ABSTRACT
Background:
Endobronchial ultrasound-guided transbronchial fine-needle aspiration (EBUS-TBNA) has replaced mediastinoscopy as the preferred investigation for evaluating mediastinum in staging lung cancer. There is little evidence of mediastinal staging by EBUS-TBNA from India.
Objectives:
To study endobronchial ultrasound’s diagnostic accuracy in staging lung cancer.
Methodology:
We retrospectively analysed patients operated on for lung cancer where EBUS was performed preoperatively for mediastinal staging. We compared the histological findings obtained from different mediastinal lymph nodes (LNs) by EBUS-TBNA with the pathology of the same LNs obtained after surgical dissection as the reference standard.
Results:
Seventy-six patients underwent curative surgery for lung cancer. The diagnostic accuracy, sensitivity, specificity, positive predictive value and negative predictive value of EBUS-TBNA in predicting mediastinal metastasis were 93.9%, 40%, 99%, 80% and 94.6%, respectively. Of the 115 LNs sampled, EBUS-TBNA was false negative in six nodes, resulting in an up-staging of six patients.
Conclusions:
EBUS-TBNA has a high diagnostic accuracy for lung cancer staging.
KEY WORDS: Diagnostic accuracy, EBUS-TBNA, lung cancer staging, mediastinoscopy
BACKGROUND
Lung cancer is the fourth (4.9%) most common cancer in India and accounts for 8.1% of all cancer-related deaths.[1] The choice of treatment for lung cancer depends on mediastinal lymph nodal involvement and distant metastasis. In the absence of distant spread on imaging, invasive sampling from mediastinal lymph nodes (LNs) is required to select patients likely to benefit from surgery.[2] After sampling, patients with N1 disease can be immediately considered for surgery. Those found to have N2 disease may be either treated by neoadjuvant chemotherapy followed by surgery or chemoradiotherapy, depending on the extent of mediastinal involvement and operability of the patient. However, for those found to have N3 disease, with involvement of contralateral mediastinal LNs, surgery is contraindicated.[3] It has been reported that in about 21% of patients with N0 disease (no lymph node enlargement) by CT and approximately 14% in the PET-CT negative group, occult metastasis is found by EBUS sampling.[4] Therefore current guidelines recommend patients with lung tumours more than 3 cm in size, central tumours and clinical N1 to N3 disease, mediastinal sampling must be performed.[5] Mediastinoscopy which was earlier the investigation of choice for this has now been replaced by endobronchial ultrasound-guided transbronchial fine-needle aspiration (EBUS-TBNA), a minimally invasive day-care procedure with high diagnostic yield. The diagnostic accuracy of mediastinal lymph node sampling in lung cancer patients can be truly ascertained if the cytology findings with EBUS-TBNA are compared with the histopathology of the same node after surgical resection. In this study, we aimed to ascertain the accuracy of EBUS-TBNA in mediastinal lung cancer staging by this method.
The current guidelines recommend that a confirmatory preoperative mediastinoscopy be performed in EBUS-TBNA negative cases, but this approach has been questioned.[6,7] We have discussed the need for this approach based on the results of the diagnostic accuracy of EBUS-TBNA in our patient population.
METHODS
The current study is a retrospective analysis of patients operated for lung cancer at a tertiary cancer hospital between 2015 and 2021. The study protocol was approved by the Institute Ethics Committee. We were granted a consent waiver as the study was analysis of anonymised patient data.
Inclusion criteria
Consecutive patients with lung cancer who underwent curative surgery and in whom a preoperative mediastinal staging was performed using EBUS-TBNA.
Exclusion criteria
Patients in whom neoadjuvant chemotherapy was administered before surgery.
Study methods
We retrieved the following information from the patient records: (1) demographic details (age and sex); (2) PET-CT chest findings, including lymph node size and station; (3) EBUS-TBNA findings of lymph node size and station; (4) EBUS-TBNA cytology reports; (5) histopathology reports from surgically resected LNs; and, (6) time interval between EBUS-TBNA and surgery in each patient.
