The application of optical coherence tomography in neurologic diseases

Multiple sclerosis

Multiple sclerosis (MS) causes both demyelination and axonal loss through a complex and incompletely understood inflammatory and neurodegenerative process. The peripapillary retinal nerve fiber layer (RNFL) thickness can be used to assess axonal loss. Prior to exiting the lamina cribrosa, the retinal ganglion cell axons (which comprise the RNFL) are unmyelinated, thereby allowing OCT measurements to avoid any confounding due to myelin and pathologic processes affecting myelin. More recently, as investigators have been able to segment the various layers of the retina, the ganglion cell layer (GCL) thickness from macular scans has been studied to directly quantify neuronal loss, which may be more accurate than measurements of total macular volume.⁵ Microcystic retinal edema without a clinically identifiable cause can be seen in the inner nuclear layer (INL) in patients with MS, and this may independently affect visual function.⁶

In patients with MS who have not experienced a clinical episode of optic neuritis, the average peripapillary RNFL thickness has been shown to be thinner than in healthy controls.⁷ Even in the absence of a symptomatic episode of optic neuritis, RNFL thickness correlates well with visual function measures, including low-contrast visual acuity and contrast sensitivity.⁸ Similarly, in patients with MS who have not experienced an episode of optic neuritis, whole brain, gray matter, and white matter volumes have been shown to correlate with both RNFL thickness and GCL thickness.⁹ OCT can be used to determine if the rate of neuronal or axonal loss can be delayed with the institution of disease-modifying drugs.¹⁰

Analysis of the segmented retinal layers has shown that while there is generalized atrophy of the inner retinal layers in patients with MS, the GCL and inner plexiform layer (IPL) are preferentially affected.¹¹ Combined GCL and IPL thickness has been shown to be superior to RNFL thickness in correlations with scores from the Expanded Disability Status Scale, vision-related quality of life metrics, as well as high- and low-contrast visual acuity testing.¹²

The progressive thinning of retinal layers as measured by OCT varies depending on the subtype of MS. In eyes with optic neuritis, when compared to clinically isolated syndrome, both secondary progressive MS and relapsing-remitting MS had significantly thinner measurements.¹³

Acute demyelinating optic neuritis

OCT has been used to study the presentation and progression of acute idiopathic inflammatory demyelinating optic neuritis (figure 2). The peripapillary RNFL is thickened even in cases in which the optic nerve appears clinically normal.¹⁴ Changes can be reliably detected as early as 2 months following an acute episode of optic neuritis and the RNFL thickness typically stabilizes by 6 months though minor changes can happen thereafter.¹⁵ Measurements of the RNFL prior to 2 months may not be an accurate representation of axonal loss due to unresolved RNFL edema.¹⁵ Investigators have shown that the GCL does not swell acutely and can be followed more accurately over time for the presence of thinning indicative of neuronal loss.¹⁶ Thinning of the GCL can be detected by 1 month, and combined GCL + IPL thickness decreases before RNFL thinning in eyes with optic neuritis.¹⁷ Eyes with a previous remote episode of optic neuritis have diffuse thinning of the peripapillary RNFL with preferential involvement of the temporal quadrant.^15,18 While an episode of optic neuritis causes a substantial reduction in the RNFL, it is not believed to hasten the rate of progressive RNFL thinning over time due to the underlying disease process in patients with MS.⁷

• Download figure
• Open in new tab
• Download powerpoint

Figure 2 Early and late optical coherence tomography features in optic neuritis

Top row: on the day of symptoms presentation with vision of 20/40. Bottom row: 4 months later with vision of hand motion. (A) System automatic segmentation, thinner in the late stage. (B) Retinal nerve fiber layer (RNFL) thickness profile compared against normative data. (C) RNFL thickness per quadrants in a circle around the optic nerve (peripapillary thickness). (D) Total retinal thickness maps that includes RNFL and ganglion cell layer. INF = inferior; N = nasal; NAS = nasal; NI = nasal-inferior; NS = nasal-superior; SUP = superior; T = temporal; TMP = temporal; TI = temporal-inferior; TS = temporal-superior.

