Ultrathin Bronchoscopy for the Diagnosis of Peripheral Pulmonary Lesions: A Meta-Analysis

Abstract

Background

Ultrathin bronchoscopy (external diameter, ≤3.5 mm) is useful for the diagnosis of peripheral pulmonary lesions because of its good accessibility.

Objectives

We performed a meta-analysis to investigate the diagnostic yield of ultrathin bronchoscopy for peripheral pulmonary lesions.

Methods

We performed a systematic search of MEDLINE and EMBASE (from inception to May 2021), and meta-analysis was performed using R software. The diagnostic yield was evaluated by dividing the number of successful diagnoses by the total number of lesions, and subgroup analysis was performed to identify related factors.

Results

Nineteen studies with a total of 1,977 peripheral pulmonary lesions were included. The pooled diagnostic yield of ultrathin bronchoscopy was 0.65 (95% confidence interval, 0.60–0.70). Significant heterogeneity was observed among studies (χ², 87.75; p < 0.01; I², 79.5%). In a subgroup analysis, ultrathin bronchoscopy with 1.2 mm channel size showed a diagnostic yield of 0.61 (95% confidence interval, 0.53–0.68), whereas ultrathin bronchoscopy with 1.7 mm channel size showed 0.70 (95% confidence interval, 0.66–0.74) (χ², 5.35; p = 0.02). In addition, there was a significant difference in diagnostic yield based on lesion size, histologic diagnosis (malignant vs. benign), bronchus sign, and lesion location from the hilum, whereas no significant difference was found based on lobar location. The overall complication rate of ultrathin bronchoscopy was 2.7% (pneumothorax, 1.1%).

Conclusions

Ultrathin bronchoscopy is an excellent tool for peripheral pulmonary lesion diagnosis with a low complication rate. The diagnostic yield of ultrathin bronchoscopy was significantly higher with larger channel size, which might be attributed to the availability of radial endobronchial ultrasound.

Introduction

Lung cancer is reportedly one of the most common cancers worldwide (2.09 million cases, 11.6% of all cancers) and is the main cause of cancer-related mortality (1.76 million deaths, 18.4%) [1]. A recent meta-analysis by Huang et al. [2] evaluated the effect of low-dose computed tomography on lung cancer screening compared with no screening or chest X-ray, and low-dose computed tomography proved to be a sensitive tool for the detection of stage I lung cancer (risk ratio [RR], 2.08). As an increase in the number of peripheral pulmonary lesions (PPLs) is anticipated, the importance of their accurate pathological diagnosis increases.

Transbronchial biopsy (TBB) and transthoracic needle biopsy (TNB) are generally recommended as non-surgical approaches for diagnosing PPLs [3]. Conventional TBB with a standard bronchoscope (external diameter [ED], 6.0 mm) and fluoroscopy showed a median sensitivity of 31% for diagnosis of malignant PPLs <20 mm, whereas TNB showed a sensitivity of ≥90% (70–82% for PPLs ≤15 mm) [3]. However, in the US population-based studies, TNB was associated with a higher risk of complications such as haemorrhage (1.0%), pneumothorax (15%), and pneumothorax requiring chest tube insertion (6.6%) [4], whereas TBB was associated with a lower risk of haemorrhage (0.58%), pneumothorax (0.97%), and pneumothorax requiring chest tube insertion (0.55%) [5].

To increase the diagnostic yield of TBB, new adjunct modalities for bronchoscopy such as radial endobronchial ultrasonography (r-EBUS) [6], guide sheath (GS) [7], computed tomography (CT)-guided [8], virtual bronchoscopic navigation (VBN) [9], and electromagnetic navigation bronchoscopy (ENB) [10] have been introduced. In a recent meta-analysis by McGuire et al. [11], TBB with either r-EBUS or ENB showed overall sensitivity of 70.7% for detecting PPLs. Although safe, TBB still needs further improvement in sensitivity compared with TNB.

