In lung cancer with enlarged or non-enlarged mediastinal lymph nodes, contrast-enhanced computed tomography (CT) and Positron emission tomography (PET) scan frequently show discrepancy in tuberculosis-endemic area. Endobronchial ultrasound guided transbronchial aspiration (EBUS-TBNA) with ability of real-time nodal sampling possibly improves the nodal diagnosis.

The purpose of this study is to compare the accuracy of nodal diagnosis of contrast-enhanced CT and PET scan with and without EBUS-TBNA, this study will be performed.

Lung cancer remains a fatal disease worldwide, and surgical treatment offers possibility for long-term survival. However, the indication and outcome of surgical resection depends on the pre-operative accurate staging and extent of intra-operative lymph node dissection. Therefore, the accurate lymph node staging in non-small cell lung cancer (NSCLC) is crucial for planning optimal treatment. Traditionally, the conventional contrast-enhanced CT essentially identifies enlarged lymph node greater than 1cm as nodal metastasis. Nevertheless, with moderate sensitivity and specificity, contrast-enhanced CT carries substantial risk to under-stage small nodal metastasis and to over-stage inflammatory lymphadenitis.

Positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose (FDG) provides functional images of tumor metabolism, and has been used as a non-invasive alternative other than contrast-enhanced CT for nodal staging in NSCLC. In the absence of detectable lymph node enlargement by CT, FDG-PET scan were increasingly used to stage the lymph node status for NSCLC in some part of world. Hence, the accuracy of FDG-PET might substantially alter the treatment strategy in an institution where the mediastinoscopy is unavailable for lymph node sampling. However, it is generally agreed that abnormal FDG uptake occurred frequently in granulomatous and inflammatory disease. In an endemic area where tuberculosis is still prevalent; such as Eastern Asia, FDG-PET scan has reportedly shown reduced sensitivity and positive predictive value in nodal staging of NSCLC. Thereby, FDG-PET scan alone does not appear to replace mediastinoscopy for nodal staging of NSCLC in a tuberculosis-endemic area, especially in potentially operable patients without enlarged mediastinal lymph nodes.

The recent development of curved ultrasound probe-equipped bronchoscope, which enables direct and real-time aspiration by endobronchial ultrasound- transbronchial needle aspiration (EBUS-TBNA) of mediastinal and hilar lymph nodes, has become an less invasive alternative for nodal staging other than mediastinoscopy. By direct nodal sampling, EBUS-TBNA improves lymph node staging from an image basis to a cytology basis; or even, pathology basis. However, the variable sensitivity and negative predictive value of EBUS-TBNA has been reported, especially in lymph node reduced in size after induction chemotherapy. Nevertheless, reports from NSCLC without significant mediastinal lymph node enlargement on CT otherwise suggested EBUS-TBNA exhibited a high sensitivity and specificity for detecting small nodal metastasis. Therefore, whether EBUS-TBNA retains the reportedly high performance of nodal staging in lung cancer patients without enlarged mediastinal lymph node on CT in a TB endemic country; a condition of FDG-PET scan reportedly showed increased false-positive rate, is still unclear.

In present study, we primarily aim at the comparison of accuracy of nodal diagnosis of contrast-enhanced CT and PET scan with and without EBUS-TBNA in a condition of mediastinal and hilar lymph nodes of lung cancer. Secondarily, we aim at the accuracy of nodal diagnosis by FDG-PET scan in the same condition, and investigate the characteristics of lymph nodes with false PET result.

Inclusion Criteria:

• NSCLC,
• Completed whole body CT or PET scan.

Exclusion Criteria:

• Pregnancy,
• Age less than 20 years old,
• Other malignancy.