Peripheral immune circadian variation, synchronisation and possible dysrhythmia in established type 1 diabetes.

AIMS/HYPOTHESIS: The circadian clock influences both diabetes and immunity. Our goal in this study was to characterise more thoroughly the circadian patterns of immune cell populations and cytokines that are particularly relevant to the immune pathology of type 1 diabetes and thus fill in a current gap in our understanding of this disease. METHODS: Ten individuals with established type 1 diabetes (mean disease duration 11 years, age 18-40 years, six female) participated in a circadian sampling protocol, each providing six blood samples over a 24 h period. RESULTS: Daily ranges of population frequencies were sometimes large and possibly clinically significant. Several immune populations, such as dendritic cells, CD4 and CD8 T cells and their effector memory subpopulations, CD4 regulatory T cells, B cells and cytokine IL-6, exhibited statistically significant circadian rhythmicity. In a comparison with historical healthy control individuals, but using shipped samples, we observed that participants with type 1 diabetes had statistically significant phase shifts occurring in the time of peak occurrence of B cells (+4.8 h), CD4 and CD8 T cells (~ +5 h) and their naive and effector memory subsets (~ +3.3 to +4.5 h), and regulatory T cells (+4.1 h). An independent streptozotocin murine experiment confirmed the phase shifting of CD8 T cells and suggests that circadian dysrhythmia in type 1 diabetes might be an effect and not a cause of the disease. CONCLUSIONS/INTERPRETATION: Future efforts investigating this newly described aspect of type 1 diabetes in human participants are warranted. Peripheral immune populations should be measured near the same time of day in order to reduce circadian-related variation.

Use of Dried Capillary Blood Sampling for Islet Autoantibody Screening in Relatives: A Feasibility Study.

BACKGROUND: Islet autoantibody testing provides the basis for assessment of risk of progression to type 1 diabetes. We set out to determine the feasibility and acceptability of dried capillary blood spot-based screening to identify islet autoantibody-positive relatives potentially eligible for inclusion in prevention trials. MATERIALS AND METHODS: Dried blood spot (DBS) and venous samples were collected from 229 relatives participating in the TrialNet Pathway to Prevention Study. Both samples were tested for glutamic acid decarboxylase, islet antigen 2, and zinc transporter 8 autoantibodies, and venous samples were additionally tested for insulin autoantibodies and islet cell antibodies. We defined multiple autoantibody positive as two or more autoantibodies in venous serum and DBS screen positive if one or more autoantibodies were detected. Participant questionnaires compared the sample collection methods. RESULTS: Of 44 relatives who were multiple autoantibody positive in venous samples, 42 (95.5%) were DBS screen positive, and DBS accurately detected 145 of 147 autoantibody-negative relatives (98.6%). Capillary blood sampling was perceived as more painful than venous blood draw, but 60% of participants would prefer initial screening using home fingerstick with clinic visits only required if autoantibodies were found. CONCLUSIONS: Capillary blood sampling could facilitate screening for type 1 diabetes prevention studies.

Islet autoantibody seroconversion in the DPT-1 study: justification for repeat screening throughout childhood.

OBJECTIVE: Although type 1 diabetes autoimmunity frequently begins in childhood, little is known about the relationship between age and autoimmunity development. Our aim was to determine the timing of seroconversion to diabetes-associated autoantibody (DAA) positivity and risk in first- and second-degree relatives of patients with type 1 diabetes. RESEARCH DESIGN AND METHODS: Study subjects were identified through the Diabetes Prevention Trial-Type 1 (DPT-1). Children 3-18 years of age (n = 42,447) were screened for DAAs; 1,454 were ICA positive (≥ 10 JDF units), 1,758 were GAD65 positive, and 899 were ICA512 positive at the time of initial screening. Subjects who were initially antibody negative (n = 39,212) were recalled for rescreening, and 11,813 returned for rescreening. RESULTS: DAA seroconversion occurred in 469 (4%) children; 258 seroconverted to ICA, 234 to GAD65, and 99 to ICA512. The median time to seroconversion was 2 years. The 2-year risk for DAAs was highest in early childhood. For each 1-year increase in age in this cohort, the risk of any autoantibody seroconversion (HR 0.95, 95% CI 0.92-0.97) decreased by 5%, and for any two autoantibodies risk decreased by 13% (0.87, 0.82-0.93). CONCLUSIONS: Risk of autoantibody seroconversion among children followed in DPT-1 is age dependent. Younger children have the highest risk for DAAs, with the majority of children seroconverting by 13 years of age (75%). This suggests that annual screenings should be started in early childhood and continued through early adolescence to identify the majority of subjects at risk for type 1 diabetes and eligible for prevention trials.

