Factors That Affect the Diagnostic Yield of Endobronchial Ultrasonography-Assisted Transbronchial Lung Biopsy

Abstract

Background:

Endobronchial ultrasound (EBUS) is a relatively noninvasive procedure used to diagnose and stage lung cancer. Although EBUS-guided transbronchial lung biopsy (TBLB) procedures for peripheral lesions have higher diagnostic yields than traditional TBLB or fluoroscopy-assisted TBLB, the diagnostic yield is not 100%, and the reasons for this are not clear.

Subjects and Methods:

This retrospective study assessed what characteristics influence the diagnostic yield of EBUS-guided TBLB. EBUS was used to locate a single peripheral lung nodule or mass, and the lesion was biopsied and pathologically characterized. Parameters that were evaluated included patient demographics, lesion location, ease of tumor sampling, location of the EBUS probe relative to the lesion, pathological volume, tumor cell type, and whether physicians were under supervision.

Results:

Thirty-nine patients received EBUS-guided TBLB, which correctly identified 76.9% of the patients as having lung cancer. For the remaining patients, subsequent surgery indicated their tumors were malignant. Univariate logistic regression modeling indicated that only the location of the probe relative to the lesion was significantly associated with diagnostic yield of EBUS-guided TBLB. When the probe was directly within the lesion, it was 8.17 times (odds ratio 8.17; 95% confidence interval 1.41, 47.22; P=.019) more likely to have a successful TBLB than when the probe was adjacent to the lesion.

Conclusions:

In this study, the position of the probe relative to a peripheral lung lesion was associated with the diagnostic yield of EBUS-guided TBLB. Larger prospective studies are required to further assess what influences the diagnostic yield of this technology.

Introduction

Accurate tumor diagnosis improves prognosis and influences the choice of treatment.^1,2 A major challenge is to correctly diagnose lesions located in the lung periphery, which account for between 25% to 30% of lung cancers.² These lesions are outside the direct view of a bronchoscope, and consequently biopsy samples are often obtained blindly by methods including transbronchial lung biopsy (TBLB), bronchoalveolar lavage, bronchial washing, or brushing.^3,4 Moreover, without direct localization, the diagnostic yields vary widely (between 20% to 75%).^5,6 Image guidance techniques such as fluoroscopy have been used to improve the diagnostic yield of bronchoscopy and TBLB, but these approaches are costly and time consuming and expose patients and staff to radiation.⁷

Endobronchial ultrasound (EBUS) is a relatively noninvasive procedure that was developed for the diagnosis of parenchymal lung lesions and the evaluation of mediastinal lymph nodes for diagnosis and staging of lung cancer.^2,8 An advantage of EBUS is that it expands the view of the physician beyond the lumen of the airway.² EBUS-guided TBLB of peripheral lung lesions (particularly for solitary pulmonary nodules <3 cm in diameter) has higher diagnostic yields than TBLB without EBUS or bronchoscopy with fluoroscopy guidance.^9–12 The diagnostic yield for EBUS-guided TBLB ranges from about 58% to 80%.^10,12,13 It is unclear what characteristics of EBUS-guided TBLB limit its diagnostic yield.

In this retrospective single-site study, we investigated what factors impact the success rate of EBUS-guided TBLB. We hypothesized factors such as lesion location, size, position of the probe with respect to the lesion, cell type, etc., may contribute to reduction in the diagnostic yield of this technique. We found that the location of the probe relative to the lesion was significantly associated with the diagnostic yield and other parameters such as tumor size were not.

Subjects and Methods

This was a retrospective study of patients admitted to the Department of Pulmonary and Critical Care Medicine, Buddhist Tzu Chi General Hospital, Taipei Branch, New Taipei City, Taiwan, between October 2005 and May 2011. This study was approved by the Institutional Review Board of Buddhist Tzu Chi General Hospital, Taipei Branch.

Study patients

During the study time period, in total, 2639 patients received bronchoscopic examinations for various indications (Fig. 1). Among the 2639 patients, 334 received EBUS during the bronchoscopy procedure. Of these 334 patients, 228 received EBUS for evaluation and biopsy of endobronchial lesions versus 106 for peripheral lung lesions. For these 106 patients diagnosed with a peripheral solitary pulmonary lesion, EBUS was able to localize the lesion in 69 of the patients but was unable to localize the lesion in the other 37 patients. Thirty-nine of these 69 patients whose lung lesion could be localized by EBUS received EBUS-guided TBLB, and their data were included in this analysis. The other 30 patients whose peripheral lung lesions were localized by EBUS did not receive EBUS-guided TBLB because of patient refusal, and their data were not used for analysis.

