The diagnosis of neurosarcoidosis often is difficult. Definitive diagnosis can only be made by biopsy (surgically removing a tissue sample). Because of the risks associated with brain biopsies, they are avoided as much as possible. Other investigations that may be performed in any of the symptoms mentioned above are computed tomography (CT) or magnetic resonance imaging (MRI) of the brain, lumbar puncture, electroencephalography (EEG) and evoked potential (EP) studies. If the diagnosis of sarcoidosis is suspected, typical X-ray or CT appearances of the chest may make the diagnosis more likely; elevations in angiotensin-converting enzyme and calcium in the blood, too, make sarcoidosis more likely. In the past, the Kveim test was used to diagnose sarcoidosis. This now obsolete test had a high (85%) sensitivity, but required spleen tissue of a known sarcoidosis patient, an extract of which was injected into the skin of a suspected case.

Only biopsy of suspicious lesions in the brain or elsewhere is considered useful for a definitive diagnosis of neurosarcoid. This would demonstrate granulomas (collections of inflammatory cells) rich in epithelioid cells and surrounded by other immune system cells (e.g. plasma cells, mast cells). Biopsy may be performed to distinguish mass lesions from tumours (e.g. gliomas).

MRI with gadolinium enhancement is the most useful neuroimaging test. This may show enhancement of the pia mater or white matter lesions that may resemble the lesions seen in multiple sclerosis.

Lumbar puncture may demonstrate raised protein level, pleiocytosis (i.e. increased presence of both lymphocytes and neutrophil granulocytes) and oligoclonal bands. Various other tests (e.g. ACE level in CSF) have little added value.

Some recent papers propose to classify neurosarcoidosis by likelihood:

- "Definite" neurosarcoidosis can only be diagnosed by plausible symptoms, a positive biopsy and no other possible causes for the symptoms

- "Probable" neurosarcoidosis can be diagnosed if the symptoms are suggestive, there is evidence of central nervous system inflammation (e.g. CSF and MRI), and other diagnoses have been excluded. A diagnosis of systemic sarcoidosis is not essential.

- "Possible" neurosarcoidosis may be diagnosed if there are symptoms not due to other conditions but other criteria are not fulfilled.

Diagnosis of sarcoidosis is a matter of exclusion, as there is no specific test for the condition. To exclude sarcoidosis in a case presenting with pulmonary symptoms might involve a chest radiograph, CT scan of chest, PET scan, CT-guided biopsy, mediastinoscopy, open lung biopsy, bronchoscopy with biopsy, endobronchial ultrasound, and endoscopic ultrasound with fine-needle aspiration of mediastinal lymph nodes (EBUS FNA). Tissue from biopsy of lymph nodes is subjected to both flow cytometry to rule out cancer and special stains (acid fast bacilli stain and Gömöri methenamine silver stain) to rule out microorganisms and fungi.

Serum markers of sarcoidosis, include: serum amyloid A, soluble interleukin-2 receptor, lysozyme, angiotensin converting enzyme, and the glycoprotein KL-6. Angiotensin-converting enzyme blood levels are used in the monitoring of sarcoidosis. A bronchoalveolar lavage can show an elevated (of at least 3.5) CD4/CD8 T cell ratio, which is indicative (but not proof) of pulmonary sarcoidosis. In at least one study the induced sputum ratio of CD4/CD8 and level of TNF was correlated to those in the lavage fluid. A sarcoidosis-like lung disease called granulomatous–lymphocytic interstitial lung disease can be seen in patients with common variable immunodeficiency (CVID) and therefore serum antibody levels should be measured to exclude CVID.

Differential diagnosis includes metastatic disease, lymphoma, septic emboli, rheumatoid nodules, granulomatosis with polyangiitis, varicella infection, tuberculosis, and atypical infections, such as "Mycobacterium avium" complex, cytomegalovirus, and cryptococcus. Sarcoidosis is confused most commonly with neoplastic diseases, such as lymphoma, or with disorders characterized also by a mononuclear cell granulomatous inflammatory process, such as the mycobacterial and fungal disorders.

Chest radiograph changes are divided into four stages:

1. bihilar lymphadenopathy

2. bihilar lymphadenopathy and reticulonodular infiltrates

3. bilateral pulmonary infiltrates

4. fibrocystic sarcoidosis typically with upward hilar retraction, cystic and bullous changes

Although people with stage 1 radiographs tend to have the acute or subacute, reversible form of the disease, those with stages 2 and 3 often have the chronic, progressive disease; these patterns do not represent consecutive "stages" of sarcoidosis. Thus, except for epidemiologic purposes, this categorization is mostly of historic interest.

In sarcoidosis presenting in the Caucasian population, hilar adenopathy and erythema nodosum are the most common initial symptoms. In this population, a biopsy of the gastrocnemius muscle is a useful tool in correctly diagnosing the person. The presence of a noncaseating epithelioid granuloma in a gastrocnemius specimen is definitive evidence of sarcoidosis, as other tuberculoid and fungal diseases extremely rarely present histologically in this muscle.

Sarcoidosis may be divided into the following types:

- Annular sarcoidosis

- Erythrodermic sarcoidosis

- Ichthyosiform sarcoidosis

- Hypopigmented sarcoidosis

- Löfgren syndrome

- Lupus pernio

- Morpheaform sarcoidosis

- Mucosal sarcoidosis

- Neurosarcoidosis

- Papular sarcoid

- Scar sarcoid

- Subcutaneous sarcoidosis

- Systemic sarcoidosis

- Ulcerative sarcoidosis

Diagnosis is made by an ophthalmologist during eye examination. Further tests such as fluorescein angiography or lumbar puncture are usually performed to confirm the diagnosis.