Combined antioxidant carotenoids and the risk of persistent human papillomavirus infection.

Persistent infection with human papillomavirus (HPV) is the primary etiologic factor for cervical cancer. The synergistic effect of carotenoids on HPV persistence has not been examined. To explore these potential synergies, we developed 2 measures of carotenoid status using circulating and dietary intake nutrients in which each nutrient was given equal weighting. We then compared persistent HPV infection with its counterpart, intermittent infection. In the analysis using the Crude Index, no association was observed between circulating nutrients and persistent infection with oncogenic HPV [odds ratio (OR)(adjusted) = 0.8, 95% confidence interval (CI) = 0.3-2.2)] or any type HPV (OR(adjusted) = 0.8, 95% CI = 0.3-2.1). Similar results were obtained using the Cumulative Index. However, associations between dietary intake and persistent infection were observed using both indexes. When the analysis was restricted to oncogenic HPV, a 50% higher risk was observed for women with low dietary carotenoid status using the Crude Index (OR(adjusted) = 1.5, 95% CI = 0.6-3.7). In the analysis using any type HPV, the adjusted OR for women with low dietary intake of combined carotenoids using the Cumulative Index was 2.4 (95% CI = 1.1-5.2). These results may be consistent with the hypothesis that low levels of carotenoids may increase the risk of persistent HPV infection.

Pioglitazone versus glimepiride on coronary artery calcium progression in patients with type 2 diabetes mellitus: a secondary end point of the CHICAGO study.

OBJECTIVE: To compare coronary artery calcium (CAC) progression between 2 treatment groups, pioglitazone versus glimepiride. METHODS AND RESULTS: The CHICAGO (Carotid Intima-Media Thickness in Atherosclerosis Using Pioglitazone) study demonstrated that pioglitazone significantly decreased carotid intima-media thickness progression compared with glimepiride in patients with type 2 diabetes mellitus. The CAC level was determined at baseline and at the end of 72 weeks of treatment in the pioglitazone (n=146) and glimepiride (n=153) treatment groups using electron beam computed tomography. There was no difference in CAC progression between the treatment groups. By using backward and forward selection models, age, race/ethnicity, and baseline apolipoprotein B level predicted CAC progression. There was no relationship between carotid intima-media thickness and CAC progression during the study. CONCLUSIONS: There was no difference in CAC progression in patients with type 2 diabetes mellitus treated with pioglitazone or glimepiride. Age, race/ethnicity, and baseline apolipoprotein B level predicted CAC progression in patients with type 2 diabetes mellitus.

Gene expression patterns in the human breast after pregnancy.

Epidemiologic studies have established that pregnancy has a bidirectional, time-dependent effect on breast cancer risk; a period of elevated risk is followed by a long-term period of protection. The purpose of the present study was to determine whether pregnancy and involution are associated with gene expression changes in the normal breast, and whether such changes are transient or persistent. We examined the expression of a customized gene set in normal breast tissue from nulliparous, recently pregnant (0-2 years since pregnancy), and distantly pregnant (5-10 years since pregnancy) age-matched premenopausal women. This gene set included breast cancer biomarkers and genes related to immune/inflammation, extracellular matrix remodeling, angiogenesis, and hormone signaling. Laser capture microdissection and RNA extraction were done from formalin-fixed paraffin-embedded reduction mammoplasty and benign biopsy specimens and analyzed using real-time PCR arrays containing 59 pathway-specific and 5 housekeeping genes. We report 14 of 64 (22%) of the selected gene set to be differentially regulated (at P < 0.05 level) in nulliparous versus parous breast tissues. Based on gene set analysis, inflammation-associated genes were significantly upregulated as a group in both parous groups compared with nulliparous women (P = 0.03). Moreover, parous subjects had significantly reduced expression of estrogen receptor alpha (ERalpha, ESR1), progesterone receptor (PGR), and ERBB2 (Her2/neu) and 2-fold higher estrogen receptor-beta (ESR2) expression compared with nulliparous subjects. These initial data, among the first on gene expression in samples of normal human breast, provide intriguing clues about the mechanisms behind the time-dependent effects of pregnancy on breast cancer risk.

