December 17, 2012
Diagnosing Peripheral Neuropathy
By David Yeager
For The Record
Vol. 24 No. 23 P. 22
The diagnosis for older adults can be particularly challenging, making it imperative that physicians become more efficient in their patient workups.
Peripheral neuropathy afflicts as many as 8% of people over the age of 55, and the numbers are increasing. There are more than 100 known causes, each with its own characteristic symptoms, development pattern, and prognosis. For this reason, confirming a diagnosis and developing a treatment plan can be highly challenging.
To complicate matters, there are only a few diagnostic tests for peripheral neuropathy that are supported by significant research-based evidence, and there is no standard diagnostic protocol. Faced with an indeterminate set of symptoms, physicians may be inclined to order a wide variety of tests, sometimes at great expense and with little clinical benefit. In fact, physicians often forgo less expensive, more reliable tests when attempting to confirm a peripheral neuropathy diagnosis.
A University of Michigan Medical School (UMMS) study of 1,031 patients with confirmed neuropathy diagnoses published in the Archives of Internal Medicine found that 23.2% of patients received an MRI of the brain or spine, contributing to significant expenditures during the diagnostic phase of neuropathy treatment. More surprisingly, only 1% of patients received inexpensive and far more reliable blood glucose tolerance tests. The study also found that physician test-ordering patterns were highly variable.
Knowing a Patient’s History
If peripheral neuropathy is suspected, it may take some detective work to determine the underlying cause. Therefore, physicians should first conduct a thorough patient history, as many underlying causes can be deduced just by asking questions.
“Far and away, the most important thing is a good history,” says Brian Callaghan, MD, an assistant professor of neurology at UMMS and lead study author. “The vast majority of causes are determined based on talking with the patient and, if you don’t figure it out based on talking to patients, the chance that testing is going to help with the diagnosis is small.”
Factors such as alcohol use, vitamin B12 deficiency, heredity, and diabetes can influence whether a person develops peripheral neuropathy. Diabetes is a particularly significant factor; it’s estimated that as many as one-half of people with diabetes develop some form of neuropathy.
Callaghan says some of the common, relatively inexpensive tests that can detect several of these problems, such as glucose tolerance testing, serum B12 testing, and serum protein electrophoresis (SPEP) with immunofixation electrophoresis (IFE), are the ones with the strongest clinical evidence to support their use in diagnosing peripheral neuropathy. Screening for common causes improves care and reduces treatment costs by diminishing reliance on more expensive tests and reducing treatment costs over time.
“If those disorders are picked up early—and that’s usually through a primary care physician—then [the patient] may never need to see a specialist. And also, if these diseases are picked up early, then the morbidity decreases,” says Annabel Wang, MD, an associate professor of neurology and director of the Neuromuscular Diagnostic Laboratory at the University of California, Irvine ALS and Neuromuscular Center. “Whereas, if you’re an unrecognized diabetic, it could take four or five years for someone to figure out that you have diabetes; then you have four or five years’ additional damage and therefore more complications and more secondary issues that develop.”
Common symptoms of peripheral neuropathy are pain; tingling; loss of sensation, usually more in the toes than the fingers; distal weakness; and difficulty walking. However, symptoms may not be as straightforward in older patients. If someone has difficulty extending his or her toes, flexing the foot, or determining whether the toes are moving up or down, that could be indicative of peripheral neuropathy. A diminished ability to feel cold or a pinprick or the absence of normal reflexes also is consistent with a neuropathy diagnosis.
If the history and exam point to peripheral neuropathy, the goal is to clinically characterize it as much as possible. Determining whether it’s acute or chronic is straightforward: If the symptoms have progressed during a period of time that exceeds three to six months, it’s chronic. The next step is to determine whether the neuropathy is axonal or demyelinating, which can partly be determined by electrodiagnostic studies. Frequently, an electromyogram (EMG) is used to test nerve conduction.
“EMG does not have to be performed in a patient who has a typical presentation and is, for example, diabetic,” says Gil I. Wolfe, MD, FAAN, chair of the department of neurology at the Jacobs Neurological Institute of the University at Buffalo School of Medicine, State University of New York. “There is a typical clinical pattern for the most common form of diabetic neuropathy, a distal axonal polyneuropathy, which is readily recognized by neuropathy experts. One can argue that you don’t have to do an EMG in this situation. Not only is this of economic benefit to the overall health system, but it also eliminates subjecting our patients to what can be a painful test. But in many other situations, EMG provides crucial information in characterizing neuropathy, especially whether it is an axonal or demyelinating process.”
Once it has been determined whether the neuropathy is axonal or demyelinating, the search for an underlying cause can continue. Wolfe says it’s important to note that many medications can cause an axonal neuropathy pattern, even commonly used ones such as antibiotics, so it’s important for physicians to obtain a detailed medication history.
Because there are different types of neuropathy, people may exhibit a wide variation in symptoms. Wolfe says most neuropathies are axonal, which generally requires more diagnostic evaluation to determine the cause. Deciding which tests are appropriate depends on the suspected cause.
