Endobronchial Ultrasonography-Guided Transbronchial Needle Aspiration Increases the Diagnostic Yield of Peripheral Pulmonary Lesions

CHEST

Original Research INTERVENTIONAL PULMONOLOGY

Endobronchial Ultrasonography-Guided Transbronchial Needle Aspiration Increases the Diagnostic Yield of Peripheral Pulmonary Lesions A Randomized Trial Tung-Ying Chao, MD; Min-Te Chien, MD; Chien-Hao Lie, MD; Yu-Hsiu Chung, MD; Jui-Long Wang, MD; and Meng-Chih Lin, MD

Background: The diagnostic yield of endobronchial ultrasonography (EBUS)-guided transbronchial needle aspiration (TBNA) for peripheral pulmonary lesions (PPLs) has not been evaluated. The diagnostic impact of TBNA when the EBUS probe is adjacent to lesions remains to be determined. Design: A prospective, randomized trial. Methods: Two hundred two patients with PPLs and positive EBUS findings were enrolled. They were randomly classified into two groups. In the EBUS conventional diagnostic procedures (CDPs) group (103 patients), both transbronchial biopsy (TBB) and bronchial washing (BW) were performed. In the EBUS-TBNA plus CDPs group (99 patients), TBNA, TBB, and BW were performed. The diagnostic yield in each group was compared. Results: A total of 182 patients (94 in the EBUS CDPs group and 88 in the EBUS-TBNA plus CDPs group) were analyzed. The yield in the EBUS-TBNA plus CDPs group (78.4%) was significantly higher than the EBUS CDPs group (60.6%, p ⴝ 0.015). Cases in which the EBUS probe was located within the lesions had a significantly higher diagnostic yield (78.3%) than when the EBUS probe was adjacent to them (47.2%, p < 0.001). Concerning the three different techniques, TBNA showed the highest diagnostic yield (62.5%) in comparison to TBB (48.9%) and to BW (19.8%). The diagnostic yield of TBNA remained unchanged even when the EBUS probe was adjacent to the lesions (p ⴝ 0.89). No additional adverse effects were observed in the EBUSTBNA plus CDPs group. Conclusions: Applying TBNA to EBUS-guided CDPs further increased the diagnostic yield of PPLs without additional risk. The diagnostic advantage of TBNA became more obvious if the EBUS probe was adjacent to the lesions. Trial registration: Clinicaltrials.gov Identifier: NCT00626587. (CHEST 2009; 136:229 –236) Abbreviations: BW ⫽ bronchial washing; CDP ⫽ conventional diagnostic procedure; EBUS ⫽ endobronchial ultrasonography; PPL ⫽ peripheral pulmonary lesion; ROSE ⫽ rapid on-site evaluation; TBB ⫽ transbronchial biopsy; TBNA ⫽ transbronchial needle aspiration

diagnosis of peripheral pulmonary lesions T he(PPLs) remains a clinical challenge for chest

physicians. Flexible bronchoscopy with sampling procedures is recognized as the “gold standard” to obtain the correct diagnosis of PPLs. Conventional diagnostic procedures (CDPs) for PPLs include transbronchial biopsy (TBB), bronchial washing (BW), or bronchial brushing, but the diagnostic yields are variable and sometimes suboptimal (diagnostic rate

www.chestjournal.org

approximately 18 to 62%).1 Many modified modalities have been effective in improving the diagnostic yield of PPLs, including combination of current sampling procedures,2 TBB by endobronchial ultrasonography (EBUS) guidance,3 sampling by guide-sheath guidance,4 virtual bronchoscopic navigation systems,5 and angled-forceps biopsy.6 Both fluoroscopy and EBUS can guide sampling procedures to gain cytopathologic specimens for PPLs.3,7 Moreover, characteristics of CHEST / 136 / 1 / JULY, 2009