FIG. 1.

Flow chart of retrospective identification of patients from the medical charts survey. EBUS, endobronchial ultrasonography; TBLB, transbronchial lung biopsy.

EBUS-guided TBLB

The probable location of the lesion was determined initially using conventional chest radiography and chest computed tomography. Following premedication of the patient with 2% lidocaine for local anesthesia, the bronchoscope (model BF-260, Olympus, Tokyo, Japan) was introduced transnasally. A lesion was considered to be in the lung periphery if it was not visible by bronchoscopy. After it was determined that the solitary pulmonary nodule or mass was not in the bronchoscopic-visible airways, an endoscopic ultrasound system (model (EU-M30; Olympus) EBUS probe (20-MHz miniature radial probe, UM-S20–20R; Olympus) was inserted through the working channel of the bronchoscope into the target bronchus and placed into all accessible segmental or subsegmental bronchi distal to the lesion. The probe was moved back and forth to obtain the best images. One physician reviewed all the EBUS images and double-checked the image features (peribronchial or parabronchial) to ensure they were adequate for localization of the lung lesion prior to TBLB.

Once the location of the lesion was identified by EBUS, the EBUS probe was slowly withdrawn, during which time the depth of the lesion was measured. Subsequently, a TBLB forceps (model FB-19C-1, Olympus) was inserted into the working channel of the bronchoscope to the same location and depth, and biopsy samples were taken. The standard procedure was to obtain six biopsy specimens per patient, but this number could be reduced depending upon the condition of the patient. All biopsy specimens were analyzed by a single independent experienced pathologist. Following the procedure, patients were observed for an additional 15 minutes before being released to monitor for complications such as chest pain and signs of pneumothorax. If chest pains occurred the patient was examined by chest X-ray.

The parameters that were evaluated included tumor size (both minimum and maximum sizes), tumor location, whether the tumor location was difficult to approach by the biopsy forceps (easy or difficult angle), EBUS feature (location of the probe relative to the lesion [for parabronchial infiltration the probe was adjacent to the lesion, and for peribronchial infiltration the probe was within the lesion]; Fig. 2), pathologic volume recorded by pathology report, number of biopsy samples taken, cancer cell type, and whether the physician performing the EBUS-guided TBLB was under supervision by a senior attending physician. The visiting physicians were all trained by a senior attending physician for 1–2 weeks prior to performing the procedures.

FIG. 2.

Graphic presentations of EBUS-assisted TBLB detection of peripheral bronchial nodules. (A) The result of a parabronchial mass identified by EBUS-assisted TBLB. (B) The result of a peribronchial mass identified by EBUS-assisted TBLB.

Statistical analysis

Comparability between the TBLB biopsies that were able to give a definitive diagnosis (positive) and those that were not (negative) was tested using independent two-sample t test and Fisher's exact test for categorical variables. Continuous variables were represented as mean±standard deviation values, and categorical data were represented by number (n) and percentage (%). The Mann–Whitney U test was used to determine the difference between the results of EBUS-guided TBLB for skewed variable (biopsy number and pathologic volume), and data were displayed as median (interquartile range, Q1–Q3). Logistic regression analysis was performed to analyze the odds ratio of significant factors associated with diagnostic yield of TBLB-guided EBUS. All statistic assessments are two sided and evaluated at the .05 level of significance. Statistic analyses were performed using SPSS version 15.0 statistics software (SPSS Inc., Chicago, IL).

Results

The patient population consisted of 19 males and 20 females with an average age of 65±13.86 years (range, 34–85 years). The bronchoscopy before EBUS was operated by experienced pulmonologists. For 31 (79.5%) of the patients the EBUS-guided TBLB was performed by a visiting staff member under supervision of a senior attending physician, and the other 8 (20.5%) patients received EBUS-guided TBLB by an visiting staff member with 2 weeks of training without the supervision of a senior physician (Table 1).

Table 1.