“I think the most important thing is to remember that there are different types of neuropathies. Neuropathies can be only sensory and only cause symptoms or they can also be only motor so they only cause weakness without sensory symptoms,” Wang says. “And then there’s an entity called small fiber neuropathy, where there may be pain and temperature loss or just a lot of pain, and the changes in vibration or the reflexes may be absent. Those are the cases that perhaps are missed and perhaps are very early neuropathies.”
Neuropathies may be classified as primarily small fiber; primarily large fiber, which includes loss of position sense, loss of vibratory sense, and some degree of loss of light touch; or both. Small vs. large fiber can be identified to some degree based on a person’s history and examination. In addition, Wolfe says fasting glucose/two-hour glucose tolerance tests, HIV tests, urine protein electrophoresis tests, hepatitis B and C serologies, thiamine/pyridoxine tests, and celiac serologies may be used to diagnose small fiber neuropathies. Those tests, as well as vitamin B12 testing with methymalonic acid/homocysteine levels, SPEP with IFE testing, metabolic panels, lipid and cholesterol levels, syphilis serologies, Lyme serologies, urine heavy metal levels, and many others can be used to test for suspected large fiber neuropathies.
However, aside from the glucose, B12, and SPEP tests, none of these tests are supported by the American Academy of Neurology (AAN) Practice Parameter for evaluating distal symmetric neuropathies. Wolfe says the only reason physicians should use some of these tests is if the neuropathy is unclassified and cancer or an immune-related condition is suspected. They should never be used on a routine basis.
Another important factor to consider is heredity. The AAN Practice Parameter supports some genetic testing but only in cases where certain hereditary causes are suspected. In older patients who develop neuropathy, heredity’s role can be easy to miss.
“There is a sector of these late-onset neuropathies that on initial inspection may look acquired but very well may be hereditary,” Wolfe says. “And you shouldn’t discard the possibility that a neuropathy is hereditary just because it started in somebody after age 50. Some of those patients may very well have a hereditary process, either because of certain ion channel mutations or because of Charcot-Marie-Tooth disease type 2.”
Beware of Diminishing Returns
Because physicians try to render treatment based on the symptoms’ origin, many will continue to order tests when a cause is not found. But despite all the tests that could potentially be ordered, Wolfe estimates that 20% to 25% of neuropathies end up being unclassified, and most unclassified neuropathies are seen in older patients. Although there are some factors associated with neuropathy that have only recently been understood, such as copper deficiency, most of the increase in cases is due to diabetes.
“We can sometimes get in more trouble by ordering more tests,” Callaghan says. “We really need to focus on diabetes and taking a good history from our patients. Those are the things that are really going to make a big difference.”
The most important consideration for physicians in deciding which tests to order is how they will affect patient care. Unfortunately, other than treating pain, there are not many treatments for neuropathy. Callaghan says most of the 40 to 50 tests that can be given for neuropathy are very low yield. Many not only have minimal influence on how treatment is rendered, but they don’t even affect the way the physician thinks about the patient’s condition. Callaghan cites MRI as the top example because it looks at the central nervous system and is rarely indicated for a patient with a peripheral nerve problem.
Thyroid and rheumatological studies also are relied on too heavily. Although these tests are frequently ordered, Callaghan says they rarely affect patient management.
Callaghan says too much testing can make it more difficult to determine an underlying cause, especially if a test has a high false-positive rate or multiple tests produce conflicting results. One problem he has noted in his research is that physicians are apt to order a battery of tests as a rule rather than as the exception. He recommends that physicians stick with the tests that have the best levels of evidence and order additional testing only if something unusual is suspected.
“For example, there are some warning signs that you might not be dealing with a garden-variety neuropathy, such as if the neuropathy comes on quickly, is very asymmetric side to side, or involves weakness more than sensory changes. Those are examples of things that might make you order more diagnostic testing,” Callaghan says. “But in most patients, the current evidence would suggest that you probably should only get a few tests.”
One factor that may help reduce unneeded testing is patient education. Wang says it’s important for physicians to address patient questions about diagnosis and treatment as completely as possible. Patients who don’t fully understand the condition and how it’s treated may have more anxiety about it, which may make them more likely to push for additional testing.
“The fear is not knowing what’s going to happen to them, and that probably creates a lot of unnecessary workups,” Wang says. “Whereas, if they had an evaluation and they feel that they understand what their disease process is, they don’t feel that they’ve been abandoned. That in itself can save a lot of money.”
However, the main responsibility in the use of testing lies with physicians. Wolfe says physicians need to become more efficient in working up patients. Many lab tests currently ordered on a routine basis have low clinical efficacy and therefore low cost-efficiency. He believes this is an area that can be improved with better research into which tests are effective.
Callaghan agrees that more research is needed to define the roles of all diagnostic tests that are ordered for peripheral neuropathy. He says the AAN guidelines are highly useful and that physicians should avoid using nonrecommended tests for routine screening. However, physicians can be slow to change the way they practice, especially if a new approach calls for doing less rather than more. Even though guidelines are available, many physicians continue to routinely order high-cost, low-yield tests.
“I think there are two big obstacles. One is defining what really should be the best diagnostic approach, and No. 2 is altering physician behavior. And those are both difficult to do,” Callaghan says. “Despite the guidelines that are out there, we’re not necessarily practicing according to those guidelines.”
— David Yeager is a freelance writer and editor based in Royersford, Pennsylvania.