Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print November 16, 2007, 10.1148/radiol.2461061298
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Supplemental Table
Right arrow All Versions of this Article:
2461061298v1
246/1/116    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Peters, N. H. G. M.
Right arrow Articles by Peeters, P. H. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Peters, N. H. G. M.
Right arrow Articles by Peeters, P. H. M.
(Radiology 2008;246:116-124.)
© RSNA, 2007

Evidence-based Practice

Meta-Analysis of MR Imaging in the Diagnosis of Breast Lesions1

Nicky H. G. M. Peters, MD, Inne H. M. Borel Rinkes, MD, PhD, Nicolaas P. A. Zuithoff, MSc, Willem P. T. M. Mali, MD, PhD, Karel G. M. Moons, MSc, PhD, and Petra H. M. Peeters, MD, PhD

1 From the Department of Radiology (N.H.G.M.P., W.P.T.M.M.); Department of Surgical Oncology (I.H.M.B.R.); and Department of Clinical Epidemiology, Julius Center for Health Sciences and Primary Care (N.H.G.M.P., N.P.A.Z., K.G.M.M., P.H.M.P.), University Medical Center Utrecht, Heidelberglaan 100, E01.132, 3584 CX Utrecht, the Netherlands. Received July 28, 2006; revision requested October 3; revision received December 21; accepted January 30, 2007; final version accepted June 1. Address correspondence to N.H.G.M.P. (e-mail: n.peters{at}umcutrecht.nl).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
 References
 
Purpose: To determine, in a meta-analysis, the diagnostic performance of contrast material–enhanced magnetic resonance (MR) imaging in patients with breast lesions.

Materials and Methods: Studies to assess the diagnostic performance of MR imaging in patients suspected of having breast cancer who underwent MR imaging and biopsy from January 1985 through March 2005 were reviewed for inclusion. A summary receiver operating characteristic curve was constructed, and pooled weighted estimates of sensitivity and specificity were calculated by using the recently developed bivariate approach for diagnostic meta-analysis.

Results: Of 251 eligible studies, 44 were included in the meta-analysis (sample size range, 14–821; cancer prevalence, 23%–84%). Pooled weighted estimates of sensitivity and specificity were 0.90 (95% confidence interval: 0.88, 0.92) and 0.72 (95% confidence interval: 0.67, 0.77), respectively. The performance of breast MR imaging was influenced by the prevalence of cancer in the studied population (P = .05) and by whether two criteria (ie, morphology, enhancement, and kinetic enhancement pattern)—versus one or three criteria—were used to differentiate benign from malignant lesions (P = .02).

Conclusion: MR imaging of the breast has high sensitivity and lower specificity in the evaluation of breast lesions.

Supplemental material: http://radiology.rsnajnls.org/cgi/content/full/2461061298/DC1

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
 References
 
The value of magnetic resonance (MR) imaging of the breast is based on the capability of this modality to depict (a) multicentric and multifocal disease (1–7), (b) an invasive component in ductal carcinoma in situ lesions (8,9), (c) the tumor in a three-dimensional way (3,5,10,11), and (d) cancer in dense breast tissue (12–14). Thus, MR imaging has the potential to facilitate improved diagnosis and treatment in patients with breast lesions. Yet, for all of these applications, the level of diagnostic performance is important.

Numerous studies have been performed to assess the diagnostic performance of MR imaging in the evaluation of breast lesions (15–58). Hrung et al (59) performed a meta-analysis of studies performed between 1994 and 1997. Sensitivity and specificity varied widely among the included studies: Sensitivity ranged from 0.63 to 1.00, and specificity ranged from 0.21 to 1.00. At a sensitivity of 0.95, the corresponding specificity was 0.67. However, the patient characteristics, MR imaging techniques, and diagnostic criteria for malignancy in the studies differed substantially, and this may have compromised the comparison of the diagnostic performance of breast MR imaging between the studies. After 1997, many additional studies of the diagnostic performance of breast MR imaging were performed (15–17,19,20,22–25,27–29,33–35,37,38,47–50,54–57). The aim of our study was to determine, in a meta-analysis, the diagnostic performance of contrast material–enhanced MR imaging in patients with breast lesions.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
 References
 
Search Strategy and Selection Criteria
A Medline search was performed (N.H.G.M.P.; February 23, 2005) by using the terms "Magnetic Resonance Imaging [MeSH] or MRI or MR mammography or MR-mammography" for the diagnostic test and "Breast Neoplasms [MeSH] or breast cancer or breast lesions or breast" for the clinical domain. To avoid missing relevant studies, we did not include terms regarding diagnostic performance in our search criteria. We limited our search to the English language, results had to apply to human female subjects older than 19 years, and the search terms had to be present in the title or abstract of the article. Since contrast-enhanced MR imaging of the breast was not introduced before 1985, we limited the search to publication dates from January 1985 through March 2005. We did not define a minimal subject sample size for inclusion. In addition, with use of the same criteria applied in the Medline search, published meta-analyses and reviews on this subject were searched for (N.H.G.M.P.; July 7, 2005) in PubMed, the DARE (Database of Abstracts of Reviews of Effectiveness) database, and the Cochrane library to find additional studies.

