Incidentalomas — Clinical Correlation and Translational Science Required
http://www.100md.com
《新英格兰医药杂志》
It should have been a moment of huge relief. The biopsy specimen of the temporal artery was positive. We had answered the riddle of the patient's year-long fatigue, limb pain, and turbine-like noises in his head. He had giant-cell arteritis — not cancer, as he had feared. His disease would respond quickly to prednisone, and we had made the diagnosis before he had lost vision. The biopsy findings constituted a eureka moment.1
There was, however, a fly in the ointment. The report of the computed tomographic scan of the patient's abdomen, performed in search of an explanation for the symptoms and high erythrocyte sedimentation rate, read a "5-mm mass in the posterior right kidney. Cannot exclude renal-cell carcinoma" (see Figure 1). My excitement at the diagnosis of giant-cell arteritis vanished as I pondered how to break this news. My clinical sense — and Ockham's razor,2 the law of averages, and all other principles of diagnostic parsimony — dictated that this renal mass should neither cause concern nor distract from the tasks at hand: helping the patient understand his disease and beginning to treat it effectively. The mass, I was sure, was an "incidentaloma." But informed of the mass's existence, the patient replied, "An `incidentaloma'? You mean, an incidental `Oh, my!'"
Figure 1. Incidentally Discovered Mass in the Posterior Right Kidney, on CT Scan of the Abdomen (Arrow).
Urgent telephone consultations with Urology and Radiology ensued. The patient, who lived out of town, delayed his return home for two days (incurring substantial expense) to accommodate the magnetic resonance imaging (MRI) study that was recommended. The MRI report, unfortunately, was not completely reassuring. True, it concluded with the statement "Consistent with a proteinaceous cyst." But it also said, "Recommend follow-up imaging in six months. Clinical correlation required." Thus, an incidentaloma ruined the patient's peace of mind and diverted our focus from the otherwise clear path to health.
How many times each day does an incidental finding lead to other tests, additional inconclusive results, and still further testing? What is the effect of incidentalomas on the costs of health care, physicians' concern about being sued, and patients' fears of serious illness? The experience of the past several decades indicates that incidental findings — phenomena of improved imaging capabilities — frequently trigger the ordering of unnecessary tests and cause untold anxiety. Although the medical literature on incidentalomas has grown enormously during the past two decades, we have little quantification of the economic impact of incidentalomas and inadequate plans to address the problem.
One study of ultrasonography indicates that simple renal cysts occur in 12 percent of persons between 50 and 70 years of age.3 Incidentalomas also commonly occur in the adrenal glands, liver, pancreas, thyroid gland, and lungs — in short, in any part of the body that can be imaged (see Figure 2). The most challenging question, regardless of the organ in which incidentalomas are found, is what to do with the small ones — the renal mass smaller than 1.5 cm, the adrenal lesion smaller than 4 cm, and the pituitary abnormality smaller than 1.0 cm. My patient's 5-mm renal mass challenges the capabilities of any radiologic test ordered in the hope of definitive reassurance. With a growing number of incidentalomas, what to tell the patient remains unclear.
Figure 2. Adenoma Found on CT Performed to Rule Out Thymoma in a Patient with Myasthenia Gravis.
The scan showed no thymoma but revealed a lesion on the right adrenal gland, leading to four years of imaging, biochemical, and endocrine workups — and an eventual conclusion that the growth was benign.
To some extent, what patients should know depends on the organ of concern. Autopsy studies suggest that adrenal masses of 2 mm to 4 cm in diameter are present in approximately 10 percent of persons.4 In the case of masses larger than 4 cm, the likelihood of adrenal carcinoma is sufficiently high that the consensus recommendation is surgical removal. Current imaging techniques, however, can easily detect much smaller adrenal lesions. The finding of an adrenal incidentaloma leads not only to additional radiologic testing, but also to biochemical assays to rule out pheochromocytomas and hypersecreting adenomas. Because such testing may not definitively rule out cancer, some patients undergo needle biopsies. And because the examination of needle-biopsy specimens often does not distinguish reliably between benign adrenal masses and small adenocarcinomas, the frequent upshot is adrenalectomy. This zealous pursuit is validated on occasion, of course, when the pathologist finds an adrenal carcinoma within the surgical specimen. Yet only 1 adrenal mass in 4000 is malignant, and the likelihood that an incidentally detected adrenal mass is cancerous (particularly if it is less than 4 cm in diameter) is substantially lower. Vastly more common are expensive workups and equivocal interpretations.
