Diagnostic Effectiveness of Magnetic Resonance Imaging Compared to Mammography - Research Paper Example

A Systematic Review of the Diagnostic Effectiveness of Magnetic Resonance Imaging (MRI) Compared to Mammography when Imaging Young, Asymptomatic, High-Risk Women For Breast Cancer. Could MRI Replace Mammograms? ABSTRACT Table of Contents Page Number Content 4 Introduction and Literature Review 6 Aims/Objectives 7 Methodology 12 Results and Discussion 19 Summary 20 Conclusions and Recommendations 21 References 23 Appendix A Introduction and Literature Review Breast cancer is the most common form of cancer in women and the second highest cause of cancer death in women. In 2005 alone, 45,947 women were diagnosed with breast cancer in the UK and in 2006, 12,392 women died from the disease (NICECG80 2009). The “average” woman has an 11% risk (1 in 9 women) of developing breast cancer during their lifetime. Breast cancer can develop from a combination of lifestyle, environment, and genetic effects. BRCA1 and BRCA2 are gene mutations that confer susceptibility to breast cancer. Some women have a higher risk of developing the disease if they have either of these mutations and have a strong family history of breast cancer. The risk is greater than 8% between the age of 40 and 50 years, and a lifetime risk of 30% or greater (NICECG14 2004). It is believed that 5 – 10% of new cases of breast cancer each year have a hereditary cause. An additional 15 – 20% of newly diagnosed cases are referred to as familial breast cancer (Lynch & Lynch 2006). A family history of breast cancer is a strong risk factor for developing the disease, although this risk factor varies considerably with the type of genetic mutation involved (NICECG14 2004). Breast screening with mammography is available every 3 years for all women aged 50 – 70 years old (NHSBSP 2008). Screening mammography is a very effective tool for women in this age group, but younger women in the high-risk grouping have denser breast tissue, which can often mask the pathology on a mammogram (Arun and Kuerer, 2006). Generally, women under the age of 50 are not routinely screened for breast cancer, as those most at risk are aged 50 and beyond. Early studies have shown the possibility of identifying young women who have high risks of getting breast cancer using physical examination and mammography (Brekelmans, et al., 2001). However, the sensitivity of these tests is quite low. Young high-risk women also need annual breast cancer surveillance, although they are the most difficult to diagnose through mammography. Younger women (those under 50) have denser breast tissue, because the breast tissues are still mainly composed of fat. This radiographically dense breast tissue can mask small lesions giving a false negative diagnosis. MRI is a good alternative to mammography in detecting breast cancer in women with dense breast tissue as shown by data from multi-modality screening studies (Schnall, 2006). MRI has been used since the 1990s as a follow-up test to suspicious mammograms. Now it would seem there could be a role for it as a first line technology in breast cancer screening of high-risk young women. MRI results are not affected by breast density and MRI can detect tissue vascularity that cannot be seen on mammography. As most breast cancers are vascular, intravenous gadolinium will enhance the tumour visualization, which makes the imaging results easier to analyze and interpret (Liberman, 2004). Many prospective studies published since 2000 show that MRI has 100% sensitivity for detecting breast cancer (Kuhl 2000; Warner 2001; Robson 2003). This is much higher than the sensitivity obtained for mammography, especially for dense breast tissue where sensitivity of mammography is inversely related to breast density (Artmann, 2006). Furthermore, many studies show that mammography has a poor sensitivity in detecting cancer in young women (MARIBS Study Group, 2005). However the specificity of MRI varies from 80% to 83%, is much lower than that for mammography which has a specificity ranging from 92-95% (MARIBS Study Group, 2005). Specificity is a measure of the capability of a technique to detect a negative result. Aims and objectives The aim of this systematic review is to answer the following question “What is the diagnostic effectiveness of magnetic resonance imaging (MRI) compared to mammography when imaging young, asymptomatic, high-risk women for breast cancer. Could MRI replace mammograms?” Mammography is considered the “gold standard” imaging modality to diagnose breast cancer. However, in young women with denser breast tissue MRI could be the most diagnostically effective modality. The objectives of the study are: 1. To critically review published studies and articles on screening young women at high risk of developing breast cancer. 