Researchers highlight refinements to CIDP diagnosis with eye toward revamped guidelines

When a patient presents with polyneuropathy and progressive or relapsing symptoms such as tingling, proximal weakness, and absent reflexes, the clinician should consider a diagnosis of chronic inflammatory demyelinating polyneuropathy (CIDP), according to the National Institutes of Health. Confirming the diagnosis requires further testing, however.

Specialists have come to regard the detailed criteria published in 2010 by the European Federation of Neurological Societies and the Peripheral Nerve Society (EFNS/PNS) as the best guidelines for confirming a diagnosis of CIDP. But in February, a survey  by Gelinas et al. indicated that no more than 13% of community neurologists use these guidelines. In addition, about half of respondents endorsed electrodiagnostic criteria that do not support a diagnosis of CIDP.

Competing demands may overshadow guidelines

Lack of awareness of the EFNS/PNS guidelines among neurologists may partly explain their restricted use. “For those who do know about them, there is a perception that they are too time consuming or hard to follow,” said Jeffrey Allen, MD, associate professor of neuromuscular medicine at the University of Minnesota, Minneapolis. “Although it can seem like a daunting task to track down a guideline and filter through recommendations during a clinic visit, recent guidelines, including those published by the EFNS/PNS, are easy to access and meant to be used during routine practice. They can really be a critical tool to improve CIDP diagnostic accuracy.”

“It takes effort to step back after you’ve done your EMG testing and look at the references; calculate the percentages on your distal latencies, velocities, and F-waves; and see if it meets EFNS/PNS criteria,” said Amro Stino, MD, assistant professor and director of the peripheral neuropathy clinic at the University of Michigan, Ann Arbor. In private practice, financial concerns promote an emphasis on diagnostic volume, rather than diagnostic quality, which “puts a lot of pressure on physicians to stick labels [on patients], rather than verifying whether those labels are accurate.”

Neurologists also may have problems accessing the guidelines. “No one is going to have memorized the guidelines, so, to be useful, they need to be readily available while the physician is seeing patients,” said Richard A. Lewis, MD, professor of neurology at Cedars Sinai in Los Angeles. “Most clinical practices don’t do that.”

The current approach to medical training has made neurologists comfortable with administering intravenous immunoglobulin (IVIg), a recommended treatment for CIDP, but has not impressed upon them the need for diagnostic criteria, said David S. Saperstein, MD, director of the Center for Complex Neurology,  Ehlers–Danlos Syndrome, and Postural Orthostatic Tachycardia Syndrome in Phoenix. Neurologists have developed a habit of not using guidelines, diagnosing “by feel” instead, he added. “This is not a problem for neuropathy experts, but by disregarding criteria, many neurologists often give too much weight to mild nerve conduction or lab abnormalities. People are more confident of their ability to make diagnoses in complicated situations than they probably should be.”

Besides the limited use of the 2010 guidelines, other factors make the diagnosis of CIDP difficult. One salient factor is the lack of reliable biomarkers, Dr. Lewis said. “The best biomarker is the EMG … but the interpretations [of this test] are not always as good as we would like them to be.”

Although favorable response to treatment supports the diagnosis of CIDP, a subjective response does not discriminate well between patients with CIDP and those without. In 2015, a retrospective study by Allen et al. found that most patients who had received a diagnosis of CIDP but subsequently were determined not to have the disease still thought that they benefited from treatment. Patients frequently say, “I’m better with treatment because it gives me more energy,’” said Dr. Lewis. “But when based on objective measures of improvement [e.g., improved walking or grip strength], only about 14% had objective improvement.”

Diagnosis has evolved since 2010

Since the publication of the EFNS/PNS guideline in 2010, scientific advancements have shed important insights into the pathobiologic underpinnings of CIDP. One important discovery is the identification of the nodal and paranodal antibodies neurofascin-155, neurofascin-140/186, and contactin-1 in subsets of patients with CIDP, Dr. Allen said. Between 10% and 15% of patients with CIDP may have these antibodies. “They tend to have a distinct phenotype that often includes tremor or ataxia. There is much to be learned on how best to treat these patients, but unlike CIDP patients that do not harbor these antibodies, [those who do] tend to respond poorly to IVIg. These patients appear to respond more favorably to corticosteroids and B-cell depletion therapy.”

