| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Pulmonary Institute (Drs. Shitrit, Yarmolovsky, Bendayan, Miller, Fink, Mazar, and Kramer), Rabin Medical Center, Beilinson Campus, Petah Tiqwa and Sackler School of Medicine, Tel Aviv University, Tel Aviv; and Pulmonary Institute (Dr. Izbicki), Shaare Zedek Medical Center and Ben-Gurion University of the Negev, Jerusalem, Israel.
Correspondence to: Gabriel Izbicki, MD, Pulmonary Institute, Shaare Zedek Medical Center, Jerusalem 91031, Israel; e-mail: izbicki{at}szmc.org.il
Abstract
Background and study objective: Pneumothorax following flexible bronchoscopy (FB) with transbronchial biopsy (TBB) occurs in 1 to 6% of cases. Routine chest radiography (CXR) following TBB is therefore requested by most pulmonologists in an attempt to detect complications, particularly pneumothorax. The objective of this study was to determine if routine CXR after bronchoscopy and TBB is necessary.
Patients and method: The study group included 350 consecutive patients who underwent FB with TBB at our institution between December 2001 and January 2004. Routine CXR was performed up to 2 h after the procedure in all cases. Additionally, the following information was recorded in all patients: sex, age, immune status, indication for bronchoscopy, total number of biopsies done, segment sampled, pulse oxygen saturation, and development of symptoms suggestive of pneumothorax.
Results: Pneumothorax was diagnosed radiologically in 10 patients (2.9%). Seven patients had symptoms strongly suggestive of pneumothorax prior to CXR, including four patients with large (> 10%) pneumothorax. The other three patients were asymptomatic, with only minimal pneumothorax (
10%), which resolved completely 24 to 48 h later.
Conclusions: We conclude that routine CXR after bronchoscopy with TBB is necessary only in patients with symptoms suggestive of pneumothorax. In asymptomatic patients, pneumothorax is rare and usually small, so routine CXR is not necessary in this category of patients.
Key Words: bronchoscopy • chest radiograph • pneumothorax • routine • transbronchial biopsies
Flexible bronchoscopy (FB) is an established diagnostic and therapeutic procedure in pulmonology, with low morbidity and very rare mortality. FB with transbronchial biopsy (TBB) is complicated by pneumothorax in 1 to 6% of cases12345678; therefore, most pulmonologists routinely request chest radiography (CXR) after TBB. However, radiographs are time consuming and may account for up to 10% of the total cost of the FB procedure.3 The aim of the present prospective study was to determine if routine CXR is indicated following TBB.
Materials and Methods
All consecutive patients referred for FB with fluoroscopically guided TBB at the Rabin Medical Center, Beilinson Campus, Petah Tiqwa, Israel, between December 2001 and January 2004 were enrolled in the study. FB was performed with midazolam, 1–2 mg IV as needed, and fentanyl, 0.5 mg, sedation. Each patient underwent CXR within 2 h after the procedure, in accordance with our departmental policy. The CXR films were examined by radiologists blinded to the study protocol and the patient’s clinical status. Before the radiologic examination, the following information was collected: sex and age, immune status, indication for the procedure, number of biopsies performed, segment from which most of the biopsy specimens were obtained, pulse oxygen saturation before and after the procedure, and symptoms suggestive of pneumothorax (dyspnea, chest pain, and cough). After the radiographic findings were obtained, the presence (or absence) and extent of pneumothorax (expressed as a percentage of the thoracic volume) were recorded, in addition to the clinical management (hospitalization, chest tube) if necessary. Informed consent was waived in accordance with the institutional ethics committee, as only routine procedures were done.
Statistical Analysis
Pearson correlation coefficient (r) and the significance for it (p values) were calculated between the variables. The significance of differences in the distribution of the categorical variables between patients with or without pneumothorax was analyzed by 
2 test or Fisher exact test, as appropriate. Student t test was used to evaluate differences in continuous variables between two groups.
A series of multivariate stepwise logistic regression models were fitted to the data in order to predict the presence of pneumothorax after bronchoscopy. Odds ratios and 95% confidence intervals were estimated from the model: p 0.05 was considered statistically significant.
Results
Of 1,227 consecutive bronchoscopies performed in our institute between December 2001 and January 2004, 350 procedures (29.5%) included TBB. Indications for FB with TBB were suspected cancer (n = 96, 27.4%); lung or heart-lung transplantation (n = 91, 26%); persistent infiltrates (n = 63, 18%); interstitial lung disease (n = 42, 12%); suspected sarcoidosis (n = 33, 9.4%); suspected tuberculosis (n = 14, 4%); and others (n = 11, 3.2%). One hundred twenty-two patients (34.8%) were considered immunosuppressed at the time of the procedure, most of them (75%) because of treatment after lung or heart-lung transplantation. Small biopsy forceps (FB 15C; diameter, 2 mm; Olympus Optical; Tokyo, Japan) were used in 37% of the bronchoscopies, and a large forceps (FB 20C; diameter 2.8 mm; Olympus Optical) was used in 63%. The number of biopsy specimens obtained were distributed as follows: one to three, n = 28 (8%); four to five, n = 288 (82%); six to seven, n = 32 (9%); more than seven, n = 2 (0.6%). Biopsy specimens were taken from the right upper lobe in 73 patients (21%), right middle lobe in 29 patients (8%), right lower lobe in 120 patients (34%), left upper lobe in 58 patients (17%), and left lower lobe in 70 patients (20%). There was no statistically significant difference in pulse oxygen saturation (mean ± SD) before (97.4 ± 2.1%) and after (96.2 ± 2.0%) the procedure (p = 0.19). Most of the FB procedures (n = 251, 72%) were done by a senior pulmonologists, and the remainder (n = 99, 28%) were done by fellows.
