SPECIAL EDITION
Collaboration

An MRI report that simply states, "No acute intracranial abnormality" doesn't help Dr. Charles H. Tegeler when he's evaluating a stroke patient. Although it attests that there's no obvious anomaly such as a major hemorrhage or brain tumor, Tegeler, head of the stroke and cerebrovascular disease section at Bowman Gray School of Medicine at Wake Forest University, is frequently looking for more information. Extremely slight changes in cerebral tissue, for example, may be small vessel subcortical ischemias that explain why a patient's memory isn't what it used to be, he said.

Tegeler also becomes impatient with arterio-grams that vaguely identify stenosis in the middle cerebral artery. Does it look like an atherosclerotic lesion? A dissection? Vasculitis? And what are the implications for its etiology?

Clinicians like Tegeler frequently find imaging studies the best way to discover minute abnormalities that explain a patient's symptomatology. And while they are well acquainted with the capabilities of imaging technology, they need the special expertise of radiologists to tease out pathology from anatomy, significant findings from background noise, actuality from artifact.

It's all part of an increasingly intimate collaboration between the physicians who order imaging studies and the radiologists who perform and interpret them, and it requires constant dialogue and clear communication. In an environment in which radiologists must increasingly compete with other specialties to retain control of certain imaging procedures, these collaborations become an important way to reinforce radiology's value in the hospital setting.

Tegeler and the neuroradiologists he works with don't hesitate to call one another to discuss the clinical issues associated with a particular patient. He recalls the time the radiologists' interpretation of an MR angiography study included nothing about a narrowing in a vessel that might explain the patient's clinical picture. When Tegeler asked them to focus on that particular artery, the radiologists found a 50% stenosis.

THE CASE FOR DETAIL

Other clinicians report that what they're looking for in image interpretations are highly specific clinical conclusions about the part of the body they're interrogating. Sometimes, of course, pathology is not apparent, but the clinician wants to know if a process is benign or malignant, or whether a kidney tumor has invaded the vena cava or the renal vein, said Dr. Martin Resnick, a professor of urology at Case Western Reserve University and incoming president of the American Urological Association.

Dr. Philip Israel, director of the Breast Center in Marietta, GA, needs to be able to precisely pinpoint the location of an abnormality. A "posterior lesion in the upper breast" needs further clarification, he said. Does it fall at 1, 2, or 3 o'clock? Is it in the subareolar region or in one of a series of concentric, 1-inch rings outside the nipple area labeled zone 1, 2, or 3?

Unless radiologists have a clear understanding of a patient's history and clinical presentation, they're in the dark about referring physicians' concerns, said Dr. James Kelly, an associate professor of clinical neuroradiology at the Feinberg School of Medicine at Northwestern University. Clinicians at Northwestern Memorial Medical Center now fill out forms to explain why they are ordering an imaging test.

For Kelly, that means determining whether there is any evidence of shearing injury or a bruise on the brain as a result of mechanical trauma to the head, or whether there is only one site of stroke or a shower of ministrokes in different parts of the brain. For the radiologists Kelly works with, it means scouring cerebral tissue in detail and fashioning imaging studies that provide specific answers.

Still, the written radiology report is sometimes not enough. Referring physicians, especially surgeons, need to view images directly. A surgeon concerned about a hepatic lesion will want to learn which blood vessels course in front, which wind behind, and which lie alongside it (see accompanying story). A radiology report stating that the anterior lateral side of a lesion is located at about 30º and extends over an area of 1 cm would go right over the head of the surgeon; it's just too ethereal, said Dr. Jonathan Silverstein, director of the Center for Clinical Information at the University of Chicago. Direct communication and interpretation in that case is more effective.

"Radiologists are better at understanding images, due to their training. Looking at the images with the surgeon, the radiologist will be able to point to an area and communicate, 'Here is where I was concerned that the lesion might be too close to this vessel,'" Silverstein said.

PUSHING THE ENVELOPE

The specialists interviewed for this article are well schooled in imaging. Israel teaches a postgraduate course on image-guided breast biopsy for the American College of Surgeons. He and other surgeons at the Marietta Breast Center conduct their own diagnostic ultrasound and mammography exams.

"If I do the ultrasound, I can solve problems immediately. If it's a cyst, I can drain it. If it's a fibrous ridge, I can tell the woman there's nothing we need to treat. If it's a solid mass, I can do a core biopsy," he said.

Tegeler is medical director of the neurosonology laboratory at Wake Forest, where he and other neurologists perform carotid ultrasound and transcranial Doppler studies. And Kelly speaks knowledgeably about gradient-echo imaging that looks for traumatic changes to the brain and software packages that target disorders such as multiple sclerosis, dementia, and atrophy.

Nevertheless, specialists look to radiologists to keep pushing them along the leading edge of imaging advancements. Kelly depends on academic neuroradiologists and medical physicists to tweak imaging sequences, hardware, and software to create more informative images.

"It's incumbent on radiologists who are doing the studies and providing the services to stay on top of not only the technology but the clinical interpretations. We rely on the radiologists to make sure we have the latest and greatest imaging available so our patients get the best they can when they go for studies," Tegeler said.

FORMING PARTNERSHIPS

The growth of international radiology procedures also has expanded interdepartmental collaboration. Referring physicians are working more closely with radiologists in direct patient care. Resnick calls upon interventional radiologists to place nephrostomies to relieve obstructions in the bladder or kidney and to ablate kidney tumors, and Tegeler consults interventionalists for cerebral arteriograms, MRA, and CT angiography. Both interact with interventionalists before a procedure to clarify why it is needed and after the fact to discuss results, ensure attentive patient follow-up, and provide an informal pick-your-brain exchange.

This type of interplay carries over into joint case conferences and grand rounds presentations at Wake Forest and Northwestern Memorial. Every physician who influences a patient's care participates in a collegial review of anything that went wrong and how it can be prevented in the future. Even if a case was handled perfectly, referring physicians and radiologists discuss how they might do even better by trying something different next time, Kelly said.

More of these efforts to "close the loop" are still needed, according to Silverstein. For example, referring physicians and radiologists could compare initial film findings with the actual status of the patient in the operating room. Such exchanges could answer questions: How predictive were the radiology readings? How can we enhance the working relationship to improve patient care? How can we increase accuracy in the future?

"I'm not sure how to improve overall practice except by doing more follow-through," Silverstein said. "That would foster a relationship so we would start asking, 'How can we do the next case better?"'

Ms. Sandrick is a freelance writer in Chicago.