2. To determine the sensitivity and specificity of both modalities for women in this group. 3. To make recommendations on how this could affect current practice and protocols. Methodology Search Parameters The PICO method (NHSCRD 2001) was used to select the appropriate studies for the critical review after searches were performed in the leading medical databases. The PICO method is a means of searching the medical databases in a systematic matter such that a more evidence-based approach is utilized. PICO stands for the four major criteria used in searching the literature: P- stands for Patient or Population studied or is of interest to the reviewer I- stands for Intervention. It answers how the therapy was administered C- means Comparison. In this study, the comparison will be between MRI and mammography O- Outcome desired or being searched. The literature search was conducted using the leading online electronic databases: Cumulative Index to Nursing and Allied Health Literature (CINAHL), Excerpta Medica Database (EMBASE), Medical Literature Analysis and Retrieval System Online (MEDLINE), and the National Health Service Centre for Reviews and Dissemination (NHS CRD). The databases were chosen because of their extensive collection of medical literature. MEDLINE is the premier bibliographic database of the US National Library of Medicine. It covers 16 million references from 5200 journals in 37 languages that primarily concern medicine. The records that are deposited in MEDLINE are indexed with the National Library of Medicine’s Medical Subject Headings or MeSH, which is a controlled vocabulary thesaurus. Terms are arranged structurally so that searches can be conducted at specific levels, increasing the ease of the search. Elsevier produces EMBASE, short for Excerpta Medica Database. EMBASE contains over 11 million records of the biomedical and pharmacological topics from more than 7000 peer-reviewed journals. Owned and operated by EBSCO Publishing, CINAHL offers the most comprehensive resource for literature on nursing and allied health. NHS CRD works exclusively in evidence synthesis in the health field. Part of the University of York in the UK, CRD systematically evaluates research evidence on health issues, findings of which have significant impacts on the formulation of health care policy. The databases were searched separately in order to make use of the search engines thesaurus to explode the search area. The following key words and Boolean commands were used: breast cancer screening OR surveillance AND MRI AND mammography AND high risk AND biopsies. These were exploded to include breast neoplasms OR breast carcinoma, magnetic resonance imaging, and familial OR hereditary breast cancer. Where available, MeSH, Medical Subject Headings, was deployed using the key words as sub-headings. Initially the search was specified for the article title and abstract, to ensure that enough data are available to give a large and sufficient sample size of the literature for the systematic review. Sample sizes of the articles were expected to be small as this population group is quite small, considering that the screenings of the women subjects are performed in tertiary care centres. The search was later expanded to articles where the full text is available or where printed copies can be found. Search Limits and Inclusion Criteria For the search strategy some limits were applied to define the boundaries of the review. Only studies on the screening of women were included, since male cancers are rare and asymptomatic men are not normally tested (American Cancer Society, 2010). Although no upper age limit of the women subjects was set, the studies that were selected were primarily about young high-risk women, since women aged over 50 are already screened under the NHS Breast Screening Programme. Finding studies exclusive for women aged 50 and below was difficult, since most studies covered a wide range of ages. Thus, the mean age was considered, which was below 50 years old in all the selected studies. Only studies in English were reviewed to avoid any problems with the logistics involved in translating foreign language work. Another criterion for inclusion was the region where these were conducted; only the studies conducted in western countries were selected. Since the incidence and mortality rates for breast cancer vary across the world, studies from Asia or Africa cannot be considered to