Purpose
To determine the effect of head tilt on retinal nerve fiber layer(RNFL) thickness measurements using the Cirrus spectral-domain optical coherence tomography(OCT) compared to the Spectralis OCT in healthy eyes. Our secondary objective was to assess the repeatability of cup disc parameters(CDP) using Cirrus OCT.
Methods
This is a prospective, observational study in healthy patients with no evidence of glaucoma. A single randomly selected eye was scanned undilated in 3 positions (neutral, 30° right and left tilt). 28 eyes were used for RNFL analysis while 29 were used for CDP analysis. RNFL thickness in each quadrant, disc area, rim area, cup volume, vertical cup-to-disc ratio(CDR) and average CDR measurements were recorded. The paired t-test was used to analyze effect on RNFL thickness. Coefficient of variation(COV) and intraclass correltion coefficient(ICC) were used to analysis the repeatability of CDP.
Results
In Cirrus OCT RNFL analysis, ipsilateral head tilt resulted in superior thinning and temporal thickening, while contralateral head tilt resulted in inferior thinning, nasal thickening as well as temporal thickening (all p <0.05). Head tilt did not cause any significant change in RNFL thickness with Spectralis OCT (p>0.05). In Cirrus OCT CDP measurements, analysis of variance did not show statistically significant differences between 3 scans taken a single position. There was good agreement between measurements (ICC 0.919-0.996), with COV 1.94% - 5.48%. Taking head tilt into account, we compared single scans taken in each position and there was good agreement as well (ICC 0.888 - 0.996), with COV 2.04% - 5.39%. The agreement was found to be highest for cup volume but lowest for rim area.
Conclusions
Head tilt affects RNFL thickness measurements for the Cirrus OCT but not the Spectralis OCT. Spectralis OCT features an alignment technology that tracks and aligns circle scans to overcome measurement errors due to head tilt. A new alignment overlay tool has been developed by Cirrus OCT but remains to be extensively analysed. Although the extent of head tilt was large for the purposes of this study, artefacts caused by minor head tilt should be considered in the interpretation of Cirrus OCT RNFL results. In addition, Cirrus OCT CDP were found to have good repeatability between multiple measurements, even in the presence of head tilt.
Keywords: 610 nerve fiber layer •
552 imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) •
629 optic nerve