Recently, the ultrathin bronchoscope (UTB) (ED, ≤3.5 mm) has been developed for diagnosis of PPLs [12], which is smaller in diameter than the thin bronchoscope (ED, 4.0 or 4.2 mm) commonly used for small PPLs in previous studies [11]. Therefore, UTB showed greater accessibility to peripheral bronchi with mobility compared with thin bronchoscope (median 5th vs. median 4th bronchial generation; p < 0.001) [13]. Although UTB seems promising for PPLs, systematic research on this modality is currently lacking. Thus, we performed a meta-analysis on the diagnostic yield of UTB for PPLs. In addition, we also evaluated the factors affecting its performance and associated complications.

Material and Methods

Literature Search

We performed a meta-analysis in accordance with the Preferred Reporting Items for Systematic Reviews and Meta Analyses (PRISMA) guidelines [14]. We searched MEDLINE and EMBASE (from their inception until May 2021) to identify all studies that employed UTB for evaluation of PPLs using a predetermined protocol (Table 1). A manual search of references cited in original and review papers was also done for relevant studies, which might have been missed by electronic search.

Table 1.

Search strategy for meta-analysis

Search strategy
1	“Bronchoscope and lung nodule(s)”
2	“Bronchoscope and lung lesion(s)”
3	“Bronchoscope and pulmonary nodule(s)”
4	“Bronchoscope and pulmonary lesion(s)”
5	“Bronchoscope and peripheral nodule(s)”
6	“Bronchoscope and peripheral lesion(s)”
7	“Bronchoscopic and lung nodule(s)”
8	“Bronchoscopic and lung lesion(s)”
9	“Bronchoscopic and pulmonary nodule(s)”
10	“Bronchoscopic and pulmonary lesion(s)”
11	“Bronchoscopic and peripheral nodule(s)”
12	“Bronchoscopic and peripheral lesion(s)”
13	“Bronchoscopy and lung nodule(s)”
14	“Bronchoscopy and lung lesion(s)”
15	“Bronchoscopy and pulmonary nodule(s)”
16	“Bronchoscopy and pulmonary lesion(s)”
17	“Bronchoscopy and peripheral nodule(s)”
18	“Bronchoscopy and peripheral lesion(s)”

Selection of Studies

All articles identified by the search strategy were independently assessed by two authors (S.H.K. and J.S.E.). Discordance was resolved by consensus. Abstracts were initially examined, and studies were selected for inclusion only after both reviewers assessed the full-text articles. Criteria for inclusion were: (1) UTB used for diagnosis of PPL and diagnostic yield was provided, (2) diagnosis confirmed histologically or by close clinical follow-up, and (3) studies enrolled at least 20 patients. We excluded review articles, letters, papers not available in English, or case reports with fewer than 20 patients. When two or more studies were published by the same author(s), the methods sections were reviewed to check if study periods were overlapping. If so, we included only one publication with the greatest number of patients to prevent duplication of the study cohorts.

Data Extraction

All data were independently extracted by S.H.K. and J.S.E and compared. Disagreements were resolved by further discussion with third investigator (K.P.). The following were retrieved: author, year of publication, study design (randomized controlled trial, prospective, retrospective, or unknown), total number of lesions, number of successful diagnoses, UTB channel size, use of guidance modalities, lesion size, prevalence of malignancy, histologic diagnosis, presence of bronchus sign, distance from the hilum, lobar location of the lesion, complications, and reference standard.

Quality of selected studies was evaluated using the Quality Assessment of Diagnostic Accuracy Studies-2 tool [15]. This validated tool contains 14 signalling questions to evaluate four main components (patient selection, index test, reference standard, and flow and timing) in two categories (risk of bias and applicability concerns).

Statistical Analysis

Meta-analysis was performed using meta package of R statistical software (version 4.0.5, http://www.R-project.org). A p value of <0.05 was considered statistically significant. The primary outcome was the diagnostic yield with 95% confidence interval (CI), which was calculated by dividing the number of successful diagnoses by the total number of lesions. Inverse variance weighting across selected studies was applied to evaluate the pooled diagnostic yield, where the weight of each study was based on the number of lesions.