Neurosarcoidosis is a similar autoimmune disorder that can be confused with APMPPE.

Owing to the self-limiting nature of the disease, treatment is generally not required. In cases where lesions appear to be interfering with the optic nerve, methyl prednisone is prescribed.

In most MS-associated optic neuritis, visual function spontaneously improves over 2–3 months, and there is evidence that corticosteroid treatment does not affect the long term outcome. However, for optic neuritis that is not MS-associated (or atypical optic neuritis) the evidence is less clear and therefore the threshold for treatment with intravenous corticosteroids is lower. Intravenous corticosteroids also reduce the risk of developing MS in the following two years in patients with MRI lesions; but this effect disappears by the third year of follow up.

Paradoxically, oral administration of corticosteroids in this situation may lead to more recurrent attacks than in non-treated patients (though oral steroids are generally prescribed after the intravenous course, to wean the patient off the medication). This effect of corticosteroids seems to be limited to optic neuritis and has not been observed in other diseases treated with corticosteroids.

A Cochrane Systematic Review studied the effect of corticosteroids for treating people with acute optic neuritis. Specific corticosteroids studied included intravenous and oral methylprednisone, and oral prednisone. The authors conclude that current evidence does not show a benefit of either intravenous or oral corticosteroids for rate of recovery of vision (in terms of visual acuity, contrast sensitivity, or visual fields)..

The repetition of an idiopathic optic neuritis is considered a distinct clinical condition, and when it shows demyelination, it has been found to be associated to anti-MOG and AQP4-negative neuromyelitis optica

When an inflammatory recurrent optic neuritis is not demyelinating, it is called "Chronic relapsing inflammatory optic neuropathy" (CRION)

When it is anti-MOG related, it is demyelinating and it is considered inside the anti-MOG associated inflammatory demyelinating diseases.

A thorough medical history and physical examination, including a neurological examination, are the first steps in making a diagnosis. This alone may be sufficient to diagnose Bell's Palsy, in the absence of other findings. Additional investigations may be pursued, including blood tests such as ESR for inflammation, and blood sugar levels for diabetes. If other specific causes, such as sarcoidosis or Lyme disease are suspected, specific tests such as angiotensin converting enzyme levels, chest x-ray or Lyme titer may be pursued. If there is a history of trauma, or a tumour is suspected, a CT scan may be used.

Facial nerve paralysis may be divided into supranuclear and infranuclear lesions.

If someone is suspected of having meningitis, blood tests are performed for markers of inflammation (e.g. C-reactive protein, complete blood count), as well as blood cultures.

The most important test in identifying or ruling out meningitis is analysis of the cerebrospinal fluid through lumbar puncture (LP, spinal tap). However, lumbar puncture is contraindicated if there is a mass in the brain (tumor or abscess) or the intracranial pressure (ICP) is elevated, as it may lead to brain herniation. If someone is at risk for either a mass or raised ICP (recent head injury, a known immune system problem, localizing neurological signs, or evidence on examination of a raised ICP), a CT or MRI scan is recommended prior to the lumbar puncture. This applies in 45% of all adult cases. If a CT or MRI is required before LP, or if LP proves difficult, professional guidelines suggest that antibiotics should be administered first to prevent delay in treatment, especially if this may be longer than 30 minutes. Often, CT or MRI scans are performed at a later stage to assess for complications of meningitis.

In severe forms of meningitis, monitoring of blood electrolytes may be important; for example, hyponatremia is common in bacterial meningitis. The cause of hyponatremia, however, is controversial and may include dehydration, the inappropriate secretion of the antidiuretic hormone (SIADH), or overly aggressive intravenous fluid administration.

Meningitis can be diagnosed after death has occurred. The findings from a post mortem are usually a widespread inflammation of the pia mater and arachnoid layers of the meninges. Neutrophil granulocytes tend to have migrated to the cerebrospinal fluid and the base of the brain, along with cranial nerves and the spinal cord, may be surrounded with pus – as may the meningeal vessels.

CES is often concurrent with congenital or degenerative diseases and represents a high cost of care to those admitted to the hospital for surgery. Hospital stays generally last 4 to 5 days, and cost an average of $100,000 to $150,000, unless the patient lives in a country where healthcare is free at the point of delivery.

The prognosis for complete recovery is dependent upon many factors. The most important of these is the severity and duration of compression upon the damaged nerve(s). Generally, the longer the time before intervention to remove the compression causing nerve damage, the greater the damage caused to the nerve(s).

Damage can be so severe that nerve regrowth is impossible, and the nerve damage will be permanent. In cases where the nerve has been damaged but is still capable of regrowth, recovery time is widely variable. Surgical intervention with decompression of the cauda equina can assist recovery. Delayed or severe nerve damage can mean up to several years' recovery time because nerve growth is exceptionally slow.

Review of the literature indicates that around 50-70% of patients have urinary retention (CES-R) on presentation with 30-50% having an incomplete syndrome (CES-I). The latter group, especially if the history is less than a few days, usually requires emergency MRI to confirm the diagnosis followed by prompt decompression. CES-I with its more favourable prognosis may become CES-R at a later stage.