In reading this morning’s headlines, several caught my eye: “Medicare will pay for Alzheimer’s scan” and “US to pay for brain scans to diagnose Alzheimer’s”. The funny thing is, the diagnosis of Alzheimer’s dementia is not based on any imaging nor laboratory tests. 87 percent of cases can be detected clinically in the absence of blood tests or scans. So, what is the role of imaging in Alzheimer’s dementia?
The answer is not clear and there is no consensus. A number of guidelines suggest imaging in specific cases:
The use of neuroimaging in patients with dementia is controversial. A number of guidelines on the clinical evaluation of dementia have been published, many of which do not recommend imaging studies routinely, but include clinical prediction rules to identify patients who might have reversible causes of dementia that can be diagnosed with imaging studies. The prediction rules vary, including factors such as young age (<60), focal signs, short duration of symptoms (<2 years), among others. However, the sensitivity and specificity of these prediction rules is low. . . Most patients do not require head CT scans or magnetic resonance imaging (MRI) to make a diagnosis of Alzheimer’s disease.
Medicare however is referring to a specific type of neuroimaging – PET scanning. The data suggests that PET scanning has a sensitivity in diagnosing progressive dementia and Alzheimer’s dementia of 93 and 94 percent respectively. However, the available treatments of dementia are only of limited benefit, and it is unclear that earlier diagnosis leads to an improved clinical outcome. Costing around $2000 per test, each machine costs about $2 million for a hospital to install.
It is worth noting that there is opposition to this movement – since there is not enough evidence to support such general screening:
Several doctors and other medical professionals wrote to oppose coverage, fearing that once PET was approved for a narrow use, it would be marketed more broadly as a diagnostic test for Alzheimer’s generally. The disease affects more than 4 million Americans today, and the number is expected to multiply rapidly as the population ages.
Recognizing that fact, Medicare coverage only refers to a very narrow range of potential cases, suggesting this is not for general use, nor for routine cases of early dementia:
. . . in the absence of convincing evidence that PET scans can, by themselves, tell whether a person has Alzheimer’s, the agency settled on a much narrower application. It will reimburse for the brain scans only for patients whose Alzheimer’s symptoms are not typical and who doctors believe may instead have one of several rare brain diseases known collectively as “fronto-temporal dementia. ”
Those conditions — the best known is Pick’s disease — generally occur at a younger age than Alzheimer’s. They tend to first become apparent as behavioral changes and language difficulties rather than memory loss. Nevertheless, they can mimic Alzheimer’s enough that distinguishing them can be difficult, and that can be important because these dementias tend to progress much more rapidly.
Bottom line: more studies are needed before routine imaging (PET scanning in particular) can be recommended for general use in dementia screening.
