Search

The aim was to characterize the computed tomographic (CT) findings from Japanese patients with lymphangioleiomyomatosis (LAM).CT scans of the chest, abdomen, and pelvis from 124 patients with sporadic LAM (S-LAM, mean age, 37.4 years) and 14 patients with tuberous sclerosis complex (TSC)-LAM (mean age, 35.6 years) were analyzed.Pulmonary nodules (18.8%) and hepatic angiomyolipoma (AML, 24.3%) were more common in our patients than those in previous reports. Compared with TSC-LAM, S-LAM group had a higher frequency of pulmonary nodules (28.6% vs 32.3%, P0.01) and lower frequencies of air-space consolidation (21.4% vs 2.4%, P0.01), pneumothorax (28.6% vs 8.1%, P=0.02), pulmonary hilar lymphadenopathy (14.3% vs 0.8%, P0.01), renal AML (85.7% vs 17.4%, P0.01), hepatic AML (71.4% vs 17.4%, P0.01), and retrocrural lymphadenopathy (14.3% vs 1.4%, P=0.04). Axial lymphatic abnormalities (i.e., thoracic duct dilatation, lymphadenopathy, and lymphangioleiomyoma) were most common in the pelvis and tended to decrease in incidence with increased distance from the pelvis.The incidence of some CT findings in Japanese patients differed from those in previous reports. Axial lymphatic abnormalities noted here suggest that the origin of LAM cells may be the pelvis.

Lymphangioleiomyomatosis (LAM) is a destructive lung disease that share clinical, physiologic, and radiologic features with chronic obstructive pulmonary disease (COPD). This study aims to identify those features that are unique to LAM by using quantitative CT analysis.We measured total cross-sectional areas of small pulmonary vessels (CSA) less than 5mm(2) and 5-10mm(2) and calculated percentages of those lung areas (%CSA), respectively, in 50 LAM and 42 COPD patients. The extent of cystic destruction (LAA%) and mean parenchymal CT value were also calculated and correlated with pulmonary function.The diffusing capacity for carbon monoxide/alveolar volume (DL(CO)/VA %predicted) was similar for both groups (LAM, 44.4 ± 19.8% vs. COPD, 45.7 ± 16.0%, p=0.763), but less tissue damage occurred in LAM than COPD (LAA% 21.7 ± 16.3% vs. 29.3 ± 17.0; p0.05). Pulmonary function correlated negatively with LAA% (p0.001) in both groups, yet the correlation with %CSA was significant only in COPD (p0.001). When the same analysis was conducted in two groups with equal levels of LAA% and DL(CO)/VA %predicted, %CSA and mean parenchymal CT value were still greater for LAM than COPD (p0.05).Quantitative CT analysis revealing a correlation between cystic destruction and CSA in COPD but not LAM indicates that this approach successfully reflects different mechanisms governing the two pathologic courses. Such determinations of small pulmonary vessel density may serve to differentiate LAM from COPD even in patients with severe lung destruction.

Backgrounds Since Birt–Hogg–Dube syndrome (BHDS) and lymphangioleiomyomatosis (LAM) share some clinical manifestations (multiple pulmonary cysts with pneumothorax, renal tumors, and sometimes skin lesions), the differential diagnosis of the two diseases becomes problem especially in female patients. This study aims to quantify pulmonary cysts in computed tomography (CT) of females with BHDS and those with LAM and also to identify the independent parameters for differentiating the two diseases. Methods Fourteen patients with BHDS and 52 with LAM were studied. In CT scans, lung fields were defined as areas with fewer than −200 Hounsfield units (HU) and pulmonary cysts as areas consisting of 10 or more consecutive pixels with fewer than −960 HU. The extent, number, size and circularity of cysts were calculated by using hand-made software and compared between the two diseases. Moreover, the lung fields were divided into six zones and analyzed for the distribution of cysts. Finally, a stepwise discriminant analysis employing quantitative measurements of cysts and clinical features was performed. Results The two diseases were significantly different in all quantitative measurements of cysts. Stepwise discriminant analysis accepted the following four variables: the family history of pneumothorax within the second degree relatives, lower-medial zone predominance of cysts, diffusing capacity and mean size of cysts in this order. Conclusion The quantitative characteristics of pulmonary cysts are significantly different between BHDS and LAM. The independent parameters for differentiating the two diseases are the family history of pneumothorax, zonal predominance of cysts, diffusing capacity and size of cysts.

Purpose To describe in detail the characteristic chest computed tomography (CT) findings of Birt–Hogg–Dube (BHD) syndrome. Materials and methods Thin-section chest CT scans of consecutive 12 patients with genetically diagnosed BHD syndrome were retrospectively evaluated by two observers, especially about the characteristics (distribution, number, size, shape and relation to pleura) of pulmonary cysts. Interobserver agreement in the identification of abnormalities on the CT images was achieved using the κ statistic, and the degree of interobserver correlation for the characterization of pulmonary cysts was assessed using the Spearman rank correlation coefficient. Results Multiple pulmonary cysts were seen in all patients. The number of cysts in each patient was various (range, 29–407), and cysts of various sizes (from a few mm to 2 cm or more) were seen in all patient. 76.6% (mean) of cysts were irregular-shaped, and 40.5% (mean) of cysts were located along the pleura. The mean extent score of cysts was 13% of the whole lung, and the distribution of cysts was predominantly in the lower medial zone. Finally, cysts abutting or including the proximal portions of lower pulmonary arteries or veins were also seen in all patients. Conclusion Multiple, irregular-shaped cysts of various sizes with lower medial lung zone predominance are characteristic CT findings of BHD syndrome. Cysts abutting or including the proximal portions of lower pulmonary arteries or veins may also exist in this syndrome in a high probability.