| Objective:We comprehensively searched the published literature and systematically evaluated the clinical value of magnetic resonance imaging(MRI)in the preoperative diagnosis of cervical cancer in terms of vagina invasion,parauterine invasion,lymph node metastasis,and deep myometrial invasion with meta-analysis.Methods:We comprehensively searched foreign databases including PubMed,EMBASE,web of science,Cochrane Library and Chinese databases including CNKI,Wanfang and VIP,and collected literatures on the value of MRI in the preoperative diagnosis of cervical cancer from inception to November 2019.According to the inclusive and exclusive criteria,we screened the literatures,used QUADAS-2 to evaluate the quality of the included literatures with Review Manager 5.3 software and extracted the data.We integrated and analyzed the data,including sensitivity(Sen),specificity(Spe),positive likelihood ratio(+LR),negative likelihood ratio(-LR)and the diagnostic odds ratio(DOR),and drew the forest map and the summary receiver operating characteristic curve(SROC),calculated the area under the curve(AUC)with Stata 14.0 software,analyzed whether there was heterogeneity among the studies and found the causes of heterogeneity.Results:1.Twenty literatures were included to study MRI diagnosis of vagina invasion of cervical cancer.The results showed that the combined Sen was 0.81(95% CI: 0.76,0.85),the Spe was 0.90(95% CI: 0.86,0.92),the + LR was 7.7(95% CI: 5.4,11.0),the-LR was 0.21(95% CI: 0.16,0.27),the DOR was 36(95% CI: 21,63),the AUC was 0.91(95% CI: 0.88,0.93).The pre-test probability is 50%,the post test probability is 89% when the MRI result of vagina invasion is positive,and 18% when the MRI result is negative.2.27 literatures were included to study MRI diagnosis of the parauterine invasion of cervical cancer.The results showed that the combined Sen was 0.81(95% CI: 0.74,0.87),the Spe was 0.92(95% CI: 0.89,0.94),the +LR was 9.7(95% CI: 7.4,12.7),the-LR was 0.20(95% CI: 0.14,0.29),the DOR was 48(95% CI: 29,78),the AUC was 0.94(95% CI: 0.92,0.96).The pre-test probability is 50%,the post test probability is 91% when the MRI result of parauterine invasion is positive,and 17% when the MRI result is negative.3.Thirty literatures were included to study MRI diagnosis of lymph node metastasis of cervical cancer.The results showed that the combined Sen was 0.73(95% CI: 0.65,0.79),the Spe was 0.94(95% CI: 0.92,0.96),the +LR was 11.9(95% CI: 8.3,17.0),the-LR was 0.29(95% CI: 0.23,0.37),the DOR was 41(95% CI: 24,68),the AUC was 0.93(95% CI: 0.90,0.95).The pre-test probability is 50%,the post test probability is 92% when the MRI result of lymph node metastasis is positive,and 23% when the MRI result is negative.4.Five articles were included to study MRI diagnosis of deep myometrial invasion of cervical cancer.The results showed that the combined Sen was 0.89(95% CI: 0.77,0.95),the Spe was 1.00(95% CI: 0.88,1.00),the +LR was 186.1(95%CI:6.9,4999.4),the-LR was 0.11(95% CI: 0.05,0.24),the DOR was 1710(95%CI:69,42254),the AUC was 0.95(95% CI: 0.93,0.97).The pre-test probability is 50%,the post test probability is 99% when the MRI result of deep myometrial invasion is positive,and 10% when the MRI result is negative.5.The AUCs of MRI in the diagnosis of cervical cancer with vagina invasion,parauterine invasion,lymph node metastasis and deep myometrial invasion were as follows: deep myometrial invasion(0.95)> parauterine invasion(0.94)> lymph node metastasis(0.93)> vaginal invasion(0.91).Conclusions:1.MRI has high accuracy in the diagnosis of cervical cancer with vagina invasion,parauterine invasion,lymph node metastasis,deep myometrial invasion,and has high clinical diagnostic value.2.MRI has high diagnostic efficiency and can be used as an important tool for preoperative evaluation of cervical cancer,which can assist the clinical stage and select a reasonable treatment plan. |
PDF Full Text Request