Skip to main content
NCBI home page
The British Journal of Radiology logoLink to The British Journal of Radiology
. 2015 Jun 24;88(1052):20150063. doi: 10.1259/bjr.20150063

Diagnostic performance of diffusion-weighted MRI for detection of pelvic metastatic lymph nodes in patients with cervical cancer: a systematic review and meta-analysis

G Shen 1, H Zhou 1, Z Jia 1,✉, H Deng 1
PMCID: PMC4651381  PMID: 26111112

Abstract

In recent years, diffusion-weighted (DW) MRI has emerged as a new technique for detecting the pelvic lymph metastases in patients with cervical cancer. The aim of this meta-analysis was to assess the diagnostic value of DW imaging (DWI) for benign/malignant discrimination of pelvic lymph nodes (LNs). Studies about DWI for the detection of metastatic LNs were searched in the PubMed, EMBASE, Web of Science, EBSCO, the Cochrane Library and three Chinese databases. Based on the extracted data, we determined pooled sensitivities, specificities and diagnostic odds ratios (DORs) across studies, calculated positive and negative likelihood ratios (LRs) and constructed summary receiver operating characteristic curves with area under the curve (AUC) and Q* obtained. We also analysed the heterogeneity between studies based on subgroup analysis, threshold effect and publication bias. In total, 15 studies involving 1021 patients met the inclusion criteria. The pooled sensitivity, specificity and DOR of DWI were 0.86 [95% confidence interval (CI), 0.84–0.89], 0.84 (95% CI, 0.83–0.86) and 47.21 (95% CI, 25.67–86.81), respectively. LR syntheses yielded overall positive LR of 6.55 (95% CI, 4.77–9.01) and negative LR of 0.17 (95% CI, 0.12–0.23). The AUC and Q* index were 0.9384 and 0.8754, respectively. The heterogeneity was relatively high between studies; however, there was no evidence for threshold effect and publication bias. DWI is beneficial in the pelvic nodal assessment in patients with cervical cancer. Large-scale, high-quality trials with standard protocols are required to evaluate its clinical value for discrimination of metastatic from non-metastatic pelvic LNs in patients with cervical cancer. Advances in knowledge include providing evidence to assess the role of DWI in nodal staging of cervical cancer.

Cervical cancer is one of the most common gynaecological malignancies, which takes third place among female genital malignancies in the world, following the uterine body and ovarian cancers.1,2 During the past few decades, there has been an obvious decline in the death rate and incidence of cervical cancer in developed countries observed, but this has not been mirrored by a similar decline in developing countries.3

The extent of lymph node (LN) metastasis is an important prognostic factor, and the survival rates of patients with nodal metastases are significantly lower than those of patients without such metastases.4–9 Surgical LN assessment is the gold standard for the diagnosis of LN metastasis;10 however, it increases the time and cost of diagnosis with an increased risk of immediate and delayed complications to the patient. Therefore, a non-invasive technique that accurately identifies LN metastasis would be valuable.

CT and MRI are widely used to assess metastatic LNs of patients with cervical cancer, which is based on the measurement of node size. However, the sensitivity of these imaging techniques for the detection of LN metastasis in uterine cervical cancer is between only 30% and 73%, while the specificity is between 44% and 93%.11–17 Fluorine-18 fludeoxyglucose positron emission tomography/CT is considered to be a useful technique in the detection of LN metastasis for tumour with high sensitivity and specificity, especially as it could provide diagnostic information of the entire body, but so far it has not been widely used for clinical application owing to high cost. Diffusion-weighted imaging (DWI) is sensitive to the diffusion of water molecules in tissue, which can make subtle abnormality more obvious and can provide better characterization of tissue and their pathological processes at microscopic level.18 Additionally, the quantitative assessment can be performed by the measurement of the apparent diffusion coefficient (ADC).19 Of late, DWI has been introduced for the differentiation of benign and malignant lesions in extracranial organs, especially for diagnosis of abdominal and pelvic organs.20,21 Although multiple previous studies have discussed the feasibility of DWI in detecting and diagnosing cervical cancer,22–24 the findings are inconclusive or conflicting. Thus, we undertook a meta-analysis to evaluate the possible benefit of DWI to differentiate metastatic from non-metastatic LNs in uterine cervical cancer.