be relevant to medical practice in the UK because of the differences in ethnic subgroups, of which there are many, in Asian and African populations (McCormack, Mangtani, Bhakta, McMichael, & Santos Silva, 2004). Studies that are less than or at least 10 years old were considered for inclusion. The possibility of reducing the number of years since the articles came out to at least five years was considered. The basis for this criterion was the assumption that MRI techniques can change rapidly. However, this requirement was not met because long-term screening studies that were selected were only published after they were finished or at most after the second round of screening. The studies selected were quantitative, and presented sensitivity and specificity data for cancer detection, which made comparison of the results easier. Nevertheless, these studies were published within the last ten years. This systematic review is not concerned with the cost-effectiveness of the diagnostic tests and any data on this was ignored. Analysis of selected articles A systematic review of specific literature topics requires an appraisal tool for the critical analysis of the selected studies (Public Health Resource Unit, England, 2006). The Public Health Resource Unit, a public health consultancy of the National Health Service of UK, developed the appraisal tool used in this study. The unit has developed the Critical Appraisal Skills Programme that is designed specifically to evaluate different types of research studies. For appraising diagnostic tests like MRI and mammography, the programme formulated a twelve-question checklist to help in assessing the methodology used, results obtained, the validity of the results and applications of the study results to the a certain patient population (Appendix A) (Public Health Resource Unit, England, 2006). A twelve-question checklist is available from the unit. The first two questions in the checklist serve as screeners. They concern the validity of the study, and if the answer is “no” to either of the two questions, then the article was discarded, and not considered for inclusion into this systematic review. The remaining 10 questions were more detailed, dwelling on the nature and clarity of the results. The final question answered was the impact of the results on the patients and the general population. One further question was added to this appraisal tool; the question “Was biopsy performed to confirm diagnosis?” was the thirteenth question in the checklist. This was a necessary addition because MRI can detect lesions that are not visualized in mammograms, the technique that is considered the “gold standard” modality for breast cancer screening (Arun & Kuerer 2006). Thus, the inclusion criteria were further broadened to include the performance of biopsy to confirm the diagnosis of cancer in MRI-detected lesions. All of the studies reviewed were diagnostic test studies as this is a comparison between two modalities. Randomised control trials, which are often considered to produce the best evidence, were not used for this review, as women would have to be unethically denied the gold standard of mammography (Parahoo, 2006). Ideally, the appraisal tool was to be used by two or more reviewers to avoid subjectivity; however, as this is a solo project, some bias may be unavoidable. Presentation of results The major results of the included studies were tabulated to highlight the differences (heterogeneity) and similarities (homogeneity). Ideally, an electronic data extraction form should be developed as an Access/Excel file to record and handle the data (NHS CRD 2001). The form will contain data such as Date, Title, and Author of the study; the length of any follow-ups; if any participants dropped out; any statistical data; p-value, sensitivity, specificity, and positive predictive value. After the results were analyzed, generalizations and recommendations for practice are given. Results and Discussion From the extensive search conducted on the different biomedical databases, only five articles were chosen. These were the following: 1. Kriege, M., Brekelman, C., Boetes, C., Besnard, P., Zonderland, H., Obdeijn, I., et al. (2004). Efficacy of MRI and mammography for breast-cancer screening in women with a familial or genetic predisposition. The New England Journal of Medicine, 351(5):427-437. 2. Kuhl, C., Schrading, S., Leutner, C., Morkkabati-Spitz, N., Wardelmann, E., Fimmers, R., et al. (2005). Mammography, breast ultrasound, and magnetic resonance imaging for surveilance of women at high familial risk for breast cancer. Journal of clinical Oncology , 23:8469-8476. 