Eligibility Criteria and Data Extraction
Studies were excluded after exclusion criteria based on quality assessment of diagnostic accuracy studies, or QUADAS (60), and standards for reporting of diagnostic accuracy, or STARD (61), guidelines were prospectively defined and applied. First, the inclusion and exclusion criteria were applied to the titles of the study articles that fulfilled the search criteria. Then, the same criteria were applied to the article abstracts of the studies that were still eligible after the titles were reviewed. Finally, for the remaining studies, we applied inclusion and exclusion criteria to the full extent of the remaining studies to obtain the final studies to be included in the meta-analysis.

Patients.—Our main interest was the diagnostic performance of MR imaging of small lesions detected at mammographic screening. Because a large proportion of these lesions are nonpalpable, we included only those studies that included at least one patient with a nonpalpable lesion. Articles with data on patients with only palpable or otherwise symptomatic lesions were excluded since sensitivity and specificity vary with lesion size and thus with disease severity (62,63). Such variation can affect the overall estimates of the sensitivity and specificity of breast MR imaging (62–64).

Index test.—The diagnostic test evaluated in this meta-analysis was contrast-enhanced MR imaging of the breast. There is substantial heterogeneity among the studies in the types of MR imaging systems, image acquisition protocols, contrast agents, contrast agent doses, and image interpretation methods, all of which may affect the sensitivity and specificity of MR imaging in breast lesion characterization. We included only those studies that were performed by using an MR system with a field strength of at least 1.5 T, in which T1-weighted images were acquired before and after contrast agent administration, and in which no complex imaging modalities such as MR spectroscopy or diffusion-weighted imaging were used. By applying these criteria, we sought to include only those studies performed according to current clinical practice standards and to increase the comparability of the studies.

Reference standard.—To be eligible for inclusion, the results of breast MR imaging had to be compared with those of an appropriate reference-standard examination. Hence, to minimize the effects of partial verification bias, we included only those studies in which a valid reference test was used in more than 75% of the included patients (60,65). Histologic analysis (of a large core-needle biopsy specimen or surgical specimen) and mammographic and clinical follow-up of more than 2 years were considered valid reference standards.

Data extraction.—We included only those studies in which the numbers of true-positive, false-positive, false-negative, and true-negative findings, as well as the numbers of studies on which these values were based, were reported. If in one article, the results were presented for several observers (to obtain the lowest number of missed cancers or the lowest number of false-positive cases), those results that were "most likely in accordance with clinical practice," as judged by two authors (P.H.M.P., W.P.T.M.M.), were abstracted. A prospectively constructed data extraction sheet based on quality assessment of diagnostic accuracy studies guidelines for diagnostic performance studies included in systematic reviews (60) extended with topic-specific items was applied to the eligible studies (45 items).

Data were extracted by two reviewers (N.H.G.M.P. and I.H.M.B.R., W.P.T.M.M., or P.H.M.P.) independently. In cases of discrepancy between the first two reviewers, a third reviewer (P.H.M.P. or W.P.T.M.M.) was consulted and a consensus was reached. The following data were abstracted from the eligible articles: country where the study was performed, year of publication, study design (case-control or patient cohort), mean patient age, cancer prevalence, prevalences of palpable lesions and invasive ductal carcinoma, contrast agent type and dose, whether the criteria for malignancy (lesion morphology, enhancement characteristics, and contrast enhancement over time) were reported, number of criteria used to differentiate benign from malignant lesions, whether the threshold for malignancy was determined a priori or post hoc during analysis of the study results, percentage of patients in whom an appropriate reference standard was used, number of different reference standards used, whether the interpreter of the MR images was blinded to clinical information and/or the reference-standard examination results, and lesion-to-patient ratio (number of included lesions divided by number of included patients).

Statistical Analyses
Overall analysis.—Sensitivity was calculated as TP/(TP + FN), where TP is the number of true-positive findings and FN is the number of false-negative findings. Specificity was calculated as TN/(FP + TN), where TN is the number of true-negative findings and FP is the number of false-positive findings. The diagnostic odds ratio (DOR) was calculated as (TP · TN)/(FP · FN). If the DOR could not be calculated because one of the cells in the 2 x 2 table was empty, 0.5 was added to all cells so that an approximate DOR could be calculated (66,67). The DOR is a single overall indicator of diagnostic performance and is the ratio of the odds of positivity in disease relative to the odds of positivity in nondiseased subjects (68). If the DOR is equal across studies, the only cause of heterogeneity is a difference in cutoff levels for malignancy. If the DOR varies across studies, factors other than cutoff differences exist as well (68). With use of statistical software (SAS, version 8; SAS Institute, Cary, NC), Forest plots for sensitivity and specificity were constructed to graphically present the sensitivity and specificity values, with corresponding 95% confidence intervals (CIs), for the individual studies.

