Cervical ultrasound in operable non-small cell lung cancer.

We read with interest the recent article [1] in the journal on the role of cervical ultrasonography in the staging workup of patients with otherwise operable non-small cell lung cancer (NSCLC). We were impressed by the simplicity of the study and the ability of cervical ultrasound to diagnose metastatic neck nodes in patients with impalpable nodes. We assume that these patients would otherwise have gone in for a mediastinoscopy and even thoracotomy and lung resection. Skip metastases are known in lung cancer and if the results of mediastinoscopy had been negative, they would have fallaciously been staged as T1–3, N0–1 tumors and treated with lung resection. This would not only have contributed to incorrect staging but also increased the number of ‘futile thoracotomies’. Even assuming a low pick up rate of 4% (other studies show a higher pick up rate for cervical lymph nodes, probably because they included higher stages of disease also), it is an important investigation in the preoperative workup of patients with operable lung cancer as it is a non-invasive, zero morbidity, low-cost investigation and spares a proportion of patients from undergoing unnecessary major surgery. As the authors rightly point out, we have readily accepted other investigations in metastatic workup (CT scans of liver and adrenals, brain) with much less yield. We would also like to emphasize that if radiological suspicion of metastasis is high, a formal lymph node biopsy should be done even if the fine needle aspiration cytology is negative. Though it is fortuitous that the procedure has proven to be cost effective also, it should be strongly recommended in the routine preoperative workup even if it is marginally cost-ineffective, as it is difficult to put a price on the morbidity of a futile thoracotomy! Cervical ultrasonography would be even more useful in stage IIIA disease as it would not only obviate the need for an unnecessary mediastinoscopy, but it would also reclassify these patients as stage IIIB, which is usually treated with chemoradiotherapy rather than with neoadjuvant chemotherapy followed by surgery. We note that the study accrued patients between August 1997 and November 1998 and wonder whether the results of this study have prompted the incorporation of cervical ultrasound as part of the routine workup of patients with operable lung cancer in the authors’ unit. If so, it would be interesting to know further updated results of the procedure with larger numbers. We have started performing routine ultrasonography of the neck in patients staged I to IIIA prior to surgical resection and/or mediastinoscopy and will be able to add to the database on occult N3 disease in otherwise operable NSCLC.