Lymphangioleiomyomatosis (LAM) is a rare disease, occurring in women, characterized by cystic degeneration of the lungs, abdominal tumors, and proliferation of abnormal smooth muscle cells. Lung function abnormalities consist of impairment of the diffusion capacity (Dl CO) and airflow obstruction. The objective of this study was to correlate the functional impairment with histologic measures of disease severity to identify predictors of disease outcome. Lung function of 143 patients and lung biopsies of 74 of these patients were reviewed for evidence of airway disease and scoring of disease severity. A positive response to bronchodilators was associated with more severe airflow obstruction, a predominantly solid pattern of LAM lesions in the lung biopsy, and greater rate of decline in expiratory flow. Airway inflammation, present in 61% of the lung specimens, was not associated with reversible airway obstruction and did not correlate with the severity of airflow obstruction. Dl CO correlated best with the LAM histologic score (LHS), a demonstrated predictor of outcome. We conclude that reversible airway obstruction is found in LAM patients with accelerated loss of lung function and a predominantly solid pattern of LAM lesions. Impairment of Dl CO correlates with LHS, a predictor of survival and time to lung transplantation.
Keywords: lymphangioleiomyomatosis; respiratory function tests; histology; outcome measures
Lymphangioleiomyomatosis (LAM), a disease affecting women of childbearing age, is characterized by recurrent spontaneous pneumothoraces, chylothorax, chylous ascites, and angiomyolipomas (1-3). The primary pathologic findings in LAM include the proliferation of immature-appearing smooth muscle cells (LAM cells) in the lung and along axial lymphatics in the thorax and abdomen. The proliferation of LAM cells leads to the formation of thin-walled cysts in the lungs and large, fluid-filled cystic structures in the axial lymphatics. Angiomyolipomas, which occur predominantly in the kidney or elsewhere in the abdomen (1-3), are characterized by abnormal smooth muscle cells, similar to those found in the lungs and along the axial lymphatics, and the presence of adipose tissue, intermixed with incompletely developed vascular structures.
The most common abnormalities of pulmonary function in LAM are airflow obstruction, caused by compression of the airways by smooth muscle cell proliferation or loss of lung elastic recoil, and decreased diffusion capacity of the lungs for carbon monoxide (Dl CO), which probably reflects loss of gas exchange area. A positive response to inhaled bronchodilators has been reported (1). However, it is not known whether a positive response to bronchodilators has prognostic significance in LAM. Although a study of the rate of decline in lung function in LAM was recently published (4), the relationship between the rate of decline in lung function, lung histology, and the presence of reversible airway obstruction was not examined.
The purpose of this research was to correlate histologic characteristics of LAM with pulmonary function, and to identify predictors of disease outcome in LAM. We found that Dl CO correlates best with histologic severity of disease. Predominance of a solid pattern of LAM lesions, reversible airway obstruction, and decline in gas exchange are all predictors of outcome in patients with LAM.
The study population consisted of 143 patients with LAM referred by their physicians to the National Institutes of Health between 1996 and 2000 (1). The research was approved by the institutional review board of the National Heart, Lung, and Blood Institute and informed consent was obtained from all participants. The diagnosis of LAM was made by lung biopsy (84 patients), retroperitoneal node biopsy, excised angiomyolipomas, and clinical and roentgenographic data. Patients were evaluated every 6 mo (Table 1). The study was both retrospective and prospective. Patients who had undergone lung transplantation were excluded from the study. Most of the patients were being treated with progesterone and inhaled bronchodilators.
Surgical lung biopsies from 74 patients were evaluated and scored, in a blinded fashion, by two pathologists, for the presence of bronchiolitis (scores 0–1) and whether the LAM lesion showed a predominantly cystic (scores 0–3) or solid pattern (scores 0–1). A LAM histologic score (LHS) was determined on the basis of the extent of involvement of the lung tissue by the cystic lesions and infiltration by LAM cells and was graded as follows: LHS-1 < 25%; LHS-2 = 25 to 50%; and LHS-3 > 50% (5).
Lung volumes, flow rates, and diffusion capacity were measured in the sitting position using a computerized system (Collins Gold Standard Plus; Warren E. Collins, Braintree, MA or Master Screen PFT; Erich Jaeger, Wuerzburg, Germany) according to the American Thoracic Society (ATS) recommendations (6-8). Values were expressed as a percentage of predicted normal values using the data of Morris and coworkers (9), Knudson and coworkers (10), and Crapo and Morris (11). FEV1 and FVC were measured before and after administration of albuterol, either 2.5 mg via nebulizer or 180 μg via a metered-dose inhaler. A positive response to bronchodilators was defined according to ATS recommendations (6). The average yearly decline in FEV1 and Dl CO was determined by subtracting the results of the most recent test from the results of the first study and dividing it by the number of years.
We used the generalized estimating equations methodology to determine which factors were associated with a positive response to bronchodilators (12, 13). Only the variables that had a value of p < 0.10 after the univariate analysis were included in a full logistic regression. After calculating the proportion of positive responses to bronchodilators and the average change per year in either FEV1 or Dl CO for each subject, we ran a linear regression model with change per year in FEV1 or Dl CO as a response variable, and the proportion of response to bronchodilators as the predictor variable. For testing whether the FEV1 and Dl CO were associated with the LHS we used the Fisher's least significance difference method. Finally, we compared the yearly change in FEV1 in subjects with a predominantly solid pattern (score of 1) and a nonsolid pattern (score of 0) using a t test. Data are presented as mean ± SEM.
Of 143 patients studied, 130 were white, seven Asian, three African-American, two Hispanic, and one African. The mean age of the patients at the time of enrollment was 42.7 ± 0.8 yr (Table 1). All patients were female. None of the patients was a current smoker, although 37 were ex-smokers. Four patients had either smoked only an occasional cigarette or had smoked for less than a year. The remaining 33 patients had smoked an average of 18.8 ± 2 pack-years and had quit 7.9 ± 1.3 yr before enrollment in the study. Of the 143 patients, 12 gave a history of being diagnosed with asthma and treated with bronchodilators.
Open lung biopsy specimens from 74 of the 143 patients were available for examination. Airway inflammation, i.e., bronchiolitis, was observed in 47 of the 74 patients. Bronchiolitis tended to occur in airways surrounded by heavy infiltrates of abnormal smooth muscle cells characteristic of LAM (Figure 1A). In 34 of the 74 patients there was a predominantly solid pattern of LAM lesions (Figure 1C). However, cystic lesions of variable severity were present in 93% of the patients (Figure 1B). Pathologic changes suggestive of bronchial asthma were observed in five patients. Four of these patients had mucus plugs in the small airways and three showed goblet cell hyperplasia. Smooth muscle hyperplasia was noted in three of the five patients and eosinophilic infiltration of the bronchiolar wall was observed in two.
The results of the initial pulmonary function tests are shown in Table 2. Decreased FEV1/FVC ratio and FEV1 (airflow obstruction) were present in 68% of the patients. Five patients (3.5%) had low lung volumes (restrictive pattern) without airflow obstruction and 22 patients (16%) had combined restrictive and obstructive abnormalities. The 27 patients with a restrictive defect had a history of pneumothoraces, chylothorax, and pleurodesis. Dl CO abnormalities were found in 62% of the patients. Twenty-seven patients (19%) had normal pulmonary function tests. The mean FEV1 and Dl CO for the 143 patients, expressed as a percentage of normal predicted values, were 69.8 ± 2% and 72 ± 2%, respectively.
Reversible airflow obstruction. Thirty-six of the 143 patients met the preset criteria for a positive response to bronchodilators. A total of 117 tests were performed in these patients and a positive response to bronchodilators was observed in 86% of the tests. A total of 301 tests were performed in the remaining 107 patients. The increase in FEV1 after treatment with bronchodilators in these two groups of patients were respectively, 20.8 ± 0.3% and 6 ± 0.3%. Table 3 shows the results of the multivariate analysis. The probability of a positive response to bronchodilators was significantly correlated with a lower FEV1 and a higher residual volume (RV) and Dl CO.
Pathophysiologic correlations. Histologic data were available for 19 of the 36 patients who met the criteria for a positive bronchodilator response and 55 of the 107 nonresponders. As shown in Figure 2, the patients who responded to bronchodilators had a significantly greater predominance of a solid pattern of LAM lesions than those who did not respond to bronchodilators (0.7 ± 0.1 versus 0.39 ± 0.06, p = 0.021). Conversely, the proportion of positive responses to bronchodilators was significantly greater in patients whose lung biopsies showed a predominantly solid pattern of LAM lesions (0.351 ± 0.07 versus 0.117 ± 0.04, p = 0.002). There was no significant difference in LHS or prevalence of bronchiolar inflammation or cysts between responders and nonresponders (Figure 2). Pulmonary function tests obtained within 6 mo before the date of the biopsy were available in 55 of the 74 patients. FEV1 in patients with a predominantly solid pattern of LAM lesions (n = 27) tended to be lower than in those patients (n = 28) without this pattern (61.9 ± 5% versus 70.4 ± 4%) but the difference was not statistically significant (p = 0.187). There was also no significant difference (p = 0.58) in FEV1 between patients with histologic evidence of bronchiolitis (n = 33) and patients without (n = 22) bronchiolar inflammation (67.6 ± 4.5 versus 64.2 ± 4.1). There was a close association between the LHS and lung function. Figure 3 shows that Dl CO in patients with an LHS of 1 (85.7 ± 6.6%, n = 17) was significantly greater than that in patients with an LHS of 2 (62.8 ± 4.1%, n = 24, p = 0.003) or 3 (47.9 ± 5.8%, n = 14, p < 0.001). As shown also in Figure 3, the FEV1 in patients with an LHS of 1 (79.2 ± 5%) was greater than that in patients with an LHS of 2 (63.8 ± 4%, p = 0.027) or 3 (58.3 ± 8%, p = 0.031).
Rates of decline in FEV1 and Dl CO for patients with a follow-up of more than 3 yr (n = 118) were respectively 84.7 ± 14.6 ml and 0.79 ml/min/mm Hg/yr (Table 4). For patients with a follow-up of more than 5 yr the rates of decline were respectively, 58 ± 9 ml/yr and 0.63 ± 0.1 ml/min/mm Hg/yr. Table 4 and Figure 4 show that the decline in FEV1 in patients with a predominantly solid pattern of LAM lesions in the biopsy specimen (n = 15) was significantly greater than the rate of decline observed in patients without a predominance of this pattern (n = 15): 110 ± 24 versus 42 ± 13 ml/yr (p = 0.0187). Interestingly, Dl CO in the patients with the predominantly solid pattern tended to be higher than in the patients without a predominance of this pattern (Table 4), but the difference was not statistically significant. The rates of decline in FEV1 and Dl CO were evaluated and correlated with the proportion of positive responses to bronchodilators in 118 and 104 patients, respectively (mean follow-up 3.8 ± 0.3 yr). A significant correlation between the rate of decline in FEV1 and the proportion of positive responses to bronchodilators was observed (r = −0.65, p < 0.001). In contrast, a positive response to bronchodilators was not a predictor of the decline in Dl CO (r = −0.067, p = 0.74).
In our study, positive responses to a bronchodilator were observed in 23% of a total of 418 tests performed in 143 patients with LAM. Patients who responded more frequently to bronchodilators had a predominantly solid pattern of LAM lesion surrounding the airways, and relaxation of these cells by β2-adrenergic receptor agonists could account for the increase in expiratory flow. Predictors of a positive response to bronchodilators were a greater severity of airflow obstruction, manifested by a lower FEV1 and a larger RV and Dl CO, and predominance of solid LAM lesions in the lung biopsy specimens. Despite our new finding of a high prevalence of bronchiolitis in airways surrounded by heavy proliferation of LAM cells, airway inflammation was not a predictor of a positive response to bronchodilators. However, a positive response to bronchodilators was a predictor of the severity of airflow obstruction because it was associated not only with a lower FEV1 but also with a greater rate of decline in FEV1. Thus, reversible airflow obstruction may have prognostic significance in LAM. In other obstructive pulmonary diseases, such as bronchial asthma and chronic bronchitis, a positive response to bronchodilators also appears to be associated with a greater rate of decline in lung function (14). These findings, however, should be interpreted with caution because histologic data were available only in 19 of the 36 patients who met the criteria for a positive response to bronchodilators and because the finding of a greater decline in FEV1 in patients with predominantly solid lesions is based on data from only 15 patients.
The presence of reversible airflow obstruction in LAM suggests that airflow limitation may result, at least in part, from airway disease and not from loss of lung elastic recoil. These data are consistent with the report by Burger and coworkers (15) who obtained maximal flow–elastic recoil curves in eight LAM patients and found that airflow limitation was due to airway obstruction and not to loss of lung elastic recoil.
Our data also show that the severity of lung damage as estimated by the LHS is functionally most closely paralleled by Dl CO. It appears, in addition, that FEV1 in LAM may be influenced by factors such as proliferation of LAM cells and bronchiolar inflammation, which may not affect Dl CO but are associated with a decline in expiratory flow. Because LHS is a predictor of survival or time to lung transplantation (5), it is very likely that, physiologically, the best test in assessing lung disease severity and time to lung transplantation in LAM is Dl CO. Our findings are consistent with the data of Kitaichi and coworkers (16), showing that predominance of cystic lesions, which are more likely to impair the Dl CO, is associated with poor prognosis whereas smooth muscle proliferation does not correlate with survival.
In our study we observed rates of decline in FEV1 and Dl CO that are lower than those reported recently by Johnson and Tattersfield (4). Further, we noted that for the patients with the longer follow-up, i.e., 5.7 ± 0.4 yr, the rate of decline in function was even lower. These data suggest that in some patients lung function either remains stable or declines very little. Currently no method is available to identify the patients who are going to have a more stable course.
We conclude that reversible airway obstruction in LAM appears to be associated with an accelerated loss of lung function and greater proliferation of abnormal smooth muscle cells, and that impairment of Dl CO correlates with the histologic severity of the disease and probably is a predictor of survival or time to transplantation.
Supported by the Division of Intramural Research, National Heart, Lung, and Blood Institute, National Institutes of Health.
|1.||Chu SC, Horiba K, Usuki J, Avila NA, Chen CC, Travis WD, Ferrans VJ, Moss JComprehensive evaluation of 35 patients with lymphangioleiomyomatosis. Chest115199910411052|
|2.||Sullivan EJLymphangioleiomyomatosis: a review. Chest114199816891703|
|3.||Urban TJ, Lazor R, Lacronique J, Murris M, Labrune S, Valeyre D, Cordier JFPulmonary lymphangioleiomyomatosis: a study of 69 patients. Medicine781999321337|
|4.||Johnson S, Tattersfield AEDecline in lung function in lymphangioleiomyomatosis: relation to menopause and progesterone treatment. Am J Respir Crit Care Med1601999628633|
|5.||Matsui K, Takeda K, Yu Z, Travis WD, Barnes JP, Bechtle J, Falk R, Ferrans VJ, Moss J, Travis WDPrognostic significance of pulmonary lymphangioleiomyomatosis histologic score. Am J Surg Path252001479484|
|6.||American Thoracic Society lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis 1991;144:1202–1218.|
|7.||American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med 1994;152:1107–1136.|
|8.||American Thoracic Society. Single-breath carbon monoxide diffusing capacity (transfer factor): recommendations for a standard technique-1995 update. Am J Respir Crit Care Med 1995;152:2185–2198.|
|9.||Morris JF, Koski A, Johnson LCSpirometric standards for healthy nonsmoking adults. Am Rev Respir Dis10319715767|
|10.||Knudson RJ, Slatin RC, Lebowitz MD, Burrows BThe maximal expiratory flow–volume curve. Am Rev Respir Dis1131976587600|
|11.||Crapo RO, Morris AHStandardized single breath normal values for carbon monoxide diffusing capacity. Am Rev Respir Dis1231991185189|
|12.||Liang KY, Zeger SLLongitudinal data analysis using generalized linear models. Biometrika7319861322|
|13.||Zeger SL, Liang KWLongitudinal data analysis for discrete and continuous outcomes. Biometrics421986121130|
|14.||Postma DS, Lebowitz MDPersistence and new onset of asthma and chronic bronchitis evaluated longitudinally in a community population sample of adults. Arch Intern Med155199513931399|
|15.||Burger CD, Hyatt RE, Staats BAPulmonary mechanics in lymphangioleiomyomatosis. Am Rev Respir Dis143199110301033|
|16.||Kitaichi M, Nishimura K, Harumi I, Izumi TPulmonary lymphangiomyomatosis: a report of 46 patients including a clinicopathologic study of prognostic factors. Am J Respir Crit Care Med1511995527533|
Dr. Matsui is currently at the Department of Pathology, Faculty of Medicine, Toyama Medical Pharmaceutical University, Toyama, Japan.