背景：良性甲状腺结节射频消融（RFA）后需要更长时间的随访，以了解再生以及延迟手术和长期并发症的其他原因。方法：这项回顾性研究纳入了 2007 年 3 月至 2010 年 12 月期间因症状性良性无功能甲状腺结节接受 RFA 治疗的连续患者。RFA 是根据标准方案进行的。我们在第 1、6 和 12 个月对患者进行随访，然后每年进行一次随访，直至 2022 年 8 月，并计算每次随访时的体积缩小率 (VRR)。我们根据三个已发表的标准、延迟手术和并发症评估了再生的发生率。使用Kaplan-Meier法评估再生的累积发生率，并进行单变量和多变量Cox回归分析以确定再生的危险因素。结果：本研究包括 421 名患者（平均年龄，47 ± 13 岁；372 名女性），有 456 个结节（平均体积，21 ± 23 mL）。中位随访期为 90 个月（四分位数范围，24-143 个月）。 2 年时的平均 VRR 为 81%，5 年时为 90%，≥10 年时为 94%。 12% (53/456) 的结节出现总体再生，并通过重复 RFA (n = 33) 或手术 (n = 4) 或留观观察 (n = 16) 进行治疗。与初始体积 <10 mL 的甲状腺结节相比，初始体积≥20 mL 的甲状腺结节再生风险显着更高（风险比，2.315 [95% 置信区间，1.183-4.530]；多变量 Cox 回归分析 p = 0.014）。 6% (26/421) 的患者因再生和/或持续症状 (n = 4) 或新发现的甲状腺肿瘤 (n = 22)（其中 1 例为良性，21 例为恶性）而延迟手术。总体并发症发生率为 2.4% (10/421)，无手术相关死亡或长期并发症。结论：RFA治疗良性甲状腺结节安全有效，长期随访VRR较高。由于消融的结节可能会重新生长以及某些患者需要延迟手术，因此在初步成功后需要定期随访。

Background: Longer follow-up after radiofrequency ablation (RFA) of benign thyroid nodules is needed to understand regrowth and other causes of delayed surgery and long-term complications. Methods: This retrospective study included consecutive patients treated with RFA for symptomatic benign nonfunctioning thyroid nodules between March 2007 and December 2010. RFA was performed according to the standard protocol. We followed up patients at 1, 6, and 12 months, then yearly, until August 2022, and calculated the volume reduction ratio (VRR) at each follow-up. We assessed the incidence of regrowth according to three published criteria, delayed surgery, and complications. The Kaplan-Meier method was used to evaluate the cumulative incidence of regrowth, and univariable and multivariable Cox regression analyses were performed to identify risk factors for regrowth. Results: This study included 421 patients (mean age, 47 ± 13 years; 372 women) with 456 nodules (mean volume, 21 ± 23 mL). The median follow-up period was 90 months (interquartile range, 24-143 months). The mean VRR was 81% at 2 years, 90% at 5 years, and 94% at ≥10 years. Overall regrowth was noted in 12% (53/456) of nodules and was treated with repeat RFA (n = 33) or surgery (n = 4) or left under observation (n = 16). Thyroid nodules with ≥20 mL initial volume had significantly higher risk of regrowth compared with nodules with <10 mL initial volume (hazard ratio, 2.315 [95% confidence interval, 1.183-4.530]; p = 0.014 on multivariable Cox regression analysis). Delayed surgery was performed in 6% (26/421) of patients because of regrowth and/or persistent symptoms (n = 4) or newly detected thyroid tumors (n = 22), one benign and 21 malignant. The overall complication rate was 2.4% (10/421), with no procedure-related deaths or long-term complications. Conclusion: RFA is safe and effective for treating benign thyroid nodules, with a high VRR at long-term follow-up. Regular follow-up after initial success is warranted because of the possibility of regrowth of ablated nodules and the need for delayed surgery in some patients.