METHODS AND MATERIALS

Literature search

The PubMed, EMBASE, Web of Science, EBSCO, the Cochrane Library and three Chinese databases—CNKI, VIP and WanFang—were searched for eligible studies about the diagnostic performance of DWI-MR for the pelvic nodal assessment in patients with cervical cancer. The search strategy was based on the combination of the following key words: (“Diffusion Magnetic Resonance Imaging” or “Diffusion MRI” or “Diffusion Weighted MRI” or “DWI” or “diffusion-weighted magnetic resonance imaging” or “MRI DWI” or “diffusion-weighted imaging” or “diffusion-weighted-MRI”) and (“Uterine Cervical Neoplasms” or “Cervical Neoplasms” or “Cervical cancer” or “cervical carcinoma” or “cervix cancer” or “Cervix Neoplasms” or “Cervix carcinoma”). In addition, reference lists of included studies were manually screened for additional eligible studies.

Study selection

Two reviewers independently judged the eligibility with disagreements resolved by discussion and consensus. The inclusion criteria were as follows: (1) DWI was identified to detect pelvic LN metastasis of cervical cancer; (2) histopathology results and/or clinical follow-up were used as the reference standard; (3) only studies with absolute numbers of true-positive (TP), false-positive (FP), false-negative (FN) and true-negative (TN) extracted could be included; (4) the studies were based on a per-lesion analysis; (5) when data or subsets of data were presented in more than one article, the article with the most details or the most recent article was chosen. Because of non-sufficient information, the reviews, letters, conference abstracts and case reports were excluded. The studies consisting of no more than 10 patients were also excluded.

Data extraction and quality assessment

From the studies finally selected, the same investigators independently extracted relevant data including basic description of studies (first author, publication year, original country), study population (sample, age, type of primary cancer), study design (prospective vs retrospective, patient enrolment and blindness), technique parameters (magnetic strength, b-value and scanning sequence) and reference standard. Obviously, TP, FP, FN and TN results were also recorded for further analysis. In addition, the quality assessment of included studies was performed with Quality Assessment of Diagnostic Accuracy Studies (QUADAS) tools, which comprised 14 questions, and each item was answered by “yes”, “no” or “unclear”.25 Score 12 was used as the cut-off of high and low quality.

Statistical analyses

We conducted all analysis based on a per-region/node data analysis. The pooled sensitivity, specificity and diagnostic odds ratio (DOR) with 95% confidence intervals (CIs) were calculated based on the bivariate analysis.26 In addition, likelihood ratios (LRs) that were the combination of sensitivity and specificity in the calculation were also obtained. The discrimination ability was better with higher DOR, higher positive LR and lower negative LR. Furthermore, we also constructed the summary receiver operating characteristic (SROC) curves with area under the curve (AUC) and Q* estimate obtained.

With regard to the heterogeneity between studies, we used the inconsistency index (I2) test to examine it. A fixed model was used if I2 value was >50%, which reflected substantial heterogeneity; otherwise, a fixed model was used. To explain the source of heterogeneity, we performed the subgroup analysis according to the study design, patient enrolment, blindness, QUADAS score and the reference standard. In addition, the meta-regression and publication bias analysis were also performed. All statistical analyses were conducted using Meta-Disc v. 1.4 (XI Cochrane Colloquium; Barcelana, Spain) and STATA v. 12.0 (StataCorp LP; College Station, TX).

RESULTS

Study selection and description

We initially identified 520 articles after electronic search and excluded 455 articles after reviewing the title and abstract. The remaining 65 articles were screened in full text, and 15 articles22,23,27–39 were finally included for further analysis. The selection process and reasons for exclusion were shown in Figure 1. In total, 1021 patients involving 3134 LNs were enrolled in this meta-analysis. Of all studies, only two studies27,28 enrolled patients prospectively, another six studies22,23,29,31,38,39 were retrospectively designed and the remaining were unknown. Patients were enrolled consecutively in five articles.28,29,31–33 The principal characteristics of included studies were presented in Table 1.

Figure 1.

Figure 1.

Open in a new tab

Flowchart for study selection.

Table 1.

