|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Pulmonary and Critical Care Medicine Service, Department of Medicine, Walter Reed Army Medical Center, Washington, DC.
Correspondence to: Andrew Shorr, MD, MPH, Pulmonary and Critical Care Medicine, Walter Reed Army Medical Center, 6900 Georgia Ave, NW, Washington, DC 20307; e-mail: AFSHORR{at}DNAMAIL.COM
| Abstract |
|---|
|
|
|---|
Design: Prospective study of consecutive patients.
Setting: Pulmonary clinic of a tertiary-care, academic medical center.
Patients: Patients consecutively referred for suspected pulmonary sarcoidosis.
Interventions: All patients having FOB performed underwent an evaluation that included history, physical examination, a chest radiograph, and spirometry. During FOB, airway appearance was recorded and both TBB and EBB were performed in a standardized fashion. Six TBB specimens were obtained, as were six EBB samples. For patients with abnormal-appearing airways, four specimens were obtained from the abnormal-appearing airways and two specimens were obtained from the main carina. In patients with normal-appearing airways, four specimens were obtained from a secondary carina and two specimens were obtained from the main carina. A biopsy finding was considered positive if it demonstrated nonnecrotizing granulomas with special stains that were negative for fungal and mycobacterial organisms.
Measurements and results: The study cohort included 34 subjects (mean ± SD age, 37.9 ± 6.8 years; 58.8% were male; 64.7% were African American). EBB findings were positive in 61.8% of patients, while TBB showed nonnecrotizing granulomas in 58.8% of subjects. The addition of EBB increased the yield of FOB by 20.6%. Although EBB findings were more frequently positive in abnormal-appearing airways (p = 0.014), EBB provided diagnostic tissue in 30.0% of patients with normal-appearing endobronchial mucosa. There were no complications resulting from the addition of EBB to TBB.
Conclusions: Endobronchial involvement is common in sarcoidosis. EBB has a yield comparable to TBB and can safely increase the diagnostic value of FOB. Pulmonologists should consider routinely performing EBB in cases of suspected sarcoidosis.
Key Words: airway biopsy bronchoscopy endobronchial sarcoidosis transbronchial
| Introduction |
|---|
|
|
|---|
Although the diagnosis of sarcoidosis may be made on the basis of the clinicoradiologic pattern alone, there are many other entities in the differential diagnosis of sarcoidosis. Similarly, serologic studies initially thought to be helpful in sarcoidosis, such as the serum angiotensin-converting enzyme (ACE) level, have been noted to be elevated in multiple, unrelated conditions.4 Nuclear imaging with 67Ga may occasionally reveal a pattern classic for sarcoidosis, but these findings are present in few individuals and therefore the yield of routine nuclear imaging is low.5 Thus, in a 1999 position statement,6 the American Thoracic Society in accordance with the European Respiratory Society concluded that the diagnosis of sarcoidosis requires "histologic demonstration of [nonnecrotizing] granulomas." Since the lungs are most commonly involved in sarcoidosis, transbronchial biopsy (TBB) via fiberoptic bronchoscopy (FOB) is routinely performed in cases of suspected sarcoidosis. The yield of TBB is high but ranges from 50 to 80% depending on the extent of parenchymal involvement.7 8 9 TBB may occasionally result in either pneumothorax or pulmonary hemorrhage. If the TBB finding is negative, either surgical lung biopsy or mediastinoscopy may be performed. Both are relatively more invasive, associated with greater mortality, and are more expensive than TBB. One alternative for increasing the yield of FOB is endobronchial biopsy (EBB). We have shown retrospectively that EBB can safely increase the yield of TBB even if the airway mucosa appears normal.10 The American Thoracic Society/European Respiratory Society position statement acknowledges that EBB may be a useful adjunct in the approach to sarcoidosis. However, little prospective data exist to support the routine use of EBB.6 Therefore, we undertook a prospective trial in consecutive patients undergoing FOB for suspected sarcoidosis to confirm the hypothesis that EBB can significantly increase the yield of FOB and to determine factors that might predict the results of EBB.