3. MARIBS Study Group. (2005). Screening with magnetic resonance imaging and mamography of a UK population at high familial risk of breast cancer: a prospective multi-cohort study. Lancet, 365:1769-1778. 4. Sardanelli, F., Podo, F., D'Agnolo, G., Verdecchia, A., Santaquilani, M., Musumeci, R., et al. (2007). Multicenter comparative multimodality surveillance of women at genetic-familial high risk for breast cancer (HIBCRIT Studu): Interim Results. Radiology , 242(3):698-715. 5. Warner, E., Plewes, D., Hill, K., Causer, P., Zubovits, J., Jong, R., et al. (2004). Surveillance of BRCA1 and BRCA2 mutation carriers with magenetic resonance imaging, ultrasound, mammography and clinical breast examination. JAMA , 292:1317-1325. The five articles described the methodology and results of prospective, non-randomized studies that compare MRI with mammography (Table 1). The studies have a similar general objective, to compare the efficacy of MRI and mammography in detecting breast cancers. However, they were different with respect to the size or number of women, the categories under which they classified the women, and the scope of screening, the methods used and result interpretation. Except for the study by Warner et al., which was conducted in Canada, all the studies were conducted in Europe. The studies by Kuhl et al. (2005) and Warner et al. (2004) were conducted in single centres, while all others were conducted in multiple centres. The MARIBS Study Group had the most number of centres (22), followed by Sardanelli et al. (17), and Kriege et al. (6). Sardanelli et al. (2007) reported the shortest study results; the results they presented were only for a year’s study period. The longest study conducted was by the MARIBS Study Group (2005) which ran from August 1997 to May 2004. Follow-up period was longest at 5.3 years by Kuhl et al. (2005). The biggest subject base was handled by Kriege et al. (2004), which included 1909 women, while the study of Warner et al. (2004) had the least (236) women subjects. Kriege and colleagues gave data on the number of women who withdrew from the screening; the reasons for the withdrawals included recurrence of cancer, claustrophobia from the MRI procedure, mastectomy, and pregnancy. The age range of women was 25-79 years, with mean age from 40 to 47 years. The youngest women group had ages from 31 to 55 years (MARIBS Study Group, 2005), while the oldest women can be found in the study conducted in Italy (Sardanelli, et al., 2007). Looking at the age data, it is obvious that the original criterion for the age of women to be less than fifty years old in the desired studies was not met. No reports were published for the exclusive screening of women who were younger than fifty. The data can be extracted from the individual studies but this will require access to the original data of the research groups, which is not within the scope of this review. The common risk criterion in all studies was being a carrier of the gene mutations BRCA1 or BRCA2. High familial risk was the second most common criterion for inclusion into the study. Two of the studies excluded patients with a history of breast cancer (MARIBS Study Group, 2005) (Kriege, et al., 2004), while the other 3 studies included women who have had breast cancer in the past but did not show signs or metastasis at the time of study recruitment. The group of Kriege classified the risks grouping of the women further based on experience with previous screening for breast cancer, menopausal status, use of hormone replacement therapy and oophorectomy (surgical removal of the ovaries). They categorized the women into mutation carriers, high risk, and moderate risk groups. Kuhl et al. categorized the subjects based on their personal history of cancer, mutation carriers, and moderate and high-risk groups. The MARIBS Study group studied only women with known mutations and with relatives having the mutations in BRCA1 and BRCA2. This was also the criteria used by Sardanelli et al. but those with history of ovarian cancer aside from breast cancer were included. Warner and colleagues included only women who tested positive for mutations in the BRCA1 and BRCA2 genes. Table 1. Details of the prospective studies on screening breast mammography and magnetic resonance imaging. Authors/ Date Published Design (#of Centres) Period Conducted/ Follow-up No. Women Studied/ /Age Range/ Mean Age Risk Criteria in Addition to Mutation Carriers Other Tests Total Number of MRI examinations MRI Plane Kriege et al., 2004 Non-randomized, prospective multi-centre study (6) Nov 1, 1999-Dec 1, 2003/2.9 years 1909/ 25-70/40 ≥ 