A summary receiver operating characteristic (sROC) curve was constructed by using Stata, version 7 (Stata, College Station, Tex), software. The sROC curve is a method of summarizing the true- and false-positive rates from different diagnostic studies (66,69). If all of the sensitivity–1 – specificity combinations in the individual studies lie on the best fit line in the sROC curve or the deviation from this line can be assumed to be a random variation, the observed differences between studies can be explained by differences in cutoff levels for malignancy (66,68). If, however, the line does not fit the data points, heterogeneity between the studies due to causes other than cutoff differences—such as differences in disease severity, study design, or mean age—could be present.

The goodness of fit of the sROC curve was evaluated by using the (unweighted) R2 statistic of the underlying model, in which the logarithm of the DOR was the dependent variable and the sum of the logit sensitivity and logit specificity was the independent variable (‘proc reg,’ SAS, version 9) (69,70). If all observed values fall on the fit regression line, R2 is 1. If there is no relationship between the observed and predicted values, R2 is 0.

Overall estimates of sensitivity and specificity cannot be calculated separately with the sROC method. Furthermore, the effect of covariates cannot be studied adequately with either the DOR or the sROC approach, as neither method takes into account the correlation between the sensitivity and specificity values within studies (66,68,69,71). Hence, we applied the random effects model by using the relatively recently introduced bivariate analysis for diagnostic meta-analysis (‘proc mixed’ routine) to obtain an overall sensitivity and an overall specificity with a 95% confidence ellipse (SAS, version 9) (71). The 95% confidence ellipse is a combination of the 95% CI for the overall sensitivity and the 95% CI for the overall specificity and can be interpreted as the region that contains the most likely combinations of summary sensitivity and specificity values.

Subgroup analysis.—Finally, we studied the effects of heterogeneity between studies on overall sensitivity and overall specificity. We included each of the following covariates separately in the bivariate model to compare the overall sensitivity and overall specificity between different strata: country where the study was performed, year of publication, study design, mean patient age, cancer prevalence, palpable lesion prevalence, invasive ductal carcinoma prevalence, contrast agent type, contrast agent dose, number of criteria used to differentiate benign from malignant lesions (ie, morphology, enhancement, kinetic enhancement pattern; any one, any two, or all three of these criteria), whether the criteria for malignancy were reported, whether the threshold for malignancy was determined before or after the study was conducted, percentage of patients in whom an appropriate reference standard was used, number of different reference standards used, whether the interpreters of the MR images were blinded to clinical information and histologic findings, and lesion-to-patient ratio. Stratum-specific sensitivities and specificities were compared for all characteristics.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
 References
 
Evaluated Studies
By using the search terms described earlier, we identified 1069 studies. After reviewing the article titles for these studies, we excluded 818 studies. After review of the article abstracts of the remaining 251 studies, 117 studies were left. After cross-referencing the references from a previous meta-analysis (59), two extra studies were included. After review of the 119 articles, 44 studies remained for inclusion in our meta-analysis (Fig 1). The abstracted data of these individual studies are summarized in Table E1 (http://radiology.rsnajnls.org/cgi/content/full/2461061298/DC1).


Figure 1
View larger version (24K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Flowchart of eligible studies. *Other clinical domains such as patients with metastatic breast cancer, patients with recurrent cancer, and patients at high risk. {dagger}Other aims such as MR-guided interventions, MR evaluation of effect chemotherapy, and MR evaluation of breast implants.

 
Overall Analysis
Specificity varied across studies more than sensitivity did (Fig 2). The fact that the fit line of the sROC curve did not precisely fit the data points (Fig 3) suggests that factors other than differences in cutoff points for malignancy caused variations in the sensitivity and specificity of breast MR imaging across the studies. The goodness-of-fit test revealed the same finding: The R2 statistic was 0.12, with a P value of .024, indicating significant heterogeneity across the studies. The overall sensitivity, overall specificity, and DOR based on the bivariate approach were 0.90 (95% CI: 0.88, 0.92), 0.72 (95% CI: 0.67, 0.77), and 23.5 (95% CI: 16.8, 32.9), respectively (Fig 3).


Figure 2
View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Forest plot of sensitivity and specificity, with corresponding 95% CIs, of included studies. Numbers 15–58 are the reference numbers of the individual studies. n/N = number of studies included/total number of studies. p/P = number of patients included/total number of patients.

 

Figure 3
View larger version (8K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3: Sensitivity and specificity plotted in receiver operating characteristic space for individual studies; an sROC curve is plotted through the data points. {blacksquare} = overall sensitivity and overall specificity, with the 95% confidence ellipse ({circ}) indicating the region that contains the most likely combinations of summary sensitivity and specificity values.

