Use of high spatial resolution perimetry to identify scotomata not apparent with conventional perimetry in the nasal field of glaucomatous subjects. (Clinical Science).

Aim: To examine whether high spatial resolution perimetry (HSRP) could identify fine scale scotomata which may not be apparent with conventional perimetry. The HSRP was performed in the nasal field, as this location is a recognised site for the early occurrence of glaucomatous defects.

Method: 16 early glaucoma eyes, 17 glaucoma suspect eyes, and 20 age matched healthy control eyes underwent conventional automated perimetry using the 24-2 program of the Humphrey field analyser (HFA) and HSRP. The HSRP was performed in the nasal field by testing 9 x 9 degrees of 100 tested points separated by 1 degree and the results compared with the HFA 24-2 program.

Results: Mean HSRP thresholds were significantly abnormal in the suspect and glaucoma eyes, with elevated levels of asymmetry between the superior and inferior nasal field. Overall, 7/17 (41%) suspect eyes (95% confidence interval 5/17 (29%) to 7/17 (41%)) had nasal scotomata on HSRP, although their HFA 24-2 fields failed to identify any defects. In glaucomatous eyes, 15/16 (94%) eyes had HSRP scotomata (95% CI 14/16 (88%) to 15/16 (94%)). In 12 these coexisted with HFA 24-2 defects at the same location, while in three eyes only HSRP identified scotomata in the nasal field.

Conclusion: HSRP can identify scotoma in glaucomatous eyes in the nasal field which may be missed with the lower spatial resolution of conventional perimetry.

**********

The nasal step is a characteristic glaucomatous field defect elicited in kinetic perimetry, arising as a result of a difference in the sensitivity above and below the horizontal midline. This gives rise to a "step" in the isoptre, a defect which was first reported by Ronne in 1909. (1) Subsequent studies of kinetic perimetry have consistently shown that this defect is an early feature of glaucoma, with reported incidences of nasal steps as isolated manifestations of glaucoma varying from 1% to as much as 50%. (2-6)

In static perimerry, the equivalent of a nasal step is a difference in the sensitivity above and below the horizontal midline.

Caprioli and Spaeth identified peripheral nasal defects in 10% of glaucoma suspects with normal central (<30 degree) fields using static automated perimetry, although their findings were difficult to interpret because of the high false positive rate in their controls. (7) A study by Stewart also reported glaucomatous defects between 30-50 degrees nasally in 11% of glaucoma suspects. (8)

Both studies identified peripheral nasal field defects in the presence of a normal central field when tested with conventional automated perimetry.

However, a major limitation with the use of conventional automated perimetry is that it is insensitive to early glaucomatous damage. Evidence for this is provided by clinical, (9,10) histological, (11) and psychophysical studies (12-18) which have identified glaucomatous abnormalities before field abnormalities can be reliably detected using conventional automated perimetry.

One hypothesis which accounts for the poor sensitivity of conventional perimetry in early glaucoma is the "selective cell death hypothesis." (19)

Quigley and colleagues hypothesised a selective cell death of the larger diameter optic nerve fibres in early glaucoma. (19) Since magnocellular cells are associated with larger mean diameters, the selective cell death hypothesis implies a preferential loss of magnocellular function in early glaucoma. Selective tests of magnocellular function include motion, flicker, and frequency doubling perimetry. These have all identified abnormalities in early glaucoma. (12,13,15,16,18)

Alternatively, the poor sensitivity of conventional perimetry may be a consequence of the non-selective nature of the stimulus used in luminance perimetry, which stimulates a broad spectrum of retinal ganglion cells. The large overlap in ganglion cell receptive fields results in considerable redundancy that may mask early losses if all classes of ganglion cells are stimulated. (20)

An additional factor contributing to the poor sensitivity of conventional perimetry in detecting early glaucoma is its relatively low spatial resolution, with 6 degrees separating adjacent test locations in the 24-2 and 30-2 programs of the Humphrey field analyser (HFA, Zeiss Humphrey Systems, Dublin, CA, USA). Several studies have addressed this problem by performing perimetry at a higher spatial resolution, and have identified scotomata in glaucoma patients beyond the resolution of conventional perimetry. 21-24 These findings suggest that the spatial resolution of conventional perimetry is not high enough to identify early glaucomatous damage.

Previous publications have reported a technique for performing high spatial resolution perimetry (HSRP) in a number of diseases. The technique obtains a threshold plot of 100 HFA test points separated by 1 degree to generate a 10 x 10 grid and allows the field to be sampled at much greater resolution than conventional perimetry. The technique is clinically practical and can be performed on an unmodified HFA.

HSRP in glaucoma has been shown to identify …