RT Journal Article SR Electronic T1 Estimation of ocular volume from axial length JF British Journal of Ophthalmology JO Br J Ophthalmol FD BMJ Publishing Group Ltd. SP 1697 OP 1701 DO 10.1136/bjophthalmol-2013-304652 VO 98 IS 12 A1 Nagra, Manbir A1 Gilmartin, Bernard A1 Logan, Nicola S YR 2014 UL http://bjo.bmj.com/content/98/12/1697.abstract AB Background/aims To determine which biometric parameters provide optimum predictive power for ocular volume. Methods Sixty-seven adult subjects were scanned with a Siemens 3-T MRI scanner. Mean spherical error (MSE) (D) was measured with a Shin-Nippon autorefractor and a Zeiss IOLMaster used to measure (mm) axial length (AL), anterior chamber depth (ACD) and corneal radius (CR). Total ocular volume (TOV) was calculated from T2-weighted MRIs (voxel size 1.0 mm3) using an automatic voxel counting and shading algorithm. Each MR slice was subsequently edited manually in the axial, sagittal and coronal plane, the latter enabling location of the posterior pole of the crystalline lens and partitioning of TOV into anterior (AV) and posterior volume (PV) regions. Results Mean values (±SD) for MSE (D), AL (mm), ACD (mm) and CR (mm) were −2.62±3.83, 24.51±1.47, 3.55±0.34 and 7.75±0.28, respectively. Mean values (±SD) for TOV, AV and PV (mm3) were 8168.21±1141.86, 1099.40±139.24 and 7068.82±1134.05, respectively. TOV showed significant correlation with MSE, AL, PV (all p<0.001), CR (p=0.043) and ACD (p=0.024). Bar CR, the correlations were shown to be wholly attributable to variation in PV. Multiple linear regression indicated that the combination of AL and CR provided optimum R2 values of 79.4% for TOV. Conclusion Clinically useful estimations of ocular volume can be obtained from measurement of AL and CR.