15% lifetime familial risk Clinical breast exam (CBE) 4169 Axial Kuhl et al., 2005 Non-randomized single-centre study (1) Feb 1996-Feb 2002/5.3 years 529/27-59/ 42 ≥ 15% lifetime familial risk ultrasound 1542 Axial MARIBS Study Group, 2005 Non-randomized, prospective multi-centre study (22) August 1997-May 2004/7 years 649/31-55/40  ≥ 0.9% annual familial risk None 1881 Coronal Sardinelli et al., 2007 Non-randomized, prospective multi-centre study (17) 2002-2003/ 1 year 278/ 25-79/46 High familial risk CBE, ultrasound 377 Coronal/ Axial Warner et al., 2004 Non-randomized, single centre study (1) November 3, 1997- March 31, 2003 236/25-65/47 None CBE, ultrasound 457 Coronal/ Axial Surveillance tests were all performed more once on the patients for tumours and signs of cancer were using mammography and MRI. Additional tests conducted were clinical breast examination (Kriege, et al., 2004), ultrasound (Kuhl, et al., 2005), and breast examination plus ultrasound (Sardanelli, et al., 2007) (Warner, et al., 2004). The most number of MRI examinations were carried out by Kriege’s group, which is expected since the group had the most number of subjects; Sardanelli et al. (2007) reported the least number of examinations averaging at only 1.36 examinations per woman. This is understandable because Sardanelli and colleagues reported only the interim results of their study. In all the five studies, the MRI and mammographic examinations were conducted within ninety days of each other. More than two rounds of screening were also conducted in all five studies evaluated. However, patients who received only a single round were excluded from the reporting of the results. Sardanelli et al. did not mention of any follow-up of the patients who underwent the first round of screening. In the study by the MARIBS Study Group, images were analyzed by two readers. It also considered only the higher BI-RADS scores 4-5 as a positive for breast cancer. All the studies used contrast enhanced MRI with gadolinium diethylenetriamine pentaacetic acid as contrast agent. Several sets of post injection images were taken to obtain information about tumour characteristics. In the studies which were conducted in Europe (Kriege, et al., 2004; Kuhl, et al., 2005; MARIBS Study Group, 2005), the MRI imaging was done on either coronal or axial plane while Warner’s group used coronal and sagittal imaging. Differences in the images produced were attributed to variability in the hardware/equipment used, and imaging software. The number of cancers detected using MRI and mammography are presented in Table 2. Kriege’s group detected 45 cancers in all, while Sardanelli et al. reported the least number of cancers detected (18). The total number of cancers detected by MRI alone was significantly higher compared to the number detected by mammography alone. Results of the MARIBS study group, which did not use other screening techniques, showed MRI detection to be 81% of all cancers. Another study was not able to detect cancer using mammography alone, but detected 35% of all cancers using MRI. For studies with two or more rounds of screening, the interval cancer rates were less than 10%. Table 2. Cancers detected in studies on screening MRI and mammography. Authors/ Date Published Number of Cancers Detected By Mammography By MRI By MRI+Mammography Total Cancers Kriege et al., 2004 18 32 45 Kuhl et al., 2005 14 39 40 43 MARIBS Study Group, 2005 14 27 33 33 Sardinelli et al., 2007 6 18* Warner et al., 2004 8 17 22 *detected by a combination of CBE, ultrasound, mammography and MRI Three studies have higher detection rates for ductal carcinoma in situ or DCIS (Kuhl, et al., 2005) (Warner, et al., 2004) (Sardanelli, et al., 2007). DCIS is the most common and earliest to develop noninvasive cancer detected in women. The cancer is contained in the milk ducts of the breast, and can be difficult to detect. MRI detected most of the non-invasive cancers in these studies only. In contrast, the studies by Kriege et al. and the MARIBS study group had lower rates of detection for DCIS, and most of the noninvasive cancers here were detected only by mammography. Except for the Kuhl study, data on the screening by year was available. Most invasive tumours were node-positive. To calculate the sensitivity of the tests, biopsy was needed to confirm the cancers. Kuhl et al., the MARIBS Study Group, and Sardinelli et al. were more conservative in declaring that cancer was positive, because they only considered cancer when the image has a BI-RADS score of 4-5. All the other studies considered a score of 3, 4 or 5 as positive. Sensitivity of a clinical test