| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, Warren G. Magnuson Clinical Center,
National Institutes of Health, Bldg. 10, Rm. 1C-660, 10 Center Dr., MSC 1182,
Bethesda, MD 20892-1182.
2 Division of Pulmonary Critical Care Medicine, University of Florida College of
Medicine, J. Hillis Miller Health Center, Rm. M-452 MSB, 1600 S.W. Archer Rd.,
Gainesville, FL 32610.
3 Section on Human Biochemical Genetics, Heritable Disorders Branch, National
Institute of Child and Human Development, National Institutes of Health, Bldg.
10, Rm. 9S-241, Bethesda, MD 20892.
Received October 17, 2001;
accepted after revision March 18, 2002.
Presented at the annual meeting of the American Roentgen Ray Society, New
Orleans, May 1999.
Abstract
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
SUBJECTS AND METHODS. Sixty-seven patients with Hermansky-Pudlak syndrome underwent high-resolution CT of the chest. A scoring system based on the extent of pulmonary involvement and specific high-resolution CT findings was used, and the findings were compared with patient age and the results of pulmonary function and genetic studies. Fifty-eight (87%) of the 67 patients also underwent chest radiography. These radiographs were compared with the high-resolution CT scans.
RESULTS. High-resolution CT was more sensitive than chest radiography in evaluating the extent of pulmonary disease in patients with Hermansky-Pudlak syndrome. All patients with mild findings on high-resolution CT scans had normal findings on chest radiographs. Common chest radiographic findings included reticulonodular interstitial pattern, perihilar fibrosis, and pleural thickening. High-resolution CT showed septal thickening, ground-glass opacities, and peribronchovascular thickening. For patients with Hermansky-Pudlak syndrome who were 30 years old or younger, high-resolution CT findings were usually minimal. Among patients who were older than 30 years, the 34 patients with HPS1 mutations had a score of 1.38±0.18 (mean±standard error of the mean) on high-resolution CT. This score is significantly greater than the score for the 11 patients without HPS1 mutations (0.36 ± 0.15) (p < 0.001). The score based on high-resolution CT findings inversely correlated with percentage of forced vital capacity and was useful in defining the progression of interstitial disease.
CONCLUSION. High-resolution CT provides a good radiologic monitor of disease status and progression in patients with Hermansky-Pudlak syndrome and correlates well with patient age, extent of pulmonary dysfunction, and genetic findings.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
|
Hermansky-Pudlak syndrome can be caused by a number of genes, including HPS1, ADTß3A, and HPS3 [8,9,10]. In northwest Puerto Rico, approximately 400 individuals have Hermansky-Pudlak syndrome, and the prevalence is one in 1800 [11]. Patients living in northwest Puerto Rico exhibit homozygosity for a 16-base pair duplication in exon 15 of the HPS1 gene [8]. Only three patients with ADTß3A mutations have been described [12, 13], and HPS3 mutations are concentrated in patients living in central Puerto Rico [10]. Hermansky-Pudlak syndrome is a model for disorders of vesicle formation [14,15,16]. Fewer than 100 patients with Hermansky-Pudlak syndrome who are not Puerto Rican have been identified in the United States, but this number probably represents an underestimation because the index of suspicion for the disorder is low [9].
Phenotypic variability has also become apparent. Patients with Hermansky-Pudlak syndrome display different degrees of hypopigmentation, bleeding diathesis, colitis, visual acuity and ophthalmic involvement, and dermatologic complications [17,18,19]. Pulmonary fibrosis associated with Hermansky-Pudlak syndrome occurs with increased frequency in patients with mutations in the HPS1 gene [17, 20].
We described the imaging and pulmonary function results of 38 patients with Hermansky-Pudlak syndrome and correlated pulmonary function with extent of disease on chest CT [20]. We now present specific high-resolution CT and radiographic findings and the incidence of these findings in a larger group of 67 patients. Follow-up high-resolution CT examinations were compared with baseline evaluations in a subset of patients. High-resolution CT scans were analyzed in relation to patient age and to the genetic locus causing Hermansky-Pudlak syndrome. Finally, the severity of pulmonary disease evident on high-resolution CT was compared with pulmonary function as represented by percentage of forced vital capacity.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Two patients were excluded from the analysis because they had concurrent bronchoalveolar cell carcinoma and pneumonia, respectively. The remaining 67 patients ranged in age from 9 to 57 years (mean, 35 years); there were 30 males and 37 females. Each patient was enrolled in a clinical research protocol approved by the institutional review board. Hermansky-Pudlak syndrome was suspected on the basis of the presence of oculocutaneous albinism and a bleeding diathesis and was confirmed by the absence of platelet dense bodies on electron microscopy [3].
