Diagnostic accuracy of multidetector-row computed tomography for common bile duct calculi: is it necessary to add non-contrast-enhanced images to contrast-enhanced images?

OBJECTIVE: To evaluate the diagnostic accuracy of 16-slice multidetector-row computed tomography (MDCT) for the detection of common bile duct (CBD) stones and to compare the diagnostic values between contrast-enhanced axial with multiplanar reformation (MPR) images and contrast-enhanced combined with non-contrast-enhanced axial images. MATERIALS AND METHODS: Fifty-four patients with suspected CBD stones underwent MDCT and endoscopic retrograde cholangiopancreatography (ERCP). Two radiologists independently evaluated contrast-enhanced axial with MPR images for the presence of CBD stones. Two weeks later, they performed a second review of contrast-enhanced and non-contrast-enhanced axial images. The ERCP was used as the criterion standard to confirm the presence of CBD stones. The McNemar test and receiver operating characteristic curve analysis were used to assess the differences in accuracy for detecting CBD stones. RESULTS: The ERCP identified bile duct stones in 27 patients and no stones in the other 27. Contrast-enhanced axial with MPR images had a sensitivity of 88.9%, specificity of 92.6%, and diagnostic accuracy of 90.7%. Contrast-enhanced combined with non-contrast-enhanced axial images had a sensitivity, specificity, and diagnostic accuracy of 88.9%. However, the difference between the 2 groups was not statistically significant. The areas under the receiver operating characteristic curve of the 2 groups were 0.929 and 0.905, respectively. CONCLUSIONS: The MDCT provides high accuracy in diagnosis for CBD stones. Adding non-contrast-enhanced images to contrast-enhanced images did not improve diagnostic accuracy, and therefore, additional non-contrast-enhanced images are not necessary.     

The “dense artery sign” — major cerebral artery thromboembolism demonstrated by computed tomography

Summary A 4 years review of high resolution thin slice (3–5 mm) computed tomography performed within 24 h after cerebral infarction revealed increased density in a major cerebral artery segment in 28 patients. Arteries affected were the vertebral and basilar artery in 6 and 8 patients, the sphenoid course of the middle cerebral artery in 13 cases and the extracranial internal carotid artery in 1 patient. In 35.7% of cases the so called dense artery sign provided earliest evidence of the ensuing infarction documented by CT controls in most patients. Angiography carried out in 8 patients, density caculations in the course of the affected vessel and resolution of the increased density on subsequent CT examinations suggest thrombembolism as the most likely etiology. In the clinical setting of acute stroke increased artery density encompassing the entire vessel diameter may serve as an early indicator of major cerebral artery occlusion and prompt angiographic investigation of a lesion potentially amenable to lysis.     

What is the value of measuring coronary artery calcification?

In this issue of Radiology, Saeed et al (1) describe the intramyocardial administration of VM202, a newly constructed plasmid human hepatocyte growth factor, in a pig model of myocardial infarction. Histopathologic findings were used to characterize and quantify neovascularization, while magnetic resonance (MR) imaging findings were used to quantify left ventricular function, perfusion, and infarct size. Compared with control animals, VM202-treated animals demonstrated an increase in number of capillaries, improved perfusion and left ventricular ejection fraction, and reduced infarct size.     

Coronary artery disease: Which degree of coronary artery stenosis is indicative of ischemia?

Purpose

To prospectively determine the best cut-off value of stenosis degree for low-dose computed tomography coronary angiography (CTCA) to predict the hemodynamic significance of coronary artery stenoses compared to catheter angiography (CA) using a cardiac magnetic resonance based approach as standard of reference.

Materials and methods

Fifty-two patients (mean age, 64 ± 10 years) scheduled for CA underwent cardiac magnetic resonance (CMR) at 1.5-T and dual-source CTCA using prospective ECG-triggering the same day. Diagnostic performance of CTCA and CA to detect myocardial ischemia was evaluated with CMR as the standard of reference. The diagnostic performance and best cut-off values to predict the hemodynamic significance of coronary were determined from receiver operating characteristics analysis (ROC).

Results

CA revealed >50% stenoses in 131/832 segments (15.7%) in 78/156 (50.0%) coronary arteries in 32/52 (62%) patients. CTCA revealed >50% stenoses in 148/807 (18.3%) segments, corresponding to 83/156 (53.2%) coronary arteries in 34/52 (65.4%) patients. CMR revealed ischemia in 118/832 (14.2%) myocardial segments corresponding to the territories of 60/156 (38.5%) coronary arteries in 29/52 (56%) patients. ROC analysis showed equal diagnostic performance for low-dose CTCA and CA with areas under the curve (AUC) of 0.82 and 0.83 (P = 0.64). The optimal cut-off value was determined at stenosis of >60% for the prediction of hemodynamically significant coronary stenosis by CTCA. Using this cut-off value, sensitivity, specificity, NPV and PPV to predict hemodynamic significance by CTCA were 100%, 83%, 100%, and 88% on a per-patient basis and 88%, 73%, 83% and 81% on a per-artery analysis, respectively.

Conclusion

By considering coronary stenosis >60%, diagnostic performance for predicting the hemodynamic significance of coronary stenosis by CTCA is optimal and equals that of CA.     

High resolution computed tomography of the chest in cystic fibrosis (CF): is simplification of scoring systems feasible?

The purpose of this study was to simplify HRCT scoring systems (SS) for CF by selecting representative HRCT parameters. Forty-two consecutive patients with CF underwent baseline and follow-up chest HRCT. Three radiologists evaluated 84 HRCTs employing five SS. “Simplified” HRCT SS were formed by selection of parameters exhibiting statistically significant relations with FEV1. Pulmonary function tests (PFTs) and nutrition (IBW%) were recorded. Regression analysis, Pearson correlation and T-test were used for statistical analysis. Three HRCT parameters were selected for the formation of “simplified” HRCT SS (severity of bronchiectasis, bronchial wall thickening, atelectasis-consolidation) using regression analysis. There was excellent correlation between each “simplified” and corresponding complete score (0.892 < r < 0.0967, p < 0.0001) or the remaining four complete scores (0.786 < r < 0.961, p < 0.0001). Strong correlation was found among the five “simplified” scores (0.803 < r < 0.997, p < 0.0001). Comparing baseline complete and “simplified” scores with corresponding follow-up ones, significant worsening was observed (p < 0.0001). PFTs and IBW% did not change significantly. HRCT scores correlated moderately with FVC and FEV1, but there was no correlation with FEF25-75 and IBW%. “Simplified” HRCT SS are as reliable as the complete ones and detect progression of lung disease earlier than clinical parameters. They are easy to use and could be adopted in clinical practice.     

Tumor lesion detection: when is integrated positron emission tomography/computed tomography more accurate than side-by-side interpretation of positron emission tomography and computed tomography?

OBJECTIVES: To determine if there is added value to oncology studies performed with a dedicated in-line positron emission tomography (PET)/computed tomography (CT) scanner as compared with PET read side by side with diagnostic CT (DCT). METHODS: Forty-one consecutive oncology patients referred for PET/CT who had contemporary DCT scans for review were enrolled. Body regions assessed on a DCT scan were assessed on PET/CT and by side-by-side reading of PET and DCT (SBS PET/DCT). Lesions identified on DCT, the CT portion of PET/CT, SBS PET/DCT, and the reading of fused PET/CT images were scored as benign or malignant. The PET portion of the PET/CT study was read by 2 teams: the first read the SBS PET/DCT scan and the other read the complete fused PET/CT scan. For discordant lesions, the final diagnosis was determined by pathologic findings (n = 6) or imaging follow-up (n = 21). RESULTS: Twenty-seven (16.1%) of the 168 lesions were discordant when comparing analysis of fused PET/CT and SBS PET/DCT. Sixteen (9.5%) were fundamentally discordant, and 11(6.6%) were discordant in degree of confidence. For all discordant lesions only, the sensitivity, specificity, negative predictive value, positive predictive value, and accuracy for PET/CT were 100%, 33%, 100%, 94%, and 78%, respectively, and for SBS PET/DCT, they were 38%, 50%, 19%, 73%, and 30%, respectively (P < 0.001 for sensitivity, P = not specific for specificity). The 2 main causes for misclassification on SBS PET/DCT were incorrect localization (n = 12) and changes occurring in the time gap between DCT and PET/CT (n = 4). CONCLUSIONS: In-line PET/CT offers better lesion localization in comparison to the visual fusion of PET and CT, especially for small lymph nodes, lesions adjacent to mobile organs, or lesions adjacent to the chest or abdominal wall.     

Accessory renal artery stenosis and hypertension: are these correlated? Evaluation using multidetector-row computed tomographic angiography

Background: Renal artery stenosis may produce hypertension, and this condition is referred to as renovascular hypertension (RVH).

Purpose: To evaluate, by using multidetector-row spiral computed tomographic angiography (MDCTA), whether a relationship between accessory renal artery stenosis and hypertension may be hypothesized.

Material and Methods: 214 patients (142 males, 72 females; mean age 66 years) who had previously undergone an MDCTA to study the abdominal vasculature were retrospectively studied. Patients with renal artery stenosis (RAS) were excluded from this analysis. The patients were studied by means of a four-detector-row CT, and scans were obtained after intravenous bolus administration of 110-140 ml of a nonionic contrast material with a 3-6 ml/s flow rate. As a second step, by means of statistical analysis, hypertension data were compared with findings of accessory artery stenosis. Two radiologists first independently reviewed the MDCTA images and then, in case of disagreement, in consensus. Interobserver agreement was calculated for all measurements.

Results: The overall number of detected accessory renal arteries was 74 in 56 of the 214 patients. Accessory renal artery stenosis was detected in 21 of the 56 patients. There was a difference in the prevalence of hypertension between patients with (n = 21) and without (n = 35) accessory renal artery stenosis (P = 0.0187). Interobserver agreement was good (kappa value 0.733).

Conclusion: Any statistical association between the presence of accessory renal artery stenosis and hypertension could not be disclosed. However, accessory renal artery stenosis, detected by MDCTA, is an important pathological sign that the radiologist has to assess in the light of its possible association with hypertension.     

Quantitative contrast-enhanced computed tomography: is there a need for system calibration?

The purpose of the study was to perform phantom studies to assess the impact of computed tomography (CT) system variability on quantitative measurements of contrast enhancement. A phantom containing tubes of contrast material at dilutions of 120, 1:35, 1:50, 1:100 and 1:200 arranged in air or water was imaged using 11 CT systems at 9 institutions. All systems had undergone routine calibration against air and water in accordance with the manufacturers’ recommendations. For a given tube voltage, the relationship between the iodine concentration and CT attenuation value on a single system varied by 17 to 24% over 46–48 weeks. The coefficients of variance for iodine calibration factors across different CT systems were 8.9% in air and 5.1% in water. Calibration of individual CT systems for iodine response is required to allow comparison of quantitative measurements of contrast enhancement across different institutions. Using the iodine calibration factor to express contrast enhancement as iodine concentration would facilitate the universal application of diagnostic enhancement thresholds, especially if the necessary calculations were performed by software installed on the CT console.     

Calcium scoring and chest pain: is it dead on arrival?

Coronary artery calcium (CAC) scoring has been shown to be a measure of overall coronary artery disease (CAD) burden and is a well-validated screening test that significantly improves cardiovascular risk prediction in asymptomatic adults beyond that provided with standard risk factors. The absence of coronary artery calcification identifies persons at very low cardiovascular risk. Among symptomatic patients, calcium scans have been shown to have high sensitivity for the presence of obstructive CAD among stable, low-intermediate risk middle-aged adults. This has prompted many to advocate for the expanded use of calcium scanning as a diagnostic test in symptomatic patients to rapidly identify patients without CAD, serving as a filter for invasive coronary angiography or hospital admission or both. However, recent studies suggest that the negative predictive value of CAC scoring to exclude obstructive CAD may be significantly decreased among patients at higher pretest likelihood for obstructive CAD, consistent with Bayesian reasoning. In a point-counterpoint format, this article discusses several considerations and potential limitations to the widespread use of CAC to exclude obstructive CAD in symptomatic patients which include (1) the effect of pretest disease prevalence on test accuracy, (2) limited clinical efficiency due to low specificity for obstructive CAD and myocardial ischemia and high background prevalence of CAC in adults, (3) occurrence of CAC relatively late in the atherosclerotic process, (4) lack of association of CAC with vulnerable and culprit coronary artery lesions, and (5) interindividual and racial heterogeneity in the process of atherosclerosis calcification.     

Detection of simulated vertical root fractures; which is better multi-detector computed tomography or cone beam computed tomography?

Objectives

Multi-detector computed tomography (MDCT) and cone beam computed tomography (CBCT) were compared regarding their ability to detect vertical root fractures.

Granulomatous disease: is it a nuisance or an asset during PET/computed tomography evaluation of lung cancers?