 
Subgroup Analysis
The mean sample size for the 44 studies was 128 patients (range, 14–821 patients), and 26 of the 44 studies were performed after 1997. The mean prevalences of cancer, palpable lesions, and invasive ductal carcinoma were 54% (range, 23%–84%), 49% (range, 2%–86%), and 70% (range, 23%–100%), respectively. The mean patient age was 51 years (range, 46–58 years). The number of criteria used to differentiate benign from malignant lesions was unknown in 11, one in 14, two in nine, and three in 10 studies (Table, Table E1; http://radiology.rsnajnls.org/cgi/content/full/2461061298/DC1). Specificity varied with or across cancer prevalence and number of criteria used to differentiate benign from malignant lesions: It decreased with increasing cancer prevalence (specificities for low, moderate, and high prevalence: 0.81, 0.71, and 0.61, respectively) and was higher when two criteria (0.81) were used to differentiate benign from malignant lesions than when one (0.74) or all three (0.67) criteria were used (Table). These findings are consistent with the larger variability in specificity across studies (Fig 2).


View this table:
[in this window]
[in a new window]

  		Results of Bivariate Analysis with Covariates

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
 References
 
In our meta-analysis, we calculated an overall sensitivity of 0.90 (95% CI: 0.88, 0.92) and an overall specificity of 0.72 (95% CI: 0.67, 0.77). Specificity varied across the individual studies more than sensitivity did, and it varied with or across cancer prevalence and number of criteria used to differentiate benign from malignant lesions.

In one study, Sherif et al (43) reported a sensitivity (0.52) that was substantially lower than the sensitivities in the other studies; however, the reported specificity was 1.00. We believe that this deviation was due to the fact that a cutoff level was chosen to optimize specificity, and this resulted in a low sensitivity. If the line of the sROC curve were extrapolated, it could be seen that this study also lay on the sROC curve; this finding suggests that there was a difference in the chosen threshold only.

In most diagnostic meta-analyses, results are summarized in an sROC curve, a DOR, or the Q point (the point where sensitivity equals specificity)—measures that are less useful to clinicians. By using the bivariate approach for diagnostic meta-analysis, we derived overall estimates of sensitivity and specificity. At a sensitivity of 0.90, we derived a specificity of 0.72, meaning that if women with a breast lesion detected at mammography, ultrasonography (US), or physical examination were referred for MR imaging, 28% of the lesions would be false-positive for cancer. However, if more precise mammography, US, and/or physical examination data were taken into account, our estimates of the diagnostic performance of breast MR imaging might have been influenced.

A previous meta-analysis included 16 studies that were published between 1994 and 1997 (59). The results were presented in an sROC curve, the Q point, and a likely optimal point (sensitivity of 0.95 with corresponding specificity of 0.67). We thought it relevant to perform a new meta-analysis of the diagnostic performance of breast MR imaging because numerous studies were performed after 1997, MR imaging technology has since improved substantially, and the bivariate approach for diagnostic meta-analysis, which facilitates more accurate summary estimates of sensitivity and specificity, has since become available. Thus, our results are difficult to compare with those of Hrung et al (59). At a sensitivity of 0.95, we derived a specificity of approximately 0.55 in our sROC curve analysis, compared with the specificity of 0.67 derived by Hrung et al. We included an additional 24 studies that were published after the previous meta-analysis. Despite improvements in MR technique and interpretation ability over time, our estimates for the diagnostic performance of breast MR imaging were slightly lower than those derived in the previous meta-analysis. We hypothesize that this can be explained by the inclusion of patients with more apparent breast lesions in the earlier studies; patients with more clinically relevant lesions were included in the later studies, and, thus, the study populations had higher proportions of nonpalpable and noninvasive lesions.

Our data show that specificity varied with or across cancer prevalence. However, from a mathematical point of view, sensitivity and specificity cannot be affected by prevalence. We believe that disease prevalence influences specificity indirectly—for example, through differences in disease severity or other characteristics across studies (62,63). We planned to perform a multivariable analysis to determine whether cancer prevalence still affected specificity after correction for other factors. We were not able to perform such an analysis, however, because of the relatively small number of included studies and because the number of missing data was too large.

We cannot fully explain why the specificity was higher when two criteria for malignancy were used than when all three criteria were used. We expected sensitivity and specificity to be highest in the subgroup of studies involving the use of all three criteria. Many studies were performed before the introduction of guidelines for interpretation of MR mammograms (72). It could be that the radiologists in the studies with higher sensitivity and specificity used all three criteria but did not describe them as three separate criteria. Furthermore, eight of the 10 studies with three criteria used were performed outside of the United States. Since specificity was lower in the non-U.S. studies, the apparent influence of the number of malignancy criteria on specificity could have been related to the difference in the country of publication between the studies. However, these effects could not be differentiated in our study because a multivariable analysis was not feasible owing to missing data and the relatively small number of included studies.

We did not make adjustments for multiple testing. The use of adjusted P values would have prevented false-positive findings—but at the cost of missed relevant associations (ie, higher rate of type 2 errors). In addition, adjustment of P values—if applied at all—should be considered in studies that are merely "fishing expeditions," "data dredging," or "hypotheses generating." This was not the case in our study. We therefore believe that the two positive findings for subgroup analyses that we report are not false (73,74).

