Test-retest variation in macular thickness measurements with the Heidelberg Retina Tomograph.

Purpose: The aim was to evaluate the test-retest variation of macular thickness measurements using the Heidelberg Retina Tomograph (HRT3) Retina Module and compare it with the Stratus optical coherence tomography (OCT). As no normative data has been published for the HRT Retina Module, a secondary purpose was to obtain reference values for this technique. Methods: Thirty healthy subjects and nine patients with macular oedema were examined. The healthy subjects underwent five examinations by two experienced examiners at two different occasions. The patients with maculopathy were examined twice by one examiner on one occasion. All measurements were made with both the HRT3 and the Stratus OCT and the macular thickness measurements were used for calculations of the test-retest variation. Results: In healthy subjects the coefficient of variation (CV) ranged from 4.5 to 8.8 per cent with the HRT3 and from 0.6 to 1.5 per cent with the Stratus OCT. In the nine patients with macular oedema, the HRT CV ranged from 7.9 to16.6 per cent and for OCT from 1.1 to 2.5 per cent. The intra-class correlation in the central area for intra-observer/intra-visit in healthy subjects was 0.78 with the HRT3 and 0.93 with the Stratus OCT. Corresponding values for the patients with macular oedema were 0.64 with the HRT3 and 1.0 with the Stratus OCT. Mean macular thickness was significantly larger in all areas measured with the HRT3 compared with the Stratus OCT. No significant correlations were found between the thickness values from the different instruments. Conclusion: In this study, macular thickness was measured with two different techniques. The HRT3 showed lower repeatability and reproducibility as well as a higher CV compared with the Stratus OCT in both healthy subjects and patients with macular oedema. This difference might be of clinical relevance, even though the nine patients with maculopathy in the current study showed abnormal results with both methods.