| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Institut de Diagnòstic per la Imatge Bellvitge 08907 Barcelona, Spain
I read with interest the article by Dr. Berlin on "Malpractice Issues in Radiology" in the June 2000 issue of AJR [1]. The first case concerning a 68-year-old woman with "clinical diagnosis of pancreatitis" raises an important point that deserves further attention. According to the interpreting radiologist, the CT scan depicts a pseudonodular appearance of the head of the pancreas, but there is not a definitive mass effect. The CT scan is apparently without IV contrast material. I agree that the radiology report is not strong enough to suggest pancreatic cancer, but I think that a positive diagnosis of pancreatic cancer is not possible with the images provided. In my opinion, the issue is not whether the radiology report was ambiguous or not, but rather why was the examination performed in that particular woman? The clinical background is of the utmost importance in designing an imaging study. First of all, what does "clinical diagnosis of pancreatitis" mean? Was the patient referring acute onset abdominal pain with elevated serum amylase levels? Had the patient any risk factors for pancreatitis, such as cholelithiasis or alcohol ingestion? What specifically was the indication for ordering CT: to detect local complications of acute pancreatitis, to establish the diagnosis in a case of acute abdomen, or to rule out pancreatic cancer in a case of chronic pancreatitis? These are extremely important questions because the radiologist requires this information to tailor an adequate study. This consideration leads me to a second issue: the quality of the radiologic study is not an independent variable. It depends on the clinical situation of the patient. In some instances, the highest standard of quality for CT would be not performing the examination because it is not the best technique in that setting (i.e., to rule out adenoma of the pituitary gland). In other clinical situations the highest standard of quality for CT will be an unenhanced helical study (i.e., for renal colic to rule out ureterolithiasis), and eventually a biphasic enhanced CT would be the highest standard in some specific clinical situations (i.e., in liver cirrhosis to rule out hepatocellular carcinoma). However, if there is a low suspicion of abdominal disease, as seems to be the case in this patient, unenhanced abdominal CT seems reasonable. Conversely, if there is a clinical suspicion of pancreatic cancer, a dedicated biphasic helical CT with thin collimation and a high dose of contrast material should have been performed. If that clinical information was available, the radiologist would be responsible for missing a pancreatic lesion with unenhanced CT. However, in view of the clinical information provided, standard abdominal CT is probably an adequate option although a small pancreatic cancer will easily be missed with this technique. I think we all agree that unenhanced abdominal CT is not a screening procedure in the diagnosis of pancreatic cancer. A dedicated biphasic thin-collimation study of the pancreas is not performed routinely if there is not a strong suspicion of pancreatic cancer. Additionally detection of incidental pancreatic cancer (i.e., asymptomatic) is extremely rare because these tumors usually have a rapid growth that leads to clinical symptoms. In the patient described by Dr. Berlin, the clinical information provided was not sufficient to warrant a dedicated pancreatic examination. The explanation is two-fold: if the patient had no symptoms that suggested pancreatic cancer, a dedicated pancreatic study was not indicated. However, if the patient actually had symptoms that suggested pancreatic cancer, the referring physician did not mention it, and, therefore, an adequate study was not possible. In both scenarios the radiologist acted correctly. Therefore when assessing a possible pitfall in radiology, it would be important to review the clinical information provided to the radiologist. In most academic hospitals, this problem does not exist because there is an adequate interaction between the radiologist and clinical colleagues. However in outpatient facilities, the information is scant if not missing for a number of patients, and an optimal radiologic examination may not be performed on the basis of the information provided. In these patients referring physicians may have a substantial part of the responsibility in potential radiologic pitfalls caused by a lack of clinically important information.
References
Rush North Shore Medical Center Skokie, IL 60076 Rush Medical College Chicago, IL 60612
I thank Dr. Valls for his thoughtful and well-reasoned letter. I certainly agree that a radiologist can optimize radiologic examinations if he knows the patient's history and clinical findings. Knowledge of this information will also result in more specific recommendations for a further diagnostic workup [1] and improved detection of radiologically abnormal findings [2, 3]. Furthermore, I have no doubt that knowledge of the patient's symptoms and clinical signs would, in turn, allow radiologists to be more definite and precise when writing radiology reports. However, even with adequate clinical information and optimized radiologic examinations, some radiologists fail to convey to the referring physicians a truly accurate representation of the radiologic findings, the radiologist's opinion of the meaning of the findings, and the radiologist's recommendations for further action. As I stated in the article [4], radiologists should strive in every circumstance to render interpretations of radiologic studies in a manner that is meaningful to the referring physician. Radiologists should make every effort to refrain from using language in their radiology reports that lulls the referring physicians into soporific conduct or inaction.
It is true, as Dr. Valls writes, that when ordering radiologic examinations, referring physicians have a responsibility to provide clinically important information to radiologists. It is also true that in some cases referring physicians conduct themselves in a lackadaisical if not a sub-standard manner. Nonetheless, the law holds radiologists responsible for lapses in their own conduct, regardless of any liability that might be imposed on other physicians [5].
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