Our study had limitations. First, the quality of a meta-analysis depends on the quality of the individual studies included in the meta-analysis (65). We did not include a quality score on our data extraction sheet because all available quality scores are highly subjective (60). By applying strict inclusion and exclusion criteria, we included only those studies that were of sufficient quality to allow estimates of the sensitivity and specificity of breast MR imaging. We searched the Medline database and cross-referenced published meta-analyses and reviews on this subject. Owing to the large number of articles identified, we did not search for nonpublished studies.

Second, predictive values, such as probability of cancer given a positive or negative MR result, are even more useful in clinical practice because patients and physicians are even more interested in the predictive value of combinations of diagnostic test results than they are in the predictive value of a single test result (64). In the case of breast cancer, such a set could include patient characteristics and conventional mammography, US, and breast MR imaging results. In the majority of studies included in our meta-analysis, combined results of different diagnostic tests were not reported, so it was impossible to report such combination data. However, there are as yet no available methods to properly pool the predictive values of test results—let alone the predictive values of combinations of test results.

Third, the results of our covariate analysis should be interpreted with caution because certain studies did not include information on several relevant characteristics, so there was a substantial number of missing values for some of the studied characteristics. Thus, further research is required to make more precise statements about the effect of covariates on sensitivity and specificity.

In conclusion, all currently available evidence indicates that MR imaging has high sensitivity (0.90) and lower specificity (0.72) in patients referred for biopsy of a breast lesion. We believe that further research is required to determine the diagnostic performance of breast MR imaging in patients with nonpalpable breast lesions.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
 References
 


   IMPLICATION FOR PATIENT CARE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
References
 


   ACKNOWLEDGMENTS
 
The authors thank Evelyn Monninkhof, PhD, for assistance in planning the meta-analysis, Cees Haaring for constructing the database and technical support, Lukas Bachmann, MD, PhD, for help with the sROC analysis, and Johannes Reitsma, MD, PhD, for help with the bivariate analysis.


   FOOTNOTES
 

Abbreviations: CI = confidence interval • DOR = diagnostic odds ratio • sROC = summary receiver operating characteristic

Guarantor of integrity of entire study, P.H.M.P.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, N.H.G.M.P., I.H.M.B.R., W.P.T.M.M., P.H.M.P.; statistical analysis, N.H.G.M.P., N.P.A.Z., K.G.M.M., P.H.M.P.; and manuscript editing, all authors

Authors stated no financial relationship to disclose.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
IMPLICATION FOR PATIENT CARE...
 References
 