All LNs with an SUV value >2.5 were deemed positive on the PET scan. EBUS-TBNA was performed using a convex-probe EBUS bronchoscope (BF-UC180F, Olympus Medical, Singapore) and dedicated 22 or 21-gauge (Olympus ViziShot TBNA) needles. A systematic sampling from LNs larger than 5mm was conducted from each accessible LN station starting from the contralateral interlobar and hilar LNs (stations 11 and 10) moving to the contralateral mediastinal (stations 2, 4, and 7) and then finally the ipsilateral mediastinal, hilar and interlobar LNs. We did not perform trans-oesophageal endoscopic ultrasound using the echobronchoscope (EUS-B) for station 8 or 9 LNs.
Rapid onsite cytological evaluation from aspirates was performed in all cases and the procedure terminated if lymph node at any station was found to be malignant. Alcohol-fixed TBNA slides and cell blocks in CytoliteR solution from each lymph node were labeled separately and sent to the pathologists for a final report.
We compared the pathology findings of mediastinal LNs for staging by EBUS-TBNA and the final pathology of the same lymph nodes after surgical resection. Nodes that were negative by EBUS-TBNA but positive for metastasis after surgery were called ‘unforeseen N1/N2 nodes’ depending on the station.
Statistical analysis
We report the EBUS-TBNA test characteristics by calculating sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). True positives (TP) were defined as LNs with the same patholgical diagnosis of metastasis on EBUS-TBNA as surgery. In contrast, true negatives (TN) were defined as LNs without pathological findings on EBUS and surgery. False negatives (FN) were defined as either EBUS-TBNA reporting the absence of disease in an LN but the presence of metastasis on surgical evaluation or EBUS failing to sample accessible pathological LNs. False positives (FP) were defined as EBUS-TBNA finding metastasis that was not seen in the pathology from surgical specimens. The accuracy of EBUS-TBNA was defined as the ability to correctly diagnose the presence or absence of pathology in LNs and calculated by an equation TP+TN/TP+TN+FP+FN. Our primary outcome was to determine the accuracy of EBUS-TBNA in detecting malignant spread in mediastinal and hilar/interlobar LNs in patients with proven lung cancer. We also looked at the number of false negative EBUS-TBNA results. We calculated the number of mediastinoscopy procedures which would have to be conducted to pick up these false negative nodes.
RESULTS
During the study period, 76 patients underwent curative surgery for lung cancer. A total of 198 LNs were sampled using EBUS-TBNA. EBUS was primarily performed under moderate sedation and local anaesthesia except for three cases under general anaesthesia. Of the 76 patients, 16 (32 LNs) were excluded as they received neoadjuvant chemotherapy. Of the remaining 166 LNs in 60 patients, 51 were not dissected at surgery and were excluded from the comparison [Figure 1]. The details of the 115 nodes sampled using EBUS-TBNA and resected during surgery are summarised in Table 1. The mean size of the LNs was 1.61 cm and 95 (82.6%) were PET positive. The LNs sampled during EBUS-TBNA were stations 7 (33.7%), 4R (22.3%), 4L (20.5%), 11R (7.2%), 11L (9.6%), 10R (3.0%), 10L (2.4), and 5 (1.2%). There were 91 N2 nodes and 24 N1 nodes.
Figure 1.
Details of the number of operated patients and lymph nodes sampled by EBUS TBNA finally included in the study
Table 1.
Data of 60 evaluable patients who underwent surgery for primary lung cancer
| Baseline characteristic of patients | Frequency | Percentage |
|---|---|---|
| Sex (n=60) | ||
| Male | 47 | 78.3% |
| Female | 13 | 21% |
| Age | ||
| Maximum | 83 | Mean age - 64.93 yrs |
| Minimum | 37 | SD +/- 8.3 yrs |
| PET results (n=115) | ||
| Negative | 20 | 17.4 |
| Positive | 95 | 82.6 |
| LN size on PET scan | ||
| Maximum | 6.1 cm | Mean size - 1.61 cm SD +/- 0.86 cm |
| Minimum | 0.6 cm | |
| LN stations sampled (n=115) | ||
| 10L | 4 | 3.5 |
| 10R | 3 | 2.6 |
| 11L | 9 | 7.8 |
| 11R | 8 | 7.0 |
| 4L | 10 | 8.7 |
| 4R | 28 | 24.3 |
| 5 | 1 | 0.9 |
| 7 | 52 | 45.2 |
| Clinical staging | ||
| 1A | 4 | 3.5 |
| 1B | 19 | 16.5 |
| 2A | 6 | 5.2 |
| 2B | 35 | 30.4 |
| 3A | 41 | 35.7 |
| 3B | 7 | 6.1 |
| Not Specified | 2 | 1.7 |
| EBUS result (n=115) | ||
| Negative | 110 | 97.0 |
| Positive | 5 | 3.0 |
| Histopathology result (Dissected=115) | ||
| Not Dissected | 51 | |
| Negative | 105 | 91.3 |
| Positive | 10 | 8.7 |
One hundred and four LNs were true negative (both EBUS-TBNA and final histopathology negative), four true positive (all N1 nodes, both EBUS-TBNA and final pathology positive), six false negative (all N2 nodes, negative on EBUS-TBNA and positive on final pathology), and one false positive (N1 node, positive on EBUS-TBNA and negative on final pathology). This patient had an ipsilateral station 11L PET-positive LN, which was also reported positive on EBUS-TBNA but was negative on the postoperative histopathology report. All the false negative nodes (station 7 [n = 3], stations 4R, 4L, and 5 [1 each]) were PET-positive and had microscopic metastasis in a single station. As a result, three patients were upstaged from 3A to 3B, two from 2A to 3A, and one from 2B to 3B after surgery [Table 2]. The diagnostic accuracy of EBUS-TBNA in predicting mediastinal metastasis was 93.9%. The sensitivity, specificity, PPV, and NPV were 40%, 99%, 80%, and 94.6%, respectively [Tables 3 and 4]. The mean interval between staging EBUS and surgery was 15.6+/- 6.5 days in the true negative patients and 11+/- 3.8 days in those with unforeseen N2 nodes, and the difference between the two groups was not statistically significant ‘p’ value- 0.302.
Table 2.
Details of patients with false negative EBUS-TBNA, Neg-Negative, Pos-Positive, LN size in mm
| Patient No | Lymph node Station | EBUS | Stage | EBUS Node stage | Postop Result | Postop Node stage | Final Stage | PET Scan | LN Size |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 7 | Neg | 3A | N1 | Pos | N2 | 3B | Pos | 5.1 |
| 2 | 5 | Neg | 2A | N0 | Pos | N2 | 3A | Pos | 2.9 |
| 3 | 4L | Neg | 3A | N0 | Pos | N2 | 3B | Pos | 1.5 |
| 4 | 4R | Neg | 2A | N0 | Pos | N2 | 3A | Pos | 1.9 |
| 5 | 7 | Neg | 3A | N0 | Pos | N2 | 3B | Pos | 1.3 |
| 6 | 7 | Neg | 2B | N0 | Pos | N2 | 3B | Pos | 2.5 |
Table 3.
EBUS-TBNA & surgical histopathology co-relation
| Postoperative histopathology | Total | ||
|---|---|---|---|
|
| |||
| Pos | Neg | ||
| EBUS result | |||
| Pos | 4 | 1 | 5 |
| Neg | 6 | 104 | 110 |
| Total | 10 | 105 | 115 |
Table 4.
Overall accuracy of EBUS-TBNA
| Accuracy of EBUS-TBNA | Percentage |
|---|---|
| Sensitivity | 40% |
| Specificity | 99% |
| NPV | 94.6% |
| PPV | 80.00% |
| Diagnostic accuracy | 93.9% |
DISCUSSION
We found a high diagnostic accuracy (93.9%) of EBUS-TBNA in predicting mediiastinal metastasis. The specificity of 99% and NPV of 94.6% are similar to that reported in the literature.[8,9,10,11,12]
We found a lower sensitivity of 40% and PPV of 80% EBUS-TBNA, than reported in the literature, in mediastinal staging of lung cancer. The lower sensitivity and PPV in our study are explained by the fact that we included only those patients who underwent surgery. Thus, patients with positive N1 nodes on EBUS-TBNA were included. However, patients with malignancy detected in the N2 and N3 nodes did not undergo lung surgery and were therefore excluded from this analysis. This has resulted in a falsely low sensitivity and PPV.
Six unforeseen N2 PET positive LNs in six different patients were found (negative on EBUS-TBNA but were positive after surgery). This resulted in patients being upstaged to either 3A or 3B. Potentially, they could have been offered neoadjuvant chemotherapy before surgery or given the option of radiotherapy and chemotherapy. There were 3 LNs from station 7, 1 each from stations 4R, 4L and 5. The stations 7, 4R and 4L are easily accessible using EBUS; thus, the false negative EBUS-TBNA was due to sampling error. Accessing the station 5 node by EBUS was technically tricky. One can argue that had mediastinoscopy been performed in these cases before being taken up for surgery, then the nodes from stations 7, 4R and 4L could have shown malignant metastasis. The metastasis in LN at station 5 would have required an anterior mediastinotomy (Chamberlain’s procedure) for sampling.
The NCCN and ACCP guidelines on diagnosis of lung cancer recommend mediastinal sampling by EBUS. If negative, additional confirmatory mediastinoscopy be conducted to look for mediastinal nodal metastasis in a clinically (PETand/or CT) positive mediastinum. We have calculated that in our cohort of 60 patients, approximately ten additional mediastinoscopy procedures would have had to be performed before picking up one false negative node on EBUS-TBNA and the node at Station 5 would have to be sampled by an anterior mediastinotomy.
A review of the literature shows, however, that the issue of the need for an additional confirmatory mediastinoscopy continues to be debated. To begin with Annema et al. compared the yield of combined EBUS and EUS followed by mediastinoscopy if negative versus mediastinoscopy alone, for detecting mediastinal nodal metastasis. He showed that the addition of combined EBUS and EUS to mediastinoscopy increased sensitivity and NPV of staging by 9% and 11%, respectively. On the contrary, a meta-analysis performed a few years later by Bousema et al.[13] suggested that the role of confirmatory mediastinoscopy is debatable. This meta-analysis comprising 3248 patients showed that the rate of unforeseen N2 disease in resected NSCLC was 9.6% in patients staged with endosonography alone versus 9.9% who underwent additional surgical staging with mediastinoscopy. Moreover, a further complication rate of 6% was seen in patients undergoing mediastinoscopy.
A recently published multicentre, randomised controlled trial (MEDIASTrial) compared endosonography with or without confirmatory mediastinoscopy for resectable lung cancer. This trial found that a negative endosonography followed by immediate resection was not inferior to additional mediastinoscopy before resection. Confirmatory Mediastinoscopy increased the rate of detection of mediastinal nodal metastasis only marginally by 1.03% at the expense of a 10-day delay for lung tumour resection, morbidity in 6.3%, mortality in 0.6%. It repeated general anaesthesia in all the patients involved.[14]
Kujawa et al.[15] studied the cost-effectiveness of mediastinoscopy after negative EBUS-TBNA and concluded that mediastinoscopy is cost-effective only in patients where the probability of metastatic pathological N2 is high (pN2 disease >57%). For pN2 disease probability between 2.5 and 57%, EBUS-TBNA is the only cost-effective staging modality. In our series, unforeseen N2 nodes were seen in only 6.5% (6/91 N2 nodes), putting a question mark on the cost-effectiveness of an additional mediastinoscopy. Moreover, single station and microscopic metastases have better outcomes than multiple stations and macroscopic unforeseen N2.[16,17] While there are no RCTs in patients with microscopic metastases, observational data suggest no survival benefit with neoadjuvant treatment versus upfront surgery in patients with minimal N2.[18,19,20] It may be noted that all six unforeseen N2 nodes in our series were single station and had microscopic metastasis, thus putting another question mark on the need for confirmatory mediastinoscopy in our patients.
Our study has significant limitations. The size of the study is small. It is a retrospective analysis of patients operated on for lung cancer but without data about the total number of patients who were taken up for lung cancer staging and of those excluded from this study due to mediastinal metastasis. Only that data would have enabled us to estimate the true sensitivity and PPV of EBUS-TBNA in our setting. Although EUS-B was not conducted in the current study, it is possible that routinely combining EBUS and EUS-B could increase the yield and diagnostic accuracy further in a larger pool of patients.[21] Finally, it is not a head-to-head comparison like the ‘Mediast’ trial, so the conclusions drawn can only provide pointers to more prospective research to answer the question about the need for preoperative confirmatory mediastinoscopy in EBUS-TBNA negative patients.
In conclusion, EBUS-TBNA, conducted systematically, can provide high diagnostic accuracy in lung cancer nodal staging. The need for additional confirmatory mediastinoscopy is debatable but may be considered in PET-positive single station nodes equal to or more than 1.5 cm in size without macroscopic metastasis. Due consideration also needs to be given to whether it is cost-effective, keeping in mind the probability of mediastinal metastasis in a given population.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
- 1.Singh N, Agrawal S, Jiwnani S, Khosla D, Malik PS, Mohan A, et al. Lung cancer in India. J Thorac Oncol. 2021;16:1250–66. doi: 10.1016/j.jtho.2021.02.004. [DOI] [PubMed] [Google Scholar]
- 2.Rankin S. PET/CT for staging and monitoring non-small lung cancer. Cancer Imaging. 2008;8:S27–31. doi: 10.1102/1470-7330.2008.9006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.NCCN Clinical Practice Guidelines in Oncology (NCCN) Non-Small Cell Lung Cancer version 3.2023-April 13 2023. Available from: www.nccn.org/guidelines .
- 4.Leoong TL, Loveland PM, Goreliik A, Irving L, Steinfort DP. Preoperative staging by EBUS in cN0/N1 Lung cancer, systematic review and meta-analysis. J Bronchol Intervent Pulmonol. 2019;26:155–65. doi: 10.1097/LBR.0000000000000545. [DOI] [PubMed] [Google Scholar]
- 5.Sivestri GA, Gonzalez AV, Jantz MA, Margolis ML, Gould MK, Tanoue LT, et al. Methods for staging non-small cell lung cancer. Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians -Evidence based clinical practice guidelines. Chest. 2013;143((5 Suppl)):e211S–50S. doi: 10.1378/chest.12-2355. [DOI] [PubMed] [Google Scholar]
- 6.Annema JT, Meerbeeck JPV, Rintoul RC, Dooms C, Deschepper E, Dekkers OM, et al. Mediastinoscopy vs endosonograpghy for mediastinal nodal staging of lung cancer. JAMA. 2010;304:2245–52. doi: 10.1001/jama.2010.1705. [DOI] [PubMed] [Google Scholar]
- 7.Czarnecka- Kujawa K, Yasufuku K. The role of endobronchial ultrasound versus mediastinoscopy for non-small cell lung cancer. J Thorac Dis. 2017;9((Suppl 2)):S83–97. doi: 10.21037/jtd.2017.03.102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hanna WC, Yasafuku K. Mediastinoscopy in the era of endobronchial ultrasound: When it should be performed? Curr Respir Care Resp. 2013;2:40–6. [Google Scholar]
- 9.Gu P, Zhao Y-Z, Jiang L-Y, Zhang W, Xin Y, Han B-H. Endo- bronchial ultrasound-guided transbronchial needle aspiration for staging of lung cancer: A systematic review and meta-analysis. Eur J Cancer. 2009;45:1389–96. doi: 10.1016/j.ejca.2008.11.043. [DOI] [PubMed] [Google Scholar]
