【作者】 郑冬梅；

【导师】 赵家军；

【作者基本信息】 山东大学 ， 内分泌与代谢病， 2009， 博士

【摘要】 甲状腺结节是临床常见疾病,通过触诊发现一般人群的患病率为4%～7%;随着高清晰度B超的普及,甲状腺结节的检出率可达20%～67%。甲状腺结节多为良性,恶性仅占5%。绝大多数恶性结节需首选手术治疗,而对于大多数甲状腺良性结节,仅需定期随访即可。因此准确判断结节性质对于指导结节的治疗至关重要。详细的病史和体格检查是诊断甲状腺结节的基础。超声检查能通过特定的超声表现鉴别甲状腺结节的良恶性,尤其是几种超声特征结合能进一步提高诊断准确性。术中准确的冰冻切片病理诊断对于防止甲状腺癌的漏诊以及确定手术范围有重要意义。术后石蜡病理诊断目前是诊断甲状腺结节的“金标准”,术前术中各种诊断方法均须与它作对比来判断准确性。目的本研究旨在通过总结甲状腺良恶性结节的临床特征、实验室检查、超声及病理特点,分析现有方法诊断甲状腺结节的临床价值。资料与方法1.对象回顾性分析山东省立医院2005年3月至2008年10月期间手术病理证实的甲状腺结节患者共1474例,根据术后石蜡病理切片结果分为良性组和恶性组。所有病例均经过临床检查,有1450例术前行二维超声检查和彩色多普勒血流成像,1460例有术中冰冻切片病理诊断,分别与术后石蜡切片病理诊断进行对比。2.方法测定甲状腺功能包括促甲状腺激素(TSH)、游离T3(FT3)、游离T4(FT4)、抗甲状腺球蛋白抗体(TgAb)和抗甲状腺过氧化物酶抗体(TPOAb),行二甲状腺超声检查,先用二维超声探查双侧甲状腺,观察结节的、大小、形态、边界、内部结构、回声、钙化、声晕、颈部淋巴结等声像图表现,然后用彩色多普勒(CDFI)观察结节内部及其周边的血流丰富程度,有无紊乱。术中取标本制备冰冻切片,术后病理为石蜡封面,苏木红—伊红染色,经有经验的病理医师诊断,以石蜡切片病理为最终确诊结果。结果(1)一般情况1474例患者中男286例,女1188例,男女性别比1:4.2。良性结节占77.40%(1141/1474),恶性者占22.60%(333/1474);男性患者结节的恶性率高于女性患者(28.3%vs 21.3%,P＜0.05)。良性结节中患者的年龄为(48.46±11.11)岁,恶性结节中患者的年龄为(43.27±12.67岁),恶性结节组的年龄小于良性组(P＜0.05)。临床检查中,恶性组声音嘶哑发生率明显高于良性组(6.3%vs 2.1%,P＜0.05);触诊结节质地硬、不活动,在良恶性组间差异有统计学意义(P＜0.001)。(2)结节类型良性结节患者中,结节性甲状腺肿1034例(90.6%),腺瘤65例,桥本病35例,其他包括肉芽肿性甲状腺炎等7例。恶性结节患者中,乳头状癌295例(88.6%)、滤泡癌17例、髓样癌12例、未分化癌2例,其他7例。333例甲状腺癌患者中,有117例为结节性甲状腺肿合并甲状腺癌(占全部结节性甲状腺肿病例的10.2%),16例为慢性淋巴细胞性甲状腺炎合并甲状腺癌(占全部慢性淋巴细胞性甲状腺炎患者的31.3%)。(3)实验室检查FT3、FT4在良恶性组患者中无统计学差异(P均＞0.05),TSH恶性组(2.20±0.20)IU/mL,高于良性组(1.76±0.08)IU/mL,有统计学差异(P＜0.05)。当TSH≥5.0IU/mL,DTC的发生率为34.3%(11/32)明显高于TSH＜0.40 IU/mL(16.4%(11/67),P＜0.05)和TSH在0.40-1.89IU/mL时(16.5%(61/368),P＜0.05)。即便TSH在正常范围(0.4-5.0IU/mL),当TSH升高时,发生DTC的危险性增加。即TSH在(1.90-2.49 IU/mL)和(2.50-4.99IU/mL)的DTC发生率分别为21.6%(16/74)和26.7%(31/116),均高于TSH在0.40-1.89 IU/mL(16.5%)时,(P=0.19和P=0.02)。恶性组TgAb、TPOAb均高于良性组[TgAb:(413.30±168.09)IU/mL vs(93.52±44.23)IU/mL;TPOAb:(141.56±45.22)IU/mL vs(49.23±15.53)IU/mL,P均＜0.001)](4)甲状腺结节的超声特征共1450例患者接受了超声检查,有1125例病理证实为恶性,325例为恶性。B超示单结节442例,多结节1008例。单结节、多发结节的甲状腺癌发生率分别为33.1%和27.6%,两者间差异无统计学意义(P=0.287)。恶性结节的最大直径为1.93±1.10cm,明显低于良性结节2.82±1.17 cm(P＜0.001)。甲状腺良、恶性结节在边缘、内部结构、回声类型、内部微钙化及血流分布等方面差异有统计学意义;而周边缺乏声晕、内部粗钙化在良、恶性结节中的差异无统计学意义。恶性结节B超表现为形态不规则139例,占恶性结节的42.%;良性结节形态不规则184例,占良性病例的16.4%,两组差别具有统计学意义(P＜0.001)。边界模糊在良恶性病例中的分布也有显著性差异(P＜0.001)。良性结节表现为实性结构的有792例,占良性病例的70.4%;恶性结节表现为实性结构的有89.2%,两组间虽有显著性差异,但实性结构诊断恶性病变的特异性较低,仅34.9%,阳性预测值仅27.1%。低回声在良恶性结节中的分布也有显著性差异,(29.0%vs 80.3%,P＜0.001),其诊断恶性甲状腺肿瘤的特异性为71.0%。如同时具有实性低回声的B超征象,则诊断甲状腺恶性结节的特异性提高到91.5%。超声共检出钙化489例,总钙化率为33.7%,恶性结节的钙化率为70.4%,微钙化253例,粗钙化236例,其中微钙化出现在良性组78例,良性组的微钙化率为24.0%。在恶性组174例,恶性组的微钙化率为53.5%,微钙化中有51例为沙砾样钙化,全部出现在乳头状甲状腺癌中。有9例针尖或颗粒状微钙化出现在滤泡癌或甲状腺髓样癌患者中。微钙化诊断甲状腺癌的特异性为92.1%,阳性预测值为66.4%,准确率为83.5%。