  1. Boetes C, Mus RD, Holland R, et al. Breast tumors: comparative accuracy of MR imaging relative to mammography and US for demonstrating extent. Radiology 1995;197(3):743–747.[Abstract/Free Full Text]
  2. Kacl GM, Liu P, Debatin JF, Garzoli E, Caduff RF, Krestin GP. Detection of breast cancer with conventional mammography and contrast-enhanced MR imaging. Eur Radiol 1998;8(2):194–200.[CrossRef][Medline]
  3. Schelfout K, Van Goethem M, Kersschot E, et al. Contrast-enhanced MR imaging of breast lesions and effect on treatment. Eur J Surg Oncol 2004;30(5):501–507.[CrossRef][Medline]
  4. Sardanelli F, Giuseppetti GM, Panizza P, et al. Sensitivity of MRI versus mammography for detecting foci of multifocal, multicentric breast cancer in fatty and dense breasts using the whole-breast pathologic examination as a gold standard. AJR Am J Roentgenol 2004;183(4):1149–1157.[Abstract/Free Full Text]
  5. Mumtaz H, Hall-Craggs MA, Davidson T, et al. Staging of symptomatic primary breast cancer with MR imaging. AJR Am J Roentgenol 1997;169(2):417–424.[Abstract/Free Full Text]
  6. Malur S, Wurdinger S, Moritz A, Michels W, Schneider A. Comparison of written reports of mammography, sonography and magnetic resonance mammography for preoperative evaluation of breast lesions, with special emphasis on magnetic resonance mammography. Breast Cancer Res 2001;3(1):55–60.[CrossRef][Medline]
  7. Hlawatsch A, Teifke A, Schmidt M, Thelen M. Preoperative assessment of breast cancer: sonography versus MR imaging. AJR Am J Roentgenol 2002;179(6):1493–1501.[Abstract/Free Full Text]
  8. Soderstrom CE, Harms SE, Copit DS, et al. Three-dimensional RODEO breast MR imaging of lesions containing ductal carcinoma in situ. Radiology 1996;201(2):427–432.[Abstract/Free Full Text]
  9. Hwang ES, Kinkel K, Esserman LJ, Lu Y, Weidner N, Hylton NM. Magnetic resonance imaging in patients diagnosed with ductal carcinoma-in-situ: value in the diagnosis of residual disease, occult invasion, and multicentricity. Ann Surg Oncol 2003;10(4):381–388.[Abstract/Free Full Text]
  10. Kristoffersen Wiberg M, Aspelin P, Sylvan M, Boné B. Comparison of lesion size estimated by dynamic MR imaging, mammography and histopathology in breast neoplasms. Eur Radiol 2003;13(6):1207–1212.[Medline]
  11. Amano G, Ohuchi N, Ishibashi T, Ishida T, Amari M, Satomi S. Correlation of three-dimensional magnetic resonance imaging with precise histopathological map concerning carcinoma extension in the breast. Breast Cancer Res Treat 2000;60(1):43–55.[CrossRef][Medline]
  12. Kriege M, Brekelmans CT, Boetes C, et al. Efficacy of MRI and mammography for breast-cancer screening in women with a familial or genetic predisposition. N Engl J Med 2004;351(5):427–437.[Abstract/Free Full Text]
  13. Leach MO, Boggis CR, Dixon AK, et al. Screening with magnetic resonance imaging and mammography of a UK population at high familial risk of breast cancer: a prospective multicentre cohort study (MARIBS). Lancet 2005;365(9473):1769–1778.[CrossRef][Medline]
  14. Kuhl CK, Schrading S, Leutner CC, et al. Mammography, breast ultrasound, and magnetic resonance imaging for surveillance of women at high familial risk for breast cancer. J Clin Oncol 2005;23(33):8469–8476.[Abstract/Free Full Text]
  15. Brix G, Henze M, Knopp MV, et al. Comparison of pharmacokinetic MRI and [18F] fluorodeoxyglucose PET in the diagnosis of breast cancer: initial experience. Eur Radiol 2001;11(10):2058–2070.[CrossRef][Medline]
  16. Choi BG, Kim HH, Kim EN, et al. New subtraction algorithms for evaluation of lesions on dynamic contrast-enhanced MR mammography. Eur Radiol 2002;12(12):3018–3022.[Medline]
  17. Trecate G, Tess JD, Vergnaghi D, et al. Breast microcalcifications studied with 3D contrast-enhanced high-field magnetic resonance imaging: more accuracy in the diagnosis of breast cancer. Tumori 2002;88(3):224–233.[Medline]
  18. Flanagan FL, Murray JG, Gilligan P, Stack JP, Ennis JT. Digital subtraction in Gd-DTPA enhanced imaging of the breast. Clin Radiol 1995;50(12):848–854.[CrossRef][Medline]
  19. Ikeda O, Yamashita Y, Morishita S, et al. Characterization of breast masses by dynamic enhanced MR imaging: a logistic regression analysis. Acta Radiol 1999;40(6):585–592.[Medline]
  20. Cecil KM, Schnall MD, Siegelman ES, Lenkinski RE. The evaluation of human breast lesions with magnetic resonance imaging and proton magnetic resonance spectroscopy. Breast Cancer Res Treat 2001;68(1):45–54.[CrossRef][Medline]
  21. Turkat TJ, Klein BD, Polan RL, Richman RH. Dynamic MR mammography: a technique for potentially reducing the biopsy rate for benign breast disease. J Magn Reson Imaging 1994;4(4):563–568.[CrossRef][Medline]
  22. Tiling R, Khalkhali I, Sommer H, et al. Limited value of scintimammography and contrast-enhanced MRI in the evaluation of microcalcification detected by mammography. Nucl Med Commun 1998;19(1):55–62.[Medline]
  23. Alamo L, Fischer U. Contrast-enhanced color Doppler ultrasound characteristics in hypervascular breast tumors: comparison with MRI. Eur Radiol 2001;11(6):970–977.[CrossRef][Medline]