or screening is the probability that a patient who tested positive has the disease; while specificity is the chance that a person who has no disease is tested to be negative. The PPV predicts if an individual who tested positive for a disease indeed has the disease. Table 3 presents the sensitivity, specificity and the positive predictive value (PPV) of cancer detection using mammography alone, MRI alone and MRI plus mammography. Most of the values presented in Table 3 were taken from the journal articles; some values were calculated based on the available data (Warner, Messersmith, Causer, Eisen, Shumak, & Plewes, 2008). Table 3. Reported sensitivity and specificity of breast screening MRI and mammography in detecting cancer in high-risk women. Reference   BI-RADS Score Mammography MRI MRI and Mammography Sensitivity % Specificity % PPV % Sensitivity % Specificity % PPV % Sensitivity % Specificity % PPV % Kriege et al., 2004 4 or 5 33.3 99 27 64 96 16 NR NR NR 0, 3, 4, 5 40 95 8 71 90 7 89* NR NR Kuhl et al., 2005 4 or 5 32.6 96.8 23.7 90.7 97.2 50 93 96.1 42.1 MARIBS Study Group, 2005 4 or 5 14* 98* 15* 51* 96* 21* 60* 95* 20* 0, 3, 4, 5 40 93 10 77 81 7 94 77 7 Sardinelli et al., 2007 4 or 5 58.8 99* 76.9 93.8 98* 62.5 100 NR NR Warner et al., 2004 4 0r 5 36 99.8 88 77 95.4 46 86* 95* 48* 0, 3, 4, 5 36* 99* 80* 82* 81* 18* 90* 80* 10* *calculated from the data (Warner, Messersmith, Causer, Eisen, Shumak, & Plewes, 2008) BI-RADS – Breast Imaging Reporting and Data System BI-RADS score as follows: 1- negative; 2- benign; 3- probably benign; 4- suspicious; 5- highly suggestive NR- not reported All five studies showed that MRI was more sensitive in detecting cancer in high-risk women; however, the mammography tests have higher specificity than MRI. When combined, the sensitivity of MRI and mammography increased, but the specificity did not improve. In most cases, the specificity was much lower than that of mammography alone. The predictive positive value was definitely higher in mammography based on the calculations done in three of the studies (Kriege, et al., 2004; Sardanelli, et al., 2007; Warner, et al., 2004). The PPV was higher when the BI-RADS score was 4-5, compared to BI-RADS score of 0,3,4,5. The higher PPV in mammography can be attributed to its greater specificity in identifying cancerous nodes. For women who have to choose between MRI and mammography, there are many considerations (aside from the cost since MRI is many times more expensive than mammography). Lower specificity should be balanced by the higher sensitivity of MRI in detection of cancers. The advantage of MRI in detecting primary breast cancers can be enhanced with improvements in the technology and additional studies. Whether cancer recurrence and death rates are reduced when the cancers detected by MRI are reduced are points that need to be further studied. Another point that needs improvement is the establishment of a positive or negative diagnosis based on the contrast enhancement from MRI; the cancer can be benign despite the contrast enhancement (Enriquez & Listinsky, 2009). Furthermore, despite the positive results for MRI screening of young high-risk women for breast cancer, the National Comprehensive Cancer Network still recommends mammography and MRI screening for very young women (aged 20-25) who are at high-risk of familial breast and ovarian cancers (Enriquez & Listinsky, 2009). In summary, five prospective, nonrandomized studies were identified from the literature that screened high-risk women for breast cancer using both mammography and MRI. The data that were presented in the studies were organized and summarized into three tables that showed study and patients’ characteristics, number of cancers detected, and the sensitivity and specificity of the tests employed. From the data presented in Table 1, it can be seen that the characteristics of the 5 studies were very variable; from the number of health centres studied to the categories of the patients. Therefore, the comparison of the studies is affected. Sensitivity and specificity of the imaging techniques could vary from patient to patient. Younger women have dense breasts, and it is more difficult to make a conclusion with the image results. Interval cancer could be higher in this women, so with late stage cancers. Thus for younger women who carry the BRCA1 and BRCA2 mutations, and have close family members with breast cancer, annual screening might not be enough to catch developing cancers early. Sensitivity estimates could also vary with the size of the study. Smaller studies will tend to overestimate the value of sensitivity. Differences