Search

Background Despite its high sensitivity in diagnosing COVID-19 in a screening population, chest CT appearances of COVID 19 pneumonia are thought to be non-specific. Purpose To assess the performance of United States (U.S.) and Chinese radiologists in differentiating COVID-19 from viral pneumonia on chest CT. Methods A total of 219 patients with both positive COVID-19 by RT-PCR and abnormal chest CT findings were retrospectively identified from 7 Chinese hospitals in Hunan Providence, China from January 6 to February 20, 2020. A total of 205 patients with positive Respiratory Pathogen Panel for viral pneumonia and CT findings consistent with or highly suspicious for pneumonia by original radiology interpretation within 7 days of each other were identified from Rhode Island Hospital in Providence, RI. Three Chinese radiologists blindly reviewed all chest CTs (n=424) to differentiate COVID-19 from viral pneumonia. A sample of 58 age-matched cases was randomly selected and evaluated by 4 U.S. radiologists in a similar fashion. Different CT features were recorded and compared between the two groups. Results For all chest CTs, three Chinese radiologists correctly differentiated COVID-19 from non-COVID-19 pneumonia 83% (350/424), 80% (338/424), and 60% (255/424) of the time, respectively. The seven radiologists had sensitivities of 80%, 67%, 97%, 93%, 83%, 73% and 70% and specificities of 100%, 93%, 7%, 100%, 93%, 93%, 100%. Compared to non-COVID-19 pneumonia, COVID-19 pneumonia was more likely to have a peripheral distribution (80% vs. 57%, p

The purpose of this study was to compare the diagnostic performance of four radiographic signs of gastric band slippage: abnormal phi angle, the "O sign," inferior displacement of the superolateral gastric band margin, and presence of an air-fluid level above the gastric band.A search of the electronic medical record identified 21 patients with a surgically proven slipped gastric band and 63 randomly-selected asymptomatic gastric band patients who had undergone barium swallow studies. These studies were evaluated for the four signs of band slippage by two independent radiologists who were blinded to clinical data. Sensitivity, specificity, and positive and negative predictive values were calculated for each radiographic sign of band slippage. Interobserver agreement between radiologists was assessed using the Fleiss kappa statistic.In evaluating for gastric band slippage, an abnormal phi angle greater than 58° was 91-95% sensitive and 52-62% specific (κ = 0.78), the O sign was 33-48% sensitive but 97% specific (κ = 0.84), inferior displacement of the superolateral band margin by more than 2.4 cm from the diaphragm was 95% sensitive and 97-98% specific (κ = 0.97), and the presence of an air-fluid level was 95% sensitive and 100% specific (κ = 1.00).We report two previously undescribed radiographic signs of gastric band slippage that are both sensitive and specific for this important surgical complication and recommend that these signs should be incorporated into the imaging evaluation of gastric band patients.

The purpose of this study was to establish practice patterns of radiology residents in regards to interpretation and reporting of outside studies for transferred patients. We performed a national survey of radiology residency chief residents, administered by email through the Association of Program Directors in Radiology (APDR). There were 81 chief resident respondents, representing 42.8 % of 187 total Accreditation Council for Graduate Medical Education (ACGME)-approved radiology residency training programs in the USA. In 97.5 % of programs, residents perform interpretations of outside studies. Up to 76.7 % of respondents state that when outside studies are reviewed by residents, an original report is available in less than one quarter of cases. While 55.1 % of respondents state that there is a mechanism for recording their findings and impressions for outside studies, only 32.1 % are aware of a policy requiring documentation. Of the respondents, 42.3 % report they have no means for documenting their findings and impressions on outside studies. Further, 65.4 % state that there is no policy requiring an attending to review and document agreement with their interpretation of outside studies. There is wide institutional variation in both policy and practice regarding reinterpretation of outside studies for patients transferred to academic hospitals. While the majority of radiology residents are providing the service of reinterpreting outside studies, only a minority of residency programs have a policy requiring (1) documentation of their impressions or (2) attending oversight and documentation of discrepant opinions.

The objective of this study was to report the clinical outcome of patients with an appendicolith on computed tomography (CT) who did not undergo appendectomy on initial presentation. Reports from 45,901 abdominal CT examinations performed between March 2000-March 2004 containing the words "appendicolith" or "fecalith" were identified. Patients with appendicoliths not initially undergoing appendectomy were followed to assess re-presentation with abdominal pain ultimately requiring appendectomy. Seventy-four patients had an appendicolith on CT report, were discharged without surgery, and had clinical follow-up. Fifty-two of 74 (70%) patients had no appendiceal symptoms, were given an alternate diagnosis, and did not return with appendicitis. Twenty-two of 74 (30%) patients were discharged without acute appendicitis but with possible appendiceal symptoms. Five of these 22 (23%) patients returned with pathologically proven acute appendicitis, and all had possible appendiceal symptoms at initial presentation. An appendicolith may be a marker of increased risk for appendicitis but is not an indication for appendectomy.

Foreign object ingestion is a common reason for visiting an emergency department; however, wire grill-cleaning brush bristles are an uncommon foreign object. This report describes a series of twelve cases identified in a single hospital system from July 2009 through June 2012. Patients included six males and six females; ages ranged from 11 to 75 (mean: 47 years). The patients all reported recent outdoor residential food grilling and use of commercially available wire grill-cleaning brushes. The severity of injury ranged from puncture of the soft tissues of the neck, causing severe pain on swallowing, to perforation of the gastrointestinal tract requiring emergent surgery. Before cooking, persons should examine the grill surface carefully for the presence of wire bristles that might have dislodged from the grill brush and could embed in cooked food. Alternative residential grill-cleaning methods or products might be considered.

To compare the effectiveness of two treatments for tunneled hemodialysis catheter malfunction: percutaneous fibrin sheath stripping (PFSS) and over-the-wire catheter exchange (EX).Adult patients with poorly functioning tunneled hemodialysis catheters (flow rates200 mL/min) were randomly assigned to receive either PFSS or EX. Over the course of 20 months, 30 patients (37 encounters) referred to a single institution met the inclusion criteria and consented to participate. PFSS employed transcatheter snares via femoral vein puncture, whereas EX was performed over a guide wire with use of fluoroscopic guidance. Patients were followed up to determine the duration of continued adequate hemodialysis via manipulated catheters for up to 4 months (primary outcome measure).Overall technical success rate was 97%. Mean catheter patency for the PFSS group was 24.5 +/- 29.3 days, and 52.2 +/- 43 days for the EX group (P.0001). After EX, patency rates at 1, 2, 3, and 4 months were 71%, 33%, 27%, and 27%, compared to 31%, 16%, 7%, and 0% after PFSS (P = .04, logrank test). Exchanged catheters were significantly more likely to be patent for as long as 4 months (23% versus 0%; P.05, chi2 test).Malfunctioning tunneled hemodialysis catheters treated by means of EX are significantly more likely to remain patent for up to 4 months than are those treated by means of PFSS. According to the results of this trial, PFSS should not be performed as a routine therapy for catheter malfunction.

Background . The purpose of this study was to document the long-term results of transcatheter embolotherapy of large pulmonary arteriovenous malformations (PAVMs). Methods . From a data base of 221 consecutive patients with PAVMs treated by embolotherapy between 1978 and 1995, 45 patients with 52 PAVMs, supplied by feeding arteries 8 mm in diameter or larger, were selected for a retrospective investigation. Results . Of 45 patients with 52 large PAVMs, 38 patients (84%) with 44 PAVMs (85%) were cured by the first embolotherapy (mean follow-up, 4.7 years). Acute periprocedural complications included self-limited pleurisy (31%), angina secondary to air embolus (2%), and paradoxical embolization of a device during deployment (4%). None of these events led to short- or long-term sequelae. Seven patients (16%) had persistence of the PAVM attributable to either recanalization (n = 4) or interim accessory artery growth (n = 3). Two of these patients presented with ischemic stroke several years after the initial treatment. Persistent PAVMs (n = 8) were retreated successfully by a second procedure (n = 7), or a third procedure (n = 1) (mean follow-up, 5.9 and 5.3 years, respectively). Conclusions . Embolotherapy of large PAVMs results in permanent occlusion in an overwhelming majority of patients. Continued patency due to recanalization or accessory artery growth is easily detected and treated.

Rationale and Objectives. We assessed how well published randomized controlled trials (RCTs) of peripheral thrombolysis have been designed and how they adjusted for patient characteristics that may influence responses to therapy. Methods. RCT studies published between 1984 and 1994 were reviewed to determine whether they included nine crucial elements: sample size calculations, description of unenrolled eligible participants, homogeneous samples, balanced experimental groups, equal potency therapies, equal follow-up of outcomes, appropriate statistical inference, multivariable or subgroup analyses, and anonymous review of subjective endpoints. Results. The nine RCTs complied with a median of three of nine standards (range = 2–6). Each report exhibited serious flaws that may affect genealizability and external validity. None adequately adjusted for sample heterogeneity or sought to identify characteristics that may affect responses to therapy. Conclusion. Failure to address prognostic or confounding variables restricts the utility of peripheral thrombolysis RCTs and helps explain why contentious debates about this therapy persist. A model for future thrombolysis trials is proposed.

Purpose: The most quoted literature on arteriographic complications is based on self-reports collected during the mid 1970s. We sought to determine whether those results remain valid despite changes in arteriographic practice and whether patient subgroups at increased risk could be identified. Methods: Five hundred forty-nine consecutive patients were examined after arteriography and twice over 72 hours. Patients were telephoned at least 2 weeks later to identify delayed complications. The sample was divided into two groups to allow independent validation of suspected prognostic factors. Results: The rate of major complications was 2.9% (16/549), but varied from 0.7% to 9.1% among three strata of relative risk. Rates were highest in patients studied for suspected aortic dissection, mesenteric ischemia, gastrointestinal bleeding, or symptomatic carotid artery stenosis and lowest in patients with trauma or aneurysmal disease. Patients studied for claudication or limb-threatening ischemia had intermediate risk (2.0%). Within these strata, congestive heart failure and furosemide use were the only variables independently associated with a significantly increased complication rate. Conclusions: Previous reports have overestimated the risk of arteriography for trauma or aneurysm but substantially underestimate the risk for patients with other common conditions. Such stratified complication rates are essential to understand relative costs and benefits of arteriography and other vascular imaging modalities in specific clinical situations. (J VASC SURG 1995;22:787-94.)

Recent reports suggest that both the nature of intravascular foreign bodies and the tools available to retrieve them have changed substantially in the past decade. We reviewed our recent experience with percutaneous retrieval of intravascular foreign bodies to determine the efficacy and safety of the procedure using currently available devices.Between 1990 and 1994, we attempted retrieval of 35 intravascular foreign bodies in 32 patients. Twelve patients (38%), including all five with intraarterial foreign bodies, were treated for complications of transcatheter interventional procedures that resulted in embolization of seven coils, four intravascular stents, an inferior vena cava filter, and a valvuloplasty balloon fragment. From procedure records, we reviewed the types of retrieval devices and methods used; the medical record was studied to determine the occurrence and treatment of any procedure-related complications.Retrieval was successful in 31 (97%) of 32 patients. All five intraarterial and 29 of 30 IV objects were removed. Nitinol goose-neck snares were used in 28 of 32 cases, but more than one retrieval system was required in eight cases (25%), often using grasping forceps, tip-deflecting wires, or stone baskets to move the foreign body into a more favorable position for snaring. In the single failure, the tip of a largely extravascular catheter fragment lay in a venous valve and could not be snared in a patient who refused surgery. Two of five patients with arterial foreign bodies suffered occlusive arterial spasm, reversible with local administration of nitroglycerine. Two large objects were repositioned to the femoral vein and removed by surgical cutdown. No other procedural complications occurred, and none of the patients required additional compression, transfusion, or surgical intervention.We conclude that use of preformed nitinol goose-neck snares facilitates retrieval of intravascular foreign bodies in most cases, although interventional radiologists must be familiar with a variety of techniques to deal with the expanding spectrum of foreign bodies currently encountered.

Purpose To evaluate the efficacy of the Wallstent endoprosthesis in the treatment of ureteral strictures. Patients and Methods Wallstents with diameters of 10 mm were placed across five malignant and six benign ureteral strictures in eight patients. All patients were believed to have poor surgical options, and their strictures were being maintained with catheter drainage. Ten lesions involved ureteroenteric anastomoses, and one malignancy involved the midureter. Results Three stents (two patients) across malignant disease remained patent until the time of patient death (3–5 months); the remaining two stents (one patient) became occluded within 1 month. Only one of six stents placed for benign disease remained patent at 11 months. All occlusions in benign strictures resulted from ingrowth of hyperplastic urothelium and granulation tissue. Complete obstruction was usually present only focally within the stent. The malignant occlusions were caused by tumor ingrowth and granulation tissue. No major complications were directly related to the stents, but two infections occurred. Conclusion Wallstent endoprostheses are ineffective in providing long-term relief in patients with benign ureteroenteric strictures. Further evaluation of their role in malignant strictures is needed.

Background Scan length reduction effectively decreases radiation dose at CT pulmonary angiography (CTPA) for pulmonary embolism (PE) but may exclude important incidental scan findings. Objective We aimed to determine the prevalence and nature of excluded findings with the use of reduced scan length CTPA. Methods We reviewed 335 consecutive emergency department CTPA studies performed on 16- or 64-detector row scanners with the use of a standard scan range. A scan length of 14.2 cm that was centered 4.1 cm below the carina has been shown to be adequate for PE diagnosis. Boundary slices for this scan range were determined. All pertinent and incidental findings within and outside the reduced scan range were noted. To determine the significance of newly detected excluded findings, we reviewed medical records and all relevant imaging studies before and 9–11 months after the reference CTPA. Results We found 374 pertinent findings in 192 patients, including 28 (8%) cases of PE. All except 3 (0.8%) were adequately seen with the reduced scan range, among which only one finding altered clinical management. There were a total of 230 incidental findings in 165 patients, 60 (26%) of which were excluded; 23 (10%) of the 60 were newly detected, including 10 thyroid nodules, 6 liver lesions, and an 8-mm pulmonary nodule. The reduced scan length decreased z-axis coverage by 49% ± 6%. Conclusion Substantial scan length reduction at CTPA may not compromise the diagnostic yield for pertinent alternative diagnoses.

Aim To determine the size and three-dimensional spatial distribution of pulmonary emboli (PE) at computed tomography angiography (CTA) to optimize the scan length. Materials and methods Two experienced radiologists jointly reviewed 100 consecutive, positive PE CTA studies performed in the Emergency Department (53 women; age 61 ± 17 years). All studies were conducted on a 16-detector row CT machine. In each case, the number of emboli was counted and the proximal and distal spatial coordinates of each embolus documented. Coordinates of the main pulmonary artery bifurcation (MPAb) and carina were recorded. For normalization, the thoracic cavity height (H)—from inlet to lowest hemidiaphragm—was measured. The minimal scan lengths for (a) capturing all emboli and (b) rendering a positive diagnosis were determined. Results Three hundred and seventy (370) emboli were detected. The average number of PE per patient was 3.7 (maximum 12, minimum 1). Their average length was 2.7 cm. Nine patients had saddle emboli (9%), and 71% of emboli were at or below the MPAb. An 18 cm (0.90×H) scan length, centred 4 cm (0.18×H) below the carina, captures all PE in this dataset while reducing z -axis coverage by 29% (34% for normalized data). Moreover, a 14.2 cm (0.78×H) scan length appropriately centred captures at least one embolus in all patients while reducing coverage by 44% (43%). Decreasing scan length to the lesser of 14.2 cm and 0.78×H per patient reduces coverage by 47%. Conclusion Scan length at CTA for PE can be reduced by up to 47% while preserving diagnostic accuracy for PE detection.

The purpose of our study was to investigate whether D-dimer screening is being used effectively to determine the need for MDCT in diagnosing acute pulmonary embolism (PE) in emergency department patients.We performed a retrospective review of all patients who underwent D-dimer testing or MDCT in the emergency department from January 1, 2003, through October 31, 2005. A D-dimer value of0.43 microg/mL was considered positive. Diagnosis of PE was made on the basis of the MDCT. Clinical algorithms for diagnosing PE mandate that patients with a low clinical suspicion for PE undergo D-dimer testing, then MDCT if positive. For patients with a high clinical suspicion for PE, MDCT should be performed without D-dimer testing.Of 3,716 D-dimer tests, 1,431 (39%) were positive and 2,285 (61%) were negative. MDCT was performed in 166 (7%) patients with negative D-dimer results and in 826 (58%) patients with positive D-dimer results. The prevalence of PE in patients with a high clinical suspicion and no D-dimer testing was 9% (139/1,628), which was higher than the rate of PE in the positive D-dimer group at 2% (19/826) (p0.0001). There was no significant difference in the prevalence of PE in the positive and negative D-dimer groups (2% vs 0.6%, respectively) (p = 0.23). The sensitivity and negative predictive value of D-dimer for PE were 95% (95% CI, 73.1-99.7%) and 99% (95% CI, 96.2-99.9%), respectively.D-dimer screening is not used according to established diagnostic algorithms to determine the need for MDCT in diagnosing acute pulmonary embolism in our emergency department.

The radiographic appearance of Ewing's sarcoma was studied retrospectively in 22 patients who survived 5 years or longer after diagnosis and treatment. Expected changes from treatment, including regression of the extraosseous soft tissue mass, periostitis, and reconstitution of the cortex, occurred in all patients. Local recurrence occurred in one patient 10 years after complete remission whereas secondary osteosarcoma occurred more than 5 years after complete remission in two other cases. Both recurrent and secondary tumors presented as new lytic foci at the site of the original primary lesion. Lytic changes from radiation (radiation osteitis) may develop more than 2 years after treatment and in this sample; such findings were widely distributed in the radiation port. The authors conclude that bone remodeling and postradiation changes occur slowly over 2 years after treatment, and that any localized lysis at the primary site is suspicious for recurrence or secondary neoplasm. Knowledge of the expected changes and patterns of local recurrence and secondary neoplasms helps one to detect any significant change in its early phase.

In a retrospective study of adrenal masses evaluated with computed tomography (CT), lesion x-ray attenuation was compared with size and radiologists' interpretations in discriminating benign lesions from malignant ones. Unenhanced CT attenuation coefficient and size were analyzed electronically in 55 patients with 66 adrenal masses. There were 38 nonhyperfunctioning adenomas in 33 patients and 28 malignant masses in 22 patients. Primary extraadrenal malignancies were present in 45 of the 55 patients. Three blinded readers characterized the adrenal masses using a seven-point scale of certainty. Results were subjected to receiver operating characteristic (ROC) analysis. The mean CT attenuation coefficient for benign adrenal masses was -2.2 HU +/- 16.0 and was significantly different from the mean for malignant lesions (28.9 HU +/- 10.6). The area under the ROC curve for CT attenuation coefficients (0.91 +/- 0.04) was significantly larger than that for lesion size (0.84 +/- 0.05) or best observer interpretation (0.84 +/- 0.05). A threshold CT attenuation value of 0 HU had a sensitivity-to-specificity ratio of 47%:100% for characterizing benign adrenal masses, whereas a threshold attenuation of 10 HU had a ratio of 79%:96%.

Objective: To compare examination volume and diagnostic yield of computed tomography (CT) pulmonary angiography (CTPA) and ventilation-perfusion (V/Q) scintigraphy for detection of suspected pulmonary embolism (PE) in emergency department patients. Methods: Every CTPA and V/Q scan result for emergency department patients between October 2001 and September 2005 were reviewed. Patients with prior PE and follow-up examinations were excluded. Results: Atotalof3421 CTPA examinations and 198 V/Q scans met inclusion criteria. Average CTPA examinations completed per month increased 227%, from 33.4 to 109.2 for the first and last 24-month periods, respectively. Ventilation-perfusion scintigraphy volume decreased 80% (from 6.9 to 1.4 per month). Total diagnoses of PE per month increased 89% from 4.0 to 7.5, whereas the percentage of positive CTPA examinations dropped from 9.8% to 6.8%. Conclusions: Availability of CT in the emergency department and lower physician thresholds for test utilization have increased the use of CT pulmonary angiography and increased detection of PE.

In light of their potential for devastating consequences, a liberalized screening approach for blunt cerebrovascular injuries (BCVI) is becoming increasingly accepted. The gold standard for diagnosis of BCVI is arteriography (ART), but noninvasive diagnostic alternatives offer clear advantages. Prospective comparative studies found that computed tomographic angiography (CTA) was unreliable in detecting BCVI. However, with advanced CTA technology, it has become more difficult to justify ART in asymptomatic patients. We implemented a liberal screening protocol for BCVI, employing 16-slice CTA. We hypothesized that CTA would detect all clinically significant BCVI.Patients undergoing computed tomography scanning for cranial or cervical trauma, as well as those with high-risk injury mechanisms or patterns, had craniocervical CTA performed. Abnormal CTA studies were confirmed with ART. Patients were followed for signs or symptoms of ischemic neurologic events.Over an 11-month period, 331 patients underwent CTA. Eighteen (5.4%) patients had a total of 20 BCVI. There were 11 carotid and 9 vertebral artery injuries. Seven of nine (78%) vertebral injuries were associated with cervical spine fractures. Two patients who did not meet screening criteria had signs or symptoms of BCVI before CTA. No patient with normal CTA developed evidence of BCVI.CTA detected all clinically significant injuries during this study period. Liberal screening with 16-slice CTA is appropriate and is likely to miss very few significant injuries. A multicenter trial will help to clarify risk factors and the accuracy of noninvasive diagnostic modalities.

Renal vascular injuries such as transection, thrombosis, dissection, and arteriovenous fistula formation are unusual but well-recognized consequences of blunt abdominal trauma. We discuss a rare case of renal artery pseudoaneurysm presenting 6 weeks after blunt abdominal trauma that was successfully treated with selective embolization.

Purpose The authors describe their experience with reinsertion of accidentally removed tunneled venous catheters using existing subcutaneous tracts. Materials and Methods Replacement of 13 dislodged tunneled venous catheters was attempted a median of 12 hours (range, 3 hours to 5 days) after accidental removal. The catheters were needed for hemodialysis ( n = 11), plasmapheresis ( n = 1), or antibiotic therapy ( n = 1). The tunnel exit was probed in the same fashion as for a dislodged nephrostomy tube, and new catheters were reinserted once a guide wire was advanced into the central veins. The medical record was reviewed to determine materials used and occurrence of complications, if any. Results Replacement was successful in 12 of 13 patients. The remaining patient had a new catheter placed through a fresh puncture during the same visit. There were no infections associated with re-use of existing tunnels. In five patients, after probing the tract with a guide wire, new catheters were simply advanced into the desired position. Seven other successes required additional manipulations with use of dilators and peel-away sheaths. Conclusions Tunneled catheters that "fall out" can be readily replaced even when reinsertion is attempted up to 5 days later. This represents an important contribution that radiologists can offer in the management of venous access cases.

To determine the accuracy and reliability of magnetic resonance (MR) angiography for identification of stenosis and patent distal vessels in patients with peripheral vascular disease.Two-dimensional time-of-flight MR angiography and conventional arteriography were performed in 22 patients. Four blinded radiologists independently graded multiple anatomic segments.MR angiography allowed detection of more patent vessel segments than did conventional arteriography. For detection of significant stenosis (75%), MR angiography had 43%-67% sensitivity and 74%-89% specificity. Discrepancies in detection of significant stenosis occurred in 39 segments for the most accurate reviewer; 27 of these discrepancies were avoidable.For detection of significant stenosis, MR angiography has low to moderate sensitivity and specificity; however, observer variability appears to be a major contributing factor to the discrepancies. Greater reviewer experience or techniques for improving reliability may improve the accuracy of MR angiography in peripheral vascular disease.

To evaluate the efficacy of transvenous systemic embolotherapy with a neuroradiologic detachable balloon.As part of a clinical trial, a detachable silicone balloon was used to occlude pulmonary arteriovenous malformations (PAVMs) in 35 patients and varicocele in 14 patients. In patients with PAVM, the indications for embolotherapy were prophylaxis against paradoxic embolization (n = 35) and dyspnea, fatigue, or both (n = 26); in patients with varicocele, they were pain or a discomforting lesion (n = 8) or infertility (n = 6).Balloon embolotherapy was successful in 79 (98%) of the 81 lesions in which it was attempted; 29 (37%) of these 79 balloons were used in conjunction with coils. Ninety-six (97%) of 99 balloons were successfully placed; the three technical failures had no substantial clinical sequelae, and in all three, occlusion was eventually achieved with either detachable balloons or coils. Of six late deflations, five occurred in balloons placed adjacent to coils; only one, which occurred between 1 day and 21 days after placement, resulted in recanalization.Transvenous embolization with this detachable balloon was relatively simple and provided cross-sectional occlusion of PAVMs and varicocele.

We studied the efficacy of T2 measurements at high field strength in distinguishing between liver hemangiomas and hepatic metastases when an ultrafast (single-excitation) MR imaging technique is used. Fourteen patients with known liver tumors were imaged in a 2.0-T prototype ultrafast MR scanner with a spin-echo (infinite TR and TE of 30-340 msec) pulse sequence. Each image was obtained with a total data acquisition time of 20 msec. T2 calculations for hepatic metastases (n = 6) showed a mean of 79.3 +/- 13.5 msec, whereas hemangiomas (n = 8) showed a T2 of 139.8 +/- 18.8 msec (p less than .0001). T2 values of lesions had a smaller relative standard deviation than previously reported, and the range of T2 values of hemangiomas (119-181 msec) and metastases (68-103 msec) did not overlap. Our preliminary results suggest that T2 calculations with ultrafast MR imaging may be useful for differentiating hemangiomas from metastases. We hypothesize that T2 values obtained from ultrafast MR images are more reliable than those obtained from conventional MR images, primarily because of the elimination of T1 information and effects of motion on image signal intensity.

Objective. —To determine whether radiologists' interpretations of images are biased by their context and by prevalence of disease in other recently observed cases. Methods. —A test set of 24 right pulmonary arteriograms with a 33% prevalence of pulmonary emboli (PE) was assembled and embedded in 2 larger groups of films. Group A contained 16 additional arteriograms, all showing PE involving the right lung, so that total prevalence was 60%. Group B contained 16 additional arteriograms without PE so that total prevalence was 20%. Six radiologists were randomly assigned to see either group first and then "cross over" to review the other group after a hiatus of at least 8 weeks. The direction of changes in a 5-point rating scale for the 2 readings of each film in the test set was compared with the sign test; mean sensitivity, specificity, and areas under receiver operating characteristic (ROC) curves were compared with the paired t test. Results. —In the context of group A's higher disease prevalence, radiologists shifted more of their diagnoses toward higher suspicion than expected by chance ( P =.03, sign test). In group A, mean sensitivity for diagnosing PE was significantly higher (75% vs 60%; P =.04), and area under the ROC curve was significantly larger (0.88 vs 0.82; P =.02) Conclusions. —Radiologists' diagnoses are significantly influenced by the context of interpretation, even when spectrum and verification bias are avoided. This "context bias" effect is unique to the evaluation of subjectively interpreted tests, and illustrates the difficulty of obtaining unbiased estimates of diagnostic accuracy for both new and existing technologies.