Recurrence of pulmonary carcinoid tumors after resection: implications for postoperative surveillance.

BACKGROUND: The current guidelines for follow-up care after treatment of non-small cell lung cancer recommend continued surveillance for detection of recurrent or metachronous disease. However, carcinoid tumors, especially those with a typical histologic profile, tend to be less aggressive. Our goal was to determine the patterns of relapse and the manner of detection of recurrences, to guide follow-up care after resection.

METHODS: Patients who underwent operations for pulmonary carcinoids at our institution were identified from a prospectively maintained database, and their medical records were reviewed for relapse patterns, detection methods, and outcomes.

RESULTS: A total of 337 patients who underwent resection between 1993 and 2010 were included, with a median follow-up time of 3.5 years. Typical and atypical carcinoids were present in 291 (86%) and 46 (14%) patients, respectively. Recurrences occurred in 21 patients (6%), with distant metastases in 20 patients (95%) and locoregional recurrence in only 1 patient. Most recurrences (15 [76%]) were not detected through scheduled surveillance imaging but after the presentation of symptoms (7 [33%]) or incidentally by studies performed for unrelated reasons (8 [38%]). The risk of recurrence increased with positive lymph nodes and atypical histologic type. Only 9 of 291 patients (3%) with typical carcinoids experienced recurrences, with a median time to recurrence of 4 years (range, 0.8-12 years). Conversely, 12 of 46 patients (26%) with atypical carcinoids experienced recurrences, with a median time to recurrence of 1.8 years (range, 0.2-7 years).

CONCLUSIONS: After complete resection, scheduled surveillance imaging failed to detect most recurrences. Recurrence was rare in patients with node-negative typical carcinoids. Given the low risk of recurrence and the unclear efficacy of surveillance imaging, routine surveillance imaging may not be warranted in this cohort.