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians

229

EBUS images can provide valuable information to differentiate the nature of PPLs.8,9 Transbronchial needle aspiration (TBNA) was first described in 1949 by Schieppati.10 It was a useful bronchoscopic technique to sample mediastinal lymphadenopathy and stage lung cancer.11,12 With the development of real-time EBUS-TBNA, TBNA could provide more accurate diagnoses of mediastinal and hilar lymphadenopathies.13 The efficacy of TBNA in the diagnosis of mediastinal benign diseases such as tuberculous lymphadenitis or sarcoidosis has also been proved.14 –16 Moreover, the utility of TBNA in the investigation of exophytic endobronchial lesions, submucosal diseases, and peribronchial diseases of the central airways has been well surveyed.17,18 However, the diagnostic role of TBNA for PPLs remains to be determined because many of the published studies19 –22 were retrospective or had small sample sizes. This may explain the fact that TBNA has always been underutilized for PPLs by bronchologists. With the popular application of EBUS-guided procedures for PPLs in clinical settings, whether EBUS-guided TBNA can contribute to a clinical impact on diagnosis of PPLs remained uncertain. No studies have been specifically aimed at the role of EBUS-guided TBNA for PPLs. To our knowledge, EBUS-guided sampling procedures were mostly successful when the EBUS probe could be put within the lesions.4 The diagnostic yield of current procedures (TBB and bronchial brushing) when the EBUS probe was adjacent to the lesions was very low. Alternative approaches such as TBNA could be more reasonable or preferable in this situation, but no study had tested this hypothesis. We designed a randomized, prospective study to evaluate the following: (1) the diagnostic yield of EBUS-guided TBNA in PPLs, and (2) the role of TBNA when the EBUS probe was adjacent to the lesions. From the Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Chang Gung Memorial Hospital-Kaohsiung Medical Center, Chang Gung University College of Medicine, Kaohsiung, Taiwan. This study was conducted in the Chang Gung Memorial HospitalKaohsiung Medical Center. The authors have no conflicts of interest to disclose. Manuscript received March 2, 2008; revision accepted August 25, 2008. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/site/misc/reprints.xhtml). Correspondence to: Meng-Chih Lin, MD, Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine, Chang Gung Memorial Hospital, 123 Dabi Rd, Niaosung Shiang, Kaohsiung, Taiwan 833, ROC; e-mail: [email protected] DOI: 10.1378/chest.08-0577

Materials and Methods Subjective Our study prospectively evaluated the clinical impact of EBUSguided TBNA on the diagnostic yield of PPLs. This randomized, interventional trial was approved by the Research Ethics and Institutional Review Board of Chang Gung Memorial Hospital in Kaohsiung, Taiwan. All patients provided informed consent before the EBUS-guided procedures. Equipment and Methods All patients underwent bronchoscopy (P260F; Olympus; Tokyo, Japan). EBUS was performed using an endoscopic ultrasound system (EU-M30; Olympus) and a 20-MHz miniature radial probe (UM-S20 –20R; Olympus). After premedication with local anesthesia lidocaine, a bronchoscope was introduced transnasally. An EBUS probe was inserted through the working channel into the target bronchus based on radiographic findings. The EBUS probe was put within the lesions as possible (avoiding being placed adjacent to the lesions) [Fig 1]. Once the location of the target lesion was identified precisely by EBUS, the EBUS probe was marked by colored tape against the orifice of the working channel of the bronchoscope. Then the EBUS probe was pulled out slowly. When the transducer of the EBUS probe reached the orifice of the subsegmental bronchus, the distance between colored tape on the probe and the orifice of the working channel was measured by an assistant. The EBUS probe was then withdrawn. TBNA and TBB followed by BW were performed without fluoroscopic guidance. No extended working channel (guide sheath) was left in situ. A more detailed description of bronchoscopic sampling procedures can be found in a previous report23 from our group. Patients Between January 1, 2005, and December 31, 2006, at the bronchoscopy unit of Chang Gung Memorial Hospital, a total of 1,816 patients underwent bronchoscopy for various indications. There were 507 patients with lung nodules or mass on chest radiograph referred for diagnostic bronchoscopy. Lesions not visible by bronchoscopy were defined as PPLs (no findings of endobronchial lesions, extrinsic compression, submucosal infiltration, or orifice narrowing). Three hundred sixty-two of the 507 patients had PPLs and received EBUS for lesion localization. Two hundred eighty-one patients whose lesions were identified on EBUS images were candidates for our study. We excluded patients who received repeated bronchoscopy, refused the sampling procedures, or refused the randomization protocol. We randomly assigned 202 patients to undergo EBUS-guided TBB and BW (EBUS CDPs group, n ⫽ 103) or EBUS-guided TBNA, TBB, and BW (EBUS-TBNA plus CDPs group, n ⫽ 99). We excluded seven patients with bacterial pneumonia because the diagnosis was made clinically rather than bronchoscopically. Our study algorithm is shown in Figure 2. Sampling Procedures Equipment TBNA (NA-2C-1; Olympus) was performed through the working channel of the bronchoscope. When the tip of the needle reached the orifice of the subsegmental bronchus, the coil was held by thumb and index finger of the bronchoscopist at the outer edge of the orifice of the working channel with the measured distance. The needle was then introduced into the target bronchial branch and advanced until the fingers touched the orifice of

