A (Not-so-Surprising) Lesson About Pain in Patients with Dementia

Pain is a subjective complaint. We can’t measure it as we can, for example, hematocrit, blood pressure, or blood glucose. If a patient doesn’t complain of pain, we generally assume that he or she isn’t experiencing it. This may in fact be true for patients who are able to communicate with those caring for them.
      
But what about patients who have pain but who are physically or mentally unable to tell anyone about it? This scenario is common among geriatric patients with dementia who are prone to suffer from disorders that can cause severe pain ranging from vertebral fractures to diabetic peripheral neuropathic pain to postherpetic neuralgia. 

A study by Husebo and colleagues¹ published online by the British Medical Journaladdresses the importance of recognizing and treating pain in patients with dementia. The authors sought to determine whether improving pain management could have a positive impact on patient behavior as measured by level of agitation. The study group consisted of 352 geriatric patients with dementia and marked agitation who lived in nursing homes in Norway. Patients were divided into 2 groups. A control group received usual treatment and a treatment group received medication management for pain based on ongoing medical treatment. 

The first subgroup of the treatment group, consisting of 68% of the patients, received acetaminophen for pain. Patients who were already receiving this medication were given oral morphine (2% of patients), buprenorphine transdermal patch (23%), or pregabalin (7%) based on their medical treatment.  All the drugs were administered on a fixed schedule. Patients who had difficulty swallowing were given a buprenorphine patch, which appears to explain why it was prescribed far more frequently than morphine or pregabalin. Patients who had been receiving drugs for dementia, psychotropic medications, aspirin for cardiac prophylaxis, or NSAIDs for at least 4 weeks before entering the study continued to take these medications. The protocol lasted 8 weeks.

The study found a significant reduction in agitation scores (an average of 17%) in the treatment group. There were also significant reductions in aggression in this group and of observable behavior that might indicate pain.

It’s not surprising!

Any well-performed study on geriatric pain is welcome-but I admit I’m a bit disheartened that these results might come as any type of a surprise. We’ve long known that patients with dementia are less likely to receive medications for pain than geriatric patients who are able to communicate. I co-authored what was probably the first study that specifically examined this problem back in 1993.² We found that pain in nursing home residents was poorly managed overall and that those with dementia were the least likely to have their pain addressed.  For the most part, pain medications were prescribed on a prn basis rather than on a fixed schedule. Patients who are unable to communicate and who can’t ask for medications probably didn’t get any. The benefit of prescribing analgesics on a fixed schedule, as demonstrated by the Husebo study, is what would be expected.

It is also not surprising that improving pain management can reduce agitation and aggression. On pain consults, I have frequently seen geriatric patients with dementia-many of whom had illnesses or who suffered from trauma that one might reasonably expect could cause marked pain- who were obviously in distress and who were usually in restraints. Inevitably, these patients were prescribed prn analgesics and hadn’t received any. I often asked the medical students who accompanied me on these consults whether they would show signs agitation and depression if they fell down a flight of stairs, and then had a gag put over their mouths (so they couldn’t speak and ask for medication), had oven mitts put on their hands (so they couldn’t write), and were tied in a chair.

I think the most significant finding of the Husebo study is that simply prescribing acetaminophen on a fixed schedule can have a significant impact on the lives of many patients.

In prescribing medications for geriatric patients, we are appropriately concerned about the increased risk of adverse events from changes in physiology associated with aging and the increased frequency of comorbid disorders. We worry about prescribing drugs such as opioids that can be associated with major adverse events, even in relatively healthy younger patients. If fixed-schedule acetaminophen alone can have such a major effect in so many patients, there is no reason this approach should not be used far more frequently than it is.

One criticism of the Husebo study: the authors do not mention whether there was a reduction in the use of antipsychotic medications, which they note might be a major benefit of improved pain treatment, or diminished use of non-study analgesics by those in the treatment group as one might expect. 

A final note: it would have been interesting if the authors had used a serotonin-norepinephrine reuptake inhibitor instead of pregabalin (Lyrica) in their study. Depression is common among geriatric patients--including those with dementia. We might have had insights as to whether a drug such as duloxetine (Cymbalta), which is FDA approved both as an analgesic and an antidepressant, would have even more of an effect than pregabalin, which provides similar analgesia but has a minimal if any antidepressant effect.
 
References:
1. Husebo BS, Ballard V, Sandvik R. Efficacy of treating pain to reduce behavioural disturbances in residents of nursing homes with dementia: a cluster randomized clinical trial. BMJ. 2011;343:d4065 doi: 10.1136/bmj.d4065  http://www.bmj.com/content/343/bmj.d4065.full
2. Sengstaken EA, King SA. The problems of pain and its detection among getiatric nursing homes residents. J Am Geriatr Soc. 1993:41:541-544