Making the diagnosis of acute appendicitis: do more preoperative CT scans mean fewer negative appendectomies? A 10-year study.

PURPOSE: To determine the frequency of preoperative computed tomography (CT) in the evaluation of patients suspected of having appendicitis at one institution during the past 10 years and to determine whether changes in CT utilization were associated with changes in the negative appendectomy rate. MATERIALS AND METHODS: Institutional review board approval was obtained, and a waiver of informed consent was granted for this HIPAA-compliant study. A surgical database search yielded medical record numbers of 925 patients (526 [ 56.9%] men and 399 [43.1%] women; mean age, 38 years (range, 18-95 years]) who underwent urgent appendectomy between January 1998 and September 2007. Patients who were younger than 18 years of age at the time of surgery were excluded. CT, pathology, and surgery reports were reviewed. By using logistic regression, changes in the proportion of patients undergoing CT and in the proportion of patients undergoing each year appendectomy in which the appendix was healthy were evaluated. Subgroup analyses based on patient age ( 45 years) and sex also were performed. RESULTS: Prior to urgent appendectomy, 18.5% of patients underwent preoperative CT in 1998 compared with 93.2% of patients in 2007. The negative appendectomy rate for women 45 years of age and younger decreased from 42.9% in 1998% to 7.1% in 2007. However, the timing of the decline in negative appendectomy rates for women 45 years and younger could not be proved to be associated with the increase in CT use. There was no significant trend toward a lower negative appendectomy rate for men regardless of age or for women older than 45 years of age with increased use of preoperative CT. The shift from single-detector CT to multidetector CT and the use of decreasing section thickness also correlated with a reduction in false-positive diagnoses. CONCLUSION: Rising utilization of preoperative CT and advances in technology coincided with a decrease in the negative appendectomy rate for women 45 years and younger but not in men of any age or women older than 45 years.

Sequential testing approach as an efficient and easier alternative for the validation of new predictive technologies in the clinic.

PURPOSE: When clinicians contemplate the use of a new predictive technology in their practice, such as a nomogram, there is always a question of whether the new test is beneficial to their own clinical population. Unfortunately, traditional validation methods require a large number of subjects for validation testing and delay the decision-making process. We present an efficient and easy-to-use method based on the concept of sequential data analysis. PATIENTS AND METHODS: We illustrate with an example determining the validity of a technology for predicting Gleason score upgrading from biopsy to postprostatectomy (the Chun nomogram) in a clinical population different from the one used to initially validate the technology. Clinical data required by the Chun nomogram were available from 201 patients from the Cooperative Prostate Cancer Tissue Resource. RESULTS: Of 124 patients predicted by the Chun nomogram to have an upgrading event, 47 actually did. The positive predictive value (PPV) of the model was therefore 38% and significantly (P < .05) less than the value of 80% which we considered to be the smallest clinically useful PPV in this situation. Had the sequential methods introduced in this article been employed prospectively in this cohort, the same conclusion would have been reached using data from only the first 15 patients. CONCLUSION: In-clinic validation of predictive technologies will help the clinician adopt truly beneficial technologies and avoid the adoption of technologies which provide no significant benefit to their local patient population. For this task, sequential methods offer clear advantages.

Expansion of human umbilical cord blood SCID-repopulating cells using chromatin-modifying agents.