Both the discovery of incidentalomas and their associated costs are likely to grow as new forms of technology are developed. The National Institutes of Health and industry have expanded funding for research on imaging methods. Some of this investment will lead to improved abilities to distinguish worrisome lesions from harmless ones, but it is naive to believe that new techniques will bring only more clarity. Imaging initiatives are also likely to inspire broad adoption of new techniques before their full implications are understood.
The issue of what to do about incidentalomas is not simple. Any discussion must acknowledge the thousands of lawsuits in the United States each year that result from a physician's failure to act on some early evidence of a problem that ultimately leads to an adverse outcome. Guided in part by fears of lawsuits, clinicians tend to order additional tests. All too often, these tests do not result in clinical clarity, diagnostic certainty, or patient satisfaction. Caught in the same bind, radiologists are inclined to suggest additional tests and to dictate in their reports "Cannot exclude . . .," "Need to rule out . . .," "Consistent with . . .," and "Clinical correlation required," implying that a sinister cause of the finding might be discovered, if only the problem were investigated with sufficient vigor. This cycle of disclaimers and protectionist test-ordering passes the buck between clinician and radiologist, adds to the financial demands on payers, heightens patients' worry, and yet offers little insight into the appropriate level of concern.
Since we cannot "unknow" the fact of an incidental finding, we need information on reasoned approaches to the findings once they occur. Defining rational approaches to incidentalomas would have important implications for society's views of medicine: Do practitioners use expensive resources responsibly, apply evidence rigorously, and validate the benefits of new technology? In the coming years, imaging techniques will grow more sensitive and, likely, less specific in many cases. Imaging can help us look, but both clinical correlation and translational science are required before any technique can help us see with optimal focus.
Source Information
Dr. Stone is an associate professor of medicine at the Johns Hopkins University School of Medicine, Baltimore.
References
Hellmann DB. Eurekapenia: a disease of medical residency training programs? Pharos Alpha Omega Alpha Honor Med Soc 2003;66:24-26.
Wudka J. What is Ockham's razor? (Accessed June 8, 2006, at http://phyun5.ucr.edu/~wudka/Physics7/Notes_www/node10.html.)
Ravine D, Gibson RN, Donlan J, Sheffield LJ. An ultrasound renal cyst prevalence survey: specificity data for inherited renal cystic diseases. Am J Kidney Dis 1993;22:803-807.
Hedeland H, ?stberg G, H?kfelt B. On the prevalence of adrenocortical adenomas in an autopsy material in relation to hypertension and diabetes. Acta Med Scand 1968;184:211-214.(John H. Stone, M.D., M.P.)
There was, however, a fly in the ointment. The report of the computed tomographic scan of the patient's abdomen, performed in search of an explanation for the symptoms and high erythrocyte sedimentation rate, read a "5-mm mass in the posterior right kidney. Cannot exclude renal-cell carcinoma" (see Figure 1). My excitement at the diagnosis of giant-cell arteritis vanished as I pondered how to break this news. My clinical sense — and Ockham's razor,2 the law of averages, and all other principles of diagnostic parsimony — dictated that this renal mass should neither cause concern nor distract from the tasks at hand: helping the patient understand his disease and beginning to treat it effectively. The mass, I was sure, was an "incidentaloma." But informed of the mass's existence, the patient replied, "An `incidentaloma'? You mean, an incidental `Oh, my!'"
Figure 1. Incidentally Discovered Mass in the Posterior Right Kidney, on CT Scan of the Abdomen (Arrow).
Urgent telephone consultations with Urology and Radiology ensued. The patient, who lived out of town, delayed his return home for two days (incurring substantial expense) to accommodate the magnetic resonance imaging (MRI) study that was recommended. The MRI report, unfortunately, was not completely reassuring. True, it concluded with the statement "Consistent with a proteinaceous cyst." But it also said, "Recommend follow-up imaging in six months. Clinical correlation required." Thus, an incidentaloma ruined the patient's peace of mind and diverted our focus from the otherwise clear path to health.
How many times each day does an incidental finding lead to other tests, additional inconclusive results, and still further testing? What is the effect of incidentalomas on the costs of health care, physicians' concern about being sued, and patients' fears of serious illness? The experience of the past several decades indicates that incidental findings — phenomena of improved imaging capabilities — frequently trigger the ordering of unnecessary tests and cause untold anxiety. Although the medical literature on incidentalomas has grown enormously during the past two decades, we have little quantification of the economic impact of incidentalomas and inadequate plans to address the problem.