RNFL thickness measurements in patients with optic neuritis correlates with visual function parameters including mean deviation on automated perimetry. GCL thickness is a stronger predictor of visual disability than RNFL. Patients with RNFL thickness below 75 μm are more likely to have persistent visual field defects.¹⁹ RNFL thickness following an episode of optic neuritis has not been shown to be useful in predicting the future risk of developing MS.²⁰ These OCT findings in acute demyelinating optic neuritis are clinically applicable and can aid a practitioner in diagnosis and providing prognostic information.

Neuromyelitis optica

CNS demyelination in neuromyelitis optica (NMO) typically follows a more aggressive and debilitating course compared to MS. Distinguishing between these 2 diseases has important prognostic and therapeutic implications. After at least 3 months, the peripapillary RNFL is typically thinner following NMO optic neuritis compared to MS optic neuritis.²¹ The pattern of RNFL loss may help distinguish NMO optic neuritis from MS optic neuritis. In NMO optic neuritis, RNFL loss is more severe, diffuse, and involves the superior and inferior quadrants. In MS optic neuritis, RNFL loss is less severe and predominantly affects the temporal quadrant that contains the papillomacular bundle. Macular volumes are significantly reduced in eyes that have had an episode of NMO optic neuritis.²² Segmentation of retinal layers has also been used to demonstrate inner retinal changes following NMO optic neuritis. Eyes with a history of NMO optic neuritis have more thinning of the combined GCL-IPL than eyes of patients with NMO without a history of optic neuritis.¹⁶ Peripapillary RNFL measurements in eyes with a history of NMO optic neuritis have been shown to correlate with visual field defects, visual acuity, contrast sensitivity, visual evoked potentials, and the Expanded Disability Status Scale score.²³ These findings may be clinically useful as the imaging features may lead a physician to more strongly consider a diagnosis of NMO in the appropriate patient.

Papilledema and idiopathic intracranial hypertension

OCT has become a valuable tool in the care of patients with papilledema and idiopathic intracranial hypertension. For example, OCT volumetric measurements correlate well with clinical grading of optic nerve head photographs, where for Frisen scale 0 to scale 4 the total retinal volumes were 11.36, 12.53, 14.42, 17.48, and 21.81, respectively (figure 3).²⁴ OCT has also been shown to correlate with increased intracranial pressure²⁵ and with visual field changes demonstrating a linear correlation of a decrease of 0.6 db per each 10 μm of axonal thickness increase.²⁶ OCT is particularly useful clinically in papilledema as it can be used to see how optic nerve structure changes over time in response to treatment.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3 Papilledema severity as seen by confocal scanning laser ophthalmoscopy (SLO)

This feature, available on the Spectralis optical coherence tomography (OCT) system, utilizes a laser light to scan the retina point by point. The light reflected from the retina goes through a small aperture, which eliminates light coming from other panes, resulting in a sharp, high-resolution image. We present SLO scans corresponding to the clinical stages of the Frisen scale. Note that the scan of grade 5 could not be reliably created due to the amount of nerve swelling that goes beyond the imaging window on OCT.

OCT can also aid in differentiation of optic nerve head drusen (i.e., pseudopapilledema) from papilledema. True papilledema has a higher head elevation, hyporeflective space at the optic nerve head, and smaller RNFL–optic nerve head angle compared to cases of pseudopapilledema due to optic nerve head drusen.²⁷ Hyperreflective foci causing irregular elevation of the optic nerve head is suggestive of optic nerve head drusen (figure e-2).

Neuroretinitis

Neuroretinitis is characterized by optic nerve swelling associated with macular exudates in a star configuration. Peripapillary RNFL shows thickening, as one would expect from clinical examination; however, changes in the macula distinguish neuroretinitis from other causes of unilateral optic nerve edema. Typically with neuroretinitis, the optic nerve edema is associated with a serous macular detachment, and the macular star develops between 9 and 12 days after the subretinal fluid resorbs.²⁸ Macular OCT performed early in the disease course may demonstrate subretinal fluid, which should prompt consideration and evaluation for underlying causes of neuroretinitis (i.e., Bartonella henselae).²⁹ Macular OCT may also be helpful in detecting subclinical subretinal fluid in the cases in which the macular star does not develop (figure e-3).³⁰ There is often abnormal foveal contour and retinal thickening along with the subretinal fluid.³¹ Fluid or exudates in the outer plexiform layer may also be seen.³² Macular OCT can be critical in diagnosing a case of neuroretinitis when the macular changes are subtle on clinical examination.

