超声和放射性核素扫描及针吸细胞学对甲状腺结节诊断的META分析
发布时间:2018-06-06 04:23
本文选题:甲状腺结节 + 超声 ; 参考:《广西医科大学》2015年硕士论文
【摘要】:目的:通过对超声、针吸细胞学、放射性核素扫描对甲状腺结节诊断的荟萃分析,探讨这三种检查的诊断价值。方法:计算机检索中国生物医学文献数据库(CBM)、中国知网(CNKI)、 PubMed、EMBASE、Cochrane Library databases数据库,时间截至2015年2月1日。严格按照纳入及排除标准对文献进行评价和筛选。使用Stata12.0 with Meta软件绘制漏斗图评价有无发表偏倚,使用meta-disc1.4软件进行阈值效应检验及合并分析,使用Chi-square、Cochran-Q及I-square检验评价异质性大小,通过计算加权灵敏度、特异度、阳性似然比、阴性似然比、诊断比值比及综合受试者综合特征曲线(SROC)下面积(AUC),评价超声、针吸细胞学、放射性核素扫描对于甲状腺结节的诊断价值。结果:纳入的2785例B超合并灵敏度(95%CI)为0.74(0.71-0.78),合并特异度(95%CI)为0.90(0.89-0.92),合并阳性似然比(95%CI)为6.84(3.69-12.66),合并阴性似然比(95%CI)为0.30(0.24-0.38),合并诊断比值比(95%CI)为24.00(11.44-50.33),AUC(SE)为0.8993(0.0266);针吸细胞学6854例,合并灵敏度(95%CI)为0.76(0.75-0.78),合并特异度(95%CI)为0.94(0.94-0.95),合并阳性似然比(95%CI)为20.37(11.14-37.24),合并阴性似然比(95%CI)为0.24(0.15-0.38),合并诊断比值比(95%CI)为84.81(51.40-139.95),AUC(SE)为0.9583(0.0113);放射性核素扫描(ECT)1462例,合并灵敏度(95%CI)为0.79(0.74-0.84),合并特异度(95%CI)为0.41(0.38-0.44),合并阳性似然比(95%CI)为1.21(1.02-1.43),合并阴性似然比(95%CI)为0.59(0.35-1.01),合并诊断比值比(95%CI)为2.29(1.08-4.82),AUC(SE)为0.6342(0.0456)。结论:1、针吸细胞学检查对甲状腺结节的诊断价值最高,B超次之,放射性核素扫描最差;2、针吸细胞学检查诊断甲状腺肿瘤准确度高,基本可明确甲状腺恶性肿瘤的诊断,B超、放射性核素扫描明显低于细针,不能依靠二者其中之一单独确定恶性肿瘤的诊断;3、针吸细胞学检查、B超、放射性核素扫描三者其中之一单独的阳性结果不能排除恶性肿瘤。
[Abstract]:Objective: to evaluate the diagnostic value of ultrasound, needle aspiration cytology and radionuclide scanning in the diagnosis of thyroid nodules. Methods: the database of Chinese biomedical literature was searched by computer. The database of Chinese biomedical literature (CBM), CNKI (China) and PubMedus (EMBASE) Cochrane Library (databases) were searched as of Feb 1, 2015. To evaluate and screen the literature strictly according to the criteria of inclusion and exclusion. Using Stata12.0 with Meta software to draw funnel chart to evaluate whether there is publication bias, using meta-disc1.4 software to test threshold effect and combining analysis, using Chi-square-Cochran-Q and I-square test to evaluate heterogeneity, calculating weighted sensitivity, specificity, positive likelihood ratio, and then using Chi-square-Cochran-Q and I-square test to evaluate heterogeneity. The negative likelihood ratio, diagnostic ratio and the area under the synthetic characteristic curve of SROC were evaluated to evaluate the diagnostic value of ultrasound, needle aspiration cytology and radionuclide scanning in thyroid nodules. Results: the sensitivity (95 CI) of 2785 cases with B-ultrasound was 0.740.71 -0.78, the combined specificity was 0.90 ~ 0.992, the positive likelihood ratio was 6.84 ~ 3.69-12.66, the negative likelihood ratio was 0.300.24-0.38, the combined diagnostic ratio was 24.000.11.44-50.33AUCSE0.89930.0266and the needle aspiration cytology 6854 cases, the combined diagnostic ratio was 24.0011.44-50.33AUCSE0.8993a 0.0266a, the negative likelihood ratio was 0.300.24-0.38, the combined diagnostic ratio was 24.000.11.44-50.33AUCSE0.966.The positive likelihood ratio was 6.84v 3.69-12.66%, the negative likelihood ratio was 0.300.24-0.38%, the combined diagnostic ratio was 24.0011.44-50.333AUCSE0.89930.266. The combined sensitivity was 0.760.5-0.78, the combined specificity was 0.94-0.95, the positive likelihood ratio was 20.3711.14-37.24, the combined negative likelihood ratio was 0.240.15-0.38, the combined diagnostic ratio was 84.81% (51.40-139.95 CI), the positive likelihood ratio was 0.958 3 ~ 0.0113, the combined likelihood ratio was 0.24 ~ 0.15-0.38, and the combined diagnostic ratio was 84.81% 51.40-139.95% AUCSE. The combined sensitivity was 0.790.74 ~ 0.84, the combined specificity was 0.41 ~ 0.38-0.44, the combined positive likelihood ratio was 1.21 ~ 1.02-1.43, the combined negative likelihood ratio was 0.590.35-1.01and the combined diagnostic ratio was 2.29 ~ 1.08-4.82AUCSE= 0.63420.0456. Conclusion the diagnostic value of needle aspiration cytology for thyroid nodule is the highest, followed by B ultrasound, and radionuclide scanning is the worst. The accuracy of needle aspiration cytology in the diagnosis of thyroid neoplasms is high. The radionuclide scan was obviously lower than that of the fine needle. The diagnosis of malignant tumor could not be determined separately by one of them. The positive results of one of them could not exclude the malignant tumor, such as needle aspiration cytology, B-ultrasound and radionuclide scanning.
【学位授予单位】:广西医科大学
【学位级别】:硕士
【学位授予年份】:2015
【分类号】:R581
【参考文献】
相关期刊论文 前5条
1 刘超;唐伟;;甲状腺结节和甲状腺癌的病因学和流行病学[J];中国实用内科杂志;2007年17期
2 屈云,何俐,刘鸣;Cochrane系统评价的基本方法[J];中国临床康复;2003年04期
3 姜军;齐晓伟;;甲状腺结节诊断方法合理选择及评价[J];中国实用外科杂志;2010年10期
4 陈万青;张思维;郑荣寿;雷正龙;李光琳;邹小农;赵平;;中国肿瘤登记地区2007年肿瘤发病和死亡分析[J];中国肿瘤;2011年03期
5 郑辉烈;王忠旭;王增珍;;Meta分析中发表偏倚的Begg’s检验、Egger’s检验及Macaskill’s检验的SAS程序实现[J];中国循证医学杂志;2009年08期
相关硕士学位论文 前1条
1 白卫兵;ECT和B超联合检查对判断甲状腺结节良恶性的价值研究[D];广西医科大学;2010年
,本文编号:1985071
本文链接:https://www.wllwen.com/yixuelunwen/nfm/1985071.html
最近更新
教材专著