Comparison of Demographics and Clinical Measurements Between Transbronchial Lung Biopsy Results

	Total (n=39)	TBLB positive (n=30)	TBLB negative (n=9)	P value
Age (years)^a	65.00±13.86	63.33±14.77	70.56±8.76	.174
Gender^b				1.000
Male	19 (48.7)	14 (44.4)	5 (55.6)
VS experience^b				.653
Inexperienced	8 (20.5)	7 (23.3)	1 (11.1)
Experienced	31 (79.5)	23 (76.7)	8 (88.9)
Location^b				.373
Left lower lobe	8 (20.5)	7 (23.3)	1 (11.1)
Right lower lobe	13 (33.3)	10 (33.3)	3 (33.3)
Right middle lobe	3 (7.7)	3 (10.0)	0 (0.0)
Left upper lobe	5 (12.8)	2 (6.7)	3 (33.3)
Right upper lobe	10 (25.6)	8 (26.7)	2 (22.2)
Difficult angle^b				.444
Easy	23 (59.0)	19 (63.3)	4 (44.4)
Difficult	16 (41.0)	11 (36.7)	5 (55.6)
CT bronchus sign^b				.716
Positive	20 (51.3)	16 (53.3)	4 (44.4)
Negative	19 (48.7)	14 (46.7)	5 (55.6)
EBUS feature^b				.019^*
Peribronchial infiltration	23 (59.0)	21 (70.0)	2 (22.2)
Parabronchial infiltration	16 (41.0)	9 (30.0)	7 (77.8)
Size of lesion^a
Minimum length (cm)	2.38±0.89	2.44±0.80	2.18±1.18	.446
Maximum length (cm)	2.91±0.98	3.04±0.91	2.48±1.12	.129
Tumor size^b				.706
≤2 cm	18 (46.2)	13 (43.3)	5 (55.6)
>2 cm	51 (53.8)	17 (56.7)	4 (44.4)
Biopsy number^c	4 (3, 6)	4 (3, 6)	3 (2, 6)	.311
Pathologic volume (mm³)^c	2 (1, 6)	2.5 (1, 7)	1 (1, 3)	.187
Type of cell^b				1.000
Adenocarcinoma	31 (79.5)	24 (80.0)	7 (77.8)
Squamous carcinoma	8 (20.5)	6 (20.0)	2 (22.2)

Data are displayed as ^amean±standard deviation values, ^bnumber (percentage), or ^cmedian (interquartile range).

P values are based on ^aindependent two-sample t test, ^bFisher's exact test, and ^cMann–Whitney U test.

Significant difference between TBLB positive and TBLB negative, P<.05.

CT, computed tomography; EBUS, endobronchial ultrasonography; TBLB, transbronchial lung biopsy; VS, visiting staff.

Of the 39 patients, 30 (76.9%) had positive EBUS-guided TBLB that indicated their solitary pulmonary nodule was malignant in nature. These patients all received surgical intervention later for tumor eradication using video-assisted thoracoscopic surgery (VATS). The pathology of the surgically removed tissue corroborated the EBUS-guided TBLB findings. For the remaining 9 patients, the EBUS-guided TBLB did not reveal the nature of the lesion. Four of the 9 patients underwent computed tomography-guided biopsy, and the lesion was diagnosed as malignant prior to VATS. The remaining 5 patients were directly treated with VATS, and their lesions were subsequently determined to be cancerous.

All the lesions were visualized by EBUS. The localizations of the lesion were the left lower lobe in 8 patients (20.5%), the right lower lobe in 13 patients (33.3%), the right middle lobe in 3 patients (7.7%), the left upper lobe in 5 (12.8%), and the right upper lobe in 10 (25.6%). The position of the EBUS probe with respect to the lesion varied: the probe had a peribronchial infiltration for 23 (59.0%) of the lesions and parabronchial infiltration for the remaining 16 tumors (Table 1).

The patients' demographics and clinical features of the tumor and EBUS technical issues were similar between the patients whose lesion were or were not successfully diagnosed via TBLB except for the position of the probe with respect to the lesion (EBUS feature). Consistent with these findings, the univariate logistic regression model indicated that only the position of the probe relative to the lesion (EBUS feature) was significantly associated with the diagnostic yield of EBUS-guided TBLB (P=.024). When the probe had a peribronchial infiltration of the lesion, it was 8.17 times (odds ratio 8.17; 95% confidence interval 1.41, 47.22; P=.019) more likely to have a successful TBLB compared with when the probe had parabronchial infiltration (Table 2).

