Evidence-based imaging in trauma radiology: where we are and how to move forward

Evidence-based imaging is an important tool for the determination of which subjects should undergo imaging and what imaging approach should be applied. This paper includes a summary of the methods of evidence-based imaging, including formulation of a clinically relevant question, searching the medical literature, critically analyzing the data, summarizing the evidence, and applying the evidence to practice. Application of the many validated clinical prediction rules to trauma imaging is also discussed. In addition, the challenges of evidence-based imaging and the needs for future research are detailed.

Correlation of MR imaging findings with intraoperative findings after cervical spine trauma

BACKGROUND AND PURPOSE

There are limited data correlating MR imaging and anatomic findings of ligamentous injury in cervical spine trauma. This study compares acute MR imaging with surgical observations of disk/ligamentous injury after blunt cervical trauma.

MATERIALS AND METHODS

Consecutive patients with acute cervical spine trauma who underwent preoperative MR imaging and surgery from 1998 to 2001 were identified. MR imaging was obtained within 48 hours of injury for most patients. All scans included sagittal T1, T2 fat-saturated, and short tau inversion recovery sequences. At surgery, extent of injury at the operated level was recorded on a standardized form for either anterior or posterior structures or both depending upon the operative approach. MR examinations were separately evaluated by 2 readers blinded to the intraoperative findings. Radiologic and surgical findings were then correlated.

RESULTS

Of 31 patients, an anterior surgical approach was chosen in 17 patients and a posterior approach in 13 patients. In one patient anterior and posterior approaches were utilized. Seventy-one percent of patients had spinal cord injury on MR imaging. MR imaging was highly sensitive for injury to disk (93%), posterior longitudinal ligament (93%), and interspinous soft tissues (100%), but it was less sensitive for injury to the anterior longitudinal ligament (71%) and ligamentum flavum (67%). For most ligamentous structures, there was limited agreement between specific MR imaging findings and injury at surgery.

CONCLUSION

In acute cervical spine trauma, MR imaging has moderate to high sensitivity for injury to specific ligamentous structures but limited agreement between specific MR imaging findings and injury at surgery. MR imaging may overestimate the extent of disruptive injury when compared with intraoperative findings, with potential clinical consequences.

Frakturen des Atlas und Axis bei älteren Patienten: Untersuchung des Radiologischen Spektrums der Frakturen und bedeutsamer Faktoren für die Bildgebende Diagnostik

OBJECTIVE

To describe patterns of atlantoaxial fractures in a population of consecutive elderly patients, including assessment of type, distribution and associated clinical and radiological findings, and to analyze any influence of the causative trauma mechanism on the individual fracture pattern.

MATERIALS AND METHODS

The distribution and type of 123 atlantoaxial fractures in 95 subjects older than 65 years (range: 65 - 102; mean age: 79 years) were retrospectively assessed. For each subject, trauma mechanism and clinical and neurological status were recorded at admission. Initial imaging studies of the cervical spine were reviewed. Preexistent degenerative changes were assessed and the atlantoaxial fractures classified. Data were evaluated for the frequency of different types of fractures of C1 and C2 and for accompanying fractures of cervical vertebrae or the occipital condyles, respectively.

RESULTS

The majority of patients with injuries of the atlantoaxial complex had fractures of C2 (90 of 95, 95 %). A large proportion of these patients (67 of 90, 74 %) had odontoid fractures. An isolated fracture of C1 was present in only 5 (5 %) patients. Associated fractures of the occipital condyles or other cervical vertebrae were rare (10 of 95, 11 %). The main trauma mechanism for atlantoaxial injuries was a fall (56 of 95, 59 %). Elderly patients injured in motor vehicle accidents were more likely to have isolated fractures of C2 and Type III fractures of the odontoid (p < 0.02).

CONCLUSION

In elderly patients, fractures of the atlantoaxial complex are mainly caused by falls and almost always involve C2. The trauma mechanism influences the fracture pattern.

Pelvic ring disruptions in emergency radiology

Pelvic ring disruptions are a result of high-energy trauma and are a significant cause of morbidity and mortality in major trauma patients. The initial pelvic radiograph, in combination with computed tomographic imaging in selected patients, provides a quick and accurate method of diagnosing pelvic fractures. Pelvic fracture classification schemes have evolved over the past few decades, and include description of the mechanism of injury and assessment of pelvic stability. Understanding these classifications is important in developing an approach to interpretation of pelvic imaging and prediction of associated injuries. Armed with these tools, the emergency radiologist can detect pelvic fractures early and guide subsequent imaging and therapy.

Cervical spine injuries in patients 65 years old and older: epidemiologic analysis regarding the effects of age and injury mechanism on distribution, type, and stability of injuries

OBJECTIVE

Our objective was to describe types and distribution of cervical spine injuries in elderly patients in regard to causative trauma mechanism and patient age.

MATERIALS AND METHODS

The distribution and type of 225 cervical spine injuries in 149 consecutive patients 65 years old and older over a 5-year interval were retrospectively assessed. For each patient, initial admission imaging studies were reviewed, and injuries were classified. Trauma mechanism (falls from standing or seated height vs higher energy mechanisms) and initial clinical and neurologic status were recorded. Data were correlated according to patients' age (65-75 years and >75 years) and causative trauma mechanism.

RESULTS

Ninety-five (64%) of 149 patients had upper cervical spine injuries. Fifty-nine (40%) of 149 patients had multilevel injuries. Main causes for cervical spine injuries were motor vehicle crashes in "young elderly" (65-75 years old; 36/59, 61%) and falls from standing or seated height in "old elderly" (>75 years old; 36/90, 40%). Fracture patterns at risk for neurologic deterioration were common (>50%), even in the absence of acute myelopathy or radiculopathy. Patients older than 75 years, independent of causative mechanism, and patients who fell from standing height, independent of age, were more likely to have injuries of the upper cervical spine (p = 0.026 and p = 0.006, respectively).

CONCLUSION

Cervical spine injuries in elderly patients tend to involve more than one level with consistent clinical instability and commonly occur at the atlantoaxial complex. Old elderly patients and patients who fall from standing height are more prone to injuries of the upper cervical spine.

Animated-simulation modeling facilitates clinical-process costing

Traditionally, the finance department has assumed responsibility for assessing process costs in healthcare organizations. To enhance process-improvement efforts, however, many healthcare providers need to include clinical staff in process cost analysis. Although clinical staff often use electronic spreadsheets to model the cost of specific processes, PC-based animated-simulation tools offer two major advantages over spreadsheets: they allow clinicians to interact more easily with the costing model so that it more closely represents the process being modeled, and they represent cost output as a cost range rather than as a single cost estimate, thereby providing more useful information for decision making.

Resource cost analysis of cervical spine trauma radiography

PURPOSE

To determine the resource costs of the technical component of cervical spine radiography in patients with trauma and the factors that drive resource costs, to provide a model for resource cost estimation, and to compare resource costs with other methods of cost estimation.

MATERIALS AND METHODS

Direct measurement was made of technologist labor and supply costs of a cohort of 409 consecutive patients with trauma who underwent cervical spine radiography. Probability of cervical spine injury was determined by reviewing emergency department medical records. An animated simulation model was used to combine cost and injury probability estimates to determine resource costs. Sensitivity analysis explored factors that determined costs and estimated uncertainty in model estimations. Comparison was made with other cost estimates.

RESULTS

The average technical resource cost for cervical spine radiography was $49.60. Both direct labor ($19.60 vs $13.33; P <.005) and film ($8.39 vs $6.76; P <.005) costs were greater in patients with high probability of injury than in those with low probability of injury. Overall costs in patients with high probability of injury exceeded those in patients with low probability of injury by 33%. Resource costs exceeded Medicare resource-based relative value unit reimbursements for all patients with trauma.

CONCLUSION

Resource costs of the technical components of cervical spine radiography varied with patient probability of injury and were higher than Medicare reimbursements.

Helical CT in the primary trauma evaluation of the cervical spine: an evidence-based approach

This review provides a summary of the cost-effectiveness, clinical utility, performance, and interpretation of screening helical cervical spine CT for trauma patients. Recent evidence supports the use of helical CT as a cost-effective method for screening the cervical spine in high-risk trauma patients. Screening cervical spine CT can be performed at the time of head CT to lower the cost of the evaluation, and when all short- and long-term costs are considered, CT may actually save money when compared with traditional radiographic screening. In addition to having higher sensitivity and specificity for cervical spine injury, CT screening also allows more rapid radiological clearance of the cervical spine than radiography. Patients who are involved in high-energy trauma, who sustain head injury, or who have neurological deficits are candidates for CT screening. Screening with CT may enhance detection of other potentially important injuries of the cervical region.

Extending simulation modeling to activity-based costing for clinical procedures

A simulation model was developed to measure costs in an Emergency Department setting for patients presenting with possible cervical-spine injury who needed radiological imaging. Simulation, a tool widely used to account for process variability but typically focused on utilization and throughput analysis, is being introduced here as a realistic means to perform an activity-based-costing (ABC) analysis, because traditional ABC methods have difficulty coping with process variation in healthcare. Though the study model has a very specific application, it can be generalized to other settings simply by changing the input parameters. In essence, simulation was found to be an accurate and viable means to conduct an ABC analysis; in fact, the output provides more complete information than could be achieved through other conventional analyses, which gives management more leverage with which to negotiate contractual reimbursements.

Cervical spine injury: a clinical decision rule to identify high-risk patients for helical CT screening

OBJECTIVE

We aimed to validate the routine use of a clinical decision rule to direct diagnostic imaging of adult blunt trauma patients at high risk for cervical spine injury.

MATERIALS AND METHODS

We previously developed and have since routinely used a prediction rule based on six clinical parameters to identify patients at greater than 5% risk of cervical spine injury to undergo screening helical CT of the cervical spine. During a 6-month period, 4285 screening imaging studies of the cervical spine were performed in adult blunt trauma patients. Six hundred one patients (398 males, 203 females; age range, 16-100 years; median age, 38 years) underwent helical CT, and the remainder underwent 3684 conventional radiographic examinations. Clinical and report data were extracted from the radiology department database, medical records, and the hospital trauma registry. Abnormal findings were independently confirmed by additional imaging studies, autopsy results, or clinical outcome.

RESULTS

The true-positive cervical spine injury rates in helical CT- and conventional radiography-screened patients who presented directly to our trauma center were 40 (8.7%) of 462 and seven (0.2%) of 3684, respectively. The cervical spine injury rate in patients who were transferred from outside institutions to our trauma center and who underwent helical CT was 37 (26.6%) of 139. This figure included 20 patients already known to have cervical spine fracture.

CONCLUSION

The clinical decision rule can distinguish patients at high and low risk of cervical spine injury, thus supporting its validity.

Communicating with the family: the risks of medical radiation to conceptuses in victims of major blunt-force torso trauma

BACKGROUND

Trauma surgeons must balance the risk and benefits of diagnostic radiographic procedures on potentially pregnant patients and should know the range and likelihood of effects that radiation might have on pregnancy.

METHODS

We present guidelines for assessing such radiation risks. Knowledge of a patient's pregnancy status and an estimate of radiation dose to the conceptus (low, < 10 mGy [milligray]; intermediate, 10-250 mGy; high, > 250 mGy) allow provisional assessment of radiation-induced risks.

RESULTS

Dose estimates may be estimated at 2 mGy per exposure (radiographs), 5 mGy per slice (computed tomography), and 10 mGy per minute of fluoroscopy, when the conceptus is within the x-ray field. A formal radiation exposure assessment is appropriate when provisional estimates exceed 10 mGy.

CONCLUSION

A simple qualitative dose assessment can inform clinical decisions and guide appropriate triage to more formal quantitative assessment.

Determining risk of traumatic aortic injury: how to optimize imaging strategy

OBJECTIVE

Our objective was to develop and validate a clinical prediction rule that determines patient probability of traumatic aortic injury to guide selection of optimal screening imaging strategy.

MATERIALS AND METHODS

A 2-year, single-institution retrospective case-control study was conducted of 31 cases of traumatic aortic injury and 171 random major trauma control subjects. The presence of potential injury predictors was determined from chart review. Logistic regression was used to determine injury predictors, and clinically similar predictors were combined into composite predictors. The composite predictors were used to develop a seven-point injury index clinical prediction rule using multivariate logistic regression. Injury probabilities were determined through Bayes' theorem. Bootstrap validation was performed.

RESULTS

Predictors of aortic injury included head injury (odds ratio, 18.3; 95% confidence interval [CI], 7.3-46), pelvic fracture (odds ratio, 27.3; 95% CI, 8.8-85), pneumothorax (odds ratio, 27.3; 95% CI, 8.8-85), and lack of seat belt use (odds ratio, 6.8; 95% CI, 2.6-17). The seven composite predictors of age, unrestrained vehicle occupant, hypotension, thoracic injury, abdominopelvic injury, extremity fracture, and head injury, were combined into the seven-point injury index. In the injury index, each composite predictor had an adjusted odds ratio of 7.1 (95% CI, 3.7-13.5), and the odds ratios were additive. The injury index prediction rule had an area under the receiver operating characteristic curve of 0.97. All injured patients had at least one composite predictor.

CONCLUSION

The probability of traumatic aortic injury can be estimated from the injury index prediction rule. Because cost-effectiveness of various imaging strategies depends on probability of injury, the prediction rule can guide imaging selection.

Cervical spine screening with CT in trauma patients: a cost-effectiveness analysis

PURPOSE

To investigate the cost-effectiveness of computed tomography (CT) relative to radiography for cervical spine screening in trauma patients.

MATERIALS AND METHODS

A decision analysis model was constructed to compare the incremental cost-effectiveness of radiography and CT as primary cervical spine screening modalities in trauma patients. Analyses were performed from a societal perspective, and probability and cost estimates from the literature and institutional experience were used. In separate cost-effectiveness analyses, hypothetical cohorts of trauma patients from three defined clinical scenarios were considered: high, moderate, and low risk for cervical spine fracture. Outcome measures included cases of paralysis prevented, total cost of screening strategies, and incremental cost-effectiveness ratios.

RESULTS

In high-risk patients, screening with CT is a dominant strategy that prevents cases of paralysis and saves money for society. In moderate-risk patients, screening with CT is cost-effective with reference-case assumptions and within the range of most sensitivity analyses. In the low-risk group, CT screening helps prevent cases of paralysis, but the incremental cost-effectiveness ratio is high (> $80,000 per quality-adjusted life year).

CONCLUSION

CT is the preferred cervical spine screening modality in trauma patients at high and moderate risk for cervical spine fracture.

Cervical spine imaging in patients with trauma: determination of fracture risk to optimize use

PURPOSE

To develop a method to use clinically apparent factors to determine cervical spine fracture risk to guide selection of optimal imaging strategies.

MATERIALS AND METHODS

Records from 472 patients with trauma (168 with fractures, 304 control patients) who visited the emergency department in 1994 and 1995 were reviewed for 20 potential predictors of cervical spine fracture in this retrospective case-control study. Simple logistic regression was used to determine predictors of cervical spine fracture. Prediction rules were formulated by using multiple logistic regression and recursive partitioning with bootstrap validation. Posttest fracture probabilities were calculated from base prevalence and likelihood ratios derived for predictors by using Bayes theorem.

RESULTS

Predictors of cervical spine fracture included severe head injury (adjusted odds ratio [OR] = 8.5, 95% CI: 4.0, 17.0), high-energy cause (OR = 11.6, 95% CI: 5.4, 25.0), and focal neurologic deficit (OR = 58, 95% CI: 12, 283). The prediction rule was used to stratify patients into groups with fracture probabilities of 0.04%-19.70%. After adjusting for overfitting, the area under the receiver operating characteristic curve was 0.87.

CONCLUSION

Clinically apparent factors, including cause of injury, associated injuries, and age, can be used to determine the probability of cervical spine fracture. Development of evidence-based imaging guidelines should incorporate knowledge of fracture probability.

A critical synopsis of the diagnostic and screening radiology outcomes literature

In summary, the radiology outcomes research literature is both extensive and broad. The methodologic quality, however, is quite variable. Overall, this quality could be improved by intervention in two areas: methodologic dissemination and development. The number of researchers investigating radiology-related outcomes is high, and presently there are over 20 journals devoted exclusively to radiology research. Even with a relatively narrow definition of "outcomes," we identified over 200 radiology outcomes studies, most from the past few years. However, the methodologic quality of most of these articles was relatively low, with important design flaws and biases. Nonetheless, a substantial number of radiology publications do employ state-of-the-art research methods and innovative approaches to methodologic challenges. The quality of radiology outcomes research overall would benefit tremendously from dissemination of such research methods. Instruction in outcomes research methods is accessible to radiologists. For example, there have been several recent articles and series of articles on outcomes research methods in JAMA, including guidelines for the performance and reporting of cost-effectiveness analyses (38-40) and for developing clinical prediction rules (57). Within radiology, several recent articles have appeared on, among other things, cost-effectiveness analysis (34,59,60), assessing quality of life (43), screening for disease (53), and defining the study population (61). The research compendium compiled for the GERRAF (General Electric-Association of University Radiologists Radiology Research Academic Fellowships) program remains a comprehensive methodologic source for many of the issues in radiology outcomes research, and outcomes research methods courses are offered every year at the Society for Health Services Research in Radiology and Society for Medical Decision Making meetings, as well as at the meeting of the Radiological Society of North America. Even so, awareness of the need for such research techniques remains limited. Dissemination of sound research methods is limited at least in part by the current incentives in radiology research. At many institutions, the number of research publications produced, rather than their quality, determines promotion or academic success. Unfortunately, more rigorous study designs often require more time and resources. Further, because peer reviewers are often as uninformed about research methods as the bulk of those who are submitting manuscripts, it may actually be more difficult to publish articles with more advanced methodologic designs. The standard in radiology is the uncontrolled case series, and deviation from the standard may make acceptance for publication more difficult. On a more optimistic note, recent publication of a number of methodology articles suggests that at least some journals are promoting improved research in methodology (43,53,59-61). We hope that time will be available for manuscript reviewers to learn to understand the strengths and weaknesses of various research approaches. If more rigorous study designs were required for publication, radiology outcomes research would probably improve drastically. Nevertheless, the current peer-review system does not effectively promote sound research design. The other great incentive in research is funding. Clearly, if advanced research design is required for funding, then there is incentive for improvement in research quality. Traditionally, National Cancer Institute and other National Institutes of Health and public sector funding has been predicated on a high level of research sophistication. Undoubtedly, availability of grants for diagnostic and screening imaging clinical trials and other research will go far to improve radiology research methods. The other traditional source of research funding is industry.

Economic analyses of radiological procedures: a methodological evaluation of the medical literature

OBJECTIVE

Increasing pressure to curb health care costs has led to considerable interest in economic analyses, including both cost-effectiveness and cost-benefit analyses. Numerous economic analyses of radiological procedures have appeared in both the radiology and non-radiology literature. The objective of this study was to evaluate the methodological quality of economic analyses of radiological procedures published in the non-radiology medical literature during the years 1990 1995.

METHODS

Original investigations from the medical (non-radiological) literature that include economic analyses of radiological interventions were identified from a computerized literature search. Each economic analysis article was evaluated by two independent reviewers for adherence to ten methodological criteria. The criteria were derived from review of the medical and radiological economic analysis methodology literature and consisted of the following: (1) Comparative options stated; (2) perspective of analysis defined; (3) outcome measure identified; (4) cost data included; (5) source of cost data stated; (6) long term costs included; (7) discounting employed; (8) summary measure provided; (9) incremental computation method used; and (10) sensitivity analysis performed. The results were compared to a previous study that evaluated the radiological literature.

RESULTS

Of the 56 articles in the medical literature that included economic analyses of radiological procedures, only eight (14%) conformed to all ten methodological criteria. The cost data (98%) and comparative options (89%) criteria exhibited high compliance, while the perspective of analysis (25%) and discounting (32%) criteria had relatively low compliance. Agreement between the reviewers was excellent (kappa = 0.88).

CONCLUSIONS

Published economic analyses of radiology procedures usually do not meet accepted methodological standards.

Economic analyses in obstetrics and gynecology: a methodologic evaluation of the literature

OBJECTIVE

To evaluate the methodology of the cost-effectiveness and cost-benefit literature in obstetrics and gynecology.

DATA SOURCES

We performed a MEDLINE search of the general and subspecialty obstetrics and gynecology journals for the years 1990 through 1996.

METHODS OF STUDY SELECTION

Original investigations including cost-effectiveness or cost-benefit were evaluated by two reviewers for adherence to ten minimum methodologic standards derived by a review of guidelines for medical economic analyses. The major criteria considered included: 1) provision of comparative options, 2) statement of analytic perspective, 3) presentation of cost data, 4) identification of outcome measure, 5) use of summary measure of economic effectiveness or benefit, and 6) performance of a sensitivity analysis. The minor criteria evaluated included: 1) statement of source of cost data, 2) inclusion of long-term costs, 3) use of discounting, and 4) calculation of an incremental summary measure.

TABULATION, INTEGRATION, AND RESULTS

Ninety-eight articles that included cost-benefit or cost-effectiveness analyses were identified. The mean number of major and minor principles adhered to were 3.6 and 1.0, respectively. Five publications (5.1%) conformed to all ten major and minor criteria, whereas nine (9.2%) articles used all six major criteria. The provision of cost data (94.8%) and statement of comparative options (96.9%) were the major principles most frequently adhered to, whereas the use of discounting (10.2%) and statement of analytic perspective (19.3%) showed the lowest compliance. Agreement between the reviewers was excellent (kappa .87).

CONCLUSION

Published economic analyses in the obstetrics and gynecology literature seldom adhere to all recommended methodologic guidelines. Further training in the methodology of cost-effectiveness analysis is needed within the specialty.

Methodologic evaluation of the radiology cost-effectiveness literature

PURPOSE

To evaluate the methodologic quality of cost-effectiveness and cost-benefit analyses reported in the radiology literature.

MATERIALS AND METHODS

Original investigations of cost-effectiveness and cost-benefit analysis were identified from an on-line search of the radiology literature from 1989 to 1995. The articles were evaluated for adherence to minimum methodologic standards for economic analysis research. Major criteria assessed were (a) provision of comparative options, (b) statement of perspective of analysis, (c) presentation of cost data, (d) measurement of outcomes, (e) use of a summary measure of economic efficiency, and (f) performance of sensitivity analysis. Minor criteria assessed were inclusion of (a) source of cost data, (b) long-term costs, (c) discounting, and (d) incremental computation of the summary measure.

RESULTS

Forty-four economic analysis articles were identified. The median numbers of major and minor principles adhered to were three and one, respectively. Five studies used all six major criteria, and three used all 10 criteria. The median number of criteria adhered to did not increase during the study period.

CONCLUSION

Adherence to methodologic standards in the radiology cost-effectiveness literature is not optimal. There are several examples from radiology journals, however, where such standards are met.

Aqueduct compression from venous angioma: MR findings

Vascular compression as the cause of aqueductal stenosis is rare. In a 16-year-old girl with hydrocephalus, MR imaging provided evidence of aqueductal stenosis caused by a venous angioma in the tectum and midbrain. This indicates the usefulness of MR imaging for the evaluation of obstructive hydrocephalus.

Pleural fluid volume estimation: a chest radiograph prediction rule

RATIONALE AND OBJECTIVES

We devised a prediction rule for estimating pleural effusion volume on the basis of posteroanterior and lateral chest radiographs.

METHODS

A prediction rule was devised for estimating pleural effusion volume on the basis of the presence or absence of a meniscus on chest radiographs. The rule was tested and validated using separate data sets obtained from a retrospective review of patients having both a chest radiograph and computed tomography (CT) scan (the gold standard) within 24 hr of each other. The accuracy of the prediction rule and the degree of interobserver agreement between the two independent readers were determined.

RESULTS

For the test and validation sets, the weighted accuracies of the prediction rule were 86% and 85%, respectively. The respective weighted interobserver agreements were 97% and 88%. Pleural effusions became visible as a meniscus on the lateral chest radiograph at a volume of approximately 50 ml; at a volume of 200 ml, the meniscus could be identified on the posteroanterior radiograph. At a volume of about 500 ml, the meniscus obscured the hemidiaphragm.

CONCLUSION

The volume of a pleural effusion can be estimated from the chest radiograph appearance with a reasonable degree of accuracy.

Verrucous carcinoma of the bladder

Verrucous carcinoma is a rare squamous cell variant that may arise in the bladder. We present a case of verrucous carcinoma of the bladder with radiologic-pathologic correlations that demonstrate the characteristic frond-like growth pattern. Verrucous carcinoma is an invasive but not a metastizing lesion, and therefore recognizing this lesion may have prognostic and therapeutic implications.

Magnetic resonance imaging of blood and clots in vitro

The effects of variations in blood clot composition on magnetic relaxation rates and magnetic resonance (MR) image have been characterized in vitro. Both 1/T1 and 1/T2 were found to be linear functions of hematocrit for blood and clot, with increases in hematocrit resulting in progressive decreases in image signal intensity. Clot formation in fully oxygenated samples produced no change in relaxation rates or MR images compared with unclotted blood, but clot retraction was associated with a significant increase in 1/T2 that resulted in a decreased signal. Retraction resulted in a heterogenous image with appearance of a hypointense peripheral rim; differences in the method of clot preparation resulted in significant image inhomogeneity. The pattern of fibrinolysis was found to depend on the type of plasminogen activator used and its site of initial application. Injection of tissue plasminogen activator into the clot resulted in lysis, primarily in the clot interior, whereas placing the enzyme in the surrounding serum caused degradation from the outside of the clot. Both observations are consistent with the high binding affinity of tissue plasminogen activator for fibrin. By comparison, streptokinase, with low fibrin binding affinity, dissolved thrombi in a peripheral pattern whether injected into the thrombus or introduced in the serum. These findings identify several variables of clot composition and structure that influence MR images of thrombi and should be considered in their interpretation.