CHARACTERIZATION OF THE PATHOGENESIS OF LYMPHANGIOLEIOMYOMATOSIS (LAM)

点击翻译按钮获取中文摘要

(A) Lymphangioleiomyomatosis (LAM), a rare disease associated with progressive cystic changes in the lung and abdominal tumors, occurs primarily in premenopausal women leading to the hypothesis that estrogen plays an important role in its pathogenesis. For this reason, hormonal manipulation, including progesterone and oophorectomy, is often used to control the disease. These therapeutic modalities may place patients at higher risk for osteoporosis and osteopenia. Using dual energy x-ray absorptiometry (DXA), we measured bone mineral density (BMD) at lumbar spine (lateral and anteroposterior) and proximal femur (total, neck, trochanter, intertrochanter, Ward's triangle) in 59 women with LAM and without a history of significant use of steroids. Among 48 women in whom BMD could be accurately measured for both proximal femur and lateral lumbar spines, 14 (29.2%) had osteoporosis and 26 (54.2%) had osteopenia (a total of 83.3% had low bone density) of the lateral spine and/or total proximal femur. Importantly, of 16 postmenopausal women, 13 (81.3%) had osteoporosis and 2 (12.5%) had osteopenia (i.e., a total of 93.8% had low bone density). Statistical analyses using logistic regression and multiple linear regression indicate that menopause (most due to oophorectomy), progesterone treatment, and poor pulmonary function (low FEV1% or RV/TLC%) were all independent factors contributing to the low mineral density seen in these patients. Patients with LAM are at very high risk for osteopenia and osteoporosis, may be appropriate candidates for early intervention to reduce bone loss, and should undergo screening for BMD. B. LAM is a rare disease, which occurs almost exclusively in women of childbearing age. The cardinal histologic feature of LAM is a proliferation of abnormal smooth muscle cells (LAM cells) in lung and along axial lymphatics. The clinical characteristics of LAM include recurrent spontaneous pneumothorax, slowly progressive dyspnea, hemoptysis, chylothorax and chylous ascites. Interestingly, serum CA125, a marker of nonmucinous epithelial ovarian cancer, was reported to be elevated in two LAM patients with chylous ascites and/or pleural effusion. It was postulated that CA125 might be produced by LAM cells. To investigate the source of elevated CA125 in LAM patients, serum CA125 levels in 36 LAM patients were measured by radioimmunoassay using a monoclonal antibody, OC125. Lung specimens from two LAM patients with high CA125 levels and normal tissue from a lung resected for lung cancer were examined microscopically after immunofluorescent standing with OC125. The serum CA125 level was highest in three patients with both chylothorax and ascites (mean+/-SD 127+/-71 U/ml), intermediate in three patients with either pleural effusion or ascites (42+/-7 U/ml), and lowest in thirty patients without pleural effusion or ascites (18+/-24 U/ml). A positive correlation between serum CA125 level and the amount of chylous pleural effusion was clearly evident in one patient. In lung tissues from patients with high serum CA125 levels, CA125 was present at the pleural surface, but not in LAM cells. No CA125 was found on the pleural surface of normal tissue. Ovarian tumors were not found in any patient by computed tomography of pelvis. The pleural and peritoneal mesothelia and not LAM cells are likely the major sources of elevated serum CA125 in LAM patients. C. Thirty-eight patients with LAM were evaluated with HRCT and pulmonary function tests. Extent of disease on HRCT was assessed independently by two radiologists and graded according to the percentage of lung volume judged abnormal as follows: Grade 0=normal; Grade 1=<30% abnormal; Grade 2=30%-60% abnormal; and Grade 3=<60% abnormal. The qualitative extent of parenchymal disease and its frequency in patients with LAM was: Grade 1 (10 patients); Grade 2 (14 patients); Grade 3 (14 patients). We used a variation of the methods of Crausman et al. (Chest 1996;109:131-137) to assess quantitatively the total area of the lungs, and the area covered by pulmonary cysts, a hallmark of LAM. The area covered by normal lung was obtained by subtracting the area covered by the cysts from the total lung area. Ratios of areas of normal lung/total area and abnormal lung/total area were then compared to the qualitative measurements of extent of lung disease and to the results of PFTs. Total lung area was positively associated with TLC (p<0.05). A negative association was observed between both cyst area and cyst area/total lung area and both FEV1 and FEV1% predicted (p<0.001). Cyst area and cyst area/total lung area were positively associated with RV/TLC (p<0.001) and negatively associated with DLCO (p<0.001). Thus, in LAM quantitative HRCT correlated well with PFTs over a broad range of disease severity.