Subgroup analysis was performed to identify the factors associated with diagnostic yield. Stratified analysis on diagnostic yield was based on channel size (1.2 mm vs. 1.7 mm), lesion size (≤20 mm vs. >20 mm), histologic diagnosis (malignant vs. benign), bronchus sign (present vs. absent), distance from the hilum (peripheral vs. non-peripheral [central or intermediate]), and location of lesion (upper lobe vs. other lobes). For each study, the diagnostic yield was determined for a single-channel size (1.2 mm or 1.7 mm). The other variables, on the other hand, were dichotomous, and the diagnostic yield was determined for each of their two possible values. Dichotomous variables were analysed using Mantel Haenszel RR, which was calculated by dividing the diagnostic yield of the former variable by that of the latter variable. RR >1 was in favour of the former variable for the diagnostic yield, while RR <1 of the latter variable.

Study heterogeneity was assessed using the Cochran Q test (χ² test) and quantified by the I² index [16]. Statistical heterogeneity was indicated in cases of p < 0.01 in χ² test [16], and I² index value of >50% was considered as significant heterogeneity [17]. Random-effect models with the inverse variance method were applied to reflect the variability of effect sizes among included studies with diversity in adjunctive modalities [18]. Publication bias was evaluated using funnel plot asymmetry [19] based on both the Egger's and Begg tests [20, 21]. We used funnel plots of standard error or diagnostic yield (logit transformed).

Results

Literature Search and Study Selection

The search algorithm revealed 1,071 potentially relevant papers after removal of duplicates (Fig. 1). Following abstract review, 32 articles were selected for full-text review. Of these, 15 articles were excluded according to the protocol, and two studies missed in the database search were added [22, 23]. Therefore, 19 studies formed the basis of our systematic review [13, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39].

Fig. 1.

Flow diagram of search and study selection.

Study Description

A total of 1,977 PPLs were included. Table 2 lists the study characteristics and summarizes their features [13, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39]. Six studies were randomized controlled trials, six were prospective, and seven were retrospective studies. UTB with 1.2 mm channel size was used in 11 studies; the others employed UTB with 1.7 mm channel size. The prevalence of malignancy was reported in 14 studies (median, 72%; interquartile range, 62–82). Among them, eight studies showed prevalence of malignancy ≤75%, whereas 6 studies showed >75%. There was a variation in additional guidance devices used among included studies, such as fluoroscopy (14 studies), VBN (11 studies), r-EBUS (eight studies), CT-fluoroscopy (six studies), and cone-beam CT (two studies).

Table 2.

Study characteristics

No.	Author	Year	Study design	Channel size, mm	Additional guidance	Lesion size, mm	Prevalence of malignancy, %	Complications	Reference/comparison test
1	Shinagawa et al. [24]	2004	Prospective	1.2	CT-Flu, VBN	13.2	69.2	No	Histology by alternative means or clinical/radiologic surveillance
2	Yamamoto et al. [25]	2004	Prospective	1.2	Flu	NE	76.1	NE	Histology by alternative means or clinical/radiologic surveillance
3	Shinagawa et al. [26]	2007	Prospective	1.2	CT-Flu, VBN	13.7	71.8	1 Bleed, 1 PNX	NE
4	Tachihara et al. [27]	2007	Retrospective	1.2	Flu, VBN	15.9	NE	No	Histology by alternative means or clinical/radiologic surveillance
5	Oki et al. [28]	2009	Prospective	1.7	Flu, R-EBUS	31.2	62.0	No	Histology by alternative means or clinical/radiologic surveillance
6	Matsuno et al. [22]	2011	Retrospective	1.2	CT-Flu, Flu, VBN	NE	60.8	1 PNX	Histology by alternative means or clinical/radiologic surveillance
7	Oki et al. [29]	2012	RCT	1.7	Flu, R-EBUS	30.6	75.2	1 Bleed, 3 PNX, 1 PNA	Histology by alternative means or clinical/radiologic surveillance
8	Asano et al. [30]	2013	RCT	1.2	Flu, VBN	17.3	88.0	2 Bleed, 2 PNX 1 PNA 2 Others (1 transient bradycardia, 1 xylocaine intoxication)	Histology by alternative means or clinical/radiologic surveillance
9	Oki et al. [31]	2015	RCT	1.7	Flu, R-EBUS, VBN	19.0	82.0	3 PNX, 1 PNA 1 Others (1 chest pain)	Histology by alternative means or clinical/radiologic surveillance
10	Franzen et al. [32]	2016	RCT	1.2	Flu	27.0	40.0	6 Others (2 coughing, 3 blocked working channel or weak suction, 1 hypertensive urgency requiring intervention)	Histology by alternative means or clinical/radiologic surveillance
11	Tokoro et al. [33]	2016	Retrospective	1.2	CT-Flu	NE	NE	NE	Histology by alternative means or clinical/radiologic surveillance
12	Ali et al. [34]	2019	Prospective	1.2	CBCT, VBN	20.0	62.5	1 PNX (requiring chest tube), 1 PNA	Histology by alternative means or clinical/radiologic surveillance
13	Diez-Ferrer et al. [35]	2019	Prospective	1.2	Flu, VBN	24.3	72.1	NE	Histology by alternative means or clinical/radiologic surveillance
14	Oki et al. [13]	2019	RCT	1.7	Flu, R-EBUS, VBN	18.9	81.4	1 Bleed, 2 PNX (1 requiring chest tube), 2 PNA	Histology by alternative means or clinical/radiologic surveillance
15	Sehgal et al. [36]	2019	Retrospective	1.7	R-EBUS	16.4	NE	1 Bleed, 1 PNX	Clinical/radiologic surveillance
16	Sumi et al. [37]	2020	Retrospective	1.7	Flu, R-EBUS	19.5	NE	1 Bleed, 1 PNX (requiring chest tube), 2 PNA	NE
17	Kawakita et al. [38]	2021	Retrospective	1.2	CBCT, CT-Flu, Flu, VBN	19.9	68.6	3 PNX, 1 others (respiratory failure)	Histology by alternative means or clinical/radiologic surveillance
18	Sumi et al. [39]	2021	Retrospective	1.7	Flu, R-EBUS	20.0	NE	2 Bleed, 1 PNA	NE
19	Zheng et al. [23]	2021	RCT	1.7	Flu, R-EBUS, VBN	27.7	83.3	No	Histology by alternative means or clinical/radiologic surveillance

CBCT, cone-beam computed tomography; CT-Flu, computed tomography-fluoroscopy; Flu, fluoroscopy; NE, not evaluable; No., number; PNA, pneumonia; PNX, pneumothorax; R-EBUS, radial-endobronchial ultrasound; RCT, randomized controlled trial; VBN, virtual bronchoscopic navigation.

Online supplementary Table S1 (for all online suppl. material, see www.karger.com/doi/10.1159/000527362) provides a quality assessment of all included studies based on Quality Assessment of Diagnostic Accuracy Studies-2. The overall analysis showed good performance in the patient selection and index test criteria. However, it showed poor performance in the reference standard in addition to flow and timing criteria, which indicates potential for significant bias. Funnel plot (Fig. 2) was not asymmetric, with both Egger's (p = 0.585) and Begg tests (p = 0.944) showing insignificant p values, indicating the absence of publication bias.

Fig. 2.

Funnel plot of publication bias.

Test Performance: Meta-analysis

The inverse variance weighted overall diagnostic yield was 0.65 (95% CI, 0.60–0.70) (Fig. 3) [13, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39]. The diagnostic yield among studies ranged from 0.38 to 0.90. χ² value of 87.75 (p < 0.01) and I² index of 79.5% (95% CI, 68.7–86.6) indicated substantial heterogeneity across studies.

Fig. 3.

Overall diagnostic yield of ultrathin bronchoscope. CI, confidence interval; IV, inverse variance.

The factors underlying diagnostic yield were further evaluated by subgroup meta-analysis. As for independent variables, the weighted diagnostic yield of UTB with 1.2 mm channel size (1,148 lesions) was 0.61 (95% CI, 0.53–0.68) [22, 24, 25, 26, 27, 30, 32, 33, 34, 35, 38], whereas that with 1.7 mm channel size (829 lesions) was 0.70 (95% CI, 0.66–0.74) (Fig. 4) [13, 23, 28, 29, 31, 36, 37, 39]. The difference of pooled diagnostic yield between these two subgroups was statistically significant (χ² value, 5.35; p = 0.02).

