【作者】 赵群仔；

【导师】 黄韬；

【作者基本信息】 华中科技大学 ， 外科学， 2013， 博士

【摘要】 第一部分甲状腺乳头状微小癌多灶性与颈淋巴结转移风险的系统综述及荟萃分析系统检索甲状腺乳头状微小癌的相关研究，对多灶性及颈淋巴结转移进行荟萃分析，并采用Mantel-Haen-szel法进行meta分析，计算总的相对危险度，以系统评价甲状腺乳头状微小癌多灶性与颈淋巴结转移风险的相关性。共14个临床研究3549名患者被纳入meta分析，所有患者均接受了颈淋巴结清扫术。与单灶微小癌相比，多灶者淋巴结转移风险显著增高，合并比值比为1.7，95%可信区间为1.1-2.7。按手术方式进行亚组分析后，研究间异质性显著降低。因而得出结论多灶甲状腺微小癌更易发生颈淋巴结转移，这部分病人可能从预防性颈淋巴结清扫获益。对于临床试验，不同的干预措施是造成异质性的主要原因。第二部分回顾性分析发现甲状腺乳头状微小癌颈淋巴结转移风险新的独立预测指标：多灶性联合癌灶总直径通过回顾性分析武汉协和医院2003-2012年所有接受手术治疗的甲状腺恶性肿瘤病例的临床病理特征，找出甲状腺乳头状微小癌颈淋巴结转移风险因素。此研究期间共有1456例甲状腺癌患者在该院接受手术治疗，选取其中初次手术且手术方式为甲状腺全切+中央区淋巴结清扫、术后病理证实为甲状腺乳头状微小癌病例，深入分析其临床病理特征。对于多灶性疾病，将其所有病灶最大径相加作为其癌灶总直径，并将癌灶总直径作为该病例的病理特征纳入分析。结果发现年龄、性别、多灶性、包膜侵犯、局部浸润均是甲状腺乳头状微小癌中央区淋巴结转移的独立危险因素；纳入癌灶总直径这一新的变量后，癌灶总直径大于1的微小癌颈淋巴结转移率明显高于单灶微小癌（60.4%vs.30%, P<0.001）；癌灶总直径<1的多灶微小癌与单灶微小癌颈淋巴结转移风险无显著性差异（37.5%vs.30%, P=0.463）。我们的研究显示多灶性联合癌灶总直径作为甲状腺微小癌危险度的分级指标更有意义；癌灶总直径>1cm的甲状腺微小癌淋巴结转移风险明显增高，我们建议对这部分病人考虑行预防性颈淋巴结清扫。第三部分统计学方法建立甲状腺微小癌颈淋巴结转移风险预测模型在甲状腺的临床诊治工作中，已有数个甲状腺癌风险分级和预后评估系统因其简单实用直观易于理解而深受临床医生和患者欢迎。然而尚没有一个简单的评分系统来指导手术方式的选择。我们期望可以建立一个简单易用的评分系统帮助临床医生评估微小癌患者颈淋巴结转移风险，从而选择哪些患者适合常规进行颈淋巴结清扫。通过回顾性分析我院2003-2012年以来所有初次手术且术后常规病理证实为微小癌的病例特点，找出影响颈淋巴结转移相关因素，统计学方法建立模型，以期在术前通过患者一般特征、影像学检查联合术中所见评估患者淋巴结转移风险，帮助术者选择手术方式。基于第二部分的发现，此部分纳入癌灶总直径这一新的临床特征，选择病例年龄、性别、癌灶最大径、癌灶总直径、多灶性共5项术前易评估的因素，通过计算其权重及对模型的贡献值赋予风险得分，纳入最终模型。数学方法评估模型的解释度。该评分系统结果易于解释，有利于临床医生预测评估微小癌患者就诊时发生颈淋巴结转移风险，对颈淋巴结清扫术的选择提供指导。第四部分甲状腺乳头状微小癌颈淋巴结转移风险预测模型的前瞻性验证研究我们通过前述回顾性研究建立了甲状腺乳头状微小癌颈淋巴结转移风险的预测模型。为了验证此模型的稳定性及准确度，于2012年12月在武汉协和医院甲状腺乳腺外科开展前瞻性验证研究。对于因甲状腺微小癌就诊于我科的初治患者，根据其临床病理特征评估其发生颈淋巴结转移风险，对于评估为中、高危患者，给予预防性颈淋巴结清扫术，术后记录其临床病理特征，与术前评分比较，结果发现评分越高者，颈淋巴结转移人数比例越高，定量评估系统评估为高危的患者，即得分>6分的患者，90%以上的病例发生了颈淋巴结转移；而对于评估为中危的患者，即得分4-6分的微小癌患者，40%左右的病例发生了颈淋巴结转移，两者间有显著统计学差异。通过前瞻性验证研究，我们发现甲状腺乳头状微小癌颈淋巴结转移风险预测模型可以帮助临床医生术前、术中评估甲状腺微小癌个体发生颈淋巴结转移的风险，以选择最佳手术方式。更多还原

【Abstract】 Part1Association between multifocality and cervical lymph node metastasis inpapillary thyroid microcarcinoma: a systematic review and meta-analysisStudies have shown multifocality significantly increases lymph node metastasis risk inpapillary thyroid carcinoma, but the association in microcarcinoma is controversial. This isa systematic review and meta-analysis of the role of multifocality in papillary thyroidmicrocarcinoma. A systematic search was performed in databases to identify all originalarticles employing the association between multifocality and lymph node metastasis amongpapillary thyroid microcarcinomas. Eligible studies were carefully reviewed and analyzed.A meta-analysis of fourteen publications with a total3549papillary thyroidmicrocarcinomas demonstrated that multifocality was significantly associated with LNMrisk with an odds ratio of1.7(95%confidence interval,1.1-2.7). Patinets who havemultifocal papillary thyroid microcarcinoma are at high risk of lymph node metastasis.Total thyroidectomy associated with routine central neck dissection may be the best surgicalapproach.Part2Risk for cervical lymph node metastasis among papillary thyroidmicrocarcinomaTo identify high-risk papillary thyroid microcarcinoma (PMC) inclined to lymph nodemetastasis (LNM), we conducted this hospital-based retrospective study. Patients whounderwent total thyroidectomy with central neck dissection and had a pathologicaldiagnosis of PMC between2003and2010at Wuhan Union Hospital were included in thisstudy. The frequency of LNM was retrospectively analyzed according to theclinicopathological features. For multifocal lesions, total tumor diameter was calculated asthe sum of the maximal diameter of each lesion. Age, gender, multifocality, capsularinvasion and local infiltration were independent factors that associated with cervical LNMin PMC. The proportion of LNM was similar between multifocal PMCs with total tumordiameter less than or equal to1cm and unifocal tumors with diameter less than or equal to1cm (37.5vs.30%, P=0.463). LNM frequency was56.89%among multifocal PMC withtotal tumor diameter greater than1cm but less than2cm, and64.9%among unifocaltumors with diameter greater than1cm but less than2cm with no significant difference between the two groups (P=0.330). However, LNM frequency was significantly higher inmultifocal PMCs with total tumor diameter greater than1cm than unifocal tumors withdiameter less than or equal to1cm (60.4vs.30%, P<0.001). Our study indicated that it wasmultifocality along with total tumor diameter that significantly affected LNM risk in PMCpatients.Part3A reliable and easy-to-use scoring system facilitating selection of prophylacticcentral neck dissection among papillary thyroid microcarcinomas based on theretrospective studyNumerous classifications have been developed to determine prognositic factors forpatients with papillary thyroid carcinoma, including the TNM (tumor, node, metastasis)classification by the International Union Against Cancer, the AGES (age, grade, extent, andsize) classification by Mayor’s clinic, and EORTC prognostic index by the EuropeanOrganization for Research and Treatment of Cancer. We conducted this study to providetables that allow thyroid surgeons to easily calculate a thyroid microcarcinoma patient’srisks of cervical lymph node metastasis preoperatively. A combined analysis was carried outof individual patient data from437papillary thyroid microcarcinoma patients whounderwent thyroid surgery for papillary thyroid microcarcinoma during2003-2012inWuhan Union Hospital. A simple scoring system was derived based on5clinical andpathological factors: sex, age, number of tumors, the maximum tumor size, total tumordiameter. The probabilities of developing cervical lymph node metastasis from less than10%to almost100%among groups based on the score. With the scoring system, the urologistcan discuss the different options with the patient to determine the most appropriatetreatment.Part4A reliable and easy-to-use scoring system facilitating selection of prophylacticcentral neck dissection among papillary thyroid microcarcinomas: a prospective studyWe conducted this prospective study to evaluate the above scoring system forpredicting cervical lymph node metastasis among papillary thyroid microcarcinoma. In thisprospecive study, we assessed104patients with papillary thyroid micocarcinoma and todetermine the predictive accuracy of the above scoring system applicable to the study population. With regard to cervical lymph node metastasis, the difference between therespective stages and/or risk groups was highly significant. The probability to developcervical lymph node metastasis among high risk group, namely those scoring more than6was more than90%. However, the probability to develop cervical lymph node metastasisamong middle risk group, namely those scoring more than4but less than6, was around40%. Moreover, among each risk group, the probability to develop cervical lymph nodemetastasis was increased along with increasing score. In conclusion, the scoring system forpredicting cervical lymph node metastasis among papillary thyroid microcarcinoma isreliable and accurate to facilitate thyroid surgeons choosing the most appropriate surgicalprocedure. Moreover, the individual factors such as age, sex, tumor size are easy to obtainand generally available.更多还原