良性病例中检出粗钙化182例,恶性病例中检出54例,粗钙化在两组间的分布无显著性差异(P=0.27)。结节内血流丰富并紊乱是恶性结节的征象之一。本组325例有243例呈现结节内血流。其诊断恶性结节的敏感性为74.7%,特异性为60.9%。良性组有35例B超发现颈部淋巴结,而恶性组有200例,合并颈部淋巴结肿大的病例占甲状腺恶性结节组病例的61.5%,明显多于良性组(P＜0.001)。颈部淋巴结诊断甲状腺癌的特异性为96.8%,阳性预测值为78.1%,准确率为83.8%。恶性结节的平均最大直径为1.93cm,明显低于良性结节2.82 cm(P＜0.001)。甲状腺良、恶性结节在边缘、内部结构、回声类型、内部微钙化及血流分布等方面差异有统计学意义;而周边缺乏声晕、内部粗钙化在良、恶性结节中的差异无统计学意义。(5)病理学检查有1460例患者有冰冻切片结果。将其与石蜡切片进行对比,结果如下:1141例良性结节患者中,结节性甲状腺肿1034例(90.6%),腺瘤65例(5.7%),桥本病35例,其他包括肉芽肿性甲状腺炎等7例。恶性结节患者中,乳头状癌295例(占全部恶性结节的88.6%)、滤泡癌17例(5.1%)、髓样癌12例、未分化癌2例,其他7例。333例甲状腺癌患者中,有117例同时合并结节性甲状腺肿,占全部结节性甲状腺肿病例的10.2%(石蜡切片显示有结节性甲状腺肿病理改变的病例,共1147例),16例为慢性淋巴细胞性甲状腺炎合并甲状腺癌,占全部慢性淋巴细胞性甲状腺炎患者的31.3%(石蜡切片显示有慢性淋巴细胞性甲状腺炎病理改变的病例,共51例)。1460例患者,冰冻切片确诊1402例,确诊率为96.0%。冰冻切片确诊甲状腺恶性病变298例,其敏感性为91.9%,恶性病变延迟诊断19例,冰冻延迟率为5.9%(19/324)。合计误诊7例,误诊率为0.5%。7例误诊病例有6例为结节性甲状腺肿合并微小乳头状癌,1例为结节性甲状腺肿合并微小滤泡乳头状癌。冰冻延迟诊断的19例恶性病例中,16例为乳头状癌(其中13例为微小癌),1例为滤泡癌,2例为髓样癌。石蜡切片共发现微小癌67例,有26例发生了淋巴结转移,转移率为38.8%。直径＞1cm的甲状腺癌淋巴结转移率为50%,两者间差异具有统计学意义(P＜0.05)。结论甲状腺结节以女性高发,但男性恶性率高。超声检查中微钙化和颈部淋巴结肿大等特征对于恶性结节的诊断具有较高的特异性,其他如实性低回声、边界模糊、形态不规则及结节内血流丰富可提示恶性,几种超声特征结合有助于鉴别甲状腺结节的良恶性。术中冰冻检查对于明确甲状腺结节的良、恶性和术式的选择具有关键性作用,是一种较可靠的检查手段。良恶性组织并存在甲状腺结节中较常见,微小癌在甲状腺结节中并不少见,临床应重视,防止漏诊。更多还原

【Abstract】 Background:Thyroid nodules are now a common clinical problem.The prevalence of thyroid nodules was estimated as 4-7%in general population on the basis of palpation.Since ultrasonography was introduced for clinical application,the prevalence of thyroid nodules has increased to 20-67%and approached 50%based on data reported in autopay.Althrough only around 5%of the clinically identified nodules are malignant, the main concern in evaluation of thyroid nodules is the verification of whether a lesion is benign of malignant,as for benign nodules,only conservative treatment is needed, but for malignancies,more aggressive treatment is needed.History and physical examination,laboratory evahaation are the basis of diagnosing thyroid nodules.Thyroid ultrasonography can detect nonpalpable nodules,differentiate simple cysts from solid nodules,determine the size of nodules during follow-up,and facilitate FNAB.Additionally,the noninvasiveness of the procedure has increased clinical application of this method for the evaluation of thyroid nodules.Intraoperative Frozen Section has value in diagnosing thyroid nodules,and is capable of determation of the extent of thyroidectomy.Paraffin section is the final pathology and is served as "gold standard" for diagnosing thyroid nodules.Pre- and intraoperative assement must be compared with paraffin section.Objective:To analyze the