Fig. 4.

Diagnostic yield of ultrathin bronchoscope based on channel size (1.2 mm vs. 1.7 mm). CI, confidence interval; IV, inverse variance.

Regarding dichotomous variables (Table 3), the pooled diagnostic yield was significantly different based on lesion size in five studies (≤20 mm vs. >20 mm; RR, 0.63; 95% CI, 0.49–0.80) (Fig. 5a) [13, 27, 28, 29, 31], histologic diagnosis in 12 studies (malignancy vs. benign; RR, 1.53; 95% CI, 1.26–1.84) (Fig. 5b) [13, 22, 23, 24, 28, 29, 30, 31, 32, 34, 35, 38], presence of bronchus sign in seven studies (present vs. absent; RR, 1.44; 95% CI, 1.21–1.72) (Fig. 5c) [13, 23, 30, 31, 34, 35, 38], and distance from the hilum in eight studies (peripheral vs. non-peripheral; RR, 0.84; 95% CI, 0.73–0.98) (Fig. 5d) [13, 23, 27, 30, 31, 35, 37, 39]. However, there was no significant difference in the pooled diagnostic yield based on lobar location of lesion in six studies (upper vs. other lobes; RR, 0.99; 95% CI, 0.91–1.09) (Fig. 5e) [13, 23, 28, 31, 37, 39].

Table 3.

Stratified meta-analysis for dichotomous variables

Variable	Studies, n	Observations, n	Events, n	Pooled sensitivity (95% CI)	Risk ratio (95% CI)	Heterogeneity			Overall effect
						x2	I2 (%)	p value	z	p value
Lesion size
≤20 mm	5	267	152	0.53 (0.41–0.64)	0.63 (0.49–0.80)	10.86	63	0.0281	−3.80	0.0001
>20 mm	5	290	233	0.80 (0.74–0.84)
Histologic diagnosis
Malignancy	12	1,163	839	0.74 (0.68–0.80)	1.53 (1.26–1.84)	26.81	59	0.0049	4.38	<0.0001
Benign	12	348	160	0.47 (0.38–0.56)
Bronchus sign
Present	7	890	625	0.72 (0.63–0.80)	1.44 (1.21–1.72)	10.37	42	0.1099	4.01	<0.0001
Absent	7	268	118	0.47 (0.36–0.59)
Distance from hilum
Peripheral	8	798	500	0.63 (0.54–0.71)	0.84 (0.73–0.98)	20.09	65	0.0054	−2.29	0.0220
Central or intermediate	8	382	268	0.71 (0.59–0.81)
Location of lesion
Upper lobe	6	348	251	0.72 (0.67–0.77)	0.99 (0.91–1.09)	3.08	0	0.6873	−0.11	0.9087
Other lobe	6	346	251	0.72 (0.67–0.77)

CI, confidence interval; n, number.

Fig. 5.

Diagnostic yield of ultrathin bronchoscope based on a lesion size (≤20 mm vs. > 20 mm), b histologic diagnosis (malignant vs. benign), c bronchus sign (present vs. absent), d distance from the hilum (peripheral vs. non-peripheral), and e location of lesion (upper lobe vs. other lobes). CI, confidence interval; MH, Mantel Haenszel.

Complication Rates

Complication rates were 2.7% (46 out of 1,708 lesions) across 16 studies [13, 22, 23, 24, 26, 27, 28, 29, 30, 31, 32, 34, 36, 37, 38, 39], whereas three studies did not report complication rates (Table 2) [25, 33, 35]. Among 16 studies, no complication was noted in four studies. The most common complication was pneumothorax with a pooled rate of 1.1% (18 out of 1,708) (range, 0.6–3.0%), whereas that of chest tube insertion was 0.2% (three out of 1,708). The pooled rate of bleeding was 0.5% (nine out of 1,708) (range, 0.6–2.9%). The pooled rate of pneumonia was 0.5% (nine out of 1,708) (range, 0.3–2.5%). No deaths were reported in any study.

