Abstract
Introduction: Endobronchial ultrasound-guided transbronchial mediastinal cryobiopsy (EBUS-TMC), a novel technique, has been reported to improve the diagnostic value of endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) for mediastinal lesions in recent studies. Current literature suggests that this procedure has greater diagnostic efficacy compared to conventional EBUS-TBNA. This systematic review and meta-analysis aimed to evaluate the diagnostic yield and complications associated with EBUS-TMC in comparison to EBUS-TBNA, thereby exploring the potential of this novel technique in enhancing the diagnostic utility for mediastinal lesions. Methods: A comprehensive literature review was conducted by searching the PubMed, Embase, and Google Scholar databases for articles published from inception to December 31, 2023. The objective of this review was to evaluate the utilization of EBUS-TMC in diagnosing mediastinal disease, while also assessing the quality of each study using the QUADAS-2 tool. The diagnostic yield estimates were subjected to a meta-analysis utilizing inverse variance weighting. Furthermore, a comprehensive analysis of the complications associated with this procedure was performed. Results: The meta-analysis included 10 studies involving a total of 538 patients. The findings of the meta-analysis demonstrated that EBUS-TMC yielded an overall diagnostic rate of 89.59% (482/538), while EBUS-TBNA yielded a rate of 77.13% (415/538). The calculated inverse variance-weighted odds ratio was 2.63 (95% confidence interval, 1.86–3.72; p < 0.0001), and I2 value was 11%, indicating a statistically significant difference between the two techniques. The associated complications consisted of pneumothorax, pneumomediastinum, mediastinitis, and bleeding, with an incidence of 0.74% (4/538), 0.37% (2/538), 0.0% (0/538), and 1.12% (6/538), respectively. Moreover, the funnel plot displayed no discernible publication bias. Further subgroup analysis revealed a notable improvement in the diagnosis value for lymphoma (86.36% vs. 27.27%, p = 0.0006) and benign disorder (87.62% vs. 60.00%, p < 0.0001). Conclusion: This review of the current available studies indicated that EBUS-TMC enhanced overall diagnostic yields compared to EBUS-TBNA, particularly for diagnosing benign disease and lymphoma. This procedure was not associated with any serious complications.
Introduction
Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) has been widely acknowledged as a highly effective and minimally invasive procedure for sampling mediastinal lesions [1‒3]. EBUS-TBNA has superseded mediastinoscopy as the preferred diagnostic approach for mediastinal diseases owing to its higher safety, noninvasive characteristics, and remarkable diagnostic accuracy [4, 5]. Several studies have demonstrated that EBUS-TBNA is a valuable tool for diagnosing primary lung cancer [1‒5]. However, this technique primarily yields cytopathology specimens rather than pathological tissues. Consequently, a definitive diagnosis is harder to achieve for uncommon tumors or benign disorders, which necessitate a thorough assessment of the overall architectural features of the lesion tissue [6, 7]. In such instances, the absence of appropriate pathological tissue evaluation may introduce diagnostic ambiguity.
Endobronchial ultrasound-guided transbronchial mediastinal cryobiopsy (EBUS-TMC), an innovative technique recently developed, is employed for harvesting pathological samples from the mediastinum. This procedure involves the insertion of a cryoprobe into the mediastinum, guided by real-time ultrasound, either through the needle aspiration site or via an electrocautery incision. This method facilitates the acquisition of tissue from mediastinal lesions through cryobiopsy. Distinct from EBUS-TBNA, EBUS-TMC offers the advantage of obtaining a larger volume of mediastinal specimens while substantially reducing avulsion artifacts. Consequently, EBUS-TMC is increasingly being recognized as a potential alternative to EBUS-TBNA for diagnosing diseases of the mediastinum [8, 9]. Over the past few years, a series of related studies have reported that EBUS-TMC for mediastinal disease sampling provides adequate high-quality samples for histological examination [10‒17]. The majority of studies supported the expeditious, minimally invasive, and secure characteristics of EBUS-TMC, which may greatly enhance the overall diagnostic efficacy of mediastinal lesions [18]. However, the findings of relevant studies exhibit inconsistency, particularly regarding the diagnosis of primary lung cancer and benign disorders, giving rise to controversy [9, 13, 14, 18].