OBJECTIVE: We investigated whether the addition of two chromatin-modifying agents, 5-aza-2'-deoxycytidine (5azaD) and trichostatin A (TSA), to cord blood (CB) CD34(+) cells in culture results in expansion of the numbers of severe combined immunodeficient (SCID) repopulating cells (SRC). MATERIALS AND METHODS: Human CB CD34(+) cells were cultured with cytokines in the presence or absence of 5azaD/TSA. After 9 days of culture, the fold expansion of CD34(+) and CD34(+)CD90(+) cell numbers, colony-forming unit (CFU)-mix, cobblestone area-forming cell (CAFC), and SRC numbers were determined. RESULTS: A 12.5-fold expansion of CD34(+)CD90(+) cells was observed in the 5azaD/TSA-treated cultures in comparison to the input cell numbers. Expansion of CD34(+)CD90(+) cells was associated with a 9.8-fold increase in the numbers of CFU-mix and 11.5-fold increase in CAFC. 5azaD/TSA treatment of the CB CD34(+) cells resulted in a 9.6-fold expansion of the absolute number of SRC following 9 days of culture as determined by limiting dilution analysis. Expansion of cells maintaining CD34(+)CD90(+) phenotype was not due to the retention of a quiescent population of cells because all of the CD34(+)CD90(+) cells in the culture had undergone cellular division. 5azaD/TSA-treated CD34(+)CD90(+) cells, but not CD34(+)CD90(-) cells were responsible for in vivo hematopoietic repopulation potential of nonobese diabetic/SCID mice. CONCLUSION: Ex vivo expansion strategy using chromatin-modifying agents provides a potential avenue by which to expand the number of hematopoietic stem cells (HSC) with a single CB unit for use as an alternative source of HSC grafts for adult recipients.

Cell adhesion to fibronectin (CAM-DR) influences acquired mitoxantrone resistance in U937 cells.

Cell adhesion to fibronectin is known to confer a temporally related cell adhesion-mediated drug resistance (CAM-DR). However, it is unknown whether cell adhesion during drug selection influences the more permanent form of acquired drug resistance. To examine this question, we compared the acquisition of mitoxantrone resistance in U937 cells adhered to fibronectin versus cells selected in a traditional suspension culture. Our data show that acquired drug resistance levels of resistance to mitoxantrone are 2- to 3-fold greater for cells adhered to fibronectin compared with cells in suspension culture. We also compared mechanism(s) of resistance associated with drug selection in suspension versus fibronectin-adherent cultures. Drug resistance in both suspension and fibronectin-adhered cultures correlated with reduced drug-induced DNA damage and diminished topoisomerase II levels and activity; however, mechanisms regulating topoisomerase II levels differed depending on culture conditions. In suspension cultures, a reduction in topoisomerase IIbeta levels was detected at both RNA and protein levels. Furthermore, the decreased expression of topoisomerase IIbeta mRNA levels correlated with decreased expression of NF-YA. In contrast, in spite of no changes in NF-YA or topoisomerase IIbeta RNA expression, topoisomerase IIbeta protein levels were decreased in fibronectin-adherent, drug-resistant cells. In addition, topoisomerase IIalpha protein levels (but not RNA levels) were reduced in drug resistance cells selected on fibronectin; however, no change in topoisomerase IIalpha was observed in cells selected with mitoxantrone in suspension culture. Taken together, our results suggest that the development of drug resistance models must consider interactions with the microenvironment to identify clinically relevant targets and mechanisms associated with acquired drug resistance.

Activation of stat3 in primary tumors from high-risk breast cancer patients is associated with elevated levels of activated SRC and survivin expression.

PURPOSE: Constitutive activation of signal transducer and activator of transcription 3 (Stat3) protein has been observed in a wide variety of tumors, including breast cancer, and contributes to oncogenesis at least in part by prevention of apoptosis. In a study of 45 patients with high-risk breast cancer enrolled in a phase II neoadjuvant chemotherapy trial with docetaxel and doxorubicin, we evaluated the levels of Stat3 activation and potentially associated molecular biomarkers in invasive breast carcinoma compared with matched nonneoplastic tissues. EXPERIMENTAL DESIGN: Using immunohistochemistry and image analysis, we quantified the levels of phospho-Stat3 (pY-Stat3), phospho-Src (pY-Src), epidermal growth factor receptor, HER2/neu, Ki-67, estrogen receptor, Bcl-2, Bcl-xL, Survivin, and apoptosis in formalin-fixed, paraffin-embedded sections from invasive carcinomas and their paired nonneoplastic parenchyma. The levels of molecular biomarkers in nonneoplastic and tumor tissues were analyzed as continuous variables for statistically significant correlations. RESULTS: Levels of activated pY-Stat3 and pY-Src measured by immunohistochemistry were significantly higher in invasive carcinoma than in nonneoplastic tissue (P < 0.001). In tumors, elevated levels of pY-Stat3 correlated with those of pY-Src and Survivin. Levels of pY-Stat3 were higher in partial pathologic responders than in complete pathologic responders. In partial pathologic responders, pY-Stat3 levels correlated with Survivin expression. CONCLUSIONS: Our findings suggest important roles for elevated activities of Stat3 and Src, as well as Survivin expression, in malignant progression of breast cancer. Furthermore, elevated Stat3 activity correlates inversely with complete pathologic response. These findings suggest that specific Stat3 or Src inhibitors could offer clinical benefits to patients with breast cancer.

Persistent activation of stat3 signaling induces survivin gene expression and confers resistance to apoptosis in human breast cancer cells.

PURPOSE: Signal transducer and activator of transcription 3 (Stat3) protein is persistently activated in breast cancer and promotes tumor cell survival. To gain a better understanding of the role of constitutive Stat3 signaling in breast cancer progression, we evaluated the expression profile of potential Stat3-regulated genes that may confer resistance to apoptosis. EXPERIMENTAL DESIGN: Stat3 signaling was blocked with antisense oligonucleotides in human MDA-MB-435s breast cancer cells and Affymetrix GeneChip microarray analysis was done. The candidate Stat3 target gene Survivin was further evaluated in molecular assays using cultured breast cancer cells and immunohistochemistry of breast tumor specimens. RESULTS: Survivin, a member of the inhibitor of apoptosis protein family, was identified as a potential Stat3-regulated gene by microarray analysis. This was confirmed in Survivin gene promoter studies and chromatin immunoprecipitation assays showing that Stat3 directly binds to and regulates the Survivin promoter. Furthermore, direct inhibition of Stat3 signaling blocked the expression of Survivin protein and induced apoptosis in breast cancer cells. Direct inhibition of Survivin expression also induced apoptosis. Increased Survivin protein expression correlates significantly (P = 0.001) with elevated Stat3 activity in primary breast tumor specimens from high-risk patients who were resistant to chemotherapy treatment. CONCLUSIONS: We identify Survivin as a direct downstream target gene of Stat3 in human breast cancer cells that is critical for their survival in culture. Our findings suggest that activated Stat3 signaling contributes to breast cancer progression and resistance to chemotherapy by, at least in part, inducing expression of the antiapoptotic protein, Survivin.

Correlation of radiologist rank as a measure of skill in screening and diagnostic interpretation of mammograms.

PURPOSE: To determine whether skill in the interpretation of screening mammograms is correlated with skill in the interpretation of diagnostic mammograms. MATERIALS AND METHODS: The institutional review board of the University of South Florida approved this study. This study was determined to be exempt from informed consent requirements because of the retrospective use of images and was conducted before HIPPA requirements were implemented. A total of 59 radiologists interpreted screening and diagnostic performance test sets of mammograms with a 1-year interval. Interpretations were recorded with modifications of the Breast Imaging and Reporting Data System. Radiologist skill was measured as the radiologist's ranking among his or her cohort in each of several measures of performance (ie, performance test receiver operating characteristic curve area, performance test screening sensitivity, performance test diagnostic sensitivity, and associated specificities). Correlations between radiologist rank in screening and rank in the diagnostic performance test measures were analyzed with the Spearman rank correlation statistical test. RESULTS: Radiologist rank in screening interpretations and in diagnostic interpretations was found to be significantly correlated in all measurements (P < .05). However, only two measurments (ie, receiver operating characteristic curve area rank correlation of 0.327 and sensitivity rank correlation of 0.402) remained significant after adjusting for multiple testing. The correlation between ranked screening specificity and ranked diagnostic specificity (0.296) was significant at only the .05 level. CONCLUSION: The interpretive performance of radiologists among their peers is moderately correlated between screening and diagnostic interpretations. Thus, proficiency in one area does not guarantee proficiency in the other area for some radiologists.

