Heterogeneous pattern of retinal nerve fiber layer in multiple sclerosis. High resolution optical coherence tomography: potential and limitations

Abstract

Background: Recently the reduction of the retinal nerve fibre layer (RNFL) was suggested to be associated with diffuse axonal damage in the whole CNS of multiple sclerosis (MS) patients. However, several points are still under discussion. (1) Is high resolution optical coherence tomography (OCT) required to detect the partly very subtle RNFL changes seen in MS patients? (2) Can a reduction of RNFL be detected in all MS patients, even in early disease courses and in all MS subtypes? (3) Does an optic neuritis (ON) or focal lesions along the visual pathways, which are both very common in MS, limit the predication of diffuse axonal degeneration in the whole CNS? The purpose of our study was to determine the baseline characteristics of clinical definite relapsing-remitting (RRMS) and secondary progressive (SPMS) MS patients with high resolution OCT technique.

Methodology: Forty-two RRMS and 17 SPMS patients with and without history of uni- or bilateral ON, and 59 age- and sex-matched healthy controls were analysed prospectively with the high resolution spectral-domain OCT device (SD-OCT) using the Spectralis 3.5mm circle scan protocol with locked reference images and eye tracking mode. Furthermore we performed tests for visual and contrast acuity and sensitivity (ETDRS, Sloan and Pelli-Robson-charts), for color vision (Lanthony D-15), the Humphrey visual field and visual evoked potential testing (VEP).

Principal findings: All 4 groups (RRMS and SPMS with or without ON) showed significantly reduced RNFL globally, or at least in one of the peripapillary sectors compared to age-/sex-matched healthy controls. In patients with previous ON additional RNFL reduction was found. However, in many RRMS patients the RNFL was found within normal range. We found no correlation between RNFL reduction and disease duration (range 9-540 months).

Conclusions: RNFL baseline characteristics of RRMS and SPMS are heterogeneous (range from normal to markedly reduced levels).

Figure 1. Comparison of the high resolution SD-OCT and standard TD-OCT.

Figure 1 shows high resolution SD-OCT scans (left column) and standard TD-OCT scans (right column) of 2 MS patients. Patient 1, 34 year old female, RRMS, disease duration 15years, prior ON (left eye, 2 years after disease onset). Patient 2, SPMS, 47 year old male, SPMS, disease duration 20 years, no prior ON. Measurements of RNFL were comparable in patient 1 but differed about 5.3 µm (OD, right eye) and 9.8 µm (OS, left eye) in patient 2 (red arrows, lower row). Baseline/reference scans of standard TD-OCT are averaged out of 3 sequential scans (asterisks) with partly substantial differences of each single scan. Low scanning speed and prolonged examination time of standard TD-OCT can often not compensate low image resolution despite good centering around the optic nerve as a result of poor fixation and motion artefacts. In contrast, with high resolution SD-OCT reliable and properly centered baseline and follow up scans can be achieved with ultrahigh speed scanning and an eye tracking mode. Moreover, with high resolution SD-OCT previous scan locations can be identified and guide the laser beam to identical scan positions repeatedly.

Figure 2. Global reduction of RNFL.

*p<0.05; **p<0.005; ***p<0.0001.

Figure 3. Segmental reduction of RNFL.

*p<0.05; **p<0.005; ***p<0.0001.