A recent systematic review conducted a comprehensive analysis of studies on mediastinal cryobiopsy across multiple databases. However, it solely presented a qualitative assessment of the pertinent findings and failed to undertake a meta-analysis of the positive diagnosis rates and complications associated with mediastinal cryobiopsy in mediastinal diseases [19]. Unlike the aforementioned systematic review, the present study performed a systematic review and meta-analysis of the literature on EBUS-TMC. The overall diagnosis rate for EBUS-TMC in mediastinal diseases was determined by pooling the relevant data; furthermore, a subgroup analysis was conducted to calculate the diagnosis rate in pulmonary cancer, lymphoma, and benign diseases. Additionally, a comprehensive summary of the safety and complications associated with EBUS-TMC was carried out.
Methods
The meta-analysis was registered with PROSPERO (CRD42023464831) (https://www.crd.york.ac.uk/prospero/). In this study, a meta-analysis of previously published data was conducted, adhering to the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA-DTA) [20]. Consequently, Ethical Committee approval was not deemed necessary for this study.
PICO Question
The Population, Intervention, Comparison, and Outcome format was employed to establish the criteria for identifying pertinent studies. The study encompassed a population of adult individuals undergoing ultrasound bronchoscopy, with interventions including EBUS-TMC and EBUS-TBNA. The comparison group was composed of patients undergoing EBUS-TBNA, and the outcome measures focused on diagnosis rates and complications.
Search Strategy
A comprehensive literature review was conducted by searching the PubMed, Embase, and Google Scholar databases for articles published from inception to December 31, 2023. The following search terms “endobronchial ultrasound” OR “endoscopic ultrasound” OR “EBUS” AND “mediastinal cryobiopsy” OR “mediastinal deep freeze” OR “lymph node cryobiopsy” OR “cryo-nodal biopsy” were used to identify studies relevant to mediastinal cryobiopsy. In addition, the reference lists from the retrieved articles were manually searched. Abstracts from related scientific conferences were also reviewed to ensure a comprehensive search.
Study Selection
For the present meta-analysis, the inclusion criteria were (1) retrospective, prospective case series and randomized clinical trials in which subjects underwent EBUS-TMC and EBUS-TBNA for mediastinal lesion sampling; (2) the studies provided the diagnostic rates from EBUS-TMC and EBUS-TBNA for the evaluation of mediastinal lesions, or alternatively, the diagnostic yield results could be derived from the data provided in the article; (3) the occurrence of complications was recorded to facilitate the calculation of the complication rate. The exclusion criteria were (1) case series containing fewer than 4 cases; (2) mediastinal cryobiopsy or needle aspiration was performed without EBUS guidance; (3) studies that contained overlapping data or review article; (4) studies that employed a non-standard method for patient inclusion.
Data Extraction and Quality Assessment
The primary objective of this study was to determine the diagnosis rates of EBUS-TMC in comparison to EBUS-TBNA. A secondary objective was to assess the diagnosis rates of EBUS-TMC specifically in cases of lung cancer, lymphoma, and benign diseases. Another secondary objective was to analyze the safety of EBUS-TMC.
The quality of the included studies was assessed by using the QUADAS-2 tool [21] and was performed by two authors (Z.Z. and S.L.) Each study was independently scored by the two authors, considering factors such as patient selection, reference standard, index test, as well as timing and flow. The risk of bias in each study was classified as low, high, or unclear. Discrepancies were first discussed by the two authors, and any disagreements were resolved by discussing with a third reviewer. Studies that exhibit a substantial risk of bias in one or more crucial domains are regarded as having a high risk of bias, whereas studies that demonstrate a low risk in all essential domains are deemed to have a low risk of bias. Other than this, studies were deemed to be at risk of unclear bias.
Statistical Analysis
The present meta-analysis was performed utilizing the Cochrane RevMan software (version 5.4) and the Stata software (version 18). The diagnostic yield in each study was summarized using inverse variance weighting, and the odds ratio was calculated. Moreover, heterogeneity was measured by the Cochran Q test and I2 index. A fixed-effects model was employed when statistical heterogeneity was low (I2<50%, p > 0.10), whereas a random-effects model was used otherwise. In addition, Egger’s test was performed to measure the asymmetry of the funnel plots, thereby assessing publication bias. A value of p < 0.05 was considered statistically significant.