Perceptually based FROC analysis.

RATIONALE AND OBJECTIVES: Analysis of reading data when cases have multiple targets and/or the reader is required to localize targets is difficult. One approach to this free-response operating characteristic (FROC) problem is for images to be segmented (eg, with quadrants) by the investigator and a segment-level analysis be conducted with the case as a nesting factor. In this report, we introduce an alternative method that uses the visual scan path of the reader to segment the image. We evaluate the new method by applying it to data from a mammography reading experiment. MATERIALS AND METHODS: The gaze scan path of one radiologist was recorded as she scanned 40 mammograms for masses and microcalcifications. The observer is an experienced mammographer and was not one of the authors. In addition, the reader provided a rating indicating the degree of suspicion for any suspected targets she identified and localized. We then established "perceptual regions" by using a clustering algorithm on the visual fixations. We combined ratings given to specific locations indicated by the reader with the segmentation from the visual scan to generate a series of ratings classified for whether the perceptually based region associated with the rating contained or did not contain a known target. We analyzed data generated by our method from all 40 cases by using the conventional maximum-likelihood method based on the binormal model. Finally, we tested goodness-of-fit of the binormal model to the data by using chi-square. RESULTS: Maximum-likelihood estimation led to a model that did not fit the data (P < .001). However, examination of the observed and expected counts suggests that the binormal assumption does not hold for segments that contain targets and a bimodal distribution model might be preferred. CONCLUSION: Our new method provides an alternative approach to analysis of the FROC experiment. It needs to be developed further. Specifically, we propose that a mixture model extension of the binormal model be developed for ratings data arising from perceptually based FROC experiments. A disadvantage to our method is the requirement to record the scan path of the reader. However, we believe that adding such information to receiver operating characteristic (ROC) curve analysis will pay off when appropriate statistical models have been identified because we believe our data support our hypothesis that the perceptual scanning of images by humans deconvolves interpretation correlation. If true, this hypothesis implies that conventional statistical methods for ROC analysis based on independent data can be applied to the analysis of FROC data after conditioning on the scan path of the observer.

Reader variability in mammography and its implications for expected utility over the population of readers and cases.

The multiple-reader, multiple-case (MRMC) approach to receiver operating characteristic (ROC) analysis is becoming the dominant assessment paradigm in medical imaging. Its most common version involves having many readers read every patient case in the study, a critical feature since differences among competing imaging modalities are often dominated by differences in reader performance. The present authors have carried out MRMC ROC analysis on a uniquely large data set for mammography. The analysis quantifies the great range of observed reader skill in that data set. It also demonstrates that the sample sizes are sufficiently large that the conclusions generalize to the populations sampled here with little uncertainty from the finite sample size. A schematic approach to bracketing the utility matrix is then used to study trends in the resulting expected utility functions that correspond to the range of observed ROC curves. This is done for both the screening and the diagnostic context. The results raise 2 hypotheses for further investigation. First, it is possible that the present ambiguity surrounding the effectiveness of mammography is due in part to the observed range of reader skills and corresponding expected utility functions. Second, it is possible that computer-assisted modalities for mammography may lead to improvements in the expected utility function not only for screening but also in the diagnostic context, especially for the lower performing readers.

Genotypic and phenotypic comparisons of de novo and acquired melphalan resistance in an isogenic multiple myeloma cell line model.