Table 2.

Results of Logistic Regression of Determining the Factors Affecting the Diagnostic Yield of Endobrachial Ultrasound-Guided Transbronchial Lung Biopsy

	Odds ratio (95% confidence interval)	P value
Age	0.96 (0.90, 1.02)	.178
Gender
Male	1.43 (0.32, 6.39)	.641
Female	1.00
VS experience
Inexperience	1.00
Experience	2.44 (0.26, 22.97)	.437
Location
Left lower lobe	1.00
Right lower lobe	0.48 (0.04, 5.58)	.555
Right middle lobe	—
Left upper lobe	0.10 (0.01, 1.50)	.094
Right upper lobe	0.57 (0.04, 7.74)	.674
Angle
Difficult	1.00
Easy	2.16 (0.48, 9.47)	.318
CT bronchus sign
Negative	1.00
Positive	1.43 (0.32, 6.39)	.641
EBUS feature
Parabronchial infiltration	1.00
Peribronchial infiltration	8.17 (1.41, 47.22)	.019^a
Size of lesion
Minimum length	1.44 (0.58, 3.60)	.437
Maximum length	2.02 (0.80, 5.08)	.135
Tumor size
≤2 cm	1.00
>2 cm	1.64 (0.37, 7.33)	.521
Biopsy number	1.20 (0.77, 1.87)	.414
Pathologic volume	1.19 (0.89, 1.59)	.252
Type of cell
Adenocarcinoma	1.00
Squamous carcinoma	0.88 (0.14, 5.34)	.85

Significant factor, P<.05.

Discussion

Previous studies have investigated whether EBUS assistance improves the diagnostic yield of TBLB. This study investigated why EBUS-guided TBLB does not achieve a diagnostic yield of 100% and, which is important, was designed to accurately measure the success rate of EBUS-guided TBLB by focusing on patients who were confirmed to have malignant lesions.

We found that the position of the probe relative to the lesion (EBUS feature) was crucial in obtaining an accurate diagnosis of the lesion. When the EBUS image showed peribronchial infiltration, there was a greater chance of a proper diagnosis than with parabronchial infiltration. This in part reflects the fact that when the probe was within the lesion (peripheral infiltration), the EBUS image showed continuous echogenicity around the probe, whereas when it was adjacent to the lesion (parabronchial infiltration), the EBUS image showed only partial echogenicity around the probe. These findings are consistent with prior results that found the position of the probe either within or adjacent to the lesion was a significant factor in predicting the diagnostic yield of EBUS-guided TBLB.^13–16 The size of the lesion has been suggested to be a factor in determining the diagnostic yield^15,17; however, in this and another study¹⁴ lesion size was not a significant predictor of yield. Parameters such as whether physicians were under supervision, gender, age, biopsy number, cancer type, and bronchus sign did not significantly affect diagnostic yield.

The diagnostic yield in this study for EBUS-guided TBLB was 76.9%, which is consistent with prior studies that found diagnostic yields to be 69%–86%.^{7,9,10,14,18,19} One retrospective study found that the diagnostic yield rose from 65% to 97% with increasing number of biopsy samples, with five samples being the optimum number.¹⁴ This may reflect the fact that guide sheath does not always remain well positioned during withdrawal of the EBUS probe, tissue between the probe and the tumor may need to be destroyed for an accurate biopsy, and for patients with benign disease not all biopsy specimens are supportive of an accurate diagnosis.

This was a small retrospective study using data from a single local hospital. Larger prospectively defined studies are required to further assess the findings of this study and to better evaluate the different parameters that may affect diagnostic yield of EBUS-guided TBLB. We only reported patients who were diagnosed with a malignant lesion, which may have induced selection bias and resulted in the high success rate.

In this study, we found that the position of the probe (EBUS feature) was significantly associated with diagnostic yield of EBUS-guided TBLB. Understanding what variables influence the diagnostic yield of EBUS-guided TBLB may help improve the usefulness of this technology in diagnosing and treating peripheral lung cancer.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Beadsmore

, Screaton

. Classification, staging and prognosis of lung cancer. Eur J Radiol, 2003; 45:8–17.

Gomez

, Silvestri

. Endobronchial ultrasound for the diagnosis and staging of lung cancer. Proc Am Thorac Soc, 2009; 6:180–186.