Principal characteristics of eligible studies included in this meta-analysis

Study Nation Patients Lesions Mean age (range) Study design Patient enrolment Blind Review Type of cervical cancer Diffusion-weighted imaging parameters (b-values s mm–2) Cut-off (×10–3 mm s–2) Reference standard
Hu et al33 China 26 77 44.6 ± 7.5 (26–58) ND C Y 2 SCC (23), AC (1), ASC (1), SCLC (1) 1.5 T
SE-EPI
(0,800)		 1.14 HP
Liao and Chen34 China 53 108 45 (25–59) ND ND Y 2 SCC (45), AC (4) , ASC (4) 1.5 T
SE-EPI
(0,800)		 1.11 HP
Ling et al35 China 36 86 45 ± 12 ND ND ND ND SCC (32), AC (4) 1.5 T
SE-EPI
(0,500,1000)		 ND HP
Liu et al36 China 49 116 44 (27–75) ND ND Y 2 SCC (44), AC (3), ASC (2) 3 T
SE-EPI
(0,600,1000)		 0.778 HP
Liu et al37 China 36 159 44.1 (32–62) ND ND ND ND SCC (26), AC (9), CS (1) 1.5 T
ASSET/SE-EPI
(0,1000)		 0.983 HP
Ren et al38 China 42 188 43.16 ± 8.95 R ND Y 2 SCC (93), AC (18) 1.5 T
SS-EPI
(0,800)		 0.879 HP
Zhang et al39 China 65 392 45.6 ± 6.5 (35–73) R ND Y 2 ND 1.5 T
SS-EPI
(0,800)		 0.881 HP
Chen et al28 China 61 153 46 (25–60) P ND ND 2 SCC (52), AC (5), ASC (4) 1.5 T
SE-EPI
(0,800)		 1.15 HP
Chen et al27 China 26 77 35 (26–58) P C ND 2 SCC (23), AC (1), ASC (1), SCLC (1) 1.5 T
STIR-EPI
(0,800)		 1.14 HP
Choi et al29 Republic of Korea 163 339 52 ± 13 (28–75) R C Y 2 ND 1.5 T
SS-SE-EPI
(0,1000)		 0.864 Clinical and imaging follow-up;
Xue et al32 China 24 66 37.9 ND C ND 2 AC (5), epitheliums (19) 1.5 T
STIR-EPI
(800)		 0.921 HP
Kim et al22 Republic of Korea 125 250 48 ± 12 R ND Y 2 ND 1.5 T
SS-EPI
(0,1000)		 0.862 HP
Kim et al23 Republic of Korea 143 680 48 ± 11 (24–73) R ND Y 2 ND 1.5 T
SE-EPI
(0,1000)		 0.911 HP
Liu et al30 China 42 188 45.3 (30–62) ND ND Y 2 ND 1.5 T
SS-SE-EPI/ASSET
(0,1000)		 0.881 HP
Park et al31 Republic of Korea 130 255 49 ± 14 R   Y 2 ND 1.5 T
SS-SE-EPI
(0,1000)		 0.790 HP
Open in a new tab

AC, adenocarcinoma; ASC, adenosquamous carcinoma; ASSET, array spatial sensitivity encoding technique; C, consecutive; CS, carcinosarcoma; HP, histopathology; ND, not documented; P, prospective; R, retrospective; SCC, squamous cell carcinoma; SCLC, small cell lung cancer; SE-EPI, spin echo echoplanar imaging; SS, single shot; SS-EPI, SS echo-planar imaging; STIR-EPI, short tau inversion recovery echoplanar imaging; Y, yes (blind).

Assessment of study quality and publication bias

The results of quality assessment were shown in Table 2. All studies fulfilled 11 or more of the 14 items in QUADAS tool. Common weaknesses were concentrated in the time interval between tests (Item 4) and the blindness between the reference standard and index test (Items 10 and 11). Based on the results of Deeks' funnel plot asymmetry test (p = 0.329), no publication bias was found (Figure 2).

Table 2.

Evaluation of quality of included studies using Quality Assessment of Diagnostic Accuracy Studies tool

Study Item
 Total score
1 2 3 4 5 6 7 8 9 10 11 12 13 14
Hu et al33 Y Y Y N Y Y Y Y Y Y Y Y Y Y 14
Liao and Chen34 Y Y Y Y Y Y Y Y Y Y N Y Y Y 13
Ling et al35 Y Y Y N Y Y Y Y Y N N Y Y Y 11
Liu et al36 Y Y Y Y Y Y Y Y Y Y Y Y Y Y 14
Liu et al37 Y Y Y Y Y Y Y Y Y N N Y Y Y 12
Ren et al38 Y Y Y Y Y Y Y Y Y Y Y Y Y Y 14
Zhang et al39 Y Y Y Y Y Y Y Y Y Y Y Y Y Y 14
Chen et al28 Y Y Y N Y Y Y Y Y N N Y Y Y 11
Chen et al27 Y Y Y N Y Y Y Y Y N N Y Y Y 11
Choi et al29 Y Y Y Y Y N N Y Y Y Y Y Y Y 12
Xue et al32 Y Y Y N Y Y Y Y Y N N Y Y Y 11
Kim et al22 Y Y Y Y Y Y Y Y Y Y Y Y Y Y 14
Kim et al23 Y Y Y Y Y Y Y Y Y N N Y Y Y 12
Liu et al30 Y Y Y Y Y Y Y Y Y Y N Y Y Y 13
Park et al31 Y Y Y Y Y Y Y Y Y Y Y Y Y Y 14
Open in a new tab

U: unclear, 0 score; Y: yes, 1 score.

Figure 2.

Figure 2.

Open in a new tab

Results of Deek's funnel plot of asymmetry test for publication bias for diffusion-weighted imaging. The non-significant slope indicated that no significant bias was found (p = 0.329). ESS, effective sample size.