230

Original Research

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians

Figure 1. EBUS probe location can be categorized into two groups: probe within the lesions, and probe adjacent to the lesions. Here are the EBUS images are from a 72-year-old man with left upper lobe non-small cell lung cancer. A: In the LB3a bronchus, the probe (arrow) was located within the lesion. B: In the LB3b bronchus, the probe (arrow) was located adjacent to the lesion. Sampling procedures were performed via LB3a bronchus to give the highest diagnostic yield.

the working channel. At that time, the needle was pushed out and negative manual suction was applied with a 20-mL syringe. The specimens were then smeared on glass slides and immersed in 95% alcohol. At least three aspirates per target lesion were obtained. A chest radiograph from all patients was obtained 1 to 2 h after bronchoscopy to determine whether pneumothorax had occurred. In order to eliminate the interperformer variability of TBNA,24 –26 procedures in the EBUS-TBNA plus CDPs group were all performed by a bronchoscopist (T.-Y.C.) with ⬎ 2 years of experience in performing TBNA. Patients in the EBUS CDPs

group underwent sampling procedures by three other bronchoscopists with ⬎ 10 years of experience in bronchoscopic procedures. TBB was performed by biopsy forceps (FB-19C-1 or FB15C-1; Olympus). At least three specimens were obtained by TBB. BW was performed by instilling 50 mL of sterile isotonic NaCl solution into the bronchus followed by immediate vacuum aspiration. Rapid on-site evaluation (ROSE) was not utilized during bronchoscopic procedures. All specimens were analyzed by two study-blinded cytopathologists.

Figure 2. Algorithm of protocol in our study and detailed characteristics of unselected patients. A total of 182 patients were analyzed. LUL ⫽ left upper lobe. www.chestjournal.org

CHEST / 136 / 1 / JULY, 2009

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians

231

Statistical Analysis Nominal variables, including gender, lesion location, complications, EBUS probe location, and diagnostic yields, were expressed as numbers and frequency (percentage). Parameters including age and lesion size were expressed as mean ⫾ SD. For categorical variables, comparisons were done with the ␹2 test. For continuous variables, comparisons were done with the Student t test. A multivariate logistic regression test was used to confirm the independent variables that influenced the diagnostic yields. Data were analyzed using the statistical software (Statistical Package for the Social Sciences version 13.0; SPSS; Chicago, IL). Values of two-sided p ⬍ 0.05 were considered statistically significant.

Results Patients Characteristics and Diagnoses A total of 182 patients (111 men and 71 women) were analyzed (Fig 2). Mean age was 62.3 years (range, 19 to 90 years). Mean diameter of the PPLs was 34.9 ⫾ 8.9 mm (range, 15.0 to 56.0 mm). Overall, the diagnostic yield was 69.2% (126 of 182 patients). If the diagnosis could not be made using bronchoscopy, further workup included chest ultrasonographyguided transthoracic biopsy, CT-guided biopsy, or operation. When no histologic diagnosis could be made, the final diagnosis was obtained by clinical follow-up and therapeutic response. Table 1 summarizes the final diagnoses in the 182 patients. Overall, 77.5% (141 of 182 patients) had malignant diseases, while the remaining 22.5% (41 of 182 patients) had benign diseases. In 131 patients with primary lung cancer, adenocarcinoma (78 patients, 59.5%) was the most common cell type. Tuberculosis was found in 75.6% (31 of 41 patients) of those with benign diseases. This reflected the endemic tuberculosis