Optic pathway glioma

In search of a noninvasive technique to detect optic pathway gliomas (OPGs), the first study utilizing OCT was published in 2010.³³ Patients with neurofibromatosis-1 (NF1) with OPGs were found to have thinner peripapillary RNFL and maculae than healthy controls, whereas patients with NF1 without OPGs had measurements that were similar to healthy controls. Eyes of most patients with OPGs (both with and without NF1) that demonstrate a reduction in peripapillary RNFL thickness also have visual field loss or decreased vision, highlighting the relationship between structure and function.³⁴ Similarly, combined macular GCL-IPL thickness was thinner in children with OPGs and vision loss.^e1 When compared to visual function assessment and optic disc evaluation, OCT measurements of the peripapillary RNFL had the highest correlation with the presence of OPGs.^e2 Changes in the peripapillary RNFL over time correlate with clinical and radiologic changes in children with OPGs, making OCT clinically useful for patients with OPGs.^e3

Compressive optic neuropathy

The clinical presentation of a compressive optic neuropathy depends on the location and chronicity of the lesion. Cupping of the optic nerve head can be seen in some cases of a compressive optic neuropathy. Numerous studies have sought to identify clinical features to help distinguish compressive optic neuropathy–related optic nerve head cupping from the more ubiquitous glaucomatous optic nerve head cupping.^e4 OCT can differentiate compressive optic neuropathy from glaucomatous optic neuropathy. Both entities show enlargement of the optic cup and thinning of the RNFL; however, there is more thinning nasally and temporally in compressive optic neuropathy.^e5

OCT can be used to predict the likelihood of postoperative visual recovery in compressive optic neuropathy (figure e-4). Patients with thinning (less than 97.5% of normal) of the RNFL were less likely to show any improvement in visual acuity or visual fields following resection of parachiasmal tumors. However, subjects with RNFL measurements above this threshold had significant improvement in visual acuity and improved visual fields postoperatively.^e6 These results were confirmed in another study that demonstrated that thicker preoperative RNFL correlated with an increased likelihood of visual field recovery regardless of a patient's age or duration of symptoms.^e7 Similar results have been found in patients with anterior visual pathway meningiomas.^e8 Patients with evidence of a chiasmopathy or an optic neuropathy who underwent surgery or fractionated stereotactic radiotherapy were more likely to have improvement in visual function if preoperative peripapillary RNFL measurements were normal. These findings have important prognostic implications in patients with compressive optic neuropathy.

Nonarteritic anterior ischemic optic neuropathy

As in other cases of optic nerve swelling, OCT can help diagnose edema in nonarteritic anterior ischemic optic neuropathy (NAION), the most common cause of an acquired optic neuropathy in adults over 50 years of age.^e9 As the edema resolves, OCT can be used to quantify RNFL thinning that stabilizes at 6 months and correlates with visual function.^e9 Both the degree of RNFL thinning and the anatomical location of thinning correlates with visual field defects.^e10 Macular thickness similarly correlates well with visual field sensitivity.^e11 Ganglion cell complex (defined as measurement between ILM to outer boundary of the IPL) measurements have also been shown to correlate well with the severity and location of visual field loss.^e12 After at least 6 months from a NAION, visual acuity correlates with temporal RNFL thickness and central macular IPL.^e13 One month following an NAION, combined GCL + IPL measurements are significantly reduced, and the thickness correlates with visual acuity and visual field mean deviation.¹⁷ OCT can be applied in NAION for both diagnostic and prognostic purposes. These applications also allow OCT to be useful investigating novel therapies.