Diagnostic accuracy of diffusion-weighted imaging

The pooled sensitivity, specificity and DOR of DWI were 0.86 (95% CI, 0.84–0.89), 0.84 (95% CI, 0.83–0.86) and 47.21 (95% CI, 25.67–86.81), respectively. LR syntheses demonstrated an overall positive LR of 6.55 (95% CI, 4.77–9.01) and nagative LR of 0.17 (95% CI, 0.12–0.23). The SROC curves showed very good diagnostic performance overall for DWI with AUC and Q* index obtained (0.9384 for AUC and 0.8754 for Q* index) (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

Summary receiver operating characteristic (SROC) curves for diffusion-weighted imaging in nodal assessment of cervical cancer. The middle line was the summarized ROC curve, and the two beside were 95% confidence intervals. AUC, area under the curve; SE, standard error.

With regard to the threshold effect, a Spearman's rank correlation test was performed and was determined to be 0.38 (p = 0.162), which indicated the absence of threshold effect. The subgroup analyses were performed based on patient enrolment, blindness and quality assessment. The results were presented in Table 3, which showed there were no significant differences among those subgroups (p > 0.05). In addition, the heterogeneity between studies was highly significant (I2 > 50%); thus, we should interpret the results with care.

Table 3.

Diagnostic accuracy of diffusion-weighted imaging in the evaluation of lymph node staging in patients with cervical cancer

Study Number Pooled sensitivity I2 (%) Pooled specificity I2 (%) Positive likelihood ratio Nagative likelihood ratio Diagnostic odds ratio
Overall 15 0.86 (0.84–0.89) 57.0 0.84 (0.83–0.86) 89.1 6.55 (4.77–9.01) 0.17 (0.12–0.23) 47.21 (25.67–86.81)
Consecutive 5 0.85 (0.80–0.90) 33.2 0.89 (0.86–0.91) 88.0 10.05 (4.76–21.23) 0.18 (0.13–0.25) 69.70 (24.77–196.13)
Inconsecutivea 10 0.87 (0.83–0.90) 65.8 0.83 (0.81–0.85) 89.4 5.81 (3.99–8.46) 0.16 (0.10–0.25) 40.46 (18.40–89.00)
Blinded 10 0.85 (0.81–0.88) 55.4 0.84 (0.82–0.85) 91.4 6.14 (4.24–8.89) 0.19 (0.13–0.27) 37.12 (18.55–74.30)
Unblindedb 5 0.90 (0.85–0.94) 57.7 0.88 (0.84–0.91) 80.9 8.13 (3.91–16.89) 0.12 (0.06–0.24) 97.39 (22.68–418.26)
High quality 11 0.86 (0.82–0.89) 57.3 0.84 (0.83–0.86) 91.0 6.52 (4.52–9.40) 0.18 (0.12–0.26) 42.58 (21.29–85.16)
Low quality 4 0.89 (0.83–0.93) 63.2 0.85 (0.80–0.90) 83.1 7.37 (3.13–17.34) 0.14 (0.07–0.28) 81.35 (15.04–440.13)
Open in a new tab
a

Inconsecutive or unclear.

b

Unblinded or unclear.

The differences in ADC measurements between metastatic LNs and non-metastatic LNs were statistically significant, with no overlapping of the 95% CIs [0.85 (95% CI, 0.84–0.87) vs 1.04 (95% CI, 1.04–1.05), p < 0.05]. This is illustrated in Figure 4.

Figure 4.

Figure 4.

Open in a new tab

Forest plot: apparent diffusion coefficient (ADC) values for metastatic lymph nodes and non-metastatic lymph nodes. CI, confidence interval; ES, effect size; ID, identification.

DISCUSSION

Compared with the surgical LN assessment that is the gold standard for the diagnosis of LN metastasis, a non-invasive technique would be of great benefit to the patients with cervical cancer, which will reduce the time and cost of diagnosis, as well as the risk of immediate and delayed complications.10,40 Traditional size criteria that the short-axis diameter of 10 mm was commonly used as the cut-off for benign and malignant lesions revealed a relevantly low sensitivity.12,41 DWI that provides information about the integrity of cell membranes and tissue consistency is non-invasive and enables the radiologist to move from morphological to functional assessment of diseases. Furthermore, quantitative assessment is possible by the ADC values, which could be measured by DWI.42 As a result of technical developments and recent research, DWI has emerged as a promising tool for detecting and characterizing gynaecologic tumours, determining the anatomic extent of disease, understanding lesion pathophysiology and predicting and monitoring treatment outcome.10,25,26

