Breast density scales: the metric matters

OBJECTIVE

Measures of percent mammographic density (PMD) are often categorized using various density scales. The purpose of this study was to examine information loss associated with the use of categorical density scales.

METHODS

Baseline PMD was assessed at 1% precision for 2,374 females. The data were used to create 21-category, 4-category and 2-category density scales. R-squared and root mean square error were used to evaluate the effect of categorizing PMD. The area under the receiver operator characteristic curves were compared between cancer risk models employing solely categorical PMD scales and solely baseline PMD for a subset of females (424 cases, 848 controls).

RESULTS

R-squared value decreased from 1.00 (1% PMD) to 0.56 (2-category scale), while root mean square error increased from 0.00 (1% PMD) to 10.83 (2-category scale). The area under the receiver operator characteristic curve decreased from 0.64 for a cancer risk model using 1% PMD to 0.58 for a risk model using a 21-category density scale (p < 0.0001), 0.55 for a 4-category Breast Imaging, Reporting and Data System-like scale (p < 0.0001) and 0.50 for a 2-category Breast Imaging, Reporting and Data System-like scale (high vs low) (p < 0.0001).

CONCLUSION

Categorizing PMD measures into categorical density scales leads to a significant loss of information. Indeed, a simple high versus low split of PMD using a 50% cut point yields a cancer risk model with no discriminatory power. Advances in knowledge: Use of categorical mammographic density scales rather than continuous percent mammographic density measures leads to significant loss of information. Breast cancer risk models using categorical mammographic density scales perform more poorly than models using continuous PMD measures.

Consistency of visual assessments of mammographic breast density from vendor-specific "for presentation" images

Discussions of percent breast density (PD) and breast cancer risk implicitly assume that visual assessments of PD are comparable between vendors despite differences in technology and display algorithms. This study examines the extent to which visual assessments of PD differ between mammograms acquired from two vendors. Pairs of "for presentation" digital mammography images were obtained from two mammography units for 146 women who had a screening mammogram on one vendor unit followed by a diagnostic mammogram on a different vendor unit. Four radiologists independently visually assessed PD from single left mediolateral oblique view images from the two vendors. Analysis of variance, intra-class correlation coefficients (ICC), scatter plots, and Bland-Altman plots were used to evaluate PD assessments between vendors. The mean radiologist PD for each image was used as a consensus PD measure. Overall agreement of the PD assessments was excellent between the two vendors with an ICC of 0.95 (95% confidence interval: 0.93 to 0.97). Bland-Altman plots demonstrated narrow upper and lower limits of agreement between the vendors with only a small bias (2.3 percentage points). The results of this study support the assumption that visual assessment of PD is consistent across mammography vendors despite vendor-specific appearances of "for presentation" images.

Utility of relative and absolute measures of mammographic density vs clinical risk factors in evaluating breast cancer risk at time of screening mammography

OBJECTIVE

Various clinical risk factors, including high breast density, have been shown to be associated with breast cancer. The utility of using relative and absolute area-based breast density-related measures was evaluated as an alternative to clinical risk factors in cancer risk assessment at the time of screening mammography.

METHODS

Contralateral mediolateral oblique digital mammography images from 392 females with unilateral breast cancer and 817 age-matched controls were analysed. Information on clinical risk factors was obtained from the provincial breast-imaging information system. Breast density-related measures were assessed using a fully automated breast density measurement software. Multivariable logistic regression was conducted, and area under the receiver-operating characteristic (AUROC) curve was used to evaluate the performance of three cancer risk models: the first using only clinical risk factors, the second using only density-related measures and the third using both clinical risk factors and density-related measures.

RESULTS

The risk factor-based model generated an AUROC of 0.535, while the model including only breast density-related measures generated a significantly higher AUROC of 0.622 (p < 0.001). The third combined model generated an AUROC of 0.632 and performed significantly better than the risk factor model (p < 0.001) but not the density-related measures model (p = 0.097).

CONCLUSION

Density-related measures from screening mammograms at the time of screen may be superior predictors of cancer compared with clinical risk factors.

ADVANCES IN KNOWLEDGE

Breast cancer risk models based on density-related measures alone can outperform risk models based on clinical factors. Such models may support the development of personalized breast-screening protocols.

Ten years of breast screening in the Nova Scotia Breast Screening Program, 1991-2001. experience: use of an adaptable stereotactic device in the diagnosis of screening-detected abnormalities

OBJECTIVE

To evaluate and present 10-year outcomes of the Nova Scotia Breast Screening Program (NSBSP), a population-based screening program in the province of Nova Scotia, Canada, total population 900 000.

SETTING

Organized Breast Screening Program in Nova Scotia, Canada.

METHODS

Rates of participation, abnormal referrals, cancer detection rates, and benign:malignant (B:M) rates for core biopsy and surgical biopsy were calculated for asymptomatic women receiving a mammogram through the NSBSP 1991-2001.

RESULTS

Of 192 454 mammograms performed on 71 317 women, 33% were aged 40 to 49 years, 39% aged 50 to 59 years, 23% aged 60 to 69 years, and 5% aged 70 years and over. Cancer detection rate increased in each age group respectively: 3.7, 5.8, 9.7, and 13.5 per 1000 population on first-time screens. The positive predictive value of an abnormal screen increased with increasing age groups. Benign breast surgery decreased with increased use of needle core breast biopsy (NCBB). Open surgery decreased from 25 to 6 surgeries per 1000 screens. Of 1519 open surgical procedures (1328 women), 878 cancers were removed, with 37% 10 mm or less, and 61% 15 mm or less. In 613 women in whom the node status was assessed, 79% were negative.

CONCLUSION

A quality screening program incorporating NCBB in the diagnostic work-up is effective in the early detection of breast cancer and results in less open surgery, particularly in younger women.

Nova Scotia Breast Screening Program experience: use of needle core biopsy in the diagnosis of screening-detected abnormalities

PURPOSE

To evaluate a mammography screening program that uses a multidisciplinary team approach and needle core biopsy.

MATERIALS AND METHODS

A total of 10,000 asymptomatic women (aged 40-74 years) underwent screening in the Nova Scotia Breast Screening Program between June 1991 and April 1994. Women aged 50-69 years were the target group. All mammograms were read by one of four radiologists. Patients with abnormal screening findings underwent work-up, which included needle core biopsy, at one of two diagnostic centers.

RESULTS

Of 10,000 women, 2,794 were aged 40-49 years; 4,097, 50-59 years; 2,941, 60-69 years; and 168, 70-74 years, with 3.9, 6.4, 12.2, and 23.8 cancers detected per 1,000 patients, respectively. The overall rate was 7.7 cancers detected per 1,000 patients. Abnormalities were detected at screening in 838 patients (8.4%), 181 of whom underwent open surgery, with malignancy diagnosed in 77 (42%). Sixty-one (79%) of 77 patients had stage 0 or I cancer. In 43 (72%) of 60 patients in whom node status was assessed, findings were negative.

CONCLUSION

A screening program with needle core biopsy and a multidisciplinary team approach to diagnosis is effective.

Noninvasive testing in women presenting with chest pain: evidence for diagnostic uncertainty

OBJECTIVE

To gain insight into the diagnostic utility of exercise stress testing in women.

DESIGN

Observational prospective cohort study.

SETTING

The Victoria General Hospital, Halifax, Nova Scotia.

PARTICIPANTS

Consecutive women with chest pain referred by cardiologists or internists for exercise stress testing between May 30, 1992 and November 30, 1992 and followed prospectively to February 28, 1993.

INTERVENTION

The proportion of patients subsequently referred for thallium scintigraphy and/or coronary angiography and their clinical profiles were determined.

MAIN RESULTS

Of 183 patients studied, stress testing was positive in 48 (26.2%), negative in 48 (26.2%) and nondiagnostic in 87 (47.5%). Women with negative results were more likely to have had normal baseline electrocardiograms (ECGs) (P = 0.002) and least likely to have undergone prior angiography (P = 0.0003). Subsequent thallium scintigraphy and/or coronary angiography was undertaken in 33.3%, 18.8% and 27.6% with positive, negative and nondiagnostic index stress tests, respectively. None of chest pain, cardiac risk factors, previous cardiac investigations or baseline ECG discriminated 33 patients with negative or nondiagnostic stress results who had additional tests from 102 who did not.

CONCLUSION

Exercise stress testing poorly screens women with chest pain for coronary artery disease (diagnostic in only 52.5%). Further study was undertaken in 27.6% with nondiagnostic tests and, surprisingly, in 18.8% with negative results. Why certain women with nondiagnostic, and so many with negative, stress tests were referred for further investigation was unclear. These results suggest diagnostic uncertainty when females presenting with chest pain are assessed.

Assessment of painful late effects of lumbar spinal fusion with SPECT

The authors reviewed planar, SPECT and other contemporaneous radiologic images of the spine and the medical records of 33 patients with back pain after lumbar fusion surgery in order to determine the value of SPECT in the assessment of painful late effects of spinal fusion surgery.

METHODS

Twenty-one patients had lateral fusion, nine patients had posterior fusion only and three patients had anterior and posterior fusions. There were 24 patients who had surgery more than 4 yr ago (late group, mean 11.8 yr) and 9 patients who had surgery less than 4 yr ago (early group, mean 17.8 mo).

RESULTS

The most common SPECT abnormality in patients in the late group were lesions in the vertebral bodies and apophyseal joints in the free motion segments adjacent to the fused segments (62.5% of patients). Such lesions occurred in 46% of patients after lateral fusion, in 87.5% of patients after posterior fusion and in 67% of patients after posterior and anterior fusions. No SPECT abnormalities were detected in the fused segments in patients in the late group with solid lateral fusion but were detected in three patients with solid posterior fusion. These results correlate with biomechanical studies that have shown posterior fusion to produce the largest amount and lateral fusion to produce the least amount of stress in the free segments adjacent to the fusion. Lateral fusion was found to have a more stabilizing effect than posterior fusion.

CONCLUSION

In addition to the already established value of SPECT in detecting painful pseudoarthrosis, our results indicate that SPECT is of value in the assessment of painful late effects of fusion.

Stereotactic fine-needle aspiration cytology of nonpalpable breast lesions: initial experience in a tertiary-care institution

To assess the accuracy of stereotactic fine-needle aspiration cytology in the diagnosis of nonpalpable breast lesions, this procedure was performed in 226 consecutive patients, all women, immediately before needle localization and excision of the lesions. The patients were treated at a tertiary-care hospital between December 1989 and October 1991. The cytologic results (lesion benign, atypical, suspicious or malignant or insufficient material for interpretation) were compared with the histologic findings and the degree of suspicion (low, slight or high) on the basis of mammographic examination. The aspiration procedures were successful in only 159 patients (70.4%), and for 155 of these, histologic findings were available for correlation. The number of unsatisfactory specimens in this group was large: 65 of 155 (42%). When unsatisfactory specimens were included in the calculations and a cytologic finding of a suspicious or malignant lesion was treated as positive, the sensitivity of fine-needle aspiration cytology for malignancy was 23% (8/35) and the specificity 53% (63/120). The positive and negative predictive values were 89% (8/9) and 78% (63/81) respectively. A cytologic finding that the lesion was benign was unreliable in lesions for which the mammographic suspicion of malignancy was slight or high (4 of 30 malignant lesions with slight mammographic suspicion and 4 of 7 with high mammographic suspicion were classified as benign on the basis of the cytologic examination). Of the 24 cases with atypical cytologic findings, 10 involved malignancy.(ABSTRACT TRUNCATED AT 250 WORDS)

Remnants of normal tissue in polycystic disease of the liver. A cause for difficulty in the interpretation of indium-111 white blood cell study

On In-111 WBC images, diffuse homogeneous uptake of the radiopharmaceutical should be present throughout the liver. The authors present a case of a febrile patient with polycystic liver disease in whom the normal diffuse uptake was not seen. Instead, the images demonstrated focal areas of uptake, which suggested infection. Tc-99m SC liver spleen scintigraphy demonstrated the In-111 WBC foci to correspond to areas of residual normal parenchyma. The patient underwent laparotomy and liver transplantation and no abscesses were found. Pathologic examination of the liver revealed multiple uninfected cysts and residual normal parenchyma in the caudate lobe, corresponding to the findings on scintigraphy.

180 degrees SPECT of the spine in patients with low back pain. Comparison with 360 degrees acquisition

Single-photon emission computed tomography has been proven to detect more abnormalities than planar bone scintigraphy in patients with low back pain (LBP). Both 180 degrees and 360 degrees acquisitions were performed in 24 patients with LBP to determine whether the shorter 180 degrees posterior SPECT is as accurate for abnormality detection as 360 degrees acquisition. The vertebral bodies and posterior elements of 193 vertebrae were scored on a five-point score (1 = normal, 5 = abnormal), independently by three experienced physicians, on three separate reconstructed image sets: standard, filtered, back projection 360 degrees acquisition, distance-weighted 360 degrees acquisition, and standard, filtered, back projection 180 degrees acquisition. With one exception, no statistically significant difference in score was found between 180 degrees and 360 degrees images. For one observer, the mean score of the posterior elements was higher on 180 degrees compared to 360 degrees standard images. These results indicate that 180 degrees acquisition SPECT may be used for abnormality detection in patients with LBP.

Role of SPECT in differentiating malignant from benign lesions in the lower thoracic and lumbar vertebrae

The authors categorized 125 spinal lesions in cancer patients and 127 lesions in patients with back pain according to their location in the vertebra on single photon emission computed tomographic (SPECT) images. Forty-four lesions were metastases, all in patients with known malignancy. Lesions in the apophyseal joints were all benign. Lesions manifesting as abnormal uptake projecting beyond the vertebral body surface were osteophytes. Thirty-seven percent of the lesions detected in cancer patients were categorized in either of these two benign categories. Lesions showing focal or diffuse uptake in the body were usually benign (96% and 87%, respectively). Lesions showing uptake in the body and pedicle were usually metastases (83%). When abnormal uptake was seen in both the body and posterior elements but with an intervening normal pedicle, benign disease was the most common cause (93%). It was concluded that the location of lesions on tomographic images provides useful information for differentiation between malignant and benign lesions in the vertebrae.

Spontaneous return of renal flow and function on technetium-99m scan in a patient with renal artery thrombosis following angioplasty

Failure to visualize a kidney on radionuclide scanning using technetium-99m (99mTc) chelates and other renal radiopharmaceuticals such as iodine-131 hippuran has been described as potentially misleading in patients who have acute renal failure due to acute tubular necrosis and urinary tract obstruction. Such failure has not been described in a single kidney nor following angioplasty. The authors report a patient in whom a kidney was not visualized on 99mTc diethylenetriamine penta-acetic acid dynamic and serial scanning 3 days after renal angioplasty. The kidney was believed to be infarcted. Three months later there was almost complete recovery of renal function.

Indium-111 chloride scintigraphy in adult osteomyelitis

Osteomyelitis is a common clinical problem that may be difficult to diagnose. We compared the performance of indium-111-labeled white cells ([111In]WBC) to 111In chloride ([111In]Cl) in two groups of adult patients suspected to have osteomyelitis or septic arthritis. Using [111In] WBC, 52 scans were performed on 51 patients. Nineteen patients had osteomyelitis. The sensitivity was 84% and specificity 82%. Using [111In]Cl, 48 scans were performed on 47 patients. Twelve had osteomyelitis. Sensitivity was 91%, and specificity was 89%. In each group, one false-negative study occurred in vertebral osteomyelitis. Three false-negative studies using [111In]WBC were due to failure to distinguish between combined bone and soft-tissue infection and soft-tissue infection alone. False-positive studies in both groups were due to overlying soft-tissue infection or inflammatory arthritis. Chi-squared test showed no significant difference in performance between the two agents. Indium-111 chloride is easier to prepare and use than [111In]WBC, which requires a time-consuming labeling process.