  24. Reinikainen H, Paakko E, Suramo I, Paivansalo M, Jauhiainen J, Rissanen T. Dynamics of contrast enhancement in MR imaging and power Doppler ultrasonography of solid breast lesions. Acta Radiol 2002;43(5):492–500.[CrossRef][Medline]
  25. Heinisch M, Gallowitsch HJ, Mikosch P, et al. Comparison of FDG-PET and dynamic contrast-enhanced MRI in the evaluation of suggestive breast lesions. Breast 2003;12(1):17–22.[CrossRef][Medline]
  26. Sinha S, Lucas-Quesada FA, DeBruhl ND, et al. Multifeature analysis of Gd-enhanced MR images of breast lesions. J Magn Reson Imaging 1997;7(6):1016–1026.[Medline]
  27. Bagni B, Franceschetto A, Casolo A, et al. Scintimammography with 99mTc-MIBI and magnetic resonance imaging in the evaluation of breast cancer. Eur J Nucl Med Mol Imaging 2003;30(10):1383–1388.[CrossRef][Medline]
  28. Gibbs P, Liney GP, Lowry M, Kneeshaw PJ, Turnbull LW. Differentiation of benign and malignant sub-1 cm breast lesions using dynamic contrast enhanced MRI. Breast 2004;13(2):115–121.[CrossRef][Medline]
  29. Huang W, Fisher PR, Dulaimy K, Tudorica LA, O'Hea B, Button TM. Detection of breast malignancy: diagnostic MR protocol for improved specificity. Radiology 2004;232(2):585–591.[Abstract/Free Full Text]
  30. Stomper PC, Herman S, Klippenstein DL, et al. Suspect breast lesions: findings at dynamic gadolinium-enhanced MR imaging correlated with mammographic and pathologic features. Radiology 1995;197(2):387–395.[Abstract/Free Full Text]
  31. Obdeijn IM, Kuijpers TJ, van Dijk P, Wiggers T, Oudkerk M. MR lesion detection in a breast cancer population. J Magn Reson Imaging 1996;6(6):849–854.[Medline]
  32. Palmedo H, Grunwald F, Bender H, et al. Scintimammography with technetium-99m methoxyisobutylisonitrile: comparison with mammography and magnetic resonance imaging. Eur J Nucl Med 1996;23(8):940–946.[CrossRef][Medline]
  33. Kinkel K, Helbich TH, Esserman LJ, et al. Dynamic high-spatial-resolution MR imaging of suspicious breast lesions: diagnostic criteria and interobserver variability. AJR Am J Roentgenol 2000;175(1):35–43.[Abstract/Free Full Text]
  34. Schedel H, Oellinger H, Kohlschein P, et al. Magnetic resonance female breast imaging (MRFBI): evaluation of the changes in signal intensity over time pre- and post-administration of 0.2 mmol/kg Gd-DTPA. Zentralbl Gynakol 2002;124(2):104–110.[CrossRef][Medline]
  35. Buadu LD, Murakami J, Murayama S, et al. Breast lesions: correlation of contrast medium enhancement patterns on MR images with histopathologic findings and tumor angiogenesis. Radiology 1996;200(3):639–649.[Abstract/Free Full Text]
  36. Daldrup-Link HE, Rydland J, Helbich TH, et al. Quantification of breast tumor microvascular permeability with feruglose-enhanced MR imaging: initial phase II multicenter trial. Radiology 2003;229(3):885–892.[Abstract/Free Full Text]
  37. Blohmer JU, Oellinger H, Schmidt C, Hufnagl P, Felix R, Lichtenegger W. Comparison of various imaging methods with particular evaluation of color Doppler sonography for planning surgery for breast tumors. Arch Gynecol Obstet 1999;262(3-4):159–171.[CrossRef][Medline]
  38. Kelcz F, Furman-Haran E, Grobgeld D, Degani H. Clinical testing of high-spatial-resolution parametric contrast-enhanced MR imaging of the breast. AJR Am J Roentgenol 2002;179(6):1485–1492.[Abstract/Free Full Text]
  39. Hulka CA, Edmister WB, Smith BL, et al. Dynamic echo-planar imaging of the breast: experience in diagnosing breast carcinoma and correlation with tumor angiogenesis. Radiology 1997;205(3):837–842.[Abstract/Free Full Text]
  40. Harms SE, Flamig DP, Hesley KL, et al. MR imaging of the breast with rotating delivery of excitation off resonance: clinical experience with pathologic correlation. Radiology 1993;187(2):493–501.[Abstract/Free Full Text]
  41. Helbich TH, Becherer A, Trattnig S, et al. Differentiation of benign and malignant breast lesions: MR imaging versus Tc-99m sestamibi scintimammography. Radiology 1997;202(2):421–429.[Abstract/Free Full Text]
  42. Kelcz F, Santyr GE, Cron GO, Mongin SJ. Application of a quantitative model to differentiate benign from malignant breast lesions detected by dynamic, gadolinium-enhanced MRI. J Magn Reson Imaging 1996;6(5):743–752.[Medline]
  43. Sherif H, Mahfouz AE, Oellinger H, et al. Peripheral washout sign on contrast-enhanced MR images of the breast. Radiology 1997;205(1):209–213.[Abstract/Free Full Text]
  44. Heiberg EV, Perman WH, Herrmann VM, Janney CG. Dynamic sequential 3D gadolinium-enhanced MRI of the whole breast. Magn Reson Imaging 1996;14(4):337–348.[CrossRef][Medline]
  45. Boetes C, Barentsz JO, Mus RD, et al. MR characterization of suspicious breast lesions with a gadolinium-enhanced TurboFLASH subtraction technique. Radiology 1994;193(3):777–781.[Abstract/Free Full Text]
  46. Fobben ES, Rubin CZ, Kalisher L, Dembner AG, Seltzer MH, Santoro EJ. Breast MR imaging with commercially available techniques: radiologic-pathologic correlation. Radiology 1995;196(1):143–152.[Abstract/Free Full Text]
  47. Lucht RE, Knopp MV, Brix G. Classification of signal-time curves from dynamic MR mammography by neural networks. Magn Reson Imaging 2001;19(1):51–57.[CrossRef][Medline]