| Materials and Methods |
|---|
|
|
|---|
Evaluation
All subjects underwent a standard evaluation that included a
history and physical examination, a CXR, pulmonary function tests
(PFTs) with measurement of the single-breath diffusion capacity of the
lung for carbon monoxide (DLCO), and an ECG. Radiographic
stages were defined as follows: stage 0, normal CXR result; stage I,
bilateral hilar lymphadenopathy alone; stage II, bilateral hilar
lymphadenopathy with interstitial infiltrates; and stage III,
interstitial infiltrates alone. The PFTs and DLCO were
interpreted in accordance with the guidelines of the American Thoracic
Society.11
Normal values were derived from Crapo et
al,12
and corrections for race were made. The
DLCO was further corrected for hemoglobin. Values for PFTs
and DLCO were considered abnormal if they fell outside the
95% confidence interval (CI) for the predicted values. Serum studies
consisted of chemistries with assessment of both renal function and
serum calcium, a CBC count, and analysis for several markers of
inflammation. Specifically, the erythrocyte sedimentation rate (ESR),
C-reactive protein (CRP), and ACE levels were measured. The presence of
pulmonary symptoms (dyspnea, cough, and wheezing) was determined from a
standard questionnaire completed by all subjects on enrollment.
Bronchoscopy
FOB was performed in standard fashion in all patients.
Initially, airway appearance was recorded as either being normal or
abnormal. Abnormal airways were defined as those that were either
erythematous or demonstrated mucosal thickening, or both. The
determination of the airway appearance was made as a consensus decision
between the bronchoscopist and the bronchoscopy technician. TBB was
then performed. The site of the TBB was left to the clinician, and for
each patient six specimens were obtained. The number of biopsies
performed during EBB was also standardized. For patients with
abnormal-appearing airways, four specimens were taken from the
abnormal-appearing airways and two were obtained from the main carina.
In patients with normal-appearing airways, four specimens were taken
from a secondary carina and two were taken from the main carina. For
the TBB, an 18-mm cup biopsy forceps (Bard Interventional Products;
Billerica, MA) was employed; for EBB, an 18-mm smooth-edged jaws
forceps (Boston Scientific; Watertown, MA) was used. Following FOB, all
patients underwent a CXR. The incidence of complications, specifically
pneumothorax and pulmonary hemorrhage, was recorded.
TBB and EBB specimens were placed in separate collection cups prior to specimen processing. Pathologists reviewing the specimens were blinded not only to the patients clinical presentation (eg, symptoms, CXR findings) but also to the appearance of the airways. A biopsy technique (TBB or EBB) finding was considered positive if it revealed nonnecrotizing granulomas with specials stains that failed to show either mycobacterial or fungal organisms. In patients for whom neither biopsy procedure produced a histologic diagnosis, surgical biopsy was offered but not mandatory. In order to determine the yield of each biopsy technique, only patients with final histologic evidence of sarcoidosis were included in the analysis.
Statistical Analysis
Patients with positive EBB findings were compared to those with
negative EBB findings for the variables noted above. The Students
t test was used for continuous variables. The
2 test was employed to compare categorical
variables except in cases when expected values were small. In these
instances, we relied on the Fishers Exact Test. The distribution of
CXR stages was analyzed via the Wilcoxin rank sum test. All tests were
two tailed, and a p value < 0.05 was assumed to represent statistical
significance. Ninety-five percent CIs are reported where appropriate.
Analyses were done using software (SPSS 7.0; SPSS; Chicago, IL).