Cancer cell adhesion confers a transient, de novo drug-resistant phenotype referred to as cell adhesion-mediated drug resistance (CAM-DR). In this report, we extend the CAM-DR phenotype to primary specimens from patients with myeloma, providing further evidence that CAM-DR is a viable clinical form of drug resistance. To examine mechanisms of cellular resistance to melphalan, we compared genotypic and phenotypic profiles of acquired and de novo melphalan resistance in an isogenic human myeloma cell line. Acquired melphalan resistance (8226/LR5) was associated with decreased drug-induced DNA damage and a complex gene expression profile showing that genes involved in the Fanconi anemia DNA repair pathway are increased in the LR5 cells compared with drug-sensitive or adherent cells. In contrast, cells adhered to fibronectin accumulate similar amounts of DNA damage compared with drug-sensitive cells but are protected from melphalan-induced mitochondrial perturbations and caspase activation. Levels of the proapoptotic protein Bim were significantly reduced in adherent cells. Gene expression changes associated with de novo resistance were significantly less complex compared with acquired resistance, but a significant overlap in gene expression was noted involving cholesterol synthesis. We propose that myeloma cell adhesion promotes a form of de novo drug resistance by protecting cells from melphalan-induced cytotoxic damage and that this transient protection allows cells to acquire a more permanent and complex drug resistance phenotype associated with a reduction in drug induced DNA damage.

Evaluation of proscriptive health care policy implementation in screening mammography.

PURPOSE: To evaluate the potential effect of proscriptive health care policies directed toward improving screening mammogram interpretation in the United States. MATERIALS AND METHODS: Percentiles of accuracy based on a random sample of 110 U.S. radiologists were used to examine the number of radiologists who would need to be restricted from providing mammographic interpretation to increase median accuracy from 66% to 67%, 71%, and 76%. In addition, reading volume data recorded for the sampled readers were used to project the percentage reduction in service volume (mammograms per year) that would result from restriction. Characteristics of participating radiologists were compared with those of nonparticipating radiologists by using chi2 testing and analysis of variance to assess the external validity of the results. RESULTS: To increase median accuracy by 1% (from 66% to 67%) would require prohibiting about 2,200 U.S. radiologists (ie, the 11% in the lowest quantile for accuracy) from performing mammographic interpretation and would result in a reduction of yearly service volume of approximately 10%. An increase in median accuracy of 5% (to 71%) would require prohibiting about 6,000 U.S. radiologists (ie, 30%) from performing this service, with an accompanying volume reduction of 25%. An increase in median accuracy of 10% (to 77%) would require prohibiting about 11,400 practicing U.S. radiologists (ie, 57%) from performing this service and would diminish the national service capacity by 50%. CONCLUSION: These data show that implementation of proscriptive health care policies based on accuracy would diminish the service capacity of screening mammography in the United States.

Association of volume and volume-independent factors with accuracy in screening mammogram interpretation.

BACKGROUND: Early detection of breast cancer is associated with the accurate reading of screening mammograms, but factors that influence reading accuracy are not well understood. We thus investigated whether reading volume and other factors were independently associated with accuracy in reading screening mammograms in a population of U.S. radiologists. METHODS: A random selection of 110 of 292 radiologists who agreed to participate, if selected, interpreted screening mammograms from 148 randomly selected women. Original index mammograms (i.e., mediolateral oblique and craniocaudal views of each breast) were used; comparison original mammograms were provided when available. Radiologist-level and facility-level factors were surveyed. Two standard metrics of screening accuracy, both based on receiver operating characteristic curves, were analyzed. The influence of volume on accuracy after controlling for other factors was assessed with multiple regression analysis. RESULTS: Current reading volume was not statistically significantly associated with interpretive accuracy. More recently trained radiologists interpreted mammograms more accurately than those trained earlier (-0.76% [95% confidence interval (CI) = -1.75% to -0.02%] reduction in sensitivity per year since residency). Facility-level factors that were statistically significantly and independently associated with better accuracy were the number of diagnostic breast imaging examinations and image-guided breast interventional procedures performed (0.55% [95% CI = 0.11% to 2.40%] increase in accuracy per examination or procedure offered), being classified as a comprehensive breast diagnostic and/or screening center or freestanding mammography center (1.39% [95% CI = 0.15% to 3.82%] higher than a hospital radiology department or multispecialty medical clinic), and being a facility that practiced double reading (1.61% [95% CI = 1.99% to 11.65%]) higher than in a facility without such practice). CONCLUSIONS: Individual radiologists' current reading volume was not statistically significantly associated with accuracy in reading screening mammograms, but several other factors were. Expertise reflects a complex multifactorial process that needs further clarification.