DWI has been reported to be capable of distinguishing metastatic from benign LNs in patients with uterine cervical cancer. Chen et al28 reported that DWI with ADC measurements could differentiate metastatic from hyperplastic nodes with a sensitivity of 83.3%, specificity of 74.7% and accuracy of 78.4%. In another study, Liu et al30 used the minimum ADC as a cut-off, and the sensitivity and specificity for differentiating metastatic from non-metastatic LNs were 95.7% and 96.5%, respectively. Kim et al22 and Choi et al29 demonstrated the similar conclusion that DWI was feasible for differentiating metastatic from non-metastatic LNs of uterine cervical cancer. However, owing to the presence of relatively small sample size and variations in study design, it is necessary to perform a meta-analysis to have a comprehensive assessment of DWI in the diagnosis of pelvic LNs of cervical cancer.

The present meta-analysis identified 15 studies that involve the diagnostic accuracy of DWI for metastatic LNs of cervical cancer. To our knowledge, this is the first meta-analysis that evaluates the role of DWI for nodal assessment in pelvic imaging. The recent meta-analysis performed by Hou et al40 investigated the diagnostic performance of DWI for differentiating normal cervical tissue from tumour tissues, not for nodal assessment. Our meta-analysis demonstrated that the pooled estimates were 0.86 (95% CI, 0.84–0.89) for sensitivity and 0.84 (95% CI, 0.83–0.86) for specificity, which indicated that DWI with ADC values could be helpful for the diagnosis of metastatic LNs. However, the heterogeneity between studies was relatively notable, and it was necessary to explore reasons for heterogeneity.

With regard to the threshold effect, the result (Spearman's correlation coefficient, 0.38; p > 0.05) indicated that no significant threshold effect existed. Based on the subgroup analysis, patient enrolment, blindness or quality assessment did not contribute to the observed heterogeneity. Although we did not perform the subgroup analysis based on the study design mainly owing to limited data, the study design might be a potential source of heterogeneity. In general, retrospective studies had higher diagnostic performance than prospective studies, for the risk that imaging observers had known the histopathological results before DWI assessment. We have examined the publication bias using Deek's funnel plot, and there was no publication bias existing. However, all of our included studies were performed in Asia, which was a potential bias.

Currently, there was no standard optimal protocol for DWI scanning; thus, the differences in the instrument manufacturers, magnetic field strength and sequences could have been a main cause of the between-study heterogeneity. For example, the most commonly used magnetic field strength was 1.5 T. Some studies27,28,32 used b-values of 0 and 800 s mm–2; others22,23,31 used 0 and 1000 s mm–2. With a low b-value, signals were significantly affected by perfusion effects, which led to an inaccurate reflection of water diffusion motion. However, high b-values have the risk of distortion and susceptibility artefacts.12,41 All of the above factors were capable of affecting ADC measurements, which led to variability in cut-off values ranging from 0.778 to 1.150. Furthermore, with regard to ADC measurements, the minimum ADC, the mean ADC or relative ADC was used as the reference standard.

The DOR, which is calculated by the combination of sensitivity and specificity, is a single indication of test accuracy.42 It ranges from zero to infinity, with higher values indicating better diagnostic performance. The pooled DOR was 47.21, which suggested high total diagnostic accuracy of DWI. In addition, both positive LR and negative LR were calculated and served as our measurements of diagnostic performance. The results (6.55 for positive LR and 0.17 for negative LR) revealed that using DWI alone is not accurate enough to rule in or out the disease.

We should acknowledge some limitations in this meta-analysis. First, the exclusion of reviews, conference abstracts and letters may lead to potential publication bias. Although it can be tested by funnel plots, the limited sample of included studies in this meta-analysis might decrease the examination power of publication bias. Second, we did not perform meta-regression analysis owing to insufficient data. The present explanation for the heterogeneity was insufficient; there were too many variables in the scanning method and acquisition protocol to conduct the subgroup analysis based on these variants. Third, most of the included studies were retrospective in design. The retrospective nature of studies can be considered a limitation, which potentially affected the diagnostic performance of DWI.

In conclusion, DWI is useful for differentiation between metastatic and benign LNs in patients with uterine cervical cancer. Large, multicentre and prospective studies with strict standardization of DWI protocols are required to evaluate its diagnostic performance for further development as a routine clinical application in the near future.

FUNDING

This study was supported by National Natural Science Foundation (grant nos. 81271532, 81171456 and 30900378) and the Fundamental Research Funds for the Central Universities (Project no. 2015SCU04B09).

Contributor Information

G Shen, Email: shengh1990@hotmail.com.

H Zhou, Email: 1017412478@qq.com.

Z Jia, Email: zhiyunjia@hotmail.com.

H Deng, Email: denghfy@126.com.

