolidation correlate with mucinous histology and isolated nodules or masses with nonmucinous histology. The pattern of multiple nodules or masses, however, did not correlate with histology. Coexisting bronchioloalveolar adenomas can contribute to apparent multicentric disease in patients with nonmucinous histology.

Spectrum of CT findings in nosocomial Pseudomonas aeruginosa pneumonia.

The purpose of this study was to evaluate CT findings in nosocomial Pseudomonas aeruginosa Pneumonia (PAP) and to compare features of PAP in patients with isolated P. aeruginosa cultures and those with coexistent infections. A retrospective database search revealed 28 patients with nosocomial PAP (12 men, 16 women; mean age, 57 years) in which thoracic CT had been performed within a mean of 1.7 days from the time of respiratory culture. Two chest radiologists blinded to culture data performed a consensus reading noting distribution and pattern of consolidation, ground-glass opacity, nodules, peribronchial infiltration, necrosis, effusions, and pleural enhancement. Coexistent respiratory cultures were recorded. Consolidation was present in all patients, involving multiple lobes in 23 (82%) and demonstrating upper zonal involvement in 23 (82%). Nodular features were present in 14 (50%), including tree-in-bud patterns with centrilobular distributions in 9 (64%) and larger, randomly distributed nodules in 5 (36%). Five of five patients with consolidations limited to the lower lung zones had associated upper lung nodules. Ground-glass opacity was seen in nine (31%) and peribronchial infiltration in 16 (57%). Necrosis was present in eight (29%). Thirteen (46%) bilateral and five (18%) unilateral pleural effusions were present with enhancement occurring in two (1%). Coexistent positive respiratory cultures were identified in 13 patients. The distribution of consolidation, frequency and distribution of nodules, and frequency of necrosis did not differ significantly between patients with and without other positive cultures. With CT, PAP most commonly presents with multifocal airspace consolidation. Nodular features were identified in half, with one-third demonstrating tree-in-bud opacities. Unsuspected necrosis occurred in one-third of cases. CT findings in patients with and without other respiratory isolates did not differ in the distribution and frequency of consolidations, nodularity, or necrosis.

Image-guided localization for video-assisted thoracic surgery

Video-assisted thoracic surgery (VATS) has become a useful diagnostic and therapeutic tool in the management of lung, pleural, and mediatstinal disease. Preoperative image-guided localization is performed to aid the surgeon in the thoracoscopic resection of small lung lesions that would otherwise be difficult to resect. This article describes the techniques of localization and reviews our experience with this procedure. While the majority of localization procedures are performed during an immediately preoperative computed tomography (CT), the use of intraoperative lesion localization using an endosonographic probe has been reported. The need for localization before resection is dependent on the skill and experience of the thoracoscopist and the characteristics of the lung lesions.

Role of thoracoscopy and preoperative localization procedures in the diagnosis and management of pulmonary pathology.

Video-assisted thoracic surgery is an important component of modern thoracic surgery, providing a safe, less invasive alternative to open thoracotomy in the evaluation of pleural, mediastinal, and parenchymal pathology. Advancements in endoscopic techniques and video-optics have permitted greater visualization of the thoracic cavity and allowed limited pulmonary resections with significantly reduced postoperative morbidity. Thoracoscopy is indicated for diagnosis of intrathoracic pathology when usual methods of diagnosis, including fine-needle aspiration and transbronchial biopsy, are inconclusive. The diagnostic accuracy of video-assisted thoracic surgery approaches 100%. Increasingly, the indications for thoracoscopy include therapeutic resections of pulmonary nodules in cases of limited lung metastases and bronchogenic carcinoma when pulmonary function is poor. Successful diagnostic and therapeutic resection by thoracoscopy requires intraoperative localization of the lesion within the collapsed lung. The indications and methods of thoracoscopic surgery and preoperative localization are discussed.

Adjacent parenchymal abnormalities in peripheral bronchogenic carcinoma: correlation of thin-section CT with histology.

Our purpose is to correlate thin section CT of peripheral bronchogenic carcinomas with histologically detected lymphatic or vascular invasion. Retrospective 3-year database search revealed 186 surgical resections for primary bronchogenic carcinoma, of which 58 had available preoperative imaging performed at our institution. Cases with prior surgery, nonconfirmatory pathology, remote imaging, or central location were excluded, resulting in a study population of 42 patients, 25 men, 17 women, with a mean age of 69 years. Imaging with 1–3 mm collimation was performed within a mean of 32 days prior to surgery. Histologic diagnoses included adenocarcinoma (n = 24, 57%), squamous cell carcinoma (n = 13, 31%), large cell carcinoma (n = 4, 10%), and small cell carcinoma (n = 1, 2%), with a mean tumor size of 27 mm. Three radiologists blindly and independently recorded bronchovascular thickening, septal and nonseptal opacities, and the extent of each beyond tumor margins: 1) <5 mm, 2) 5–10 mm, and 3) >10 mm. Lymphangio-invasion was correlated with imaging findings, tumor size, and histology. Adjacent parenchymal abnormalities were recorded in 40 (95%) of 42 masses, with isolated nonseptal opacities representing the most frequent abnormality in 21 (50%), followed by bronchovascular thickening in 16 (38%), and septal opacities in 12 (29%). Lymphangio-invasion was present in 16 (38%) of cases. The frequency of lymphangio-invasion was highest (53%) in cases with 2 or more positive findings, and extension beyond 10mm from the tumor margin. This trend did not achieve statistical significance by ROC analysis. Lymphangio-invasion was positively correlated with tumor size, P = .03, but not histology. In conclusion, parenchymal abnormalities beyond tumor margins shown by CT may be due to lymphangio-invasion but imaging findings did not reliably distinguish cases with and without lymphangio-invasion.

CT manifestations of human immunodeficiency virus (HIV)-related pulmonary infections

The infectious pulmonary complications of acquired immunodeficiency syndrome (AIDS) are reviewed, with emphasis on the spectrum of CT imaging findings and diagnostic accuracy and limitations as reported in the current literature. Changes in epidemiologic trends for common AIDS-related infections and the associated ranges of CD4 lymphocyte counts, when these infections are typically encountered, are discussed.