For sample size determination, based on our earlier study we assumed
that the yield of TBB would be approximately 60% in the study
population. Using this and controlling the
error at 0.05, we
calculated that the study would require 34 patients in order to have an
80% power to detect if EBB could increase the yield of FOB by at least
20%.
| Results |
|---|
|
|
|---|
Table 1 shows the demographic composition of the entire cohort. For the 34 subjects, the mean ± SD age was 37.9 ± 6.8 years and 58.8% were males. Most patients (64.7%) were African American. A stage I CXR was noted in 23 patients, and stage II and stage III CXRs were seen in 9 patients and 2 patients, respectively. Four patients (11.8%) had extrapulmonary sarcoidosis at the time of presentation (two uveitis, one skin, one extrapulmonary lymph nodes).
|
|
|
|
| Discussion |
|---|
|
|
|---|
Several prior retrospective reports have reported on the yield of EBB in sarcoidosis. In these studies,10 13 14 15 16 the yield of EBB has ranged from 45 to 71%. An early report by Bybee et al13 described the utility of EBB via rigid bronchoscopy; they noted that the EBB result was positive in 24% of patients and that endobronchial abnormalities may resolve with corticosteroid therapy. In a review of 85 patients, Armstrong and colleagues14 noted endobronchial abnormalities in approximately half of their patients and EBB resulted in diagnostic tissue in 57% of cases. As in our study, the results of EBB correlated with airway appearance, but EBB was still positive in approximately 35% of subjects with normal airway mucosa. These authors did not perform both TBB and EBB in all subjects and did not provide data describing the degree to which EBB added to the value of TBB.14 Armstrong et al14 also noted that EBB findings were more frequently positive in both stage I and II disease as compared to patients with stage III CXRs. This finding is in contrast to our observation that CXR stage did not impact the yield of EBB and likely reflects differences in study design. Bjemer et al15 described EBB in 62 Scandinavian patients and found nonnecrotizing granulomas in 28 instances (45%). Their bronchoscopy protocol differed from ours in that they obtained four EBB specimens from four different sites (one specimen per site) irrespective of airway appearance.15 Despite this, they also observed that the EBB result was more likely to be positive if the endobronchial mucosa were abnormal. When the endobronchial mucosa seemed normal, diagnostic tissue was obtained in 36% of patients. They did not report data regarding initial symptoms, PFTs, or serum makers of inflammation.14 Bilaceroglu and colleagues16 examined the value of EBB as an adjunct to TBB with or without transbronchial needle aspiration (TBNA) in 74 patients suspected of having sarcoidosis. Their retrospective series confirmed the findings of earlier reports with EBB proving diagnostic in 45% of patients.16 Finally, we previously reviewed our experience with EBB and observed a yield of EBB that was higher than has been described.10 Of 59 patients, EBB was diagnostic for 71% and the EBB finding was positive in 2 of 4 patients in whom the endobronchial mucosa was without abnormality.10 The relatively high value of EBB in this study likely reflects that EBB was essentially confined to patients who had evident endobronchial abnormalities.
The differences we noted in spirometry based on EBB results are of unclear clinical significance. Although both the FVC and FEV1 were approximately 10% lower in subjects with positive EBB results, the majority of patients, irrespective of EBB status, had normal or nearly normal PFT findings. These PFT findings are also unlikely to represent differences as to the extent of parenchymal involvement. Parenchymal disease as assessed by CXR stage did not vary based on the yield of EBB. Moreover, neither the DLCO nor the DLCO/alveolar volume differed based on the results of EBB. Endobronchial disease may not play a role in the development of airflow obstruction in sarcoidosis. For example, EBB status was not associated with airflow obstruction. Our study, though, was not powered to investigate this particular issue. Further prospective studies are needed to investigate the relationship between the presence of endobronchial nonnecrotizing granulomas, airflow obstruction, and airway hyperactivity in sarcoidosis.