Results
Results of the Literature Search
As shown in Figure 1, the initial search identified 1,180 citations, among which 1,151 were determined to be clearly irrelevant based on their titles. Subsequently, a thorough examination of the remaining 29 studies was conducted. Next, 19 studies were further excluded, and only 10 publications were considered eligible for inclusion in the meta-analysis. Ultimately, a total of 10 studies comprising 538 patients were included in the current analysis. The absence of obvious asymmetry in the funnel plot suggests no publication bias (Egger’s test, p = 0.49).
Study Characteristics
EBUS-TMC and EBUS-TBNA were performed on all patients in the 10 studies. As shown in Figure 2, the quality of the studies incorporated in this analysis was thoroughly evaluated. Within the selected studies, the risk of bias was low in two studies, unclear in four studies, and high in four studies. The clinical heterogeneity observed in these studies was attributed to variations in the number of needle aspiration passes, the size of cryoprobe employed, the number of cryobiopsies, the cooling time of cryoprobe, and the technique employed for cryobiopsy.
The QUADAS-2 methodology is used to assess bias risks for included studies.
Table 1 presents a comprehensive overview of the data derived from all the studies incorporated in this analysis. The studies incorporated in this analysis utilized a 19G, 21G, and 22G needle for the aspiration component of the procedure, and two to four passes were performed. In each study, cryoprobes from the same manufacturer (ERBE, Tubingen, Germany) were employed. Eight studies reported that the size of the cryoprobe used was 1.1 or 1.7 mm, while the remaining two articles did not mention the size of the cryoprobe. A cryoprobe cooling time of three to 7 s was allowed in the nine included studies, and one to four cryobiopsy passes were performed at each lymph node station. All cryoprobes were appropriately inserted through the previous puncture site or incision. Of all the studies in this analysis, two used a high-frequency needle-knife for incision followed by cryoprobe insertion into the lesion, one employed puncture holes to enlarge the airway wall using either the sheath or a laser, and the last seven performed cryobiopsy directly through the puncture needle site.
The comprehensive summaries of the studies included in the present meta-analysis
Author (year) . | Country . | Study design . | Inclusion criteria . | Patients, n . | Reference standard . | . |
---|---|---|---|---|---|---|
Gonuguntla et al. [8] (2021) | India | Case series | Mediastinal lesions ≥1 cm | 4 | Pathologic diagnosis | |
Zhang et al. [9] (2021) | China | Randomized control trial | Mediastinal lesions >1 cm | 197 (3 excluded) | Pathologic diagnosis/follow-up | |
Ariza-Prota et al. [10] (2022) | Spain | Case series | Mediastinal lesions >1 cm | 4 | Pathologic diagnosis | |
Genova et al. [11] (2022) | Italy | Case series | Mediastinal lesions >2 cm | 5 | Pathologic diagnosis/follow-up | |
Gershman et al. [12] (2022) | Israel | Prospective case series | Mediastinal lesions ≥1 cm | 24 | Pathologic diagnosis | |
Fan et al. [13] (2023) | China | Randomized control trial | Mediastinal lesions ≥1 cm | 136 (2 lost to follow-up) | Pathologic diagnosis/follow-up | |
Ariza-Prota et al. [14] (2023) | Spain | Prospective case series | Mediastinal lesions >1 cm | 50 | Pathologic diagnosis/follow-up | |
Salcedo Lobera et al. [15] (2023) | Spain | Prospective case series | Mediastinal lesions >1 cm | 50 | Pathologic diagnosis/follow-up | |
Velasco-Albendea et al. [16] (2023) | Spain | Prospective case series | Mediastinal lesions >1 cm | 27 | Pathologic diagnosis | |
Maturu et al. [17] (2023) | India | Prospective case series | Mediastinal lesions >1 cm | 46 | Pathologic diagnosis/follow-up |
Author (year) . | Country . | Study design . | Inclusion criteria . | Patients, n . | Reference standard . | . |
---|---|---|---|---|---|---|
Gonuguntla et al. [8] (2021) | India | Case series | Mediastinal lesions ≥1 cm | 4 | Pathologic diagnosis | |
Zhang et al. [9] (2021) | China | Randomized control trial | Mediastinal lesions >1 cm | 197 (3 excluded) | Pathologic diagnosis/follow-up | |
Ariza-Prota et al. [10] (2022) | Spain | Case series | Mediastinal lesions >1 cm | 4 | Pathologic diagnosis | |
Genova et al. [11] (2022) | Italy | Case series | Mediastinal lesions >2 cm | 5 | Pathologic diagnosis/follow-up | |
Gershman et al. [12] (2022) | Israel | Prospective case series | Mediastinal lesions ≥1 cm | 24 | Pathologic diagnosis | |
Fan et al. [13] (2023) | China | Randomized control trial | Mediastinal lesions ≥1 cm | 136 (2 lost to follow-up) | Pathologic diagnosis/follow-up | |
Ariza-Prota et al. [14] (2023) | Spain | Prospective case series | Mediastinal lesions >1 cm | 50 | Pathologic diagnosis/follow-up | |
Salcedo Lobera et al. [15] (2023) | Spain | Prospective case series | Mediastinal lesions >1 cm | 50 | Pathologic diagnosis/follow-up | |
Velasco-Albendea et al. [16] (2023) | Spain | Prospective case series | Mediastinal lesions >1 cm | 27 | Pathologic diagnosis | |
Maturu et al. [17] (2023) | India | Prospective case series | Mediastinal lesions >1 cm | 46 | Pathologic diagnosis/follow-up |
Needle type . | Number of passes, n . | Cryoprobe type . | Cooling time . | Number of cryobiopsies, n/mean . | Cryoprobe entry point . | EBUS-TMC-induced complications . |
---|---|---|---|---|---|---|
19G/21G/22G | N/A | 1.1 mm | 3 s | 1–2 | Puncture site of TBNA | None |
19G/22G | 4 | N/A | 7 s | 3 | A needle-knife was used for incision | 2 pneumothorax |
1 pneumomediastinum | ||||||
2 bleeding | ||||||
22G | 4 | 1.1 mm | 3 s | 3 | Puncture site of TBNA | None |
19G | 3 | 1.1 mm | 4 s | 2 | Puncture site of TBNA | None |
22G | 2–4 | 1.1/1.7 mm | 3–4 s | 2–4 | Puncture site was expanded via needle sheath or Laser | None |
19G/22G | 4 | N/A | 7 s | 1 | A needle-knife was used for incision | 2 pneumothorax |
1 pneumomediastinum | ||||||
3 bleeding | ||||||
22G | 3 | 1.1 mm | 4 s | 3 | Puncture site of TBNA | None |
22G | 2 | 1.1 mm | N/A | 4 | Puncture site of TBNA | None |
19G/21G/22G | 3 | 1.1 mm | 3–4 s | 3.4±1.4 | Puncture site of TBNA | None |
19G | 3 | 1.1 mm | 5–6 s | 4 | Puncture site of TBNA | 1 bleeding |
Needle type . | Number of passes, n . | Cryoprobe type . | Cooling time . | Number of cryobiopsies, n/mean . | Cryoprobe entry point . | EBUS-TMC-induced complications . |
---|---|---|---|---|---|---|
19G/21G/22G | N/A | 1.1 mm | 3 s | 1–2 | Puncture site of TBNA | None |
19G/22G | 4 | N/A | 7 s | 3 | A needle-knife was used for incision | 2 pneumothorax |
1 pneumomediastinum | ||||||
2 bleeding | ||||||
22G | 4 | 1.1 mm | 3 s | 3 | Puncture site of TBNA | None |
19G | 3 | 1.1 mm | 4 s | 2 | Puncture site of TBNA | None |
22G | 2–4 | 1.1/1.7 mm | 3–4 s | 2–4 | Puncture site was expanded via needle sheath or Laser | None |
19G/22G | 4 | N/A | 7 s | 1 | A needle-knife was used for incision | 2 pneumothorax |
1 pneumomediastinum | ||||||
3 bleeding | ||||||
22G | 3 | 1.1 mm | 4 s | 3 | Puncture site of TBNA | None |
22G | 2 | 1.1 mm | N/A | 4 | Puncture site of TBNA | None |
19G/21G/22G | 3 | 1.1 mm | 3–4 s | 3.4±1.4 | Puncture site of TBNA | None |
19G | 3 | 1.1 mm | 5–6 s | 4 | Puncture site of TBNA | 1 bleeding |
EBUS-TMC, endobronchial ultrasound-guided transbronchial mediastinal cryobiopsy; TBNA, transbronchial needle aspiration; N/A, not available.
