Parkinson disease with tremor, rigidity, and loss of postural control can be difficult to diagnose because the symptoms overlap with other parkinsonian syndromes. Some studies have found that 5 to 10 percent of patients with PD are misdiagnosed. Conversely, 20 percent of patients diagnosed with PD were found, on autopsy, to have some other disorder. The disease is also difficult to treat. Even patients taking levodopa, the most effective therapy, obtain only intermittent relief and have to deal with adverse effects of medication.
In an effort to clarify diagnostic techniques and treatment options, the AAN Quality Standards Subcommittee surveyed the literature and created four practice parameters on diagnosis, treatment, neuroprotective strategies, and the psychological problems that may accompany PD. The guidelines will appear in the April 11 edition of Neurology.
The effort is the third of five broad categories the AAN has committed to investigate. Practice parameters already have been created for epilepsy and dementia. Multiple sclerosis and stroke will be studied in the years ahead.
The goal is to provide the best evidence-based approaches to these disorders. After reviewing and grading the literature, however, members of the AAN Quality Standards Subcommittees of the AAN often come away longing for stronger evidence.
“Once you do one of these guidelines, you come to the conclusion that there's a paucity of good scientific evidence to guide care, and Parkinson disease was not an exception,” said Gary Gronseth, MD, Vice Chair and Associate Professor of Neurology at the University of Kansas, one of the authors of the practice parameters. “There are studies that are valuable and can guide care, but after you've finished this process, you still don't know how to answer many of the clinical questions you've asked. But the process, in and of itself, is valuable because it can help direct future research by identifying the gaps.”
The practice parameters are just that – parameters, or guidelines, he said.
“Evidence-based medicine is not cookbook medicine,” said guidelines co-author William J. Weiner, MD, Professor of Neurology and Chair of the Neurology Department and Director of the Parkinson's Disease and Movement Disorders Center at the University of Maryland. “These practice parameters are written for people who have experience taking care of patients, and can interpret guidelines for their individual patients.”
The practice parameters often offer only tentative guidance – or none at all – because the available literature doesn't provide definitive answers. The members of the Subcommittee grade the quality of the literature they deem worthy of inclusion, and then draw the best conclusions possible.
This bothers some physicians who look to these parameters for more practical guidance. Some have criticized discussions of drugs, for example, because they report only what the literature says about the drugs, and not what many doctors know about the drugs.
“What I'm used to with practice parameters is more guidance on individual patients,” said Christopher Goetz, MD, Associate Chair of the Department of Neurological Sciences at Rush Presbyterian St. Luke's Medical Center in Chicago. “Parameters evaluate populations of patients, so it will not help me as I look at one patient. This is a good evidence-based review, but I would rather have had that be the title – an ‘evidence-based review’ – rather than ‘practice parameters’.”
DIAGNOSIS AND PROGNOSIS
Members of the Quality Standards Subcommittee attempted to identify “red flags” that suggest the proper diagnosis may not be Parkinson disease. For example, patients with forms of parkinsonism other than PD are more likely to fall often early in the disease, show little or no response to levodopa, have symmetric motor difficulties but lack tremor, and display relatively rapid progression. In addition, they may have urinary problems, erectile dysfunction, or orthostatic hypotension.
Patients who improve greatly when challenged with a large dose of levodopa and apomorphine probably have PD rather than one of the parkinsonian syndromes, they determined.
But one expert who was not involved in the evaluation process warns against this technique. Joseph Jankovic, MD, Professor of Neurology at Baylor College of Medicine in Houston, said: “Both drugs can prime patients to develop levodopa-induced dyskinesias down the road, so the drugs can potentially be dangerous.”
“I doubt that a neurologist would use such a drug challenge,” he continued. “But then if a physician does not follow the published parameters, an insurance company might say, ‘You didn't do the levodopa-apomorphine challenge, so how do you know the patient has Parkinson disease?’ Sometimes sound clinical judgment and experience are more important than a double-blind study.”
In another section, the Quality Standards Subcommittee wrote that testing the sense of smell may distinguish PD from progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD), because loss of smell is more pronounced with PSP and CBD. This feature, however, is not helpful in distinguishing PD from multiple system atrophy.
People who are older at the onset of symptoms, and who show rigidity and hypokinesia, are likely to experience more rapid progression of symptoms, especially if they are male, have had a stroke, or have auditory or visual problems, the committee wrote.
Tremor as an initial symptom predicts longer benefit from levodopa, the Subcommittee found. On this finding, Dr. Jankovic commented: “We reported this some time ago, and this report confirms that tremor-dominant patients have a better prognosis than PD patients who start with postural instability and gait difficulty.”