Discussion

To our knowledge, this is the first systematic meta-analysis investigating the diagnostic yield of UTB for diagnosing PPLs. Subgroup analysis showed significantly higher diagnostic yield of UTB with 1.7 mm channel size compared to 1.2 mm channel size. In addition, the funnel plot of included studies was symmetrical, implying a low risk of publication bias. Lastly, our study reported an excellent safety profile of UTB, with 2.7% overall complication rate.

The inverse variance-weighted pooled diagnostic yield was 0.65, which was slightly inferior to previous studies on other bronchoscopes. In a meta-analysis by Steinfort et al. [40], the pooled diagnostic yield of standard or thin bronchoscope was 0.73 (95% CI, 0.70–0.76) when combined with r-EBUS. This difference might be attributed to the median lesion size of the included studies, which is a major determinant of the diagnostic yield [41]. Steinfort et al. [40] included seven studies (580 lesions) with lesion size <25 mm and 6 studies (510 lesions) ≥25 mm, whereas our study included 12 studies (1,403 lesions) <25 mm and four studies (312 lesions) ≥25 mm. Furthermore, there is a possibility of selection bias in our study, since PPLs not observable on fluoroscopy, undiagnosed by previous conventional bronchoscopy under fluoroscopy, or difficult to access were included in some studies [22, 25, 33]. In addition, the smaller working channel of UTB prohibits the suction capability compared to thin bronchoscopy [12, 29, 31]. However, due to a small ED, UTB can pass through small airways with less resistance, thereby increasing performance and ease for operators. Furthermore, UTB provides good accessibility with visibility for PPL [13] and can play a role as a thick GS [23]. Lastly, UTB combined with fluoroscopy, r-EBUS, and VBN showed a higher diagnostic yield of 0.70 (95% CI, 0.63–0.77) than that of thin bronchoscopy (0.59; 95% CI, 0.51–0.66; p = 0.027) for PPL less than ≤30 mm [13].

In our study, UTB with 1.7 mm channel size showed a significantly higher diagnostic yield compared to that with 1.2 mm channel size. Conventional UTB mostly consisted of 1.2 mm channel size, which allowed the passage of mini-forceps and cytology brush <1.2 mm in diameter [42]. This small channel size limits adequate specimen collection, suction capability, and the use of r-EBUS [9, 28, 42], which might have led to the lower diagnostic yield observed with 1.2 mm channel size compared to that of thin bronchoscope. The recent development of 1.7 mm channel size allows the passage of r-EBUS [28], which provides a 360° view of the surrounding lesion by the ultrasound transducer [43] and shows a pooled diagnostic yield of 0.73 (95% CI, 0.70–0.76) [40]. The r-EBUS probe position relative to the lesion was known to be associated with the diagnostic yield of TBB, where the diagnostic yield of within, adjacent to, and outside the lesion was 0.83, 0.61, and 0.04, respectively (p < 0.001) [44]. In line with this, Nishi et al. [45] reported that UTB showed improved r-EBUS probe position for PPL after substituting thin bronchoscope (adjacent, 12; not visible, 32) by UTB (within, 26; adjacent, 11; not visible, 7) combined with fluoroscopy and VBN. This implies better applicability of r-EBUS in UTB compared with thin bronchoscope, which might be attributed to UTB's greater accessibility. Lastly, UTB with 1.7 mm channel size allows the passage of 1.5 mm biopsy forceps for adequate specimens and enhances the suction capability, which improves visualization around obstacles such as secretions and bleeding during the procedure [31]. These findings support the use of UTB with 1.7 mm over 1.2 mm channel size, as well as thin bronchoscope, emphasizing the importance of r-EBUS for the diagnosis of PPL.

