Background: The aim of this study was to determine the intra-session repeatability and inter-examiner reproducibility of the colour perimetry technique when assessing in vivo macular pigment optical density in age-related macular degeneration patients. Methods: Age-related macular degeneration patients were classified into four groups: early age-related macular degeneration, intermediate age-related macular degeneration, atrophic age-related macular degeneration and neovascular age-related macular degeneration after undergoing fundus photography (TRC 50DX type IA) and spectral-domain optical coherence tomography analysis (Topcon 3D-2000). Central fixation was confirmed in all patients using the MP-1 microperimeter (Nidek, Padua, Italy). To analyse repeatability, one examiner obtained three consecutive macular pigment optical density measures with MonCV3 device (Metrovision, Perenchies, France). To study agreement between two observers, a second examiner performed another macular pigment optical density measurement in random order. Within-subject standard deviation, coefficient of variation, and intraclass correlation coefficient data were obtained. Results: Fifty two (32 females and 20 males) consecutive age-related macular degeneration patients having a mean age of 71.5 ± 8.2 years were recruited. Six had early age-related macular degeneration, 25 had intermediate age-related macular degeneration, 10 had atrophic age-related macular degeneration and 11 had neovascular age-related macular degeneration. For repeatability, coefficient of variation values ranged from 22.3% (neovascular age-related macular degeneration) to 41.0% (atrophic age-related macular degeneration) and intraclass correlation coefficient values from 0.52 (intermediate age-related macular degeneration) to 0.79 (neovascular age-related macular degeneration). For agreement between two examiners, coefficient of variation values ranged from 20.1% (intermediate age-related macular degeneration) to 37.8% (neovascular age-related macular degeneration) and intraclass correlation coefficient values from 0.61 (neovascular age-related macular degeneration) to 0.80 (atrophic age-related macular degeneration). Conclusion: The reliability (intra-session repeatability and inter-examiner reproducibility) of colour perimetry technique to assess macular pigment optical density in age-related macular degeneration patients is only moderate. Thus, it cannot be recommended to be performed when evaluating and monitoring age-related macular degeneration patients in the daily clinic.