In addition, ultrasound of nerves is emerging as a potential diagnostic biomarker for CIDP. Data indicate that nerves are enlarged in CIDP, and a study by Goedee and colleagues suggests that ultrasound can help neurologists examine nerve thickening in more detail. The technique enables multiple nerves in the arms and legs to be assessed quickly. Nerve ultrasound is emerging as a complement to EMG in peripheral neuropathy, Dr. Stino said.

The technique has become more advanced and standardized, but neurologists have not yet determined the ideal area of nerves to examine or how many nerves should be examined, Dr. Saperstein said. “In the not-too-distant future, just as there are nerve conduction criteria, I suspect there will be formalized and accepted ultrasound criteria,” he added. Just as with nerve conduction studies, such criteria will support, but not by themselves establish, a diagnosis of CIDP. “Following changes to those nerves could be a useful treatment biomarker, which is sorely needed,” he said.

Another development relates to cerebrospinal fluid (CSF) testing. Previous studies have suggested that the upper limit of CSF total protein was 0.45 g/L in healthy individuals. Higher levels were considered possible indications of a disorder such as CIDP. But peripheral nerve specialists and neuromuscular specialists have long known that the level of CSF total protein increases with age. “The normal value is too low for many patients,” Dr. Saperstein said. Specialists know not to consider mildly elevated CSF total protein levels significant in older patients who do not have the signature features of CIDP.

Recent studies have spread awareness of this information among general neurologists. A study by Breiner and colleagues reviewed 57 years’ worth of medical records and found that the weighted average of CSF total protein upper reference limit was 0.55 g/L. They also found a consistent correlation between age and CSF total protein level. Values gradually exceeded 0.60 g/L after age 50 years, they wrote. Breiner’s group subsequently applied their findings to the analysis of patients with correctly and incorrectly diagnosed CIDP. They concluded that adopting higher, age-dependent upper limits of CSF total protein slightly reduced the sensitivity but increased the specificity of CSF analysis in the diagnosis of CIDP.

Clinical and electrophysiologic definitions could be improved

No guideline is perfect, and an updated guideline for the diagnosis of CIDP could include several improvements. A new guideline should contain “better, more concise clinical definitions of CIDP,” Dr. Allen said. Especially needed are improved definitions of the asymmetric, distal, sensory, and motor atypical CIDP variants. “We know that atypical CIDP is hardest to diagnose and [is] frequently misdiagnosed as CIDP. The new guideline really needs to focus in on atypical CIDP so that the clinical and electrophysiologic features are more clearly defined.”

“The atypical cases have a lot of potential mimics, and that’s where the problem with diagnosis occurs,” Dr. Lewis said. DADS syndrome, for example, has a similar presentation to that of diabetic neuropathy. “The guidelines from 2010 couldn’t address the specifics of all those atypical cases.”

The electrophysiologic testing required to diagnose typical CIDP also could be made more stringent. The current criteria require that only two nerves display features of demyelination. “In my experience, testing of bilateral upper limbs, in addition to one lower limb, generally has been helpful in further improving diagnostic accuracy,” Dr. Stino said. “Putting an emphasis on more upper limb bilateral testing would be helpful, in addition to at least one lower limb test.”

Recommendations for standard of care treatment also are in need of updates. “The most notable treatment update will be inclusion of subcutaneous immunoglobulin (SCIg) into the guideline,” Dr. Allen said. In 2018, van Schaik and colleagues published results of a trial of SCIg for maintenance therapy in CIDP. The drug was generally well tolerated. Compared with placebo, patients treated with SCIg had fewer relapses during 24 weeks of treatment, thus establishing the treatment’s efficacy for CIDP maintenance treatment.

Ideally, revised guidelines also would include objective measures of treatment outcome. Depending exclusively on the standard neurologic examination and patient report is not optimal, Dr. Saperstein said. “If you don’t have good, reliable, objective measures of outcome, you’re at risk of things being murky when you treat a patient,” he added.

“It is extremely important to improve diagnostic accuracy,” said Dr. Lewis. “Misdiagnosis leads to mistreatment, which means that the patient may not be getting the appropriate therapy for his or her true disorder. In addition, there is a significant health care cost of treatments when they are not appropriate.”

In the end, the goal of all revisions to the guideline should be to help neurologists “do a better job at identifying patients with CIDP, both typical and atypical types, while at the same time doing a better job at excluding patients without CIDP – basically optimizing the sensitivity and specificity of the guideline,” Dr. Allen said. “The key is to use the guideline to help us make better use of existing tools.”