- 10.Adams K, Shah PL, Edmonds L, Lim E. Test performance of endobronchial ultrasound and transbronchial needle aspiration biopsy for mediastinal staging in patients with lung cancer: Systematic review and meta-analysis. Thorax. 2009;64:757–62. doi: 10.1136/thx.2008.109868. [DOI] [PubMed] [Google Scholar]
- 11.Sehgal IS, Agarwal R, Dhooria S, Prasad KT, Aggarwal AN. Role of EBUS TBNA in staging of lung cancer: A clinician's perspective. J Cytol. 2019;36:61–4. doi: 10.4103/JOC.JOC_172_18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yasufuku K, Pierre A, Darling G, de Perrot M, Waddell T, Johnston M, et al. A prospective controlled trial of endobronchial ultrasound-guided transbronchial needle aspiration compared with mediastinoscopy for mediastinal lymph node staging of lung cancer. J Thorac Cardiovasc Surg. 2011;142:1393–400. doi: 10.1016/j.jtcvs.2011.08.037. [DOI] [PubMed] [Google Scholar]
- 13.Bousema JE, Dorp MV, Noyez VJJM, Dijkgraaf MGW, Annema JT, van den Broek FJC. Unforeseen N2 disease after negative endosonography findings with or without confirmatory mediastinoscopy in resectable No-small lung cancer: A systematic review and Meta-Analysis. J Thorac Oncol. 2019;14:979–92. doi: 10.1016/j.jtho.2019.02.032. [DOI] [PubMed] [Google Scholar]
- 14.Bousema JE, Dijikgraaf MGW, van der Heijden EHFM, Verhagen AFTM, Annema JT, van den Broek FJC, et al. Endosonography with or without confirmatory Mediastinoscopy for resectable lung cancer: A randomized clinical trial. J Clin Oncol. 2023;41:3805–15. doi: 10.1200/JCO.22.01728. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kujawa KC, Rochau U, Siebert U, Atenafu E, Darling G, Waddell TK, et al. Cost-effectiveness of mediastinal lymph node staging in non-small cell lung cancer. J Thorac Cardiovasc Surg. 2017;153:1567–78. doi: 10.1016/j.jtcvs.2016.12.048. [DOI] [PubMed] [Google Scholar]
- 16.Garelli E, Renaud S, Falcoz PE, Weingertner N, Olland A, Santelmo N, et al. Microscopic N2 disease exhibits a better prognosis in resected non-small-cell cancer. Eur J Cardiothoracic Surg. 2016;50:322–8. doi: 10.1093/ejcts/ezw036. [DOI] [PubMed] [Google Scholar]
- 17.Xu L, Su H, She Y, Dai C, Zhao M, Gao J, et al. Which descriptor is more predictive of prognosis in resected non-small cell lung cancer: The number of involved nodal stations or the location based pathological stage? Chest. 2021;159:2548–69. doi: 10.1016/j.chest.2020.12.012. [DOI] [PubMed] [Google Scholar]
- 18.Thomas DC, Arnold BN, Rosen JE, Salazar MC, Detterbeck FC, Blasberg JD, et al. The significance of upfront knowledge of N2 disease in non-small cell lung cancer. World J Surg. 2018;42:161–71. doi: 10.1007/s00268-017-4165-6. [DOI] [PubMed] [Google Scholar]
- 19.Keshava HB, Tan KS, Dycoco J, Livschitz J, Bott MJ, Huang J, et al. How effective is neoadjuvant therapy followed surgery for pathologic single station N2 non-small cell lung cancer? Semin Thorac Cardiovasc Surg. 2021;33:206–16. doi: 10.1053/j.semtcvs.2020.08.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kuijvenhoven JC, Korevaar DA, Tournoy KG, Malfait TL, Dooms C, Rintoul RC, et al. Five- 84. Year survival after endosonography vs mediastinoscopy for mediastinal nodal staging of lung cancer. JAMA. 2016;316:1110–2. doi: 10.1001/jama.2016.10349. [DOI] [PubMed] [Google Scholar]
- 21.Herth FJ F, Krasnik M, Kahn N, Eberhardt R, Ernst A. Combined endoscopic -endobronchial ultrasound-guided fine-needle aspiration of mediastinal lymph nodes through a single bronchoscope in 150 patients with suspected lung cancer. Chest. 2010;138:790–4. doi: 10.1378/chest.09-2149. [DOI] [PubMed] [Google Scholar]