Contrast-enhanced spectral mammography versus MRI: Initial results in the detection of breast cancer and assessment of tumour size.

OBJECTIVES: To compare mammography (MG), contrast-enhanced spectral mammography (CESM), and magnetic resonance imaging (MRI) in the detection and size estimation of histologically proven breast cancers using postoperative histology as the gold standard. METHODS: After ethical approval, 80 women with newly diagnosed breast cancer underwent MG, CESM, and MRI examinations. CESM was reviewed by an independent experienced radiologist, and the maximum dimension of suspicious lesions was measured. For MG and MRI, routine clinical reports of breast specialists, with judgment based on the BI-RADS lexicon, were used. Results of each imaging technique were correlated to define the index cancer. Fifty-nine cases could be compared to postoperative histology for size estimation. RESULTS: Breast cancer was visible in 66/80 MG, 80/80 CESM, and 77/79 MRI examinations. Average lesion largest dimension was 27.31 mm (SD 22.18) in MG, 31.62 mm (SD 24.41) in CESM, and 27.72 mm (SD 21.51) in MRI versus 32.51 mm (SD 29.03) in postoperative histology. No significant difference was found between lesion size measurement on MRI and CESM compared with histopathology. CONCLUSION: Our initial results show a better sensitivity of CESM and MRI in breast cancer detection than MG and a good correlation with postoperative histology in size assessment. KEY POINTS: • Contrast-enhanced spectral mammography (CESM) is slowly being introduced into clinical practice. • Access to breast MRI is limited by availability and lack of reimbursement. • Initial results show a better sensitivity of CESM and MRI than conventional mammography. • CESM showed a good correlation with postoperative histology in size assessment. • Contrast-enhanced spectral mammography offers promise, seemingly providing information comparable to MRI.

[Sonographic detection of microcalcifications - potential of new method]. / Sonografische Mikrokalkdetektion - Potenzial einer neuen Methode.

PURPOSE: Does the easier microcalcification detection (EMD) method enable sonographic visualization of microcalcifications in breast core biopsy specimens compared with mammography? MATERIALS AND METHODS: In a prospective randomized study, 105 core biopsy specimens obtained with stereotactic guidance were examined by mammography and ultrasound. EMD is integrated in a high-end ultrasound system and uses three level settings (0 - 2 blue, 3 - 5 violet, and 6 - 8 black-and-white; 14 MHz). Detection of microcalcifications per core specimen was determined for ultrasound and mammography. EMD image quality was rated on a scale of 1 - 9. ANOVA and Sidak post-hoc testing, Pearson regression analysis (r), and Spearman rank correlation (rho) were performed. The intraclass correlation coefficient (ICC) was calculated, and an ROC analysis was conducted. RESULTS: The blue color map 1 was assigned the highest mean score of 1.5 ± 0.7 (p< 0.05 compared with black-and-white and violet). There was good correlation between the two modalities (r= 0.708 and rho= 0.694) with detection of 3.5 ± 3.1 microcalcifications per specimen by ultrasound versus 4.3 ± 4.8 by mammography (p> 0.05). The ICC of 0.773 indicates little disagreement between the two modalities. ROC analysis showed mammography to be superior to ultrasound compared with histological detection of microcalcifications (AUC= 0.837 vs. AUC= 0.728). CONCLUSION: Sonographic detection of microcalcifications in stereotactic biopsy specimens using the EMD method correlates well with digital mammography. Mammography is slightly superior.

[Teleradiology and mammography screening: evaluation of a network with dedicated workstations for reporting]. / Teleradiologie im Mammographie-Screening: Evaluation eines Testnetzes mit dedizierten Befundungsstationen.

PURPOSE: Evaluation of a system that supports a workflow for breast cancer screening by mammography. The time of installation, system reliability and workstation operation were evaluated. MATERIALS AND METHODS: The evaluated system (Image Diagnost, Munich, Germany) contains 2 diagnostic mammography workstations, a centralized server, and 2 Dicom shuttles for exchanging images via a physically existing network structure. Temporary archiving is possible. A mask designed for the needs of mammography screening facilitates assignment of BIRADS categories. The system automatically compares the categories assigned by a first and second reviewer and decides whether a consensus conference should be held. In the event that a conference is needed, the reviews are transmitted to the mammography expert responsible for the screening program and the consensus conference. Images are transferred via ISDN, Germany's National Research and Education Network (in the following DFN) and a central server between 2 sites which are approx. 100 km apart. We evaluated the duration of installation, the reliability of the system, and the usability of the workstation. Since we used curative mammography for evaluating the system, the patient age was noted for comparison. RESULTS: The system was installed in five days. Once installed, the system functioned without any major breakdowns. Mammography units of 2 manufacturers were able to be connected to the system without difficulty. Mammographies of 151 patients were exchanged between the sites and evaluated by 2 radiologists. 57 % of the patients were in the screening age (50-69 years). 9 exams were classified BIRADS 4a, 2 were 4b and 3 were BIRADS 5. The evaluations were technically perfect in 146/151 cases; hanging protocols had to be altered manually in 6 cases; the window/level had to be manually adjusted in 26/151 cases. Magnification was sufficient in all cases. The system exchanges examinations extremely quickly; up to 100 mammography exams may be evaluated in 1 hr. CONCLUSION: The system supports the workflow given by the German Mammography Screening program both locally and in a network.

[Mammography screening in Germany: how, when and why?]. / Mammographie-Screening in Deutschland: Wie, wann und warum?

It is well-known from several large randomized trials that regular mammography screening can reduce breast cancer mortality. While in many countries mammography screening programs have been in existence for quite some time, an organized population-based screening program is only now being implemented in Germany. In this review article, the different elements of a mammography screening program and their effect on the cost-benefit ratio are discussed and the planned German screening program is compared to the international programs.

Experimental study of X-ray mammography in a fluid bath: reduced radiation dose and improved detail resolution.

OBJECTIVE: To determine the effects on x-ray mammography of embedding the breast in a fluid bath. METHODS: A plexiglass phantom in the shape of a compressed breast was x-rayed in air and water with a mammography unit, and the radiation dose was measured by 20 thermoluminescence dosimeters placed in 20 representative positions on the surface of the phantom. Udders from slaughtered sheep as an animal model and phantoms were examined by mammography in identical positions surrounded by air and different fluids. The images were evaluated for detail resolution subjectively (animal model) and objectively (phantom containing quantifiable structures) by 6 and 15 blinded readers, respectively. The readers' results were analyzed and compared for visualization of the objects in air versus fluid. RESULTS: Mammographic examination in the fluid bath reduced the radiation dose and improved detail resolution in the rounded margin of the breast. Also, viewing conditions in the central region were improved as a result of the uniform optical density across the entire film area. CONCLUSIONS: Performing x-ray mammography in a fluid bath has advantages that make the technical realization of this method desirable.

Kinetics of phenylalanine transport at the human blood-brain barrier investigated in vivo.

In vivo proton magnetic resonance spectroscopy was used to investigate intracerebral phenylalanine (Phe) concentrations in nine patients with classical phenylketonuria (PKU). The study included serial examinations (n = 31; plasma Phe levels: 0.47-2.24 mmol/l) of patients either receiving a Phe-restricted diet (200 mg Phe per day; four patients) or a diet rich in Phe (1000 mg Phe per day; three patients). No spectrum showed metabolic abnormalities besides elevated Phe. Difference spectroscopy yielded intracerebral Phe concentrations between 0.20 and 0.76 mmol/l. Regional variations between parieto-occipital periventricular brain, frontal brain, and cerebellum were not statistically significant. Data could be fitted assuming saturable Phe transport into the brain, based on a symmetric Michaelis-Menten model (characterized by an apparent Michaelis transport constant, K(t,app), and a maximum transport velocity, Tmax) and constant Phe consumption in the brain cells (described by a velocity Vmax). Non-linear least-squares fitting of the combined data from all patients yielded K(t,app) = 0.16 +/- 0.11 mmol/l and (Tmax / Vmax) = 9.0 +/- 4.1. Carrier saturation and competitive inhibition of the influx of other large neutral amino acids can be expected at plasma Phe levels usually found in PKU patients.

Computerized analysis of breast lesions in three dimensions using dynamic magnetic-resonance imaging.

Contrast-enhanced magnetic resonance imaging (MRI) of the breast is known to reveal breast cancer with higher sensitivity than mammography alone. The specificity is, however, compromised by the observation that several benign masses take up contrast agent in addition to malignant lesions. The aim of this study is to increase the objectivity of breast cancer diagnosis in contrast-enhanced MRI by developing automated methods for computer-aided diagnosis. Our database consists of 27 MR studies from 27 patients. In each study, at least four MR series of both breasts are obtained using FLASH three-dimensional (3D) acquisition at 90 s time intervals after injection of Gadopentetate dimeglumine (Gd-DTPA) contrast agent. Each series consists of 64 coronal slices with a typical thickness of 2 mm, and a pixel size of 1.25 mm. The study contains 13 benign and 15 malignant lesions from which features are automatically extracted in 3D. These features include margin descriptors and radial gradient analysis as a function of time and space. Stepwise multiple regression is employed to obtain an effective subset of combined features. A final estimate of likelihood of malignancy is determined by linear discriminant analysis, and the performance of classification by round-robin testing and receiver operating characteristics (ROC) analysis. To assess the efficacy of 3D analysis, the study is repeated in two-dimensions (2D) using a representative slice through the middle of the lesion. In 2D and in 3D, radial gradient analysis and analysis of margin sharpness were found to be an effective combination to distinguish between benign and malignant masses (resulting area under the ROC curve: 0.96). Feature analysis in 3D was found to result in higher performance of lesion characterization than 2D feature analysis for the majority of single and combined features. In conclusion, automated feature extraction and classification has the potential to complement the interpretation of radiologists in an objective, consistent, and accurate way.

Analysis of spiculation in the computerized classification of mammographic masses.

Spiculation is a primary sign of malignancy for masses detected by mammography. In this study, we developed a technique that analyzes patterns and quantifies the degree of spiculation present. Our current approach involves (1) automatic lesion extraction using region growing and (2) feature extraction using radial edge-gradient analysis. Two spiculation measures are obtained from an analysis of radial edge gradients. These measures are evaluated in four different neighborhoods about the extracted mammographic mass. The performance of each of the two measures of spiculation was tested on a database of 95 mammographic masses using ROC analysis that evaluates their individual ability to determine the likelihood of malignancy of a mass. The dependence of the performance of these measures on the choice of neighborhood was analyzed. We have found that it is only necessary to accurately extract an approximate outline of a mass lesion for the purposes of this analysis since the choice of a neighborhood that accommodates the thin spicules at the margin allows for the assessment of margin spiculation with the radial edge-gradient analysis technique. The two measures performed at their highest level when the surrounding periphery of the extracted region is used for feature extraction, yielding Az values of 0.83 and 0.85, respectively, for the determination of malignancy. These are similar to that achieved when a radiologist's ratings of spiculation (Az = 0.85) are used alone. The maximum value of one of the two spiculation measures (FWHM) from the four neighborhoods yielded an Az of 0.88 in the classification of mammographic mass lesions.

Automated segmentation of digitized mammograms.

RATIONALE AND OBJECTIVES: Fast and reliable segmentation of digital mammograms into breast and nonbreast regions is an important prerequisite for further image analysis. We are developing a segmentation algorithm that is fully automated and can operate independent of type of digitizing system, image orientation, and image projection. METHODS: The algorithm identifies unexposed and direct-exposure image regions and generates a border surrounding the valid breast region, which can then be used as input for further image analysis. The program was tested on 740 digitized mammograms; the segmentation results were evaluated by two expert mammographers and two medical physicists. RESULTS: In 97% of the mammograms, the segmentation results were rated as acceptable for use in computer-aided diagnostic schemes. Segmentation problems encountered in the remaining 22 images (2.9%) were most often caused by digitization artifacts or poor mammographic technique. CONCLUSION: The developed algorithm can serve as a component of an "intelligent" workstation for computer-aided diagnosis in mammography.

[Full-field digital mammography]. / Digitale Vollfeldmammographie.

Due to the extremely high image quality requirements in mammography, there has for a long time been no adequate digital alternative to conventional film-screen mammography. The longest experience so far exists with digital mammography on the basis of storage phosphor (CR) systems. However, at normal dose this technique has a relatively poor signal-to-noise ratio and has not found general acceptance. Recently three novel systems for digital mammography by the companies Fischer (slot-scan detector), Trex (CCD-array), and GE (amorphous silicon detector) have been introduced and are currently under clinical investigation. The main advantage of digital mammography is the linear relationship between dose and detector signal with the possibility of a tailored optimization of image contrast. Other advantages include digital image storage, telemammography, and computer-assisted diagnosis.

[Typical and unusual findings in MR mammography]. / Typische und ungewöhnliche Befunde in der MR-Mammographie.

In recent years, MR mammography has gained increasing importance in breast diagnostics. The main advantage of this technique is its high sensitivity for invasive breast cancer. The two main indications for MR mammography are preoperative staging of breast cancer and the differentiation between postoperative changes and recurrent tumor. In addition, MR mammography is increasingly used for problem solving in cases of questionable clinical, mammographic or sonographic findings. At this it is important to know not only the different manifestations of breast cancer, but also important benign and malignant diagnostic alternatives. Furthermore, it is necessary to be familiar with therapy-related changes. The most important criterion to differentiate between benign and malignant lesions in MR mammography is the extent and temporal course of contrast enhancement. In addition, the lesion morphology and the signal intensity on T2-weighted images can be used to distinguish between different disease entities. The following review article will discuss typical and unusual findings in breast MRI, including rare entities as well as changes after breast conserving therapy and chemotherapy.

[Visualization of microcalcifications on mammographies obtained by digital full-field mammography in comparison to conventional film-screen mammography]. / Visualisierung von Mikrokalzifikationen in digitaler Vollfeldmammographie im Vergleich zu konventioneller Film-Folien-Mammographie.

PURPOSE: To evaluate the visualization of microcalcifications on mammographies obtained by full-field digital mammography (FFDM) in comparison to conventional film-screen mammography (FSM). MATERIAL AND METHODS: Forty-seven digital and film-screen mammographies depicting histologically proven lesions (27 benign, 20 malignant) were assessed by 4 readers. The images obtained with the different systems were comparable in terms of positioning. Maximum time interval between film-screen mammography and digital mammography was three months. Using a questionnaire, the readers evaluated the number of microcalcifications and their subjective conspicuity for FFDM (Senographe 2000D, GE Medical Systems, Milwaukee, USA) and FSM. A 7-point scale based on the BIRADS classification was used to characterize the calcifications by means of ROC analysis. RESULTS: No statistically significant differences were seen between the two types of mammography among the readers in assessing the number of microcalcifications. The subjective conspicuity of microcalcifications was found to be significantly better for digital mammographies. The diagnosis assigned by the readers did not show significant differences between the two systems. CONCLUSION: Although the subjective conspicuity of microcalcifications was found to be significantly better on digital mammography compared to film-screen mammography, there was no significant advantage of digital mammography resulting from the higher contrast resolution nor a disadvantage in terms of characterization of microcalcifications resulting from the lower spacial resolution. The advantages of digital mammography (e. g. CAD-systems, archiving, dose reduction) can be used without a loss of diagnostic quality.

[The visualization of metal implants with digital luminescence radiography]. / Darstellung von Metallimplantaten mit der digitalen Lumineszenzradiographie.

DLR makes it possible to integrate conventional radiography into digital communication and storage methods (PACS) into radiology. The use of DLR has been compared with conventional film methods in the demonstration of hip prostheses. The high contrast differences at the edge of the metal implant leads to artifacts, which could result in erroneous interpretation. Suitable image manipulation makes it possible, however, to eliminate these artifacts almost completely. DLR leads to an improvement in diagnosis in those complications not specifically related to the prostheses.

[Dose reduction through gridless technique in digital full-field mammography]. / Dosiseinsparung durch rasterlose Technik in der digitalen Vollfeldmammographie.

PURPOSE: To determine the role of the scatter grid in digital full-field mammography with respect to image quality and dose and to compare the experimental results with initial clinical experience. MATERIALS AND METHODS: A phantom consisting of 205 fields that enclose gold dots of different thickness and size (CD-Mam phantom, Medical Department, Nijmegen, Netherlands) was used for digital full-field mammography with the conventional grid module and a special gridless module. Four different breast thicknesses were simulated using Plexiglas as scatter material. First, the phantom was exposed at the parameter and dose settings automatically selected in each experimental setup (with and without grid). Subsequently, the phantom was exposed at the different simulated breast thicknesses using the gridless module in combination with the parameters automatically selected for the grid module. This was followed by a series of phantom mammograms obtained with the experimental setup reversed. The 16 mammograms were evaluated by 3 readers and the results compared considering breast thickness, radiation dose, and quality. The gridless module was used for preoperative labeling in 16 patients for comparison of mammograms obtained with and without a grid. RESULTS: For the same entrance dose used in routine mammography, digital mammography without grid is superior to digital mammography with grid when performed on simulated thin breasts (Plexiglas less than 3 cm), with no difference found when performed on simulated large breasts. The advantages of gridless mammography are more pronounced at a markedly reduced entrance dose (identical parenchymal dose without and with grid using the dose automatically selected for the gridless module). This tendency is confirmed by the initial clinical comparison. CONCLUSION: Gridless digital full-field mammography has the potential to reduce the dose further, in addition to already known mechanisms of dose reduction. However, the gridless system must be adjusted (markedly higher "switch-off dose" at the detector) to achieve an overall dose reduction since gridless mammography is inferior to grid mammography for identical doses at the detector plane.

[Intraoperative digital radiography for diagnosis of nonpalpable breast lesions]. / Digitale intraoperative Präparate-radiographie bei diagnostischen Exzisionen nicht palpabler Läsionen der Brust.

PURPOSE: A procedure for performing intraoperative digital radiography of diagnostic breast specimens directly in the operating suite with teleradiologic assessment by a radiologist is presented. The efficiency of this procedure is compared with that of conventional magnification mammography performed in the radiology department. MATERIAL AND METHODS: Thirty-six specimen radiographs obtained by conventional magnification mammography were compared with 38 intraoperative digital magnification radiographs (DIMA Soft P42 prototype, Feinfocus Inc., Garbsen). The radiographs were assessed for lesion conspicuity and time savings for the surgeon, anesthesiologist, and radiologist. RESULTS: The new procedure identified all 38 labeled pathological lesions, and the conventional technique likewise had a detection rate of 100% (36/36). The new technique resulted in considerable time savings for the surgeon and the radiologist. The duration of surgery was shorter and the time interval from removal of the specimen to reporting of the results was reduced from about 23 min to about 13 min. A single radiograph was sufficient for complete visualization of the specimen in all cases. CONCLUSION: Digital intraoperative specimen radiography considerably reduces the time of surgery depending on the local conditions and is highly accurate in locating a suspicious area within the tissue.

[Comparing the visualization of microcalcifications with direct magnification in digital full-field mammography vs. film-screen mammography]. / Vergleich der Visualisierung von Mikroverkalkungen durch Vergrösserungsaufnahmen in digitaler Vollfeldmammographie und konventioneller Mammographie.

PURPOSE: To evaluate the conspicuity of microcalcifications in magnified mammographic views of preparations obtained with full field digital mammography (FFDM), film-screen mammography (FSM), and the DIMA technique. MATERIAL AND METHODS: Twelve preparations were examined by FFDM and FSM using 1.8 x magnification and DIMA using 7 x magnification. Parameter settings were identical for all three techniques. The number of visible microcalcifications was then determined for each modality by three radiologists. As far as possible, all preparations were X-rayed at 22 kV and 10 mAS. RESULTS: Altogether 9705 calcifications were counted (DIMA: 1609/1542/1534; FFDM: 1020/753/881; FSM: 901/643/822). The total number of microcalcifications identified with the DIMA technique was 4685 as compared to 2654 with FFDM and 2366 with FSM. The calcifications counted with FFDM and FSM thus corresponded to 56.6 % and 50.5 %, respectively, of those identified with DIMA. The differences between the groups were statistically significant (F-Test, p < 0.05). CONCLUSION: Significantly more calcifications are identified when magnified mammographic views of preparations containing microcalcifications are obtained with the DIMA technique compared to FFDM or FSM. FFDM depicts markedly more calcifications than FSM. This means one should increase spatial resolution. Digital mammography offers the potential for improved visualization of microcalcifications with advanced applications.

[Preoperative MRT of the breast in invasive lobular carcinoma in comparison with invasive ductal carcinoma]. / Präoperative MRT der Brust beim invasiv-lobulären im Vergleich zum invasiv-duktalen Karzinom.

PURPOSE: To evaluate the role of preoperative MRI of the breast in invasive lobular carcinoma (ILC) compared to invasive ductal carcinoma (IDC). MATERIALS AND METHODS: For one year, all patients transferred by the hospital's gynecologic outpatient service for suspicious findings in routine mammography and/or ultrasound (conventional modalities = CM) underwent preoperative MRI of the breast. Retrospective analysis of the histologic findings identified 17 patients with ILC. These were compared with 30 proven IDC patients, chosen by random. The MRI findings of these 2 patient groups were compared with regard to the detection of additional lesions. The average number of additional lesions detected by MRI was compared for significant differences between both groups using the T-test for paired samples. RESULTS: In the 17 patients with ILC, conventional modalities (CM) identified 21malignant lesions whereas MRI detected a total of 30 lesions. At least one additional lesion was detected by MRI in 7 of the 17 patients with ILC. In the 30 patients with IDC, on the other hand, MRI detected an additional lesion in three instances only. In one patient of the ILC group, MRI identified an additional lesion in the contralateral breast that had escaped detection by CM. No additional contralateral lesion was detected by MRI in any of the IDC patients. Benefit of MRI in ILC-Group: The mean numbers of detected malignant lesions differed significantly between diagnosis by MRI and CM in the ILC group (1.77 carcinomas per patient with MRI versus 1.24 with conventional modalities, T-test, p = 0.0078). Benefit of MRI in IDC-Group: although it was possible to find 1.27 carcinomas vs. 1.17 carcinomas per patient in the IDC-Group, this benefit was not statistical significant (T-test, p = 0.0831). CONCLUSION: Preoperative MRI detects multiple additional lesions compared to the ones already known by CM. The higher incidence of multiple lesions in ILC compared to IDC and the difficult diagnosis of ILC in CM might be the reason for the fact that preoperative MRI is particularly useful in patients with ILC.