“In the future, genetic testing will play an increasingly important role in the diagnosis and prognostication of patients with PD symptoms,” Dr. Jankovic said, “and neurologists must learn how to interpret the various DNA tests.”
MOTOR FLUCTUATIONS AND DYSKINESIA
The Subcommittee addressed five questions: Which medications reduce the recurrence of symptoms? What is the relative efficacy of medications in reducing off time? Which medications reduce dyskinesia? What surgical therapies are most effective? Which factors predict improvement after deep brain stimulation (DBS)?
The evidence suggested, they wrote, that entacapone and rasagiline reduce “off time” – the periods when symptoms return. And they concluded that pergolide, pramipexole, ropinirole, and tolcapone are probably effective. Apomorphine subcutaneously injected, cabergoline and selegiline are possibly effective, and sustained release carbidopa/levodopa and bromocriptine do not decrease off time.
Ropinirole may be chosen over bromocriptine for reducing off time, they suggested, but there is insufficient evidence to recommend one agent over another.
For reducing dyskinesia, they wrote that amantadine is possibly effective, and there is insufficient evidence regarding the effectiveness of clozapine in reducing dyskinesia.
DBS of the subthalamic nucleus (STN) is possibly effective in improving motor function and reducing motor fluctuations, dyskinesia and the need for antiparkinsonian medication in PD patients, the Subcommittee wrote, but there is insufficient evidence to determine if other surgical options are effective. Moreover, they suggested that the preoperative response to levodopa probably predicts postsurgical improvement, and that younger patients may improve more than older patients who have had the disease longer.
NEUROPROTECTIVE STRATEGIES AND ALTERNATIVE THERAPIES
Levodopa is the only therapy found that reliably subdues symptoms, but there is no long-term evidence that it slows progression of the disease. However, the adverse effects of the drug are ultimately reversible, according to the Subcommittee.
The seeds of the legume, Mucuma pruriens (known commonly as velvet bean, cowitch, and cowhag) contain levodopa and improve motor problems at least temporarily. Also, vitamin C can prolong the action of levodopa, but vitamin E is probably of no value.
One study found that acupuncture relieved symptoms, but did not affect disease progression. Another study showed that the Alexander technique, a system of exercises that promotes better posture, provided some lasting benefit. However, committee members concluded that evidence is insufficient to support or refute the use of either technique.
In addition, the committee found that speech therapy may be effective in improving speech volume and exercise therapy may help improve physical functions.
Depression, Psychosis, and Dementia
The Subcommittee concluded that these scales – The Beck Depression Inventory, the Hamilton Depression Rating Scale (HDRS) and the Montgomery Asberg Depression Rating Scale (MADRS), may be useful for screening depression in patients with PD. However, the HDRS and MADRS require a trained administrator.
There is insufficient evidence to support or refute the Parkinson Psychosis Rating Scale as a screening tool, they wrote. But for PD patients with dementia, the Mini-Mental State Examination and the Cambridge Cognitive Examination are useful.
To treat depression in PD patients, amitriptyline may be effective, but its anticholinergic effects may worsen cognition, according to the evidence. The committee said that amitriptyline “is not necessarily the first choice for treatment associated with PD.” But, Dr. Jankovic noted, it did not mention specific selective serotonin reuptake inhibitors (SSRIs), which are by far the most commonly recommended anti-depressants prescribed for depression in PD patients.
“I'm sure one of the reasons they didn't mention SSRIs is that there are not robust, double-blind, placebo-controlled studies,” he said.
Dr. Jankovic agrees that consensus parameters provide useful information, but he cautioned that they may lead to a “cookbook approach” to treating individual patients.
“Some physicians, third-party payers, and lawyers may misinterpret any recommendation lower than ‘class A,’ or ‘insufficient evidence,’ as not efficacious,” Dr. Jankovic said. “I have a dozen numerous double-blind, placebo-controlled trials in my 30 years of clinical research and it is not until I use the drug in a long-term open-label study that I get a feel for its safety, tolerability, and efficacy.”
The Subcommittee weighed in on the evidence in treating psychosis in PD patients. They found that clozapine appears to be effective, but patients must be monitored carefully for signs of agranulocytosis, a serious side-effect.
“Although quetiapine received only class C recommendation, in practice it is clearly the most useful drug for management of psychosis in patients with PD,” said Dr. Jankovic. Patients with PD dementia or dementia with Lewy bodies seem to improve slightly with donepezil.
Rivastigmine appears to be slightly effective, but may exacerbate tremor. Memantine was not mentioned, but Dr. Jankovic said he is doing a study at Baylor. The investigators hope it will show that the drug is effective in PD patients who have mild cognitive deficits.
For more highlights on the four new evidence-based reviews, see “Parkinson Disease Practice Parameters from the AAN.”