Background/aims Children and adults with neurological impairments are often not able to access conventional perimetry; however, information about the visual field is valuable. A new technology, saccadic vector optokinetic perimetry (SVOP), may have improved accessibility, but its accuracy has not been evaluated. We aimed to explore accessibility, testability and accuracy of SVOP in children with neurodisability or isolated visual pathway deficits.

Methods Cohort study; recruitment October 2013–May 2014, at children's eye clinics at a tertiary referral centre and a regional Child Development Centre; full orthoptic assessment, SVOP (central 30° of the visual field) and confrontation visual fields (CVF). Group 1: age 1–16 years, neurodisability (n=16), group 2: age 10–16 years, confirmed or suspected visual field defect (n=21); group 2 also completed Goldmann visual field testing (GVFT).

Results Group 1: testability with a full 40-point test protocol is 12.5%; with reduced test protocols, testability is 100%, but plots may be clinically meaningless. Children (44%) and parents/carers (62.5%) find the test easy. SVOP and CVF agree in 50%. Group 2: testability is 62% for the 40-point protocol, and 90.5% for reduced protocols. Corneal changes in childhood glaucoma interfere with SVOP testing. All children and parents/carers find SVOP easy. Overall agreement with GVFT is 64.7%.

Conclusions While SVOP is highly accessible to children, many cannot complete a full 40-point test. Agreement with current standard tests is moderate to poor. Abnormal saccades cause an apparent non-specific visual field defect. In children with glaucoma or nystagmus SVOP calibration often fails.