  48. Kristoffersen Wiberg M, Aspelin P, Perbeck L, Boné B. Value of MR imaging in clinical evaluation of breast lesions. Acta Radiol 2002;43(3):275–281.[CrossRef][Medline]
  49. Liu PF, Debatin JF, Caduff RF, Kacl G, Garzoli E, Krestin GP. Improved diagnostic accuracy in dynamic contrast enhanced MRI of the breast by combined quantitative and qualitative analysis. Br J Radiol 1998;71(845):501–509.[Abstract]
  50. Abdolmaleki P, Buadu LD, Naderimansh H. Feature extraction and classification of breast cancer on dynamic magnetic resonance imaging using artificial neural network. Cancer Lett 2001;171(2):183–191.[CrossRef][Medline]
  51. Gilles R, Meunier M, Lucidarme O, et al. Clustered breast microcalcifications: evaluation by dynamic contrast-enhanced subtraction MRI. J Comput Assist Tomogr 1996;20(1):9–14.[CrossRef][Medline]
  52. Orel SG, Schnall MD, Powell CM, et al. Staging of suspected breast cancer: effect of MR imaging and MR-guided biopsy. Radiology 1995;196(1):115–122.[Abstract/Free Full Text]
  53. Boné B, Péntek Z, Perbeck L, Veress B. Diagnostic accuracy of mammography and contrast-enhanced MR imaging in 238 histologically verified breast lesions. Acta Radiol 1997;38(4 pt 1):489–496.[Medline]
  54. Kuhl CK, Mielcareck P, Klaschik S, et al. Dynamic breast MR imaging: are signal intensity time course data useful for differential diagnosis of enhancing lesions? Radiology 1999;211(1):101–110.[Abstract/Free Full Text]
  55. Nunes LW, Schnall MD, Orel SG. Update of breast MR imaging architectural interpretation model. Radiology 2001;219(2):484–494.[Abstract/Free Full Text]
  56. Fischer U, Kopka L, Grabbe E. Breast carcinoma: effect of preoperative contrast-enhanced MR imaging on the therapeutic approach. Radiology 1999;213(3):881–888.[Abstract/Free Full Text]
  57. Fischer DR, Baltzer P, Malich A, et al. Is the "blooming sign" a promising additional tool to determine malignancy in MR mammography? Eur Radiol 2004;14(3):394–401.[CrossRef][Medline]
  58. Bluemke DA, Gatsonis CA, Chen MH, et al. Magnetic resonance imaging of the breast prior to biopsy. JAMA 2004;292(22):2735–2742.[Abstract/Free Full Text]
  59. Hrung JM, Sonnad SS, Schwartz JS, Langlotz CP. Accuracy of MR imaging in the work-up of suspicious breast lesions: a diagnostic meta-analysis. Acad Radiol 1999;6(7):387–397.[CrossRef][Medline]
  60. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC Med Res Methodol 2003;3:25.[CrossRef][Medline]
  61. Bossuyt PM, Reitsma JB, Bruns DE, et al. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD initiative. Radiology 2003;226(1):24–28.[Abstract/Free Full Text]
  62. Fletcher RH. Carcinoembryonic antigen. Ann Intern Med 1986;104(1):66–73.[CrossRef][Medline]
  63. Hlatky MA, Pryor DB, Harrell FE Jr, Califf RM, Mark DB, Rosati RA. Factors affecting sensitivity and specificity of exercise electrocardiography: multivariable analysis. Am J Med 1984;77(1):64–71.[CrossRef][Medline]
  64. Moons KG, van Es GA, Deckers JW, Habbema JD, Grobbee DE. Limitations of sensitivity, specificity, likelihood ratio, and Bayes' theorem in assessing diagnostic probabilities: a clinical example. Epidemiology 1997;8(1):12–17.[Medline]
  65. Lijmer JG, Mol BW, Heisterkamp S, et al. Empirical evidence of design-related bias in studies of diagnostic tests. JAMA 1999;282(11):1061–1066.[Abstract/Free Full Text]
  66. Littenberg B, Moses LE. Estimating diagnostic accuracy from multiple conflicting reports: a new meta-analytic method. Med Decis Making 1993;13(4):313–321.[Abstract/Free Full Text]
  67. Sweeting MJ, Sutton AJ, Lambert PC. What to add to nothing? use and avoidance of continuity corrections in meta-analysis of sparse data. Stat Med 2004;23(9):1351–1375.[CrossRef][Medline]
  68. 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(11):1129–1135.[CrossRef][Medline]
  69. Irwig L, Macaskill P, Glasziou P, Fahey M. Meta-analytic methods for diagnostic test accuracy. J Clin Epidemiol 1995;48(1):119–130.[CrossRef][Medline]
  70. Moses LE, Shapiro D, Littenberg B. Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. Stat Med 1993;12(14):1293–1316.[Medline]
  71. Reitsma JB, Glas AS, Rutjes AW, Scholten RJ, Bossuyt PM, Zwinderman AH. Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. J Clin Epidemiol 2005;58(10):982–990.[CrossRef][Medline]
  72. Schnall MD, Ikeda DM. Lesion Diagnosis Working Group report. J Magn Reson Imaging 1999;10(6):982–990.[CrossRef][Medline]
  73. Feise RJ. Do multiple outcome measures require p-value adjustment? BMC Med Res Methodol 2002;2:8.[CrossRef][Medline]
  74. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology 1990;1(1):43–46.[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Supplemental Table
Right arrow All Versions of this Article:
2461061298v1
246/1/116    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Peters, N. H. G. M.
Right arrow Articles by Peeters, P. H. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Peters, N. H. G. M.
Right arrow Articles by Peeters, P. H. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE