Understanding interpretive errors in radiologists learning computed tomography colonography.

Jeff L Fidler; Joel G Fletcher; C Daniel Johnson; James E Huprich; John M Barlow; Franklin Earnest; Brian J Bartholmai

doi:10.1016/j.acra.2004.03.052

Home/Archive/PubMed

PubMed · 15217592

Understanding interpretive errors in radiologists learning computed tomography colonography.

Jeff L Fidler·Joel G Fletcher·C Daniel Johnson·James E Huprich·John M Barlow·Franklin Earnest·Brian J Bartholmai

View original source ↗Export metadata ↓

Abstract

RATIONALE AND OBJECTIVES: To determine if interpretive errors in the course of learning CT colonography are secondary to failures in detection or in characterization and determine the types of lesions frequently missed. MATERIALS AND METHODS: Fifteen radiologists completed an electronic CTC training module consisting of two parts: 1) a teaching file demonstrating the varied appearances of polyps, cancers, and pitfalls in interpreting exams; and 2) a test of 50 complete CTC datasets. Following review of each test case, radiologists were asked to indicate if and where a polyp was visualized. The module then showed each neoplasm (if any) located within the dataset. For false negative examinations, radiologists indicated if the lesion was not seen, was seen but interpreted as colonic wall or fold, or was seen but interpreted as stool or fluid. RESULTS: The average sensitivity for sessile, pedunculated, and flat polyps for these novice readers was 76%, 63%, and 32%, respectively. Average sensitivity for all morphologies of cancers (annular, polypoid, flat) was high (93%, 85%, 95%), with 8/11 missed cancers being secondary to failure in detection. The most frequently missed cancer was an annular constricting tumor (5/11). Overall, 55% (73/132) of errors were failures of detection and 45% (59/132) were errors in characterization. CONCLUSION: Radiologists learning CT colonography had slightly more errors of detection than characterization, but this difference was not statistically significant. Flat and pedunculated polyps and annular constricting cancers were the most frequently missed morphologies. Examples of these abnormalities should be emphasized in CTC training programs.

Explore related subjects

Colonic Neoplasms Colonic Polyps Colonography, Computed Tomographic Diagnostic Errors Education, Medical, Continuing Humans Radiology Sensitivity and Specificity Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Jeff L Fidler, Joel G Fletcher, C Daniel Johnson, James E Huprich, John M Barlow, Franklin Earnest, Brian J Bartholmai. 2004. Understanding interpretive errors in radiologists learning computed tomography colonography.. https://doi.org/10.1016/j.acra.2004.03.052

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2004

Epigenetic inactivation of RUNX3 in microsatellite unstable sporadic colon cancers.

Runt domain transcription factors are important targets of TGF-beta superfamily proteins and play a crucial role in mammalian development. Three mammalian runt-related genes, RUNX1, RUNX2 and RUNX3, have been described. RUNX3 has been shown to be a putative tumor suppressor gene localized to chromosome 1p36, a region showing frequent loss of heterozygosity events in colon, gastric, breast and ovarian cancers. Because of the important role of TGF-beta signaling in the human colon, we hypothesized that RUNX3 may serve as a key tumor suppressor in human colon cancers and colon cancer-derived cell lines. We examined RUNX3 expression and the frequency of RUNX3 promoter hypermethylation in 17 colon cancer cell lines and 91 sporadic colorectal cancers. Semiquantitative analysis of RUNX3 transcripts was performed by RT-PCR and de novo methylation of the RUNX3 promoter was studied by a methylation-specific PCR (MSP) assay. Nineteen of 91 informative tumors (21%) and 11 of 17 (65%) colon cancer cell lines exhibited hypermethylation of the RUNX3 promoter. Interestingly, RUNX3 promoter hypermethylation was more common in tumors exhibiting high frequency of microsatellite instability (MSI-H) (33% of MSI-H vs. 12% of MSI-L/MSS tumors; p = 0.012). Hypermethylation of the RUNX3 promoter correlated with loss of mRNA transcripts in all cell lines. RUNX3 promoter methylation was reversed and its expression restored in SW48 and HCT15 colon cancer cells after treatment with the demethylating agent 5-aza-2'-deoxycytidine, indicating that loss of expression is caused by epigenetic inactivation in colon carcinogenesis. This is the first demonstration of frequent de novo hypermethylation of the RUNX3 promoter in sporadic colon cancers. The significant association of RUNX3 promoter hypermethylation with MSI-H colon cancers suggests that RUNX3 is a novel target of methylation, along with the hMLH1 gene, in the evolution of MSI-H colorectal cancers.

Colonic Neoplasms↗

PubMed2004

Colon cancer survival rates with the new American Joint Committee on Cancer sixth edition staging.

BACKGROUND: The recently revised American Joint Committee on Cancer (AJCC) sixth edition cancer staging system increased the stratification within colon cancer stages II and III defined by the AJCC fifth edition system. Using nationally representative Surveillance, Epidemiology, and End Results (SEER) data, we compared survival rates associated with colon cancer stages defined according to both AJCC systems. METHODS: Using SEER data (from January 1, 1991, through December 31, 2000), we identified 119,363 patients with colon adenocarcinoma and included all patients in two analyses by stages defined by AJCC fifth and sixth edition systems. Tumors were stratified by SEER's "extent of disease" and "number of positive [lymph] nodes" coding schemes. Kaplan-Meier analyses were used to compare overall and stage-specific 5-year survival. All statistical tests were two-sided. RESULTS: Overall 5-year survival was 65.2%. According to stages defined by the AJCC fifth edition system, 5-year stage-specific survivals were 93.2% for stage I, 82.5% for stage II, 59.5% for stage III, and 8.1% for stage IV. According to stages defined by the AJCC sixth edition system, 5-year stage-specific survivals were 93.2% for stage I, 84.7% for stage IIa, 72.2% for stage IIb, 83.4% for stage IIIa, 64.1% for stage IIIb, 44.3% for stage IIIc, and 8.1% for stage IV. Under the sixth edition system, 5-year survival was statistically significantly better for patients with stage IIIa colon cancer (83.4%) than for patients with stage IIb disease (72.2%) (P<.001). CONCLUSIONS: The AJCC sixth edition system for colon cancer stratifies survival more distinctly than the fifth edition system by providing more substages. The association of stage IIIa colon cancer with statistically significantly better survival than stage IIb in the new system may reflect current clinical practice, in which stage III patients receive chemotherapy but stage II patients generally do not.

Colonic Neoplasms↗

PubMed2004

A case of leptospirosis simulating colon cancer with liver metastases.

We report a case of a 61-year-old man who presented with fatigue, abdominal pain and hepatomegaly. Computed tomography (CT) of the abdomen showed hepatomegaly and multiple hepatic lesions highly suggestive of metastatic diseases. Due to the endoscopic finding of colon ulcer, colon cancer with liver metastases was suspected. Biochemically a slight increase of transaminases, alkaline phosphatase and gammaglutamyl transpeptidase were present; alpha-fetoprotein, carcinoembryogenic antigen and carbohydrate 19-9 antigen serum levels were normal. Laboratory and instrumental investigations, including colon and liver biopsies revealed no signs of malignancy. In the light of spontaneous improvement of symptoms and CT findings, his personal history was reevaluated revealing direct contact with pigs and their tissues. Diagnosis of leptospirosis was considered and confirmed by detection of an elevated titer of antibodies to leptospira. After two mo, biochemical data, CT and colonoscopy were totally normal.

Colonic Neoplasms↗