in the results could also be attributed to the variability in experience and skill in coming out with the results of MRI and mammography. This could be the reason why the MARIBS Study Group used two readers to analyze and interpret the results. Another factor which affecting the sensitivity estimates is the adherence of patients to protocol; their return for the follow-up tests is important in identifying previously unseen cancerous lesions. Nevertheless, the sensitivity and specificity estimates are relatively consistent in the five studies. However, are the data enough to come to a conclusion about the sensitivity of MRI in screening high-risk women? Another question that was not answered by the 5 studies was when to start MRI screening in these women. MRI screening could be more appropriate for younger women who belong to the high-risk group, but it may not be suitable for low-risk women. Mammography is not ideal for younger women because of their higher breast density. From the data, it also appears that MRI with mammography increased specificity, therefore for high-risk women, the combination of the two techniques is the most accurate means for early cancer detection and treatment. Conclusion and Recommendations In spite of the heterogeneity in the five studies reviewed, they all came up with the same conclusion: that MRI is more sensitive in the early detection and diagnosis of cancer in women who are positive for mutations in the BRCA1 and BRCA2, and have a high familial history of breast cancer. Even for women who are not known to be carriers of the mutation but have increased familial risks, the choice of mammography alone, and the combination of mammography and breast ultrasound, is not enough for early diagnosis of breast cancer (Kuhl, et al., 2005). The use of MRI for surveillance is recommended in order to diagnose invasive and intraductal familial cancer with a higher sensitivity at an earlier time (Kuhl, et al., 2005). In mutation carriers, MRI was also found to be more sensitive in detecting cancers, compared to ultrasound, clinical breast examination, and mammography (Warner, et al., 2004). The same conclusions were arrived at by the three other studies. All studies recommend the inclusion of MRI in screening high-risk women. The use of surveillance regimens to detect cancer early in high-risk women also needs further study and should be related to the actual reduction in mortality attributed to breast cancer. However, this type of study will take years to come to a full understanding of the future benefits of MRI in breast cancer detection. The present studies point to the high potential of MRI as a sensitive tool for breast cancer protection. Its use as a routine screen for all risk groups is recommended and is possible in the near future not only for high-risk women, but for all risk groups. However, the issues identified with the routine use of MRI need to be addressed, including its low specificity and the need for biopsy to confirm findings. Nevertheless, with the current state of the art, MRI could replace mammography as a screening tool for high-risk, young women. References American Cancer Society. (2010, January 14). Detailed guide: breast cancer in men [Internet].Available from: American Cancer Society< http://www.cancer.org/docroot/cri/content/cri_2_4_3x_can_male_breast_cancer_be_found_early_28.asp> [Accessed 22 April 2010]. Artmann, A. et al. (2006) Screening in women with increased breast cancer risk. Breast Care, 1, pp. 22-25. Arun, B. and Kuerer, H. (2006) Breast Cancer Risk Assessment and Management” In Breast Cancer 2nd Ed. Winchester, D.J. et al BC Decker Inc. Brekelmans, C., Seynaeve, C., Bartels, C., Tilanus-Linthorst, E., Meijers-Heijboer, E., Crepin, C., et al. (2001) Effectiveness of breast cancer surveillance in BRCA1/2 gene mutation carriers and women with high familial risk. Journal of Clinical Oncology, 19, pp. 924-930. Cormack, D. (2000) The research process in nursing. 4th Edition. Blackwell Publishing. Enriquez, L., and Listinsky, J. (2009). Role of MRI in breast cancer enhancement. Cleveland Clinic Journal of Medicine, 76 (9), pp. 525-532. Greenhalgh, T. (2006) How to read a paper: the basics of evidence based medicine. 