Does video-mediastinoscopy improve the results of conventional mediastinoscopy?

OBJECTIVE: Despite new technologies, mediastinoscopy remains the gold standard for mediastinal staging of lung cancer even though the procedure is not standardised. Introduction of video-mediastinoscopy (VM) may help to overcome this problem as it better visualises the anatomy and allows a more uniform dissection than conventional mediastinoscopy (CM). Does the use of VM result in more lymph node tissue, higher accuracy and lower complication rates as compared to CM?

METHODS: All mediastinoscopies from June 2003 to December 2005 were analysed. In a protocol surgeons documented location of lymph node stations, number of lymph nodes resected or biopsied and technique (VM or CM). Two groups were created for analysis: group 1 (n=366) consisting of all mediastinoscopies was reviewed for complication rates; group 2 included all patients with lung cancer who had a pN0 status by mediastinoscopy and underwent subsequent thoracotomy (n=171). This group was studied for the number of lymph nodes resected or biopsied according to the technique (VM or CM), on accuracy and negative predictive value.

RESULTS: Of 366 mediastinoscopies, 132 were CM (36.1%) and 234 VM (63.9%). Complications occurred in 17 patients (4.6%): 9 recurrent laryngeal nerve palsies (VM 2.1%, CM 3.0%), 5 mediastinal enlargement on routine chest radiography interpreted as postoperative bleeding (VM 0.9%, CM 2.3%), pneumonia (1), intraoperative laceration of the pleura (1) and main bronchus (1), both corrected during the procedure (all VM 1.3%). No intraoperative haemorrhage or death occurred. VM resected more lymph nodes (mean 8.1, range 3-25) then CM (mean 6.0, range 3-11), for all mediastinoscopies the mean lymph node yield was 7.6 (range 3-25). Comparison of lymphadenectomy via thoracotomy in patients classified pN0 by mediastinoscopy (n=171) showed an accuracy of 87.9% for VM versus 83.8% for CM (85.8% for all mediastinoscopies) with a negative predictive value of 0.83 for VM and 0.81 for CM (0.82 for all mediastinoscopies).

CONCLUSION: This study demonstrates that in comparison with CM, VM routinely yields more lymph nodes with fewer complications with a tendency towards better accuracy and negative predictive value. For these reasons, we believe that VM should replace CM as the method of choice. Furthermore VM would allow standardisation, thereby having an advantage in comparison to the less invasive newer staging techniques. This way mediastinoscopy could remain the gold standard despite its invasiveness.