REFERENCES

  • 1.Anttila A, Ronco G, Clifford G, Bray F, Hakama M, Arbyn M, et al. Cervical cancer screening programmes and policies in 18 European countries. Br J Cancer 2004; 91: 935–41. doi: 10.1038/sj.bjc.6602069 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Arbyn M, Antoine J, Valerianova Z, Mägi M, Stengrevics A, Smailyte G, et al. Trends in cervical cancer incidence and mortality in Bulgaria, Estonia, Latvia, Lithuania and Romania. Tumori 2010; 96: 517–23. [DOI] [PubMed] [Google Scholar]
  • 3.Arbyn M, Castellsagué X, de Sanjosé S, Bruni L, Saraiya M, Bray F, et al. Worldwide burden of cervical cancer in 2008. Ann Oncol 2011; 22: 2675–86. doi: 10.1093/annonc/mdr015 [DOI] [PubMed] [Google Scholar]
  • 4.Manetta A, Delgado G, Petrilli E, Hummel S, Barnes W. The significance of paraaortic node status in carcinoma of the cervix and endometrium. Gynecol Oncol 1986; 23: 284–90. doi: 10.1016/0090-8258(86)90128-9 [DOI] [PubMed] [Google Scholar]
  • 5.Creasman WT, Morrow CP, Bundy BN, Homesley HD, Graham JE, Heller PB. Surgical pathologic spread patterns of endometrial cancer. A Gynecologic Oncology Group Study. Cancer 1987; 60(Suppl. 8): 2035–41. doi: [DOI] [PubMed] [Google Scholar]
  • 6.Inoue T, Morita K. The prognostic significance of number of positive nodes in cervical carcinoma stages IB, IIA, and IIB. Cancer 1990; 65: 1923–7. doi: [DOI] [PubMed] [Google Scholar]
  • 7.Gal D, Recio FO, Zamurovic D, Tancer ML. Lymphvascular space involvement—a prognostic indicator in endometrial adenocarcinoma. Gynecol Oncol 1991; 42: 142–5. doi: 10.1016/0090-8258(91)90334-2 [DOI] [PubMed] [Google Scholar]
  • 8.Stehman FB, Bundy BN, DiSaia PJ, Keys HM, Larson JE, Fowler WC. Carcinoma of the cervix treated with radiation therapy. I. A multi-variate analysis of prognostic variables in the Gynecologic Oncology Group. Cancer 1991; 67: 2776–85. doi: [DOI] [PubMed] [Google Scholar]
  • 9.Lai CH, Hong JH, Hsueh S, Ng KK, Chang TC, Tseng CJ, et al. Preoperative prognostic variables and the impact of postoperative adjuvant therapy on the outcomes of Stage IB or II cervical carcinoma patients with or without pelvic lymph node metastases: an analysis of 891 cases. Cancer 1999; 85: 1537–46. doi: [DOI] [PubMed] [Google Scholar]
  • 10.Malayeri AA, El Khouli RH, Zaheer A, Jacobs MA, Corona-Villalobos CP, Kamel IR, et al. Principles and applications of diffusion-weighted imaging in cancer detection, staging, and treatment follow-up. Radiographics 2011; 31: 1773–91. doi: 10.1148/rg.316115515 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Rockall AG, Sohaib SA, Harisinghani MG, Babar SA, Singh N, Jeyarajah AR, et al. Diagnostic performance of nanoparticle-enhanced magnetic resonance imaging in the diagnosis of lymph node metastases in patients with endometrial and cervical cancer. J Clin Oncol 2005; 23: 2813–21. doi: 10.1200/JCO.2005.07.166 [DOI] [PubMed] [Google Scholar]
  • 12.Bastin ME. Correction of eddy current-induced artefacts in diffusion tensor imaging using iterative cross-correlation. Magn Reson Imaging 1999; 17: 1011–24. doi: 10.1016/S0730-725X(99)00026-0 [DOI] [PubMed] [Google Scholar]
  • 13.Scheidler J, Hricak H, Yu KK, Subak L, Segal MR. Radiological evaluation of lymph node metastases in patients with cervical cancer. A meta-analysis. JAMA 1997; 278: 1096–101. doi: 10.1001/jama.1997.03550130070040 [DOI] [PubMed] [Google Scholar]
  • 14.Hawighorst H, Schoenberg SO, Knapstein PG, Knopp MV, Schaeffer U, Essig M, et al. Staging of invasive cervical carcinoma and of pelvic lymph nodes by high resolution MRI with a phased-array coil in comparison with pathological findings. J Comput Assist Tomogr 1998; 22: 75–81. doi: 10.1097/00004728-199801000-00013 [DOI] [PubMed] [Google Scholar]