Diagnostic Yield
The meta-analysis revealed a pooled overall diagnostic yield of 89.59% (482/538) for EBUS-TMC and 77.13% (415/538) for EBUS-TBNA. The inverse variance-weighted odds ratio was calculated as 2.63 (95% confidence interval, 1.86–3.72; p < 0.0001), with an I2 index of 11%. Figure 3 displays the forest plot for this meta-analysis, comparing EBUS-TMC and EBUS-TBNA.
The forest plot illustrates the aggregated diagnostic yield of EBUS-TMC in comparison to EBUS-TBNA.
The forest plot illustrates the aggregated diagnostic yield of EBUS-TMC in comparison to EBUS-TBNA.
Subgroup Analyses
A subgroup meta-analysis was conducted in three studies on lung cancer, which revealed a pooled diagnostic yield of 94.76% (217/229) for EBUS-TMC and 93.01% (213/229) for EBUS-TBNA. The inverse variance-weighted odds ratio was calculated to be 1.36 (95% confidence interval, 0.63–2.94; p = 0.44), with an I2 index of 45% (shown in online suppl. Fig. 1; for all online suppl. material, see https://doi.org/10.1159/000538609).
A subgroup meta-analysis was conducted in three studies on lymphoma, which revealed a pooled diagnostic yield of 86.36% (19/22) for EBUS-TMC, whereas a yield of 27.27% (6/22) was achieved with EBUS-TBNA. The inverse variance-weighted odds ratio was calculated to be 12.06 (95% confidence interval, 2.90–50.18; p = 0.0006), with an I2 index of 30% (shown in online suppl. Fig. 2).
A subgroup meta-analysis was conducted in three studies to examine the diagnostic yield of EBUS-TMC and EBUS-TBNA for benign disorders. The results showed that EBUS-TMC had a higher pooled diagnostic yield of 87.62% (92/105) compared to EBUS-TBNA, which showed a yield of 60.00% (63/105). The inverse variance-weighted odds ratio was calculated to be 4.74 (95% confidence interval, 2.35–9.57; p < 0.0001), and the I2 index was found to be 0% (shown in online suppl. Fig. 3).
Complications
The complication rates for each study are summarized in Table 1. Pneumothorax, pneumomediastinum, and mediastinitis occurred in 0.74% (4/538), 0.37% (2/538), and 0.0% (0/538) of cases, respectively. According to a previously defined bleeding severity grading scheme, 6 patients (1.12%) had major airway bleeding (grade 3 or worse) [22]. However, there were no deaths following aggressive endoscopic hemostasis. No additional complications were observed in these studies.
Discussion
According to current clinical guidelines, patients exhibiting abnormally enlarged mediastinal lymph nodes or mediastinal masses as indicated by imaging examination should undergo invasive sampling to ascertain the nature of these lesions. Among the available invasive techniques, EBUS-TBNA is regarded as the preferred method for sampling mediastinal lesions. Although EBUS-TBNA has demonstrated high efficacy and safety in diagnosing lung carcinoma, its usage is currently limited to cytological evaluation due to the restricted volume of specimens obtained [23]. Consequently, its diagnostic utility is constrained when it comes to lymphoma or benign diseases [24, 25]. Furthermore, in light of the emergence of targeted therapy and immunotherapy as viable treatment options for lung cancer, alternative sampling techniques that can yield a larger volume of pathological samples should be explored for immunohistochemistry and genetic testing [26].
To meet the increasing demand for superior specimens in clinical diagnosis and treatment, numerous prior investigations have strived to integrate mediastinum forceps biopsies into transbronchial needle aspiration with the aid of real-time ultrasound guidance [27‒29]. These studies indicate promising potential in augmenting specimen quality and enhancing diagnostic rates among sarcoidosis and lymphoma patients. Nevertheless, a comprehensive prospective trial has revealed that this methodology demonstrates diminished precision when compared to conventional EBUS-TBNA in the identification of malignant mediastinal lymph node metastasis [30].
The cryoprobe is currently predominantly utilized for obtaining pulmonary tissue in cases of interstitial lung diseases. It is favored over forceps biopsy due to its ability to collect samples of adequate volume and higher quality for pathological analysis [31‒33]. This advantage broadens the applications of this technique, especially for peripheral tumor lesions and mediastinum diseases [9, 12‒14, 34]. Furthermore, researchers have also published reviews and meta-analyses on the effectiveness and security of cryobiopsy in the detection of benign or malignant lung conditions [19, 35‒37]. However, the existing literature on mediastinal cryobiopsy primarily consists of descriptive or qualitative analyses of the outcomes, lacking statistical methods for meta-analysis. To the best of our knowledge, this study is the first to conduct a meta-analysis assessing the efficacy and safety of EBUS-TMC in mediastinal disorders.
In this study, a systematic literature review and meta-analysis were conducted to assess the efficacy and safety of EBUS-TMC compared to EBUS-TBNA for diagnosing patients with mediastinal lesions. The findings indicate that EBUS-TMC yields a higher overall diagnostic rate for mediastinal disorders than EBUS-TBNA. In addition, subgroup analysis revealed that EBUS-TMC does not confer a significant advantage in diagnosing lung cancer when compared to EBUS-TBNA. However, EBUS-TMC demonstrated a significantly higher diagnostic accuracy for lymphoma and benign diseases. In general, bleeding was the most commonly reported complication, although many cases were mild or minor. None of the included studies reported any serious infections or consequences, such as breathing difficulties or death. The technique is deemed safe as no significant postoperative complications were observed. Therefore, EBUS-TMC may provide a promising and novel approach to improve the diagnostic efficiency of mediastinal lesions.
Despite its strengths, the shortcomings of this meta-analysis should be acknowledged. The constraint imposed by the limited number of existing studies significantly diminishes the statistical accuracy of our analysis, thus warranting careful consideration. Moreover, the variety of cryoprobe sizes and cooling times employed in each study may have significantly affected the overall diagnostic performance and safety profile. In addition, potential variations in the amount of cryobiopsy passes per lymph node station and the specific cryobiopsy procedure could not be standardized, which could potentially influence the diagnostic accuracy. Finally, regardless of the outstanding outcomes, a number of challenges prevent the widespread adoption of this procedure. Presently, cryobiopsy is only accessible in a few specialized centers. The procedure involves a steep learning curve when performed in the mediastinum and the most appropriate sampling approach remains unknown.
Conclusion
In summary, this comprehensive analysis indicates the improved diagnostic outcomes of EBUS-TMC on overall yield, as well as in the diagnosis of benign disease and lymphoma compared to EBUS-TBNA. Notably, the procedure did not present any major complications. Nevertheless, additional prospective multicenter randomized controlled trials should be conducted to establish the relevance of these results.
Acknowledgments
We thank Home for Researchers editorial team (www.home-for-researchers.com) for language editing service.
Statement of Ethics
The present study was conducted in Sichuan Cancer Hospital in full accordance with the ethical principles of the World Medical Association Declaration of Helsinki and the additional requirements of Chinese law. Furthermore, Sichuan Cancer Hospital, China, classified the study as being exempt from ethical review as it carries only negligible risk and involves the use of existing non-identifiable human data. The need for written informed consent was waived by the Ethical Committee of Sichuan Cancer Hospital. Written informed consent was obtained from the authors for publication of this meta-analysis.
Conflict of Interest Statement
The authors have no conflicts of interest to declare.
Funding Sources
This study was not supported by any sponsor or funder.
Author Contributions
Zhenming Zhang, Shengping Li, and Yu Bao proposed the conception, analyzed the bibliography, and revised the final manuscript. Zhenming Zhang and Shengping Li drafted the initial manuscript. All coauthors edited and approved the final version.
Additional Information
Zhenming Zhang and Shengping Li contributed equally to this work.
Data Availability Statement
All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.