Our subgroup analysis showed that UTB provided a significantly lower diagnostic yield of lesions located within the outer third of the lung field. On the contrary, Oki et al. [13] reported no significant difference in the diagnostic yield of UTB with multimodalities between the lesion in intermediate and peripheral location from the hilum (0.73 vs. 0.69; p = 0.701). In the same study, UTB showed significantly higher diagnostic yield than that of thin bronchoscope in the peripheral (0.69 vs. 0.56; p = 0.019) but not in the intermediate location (0.73 vs. 0.70; p > 0.999) [13]. This might be attributed to the good accessibility, bronchial selectivity, and ability of the thick GS of UTB in peripheral bronchi [12, 23, 30]. The discrepancy between these findings with our study might be attributed to inclusion of studies using 1.2 mm channel size (three out of eight) and the variability of adjunct modalities among included studies. Thus, UTB might be a good option during TBB for peripheral as well as central or intermediate lesions.

Furthermore, we found no difference in the diagnostic yield of UTB based on the upper lobe position of PPL, contrary to previous study findings on other bronchoscope [46]. When combined with fluoroscopy, GS, and r-EBUS, Kurimoto et al. [46] reported the lowest diagnostic yield of standard bronchoscope in the apico-posterior segment of left upper lobe compared with all other locations (0.40 vs. 0.76, p = 0.003). This might be attributed to the limitation of previous bronchoscope insertion in the upper lobes due to their acute angle, especially in the apical segment. On the contrary, UTB permits relatively easy access to upper lobes because of its wide angulation range (210° up and 130° down in BF-MP190F; Olympus Medical Systems, Tokyo, Japan). Thus, UTB might be the best option for TBB for lesions in the upper lobes. In addition, lesions in the superior segment of lower lobes with their acute angles might also be a good candidate for UTB, which warrants further investigation.

However, our study has some limitations. First, there was substantial heterogeneity in diagnostic yield among the included studies, as their quality was not uniform. To overcome this issue, we performed subgroup analysis to explore the factors regarding heterogeneity. However, we could not evaluate the diagnostic yield according to each guidance system due to different combinations of guidance use and study design among included studies, which might limit interpretation of our study results. Second, information on the other predictive factors for successful diagnosis was lacking due to variability among included study designs, such as r-EBUS findings, experience of operators, radiologic appearance on CT, distance from the pleura, and so on. Third, the reference standard of diagnostic yield was inconsistent among included studies, perhaps leading to different interpretations of successful diagnosis, which might have impacted the overall diagnostic yield of UTB. Fourth, the number of studies in subgroup analysis on PPL size was relatively small. This is attributed to the different method of measurement on lesion size (longest or mean) and lack of diagnostic yield based on lesion size (20 mm) among included studies. Fifth, our study excluded systematic reviews and conference abstracts, which are likely to show a low study quality. However, this may predispose to publication bias. Finally, our study was not registered prior to the review, but we followed the suggested PRISMA protocol.

Conclusion

In conclusion, our study showed that UTB is an excellent tool for PPL diagnosis with a low complication rate. The diagnostic yield of UTB with 1.7 mm channel size was significantly higher compared to that with 1.2 mm, which might be attributed to availability of r-EBUS in 1.7 mm channel size.

Statement of Ethics

An ethics statement is not applicable because this study is based exclusively on the published literature.

Conflict of Interest Statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding Sources

This work was supported by a clinical research grant from Pusan National University Hospital in 2022, and also supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (Ministry of Science and ICT) (NRF-2022R1F1A1074117).

Author Contribution

Soo Han Kim contributed to conceptualization, methodology, study search, study selection, data acquisition, data curation, formal analysis, validation, and writing (original draft). Jinmi Kim contributed to formal analysis, validation, and writing (review and editing). Kyoungjune Pak contributed to conceptualization, methodology, data curation, validation, and writing (review and editing). Jung Seop Eom contributed to conceptualization, methodology, study search, study selection, data acquisition, data curation, writing (review and editing), and supervised the project. All authors read and approved the final manuscript.

Data Availability Statement

All data generated or analysed during this study are included in this article and its online supplementary material. Further enquiries can be directed to the corresponding author.

Supplementary Material

Supplementary data

Supplementary data

Funding Statement

This work was supported by a clinical research grant from Pusan National University Hospital in 2022, and also supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (Ministry of Science and ICT) (NRF-2022R1F1A1074117).