  • 15.Yang WT, Lam WW, Yu MY, Cheung TH, Metreweli C. Comparison of dynamic helical CT and dynamic MR imaging in the evaluation of pelvic lymph nodes in cervical carcinoma. AJR Am J Roentgenol 2000; 175: 759–66. doi: 10.2214/ajr.175.3.1750759 [DOI] [PubMed] [Google Scholar]
  • 16.Reinhardt MJ, Ehritt-Braun C, Vogelgesang D, Ihling C, Högerle S, Mix M, et al. Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET. Radiology 2001; 218: 776–82. doi: 10.1148/radiology.218.3.r01mr19776 [DOI] [PubMed] [Google Scholar]
  • 17.Choi HJ, Roh JW, Seo SS, Lee S, Kim JY, Kim SK, et al. Comparison of the accuracy of magnetic resonance imaging and positron emission tomography/computed tomography in the presurgical detection of lymph node metastases in patients with uterine cervical carcinoma: a prospective study. Cancer 2006; 106: 914–22. doi: 10.1002/cncr.21641 [DOI] [PubMed] [Google Scholar]
  • 18.Abdel Razek AA, Soliman NY, Elkhamary S, Alsharaway MK, Tawfik A. Role of diffusion-weighted MR imaging in cervical lymphadenopathy. Eur Radiol 2006; 16: 1468–77. doi: 10.1007/s00330-005-0133-x [DOI] [PubMed] [Google Scholar]
  • 19.Koc Z, Erbay G, Ulusan S, Seydaoglu G, Aka-Bolat F. Optimization of b value in diffusion-weighted MRI for characterization of benign and malignant gynecological lesions. J Magn Reson Imaging 2012; 35: 650–9. doi: 10.1002/jmri.22871 [DOI] [PubMed] [Google Scholar]
  • 20.Liu Y, Ye Z, Sun H, Bai R. Grading of uterine cervical cancer by using the ADC difference value and its correlation with microvascular density and vascular endothelial growth factor. Eur Radiol 2013; 23: 757–65. doi: 10.1007/s00330-012-2657-1 [DOI] [PubMed] [Google Scholar]
  • 21.Tam H, Collins D, Brown G, Chau I, Cunningham D, Leach MO, et al. The role of pre-treatment diffusion-weighted MRI in predicting long-term outcome of colorectal liver metastasis. Br J Radiol 2013; 86: 20130281. doi: 10.1259/bjr.20130281 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kim JK, Kim KA, Park BW, Kim N, Cho KS. Feasibility of diffusion-weighted imaging in the differentiation of metastatic from nonmetastatic lymph nodes: early experience. J Magn Reson Imaging 2008; 28: 714–19. doi: 10.1002/jmri.21480 [DOI] [PubMed] [Google Scholar]
  • 23.Kim MH, Kim JK, Lee Y, Park BW, Lee CK, Kim N, et al. Diagnosis of lymph node metastasis in uterine cervical cancer: usefulness of computer-aided diagnosis with comprehensive evaluation of MR images and clinical findings. Acta Radiol 2011; 52: 1175–83. doi: 10.1258/ar.2011.110202 [DOI] [PubMed] [Google Scholar]
  • 24.Klerkx WM, Veldhuis WB, Spijkerboer AM, van den Bosch MA, Mali WP, Heintz AP, et al. The value of 3.0Tesla diffusion-weighted MRI for pelvic nodal staging in patients with early stage cervical cancer. Eur J Cancer 2012; 48: 3414–21. doi: 10.1016/j.ejca.2012.06.022 [DOI] [PubMed] [Google Scholar]
  • 25.Punwani S. Diffusion weighted imaging of female pelvic cancers: concepts and clinical applications. Eur J Radiol 2011; 78: 21–9. doi: 10.1016/j.ejrad.2010.07.028 [DOI] [PubMed] [Google Scholar]
  • 26.Nougaret S, Tirumani SH, Addley H, Pandey H, Sala E, Reinhold C. Pearls and pitfalls in MRI of gynecologic malignancy with diffusion-weighted technique. AJR Am J Roentgenol 2013; 200: 261–76. doi: 10.2214/AJR.12.9713 [DOI] [PubMed] [Google Scholar]
  • 27.Chen YB, Hu CM, Chen GL, Hu D, Liao J. Staging of uterine cervical carcinoma: whole-body diffusion-weighted magnetic resonance imaging. Abdom Imaging 2011; 36: 619–26. doi: 10.1007/s00261-010-9642-4 [DOI] [PubMed] [Google Scholar]
  • 28.Chen YB, Liao J, Xie R, Chen GL, Chen G. Discrimination of metastatic from hyperplastic pelvic lymph nodes in patients with cervical cancer by diffusion-weighted magnetic resonance imaging. Abdom Imaging 2011; 36: 102–9. doi: 10.1007/s00261-009-9590-z [DOI] [PubMed] [Google Scholar]