Table 1—Final Diagnosis in 182 Patients With PPLs Diagnosis

No. (%)

Malignancy Adenocarcinoma Squamous cell carcinoma Small cell carcinoma Adenosquamous carcinoma Bronchoalveolar carcinoma Undifferentiated non-small cell lung cancer Mucoepidermoid carcinoma Metastasis Benign Tuberculosis Nontuberculous mycobacterium Cryptococcosis Actinomycoses Angiofibroma Sclerosing hemangioma Total

141 (77.5) 78 (42.9) 32 (17.6) 3 (1.6) 1 (0.5) 2 (1.1) 15 (8.2) 1 (0.5) 9 (4.9) 41 (22.5) 31 (17) 2 (1.1) 5 (2.7) 1 (0.5) 1 (0.5) 1 (0.5) 182 (100)

burden in Taiwan. Tuberculosis was diagnosed clinically in 5 of 31 patients (with improvements of symptoms or radiologic findings after antituberculosis medication for 6 months). Results of Different Bronchoscopic Procedures Eighty-eight patients were in the EBUS-TBNA plus CDPs group and 94 patients were in the EBUS CDPs group. Characteristics of the patients in the two groups are compared in Table 2. There were no clinically significant differences in baseline characteristics between the two groups, including age, gender, lesion size, lesion location, EBUS probe location, and final diagnoses. Overall, four patients had pneumothorax and six patients had bleeding after sampling procedures. The complication rates between the two groups did not differ. All complications in our study were self-limited, and none required tube thoracostomy or endotracheal intubation. The diagnostic yield in the EBUS-TBNA plus CDPs group (69 of 88 patients, 78.4%) was significantly higher than in the EBUS CDPs group (57 of 94 patients, 60.6%) [p ⫽ 0.015]. The application of TBNA increased the overall diagnostic yield of PPLs.

Table 2—Characteristics of Patients in Two Groups of EBUS-Guided Sampling Procedures

Characteristics Age, yr Male/female gender Size, mm Location of lesion, No. Right upper lobe Right middle lobe Right lower lobe Left upper lobe Lingular lobe Left lower lobe EBUS probe location, No. Within the lesions Adjacent to the lesions Complications, No. Pneumothorax Bleeding Diagnostic yield (sensitivity) Overall Malignancy Benign Final diagnoses, No. Malignancy Benign

EBUS-TBNA EBUS CDPs Plus CDPs Group Group (n ⫽ 88) (n ⫽ 94) p Value 61.7 ⫾ 12.5 57/31 34.6 ⫾ 9.5

62.9 ⫾ 12.5 54/40 35.1 ⫾ 8.3

30 9 21 12 6 10

27 14 16 18 5 14

NS NS NS NS

NS 66 22

63 31

2 4

2 2

78.4 (69/88) 79.2 (57/72) 75.0 (12/16)

60.6 (57/94) 56.5 (39/69) 72.0 (18/25)

0.015 0.006 NS

72 16

69 25

NS

NS

Data are presented as mean ⫾ SD or % (No./total) unless otherwise indicted. NS ⫽ not significant.

232

Original Research

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians

Table 3—Comparisons of Diagnostic Yields Among Three Different Procedures Variables

BW

TBB

TBNA

Total samples, No. 182 182 88 Positive samples, No. 36 89 55 Diagnostic rates, % 19.8 48.9 62.5 p Value ⬍ 0.001 0.049 Diagnostic sensitivity for 19/141 (13.5) 71/141 (50.4) 52/72 (72.2) malignancy p Value ⬍ 0.001 0.004 Diagnostic sensitivity for 17/41 (41.5) 18/41 (43.9) 3/16 (18.8) benign p Value NS NS Data are presented as No./total (%) unless otherwise indicated. All p values are comparison with TBNA. See Table 2 for expansion of abbreviation.