Ethambutol

In a small case series of 3 patients with ethambutol optic neuropathy, thinning of the peripapillary RNFL was predictive of poor visual recovery.^e14 The temporal quadrant of the peripapillary RNFL, corresponding to the papillomacular bundle, showed the greatest average decrease in thickness in a separate study.^e15 However, it should be noted that OCT changes may not be evident at the time of presentation.^e16

Vitamin B₁₂ deficiency

OCT has shown thinning of the temporal peripapillary RNFL correlating to e serum B₁₂ levels.^e17 Patients with vitamin B₁₂ supplementation have demonstrated normalization of visual evoked potentials following supplementation, but the effect of supplementation on the peripapillary RNFL thickness as measured by OCT has not been reported.^e18

Vigabatrin

Vigabatrin (Sabril®; Lundbeck Inc., Deerfield, IL) is indicated for infantile spasms as well as other seizure disorders including simple and complex partial epilepsy. Clinically, patients experience characteristic nasal RNFL thinning prior to developing optic nerve pallor.^e19 OCT has been investigated as a tool to identify structural retinal changes due to vigabatrin. In a prospective cross-sectional observational study, thinning of the peripapillary RNFL correlated with mean deviation loss on automated perimetry. The study also found that thinning of the peripapillary RNFL might also occur before changes on automated perimetry. This suggests that there is utility of OCT in screening patients for vigabatrin-related retinal toxicity prior to the loss of visual function.^e20

Fingolimod

Fingolimod (FTY720, Gilenya®; Novartis Pharma AG, Basel, Switzerland) is a sphingosine-1-phosphate (S1P) receptor modulator and the first approved oral therapy for MS. The S1P receptor is involved in the regulation of vascular permeability, which may be a part of the pathophysiology of fingolimod-associated macular edema (FAME).^e21 Common OCT findings in FAME include increased central foveal thickness and hyporeflective areas affecting both the INL and ONL often within 1,000 µm of the foveal center (figure e-5).^e22 FAME typically occurs within the first 3–4 months after initiation of treatment.

OCT is useful in the detection of FAME, particularly as patients may be initially asymptomatic and findings on clinical examination may be subtle. In some cases, OCT may be helpful in differentiating FAME from an attack of optic neuritis, though the latter is more commonly associated with pain with eye movements and a relative afferent pupillary defect if the fellow eye has not been previously affected. OCT also demonstrates resolution of morphologic changes of the macula with successful management of FAME.

Alzheimer disease

Alzheimer disease (AD) can lead to decreased RNFL thickness, increased optic disc cupping, vascular tortuosity, and vision impairment.^e23 Decreased macular thickness and RNFL thinning are associated with abnormalities on electroretinogram even in patients with AD with normal visual acuity, normal visual fields, and normal color vision.^e24 More recently it has been shown that there is decreased choroidal thickness in patients with AD compared to healthy controls.^e25 OCT may play a role in early detection of AD but more studies are required to validate its utility as a biomarker of disease.

Parkinson disease

Ophthalmologic manifestations of Parkinson disease (PD) include decreased visual acuity, decreased contrast sensitivity, prolonged visual evoked potential latency, hallucinations, altered color perception, and electroretinographic changes.^e26

A systematic review of 13 case-control studies totaling 644 eyes of patients with PD and 604 eyes of healthy controls was recently performed.^e27 The meta-analysis found a higher prevalence of RNFL thinning at the temporal quadrant compared to other quadrants. Furthermore, another study found inner retinal layers (IRL = RNFL + GCL + IPL) thinning in patients with PD and decreased contrast sensitivity compared to healthy controls.^e28 A correlation between IRL thinning and decreased contrast sensitivity in healthy controls was also found. Investigators enrolled 129 patients with PD and 129 healthy controls in an observational prospective study and found that GCL thickness was inversely correlated with disease duration and severity.^e29 OCT may have a role in identifying patients with PD early in the disease or in those patients who present with a paucity of symptoms or atypical clinical manifestations.

Amyotrophic lateral sclerosis

Researchers found no difference in measures of RNFL, GCL, INL, and ONL between 76 patients with amyotrophic lateral sclerosis (ALS) and healthy controls.^e30 In a study of 24 patients with ALS, there were significantly lower measures of average RNFL, INL, and macular thickness compared to controls.^e31 While these findings are highly suggestive of nonmotor involvement and visual pathway involvement in ALS, further studies are needed to clarify the structural changes and determine the clinical relevance.

Migraine

RNFL measurements have been shown to be thinner in migraineurs compared to controls.^e32 Measurements of choroidal thickness were found to be increased during an acute attack in migraine patients when compared to baseline level in the same patient and when compared to a control group.^e33 Further studies are needed to determine the significance and clinical utility of these findings.