The Diagnosis Performance Of Superb Microvascular Imaging In Differential Diagnosis Of Benign And Malignant Thyroid Nodules: A Meta-analysis

Background:In recent years,the incidence rate of thyroid cancer is on the rise.With the wide application of ultrasound technology,ultrasound has become an important means of thyroid detection.Fine needle aspiration biopsy(FNA)is the best non-surgical diagnostic method for thyroid cancer,but it is invasive.To improve the accuracy of noninvasive ultrasound diagnosis is helpful to reduce the number of FNA,which is of great significance in clinic.Conventional Doppler ultrasound techniques use a one-dimensional wall filter to remove hetero waves,resulting in the loss of low velocity blood flow.Superb microvascular imaging(Superb microvascular imaging,SMI)is a new ultrasound imaging method proposed by Toshiba(Canon)in 2014.It uses multi-dimensional filter to separate the blood flow signal from the clutter,so as to retain the low-speed blood flow signal,provide more vascular information,make up for the deficiency of color Doppler imaging(CDFI)and power Doppler imaging(PDI),and provide more abundant information for exploring the blood vessel characteristics of malignant tumors.Since the birth of SMI technology,many experts have applied it to the differential diagnosis of benign and malignant thyroid nodules.So far,no meta-analysis has been found to evaluate the diagnostic ability of SMI in differentiating benign and malignant thyroid nodules.Purpose:We aimed to investigate the diagnostic performance of superb microvascular imaging(SMI)in distinguishing benign from malignant thyroid nodules.Methods:Use"thyroid neoplasm,thyroid neoplasms,thyroid cancer,thyroid tumor,thyroid tumors,thyroid nodule,thyroid nodules"and"superb microvascular imaging,SMI"as search terms.Use a combination of subject headings and free words to search for related articles published in Pub Med,Cochrane and EMBASE,Wan fang and CNKI databases until March 2021.According to the established inclusion and exclusion criteria,the literature was strictly screened,and the review manager 5.3 was used to evaluate the quality of the literature.Extract basic information and related data from documents.Statistical analysis was performed on the extracted data using Meta-Disc 1.4 and Stata15.0 to calculate the combined sensitivity,specificity and area under the receptor operating characteristic curve(AUC).The Cochrane-Q test in Stata 15.0 was used to analyze the heterogeneity,and the I~2index was calculated to evaluate the heterogeneity.When the I~2 value is greater than 50%,it indicates that there is heterogeneity.Use subgroup analysis to explore other sources of heterogeneity.The corresponding 95%confidence interval(95%CI)was generated through the bivariate mixed effects model.Subgroup analysis was performed to test the source of heterogeneity,and a Deeks funnel chart was drawn to test publication bias.Results:This Meta-analysis included 13 articles,including 2047 nodules from 1785patients.The result of I~2 index assessing heterogeneity is I~2=78%,indicating that the included studies are heterogeneous.The results showed that the combined sensitivity of SMI was 82%,the specificity was 81%,the positive likelihood ratio was 4.3,the negative likelihood ratio was 0.23,the diagnostic odds ratio was 19 and the area under the SROC curve(AUC)was 0.88.According to the different diagnostic criteria(perforating blood vessels,blood flow classification,intra-nodular blood vessels),subgroup analysis was performed.The sensitivity was 84%,82%,79%,and the specificity was 83%,70%,87%,respectively.Conclusion:1.Meta-analysis results show that SMI has high sensitivity,specificity and AUC value for distinguishing benign and malignant thyroid nodules.2.At present,there is no unified standard for the differential diagnosis of thyroid nodules by using SMI to assess the blood flow characteristics of thyroid nodules.In this study,three diagnostic criteria were used to differentially diagnose thyroid nodules in the literature.Blood flow classification>blood flow classification>intra-nodular blood vessel(84%>82%>79%);in terms of specificity,intra-nodular blood vessel>blood flow classification>blood flow classification(87%>83%>70%).It can be seen that using the blood flow information provided by SMI to explore the blood flow characteristics of thyroid nodules has important value in the early judgment of thyroid malignant tumors.