clinical and pathological characteristics of thyroid nodules and to evaluate the value of the present methods for diagnosing thyroid nodules.Research Design and Methods:SubjectsData from 1,474 consecutive patients,who underwent a thyroidectomy for nodular thyroid disease between March 2005 and October 2008 in the Shandong Provincial Hospital,were studied retrospectively.Subjects were 286 male and 1188 female patients(with a male-to-female ratio of 1:4.2),and had a mean age of(45.5±11.1) years.1450 of the 1474 patients were analyzed by US before surgery,others were not included in the study because of incomplete ultrasonographical information of their thyroid nodules or the absence of the US examination.Clinical evaluationHistory and physical examination such as family history,age,gender,duriation of thyroid nodules,symptom,careful thyroid palpation are made.Measurement of serum TSH,FT3,FT4,TPOAb and TGAb is also needed.UltrasonographyComplete high-resolution ultrasonographical information before surgery was collected from 1450 of the 1474 patients.All scans were performed by experienced radiologists with ultrasonographical scanners(IU 22 or Sonos 4500;Philips,Bothell, Wash.,USA) equipped with a 5- to 12-MHz linear-array transducer.Records for each patient included the number,size,location,morphology,boundaries and echogenic patterns of the nodules,the presence and patterns of calcification,as well as the internal blood flow and blood flow at the perimeter of the nodules.Both hyperechoic structures with acoustic shadowing(coarse calcifications) and very bright echoes without shadowing(microcalcifications) were considered evidence of calcifications.Pathology1460 patients had a intraoperative frozen section.All 1474 patients had a conclusive pathological diagnosis after surgery.A detailed gross description was made of all thyroid specimens before they were fixed in 10%neutral formalin.Tissue sections of 4 mm thickness were stained with HE.The pathological characteristics of all the nodules were confirmed by senior pathologists.Patients with a benign lesion accompanying the cancer were included in the malignant group.Results(1) There were 286 males and 1188 females.The ratio of male to female was 1:4.2. 1141(77.40%) patients were diagnosed with benign nodules and 333(22.60%) malignant nodules.The malignant rate was higher in males than in females(28.3%vs 21.3%,P＜0.05).The average age of malignant samples was younger than that of benign ones,the difference between which was significant(P＜0.05).