Just because someone has Alzheimer's, doesn't mean they can't perceive pain. They may be unable to tell people about it, though. Different tests may be needed to help find out how bad the pain is in such patients.
Old age is associated with the development of a number of painful conditions - arthritis, cancer, neuropathies, and so on. People with dementia of any type are not immune to these conditions, but they may have trouble in describing their pain properly. This means that sometimes it goes untreated. Commonly, two types of pain assessments are made: 'sensory-discriminatory', meaning the presence and intensity of the pain, and 'motivational-affective', as shown by the patient's behavior, facial expressions, etc. Clearly, it can be difficult to use the first type of assessment in severely demented patients.
Scientists in the Netherlands have recently reviewed the whole question of pain in dementia patients, and published their findings in the British Medical Journal. Here's a summary of their report.
The information used for this review was obtained from published reports in medical journals, as well as the scientists' personal files. It soon became apparent that pain in older patients with impaired mental functioning is under-treated. Fewer analgesics are prescribed in such patients; this is particularly the case in patients with a fractured hip.
Patients with Alzheimer's disease receive fewer analgesics than those with vascular dementia. This may be because impairment in verbal communication is more common in Alzheimer's than in vascular dementia.
In clinical practice, the bulk of available pain assessment 'tests' require that the patient can communicate clearly, by speaking, writing, or pointing to a scale of some sort (e.g. choosing a smiley-face, a point on a line labeled from 1 to 10, or checking a box). Such tests allow pain intensity to be rated, and can be used to compare the effectiveness of pain therapy. Unfortunately, only mildly demented patients can use them.
Observational scales have more application for demented patients. A nurse or caregiver scores or checks such behaviors as rapid breathing, frowning, grimacing, bracing the body, and moaning or vocalization. Unfortunately, these 'motivational-affective' assessments are sometimes part of the dementia rather than representing perceived pain; it may require considerable skill and experience to differentiate between the two. The example given in the review is that of assessment of "absence of a relaxed body posture"; while this can be indicative of pain, it can also occur in some Alzheimer's patients with Parkinsonian symptoms.
It may be possible to get some idea of a patient's pain by increases in blood pressure and heart rate, but these are not very sensitive measures, and they don't respond as strongly in demented patients as in others. Moreover, it's been shown that the worse the dementia, the less the cardiovascular response to pain.
The first link below gives more information on actual test systems used for dementia patients.
The two types of pain assessed are, in fact, processed in different areas of the brain. The thalamus nuclei, which process the motivational-affective perception of pain, are severely affected in Alzheimer's disease. In contrast, sensory-discriminatory pain perception is mainly a function of the sensory areas, which are relatively well-preserved in Alzheimer's. This means that while Alzheimer patients can perceive pain (although they may experience its intensity to a lesser extent), they may have difficulty in placing the pain in context (e.g. exact location, timing) and thus exhibit atypical behavioral responses.
Patients with vascular dementia, where there is atrophy of the white matter of the brain, are more sensitive to motivational-affective aspects of pain. Other types of dementia have not been studied enough for one to make statements about the type of pain most readily perceived. Unfortunately, most clinical studies have not distinguished between the pain responses in the different types of dementia.
People with Parkinson's disease and multiple sclerosis are at high risk of developing mental impairment. Often pain is a prominent symptom at a stage before the mental status declines. It would be important to define the type of pain perception in such patients, and then follow them into their dementia, should this develop, to see how pain assessment pain is best determined in these conditions. It's known that two common forms of treatment - levodopa for Parkinson's and interferon-beta for multiple sclerosis - can increase pain, in some cases; being able to assess pain accurately would be useful in determining how best to continue treatment.
This review pointed to the need to try to assess pain in dementia patients, to avoid unnecessary suffering. It seems the lack of good testing is recognized, and will probably be addressed in future studies. There is no mention of the best ways of treating such pain, but the general principles apply - pain should be treated adequately to ensure that it is gone, or is at least tolerable (see the second link below). What's important is to recognize that someone who cannot communicate normally may still be experiencing considerable pain.