Twenty years ago, emergency radiology meant a plain chest or ankle film or perhaps an intravenous urogram, with all of the studies being interpreted by a clinician. Today, the field is one of the most sophisticated imaging practices in the hospital. Emergency radiologists are called on to make a rapid diagnosis of a multitude of conditions from the head (eg, stroke, intracerebral hemorrhage from bleeding aneurysms) to the chest (pulmonary embolism) to the abdomen (appendicitis, bowel obstruction) to the legs (deep venous thrombosis), as well as injuries to every part of the body.
Predictably, the simple x-ray machine is far from enough. “The emergency radiology department has to have the most up-to-date state-of-the-art equipment,” stresses Robert A. Novelline, MD, director of emergency radiology, Massachusetts General Hospital, Boston, where 80,000 patients are seen in the emergency department a year. “That means digital radiography, the latest in ultrasound equipment, multidetector CT scanners, and magnetic resonance imaging. And you may well need more than one: ideally, all required imaging should be obtained within an hour of the patient’s admission. Such rapid throughput improves patient outcome, and it also saves money, because if a patient has to wait 4 hours for a CT scan, the hospital bears the cost of providing nursing care for that period. The need for immediate sophisticated procedures also means that radiology is no longer a 9-to-5 proposition, with other physicians being called on to interpret studies obtained after hours. Today’s emergency radiology department requires around-the-clock staffing, especially in Level I trauma centers, with radiologists being immediately available either in-house or by teleradiology.”
The CT scanner has displaced the x-ray machine at the heart of the department.
“CT has revolutionized the work-up of the emergency patient,” Novelline points out. “With CT, diagnoses are obtained faster, the study is easier on the patient and more accurate, and it can be done in a more cost-efficient way. As an example, detecting conditions such as a dissection of the aorta or an aortic aneurysm used to require an arteriogram. Now, we can obtain the same information with a 30-second CT scan, avoiding angiography, which is more expensive and more morbid.” Especially desirable is the multidetector-array scanner, which provides greater speed and coverage and higher spatial resolution, along with many options for three-dimension reconstruction of the images.1,2
Massachusetts General Hospital’s emergency department has two such scanners, and every night shift (10 PM to 8 AM), they are used to perform 35 to 50 scans. According to Novelline, the number of imaging studies of all types averages about one per patient, with some patients needing no studies and others needing several. One to one is the national average for an emergency practice, he notes.
The high level of sophistication of emergency radiology implies a need for specialized training. All radiology residents at Massachusetts General Hospital are required to spend at least 2 months in the emergency department. The hospital also offers a fellowship in the specialty, as do some other medical centers, and the American Society of Emergency Radiology has been established to address the issues of specific concern in this field.
Chest pain is one of the most common reasons for a visit to an emergency department and the cause of between 20% and 30% of hospital admissions from the department.3 The initial study usually is a chest film, which can quickly diagnose many acute conditions such as pneumonia, rib fracture, or congestive heart failure. The patient history and a detailed physical examination dictate the next study (if any). If aortic dissection is suspected, there will be a CT study with a specialized aortic dissection protocol. If the patient is allergic to iodinated contrast medium or has bad renal function, an MRI scan is performed instead.
A diagnosis of myocardial infarction usually is based on the clinical history, the physical examination, the electrocardiogram, and the cardiac enzymes. The 12-lead ECG, which may be obtained even before the patient reaches the hospital, has reasonable sensitivity and specificity and can hasten the delivery of specific treatments, such as thrombolytics. Serial measurements of biochemical markers increase the sensitivity.4
If these studies do not clearly diagnose or rule out infarction, the patient is said to have “nonspecific” chest pain, and further imaging studies may be indicated. One possibility is echocardiography. Another is radionuclide scanning, which has excellent sensitivity in detecting early infarction; a normal resting perfusion study has a negative predictive value for myocardial infarction exceeding 99%.5 Moreover, the study performed in conjunction with echocardiography, perhaps with harmonic imaging and myocardial contrast agents, is accurate in women, as other studies sometimes are not.6 In a randomized trial, it was possible to reduce hospitalizations and conserve costs using radionuclide scans.5 New radiopharmaceuticals that identify infarction sites may make the study even more useful in the future.5 Exercise-gated SPECT is another approach. In one series of 306 patients presenting to the emergency department with chest pain, a team from Careggi General Hospital in Florence, Italy, found that the scans were effective for early triage.7
Novelline expects that many of these studies will be unnecessary when the greater resolution and speed of the 16-slice CT scanners become more widely available.
“We are expecting to do CT coronary angiography on such a scanner,” he reports. “Then, someone whose clinical examination and ECG are strongly suggestive of acute infarction will go right to cardiac angiography for confirmation of the diagnosis and treatment with dilation with or without a stent, but if there is a question whether the patient has had an infarction, you could screen them with CTA and send to the catheterization laboratory only those patients in whom it will truly help. Certainly, CT would be faster and easier to arrange than a radionuclide study, but we need comparative trials to see which is truly better.” He expects such data to be available within 2 years.
Another cause of chest pain necessitating prompt treatment is pulmonary embolism. The traditional study, a ventilationperfusion scan, has well-known drawbacks. Indeed, no element of the patient’s history, physical finding, or diagnostic study is absolutely reliable in ruling out the potentially fatal condition.8 Radiologists at the Department of Emergency Medicine at the Carolinas Medical Center in Charlotte, NC, have proposed a “rule-out” protocol in which the pretest probability of an embolus, as judged by a negative D-dimer assay or a D-dimer assay with a normal alveolar dead-space measurement, is used to decide for or against imaging.9 Their analysis suggests that the rule-out protocol would enhance emergency department throughput while making more efficient use of imaging. Massachusetts General Hospital has a specialized CT protocol for pulmonary angiography. Still another approach is CT venography of the legs and abdominal veins to search for source clots together with CT angiography of the pulmonary arteries.10
Other CT Uses
Flank pain. The classic study for a patient with flank pain has been an intravenous urogram to rule out stone. However, CT has “unprecedented sensitivity, specificity, and accuracy” for detecting ureteral and renal calculi and does not require contrast medium.11 Moreover, CT enables identification of causes of flank pain that are outside the urinary tract.
Imaging for stroke. In the patient with a possible stroke, confirmation of the diagnosis and determination of the type of lesion (ischemic versus hemorrhagic) is a matter of urgency if any effective therapy is to be given in time. Magnetic resonance imaging with diffusion and perfusion sequences has been a popular method, as it not only localizes the stroke but indicates how much tissue is dead and how much is merely stunned and thus potentially salvageable. However, interest in CT is returning with the introduction of functional (perfusion) protocols with or without angiography.12 These tools have the advantages of being faster and more widely available in the emergency setting than is MRI. The National Institute of Neurological Disorders and Stroke (NINDS) has issued guidelines for initial evaluation and care of these patients that include evaluation by a specialist within 10 minutes of admission, completion of CT scan interpretation within 45 minutes, and administration of tissue plasminogen activator within an hour in patients with ischemic stroke. Researchers at the Institute for Technology Assessment and the Department of Radiology at Massachusetts General Hospital have demonstrated that adherence to the NINDS guidelines increases the proportion of treatment candidates from 1.4% to 3.0% at a per-patient cost of approximately $434.13
Because of the tremendous complexity of the possible abnormalities, the greatest challenge to the diagnostic acumen of the emergency department staff is the trauma patient. Here again, helical CTwith its ability to survey the whole patient quickly and the ability to reformat the images to answer a particular question after the patient has left the radiology suiteis often critical. In a series of more than 1,900 patients with blunt trauma seen at Massachusetts General Hospital, reformatted CT scans obtained with a visceral protocol were 97% sensitive for thoracic spinal fractures and 95% sensitive for lumbar spine fractures. The traditional imaging method, namely conventional radiography, had specificities of only 62% and 86%, respectively.14 Plain films therefore are not required.
However, many other modalities may be called on for the trauma victim. At the 2002 Annual Meeting of the Radiological Society of North America, Chicago, an entire session was devoted to the applications of the readily available and radiation-free ultrasonography scanner for the evaluation of the patient with blunt trauma. A team from the University of California at San Diego demonstrated that patients who have sustained blunt abdominal trauma and do not have rib, lumbar spine, or pelvic fractures or hematuria and have a negative ultrasound, do not require further abdominal imaging, although additional studies are the (more expensive) policy in some departments.15 Investigators from Naples, Italy, described the use of ultrasonography as the “preferred screening method” for patients with blunt abdominal trauma.16 The use of contrast medium may further improve the ability to assess the liver, spleen, and kidney.17
Magnetic resonance imaging of the brain has utility in evaluating trauma cases.18,19 For example, the team at Massachusetts General Hospital has tested a cerebral fractional anisotropy score (C-FAST) to assess injury to the white matter of the brain, finding a good correlation between the score and length of the ICU and hospital stay, death, and discharge to rehabilitation.19
The Payoff for the Hospital
For a busy emergency department, there is a clear payoff for this enormous investment in staff and equipment. “Rapid diagnosis of acute medical and surgical problems, both trauma and nontrauma, will quickly identify those patients who need urgent surgery or hospital admission, decrease morbidity and mortality, and save health care dollars because it averts unnecessary admissions and surgery,” Novelline asserts. “For example, 10 or 15 years ago, all 60-year-old men and women presenting with left lower-quadrant pain would be presumptively considered to have diverticulitis and be admitted for diagnostic tests in the morning or the next day. Now, the same patient can have a CT scan in the emergency department. If it rules out diverticulitis, the patient probably can go home. If it shows perforation, the patient goes right to the operating room, so there are fewer complications, and if it shows diverticulitis without perforation, the patient is admitted for administration of antibiotics. So triage is available with one quick scan.” Massachusetts General Hospital has found a similar benefit from limited helical CT with colonic contrast in children with right lower-quadrant pain.20 In a series of 199 patients, there were 64 true-positive scans and 128 true-negative scans, and the negative appendectomy rate was 9%. In 62 patients without appendicitis, the CT study provided an alternative diagnosis. In another series of patients with nontraumatic abdominal pain, CT showed a high rate of error in the presumptive diagnoses and often suggested unsuspected problems, thus quickly directing the patient toward appropriate treatment.21 The diagnosis made by CT proved to be in error in only 6% of the patients who required hospital admission.
As the popular press focuses increasingly on the tendency of the general public to turn to emergency departments for routine medical care, it is easy to lose sight of the severely ill and injured patients to whom such departments mean the difference between living and dying. As in so many other areas of medicine in the last few years, these greater challenges have created another specialty: emergency radiology.
Judith Gunn Bronson, MS, is a contributing writer for Decisions in Axis Imaging News.
- Philipp MO, Kubin K, Hormann M, Metz VM. Radiological emergency room management with emphasis on multidetector-row CT. Eur J Radiol. 2003;48:2â€”4.
- Philipp MO, Kubin K, Mang T, Hormann M, Metz VM. Three-dimensional volume rendering of multidetector-row CT data: applicable for emergency radiology. Eur J Radiol. 2003;48:33â€”38.
- Obrador D, Santalo M. Evaluation of patients with suspected acute coronary syndromes in the emergency department. Nucl Med Commun. 2003;24:1041â€”1048.
- Pope JH, Selker HP. Diagnosis of acute cardiac ischemia. Emerg Med Clin North Am. 2003;21:27â€”59.
- Abbott BG, Wackers FJ. Use of radionuclide imaging in acute coronary syndromes. Curr Cardiol Rep. 2003;5:25â€”31.
- Mobasseri S, Hendel RC. Cardiac imaging in women: use of radionuclide myocardial perfusion imaging and echocardiography for acute chest pain. Cardiol Rev. 2002;10:149â€”160.
- Conti A, Zanobetti M, Grifoni S, et al. Implementation of myocardial perfusion imaging in the early triage of patients with suspected acute coronary syndromes. Nucl Med Commun. 2003;24:1055â€”1060.
- Sadosty AT, Boie ET, Stead LG. Pulmonary embolism. Emerg Med Clin North Am. 2003;21:363â€”384.
- Kline JA, Wells PS. Methodology for a rapid protocol to rule out pulmonary embolism in the emergency department. Ann Emerg Med. 2003;42:266â€”275.
- Ghaye B, Dondelinger RF. Non-traumatic thoracic emergencies: CT venography in an integrated diagnostic strategy of acute pulmonary embolism and venous thrombosis. Eur Radiol. 2002;12:
- Tamm EP, Silverman PM, Shuman WP. Evaluation of the patient with flank pain and possible ureteral calculus. Radiology. 2003;228:319â€”329.
- Wintermark M, Bogousslavsky J. Imaging of acute ischemic brain injury: the return of computed tomography. Curr Opin Neurol. 2003;16:59â€”63.
- Stahl JE, Furie KL, Gleason S, Gazelle GS. Stroke: effect of implementing an evaluation and treatment protocol compliant with NINDS recommendations. Radiology. 2003;228:659â€”668.
- Sheridan R, Peralta R, Rhea J, Ptak T, Novelline R. Reformatted visceral protocol helical computed tomographic scanning allows conventional radiographs of the thoracic and lumbar spine to be eliminated in the evaluation of blunt trauma patients. J Trauma. 2003;55:665â€”669.
- Sirlin CB, Brown MA, Andrade-Barreto OA, Hoyt DB, Fortlage DA, Cascia G. Screening ultrasound in blunt abdominal trauma: when is negative ultrasound not enough? [abstract 470]. Radiology. 2002;225(P):357.
- Pinto F, Scaglione M, Pinto A, Ritzo A, Bignardi E, Romano L. Ultrasound in the triage of patients after blunt abdominal trauma: our experience in 3500 consecutive patients [abstract 473]. Radiology. 2002;225(P):358.
- Martegani A, Cosgrove DO, Del Favero C, Alani L, Harvey CJ. Contrast enhanced abdominal ultrasound in trauma using SonoVueTM [abstract 474]. Radiology. 2002;225(P):358.
- Imhof H, Fuchsjager M. Traumatic injuries: imaging of spinal injuries. Eur Radiol. 2002;12:1262â€”1272.
- Ptak T, Sheridan RL, Rhea JT, et al. Cerebral fractional anisotropy score in trauma patients: a new indicator of white matter injury after trauma. AJR Am J Roentgenol. 2003;181:1401â€”1407.
- Mullins ME, Kircher MF, Ryan DP, et al. Evaluation of suspected appendicitis in children using limited helical CT and colonic contrast material. AJR Am J Roentgenol. 2001;176:37â€”41.
- Brown DF, Fischer RH, Novelline RA, Kim J, Nagurney JT. The role of abdominal computed tomography scanning in patients with non-traumatic abdominal symptoms. Eur J Emerg Med. 2002; 9:330â€”333.