br ACCEPTED MANUSCRIPT br The current study was performed with

ACCEPTED MANUSCRIPT
The current study was performed with the goal of defining the prevalence of occult LN metastases discovered on combined endosonography during a single procedure performed with endobronchial ultrasound (EBUS) and endoscopic ultrasound (EUS) fine needle aspiration in the radiologically normal mediastinum in patients undergoing pre-operative lung cancer staging.
Methods
The study was approved by the institutional review board at the Centre de Recherche, Centre Hospitalier de l’Université de Montréal (IRB # 15.150). The study consists of a retrospective single tertiary referral center review of a prospectively maintained thoracic surgery interventional endoscopy database. Records of all consecutive patients who underwent combined endosonographic staging with EBUS and EUS – fine needle aspiration (FNA) from 2009 to 2014 were reviewed. Patients with biopsy proven non-small cell lung cancer (NSCLC) with radiologically N0 and N1 LNs and radiographically resectable tumor assessed with CT and PET
– CT whom underwent combined EBUS/EUS as initial pre-operative staging were included in the study.
radiologically normal. The literature and American College of Chest Physicians guidelines
define a high risk tumor for occult mediastinal lymph node metastases as: central tumor, tumor
is considered high risk because in this situation, the lymph nodes will typically also not show
metabolic activity on PET, even in the presence of occult metastases. Patients were staged if they
had central tumor, tumor size > 3cm, N1 lymph node involvement on PET–CT/CT, or if there
international association for the study of lung cancer (IASLC) map (36).
ACCEPTED MANUSCRIPT
Presence of malignancy or adequate lymph node tissue reported by an experienced cytopathologist on final official cytopathology report was used as the final result for analysis. Positive, negative and insufficient results on all the lymph node aspirations and adrenal gland biopsies were documented. Isolated adrenal metastasis from lung cancer has been reported in the literature (37). Hence, we also collected the data on adrenal biopsies that were performed in this cohort. Insufficient sampling was defined as a lack of cancer Necrosulfonamide or normal lymphocytes on cytopathology. Adequate lymph node sampling confirmed by cytopathologists was obtained in all specimens (Table – 1 and 2).
EBUS – TBNA was performed using a real time – curvilinear bronchoscope (Olympus BF-
UC180F). EUS – FNA was performed using a real time – curvilinear endoscope (Olympus GF-
gauge needle. Rapid on-site cytology was not utilized in any case. Lymph nodes equal to or
greater than 4 mm in short axis were biopsied. The criteria that we used for suspicious lymph
nodes were: round with a distinct margin, heterogeneous and absence of central hilum as
described by Fujiwara et al (38).
All suspicious lymph nodes greater than or equal to 4 mm were sampled in the standard N3 to
N2 to N1 order. The adrenal gland was sampled if it was more than 1 cm in size or PET – CT
avid. A minimum of 3 needle aspirations were performed in each target. Procedures were
performed either under moderate sedation or general anesthesia depending on patient
comorbidities and ability to tolerate moderate sedation.
ACCEPTED MANUSCRIPT
procedure is standardized and starts with white light bronchoscopy followed by EBUS and then EUS. There were no procedure related complications.
Patients were considered high risk for pulmonary resection if the FEV1 was < 60% of predicted /
< 10 ml/kg/min (if results of cardiopulmonary exercise tests area available), or if ECOG performance score was > 2.
Statistical Analysis
Analysis was performed using IBM SPSS Statistics software version 22.
Results
A total of 161 patients with non-small cell lung cancer were identified in whom both the pre-endosonography CT and PET-CT were negative for mediastinal LN metastases. 40/161 (25%) were considered to be high risk for surgical resection due to poor lung functions or performance status. The median age of the cohort was 65 years, 53% of patients were male (n=85). The histology, location and T size of primary tumor is depicted in table 3. A total of 416 lymph nodes were biopsied in the 161 patients by combined endosonography, 147 by EBUS and 269 by EUS. Mean and median number of lymph nodes biopsied per patient using combined EBUS/EUS was 2.5 and 3, respectively (mean and median EBUS: 0.91 and 2.5; mean and median EUS 1.6 and 3).
ACCEPTED MANUSCRIPT
– CT was negative for adrenal involvement in both cases, however, the adrenal gland was biopsied since it was larger than 1 cm on CT.
Of the 6 patients with N1 disease on endosonographic staging who underwent surgical resection, no patients were upstaged on surgical lymph node dissection. The clinical characteristics of upstaged patients is shown in tables 4 and 5. The rate of unsuspected N2 disease was higher in patients with adenocarcinoma given the higher prevalence of adenocarcinoma in the cohort. Out of the 13 patients that upstaged to N2 disease on staging endosonography, 3 patients underwent surgical resection. Two out of three underwent neo-adjuvant chemotherapy followed by resection one of these patients was alive at one year follow up. One out of these three patients had a bone metastases detected after surgery and underwent palliative chemotherapy. This patient was not alive at one year follow up. Ten patients underwent definitive chemo-radiation therapy as they were deemed inoperable due to comorbidities. Six patients underwent follow up within our system and 5 of these patients were alive at 18 months. Four patients were treated at an outside facility and follow- up data is unavailable for these patients.