  • 29.Choi EK, Kim JK, Choi HJ, Park SH, Park BW, Kim N, et al. Node-by-node correlation between MR and PET/CT in patients with uterine cervical cancer: diffusion-weighted imaging versus size-based criteria on T2WI. Eur Radiol 2009; 19: 2024–32. doi: 10.1007/s00330-009-1350-5 [DOI] [PubMed] [Google Scholar]
  • 30.Liu Y, Liu H, Bai X, Ye Z, Sun H, Bai R, et al. Differentiation of metastatic from non-metastatic lymph nodes in patients with uterine cervical cancer using diffusion-weighted imaging. Gynecol Oncol 2011; 122: 19–24. doi: 10.1016/j.ygyno.2011.03.023 [DOI] [PubMed] [Google Scholar]
  • 31.Park SO, Kim JK, Kim KA, Park BW, Kim N, Cho G, et al. Relative apparent diffusion coefficient: determination of reference site and validation of benefit for detecting metastatic lymph nodes in uterine cervical cancer. J Magn Reson Imaging 2009; 29: 383–90. doi: 10.1002/jmri.21635 [DOI] [PubMed] [Google Scholar]
  • 32.Xue HD, Li S, Sun F, Sun HY, Jin ZY, Yang JX, et al. Clinical application of body diffusion weighted MR imaging in the diagnosis and preoperative N staging of cervical cancer. Chin Med Sci J 2008; 23: 133–7. doi: 10.1016/S1001-9294(09)60027-4 [DOI] [PubMed] [Google Scholar]
  • 33.Hu CM, Chen YB, Chen JL. Study of the application of whole body diffusion weighted magnetic resonance imaging in staging of uterine cervical carcinoma. MD thesis. Fujian: Fujian Medical University; 2009.
  • 34.Liao J, Chen YB. Study of diffusion-weighted MR imaging for pelvic lymph node in patients with cervical carcinoma. MD thesis. Fujian: Fujian Medical University; 2008.
  • 35.Ling RN, Gong JS, Wang XM, Yang L, Wu MX. The application of diffusion weighted imaging (DWI) in the diagnosis of lymph node metastases in patients with uterine cervical cancer. [In Chinese.] J Qiqihar Univ Med 2012; 33: 441–3. [Google Scholar]
  • 36.Liu L, Pan Y, Ning G, Guo YK, Li CX, Ou YQ. Diagnostic value of 3T diffusion-weighted MR imaging for metastatic pelvic lymph nodes in cervical cancer patients. [In Chinese.] J Sichuan Univ 2014; 45: 159–63. [Google Scholar]
  • 37.Liu Y, Bai RJ, Liu HD, Wang DH, Ye ZX. Evaluation of metastatic lymph nodes in patients with uterine cervical cancer by DWI. [In Chinese.] J Clin Radiol 2011; 30: 834–8. [Google Scholar]
  • 38.Ren C, Jin ZY, Xue HD, Shen J. The application of DWI in preoperative evaluation and accessing therapeutic response to concurrent chemoradiotherapy in patients with cervical cancer. MD thesis. Beijing: Peking Union Medical College; 2013.
  • 39.Zhang J, Ren C, Xue HD, Zhou HL, Sun ZY, Jin ZY. Value of diffusion-weighted imaging in diagnosis of lymph node metastasis in patients with cervical cancer. [In Chinese.] Acta Academiae Med Sci 2014; 36: 73–8. doi: 10.3881/j.issn.1000-503X.2014.01.014 [DOI] [PubMed] [Google Scholar]
  • 40.Hou B, Xiang SF, Yao GD, Yang SJ, Wang YF, Zhang YX, et al. Diagnostic significance of diffusion-weighted MRI in patients with cervical cancer: a meta-analysis. Tumour Biol 2014; 35: 11761–9. doi: 10.1007/s13277-014-2290-5 [DOI] [PubMed] [Google Scholar]
  • 41.Jezzard P, Barnett AS, Pierpaoli C. Characterization of and correction for eddy current artifacts in echo planar diffusion imaging. Magn Reson Med 1998; 39: 801–12. doi: 10.1002/mrm.1910390518 [DOI] [PubMed] [Google Scholar]
  • 42.Glas AS, Lijmer JG, Prins MH, Bonsel GJ, Bossuyt PM. The diagnostic odds ratio: a single indicator of test performance. J Clin Epidemiol 2003; 56: 1129–35. doi: 10.1016/S0895-4356(03)00177-X [DOI] [PubMed] [Google Scholar]

Articles from The British Journal of Radiology are provided here courtesy of Oxford University Press

RESOURCES