Use of iodine-based contrast media in digital full-field mammography--initial experience.

AIM: To investigate the use of iodine-based contrast media in digital full-field mammography. METHODS: After performing initial phantom studies, seven patients underwent digital mammography (Senographe 2000D, GE Medical Systems, Milwaukee, USA) using a specially filtered beam before as well as 60, 120, and 180 seconds after injection of 80 ml of iodine contrast medium (Ultravist 370, Schering AG, Germany). The precontrast mammograms were then subtracted from the postcontrast mammograms and the resulting images compared with a contrast-enhanced dynamic MRI study, performed on all women. RESULTS: Contrast medium accumulation within the tumors was visualized with a good quality in all cases. The conditions under which successful contrast-enhanced digital mammography can be performed were determined in phantom studies. CONCLUSIONS: Contrast-enhanced digital mammography has a potential for improving the visualization of breast tumors in mammography using special beam filtering, adjusted x-ray parameters, proper timing, and suitable subtraction software.

White matter abnormalities in phenylketonuria: results of magnetic resonance measurements.

In adolescents and adults with PKU, blood phenylalanine levels above 10 mg/dl are generally associated with white matter changes in MRI. The grade of these changes is correlated to most recent blood phenylalanine levels. Based on studies using T2 relaxometry the MRI changes seem to be the consequence of a reversible dysmyelination. The clinical relevance of these white matter changes remains unclear as the extent of MRI alterations did not correlate with IQ, neurological and electrophysiological deficits of the patients. The intracerebral phenylalanine concentration as measured by protonspectroscopy amounts to about 50% of blood phenylalanine concentrations. Preliminary data indicate that brain phenylalanine levels remain constant if blood concentrations exceed 20 mg/dl. This might be of clinical relevance for the treatment of adolescent and adult PKU patients.

Volumetric quantification of the effect of aging and hormone replacement therapy on breast composition from digital mammograms.

OBJECTIVE: To assess the physiological changes in breast composition with aging using volumetric breast composition measurement from digital mammograms and to assess the effect of hormone replacement therapy (HRT). METHODS: A total of 764 consecutive mammograms of 208 non-HRT using women and 508 mammograms of 134 HRT-using women were analyzed using a volumetric breast composition assessment software (Quantra™, Hologic Inc.). Fibroglandular tissue volume (FTV), breast volume (BV), and percent density (PD) were measured. For statistical analysis, women were divided into a premenopausal (<46 years), a perimenopausal (46-55 years), and a postmenopausal (>55 years) age group. More detailed graphical analysis was performed using smaller age brackets. Women using HRT were compared to age-matched controls not using HRT. RESULTS: Women in the postmenopausal age group had a significantly lower FTV and PD and a significantly higher BV than women in the premenopausal age group (FTV: 77 vs. 120 cm(3), respectively; PD: 16% vs. 28%, respectively; BV 478 vs. 406 cm(3), respectively; p<0.01 for all). Median FTV was nearly stable in consecutive mammograms in the premenopausal and postmenopausal age groups, but declined at a rate of 3.9% per year in the perimenopausal period. Median PD was constant in the premenopausal and postmenopausal age groups and declined at a rate of 0.57% per year in the perimenopausal age group. BV continuously increased with age. Women using HRT throughout the study had a 5% higher PD than women not using HRT (22% vs. 17%, respectively; p<0.001). CONCLUSIONS: Accurate knowledge of normal changes in breast composition are of particular interest nowadays due to the importance of breast density for breast cancer risk evaluation. FTV and PD change significantly during the perimenopausal period but remain relatively constant before and thereafter. Median total breast volume consistently increases with age and further contributes to changes in breast density. HRT use is associated with a significantly higher PD.