Mesopic conditions elicit both rod and cone responses, and they are more commonly encountered in daily life than are scotopic conditions; yet visual function outcome measures of mesopic visual acuity (VA) or contrast sensitivity (CS) are rarely evaluated. In retinitis pigmentosa (RP), we explored whether visual reductions in mesopic versus photopic conditions were correlated with cone or rod function, as well as the between-visit test-retest variability in mesopic measures. At each of two

more »

... , 22 RP subjects completed mesopic and photopic ETDRS VA and Pelli-Robson chart CS tests obtained with and without a U23 NoIR 4% transmission filter; testing of perifoveal scotopic cone or rod sensitivity with the AdaptDx; and the Rabin Cone Contrast Test (CCT). A greater CS reduction in mesopic versus photopic conditions was significantly related to absence of scotopic rod function (p = 0.038) or longer self-reported duration of night vision loss (p = 0.044). VA reductions >0.2 logMAR in mesopic versus photopic conditions were significantly related to reduced cone-mediated scotopic sensitivity (p = 0.038). Significant predictors of the CCT ratio of S-cone to M- and L-cone sensitivity were mesopic VA (p = 0.038) and absence of AdaptDx rod function (p = 0.008). Test-retest 95% coefficients of repeatability were not significantly different when comparing between photopic and mesopic tests of VA (0.16 and 0.12 logMAR, respectively) or CS (0.21 and 0.24 logCS, respectively). Perifoveal scotopic rod and cone function measured with the AdaptDx was significantly correlated with mesopic CS and VA, respectively, which had good, acceptable test-retest repeatability; thus, they appear to be suitable outcome measures to monitor mesopic visual function in clinical practice or trials. RP subjects with reduced mesopic VA and no perifoveal rod function had a greater loss of sensitivity for S-cones than for L-/M-cones.