Topographical Correlation Between
Disc Haemorrhage and Retinal Nerve
Fibre Layer Defect

We demonstrated that disc haemorrhages were significantly associated with localised wedge-shaped defects of the retinal nerve fibre layer in patients with NTG.¹⁰ Disc haemorrhage has also been described to be correlated with retinal nerve fibre layer defects in location,^11,12 and disc haemorrhages have been observed in some eyes before clinically detectable localised retinal nerve fibre layer defects were noted.¹¹ However, no attempt has been made to clarify the exact topograph-ical relationship between disc haemorrhages and retinal nerve fibre layer defects, either by SLO, clearly visualising retinal nerve fibre layer defects, or in a prospective manner enrolling patients who developed new disc haemorrhage during the follow-up period. Using SLO, one of the study group determined the topographic correlation between disc haemorrhages and retinal nerve fibre layer defects in patients with NTG.¹³

42 patients with NTG who developed new disc haemorrhages were pro-spectively enrolled. We detected 64 disc haemorrhages in 48 eyes of 42 patients; retinal nerve fibre layer defects were demonstrated in 47 of the 48 eyes (97.9%) by SLO. Of the 64 disc haemorrhages, 51 (79.7%) coincided with retinal nerve fibre layer defects in location. These 51 disc haemorrhages were classified into 3 types by site:

• 21 disc haemorrhages (41.2%) were present on the border between the retinal nerve fibre layer defect and the apparently healthy-looking retinal nerve fibre layer (figure 1)
  
  
• 12 disc haemorrhages (23.5%) were present in the retinal nerve fibre layer defect adjacent to the border (figure 4a)
  
  
• 18 disc haemorrhages (35.3%) were pre-sent in the apparently healthy-looking retinal nerve fibre layer adjacent to the border (figure 4b).

Figure 4a. Disc haemorrhage located in the
retinal nerve fibre layer defect adjacent to
the border.¹³

Figure 4b. Disc haemorrhage located in the
apparently healthy-looking retinal nerve fibre
layer adjacent to the border.¹³

This study clearly shows that approximately 80% of disc haemorrhages in patients with NTG were located in the vicinity of the border between the retinal nerve fibre layer defects and the apparently healthy-looking retinal nerve fibre layer. The clinical significance of disc haemorrhage in NTG is that the haemorrhage may indicate the presence of retinal nerve fibre layer defect, even in patients where such a defect may be difficult to discern by ophthalmoscopy. Airaksinen et al. showed that disc haemorrhages often precede the development of a localised retinal nerve fibre layer defect.¹¹ If disc haemorrhage is a precursor to the further development of a localised retinal nerve fibre layer defect, the disc haemorrhage should be located very close to the border between the retinal nerve fibre layer defect and the relatively healthy-looking retinal nerve fibre layer. Our results support this notion. The close topographic association of disc haemorrhage with a retinal nerve fibre layer defect may indicate that both share a common pathogenic factor or that one may be responsible for the other.

Disc Haemorrhage and Peripapillary
Atrophy

Peripapillary atrophy or crescent has been reported to be more frequent and larger in patients with glaucoma than in healthy eyes.^14,15 Buus and Anderson reported that the extent and prevalence of peripapillary atrophy was significantly greater in NTG than in ocular hypertension.⁷ Geijssen found that the extent of peripapillary atrophy in NTG was significantly greater than in high tension glaucoma.¹⁶ These studies indicate that the greatest prevalence and extent of peripapillary atrophy are found in NTG. Our previous study demonstrated that the size of peripapillary atrophy significantly correlated with visual field defects and optic disc configuration in patients with NTG.¹⁷ We investigated the prevalence and the size of peripapillary atrophy in eyes of patients with NTG with and without a history of disc haemorrhages.¹³

40 patients with NTG who developed new disc haemorrhages were prospectively enrolled. 51 randomly selected age-matched patients with NTG without a history of disc haemorrhage were examined. There were no significant differences between the 2 groups in terms of gender, age, refractive errors in spherical equivalent, global indices of visual field (MD and CPSD), follow-up period, and the maximum, minimum, and variation of IOP in a 24-hour diurnal profile without medication (table 5).

Table 5. Patients' characteristics (mean ± SD)¹³

One of the study group determined the prevalence of peripapillary atrophy in NTG eyes with and without a history of disc haemorrhages. Using a confocal scanning laser tomograph (Heidelberg Retina Tomograph®, Heidelberg Engineering, GmBH, Germany), one of the study group measured the area, angular extent, and radial extent of zone Beta of the peripapillary atrophy (figures 5a and b).

Figure 5a. Peripapillary atrophy (zone Beta) was measured
by drawing a new contour line around the zone using an image
produced by a confocal scanning laser tomograph. The broken
lines were automatically drawn by a computer for determining
the angular extent and the largest radial extent.¹³

Figure 5b. The parameters used for detailed measurements
of peripapillary atrophy (zone Beta) were (a) area (mm²),
(b) angular extent (°), and (c) largest radial extent (mm).¹³

In the group with disc haemorrhage, 39 of 40 eyes (97.5%) showed peripapillary atrophy (zone Beta), while 42 of 51 eyes (82.4%) in the group without haemorrhage had peripapillary atrophy. The difference was statistically significant (p = 0.0385, Fisher's exact test). Among the parameters of peripapillary atrophy, the area, the angular extent, and the ratio of the areas of peripapillary atrophy to disc were significantly larger in the group with disc haemorrhage than in the group without haemorrhage (table 6).

Table 6. Disc haemorrhage and peripapillary parameters (mean ± SD)¹³

Thus, disc haemorrhage is closely associated with the prevalence and size of peripapillary atrophy in patients with NTG.

The peripapillary atrophy (zone Beta) histologically corresponded to the absence of retinal pigment epithelium and the thinning or absence of choriocapil-laries adjacent to the disc.¹⁸ Using fluorescent angiography, zone Beta did not fluoresce in the choroidal filling phase.¹⁹ Since the prelaminar and laminar portions of the optic nerve head receive their main blood supply from the peripapil- lary choroid via branches of the short posterior ciliary arteries,²⁰ the absence or obliteration of centripetal branches in the area of peripapillary atrophy may result in a vascular insufficiency in that segment of the optic nerve head. This study indicates that disc haemorrhage and peripapillary atrophy have a significant association. Hence, we could not rule out the possibility that disc haemorrhage may occur as a consequence of micro-vascular ischaemia of the optic disc.

Conclusions

Our studies reveal that disc haemorrhage is a distinct risk factor for visual field deterioration, and is closely associated with localised defects of the retinal nerve fibre layer both in frequency and location in NTG. Disc haemorrhage is considered to be a sign of an active process at the optic disc, which may lead to development of a retinal nerve fibre layer defect over time and therefore should be regarded as an indicator of possibly progressive glaucomatous damage in NTG.