Tumor spread through air spaces (STAS) is an important prognostic factor for lung cancer, including lung neuroendocrine tumors (NETs). The comparative oncological efficacy of lobar resection (LR) and sub-lobar resection (SR) for clinical stage IA (cIA) STAS-positive NETs remains unclear. This retrospective study aimed to review outcomes in patients with consecutive peripheral cIA (tumors <= 2 cm) STAS-positive NET patients (excluding small cell lung cancer) who underwent surgery between 2012 and 2020 at six high-volume tertiary Chinese institutions. SR was commonly performed for patients ineligible for LR for various reasons. The primary endpoint was recurrence-free survival (RFS), while the secondary endpoint was overall survival (OS). Propensity score matching (PSM) was utilized to minimize selection bias. Among the 599 eligible patients, PSM resulted in 328 LR and 164 SR cases with well-balanced baseline characteristics. During a median follow-up of 6.8 years, LR significantly improved 5-year RFS (68.3% versus 54.3%, hazard ratio (HR) [95% confidence interval (CI)] = 0.595 [0.449-0.787], P < 0.001) and OS (79.0% versus 72.6%, HR [95%CI] = 0.650 [0.484-0.872], P = 0.025), and reduced postoperative recurrence (34.5% versus 51.2%, P < 0.001) and mortality (32.3% versus 46.3%, P = 0.002) rates compared to SR. Subgroup analysis revealed that LR improved RFS (59.2% versus 42.5%, HR [95%CI] = 0.553 [0.408-0.749], P < 0.001) and OS (74.0% versus 63.7%, HR [95%CI] = 0.710 [0.509-0.989], P = 0.043) compared to SR in patients with large cell neuroendocrine carcinoma (LCNEC), while both approaches yielded comparable prognoses in patients with carcinoid. Furthermore, segmentectomy exhibited survival outcomes comparable to those of wedge resection. Collectively, LR demonstrated better survival outcomes than SR for cIA STAS-positive NETs, with the benefit being influenced by the tumor histology. LCNEC was associated with a 3.3-fold higher risk of recurrence and a 4.0-fold higher risk of mortality than carcinoid.