摘要目的:探讨喉返神经受累的甲状腺乳头状癌患者客观声学等临床特征及术后复发的危险因素。方法:回顾性分析首都医科大学附属北京同仁医院2006年1月至2019年12月喉返神经受累的甲状腺乳头状癌患者资料，对客观声学指标进行亚组比较。采用Kaplan-Meier法计算复发率和总生存率，采用单因素及多因素Cox回归模型分析术后复发的危险因素。结果:共纳入150例喉返神经受累患者，男48例（32.0%），女102例（68.0%），年龄（53.5 ±13.7）岁，≥55岁患者62例（41.3%）。Ⅰ期88例，Ⅲ期62例。55例患者术前即存在声带麻痹。肿物或淋巴结与喉返神经粘连75例，直接侵犯75例。客观声学指标比较结果显示，喉返神经受累的患者jitter值高于喉返神经未受累的患者［2.3%（1.4%，3.2%）比1.8%（0.8%，2.6%）， P<0.001］；术前存在声带麻痹的患者jitter值［3.1%（2.2%，4.6%）比2.0%（1.1%，2.8%）， P<0.001］及shimmer值［7.1%（4.9%，9.9%）比5.5%（4.2%，7.3%）， P<0.001］高于声带活动正常的患者，最长发音时间（MPT）短于声带活动正常的患者［8.0（6.0，11.0）s比12.0（10.0，15.3）s， P<0.001］；喉返神经粘连和喉返神经侵犯患者客观声学指标差异均无统计学意义（均 P>0.05）。随访时间12~196（65.0±35.9）个月。共16例（10.7%）患者随访期间出现了肿瘤复发或转移，8例（5.3%）患者因甲状腺癌复发或转移死亡。5年总生存率为95.1%，10年总生存率为92.8%；5年无复发生存率为88.9%，10年无复发生存率为86.2%。单因素Cox回归分析结果显示，发病年龄≥55岁、术前喉返神经麻痹、合并喉气管食管结构受累是患者术后复发的危险因素（均 P<0.05）。多因素Cox回归分析结果显示，发病年龄≥55岁（ OR=1.060，95% CI：1.011~1.110， P=0.015）是患者术后复发的危险因素。 结论:发病年龄≥55岁是甲状腺乳头状癌喉返神经受累患者术后复发的危险因素；术前客观声学指标评估可对喉返神经功能状态提供一定的参考。

abstractsObjective:To investigate the clinical characteristics and risk factors of postoperative recurrence in papillary thyroid carcinoma (PTC) patients with recurrent laryngeal nerve (RLN) invasion.Methods:The data of PTC patients with recurrent laryngeal nerve invasion treated in Beijing Tongren Hospital, Capital Medical University from January 2006 to December 2019 were retrospectively analyzed. The acoustic parameters were compared between different subgroups. Kaplan-Meier method was used to calculate the overall survival (OS) and the recurrence-free rate (RFS), and univariate and multivariate Cox regression analyses were performed to determine the risk factors for postoperative recurrence.Results:A total of 150 PTC patients were enrolled in the final analysis, including 102 females and 48 males, with an average age of (53.5±13.7) years, and 62 patients (41.3%) aged over 55 years. There were 88 cases with stage Ⅰ, and 62 cases with stage Ⅲ. Fifty-five patients presented with preoperative vocal cord paralysis. There were 75 cases appearing adhesion between tumor or lymph node and recurrent laryngeal nerve while 75 cases presented with direct invasion. The comparisons of acoustic parameters showed that patients with RLN invasion had higher jitter compared with patients without RLN invasion [2.3% (1.4%, 3.2%) vs 1.8% (0.8%, 2.6%), P<0.001]. Moreover, patients with preoperative vocal cord paralysis (VCP) had higher jitter[3.1% (2.2%, 4.6%) vs 2.0% (1.1%, 2.8%), P<0.001] and shimmer [7.1% (4.9%, 9.9%) vs 5.5% (4.2%, 7.3%), P<0.001] and shorter maximum phonation time (MPT) [8.0 (6.0, 10.0) s vs 12.0 (10.0, 15.3) s, P<0.001] compared with patients without preoperative VCP. However, there was no statistical difference in acoustic parameters between cases with RLN adhesion and RLN invasion (all P>0.05). Postoperative follow-up time ranged between 12-196 months, with an average of (65.0±35.9) months. Sixteen patients (10.7%) had recurrence or metastasis, and 8 cases (5.3%) died of recurrence or metastasis. The 5-year OS rate was 95.1%, and the 10-year OS rate was 92.8%. The 5-year RFS rate was 88.9%, and the 10-year RFS rate was 86.2%. Univariate Cox analysis showed that age of onset ≥ 55 years, preoperative recurrent laryngeal nerve palsy, laryngeal, trachea or esophageal invasion were the risk factors for postoperative recurrence of PTC with RLN invasion (all P<0.05). Multivariate Cox analysis showed that age of onset ≥ 55 years ( OR=1.060, 95% CI: 1.011-1.110, P=0.015) was an independent risk factor. Conclusions:Age of onset ≥ 55 years is an independent risk factor for postoperative recurrence in PTC patients with RLN invasion. Preoperative acoustic parameters may provide reference for evaluation of RLN function.