Bronchus sign on thin-section computed tomography is a powerful predictive factor for successful transbronchial biopsy using endobronchial ultrasound with a guide sheath for small peripheral lung lesions: a retrospective observational study

Patients

The medical records of consecutive patients who underwent bronchoscopy in Fujita Health
University Hospital from April 2012 to March 2013 were retrospectively reviewed. The
institutional review committee (Fujita Health University Institutional Review Board)
approved this study protocol, which was conducted in accordance with the tenets of
the Declaration of Helsinki (approval number Fujita 14-152). All patients provided
written informed consent before undergoing bronchoscopy.

Evaluation before bronchoscopy

All patients underwent chest TSCT (0.5 mm slice) within 1 month before bronchoscopy.
TSCT images were generated using a nonenhanced multidetector CT system (Aquilion One
Vision Edition; Toshiba Medical Systems, Tokyo, Japan). The CT scan parameters were
as follows: tube current, automatic exposure control SD7; tube voltage, 120 kV; rotation
speed, 0.5 s/rot; table speed, BP 1.39; reconstruction filter, FC51 AIDR 3D WEAK;
window width, 1600; and window level, ?600. We used automatic exposure control, the
volume helical scan mode, and multi-planar reconstruction images.

We did not use a virtual bronchoscopic navigation system. Each bronchoscopist studied
the patient’s chest X-ray and TSCT images before the procedures and identified the
size and location of the lesions and the respective bronchus. In every case, we confirmed
these findings and identified the target bronchus to approach by group discussion.
We investigated the bronchus sign on CT, which was identified as the presence of a
bronchus directly leading to the target lesion (CT bronchus sign) 10]. In each case, we categorized the relationship between the target lesion and the
nearest bronchus into three types of CT bronchus signs (A to C). In type A, the responsible
bronchus clearly reached the inside of the target lesion. In type C, no bronchus could
be detected in relation to the lesion. When the CT findings could be categorized into
neither type A nor C, the CT bronchus sign was categorized as type B (Fig. 1). Two bronchoscopists (T.M. and K.I.) independently assessed each TSCT scan and determined
the type of each CT bronchus sign (A, B, or C). When the two bronchoscopists recommended
different types, the final type was determined by discussion.

Fig. 1. CT bronchus sign. According to TSCT (0.5-mm reconstruction), we categorized the CT
bronchus sign in each case into three types a to c according to the relationship between the nearest bronchus and the target lesion.
In type A, the responsible bronchus clearly reached the inside of the target lesion. In type
C, no bronchus could be detected in relation to the lesion. When the CT findings could
be categorized into neither type A nor C, the CT bronchus sign was categorized as type B

Bronchoscopic procedures

After administration of local pharyngeal anesthesia, all patients were lightly sedated
with individually calculated dose of intravenous midazolam as reported elsewhere 11]. Transbronchial biopsy (TBB) using EBUS with a guide sheath (EBUS-GS) was performed
according to the standard Kurimoto method 12]. A 20-MHz radial type ultrasound probe with an external diameter of 1.4 mm (UM-S20-17S;
Olympus Medical Systems, Tokyo, Japan) connected to an endoscopic ultrasonography
system (EU-M30S; Olympus Medical Systems) was used in all cases. We mainly used a
bronchoscope with a 2.0-mm-diameter working channel (BF-P260F; Olympus Medical Systems)
for a guide sheath with an external diameter of 1.9 mm (K-201 kit equipped with a
biopsy forceps and cytological brush; Olympus Medical Systems). According to the size
or location of the lesion, the operator could use a 1T260 bronchoscope with a 2.6-mm-diameter
working channel with the guide sheath kit K203 (external diameter of the guide sheath:
2.55 mm) (Olympus Medical Systems). We took 12 biopsy samples when using the K201
kit and nine samples using the K203 kit. Nine bronchoscopists performed the procedures
in this study. Five of them had less than 7 years of experience as a bronchoscopist
(range: 2–6 years), whereas four had more experience (range: 7–21 years).

Diagnostic evaluation of TBB results (positive diagnostic criteria by bronchoscopy)

In this study, we evaluated the bronchoscopic diagnostic yield in both malignant and
non-malignant diseases (including inflammatory disease). Thus, we determined the evaluation
criteria for judgment of successful (diagnostic) bronchoscopy according to previous
reports 8], 13]. Lesions with evident malignant findings on histological examination or class IV/V
findings on cytological examination were defined as malignant. For benign lesions,
when bronchoscopy demonstrated a distinct histologic pattern (such as epithelioid
granuloma or intra-alveolar organization) or the presence of bacteria accompanied
by reasonable radiologic and clinical findings, we determined that the bronchoscopy
was successful. All benign lesions were followed for more than 12 months to confirm
that the lesions had remained stable or improved by appropriate therapy.

Statistical methods

Statistical analyses were carried out using JMP software (ver. 8.0; SAS Institute,
Inc., Cary, NC). A binomial proportion confidence interval (CI) was calculated for
diagnostic yield. Differences in proportions were evaluated with the ?²
test. Spearman rank correlation was used to identify the association between two ordinal
variables. Multivariate logistic regression analysis was used to identify factors
associated with diagnostic yield independent of other variables. We entered all variables
that were significantly (p??0.05) associated with diagnostic yield in the univariate analyses into the multivariate
analysis. The final model was determined by a backward variable selection method,
and odds ratios and the 95 % CIs are presented. A p-value of 0.05 was considered to indicate statistical significance. All analyses
were two-sided.