If we take the diagnostic sensitivity for malignancy into account, the addition of TBNA can increase the diagnostic sensitivity for malignancy from 56.5% (39 of 69 cases) to 79.2% (57 of 72 cases) [p ⫽ 0.006]. Concerning the three different bronchoscopic techniques, TBNA showed the highest diagnostic yield (55 of 88 patients, 62.5%), in comparison to TBB (89 of 182 patients, 48.9%; p ⫽ 0.049, ␹2), and to BW (36 of 182 patients, 19.8%; p ⬍ 0.001, ␹2) [Table 3]. If we only took the malignant lesions into account, TBNA also showed the highest diagnostic sensitivity (52 of 72 patients, 72.2%), in comparison to TBB (71 of 141 patients, 50.4%; p ⫽ 0.004, ␹2), and to BW (19 of 141 patients, 13.5%; p ⬍ 0.001, ␹2). For benign lesions, the diagnostic yields (sensitivity) of the three procedures did not differ. EBUS Probe Location and Diagnostic Yields We categorized the 182 PPLs into two groups by EBUS probe location (Table 4). Probe location was described as within the lesions in 129 patients (70.9%) and adjacent to the lesions in 53 patients (29.1%). There were no difference in lesion size and

final diagnoses (malignancy or benign) between the two groups. Moreover, the pathologic types in malignant lesions (primary lung cancer or metastasis) did not differ between the two groups (p ⫽ 0.46). Cases in which the probe was located within the lesions had a significantly higher diagnostic yield (101 of 129 patients, 78.3%) than when the probe was located adjacent to the lesions (25 of 53 patients, 47.2%) [p ⬍ 0.001]. Furthermore, we focused on the diagnostic yields of three different techniques. If the probe could only be put adjacent to the lesions rather than in the lesions, the yields of TBB and BW both decreased significantly (p ⬍ 0.001 and p ⫽ 0.041, respectively). However the yield of TBNA did not differ between a group as probe within lesions and probe adjacent to lesions (p ⫽ 0.89). While the EBUS probe was adjacent to the lesions, TBNA had the highest possibility to get a positive diagnostic result. Factors Influencing the Diagnostic Yield of TBNA We tried to identify the clinical factors that influenced the diagnostic yield of EBUS-guided TBNA (Table 5). A multivariate logistic regression model was used to search for the independent factors. Lesion size (p ⫽ 0.015) and nature of the lesions (p ⬍ 0.001) were statistically significant to influence the diagnostic yield of TBNA. Whether the EBUS probe was within or adjacent to the lesions did not influence the diagnostic yield of TBNA (p ⫽ 0.312). Discussion The efficacy of EBUS in evaluating the internal structures,8,9 identification, and guiding the sampling procedures of PPLs3,7,27,28 has been proved. To our knowledge, EBUS-guided CDPs for PPLs have always been limited to TBB and bronchial brushing. Since the diagnostic yields of TBB and bronchial brushing are sometimes low (especially when lesion

Table 4 —Comparison Between Groups of EBUS Probe Within or Adjacent to the Lesions Variables

Probe Within Lesions (n ⫽ 129)

Probe Adjacent to Lesions (n ⫽ 53)

p Value

Lesion size, mm Final diagnosis, No. Malignancy Lung cancer Metastasis Benign Diagnostic yield TBNA-positive rate TBB-positive rate BW-positive rate

35.5 ⫾ 9.1

33.3 ⫾ 7.9

NS NS

97 92 5 32 78.3 (101/129) 63.6 (42/66) 59.7 (77/129) 24.0 (31/129)

44 40 4 9 47.2 (25/53) 59.1 (13/22) 22.6 (12/53) 9.4 (5/53)

⬍ 0.001 NS ⬍ 0.001 0.041

Data are presented as mean ⫾ SD or % (No./total) unless otherwise indicted. See Table 2 for expansion of abbreviation. www.chestjournal.org

CHEST / 136 / 1 / JULY, 2009

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians

233

Table 5—Factors Influencing the Diagnostic Yield of EBUS-Guided TBNA (n ⴝ 88) Variables Size, mm Probe location Within lesion Adjacent to lesion Final diagnosis Malignancy Benign