OBJECTIVES: To evaluate combined PET-computed tomography (CT) criteria for differentiating between granulomatous disease (GD) and malignancy (CA) in oncologic PET-CT studies. METHODS: Sixty-two patients who were referred for fluoro-2-deoxyglucose (FDG) PET-CT evaluation of pulmonary lesion(s) without a history of concurrent infection were studied. PET-CT was performed 1.5 h after intravenous administration of 555 MBq 18F-FDG in the fasting state with oral contrast. Combined PET-CT criteria including (i) calcifications (Ca2+) within lymph nodes, (ii) Ca2+ in lung nodules, (iii) liver and/or spleen Ca2+, (iv) locations of lung lesion(s), (v) hilar FDG uptake, (vi) comparison of lung versus maximum mediastinal FDG uptake, (vii) lymph node uptake not in the most probable lymphatic drainage pathway from a particular lung lesion, and (viii) extra pulmonary abnormal FDG uptake were each assigned a numerical score (0-3) with progressively higher score and sum of scores toward the increasing likelihood of GD. These patients either had pathological confirmation by biopsy/resection or were followed radiographically for a period of 2 years (CA=13; GD=49). Discriminant analysis was performed on all the above criteria with this gold standard. Simple t-test and box plot analysis were also performed on the summation of the scores (from 0 in CA to 13 in GD). RESULTS: When all eight criteria were entered into discriminant analysis, the combined PET-CT criteria classified correctly 71% of patients with a sensitivity of 65% and specificity of 92% for GD. The most significant discriminating criterion was FDG uptake in the lung lesion(s) less than maximum mediastinal uptake (P=0.01). The sum scores in GD and CA were significantly different (4.9+/-2.4 vs. 3.2+/-1.5, respectively, P=0.014). Box plots showed a clear separation at a cut-off value of around 3.5. CONCLUSION: Results show that the set of combined PET-CT criteria are highly specific for GD, which is not necessarily a nuisance during oncologic evaluation. Knowledge of these criteria may attribute some of the abnormal PET findings to GD, which is a useful asset for quick recognition and clinical interpretation.     

Coronary artery calcium score: has anything changed?

Calcium deposition along the coronary artery walls is a surrogate biomarker for atherosclerosis, and its presence in the coronary arteries could reflect the severity of coronary artery disease (CAD) High coronary artery calcium score (CACS) correlates with advanced disease and a higher likelihood of coronary stenoses. Many studies have supported the role of CACS as a screening tool for CAD. Historically, CACS was introduced with electron beam computed tomography (EBCT), but in the last 30 years, many changes have occurred in CT, where the development of multidetector spiral technology has made reliable the noninvasive study of the heart and coronary arteries. Correlation studies with intravascular ultrasound (IVUS) and histology have demonstrated the capability of multidetector CT (MDCT) to provide information useful for characterising atherosclerotic plaque in a noninvasive manner. This has shifted the interest from heavily calcified deposits to plaque with a low-density core and small, superficial calcified nodules, features more frequently present in atherosclerotic plaque prone to rupture and responsible for acute coronary events (culprit lesions). The purpose of this review article is to summarise the recent evolution and revolution in the field of CT, strengthen the importance of a coronary CT study not limited to CACS evaluation and CAD grading but also used to obtain information about plaque composition, and to improve stratification of the patient at risk for acute coronary events.     

Benign or malignant?: differentiating breast lesions with computed tomography attenuation values on dynamic computed tomography mammography

OBJECTIVE: To evaluate enhancement effects of breast lesions with dynamic computed tomography (CT) and to determine characteristics to aid in differential diagnosis of benign and malignant lesions. METHODS: One hundred seventy-six women underwent preoperative dynamic CT, in which they were scanned with rapid injection of contrast media (3 mL/s) after 30 seconds and 2 minutes (early and delayed phases, respectively). The CT values and enhancement patterns of malignant lesions (n = 154) were compared with those of benign lesions (n = 22), and the cut point of CT values with the best validity was analyzed. RESULTS: The CT values of malignant lesions were higher than those of benign lesions in both phases (P < 0.0001). The cut point was determined to be 60 Hounsfield units (HU) in the early phase (44% validity, 90% sensitivity). Washout and plateau patterns were more commonly seen in the malignant group (73% specificity). CONCLUSIONS: The analysis of enhancement effects on CT may lead to more appropriate differentiation of benign and malignant lesions.     

Trainee reporting of computed tomography examinations: do they make mistakes and does it matter?

AIM: To determine the accuracy of trainees reporting computed tomography (CT) examinations. MATERIAL AND METHODS: Over a 6-month period a single consultant reviewed all the CT examinations reported by registrars in one radiology department. After recording a provisional registrar report each examination was jointly reviewed by the consultant and registrar. The consultant's opinion was regarded as the gold standard. Data collected included: the error rate, whether an error was significant, leading to a change in patient management, and whether the mistake was a false-negative or positive. RESULTS: Three hundred and thirty-one patients were included in the study. There was an overall error rate of 21.5%. A significant error leading to a change in management was made in 10% of reports, and a significant error that did not lead to a change in management was made in 9.3%; 2.1% of reports had insignificant errors; and 69% of errors were false-negatives. CONCLUSION: Registrars make a significant number of errors affecting patient management when reporting CT and ideally all examinations should be reviewed by a consultant.