Clinical Challenges: Diagnosing NMOSD

Neuromyelitis optica spectrum disorder (NMOSD) is a rare, disabling, sometimes fatal central nervous system inflammatory demyelinating disease, with features that make a diagnosis challenging.

First, noted Bruce Cree, MD, clinical research director of the Multiple Sclerosis Center at the University of California San Francisco, there is a lot of clinical overlap between neuromyelitis optica and multiple sclerosis (MS). “And for the better part of the preceding century there was ongoing debate back and forth as to whether the two disease states were one disease or two,” he told MedPage Today.

More recently, because of the improved understanding of the pathogenesis of NMOSD, and the development of an antibody test that is very specific for neuromyelitis optica, clinicians are able to distinguish the two disease states very clearly, he added. “So they have very different diagnostic criteria.”

“Because of its rarity not everybody knows about [NMOSD], whereas just about everybody has heard of MS,” said Cree. “So if a patient shows up in an emergency department in a community hospital where there aren’t neurology experts, and the person is receiving care for something like optic neuritis or a myelitis, the physicians might think this looks like MS, and could proceed down that pathway, without knowing that sometimes the patient comes into the emergency department with an optic neuritis or myelitis — which are the two most common presentations of neuromyelitis optica — that they might actually have neuromyelitis optica. So we do actually see a delay in diagnosis still to this day in many cases.”

Neuromyelitis optica almost always presents as sudden attacks, noted Brian G. Weinshenker, MD, of the Mayo Clinic in Rochester, Minnesota. “And in approximately 90% of patients, the first attack is optic neuritis or myelitis.”

With optic neuritis, he told MedPage Today, the attacks tend to be severe and can involve anything from vision impairment to complete visual loss, while myelitis can begin with neuropathic pain and progress to weakness, gait impairment, impaired bowel and bladder function, and even sensory loss.

“Typically symptoms of optic neuritis and myelitis build up for days to a week and then will last several weeks and gradually improve, usually with corticosteroid therapy, which is almost always administered in high doses intravenously,” Weinshenker said. “We used to think those were the only symptoms of neuromyelitis optica, but now we recognize that there are other symptoms as well.”

These may include intractable vomiting and hiccups that can come on quite suddenly and last for weeks, and even months, if the underlying problem is not recognized. “This is an unusual presentation and is usually associated with inflammation in the area postrema,” a structure in the dorsal medulla oblongata of the brainstem, he continued.

“It is also known as the vomiting center. And there are probably some specific characteristics of that part of the brain that make it particularly susceptible. The blood-brain barrier is leaky in that area, which probably allows the antibodies that are responsible for neuromyelitis optica to leak into that area of the brain, and is also the area with a relatively high expression of the target protein aquaporin-4, which is attacked by antibodies in this condition. So this is now a well-recognized and very characteristic syndrome of neuromyelitis optica,” Weinshenker said.

He noted that other syndromes associated with this condition include hypothalamic dysfunction, with symptoms that can include eating disorders or narcolepsy. “If you have a patient who looks like they have narcolepsy, and has low hypocretin levels in their spinal fluid and an isolated lesion in the hypothalamus on MRI, that is a recognized presentation of neuromyelitis optica,” Weinshenker said. “Again, the hypothalamus is an area of high expression of aquaporin-4 with a leaky blood-brain barrier.”

“These are syndromes that don’t involve the optic nerve and spinal cord, and that weren’t recognized in the past,” Weinshenker said, adding that it was only in 2004 that Mayo Clinic researchers were able to identify the first biomarker for the condition — serum aquaporin-4 immunoglobulin G (AQP4-IgG) antibodies — which enabled clinicians to recognize that some patients who were presenting with unexplained symptoms, such as vomiting and hiccups, actually had neuromyelitis optica, because they tested seropositive for this antibody.

“And this led to an expansion of the spectrum of the disease, which is now referred to as neuromyelitis optica spectrum disorder, in recognition that it does have a wider spectrum than just optic neuritis and myelitis,” said Weinshenker.

As for actually diagnosing NMOSD, there are clues that differentiate it from multiple sclerosis, “which is, by far, its most common mimic,” he said.

For example, if the patient has severe optic neuritis, brain magnetic resonance imaging (MRI) doesn’t show any other lesions suggestive of MS, and a spinal fluid test doesn’t show the oligoclonal bands present in most patients with MS, those are clues that the condition may be NMOSD.

In addition, MS is more common in white patients, whereas neuromyelitis optica is more common in non-Caucasian populations. So, if optic neuritis develops in an Asian or African-American patient, one should be suspicious of NMOSD, Weinshenker said.

In terms of the spinal cord, if a patient has severe myelitis, especially if the MRI reveals a spinal cord lesion that is longer than three vertebral segments, “neuromyelitis optica should certainly be on your radar,” Weinshenker said. “If you see a patient with intractable vomiting and hiccups, and especially if they’ve also had optic neuritis, that is not a syndrome we see in MS, and is a very characteristic syndrome for NMOSD. Having a high index of suspicion is very important in deciding whom to test for AQP4-IgG and how to interpret the results.”

At this point, testing for AQP4-IgG should be considered, he said. However, just 70% of patients with clinical NMOSD have these antibodies, which means that 30% of patients with NMOSD are AQP4-IgG seronegative. He noted that a recently discovered antibody — myelin oligodendrocyte glycoprotein (MOG-IgG) — is present in about 25-50% of those patients.

While testing for AQP4-IgG is quite sensitive and highly specific, said Weinshenker, testing for MOG-IgG is highly specific in cases of serum MOG-IgG at high titers, but less specific at low titers.

“If you have a low level of this MOG antibody, it’s a little hard to know what that means,” said Weinshenker. “It is very important to consider pretest probability — how typical the condition is for this MOG syndrome. If the clinical and radiologic presentation really looks like MS, and there is a low positive MOG antibody, detection of the antibody may be nonspecific. But when the MOG antibody titers are very high — over 100 — then that is quite specific.”

The upshot of these developments, he said, is that the majority of patients “can be reliably diagnosed from their first symptom.” He pointed out that the revised international consensus diagnostic criteria for NMOSD, published in Neurology in 2015, allow for patients who have a positive test for AQP4-IgG and have at least one core clinical characteristic (including optic neuritis, acute myelitis, area postrema syndrome, acute brain stem syndrome, symptomatic narcolepsy, or acute diencephalic syndrome consistent with NMOSD with characteristic brain imaging, or a symptomatic cerebral syndrome consistent with NMOSD with characteristic brain imaging) to meet the criteria for NMOSD.

“There is generally universal agreement among neurologists and experts that patients with AQP4-IgG antibodies and one of these symptoms should be treated indefinitely for prevention of further attacks, because we know the vast majority of patients with AQP4-IgG are going to relapse, and the relapses are often severe,” said Weinshenker.