(2) The mean TSH was 2.20±0.20mIU/L in patients with malignant nodules vs 1.76±0.08mIU/L in patients with benign nodules(P＜0.05)。The prevalence of malignance was 16.4%(11 of 67) when TSH was less than 0.40 mIU/L vs 34.3%(11 of 32) when TSH was 5.0mIU/L or greater(P＜0.05).Even in normal range,high rates of malignancy were found in patients with higher TSH levels.The prevalence of malignance were 21.6%(16 of 74) when TSH was between 1.90 and 2.49 mIU/L and 26.7%(31 Of 116) when TSH was between 2.50 and 4.99 mIU/L vs 16.5%(61 of 368) when TSH 0.40-1.89 mIU/L(P=0.34 and P=0.02,respectively).The mean TgAb and TPOAb were much higher in malignant nodules than that in benign nodules(TgAb:413.30±168.09 IU/mL vs 93.52±44.23IU/mL,P＜0.001; TPOAb:141.56±45.22 IU/mL vs 49.23±15.53 IU/mL,P＜0.001)(3) Most benign cases were diagnosed as nodular goiter(1034,90.6%),while most malignant nodules were diagnosed as papillary thyroid carcinoma(295, 88.6%)(4) The prevalence of thyroid cancer did not differ between patients with a solitary nodule and patients with multiple nodules(33.1%vs 27.6%,P=0.287).Those individual sonographic characteristics that showed a statistically significant difference in the malignant group were:ill defined margin,solid component,hypoechoic nodule, microcalcifications,internal blood flow and enlarged cervical lymph node(P＜0.001). There were no differences between the benign and malignant groups for absent of halo and coase calcifications(P＞0.05). (5) The accuracy,sensitivity,and disaccording rate of frozen section examinations was 96.0%,91.9%,0.5%,respectively.(6) Microcarcinoma was found in 68 patients by paraffin section examination,of which 26(38.8%) had lymph node metastases.Conclusions:(1) Thyroid nodules were found more often in women than in men,but the malignant rate was higher in men than in women.(2) The risk of malignancy in a thyroid nodule increases with higher serum TSH concentration.Even within normal ranges,a TSH level above the population mean is associated with greater llikelyhood of thyroid cancer than a TSH below the mean.(3) Ultrasonographic features could be helpful in differentiating benign and malignant nodules.Calcification detected by thyroid ultrasound represents a risk factor for malignancy,but is of limited use as a sole marker of malignancy.(4) Frozen section examination is of value in diagnosing thyroid nodules and play a key role in determination of extent of thyroidectomy.(5) Thyroid carcinoma often coexists with benign thyroid disease.Microcarcinoma can lead to misdiagnosis and disdiagnosis in ulrtasonograms and frozen section.更多还原