In 46 of the 67 patients, a mutation in HPS1 was identified by either a polymerase chain reaction—based assay of the 16—base pair duplication in exon 15 (n = 42) or by sequencing each HPS1 exon (n = 4), as previously described [8, 21,22,23,24]. All 67 patients underwent high-resolution CT of the chest, genetic tests, and pulmonary function tests. Fifty-eight of the 67 patients had chest radiographs; however, the chest radiographs were not retrievable for nine of these patients. Seventeen of the 67 patients underwent follow-up high-resolution CT of the chest.
Chest Radiography
Posteroanterior and lateral chest radiographs were obtained using a
standard technique. The chest radiographs were analyzed by two radiologists in
consensus. Specific abnormalities recorded included reticulonodular pattern of
interstitial lung disease, perihilar fibrosis, pleural thickening,
interstitial infiltrates, areas of consolidation, and pulmonary nodules.
CT
All 67 patients underwent conventional CT and high-resolution CT of the
chest. CT was performed with HiSpeed Advantage and Ct/i scanners (General
Electric Medical Systems, Milwaukee, WI). Conventional CT was performed during
end-inspiration with the patients supine using contiguous 10-mm sections.
High-resolution CT was performed during end-inspiration with the patients
prone using a 1-mm collimation at 30-mm intervals and a high-spatial-frequency
reconstruction algorithm. No IV contrast material was administered.
The following high-resolution CT features were used to evaluate the extent of pulmonary disease of patients with Hermansky-Pudlak syndrome. Reticular opacities were categorized as septal lines or reticulation. Septal lines were defined as linear densities representing thickened interlobular septa that were usually most evident in the lung periphery as lines running perpendicular to the pleura and typically 1-2 cm long and 0.5-2 cm apart. Reticulation was a mesh formed by linear opacities that were only a few millimeters long and separated from one another by only a few millimeters. Findings associated with interstitial fibrosis included peribronchial cuffing (peribronchial interstitial thickening); traction bronchiectasis (bronchial dilatation resulting from peribronchial fibrosis); thickening of the fissures and pleura; cysts, which form because alveoli are destroyed as a result of lung fibrosis; and ground-glass attenuation (bilateral patchy areas of hazy increased attenuation that did not obscure the underlying vascular markings associated with airways obstruction).
The thorax was divided into upper, middle, and lower zones by dividing the total number of images into three equal subsets. The scans were graded independently by two board-certified radiologists and then together to arrive at a consensus interpretation. Interobserver variability of the initial independent interpretations was determined using the kappa statistic [25]. The CT findings were graded according to the severity and extent of pulmonary disease as follows: grade 0, normal findings; grade 1, minimal disease (thickened interlobular septa, reticular disease, subpleural cysts, and areas of ground-glass pattern); grade 2, moderate disease (traction bronchiectasis, peribronchovascular thickening, tracheal retraction involving one or two zones of the lungs, and the findings present in grade 1); and grade 3, severe disease (the findings in grades 1 and 2 involving all three zones).
We recorded the distribution of abnormalities (upper, middle, or lower lung zones; peripheral or central). Follow-up high-resolution CT was performed in 17 patients, and these scans were compared with the baseline studies. Lung disease, if present, was categorized as unchanged, improved, or worsened.
Pulmonary Function Test
Pulmonary function was assessed using established normal values
[20]. The grades assigned from
high-resolution CT were correlated with the pulmonary function test results.
The Student's t test was used for comparisons. Analysis of variance
was used to assess the association between the grades on high-resolution CT
and the percentage of forced vital capacity.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
CT
All 67 patients in the study underwent conventional and high-resolution CT
of the chest. The distribution of disease severity was as follows: grade 0
(normal), 31 patients (46%); grade 1 (minimal), 22 (33%); grade 2 (moderate),
7 (10%); and grade 3 (severe), 7 (10%). Agreement between the two observers
was excellent: the kappa value was 0.72 (p < 0.001). In 54 (81%)
of the 67 cases, agreement between the observers was perfect, whereas in 64
(96%) of the 67, the observers agreed within one category.
The CT findings were cross-tabulated with the grade of disease on high-resolution CT (Table 1 and Figs. 3 and 4). Most of the abnormalities were distributed diffusely with a slight predominance in the middle (26 patients) and lower (30 patients) lung zones compared with the upper zone (24 patients). Most of the patients with interstitial lung disease (35/36) had a peripheral pattern of distribution. The remaining patient had an atypical pattern of interstitial lung disease with sparing of the periphery of the lungs. As the severity of disease progressed, the central portions of the lungs became increasingly involved. The central portions were involved in none of the patients with grade 1 disease, whereas the central portions were involved in four of the seven patients with grade 2 disease and in seven of the seven patients with grade 3 disease. The central portions become involved as disease progresses as a result of the peribronchovascular thickening and bronchiectasis caused by interstitial fibrosis.
|
|
|
Comparison of Radiographic Findings and High-Resolution CT
Findings
Of the 58 patients who underwent chest radiography, all those with normal
findings on high-resolution CT (27 patients) or with grade 1 findings on
high-resolution CT (19 patients) had normal findings on chest radiography. In
patients with moderate or severe disease on high-resolution CT, the following
changes were detected on radiography: reticulonodular pattern, perihilar
fibrosis, pleural thickening, and patchy interstitial infiltrates. Some
patients had a combination of radiographic findings.