Our observation that race did not impact on the results of EBB stands in contrast to our earlier study.10 In that analysis, we found that African Americans were 2.2 times more likely to have a positive EBB result (p < 0.001).10 Since the present study is prospective, focused on consecutive patients, and employed a standardized protocol for the use of EBB, we believe that race is unlikely to correlate with the results of EBB. This conclusion is supported by the other retrospective studies13 14 15 16 cited above that predominantly described white patients and found a clinically meaningful yield for EBB. The trend toward a correlation between EBB and extrapulmonary sarcoidosis implies that endobronchial biopsy may provide insight as to the expected course of sarcoidosis. Extrapulmonary sarcoidosis is known to have a poor prognosis and identifying patients at the time of diagnosis who might have a complicated course could aid in their evaluation and management. Longitudinal follow-up will be needed to confirm this hypothesis.
In comparison to other procedures, EBB results in diagnostic tissue as often as do both TBB and TBNA. Although the yield of TBB may reach 90%, studies describing heterogeneous populations in which the extent of parenchymal involvement varies report a yield for TBB of approximately 70%.7 8 9 For example, Morales et al17 reported that TBB findings were positive in 34 of 51 consecutive patients suspected of having sarcoidosis. In the subgroup with stage I disease, the diagnostic tissue was obtained in only 60% of the subjects. Similarly Puar et al18 determined the overall yield of TBB to be 70% but varied depending on the CXR stage. In the original study8 describing the role for TBB in sarcoidosis, TBB resulted in a histologic diagnosis in > 90% of the patients. However, 62% of the individuals in this cohort had parenchymal disease. For TBNA, Wang and colleagues,19 in a retrospective series, found that this procedure demonstrated nonnecrotizing granulomas in 90% of cases. In this study, though, TBB was performed in fewer than half of the individuals. Other have noted that the yield of TBNA for sarcoidosis is lower and ranges from 50 to 60%.16 17 20 As an adjunct to TBB, TBNA and EBB increase the value of FOB to similar degrees. One study17 found that with the addition of TBNA, the yield of FOB rose by 18%. Correspondingly, Bilaceroglu and colleagues16 concluded that TBNA increased the diagnostic value of TBB by 14%. EBB, however, is technically less difficult than TBNA.
Our study has several limitations. The first limitation is that the majority of patients in this study were male. Epidemiologically, sarcoidosis is seen more frequently in women.21 Therefore, the results of this study may not be generalizable. With respect to other demographic variables, though, our study cohort was similar to the typical sarcoidosis population in the United States in that the study predominantly consisted of young, African-American patients with stage I CXRs.21 Second, we did not require patients with nondiagnostic bronchoscopies to undergo further evaluation. Although this could have biased our findings, this is unlikely because only two patients declined additional diagnostic testing. Also, the prospective nature of the study coupled with the reliance on a standardized bronchoscopy protocol should have minimized the impact of this. Unlike many earlier studies on sarcoidosis, we also enrolled consecutive subjects suspected of suffering from sarcoidosis and accounted for other individuals diagnosed with pulmonary sarcoidosis during the study period. Finally, there was no prospective definition of airway appearance. Thus, implicit definitions as to appearance of the airways may differ between bronchoscopists.
In conclusion, endobronchial involvement is common in sarcoidosis. We recommend that pulmonologists consider routinely adding EBB to TBB in cases of suspected sarcoidosis. Inclusion of this technique will significantly increase the yield of FOB and avoid the need for either repeated or more invasive procedures. A normal appearance to the endobronchial mucosa should not deter the bronchoscopist from performing EBB.
| Acknowledgements |
|---|
| Footnotes |
|---|
Received for publication August 25, 2000. Accepted for publication January 30, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
S. Garwood, M. A. Judson, G. Silvestri, R. Hoda, M. Fraig, and P. Doelken Endobronchial Ultrasound for the Diagnosis of Pulmonary Sarcoidosis Chest, October 1, 2007; 132(4): 1298 - 1304. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. R. Pesola, M. Kurdi, M. Olibrice, A. F. Shorr, K. G. Torrington, and O. W. Hnatiuk Endobronchial Sarcoidosis and Hyperreactive Airways Disease Chest, June 1, 2002; 121(6): 2081 - 2081. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |