A new study published this week in Archives of Internal Medicine indicates that incidental findings frequently occur on imaging studies performed for medical research. Although only a few of these findings turn out to be clinically significant, it's important for research centers to have a plan for dealing with them.
Researchers at the Mayo Clinic in Rochester, MN, took a look at three months of research activity in 2004 to determine the number and type of incidental findings identified in its research subjects. In total, incidental findings were identified in 39.8% of the diagnostic imaging exams. Thirty-five of these generated clinical action at Mayo Clinic, leading to a medical benefit in six patients (Arch Intern Med, September 27, 2010, Vol. 170:17, pp. 1525-1532).
The Mayo Clinic benefited from well-established protocols for following up on incidental findings. But what if a research center isn't quite so well-prepared? An accompanying editorial by Bernard Lo, MD, chairman of the University of California, San Francisco's program in medical ethics, provides useful advice for dealing with unexpected incidental findings (pp. 1523-1524).
Dealing with incidental findings
A multidisciplinary research team from Mayo retrospectively analyzed 1,426 research imaging exams used in 91 different research studies performed over a three-month period, commencing January 2004. Examples of research being undertaken included the use of CT to study treatment options for osteoporosis in women, a comparison of MRI and other imaging modalities for monitoring various consequences of liver disease, and using medical imaging to track the effects of an experimental chemotherapy medication for multiple myeloma.
The majority if not all of the research participants were already patients at the Mayo Clinic, and they were recruited to participate in a research study based on their known medical conditions, principal investigator Nicholas Orme, MD,informed AuntMinnie.com. Who paid for the additional clinical studies and treatment was outside the scope of the study.
The Mayo Clinic has an established protocol requiring a staff radiologist to review all imaging studies performed exclusively for research purposes on the day they are conducted. Reports are dictated and entered into the subject's electronic medical record. If an incidental finding of potential clinical significance is discovered, the interpreting radiologist is expected to notify the patient's primary care physician by telephone.
If the patient did not have a primary care physician, the primary investigator who ordered the study would be notified, and it would be that physician's responsibility to advise the patient and, if requested, to make arrangements for additional workup of suspicious incidental findings, Orme said.
Incidental findings were identified in 567 imaging studies. CT exams of the abdomen/pelvis and thorax produced the highest percentages of incidental findings, at 61% and 55%, respectively. These most commonly were aortoiliac calcifications, renal cysts, and diverticulosis. Forty-three percent of MRI head exams also had incidental findings, as did 38.9% of x-ray exams.
The older the research subject was, the more likely the chance of identifying an incidental finding, Orme and colleagues discovered. For every year of age, there was a 4.2% increase in the odds of having an incidental finding.
A total of 35 research subjects underwent some type of clinical treatment. The most frequent incidental findings requiring further clinical action were indeterminate lung nodules in the chest and ovarian/adnexal masses in the abdomen/pelvis.
Thirty-two patients received follow-up imaging for a total of 71 additional studies, more than half of which were CT exams. One individual had a PET study, and five underwent noninvasive diagnostic tests. Six individuals underwent invasive diagnostic procedures, including biopsies, bronchoscopies, fine-needle aspiration, and flexible nasopharyngoscopy.
A total of six individuals clearly benefited from enrolling in their research projects. Conditions for which they were treated included renal cell carcinoma, small-bowel carcinoid tumor, rib osteomyelitis, sphenoid sinus Aspergillus colonization, grade 2 ependymoma, and ovarian mucinous cystadenoma.
Three patients, two of whom underwent laparoscopies and one who underwent surgery, had benign conditions. The research team could not determine whether the additional clinical treatment benefited the remaining individuals.
Recommendations
The Mayo Clinic had a clear-cut protocol. What responsibilities exist for other researchers who perform diagnostic imaging?
Not every research program has the resources for a radiologist to review the studies in a timely manner and prepare a radiology report. Lo pointed out that investigators who are not radiologists may be responsible for reviewing the studies, which, in fact, may not have been conducted with up-to-date equipment. Although it is not the responsibility of a researcher to detect previously undiagnosed conditions beyond the scope of the clinical study, the researcher should have enough clinical knowledge to identify a gross abnormality and has an ethical responsibility to take action.
After identifying an incidental finding, researchers need to assess its significance; if they're uncertain, they should obtain a professional consultation from a clinician who can make an accurate interpretation. If a finding lacks analytic and clinical validity, there may be no reason to notify the participant or his or her primary care physician. But the greater the clinical benefit, the stronger the obligation to disclose the finding, Lo wrote.
Prior to implementing a research study that involves diagnostic imaging, the research team needs to determine the likelihood of the study type showing an incidental finding. If radiologists are not involved, researchers need to determine who will read the study and the protocol for making further assessments. They also need to determine who will pay for a radiology consultation, and whether the research participant should be notified before, after, or not at all. The researchers also need to specify what kinds of findings they will disclose to the research participant.
Finally, a protocol needs to be established to determine when and how a primary care physician should be contacted, what to do if a research subject does not have a primary care physician, and who will pay for subsequent recommended clinical action if the research subject does not have health insurance coverage for additional treatment.
All of this should be spelled out in the consent agreement. Creating management plans on how to handle incidental findings before initiating the research projects will make life easier for all concerned.
By Cynthia E. Keen
AuntMinnie.com staff writer
September 29, 2010
Related Reading
Study offers strategy for managing incidental findings on fMRI, July 6, 2010
Calcium findings on abdominal CT linked to heart attack risk, March 15, 2010
Plenty of incidental extracardiac findings at screening CT, few interventions, November 16, 2009
Incidental findings are common on brain MRI: meta-analysis, August 26, 2009
Incidental findings on brain MRI common as technology advances, November 1, 2007
Copyright © 2010 AuntMinnie.com
