The effects of the type of anesthesia (spinal (SA) vs. general (GA)) used for transurethral resection of bladder tumor (TURBT) on non-muscle invasive bladder cancer (NMIBC) recurrence and progression are controversial and our aim is to investigate their associations. We identified 300 NMIBC patients who underwent initial TURBT with SA or GA. Cox's regression analysis was performed to examine the effects of anesthesia on tumor recurrence. Among 300 patients, 153 (51.0%) received GA and 147 (49.0%) SA. The 5-year recurrence-free survival (RFS) rate was 59.9% in the GA group, which was significantly lower than that in the SA group (74.4%, p = 0.029). GA (HR 1.57, p = 0.048), male sex (HR 2.72, p = 0.012), and tumor multiplicity (HR 1.96, p = 0.006) were independently associated with tumor recurrence. In a subgroup of 137 patients with high-risk NMIBC, the 5-year RFS rate was 50.3% in the GA group, which was significantly lower than that in the SA group (77.6%, p = 0.020), and GA remained an independent indicator of tumor recurrence (HR 2.35, p = 0.016). However, no significant differences were observed in the RFS rates of low- to intermediate-risk NMIBC patients between the GA and SA groups. The RFS rate was lower in NMIBC patients who received GA during TURBT than in those who received SA. Volatile anesthesia may increase tumor recurrence, particularly in high-risk NMIBC patients, which may be due to the inhibition of the immune response system during surgery.