One study of ultrasonography indicates that simple renal cysts occur in 12 percent of persons between 50 and 70 years of age.3 Incidentalomas also commonly occur in the adrenal glands, liver, pancreas, thyroid gland, and lungs — in short, in any part of the body that can be imaged (see Figure 2). The most challenging question, regardless of the organ in which incidentalomas are found, is what to do with the small ones — the renal mass smaller than 1.5 cm, the adrenal lesion smaller than 4 cm, and the pituitary abnormality smaller than 1.0 cm. My patient's 5-mm renal mass challenges the capabilities of any radiologic test ordered in the hope of definitive reassurance. With a growing number of incidentalomas, what to tell the patient remains unclear.
Figure 2. Adenoma Found on CT Performed to Rule Out Thymoma in a Patient with Myasthenia Gravis.
The scan showed no thymoma but revealed a lesion on the right adrenal gland, leading to four years of imaging, biochemical, and endocrine workups — and an eventual conclusion that the growth was benign.
To some extent, what patients should know depends on the organ of concern. Autopsy studies suggest that adrenal masses of 2 mm to 4 cm in diameter are present in approximately 10 percent of persons.4 In the case of masses larger than 4 cm, the likelihood of adrenal carcinoma is sufficiently high that the consensus recommendation is surgical removal. Current imaging techniques, however, can easily detect much smaller adrenal lesions. The finding of an adrenal incidentaloma leads not only to additional radiologic testing, but also to biochemical assays to rule out pheochromocytomas and hypersecreting adenomas. Because such testing may not definitively rule out cancer, some patients undergo needle biopsies. And because the examination of needle-biopsy specimens often does not distinguish reliably between benign adrenal masses and small adenocarcinomas, the frequent upshot is adrenalectomy. This zealous pursuit is validated on occasion, of course, when the pathologist finds an adrenal carcinoma within the surgical specimen. Yet only 1 adrenal mass in 4000 is malignant, and the likelihood that an incidentally detected adrenal mass is cancerous (particularly if it is less than 4 cm in diameter) is substantially lower. Vastly more common are expensive workups and equivocal interpretations.
Both the discovery of incidentalomas and their associated costs are likely to grow as new forms of technology are developed. The National Institutes of Health and industry have expanded funding for research on imaging methods. Some of this investment will lead to improved abilities to distinguish worrisome lesions from harmless ones, but it is naive to believe that new techniques will bring only more clarity. Imaging initiatives are also likely to inspire broad adoption of new techniques before their full implications are understood.
The issue of what to do about incidentalomas is not simple. Any discussion must acknowledge the thousands of lawsuits in the United States each year that result from a physician's failure to act on some early evidence of a problem that ultimately leads to an adverse outcome. Guided in part by fears of lawsuits, clinicians tend to order additional tests. All too often, these tests do not result in clinical clarity, diagnostic certainty, or patient satisfaction. Caught in the same bind, radiologists are inclined to suggest additional tests and to dictate in their reports "Cannot exclude . . .," "Need to rule out . . .," "Consistent with . . .," and "Clinical correlation required," implying that a sinister cause of the finding might be discovered, if only the problem were investigated with sufficient vigor. This cycle of disclaimers and protectionist test-ordering passes the buck between clinician and radiologist, adds to the financial demands on payers, heightens patients' worry, and yet offers little insight into the appropriate level of concern.
Since we cannot "unknow" the fact of an incidental finding, we need information on reasoned approaches to the findings once they occur. Defining rational approaches to incidentalomas would have important implications for society's views of medicine: Do practitioners use expensive resources responsibly, apply evidence rigorously, and validate the benefits of new technology? In the coming years, imaging techniques will grow more sensitive and, likely, less specific in many cases. Imaging can help us look, but both clinical correlation and translational science are required before any technique can help us see with optimal focus.
Source Information
Dr. Stone is an associate professor of medicine at the Johns Hopkins University School of Medicine, Baltimore.
References
Hellmann DB. Eurekapenia: a disease of medical residency training programs? Pharos Alpha Omega Alpha Honor Med Soc 2003;66:24-26.
Wudka J. What is Ockham's razor? (Accessed June 8, 2006, at http://phyun5.ucr.edu/~wudka/Physics7/Notes_www/node10.html.)
Ravine D, Gibson RN, Donlan J, Sheffield LJ. An ultrasound renal cyst prevalence survey: specificity data for inherited renal cystic diseases. Am J Kidney Dis 1993;22:803-807.
Hedeland H, ?stberg G, H?kfelt B. On the prevalence of adrenocortical adenomas in an autopsy material in relation to hypertension and diabetes. Acta Med Scand 1968;184:211-214.(John H. Stone, M.D., M.P.)