IPND Diagnostic Criteria for Neuromyelitis Optica Spectrum Disorder
The International Panel for NMO Diagnosis (IPND) 2015 criteria stratify NMOSD diagnosis based on AQP4-IgG serostatus, requiring at least one core clinical characteristic for seropositive patients, while seronegative or unknown serostatus patients need more stringent clinical and neuroimaging criteria. 1
Core Clinical Characteristics (Required for Diagnosis)
The IPND criteria recognize six core clinical syndromes or MRI findings that define NMOSD: 1
- Optic neuritis (unilateral or bilateral) 1
- Acute myelitis 1
- Area postrema syndrome (intractable nausea, vomiting, or hiccups) 1
- Acute brainstem syndrome 1
- Symptomatic narcolepsy or acute diencephalic syndrome with NMOSD-typical diencephalic MRI lesions 1
- Symptomatic cerebral syndrome with NMOSD-typical brain lesions 1
Diagnostic Algorithm Based on AQP4-IgG Status
For AQP4-IgG Seropositive Patients
Only ONE core clinical characteristic is required for diagnosis. 2, 1 This represents the least stringent pathway because the antibody is highly specific for NMOSD. 3
For AQP4-IgG Seronegative or Unknown Status Patients
More restrictive criteria apply, requiring at least TWO core clinical characteristics occurring as separate clinical events, with at least one being optic neuritis, acute myelitis, or area postrema syndrome. 1 Additionally, specific MRI requirements must be met: 2
For acute optic neuritis:
- Brain MRI showing normal findings or only nonspecific white matter lesions, OR 2
- Optic nerve MRI with T2-hyperintense or T1-weighted gadolinium-enhancing lesion extending over half the optic nerve length or involving the optic chiasm 2
For acute myelitis:
- Intramedullary MRI lesion extending over 3 contiguous vertebral segments (longitudinally extensive transverse myelitis/LETM), OR 2
- Three contiguous segments of focal spinal cord atrophy in patients with history compatible with acute myelitis 2
For area postrema syndrome:
- Associated dorsal medulla/area postrema lesions on MRI 2
For acute brainstem syndrome:
- Associated periependymal brainstem lesions on MRI 2
Critical Diagnostic Considerations
Antibody Testing Methodology
AQP4-IgG testing must use cell-based assays with full-length human MOG as the target antigen, not peptides or denatured protein. 4 Serum is the specimen of choice over CSF because MOG-IgG is produced mostly extrathecally. 4
Differential Diagnosis from Multiple Sclerosis
The IPND criteria specifically address distinguishing NMOSD from MS, which is critical because up to 70% of NMOSD patients have brain MRI lesions at onset. 2, 3 Key distinguishing features include:
- AQP4-IgG antibodies are highly specific for NMOSD and absent in MS 3
- NMOSD brain lesions tend to be periependymal (especially around brainstem and area postrema) rather than periventricular 3
- MS characteristically shows perivenular lesions, Dawson's fingers, ovoid periventricular lesions, and juxtacortical U-fiber lesions that are absent or rare in NMOSD 3
- Initial brain MRI is often normal or shows only nonspecific white matter lesions in NMOSD 3
MOG-IgG Testing Considerations
When AQP4-IgG is negative, MOG-IgG testing should be performed because these represent distinct disease entities with different treatment implications. 2, 3 AQP4-IgG/MOG-IgG "double-positive" results are extremely rare and should prompt retesting using alternative methodologies. 3
Impact on Diagnostic Rates
Application of the 2015 IPND criteria increased NMOSD diagnosis rates by 76% compared to the 2006 criteria, with the majority of patients diagnosed within 2 years of onset (73%) or after a second attack (72%). 5, 6 The criteria improved diagnostic yield even in patients with unknown AQP4-IgG status, with 72% of confirmed seropositive patients meeting criteria when antibody status was assumed unknown. 6
Common Pitfalls to Avoid
Do not apply progressive disease criteria - the IPND criteria are designed for monophasic or relapsing disease courses, not progressive presentations. 2 Testing for MOG-IgG is not recommended in patients with progressive disease due to very low pre-test probability. 2
Do not misclassify based on brain MRI alone - the presence of brain lesions does not exclude NMOSD, as 70% of patients have brain involvement at onset. 2, 3 Focus on lesion distribution patterns (periependymal vs. periventricular) and the presence of characteristic spinal cord or optic nerve involvement. 3