RESUMEN
OBJECTIVES: Typical carcinoids (TCs) are rare, slow-growing neoplasms, usually characterized by satisfactory surgical outcomes. Due to the rarity of TCs, international guidelines for the management of particular clinical presentations currently do not exist. In particular, non-anatomical resections (wedges) are sometimes advocated for Stage 1 TCs because of their indolent behaviour. The aim of this paper was to evaluate the most effective type of surgery for Stage 1 TCs, using the European Society of Thoracic Surgeons retrospective database of the Neuroendocrine Tumors of the Lung Working Group. METHODS: We analysed the effect of surgical procedure on the survival of patients with Stage 1 TCs. Overall survival (OS) was calculated from the date of intervention. The cumulative incidence of cause-specific death (tumour- and non-tumour-related) was also estimated. The impact of the surgical procedure (i.e. lobectomy vs segmentectomy vs wedge resection) on survival was investigated using the Cox model with shared frailty (for OS, accounting for the within-centre correlation) and the Fine and Gray model (for cause-specific mortality) using the approach based on the multinomial propensity score. Effects were estimated including in the model the logit-transformed propensity scores of segmentectomy and wedge resection as covariates. RESULTS: A total of 876 patients with Stage 1 TCs (569 women, 65%) were included in this study. The median age was 60 years (interquartile range 47-69). At the last follow-up, 66 patients had died: The 5-year OS rate was 94.3% [95% confidence interval (CI) 92.2-95.9]. The 5-year cumulative incidences of tumour- and non-tumour-related deaths were 2.4% (95% CI 1.4-3.9) and 3.9% (95% CI 2.5-5.6%), respectively. The analysis performed using the multinomial propensity score approach confirmed the significantly worse survival of patients treated with a wedge resection compared to those treated with a lobectomy (hazard ratio 2.01, 95% CI 1.09-3.69; P = 0.024). Similar effects of wedge resection are detectable for cause-specific deaths: tumour-related (hazard ratio 2.28, 95% CI 0.86-6.02; P = 0.096) and non-tumour-related (hazard ratio 1.74, 95% CI 0.89-3.40; P = 0.105). CONCLUSIONS: In a large cohort of patients, we were able to demonstrate the superiority of anatomical surgical resection in Stage 1 TCs in terms of OS. This result should therefore be considered for future clinical guidelines for the management of TCs.