Positive Diagnosis (n ⫽ 55)

Negative Diagnosis (n ⫽ 33)

36.1 ⫾ 9.9

32.0 ⫾ 8.2

42 13

24 9

52 3

20 13

Univariate p Value

Multivariate p Value

Odds Ratio

95% Confidence Interval

0.052 0.89

0.015 0.312

0.925 0.567

0.869–0.985 0.189–1.704

⬍ 0.001

⬍ 0.001

0.046

0.009–0.226

In this logistic regression model, the odds ratio for size increases per millimeter. For probe location and final diagnosis, the odds ratio is dichotomous. Data are presented as mean ⫾ SD or No.

size was small or the EBUS probe was adjacent to the lesions),4,23 it is reasonable to combine current available sampling procedures for PPLs. In previous small or retrospective studies, the diagnostic yield of PPLs increased from 35 to 51%21 and from 46 to 70%22 when TBNA was added to CDPs (with fluoroscopic guidance). Moreover, the diagnostic yield of TBNA for PPLs has also been reported to be higher than CDPs.20 –22 While previous studies were restricted to fluoroscopy-guided TBNA, we first evaluated the clinical impact of EBUS-guided TBNA in a larger consecutive series of patients with PPLs. In our consecutive series of 182 patients, the overall diagnostic yield of bronchoscopy increased from 60.6 to 78.4% (p ⫽ 0.015) by adding EBUSguided TBNA to the CDPs (TBB plus BW) [Table 2]. No additional complications were noted while performing TBNA. Only one case with a left upper lobe nodule was not accessible by TBNA. We found that EBUS-guided TBNA was feasible and effective to diagnose PPLs. Surprisingly, the diagnostic yield of the EBUS-TBNA plus CDPs group reached 78.4% even in the absence of fluoroscopic guidance. Some studies3,8,23,29 have also concluded that EBUSguided bronchoscopy without fluoroscopic guidance was effective to diagnose PPLs. We believed that all available bronchoscopic procedures (bronchial brushing, BW, TBB, and TBNA) should be used to give the best diagnostic yield for PPLs. Some studies24 –26 have declared that a training period was necessary for a bronchoscopist to get the best yield of TBNA. In order to eliminate the interperformer variability, all procedures in the EBUS-TBNA plus CDPs group were performed by the same bronchoscopist. As a result, the yield of TBNA was 62%, considerably higher than TBB and BW (Table 3). The diagnostic yield of BW in malignant lesions (19 of 141 cases, 13.5%) was significant lower than benign lesions (17 of 41 cases, 41.5%) [p ⬍ 0.001]. The value of routine BW cytologic examination for PPLs must be balanced with the cost-effectiveness.30

Kurimoto et al4 found that the yield of sampling procedures (bronchial brushing and TBB) when the EBUS probe was adjacent to the lesions was very low (37% and 7%, respectively). The author hypothesized that the lesions which were adjacent to the EBUS probe only compressed or contacted the outer surface of the bronchi rather than penetrating. This hypothesis provided a reasonable diagnostic role of TBNA to get accurate specimens in this circumstance. Theoretically, the TBNA needle can penetrate into the submucosa or bronchial wall and sample the submucosal or peribronchial tissues.17,18 This fact can explain why current “intrabronchial” sampling techniques (bronchial brushing, BW, and TBB) often fail to diagnose the lesions that are adjacent to the EBUS probe. In our study, the overall diagnostic yield when the EBUS probe was located within the lesions was significantly higher (78.3%) than when the probe was located adjacent to the lesions (47.2%, p ⬍ 0.001) [Table 4]. Furthermore, if the EBUS probe could only be put adjacent to the lesions rather than within the lesions, the diagnostic yields of TBB and BW decreased significantly (p ⬍ 0.001 and p ⫽ 0.041, respectively) [Table 4]. However, the diagnostic yield of TBNA remained unchanged (p ⫽ 0.89). In the probe adjacent to lesions group (53 patients), 22 patients received TBNA and 13 patients (59.1%) had positive results. Eleven of these 13 patients (all were malignant lesions) had TBNA as the exclusive diagnostic test. When the EBUS probe was adjacent to the lesions, TBNA was the best diagnostic procedure to diagnose PPLs. Under multivariate logistic regression analysis, the yield of TBNA was mainly determined by lesion size and nature of the lesions (malignancy or benign) [Table 5]. In other words, if we only took TBNA into consideration, the location of the EBUS probe was no longer the independent factor to influence the diagnostic yield. In 16 patients with benign lesions (13 patients with tuberculosis) receiving TBNA, a diagnosis using TBNA was made in only 3 patients.