3rd Edition. Blackwell Publishing. Kriege, M., Brekelman, C., Boetes, C., Besnard, P., Zonderland, H., Obdeijn, I., et al. (2004) Efficacy of MRI and mammography for breast-cancer screening in women with a familial or genetic predisposition. The New England Journal of Medicine, 351(5), pp. 427-437. Kuhl CK, et al. (2007) MRI for diagnosis of pure ductal carcinoma in situ: a prospective observational study. Lancet, 370, pp. 485–492. Kuhl, C., Schrading, S., Leutner, C., Morkkabati-Spitz, N., Wardelmann, E., Fimmers, R., et al. (2005). Mammography, breast ultrasound, and magnetic resonance imaging for surveillance of women at high familial risk for breast cancer. Journal of Clinical Oncology, 23, pp. 8469-8476. Liberman, L. (2004) Breast cancer screening with MRI – what are the data for patients at high risk? The New England Journal of Medicine, 351, pp. 497-500. Lynch, H.T. and Lynch, J.F. (2006) Genetics, natural history, and DNA-based genetic counseling in hereditary breast cancer. In: Breast cancer 2nd Ed., BC Decker Inc. MARIBS Study Group (2005) Screening with magnetic resonance imaging and mammography of a UK population at high familial risk of breast cancer: a prospective multicentre cohort study. Lancet, 365, pp. 1769-1778 McCormack, V., Mangtani, P., Bhakta, D., McMichael, A., and Santos Silva, I. (2004) Hetergeneity of breast cancer risk wihtin the South Asian female population in England: a population-based case control study of first generation migrants. British Journal of Cancer, 90(1), pp.160-166. National Health Service Centre for Reviews and Dissemination. (2001) Undertaking systematic reviews of research on effectiveness. [Internet], Available from: University of York. [Accessed 14 May 2009] National Institute for Health and Clinical Excellence NICE CG14 (2004) The classification and care of women at risk of familial breast cancer in primary, secondary and tertiary care [Internet]. Available from: [Accessed 10 April 2009]. National Institute for Health and Clinical Excellence NICE CG41 (2006) Familial breast cancer [Internet]. Available from: [Accessed 10 April 2009]. National Institute for Health and Clinical Excellence NICE CG80 (2009) Early and locally advanced breast cancer [Internet]. Available from: [Accessed 10 April 2009]. NHSBSP National Health Service Breast Screening Programme (2008) [Internet]. Available from: [Accessed on 10 April 2009]. Parahoo, K. (2006) Nursing research principles, process and issues. 2nd Ed, Palgrave Macmillan. Public Health Resource Unit (2006) Critical appraisal skills programme [Internet]. Available from: [Accessed 23 March 2009]. Punch, K.F. (2000) Developing effective research proposals. London, Sage Publications. Robson, M.E. and Offit, K. (2004) Breast MRI for women with hereditary cancer risk. The Journal of the American Medical Association, 292, pp. 1368-1370. Sardanelli, F., Podo, F., D'Agnolo, G., Verdecchia, A., Santaquilani, M., Musumeci, R., et al. (2007) Multicenter comparative multimodality surveillance of women at genetic-familial high risk for breast cancer (HIBCRIT Study): interim results. Radiology, 242(3), pp.698-715. Schnall, M. (2006) Magnetic resonance imaging: In: Winchester, D.J. et al., eds. Breast cancer 2nd Ed. BC Decker Inc. Warner, E. et al (2001) Comparison of breast magnetic resonance imaging, mammography, and ultrasound for surveillance of women at high risk for hereditary breast cancer. Journal of Clinical Oncology, 19(15), pp. 3524-3531. Warner, E., Messersmith, H., Causer, P., Eisen, A., Shumak, R., and Plewes, D. (2008). Systematic review: using magnetic resonance imaging to screen women at high risk breast cancer. Annals of Internal Medicine, 148, pp. 671-679. Warner, E., Plewes, D., Hill, K., Causer, P., Zubovits, J., Jong, R., et al. (2004). Surveillance of BRCA1 and BRCA2 mutation carriers with magnetic resonance imaging, ultrasound, mammography and clinical breast examination. JAMA, 292, pp.1317-1325. Appendix A “Was there a clear question for the study to address? “ “Was a comparison made with an appropriate reference standard?” “Did all patients get the diagnostic test and the reference standard?” “Could the results of the test of interest have been influenced by the results of the reference standard?” “Is the disease status of the tested population clearly described?” “Were the methods for performing the test described in sufficient detail?” “What are the results?” “How sure are we about these results?” “Can the results be applied to your patients/population?” “Can the test be applied to you patient or population of interest?” “Were all outcomes important to the individual or population considered?” “What would be the impact of using this test on your patients/population?” The 12 questions are adapted from: Jaesche R, Guyatt GH, Sackett DL. Users guides to the medical literature, VI. How to use an article about diagnostic test. JAMA 1994; 271 (5) 389-391. Source: Critical Appraisal Skills Programme of Public Health Resource Unit, National Health Service, England, 2006. Read More