Silvestri

, Gould

, Margolis

, Tanou

, McCrory

, Toloza

, Detterbeck

. Noninvasive staging of non-small cell lung cancer: ACCP evidenced-based clinical practice guidelines (2nd edition) Chest, 2007; 132,3 Suppl:178S–201S.

Gould

, Fletcher

, Iannettoni

, Lynch

, Midthun

, Naidich

, Ost

. Evaluation of patients with pulmonary nodules: When is it lung cancer?: ACCP evidence-based clinical practice guidelines (2nd edition) Chest, 2007; 132:108S–130S.

Torrington

, Kern

. The utility of fiberbronchoscopy in the evaluation of solitary pulmonary nodules. Chest, 1993; 104:1021–1024.

Gasparini

, Ferretti

, Bichi Secchi

, Baldelli

, Zuccatosta

, Gusella

. Integration of transbronchial and percutaneous approach in the diagnosis of peripheral pulmonary nodules or masses. Experience with 1027 cases. Chest, 1995; 108:131–137.

Paone

, Nicastri

, Lucantoni

, Iacono

, Battistoni

, D'Angeli

, Galluccio

. Endobronchial ultrasound-driven biopsy in the diagnosis of peripheral lung lesions. Chest, 2005; 128:3551–3557.

Hurter

, Hanrath

. Endobronchial sonography: Feasibility and preliminary results. Thorax, 1992; 47:565–567.

Herth

FJF

, Ernst

, Becker

. Endobronchial ultrasound-guided transbronchial lung biopsy in solitary pulmonary nodules and peripheral lesions. Eur Respir J, 2002; 20:972–974.

10.

Herth

, Eberhardt

, Becker

, Ernst

. Endobronchial ultrasound-guided transbronchial lung biopsy in fluoroscopically invisible solitary pulmonary nodules: A prospective trial. Chest, 2006; 129:147–150.

11.

Baaklini

, Reinoso

, Gorin

, Sharafkaneh

, Manian

. Diagnostic yield of fiberoptic bronchoscopy in evaluating solitary pulmonary nodules. Chest, 2000; 117:1049–1054.

12.

Yang

, Liu

, Wang

, Lin

, Chen

, Chou

, Hsueh

, Kuo

. Diagnostic value of endobronchial ultrasound-guided transbronchial lung biopsy in peripheral lung cancers. J Formos Med Assoc, 2004; 103:124–129.

13.

Shirakawa

, Imamura

, Hamamoto

, Honda

, Fukushima

, Sugimoto

, Shirkakusa

. Usefulness of endobronchial ultrasonography for transbronchial lung biopsies of peripheral lung lesions. Respiration, 2004; 71:260–268.

14.

Yamada

, Yamazaki

, Kurimoto

, Asahina

, Kikuchi

, Shinagawa

, Oizumi

, Nishimura

. Factors related to diagnostic yield of transbronchial biopsy using endobronchial ultrasonography with a guide sheath in small peripheral pulmonary lesions. Chest, 2007; 132:603–608.

15.

Huang

, Ho

, Tsai

, Yu

, Yang

. Factors influencing visibility and diagnostic yield of transbronchial biopsy using endobronchial ultrasound in peripheral pulmonary lesions. Respirology, 2009; 14:859–864.

16.

Kurimoto

, Miyazawa

, Okimasa

, Maeda

, Oiwa

, Miyazu

, Murayama

. Endobronchial ultrasonography using a guide-sheath increases the ability to diagnose peripheral pulmonary lesions endoscopically. Chest, 2004; 126:959–965.

17.

Kikuchi

, Yamazki

, Sukoh

, Kikuchi

, Asahina

, Imura

, Onodera

, Kurimoto

, Kinoshita

, Nishimura

. Endobronchial ultrasonography with guide-sheath for peripheral pulmonary lesions. Eur Respir J, 2004; 24:533–537.

18.

Eberhardt

, Ernst

, Herth

FJF

. Ultrasound-guided transbronchial biopsy of solitary pulmonary nodules less than 20 mm. Eur Respir J, 2009; 34:1284–1287.

19.

Disayabutr

, Tscheikuna

, Nana

. The endobronchial ultrasound-guided transbronchial lung biopsy in periphery pulmonary lesions. J Med Assoc Thai, 2010; 93,Suppl 1:S94–S101.