234

Original Research

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians

In the remaining 13 patients, TBNA revealed nonspecific inflammation or normal epithelial cells, and failed to determine the nature of the lesions. We believed that TBNA could not give enough tissue to establish the diagnoses of benign diseases (such as tuberculosis or fungal infections). There are three potential limitations in our study. First, interperformer variability could exist if the physicians performing the procedures in two groups were different. In order to minimize the interperformer variability, bronchoscopic procedures were performed by four well-trained bronchoscopists who had ⬎ 10 years of experience. Second, we did not know whether we returned to the same target lesion for the subsequent three TBNA (or TBB) procedures when the EBUS probe was withdrawn. In our bronchoscopic center, guide sheaths and fluoroscopy were not available. Our practice (EBUS with measurement of distance) was an indirect confirmation of target site. The aim of our study was to test whether EBUS-guided TBNA could increase the diagnostic yield of PPLs (especially when the EBUS probe was adjacent to the lesions). Further prospective studies are crucial to test whether EBUS with guide sheath TBNA can further increase the diagnostic yield of PPLs. Third, ROSE was not utilized during bronchoscopic procedures. We performed at least three serial TBNA passes per target site for PPLs. Further prospective studies with ROSE must be conducted to assess how many passes of TBNA (with or without guide sheath) should be performed.

Conclusion EBUS-guided TBNA was a promising modality for diagnosing PPLs without additional risk. The diagnostic advantage of TBNA became more obvious in the following two circumstances: (1) when the EBUS probe was adjacent to the lesions; and (2) when the lesions were malignant in nature.

References 1 Baaklini WA, Reinoso MA, Gorin AB, et al. Diagnostic yield of fiberoptic bronchoscopy in evaluating solitary pulmonary nodules. Chest 2000; 117:1049 –1054 2 Chechani V. Bronchoscopic diagnosis of solitary pulmonary nodules and lung masses in the absence of endobronchial abnormality. Chest 1996; 109:620 – 625 3 Herth F, Ernst A, Becker HD. Endobronchial ultrasound guided transbronchial lung biopsy in solitary pulmonary nodules and peripheral lesions. Eur Respir J 2002; 20:972– 974 4 Kurimoto N, Miyazawa T, Okimasa S, et al. Endobronchial ultrasonography using a guide sheath increases the ability to diagnose peripheral pulmonary lesions endoscopically. Chest 2004; 126:959 –965 www.chestjournal.org