Factors affecting radiologist inconsistency in screening mammography.

RATIONALE AND OBJECTIVES: Although research has successfully documented variability in radiologists' interpretation of mammograms, it has failed to determine the relative contributions of case-specific features and reader inconsistency. Training interventions to improve consistency will be ineffectual if they do not target the principal determinants of disagreement among radiologists. The current study assessed the relative contributions of the case and the interpreter to the problem of inconsistent interpretation. MATERIALS AND METHODS: One hundred ten radiologists independently interpreted mammograms from the same 148 screening cases (43% with biopsy-proved cancers) and reported the presence or absence of calcifications, mass, architectural distortion, and asymmetric density in each of 296 breasts. The radiologists were blinded to disease status (established at biopsy or follow-up). RESULTS: Case-related differences accounted for a greater proportion of interpretation disagreement than did differences between interpreters. The presence of cancer was associated with increased disagreement, perhaps because of the multiplicity of findings. Patient age was also associated with increased disagreement in the reporting of calcifications. CONCLUSION: For screening mammography, increased consistency between radiologists in their recognition and reporting of clinically important findings will best be achieved by reducing disagreement in difficult cases. Current training in the United States addresses difficult cases only as they have been defined intuitively or experientially. The authors' population-based method provides an objective metric to measure case difficulty and basis from which to identify difficult cases for targeted training.

Does training in the Breast Imaging Reporting and Data System (BI-RADS) improve biopsy recommendations or feature analysis agreement with experienced breast imagers at mammography?

PURPOSE: To determine whether training in the Breast Imaging Reporting and Data System (BI-RADS) improves observer performance and agreement with the consensus of experienced breast imagers with regard to mammographic feature analysis and final assessment. MATERIALS AND METHODS: A test set of mammograms was developed, with 54 proven lesions consisting of 28 masses (nine [32%] malignancies) and 26 microcalcifications (10 [38%] malignancies). Three experienced breast imagers reviewed cases independently and by means of consensus. Twenty-three practicing mammogram-interpreting physicians reviewed mammograms before and after a day's lectures on BI-RADS. Observer performance before and after training was measured by means of agreement (kappa) with consensus description and assessments, rate of biopsy of malignant and benign lesions, and areas under receiver operating characteristic (ROC) curves. Performance was also measured for 11 participants 2-3 months after training. RESULTS: Improved agreement with consensus feature analysis was found for mass margins and/or asymmetries, with a pretraining generalized kappa value of 0.36 and a posttraining generalized kappa value of 0.41. Similar improvement was seen for description of calcification morphology (pretraining kappa value of 0.36 improving to 0.44 after training). No improvement was seen in describing calcification distribution. Final assessments were more consistent after training, with a pretraining kappa value of 0.31, as compared with 0.45 after training. The mean biopsy rate for malignant lesions improved from 73% (range, 53%-89%) before training to 88% (range, 74%-100%) after training, with minimal increase in mean biopsy rate of benign lesions (43% [range, 26%-60%] before to 51% [range, 31%-63%] after training), and no net change in area under the ROC curve, as compared with histopathologic findings. For the subset of participants with delayed follow-up, no significant decline in posttraining results was seen. CONCLUSION: BI-RADS training resulted in improved agreement with the consensus of experienced breast imagers for feature analysis and final assessment. It is important that trainees showed improved rates of recommending biopsy for malignant lesions. This effect was maintained over 2-3 months.