For Patients in Pain, Scans May Not Offer a Diagnosis
BY LAUREN UZDIENSKI, DECEMBER 9, 2008
Imaging technology may be an ubiquitous part of modern medical practice, but a New York Times article suggests that the frequency and sensitivity of MRI and CT scans may reveal musculoskeletal abnormalities without showing what causes a patient's pain.
These concerns about scans as a tool for diagnosis are particularly relevant to patients with knee and back pain. The Times reports that a torn meniscus is a common finding in patients with arthritis, but a study published recently in the New England Journal of Medicine found that meniscal tears were just as common in people with arthritis who were not in pain as they were in people with arthritis who did have pain. Thus surgery on the meniscus, the most common procedure orthopedic surgeons perform, may not help the pain.
Similarly, a spine study found that, among asymptomatic patients, 20-25% will have a herniated disc show up on a scan. Further, when patients reporting back pain underwent a scan at the time of the initial complaint and again six weeks later, 13% of patients who had a herniated disc at the first visit had it disappear by the second scan. However, their pain did not always go away. In some cases, the size of the herniation had increased by the second scan when patients reported feeling better.
Among the shortcomings of highly sensitive imaging and the frequency with which scans are run (the article points out that many surgeons have their own scanners, and CMS pays about $750 to $950 for an MRI) is the fact that there is little data on what a normal knee or spine looks like. An abnormality could be interpreted as causing pain, and that could lead to a surgery that may not resolve the complaint. To mitigate some of this, the article outlined a proposal whereby radiologists could put scan findings into context. For example, if a deteriorated disc is identified, the radiologist could note in his report the percentage of patients with the condition that have no pain according to the literature, which would highlight the fact that the finding could be unrelated to the patient's complaint.
The physicians interviewed by the Times suggested that a scan may not be that helpful in assessing the best treatment plan for a patient, raising the questions of why the scan would be performed at all or recommending that they only be performed to provide a guide for surgery. However, many patients come to a surgeon expecting a scan, or, in a case highlighted by the article, convinced that a particular procedure (e.g., meniscectomy) will relieve their pain. Managing patient expectations would be one of the challenges of determining who would benefit from a scan.
Ideally, the questions raised by highly sensitive scans would be resolved by more baseline data--screening asymptomatic people to better understand common abnormalities and how they relate to pain. However, a main obstacle to collecting these scans is that many people wouldn't look forward to needless time in an MRI chamber or to the radiation exposure inherent in a CT scan. In the absence of this data, the Times suggests that, for a lot of patients, a scan may play only a small role in diagnosis. This reality also highlights the need for more refined diagnostics to help identify the elusive causes of pain.