5 Asano F, Matsuno Y, Shinagawa N, et al. A virtual bronchoscopic navigation system for pulmonary peripheral lesions. Chest 2006; 130:559 –566 6 Sasada S, Ogata Y, Kobayashi M, et al. Angled forceps used for transbronchial biopsy in which standard forceps are difficult to manipulate: a comparative study. Chest 2006; 129:725–733 7 Herth FJ, Eberhardt R, Becker HD, et al. Endobronchial ultrasound-guided transbronchial lung biopsy in fluoroscopically invisible solitary pulmonary nodules: a prospective trial. Chest 2006; 129:147–150 8 Chao TY, Lie CH, Chung YH, et al. Differentiating peripheral pulmonary lesions based on images of endobronchial ultrasonography. Chest 2006; 130:1191–1197 9 Kurimoto N, Murayama M, Yoshioka S, et al. Analysis of the internal structure of peripheral pulmonary lesions using endobronchial ultrasonography. Chest 2002; 122:1887–1894 10 Schenk DA, Bryan CL, Bower JH, et al. Transbronchial needle aspiration in the diagnosis of bronchogenic carcinoma. Chest 1987; 92:83– 85 11 Wang KP, Terry PB. Transbronchial needle aspiration in the diagnosis and staging of bronchogenic carcinoma. Am Rev Respir Dis 1983; 127:344 –347 12 Harrow EM, Abi-Saleh W, Blum J, et al. The utility of transbronchial needle aspiration in the staging of bronchogenic carcinoma. Am J Respir Crit Care Med 2000; 161:601– 607 13 Yasufuku K, Chiyo M, Sekine Y, et al. Real-time endobronchial ultrasound-guided transbronchial needle aspiration of mediastinal and hilar lymph nodes. Chest 2004; 126:122–128 14 Bilac¸erog˘lu S, Gu¨nel O, Eriçs N, et al. Transbronchial needle aspiration in diagnosing intrathoracic tuberculous lymphadenitis. Chest 2004; 126:259 –267 15 Morales CF, Patefield AJ, Strollo PJ Jr, et al. Flexible transbronchial needle aspiration in the diagnosis of sarcoidosis. Chest 1994; 106:709 –711 16 Trisolini R, Agli LL, Cancellieri A, et al. The value of flexible transbronchial needle aspiration in the diagnosis of stage I sarcoidosis. Chest 2003; 124:2126 –2130 17 Dasgupta A, Jain P, Minai OA, et al. Utility of transbronchial needle aspiration in the diagnosis of endobronchial lesions. Chest 1999; 115:1237–1241 18 Caglayan B, Akturk UA, Fidan A, et al. Transbronchial needle aspiration in the diagnosis of endobronchial malignant lesions: a 3-year experience. Chest 2005; 128:704 –708 19 Shure DJ, Fedullo PF. Transbronchial needle aspiration of peripheral masses. Am Rev Respir Dis 1983; 128:1090 –1092 20 Wang KP, Haponik EF, Britt EJ, et al. Transbronchial needle aspiration of peripheral pulmonary nodules. Chest 1984; 86:819 – 823 21 Reichenberger F, Weber J, Tamm M, et al. The value of transbronchial needle aspiration in the diagnosis of peripheral pulmonary lesions. Chest 1999; 116:704 –708 22 Katis K, Inglesos E, Zachariadis E, et al. The role of transbronchial needle aspiration in the diagnosis of peripheral lung masses or nodules. Eur Respir J 1995; 8:963–966 23 Chung YH, Lie CH, Chao TY, et al. Endobronchial ultrasonography with distance for peripheral pulmonary lesions. Respir Med 2007; 101:738 –745 24 Haponik EF, Cappellari JO, Chin R, et al. Education and experience improve transbronchial needle aspiration performance. Am J Respir Crit Care Med 1995; 151:1998 – 2002 25 Rodríguez de Castro F, Díaz Lo´pez F, Serda` GJ, et al. Relevance of training in transbronchial fine-needle aspiration technique. Chest 1997; 111:103–105 26 Hsu LH, Liu CC, Ko JS. Education and experience improve CHEST / 136 / 1 / JULY, 2009

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians

235

the performance of transbronchial needle aspiration: a learning curve at a cancer center. Chest 2004; 125:532–540 27 Shirakawa T, Imamura F, Hamamoto J, et al. Usefulness of endobronchial ultrasonography for transbronchial lung biopsies of peripheral lung lesions. Respiration 2004; 71:260 –268 28 Paone G, Nicastri E, Lucantoni G, et al. Endobronchial ultrasound-driven biopsy in the diagnosis of peripheral lung lesions. Chest 2005; 128:3551–3557

29 Yoshikawa M, Sukoh N, Yamazaki K, et al. Diagnostic value of endobronchial ultrasonography with a guide sheath for peripheral pulmonary lesions without X-ray fluoroscopy. Chest 2007; 131:1788 –1793 30 van der Drift MA, van der Wilt GJ, Thunnissen FB, et al. A prospective study of the timing and cost-effectiveness of bronchial washing during bronchoscopy for pulmonary malignant tumors. Chest 2005; 128:394 – 400

236

Original Research

Downloaded from www.chestjournal.org by Kimberly Henricks on August 8, 2009 Copyright © 2009 American College of Chest Physicians