Reliability of a commercially available heterochromatic flicker photometer, the MPS2, for measuring the macular pigment optical density of a Japanese population

• Published in: Japanese journal of ophthalmology Vol. 62; no. 4; pp. 473 - 480
• Main Authors: Obana, Akira, Gohto, Yuko, Moriyama, Takatoshi, Seto, Takahiko, Sasano, Hiroyuki, Okazaki, Shigetoshi
• Format: Journal Article
• Language: English
• Published: Tokyo Springer Japan 01.07.2018; Springer Nature B.V
• Subjects: Clinical Investigation; Coefficient of variation; Correlation analysis; Correlation coefficient; Correlation coefficients; Eye diseases; Flicker; Medicine; Medicine & Public Health; Ophthalmology; Optical density; Photometers; Pigments; Population density; Reliability analysis; Systematic errors; Heterochromatic flicker photometry; Macular pigment optical density; MPS2; validation; Asian pigmented eyes
• ISSN: 0021-5155; 1613-2246
• DOI: 10.1007/s10384-018-0588-3

Purpose The macular pigment optical density (MPOD) of a Japanese population was measured using a commercially based heterochromatic flicker photometer, the Macular Pigment Screener (MPS2). The objective of the study was to evaluate the accuracy and test-retest reliability of the MPS2 in Asian pigmented eyes. Study design Experimental study to validate the medical instrument in humans. Methods Twenty-four healthy Japanese participants with no systemic or eye diseases (men: 13, women: 11; mean [SD] age 38.6 [10.9 years]) were included. The concordance of the MPOD, obtained using the MPS2 and Macular Metrics II (MM2), and the test-retest reliability were examined. Results Determination of the MPOD was unsuccessful in 1 participant; thus, the MPOD of 23 participants was analyzed. The mean (SD) MPOD measured with the detail-mode of the MPS2 was 0.63 (0.18) and with that of the MM2, it was 0.72 (0.23). The former was significantly lower than the latter ( P = .003, paired t test). The MPOD measured with the MPS2 and the MM2 showed good concordance ( r = 0.79, P < .001, Pearson product moment correlation). Bland-Altman analyses showed no systematic errors between the MPS2 and the MM2. The intraclass correlation coefficient over 5 measurement times with the detail-mode of the MPS2 was 0.80, and the mean coefficient of variation was 9.4%. Conclusion The high concordance with the MM2 and good test-retest reliability found by this study suggest that the MPS2 is acceptable for use in a Japanese population. However, the mean MPOD yielded by the MPS2 was significantly lower than that yielded by the MM2. Therefore, the MPS2 and MM2 are not interchangeable in a single study.