Pneumothorax was diagnosed radiographically in 10 of the 350 patients (2.9% of all TBB procedures). When not considering the patients undergoing transplantation, the pneumothorax rate was 3.9% (10 of 259 procedures). The characteristics of the 10 patients are presented in Table 1 . Large forceps were used in six patients (60%). Bronchoscopy was done by a senior pulmonologist in seven cases and a fellow in three cases. There was a weak but statistically significant correlation between the presence of pneumothorax and the number of biopsy specimens obtained (r = 0.14, p = 0.008).
|
10%) had only mild symptoms, while all four patients with large pneumothoraces (> 10%) had moderate-to-severe symptoms. Symptoms resolved with chest tube drainage in all four patients. Three of the 10 patients were asymptomatic. In none of the asymptomatic patients was pneumothorax suspected by the clinician after FB. All remained symptom free after discharge, and all showed spontaneous resorption of the pneumothorax on the follow-up CXR 24 to 48 h later. When the data were fitted to a multivariate analysis, the number of biopsy specimens obtained was the only predictive factor for development of pneumothorax (p = 0.01; odds ratio, 3.33; confidence interval, 1.29 to 8.56). None of the patients in whom only one to three biopsy specimens were obtained had pneumothorax.
Discussion
The present study is the largest prospective investigation examining the role of routine CXR following FB with TBB. In this study, the large majority of patients (97%) had no radiographic pneumothorax and no clinical symptoms. The four patients found by CXR to have large pneumothoraces requiring treatment were all symptomatic, and physicians suspected pneumothorax before CXR in all of them. Only three patients who were asymptomatic had pneumothorax, but the pneumothoraces were small ( 10%) and resolved spontaneously within 24 to 48 h. These findings show that the CXR performed routinely after TBB is not necessary in asymptomatic patients.
Although we used midazolam and narcotic analgesics during FB, the symptoms were evaluated just before the CXR was performed, up to 2 h after the procedure. The fact that all patients who had pneumothoraces large enough to require treatment were symptomatic shows that the premedication did not mask symptoms.
In a retrospective study, Milam et al3 reviewed 207 FB procedures; 130 CXRs (67%) were obtained after FB, and the authors found only one pneumothorax. However, TBB was done just in 47% and transbronchial needle aspiration in 7% of all the FBs. The only patient with pneumothorax was symptomatic, and the pneumothorax had been detected already by fluoroscopy, prior to CXR. The authors3 concluded that postbronchoscopic CXR rarely provides clinically useful information or detects a complication that is not suspected clinically. This was, however, a retrospective study with an inadequate sample size to allow the risk to be quantified.
Frazier et al,4 in a retrospective study of 305 routine CXRs following TBB, found no pneumothoraces in patients who did not have both chest pain and the appearance of lung collapse on pre-CXR fluoroscopy. They concluded that the combination of these two factors is indicative of post-TBB pneumothorax, and the absence of both effectively rules out pneumothorax. However, this was also a retrospective study, and the authors4 did not investigate the number of patients who had chest pain and abnormal fluoroscopy findings in the absence of pneumothorax.
Recently, a study5 was conducted to examine whether transthoracic sonography could replace CXR for the diagnosis of postinterventional pneumothorax and hydropneumothorax. In this prospective study,5 1 of 35 patients had pneumothorax after TBB (2.9%). Although the study was designed to evaluate ultrasound as a potentially useful tool in detecting pneumothorax, rather than to measure the risk of pneumothorax, the authors5 concluded that the sensitivity, specificity, and accuracy of transthoracic sonography are 100% in excluding postinterventional pneumothorax/hydropneumothorax. Clearly, the small sample size was insufficient to provide an accurate estimate of the risk. Furthermore, transthoracic sonography is time consuming and no less expensive than CXR.
On multivariate analysis in the present study, the number of biopsy specimens obtained was the only statistically significant prediction factor for the development of pneumothorax. Other factors investigated, namely, suspected diagnosis at time of FB, segment from which biopsy specimens were obtained, biopsy forceps diameter, immune status, and experience of the clinician (fellow vs senior pulmonologist) had no predictive value. However, despite our relatively large sample size, the small number of pneumothoraces may well have masked an effect of one or more of these variables.
Table 2 shows a review of the literature on the rate of pneumothorax following TBB. In our study, 10 of the 350 patients (2.9% of all TBB procedures) had a pneumothorax. In 91 patients (26%), the indication for FB with TBB was lung transplantation. This high percentage of patients undergoing transplantation might not be representative of the patient profile in most institutions. Moreover, as many of the patients undergoing transplantation have obliterated pleural space, this might explain that no pneumothorax was observed in that population. When not considering the patients undergoing transplantation, the pneumothorax rate was 3.9% (10 of 259 cases). This rate possibly represents the true percentage of pneumothorax in our general nontransplanted population.
|
We conclude that routine CXR after TBB is not necessary in asymptomatic patients. Nonetheless, taking our data together with previous studies,67 routine CXR following TBB should be recommended in the restricted group of individuals at high risk for pneumothorax, such as patients requiring positive pressure ventilation, and patients unable to report symptoms of pneumothorax.
Acknowledgements
The authors thanks Dr. T. K. Aldrich, Chief, Unified Pulmonary Division of the Albert Einstein College of Medicine, Bronx, NY, for critical review of this article.
Footnotes
Abbreviations: CXR = chest radiograph; FB = flexible bronchoscopy; TBB = transbronchial biopsy
No financial support was received for this study.
Received for publication September 21, 2005. Accepted for publication November 24, 2005.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