Review of Follow-Up High-Resolution CT Studies
Seventeen patients had follow-up high-resolution CT studies 12-51 months
after the initial examination at our institution. Thirteen patients had no
change during the period of follow-up (12-51 months). In four patients,
interstitial lung disease had worsened during the time of observation (13-49
months) (Fig.
5A,5B,5C).
|
|
|
Variation of CT Score with Age and Genetic Locus
The 67 patients who underwent high-resolution CT of the chest were grouped
according to age and whether they had mutations in HPS1. Forty-six
had mutations in HPS1, and 21 did not.
Patients with mutations in the HPS1 gene had more severe lung disease than patients without HPS1 gene mutations. No patient younger than 20 years old in either group had any evidence of pulmonary fibrosis on high-resolution CT (score = 0). Patients who ranged in age from 20 to 29 years in the group with HPS1 mutations had a high-resolution CT score of 0.25 ± 0.16 (mean ± standard error of the mean [SEM]), which is not significantly different from that for the patients without HPS1 mutations (mean ± SEM, 0.50 ± 0.22) (p > 0.3). Patients with HPS1 mutations who ranged in age from 30 to 39 years had a high-resolution CT score of 1.47 ± 0.27 (mean ± SEM) compared with 0.25 ± 0.25 (mean ± SEM) for the patients in the same age group without these mutations (p < 0.005). For the patients who were 40 years old or older, those with HPS1 mutations had a higher high-resolution CT score (mean ± SEM, 1.33 ± 0.23) than those without HPS1 mutations (mean ± SEM, 0.43 ± 0.20) (p < 0.001). The 34 patients older than 30 years who had the HPS1 mutations had a mean high-resolution CT score of 1.38 ± 0.18, which is significantly greater than that for the 11 patients older than 30 years who did not have HPS1 mutations (mean ± SEM, 0.36 ± 0.15) (p < 0.001).
Comparison of CT Findings and Pulmonary Function Test Results
For the high-resolution CT scans grouped by severity of disease, average
pulmonary function of each patient was estimated using forced vital capacity
as the representative parameter. As the high-resolution CT severity score
increased, the forced vital capacity, expressed as percentage of predicted,
fell dramatically. This decrease was found to be statistically significant at
a p value of less than 0.001 using the analysis of variance.
Prediction of Mortality
The score based on high-resolution CT findings also enabled a general
prediction of mortality. Four of the seven patients with a score of grade 3
disease on high-resolution CT died from complications of pulmonary fibrosis
within 4 months of the examination, and a fifth died within 8 months. Two of
the seven patients with a score of grade 2 disease died as a result of
pulmonary fibrosis 12 and 32 months after the examination. One of the 22
patients with a score of grade 1 disease died, also of pulmonary fibrosis, 28
months after the examination.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
High-resolution CT is the imaging method of choice in evaluating the extent of pulmonary disease in patients with Hermansky-Pudlak syndrome because chest radiographs can underestimate the extent of disease. Findings on high-resolution CT during the earlier stages of disease include septal thickening, ground-glass pattern, and mild reticulation, whereas in the more advanced stages of disease findings on high-resolution CT include moderate to severe reticulation, bronchiectasis, subpleural cysts, and peribronchovascular thickening. The abnormalities on high-resolution CT are fairly evenly distributed throughout the lungs among the upper, middle, and lower zones, with a slight predilection for the middle and lower zones. The abnormal findings associated with Hermansky-Pudlak syndrome predominate in the periphery of the lungs. However, as the disease progresses, the central portions of the lungs become increasingly involved because peribronchovascular thickening and bronchiectasis develop from interstitial fibrosis.
Using a scoring system that accounts for extent of disease and specific CT findings, one can follow the natural history of pulmonary fibrosis in patients with Hermansky-Pudlak syndrome. Our results reflect the influence of the HPS1 mutations on the rapidity of progression of the disease. In fact, high-resolution CT scores indicate minimal involvement of pulmonary fibrosis in patients with Hermansky-Pudlak syndrome who are 30 years old or younger. In patients older than 30 years, the abnormalities on high-resolution CT are significant, but in this age group, most abnormal findings were seen in 34 patients with mutations in HPS1. The 11 patients older than 30 years and lacking HPS1 mutations were relatively, although not absolutely, spared. This observation supports previous findings that the greatest risk of pulmonary fibrosis in patients with Hermansky-Pudlak is associated with abnormalities in the HPS1 gene [17, 20]. We found that the HPS1 mutations need not be the 16—base pair duplication in exon 15 that is present in all members of the founder population in northwest Puerto Rico [20].
We also found a direct correlation between high-resolution CT findings indicative of worsening disease and decreasing pulmonary function, as gauged by the percentage of predicted forced vital capacity. A correlation between high-resolution CT and pulmonary function has been previously described for a smaller group of patients with Hermansky-Pudlak syndrome and for patients with fibrosing alveolitis [20, 28]. A potential clinical use of the high correlation between high-resolution CT and pulmonary dysfunction is that patients may be advised about the degree of severity of pulmonary involvement. Individuals with disease characterized as severe on the basis of high-resolution CT findings are at a high risk of dying in a few months.
As therapies for pulmonary fibrosis emerge, high-resolution CT may serve as an outcome parameter for evaluating treatments. Moreover, Hermansky-Pudlak syndrome provides a valuable model for studies of other forms of pulmonary fibrosis because most patients with pulmonary fibrosis are genetically uniform for the 16—base pair duplication in HPS1. The idiopathic forms of pulmonary fibrosis are more difficult to investigate because they are genetically and pathogenetically heterogeneous.
A limitation of this study is the low number of patients with follow-up examinations, thus making it difficult to prove statistically our observation of lowered life expectancy in patients with high-resolution CT findings characteristic of severe disease.
High-resolution CT provides a good measure of the extent of disease at presentation and its progression through the assessment of pulmonary fibrosis in patients with Hermansky-Pudlak syndrome. High-resolution CT is more sensitive than chest radiography and correlates with pulmonary dysfunction. A description of the specific high-resolution CT findings in patients with Hermansky-Pudlak syndrome and correlations with patient age, pulmonary function, and genetic mutation establish the role of high-resolution CT in the evaluation of patients with pulmonary fibrosis in general and in Hermansky-Pudlak syndrome in particular.
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  I. F. Pedraza, K. Scher, E. Fajardo, and N. Kamangar A CASE OF A 29-YEAR-OLD MALE WITH HISTORY OF OCULOCUTANEOUS ALBINISM, EPISTAXIS AND PROGRESSIVE DYSPNEA Chest Meeting Abstracts, October 1, 2008; 134(4): c65001 - c65001. [Abstract]
|
|
|
|
 |
|
 B. R. Gochuico, N. A. Avila, C. K. Chow, L. J. Novero, H.-P. Wu, P. Ren, S. D. MacDonald, W. D. Travis, M. P. Stylianou, and I. O. Rosas Progressive Preclinical Interstitial Lung Disease in Rheumatoid Arthritis Arch Intern Med, January 28, 2008; 168(2): 159 - 166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. M. Berkmen and B. M. Dsouza Case 124: Hermansky-Pudlak Syndrome Radiology, November 1, 2007; 245(2): 595 - 599. [Full Text] [PDF]
|
|
|
|
 |
|
 I. O. Rosas, P. Ren, N. A. Avila, C. K. Chow, T. J. Franks, W. D. Travis, J. P. McCoy Jr., R. M. May, H.-P. Wu, D. M. Nguyen, et al. Early Interstitial Lung Disease in Familial Pulmonary Fibrosis Am. J. Respir. Crit. Care Med., October 1, 2007; 176(7): 698 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. R. Young, R. Pasula, P. M. Gulleman, G. H. Deutsch, and F. X. McCormack Susceptibility of Hermansky-Pudlak Mice to Bleomycin-Induced Type II Cell Apoptosis and Fibrosis Am. J. Respir. Cell Mol. Biol., July 1, 2007; 37(1): 67 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Anthi, R. F. Machado, M. L. Jison, A. M. Taveira-DaSilva, L. J. Rubin, L. Hunter, C. J. Hunter, W. Coles, J. Nichols, N. A. Avila, et al. Hemodynamic and Functional Assessment of Patients with Sickle Cell Disease and Pulmonary Hypertension Am. J. Respir. Crit. Care Med., June 15, 2007; 175(12): 1272 - 1279. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. R. Young, M. T. Borchers, H. L. Allen, R. S. Gibbons, and F. X. McCormack Lung-Restricted Macrophage Activation in the Pearl Mouse Model of Hermansky-Pudlak Syndrome J. Immunol., April 1, 2006; 176(7): 4361 - 4368. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
