To evaluate the clinical utility of spherocylindrical automated refraction (AR) compared with subjective manifest refraction (MR) after cataract surgery with implantation of ReZoom refractive multifocal intraocular lenses (IOLs) (Advanced Medical Optics, Inc.). Private Practice. This prospective nonrandomized interventional study consisted of 72 patients with bilateral cataract and a potential visual acuity of at least 20/40. Patients had bilateral nonsimultaneous cataract surgery and implantation of a ReZoom IOL. Manifest refraction was performed in all patients followed by 3 consecutive measurements using the Topcon KR-8000 autorefractor with nondilated and dilated pupils. Assessment of repeatability of multiple consecutive ARs and comparison of the AR and MR using power vector analysis were performed at the 6-month follow-up. The main outcome measures were the correlation between AR and MR with sphere, spherical equivalent, and astigmatism as well as the repeatability of AR before and after dilation with phenylephrine 10%. Repeatability analysis showed that the initial nondilated AR was not significantly different from the mean of the 3 refractions for nondilated and dilated measurements. The mean difference between the initial AR and the MR was -0.84 +/- 0.62 diopters (D) for sphere (SD), -1.00 +/- 0.61 D for spherical equivalent, and -0.06 +/- 0.19 D and -0.01 +/- 0.17 D, respectively, for J(0) and J(45), the 2 components of astigmatism. Linear regression of AR versus MR data showed poor correlation for sphere (R(2) = 0.4852) and spherical equivalent (R(2) = 0.5529), whereas the correlation for the 2 astigmatic components of vector analysis was excellent (J(0), R(2) = 0.8881; J(45), R(2) = 0.8640). Correlation was better when the MR residual refractive defect was higher. Although autorefraction showed excellent agreement with subjective refractive astigmatism, correlation with spherical values was poor, with a trend toward more negative values. Autorefraction after ReZoom IOL implantation can be used as a good starting point for subjective refraction of astigmatism; however, spherical values should be underestimated.