Visual Field Loss in Patients with Glaucoma Who Have Asymmetric Peripapillary Focal Arteriolar Narrowing

CLINICAL SCIENCES Visual Field Loss in Patients With Glaucoma Who Have Asymmetric Peripapillary Focal Arteriolar Narrowing

Andrew Lam, MD; Vatinee Bunya, MD; Jody R. Piltz-Seymour, MD

Objective: To evaluate the relationship between peri- eyes with peripapillary focal arteriolar narrowing. The papillary focal arteriolar narrowing and visual field eyes with narrowing exhibited a mean MD of −8.77±8.27 defects. dB and a mean CPSD of 5.01±3.42 dB. Eyes without narrowing displayed a mean MD of −4.52±6.64 dB and a Methods: From our institutional practice, we identi- mean CPSD of 3.01±2.68 dB (P=.003 for both). There fied 31 patients with glaucoma who had peripapillary fo- was no significant difference in severity of the visual field cal arteriolar narrowing in only one eye and compared defect between eyes with mild and severe arteriolar nar- visual field data between the two eyes. Mean deviation rowing. (MD) and corrected pattern standard deviation (CPSD) were recorded using Humphrey visual field testing at the Conclusion: To our knowledge, this is the first study time proximal narrowing was apparent on the fundus pho- to show that the presence of peripapillary focal arterio- tograph. Visual field data from subsets of patients with lar narrowing is related to the severity of visual field loss mild and severe narrowing were also compared. in patients with glaucoma.

Results: The MD and CPSD were significantly worse in Arch Ophthalmol. 2002;120:1494-1497

N 1994, Rader et al1 described the illary vessels in the hemidisc correspond- focal narrowing of retinal ves- ing to this defect. These findings indicate sels adjacent to the optic disc in that peripapillary focal arteriolar narrow- eyes with glaucoma. They re- ing may be related to the severity of glau- ported that these proximal con- coma. Istrictions were present in 42% of eyes with In this study, we evaluated the rela- glaucoma but in only 5% that were nor- tionship between the presence of proximal. In glaucomatous eyes, the location mal narrowing and the severity of visual of these arteriolar narrowings corre- field defects. We selected patients who ex- sponded to the sector of the disc, with the hibited peripapillary focal arteriolar nar- greatest cupping in 91% of cases and to rowing in only one eye and compared that areas of absent retinal pigment epithe- eye’s visual field information with its fel- lium in 87% of cases (␤-peripapillary low eye. We also hoped to determine atrophy). Peripapillary focal arteriolar nar- whether the degree of narrowing corre- rowing was not exclusive to eyes with glau- lated with exacerbation of the visual field comatous optic nerve damage; 68% of eyes defect. with nonarteritic anterior ischemic optic neuropathy also had proximal narrow- METHODS ing. Papastathopoulos and Jonas2 con- firmed that this finding represents a true We examined the fundus photographs of 325 stenosis of the vessel lumen and is not an consecutive patients seen in a university- ophthalmoscopic artifact. Rankin and based glaucoma referral practice. All photo- Drance3 noted that the location of the peri- graphs were obtained from the medical rec- papillary narrowing correlated with the ords of patients at the Scheie Eye Institute in Philadelphia, Pa. For this review, written con- presence of a visual field defect in the cor- sent for the release of medical records was ob- From the Scheie Eye Institute, responding superior or inferior hemi- tained from all patients. Photographs were ex- University of Pennsylvania field in 89% of cases. Moreover, in eyes amined for any evidence of focal arteriolar School of Medicine, with worse visual field loss in 1 hemi- narrowing in the area spanning the periph- Philadelphia. field, Hall et al4 found narrower peripap- eral optic disc rim and half the diameter of the

(REPRINTED) ARCH OPHTHALMOL / VOL 120, NOV 2002 WWW.ARCHOPHTHALMOL.COM 1494

–5

–10

–15 Mean Deviation, dB

Figure 1. Peripapillary focal arteriolar narrowing at the 5-o’clock position in the left eye of a 67-year-old man with primary open-angle glaucoma. –20

–25 Table 1. Comparison of Mean Deviation and Corrected Pattern Standard Deviation in Eyes With and Without Peripapillary Focal Arteriolar Narrowing* –30 Eyes With Narrowing Eyes Without Narrowing Eyes With Eyes Without P Narrowing Narrowing Value† Figure 2. Box plot of mean deviation scores in eyes with and without peripapillary narrowing. The line within the box, the limits of the box, and the Mean deviation, dB −8.77 ± 8.3 −4.53 ± 6.6 .003 upper and lower bars indicate the mean, 25th to 75th percentiles, and 10th Corrected pattern standard 5.01 ± 3.4 3.02 ± 2.7 .003 to 90th percentiles, respectively. deviation, dB teen of these patients had constrictions graded 1, whereas *Data are presented as mean ± SD. †Determined using a paired 2-tailed t test. the remaining 18 had constrictions graded 2 to 4. In all patients, arteriolar narrowing was static, or present in multiple photographs across time in the same location. Fur- disc away from the rim. A focal narrowing was defined as an arterial segment that appeared narrower than the same artery thermore, the degree of stenosis either remained con- immediately distal and proximal to the narrowing (Figure 1). stant or grew worse in all cases (ie, we did not observe The position of any focal narrowing was described using clock any reversibility in the arteriolar constrictions). hours relative to the disc, and the degree of constriction was Eyes with proximal narrowing had significantly graded on a scale of 0 to 4. A score of 1 denoted mild narrow- greater visual field defects than contralateral eyes with- ing, and 4 indicated very severe narrowing. The presence and out proximal narrowing (Table 1). The mean MD and grading of narrowings were agreed on by 2 investigators (V.B. CPSD were significantly worse in eyes with narrowing and J.R.P.-S.). than in eyes without narrowing (mean difference in MD, We subsequently identified patients with primary open- 4.26; P=.003; mean difference in CPSD, 1.99; P=.003). angle glaucoma who had peripapillary focal arteriolar narrow- Box plots for MD and CPSD illustrating the data spread ing in only one eye. Visual field information including mean deviation (MD) and corrected pattern standard deviation (CPSD) are shown in Figure 2 and Figure 3. was collected and compared between eyes with narrowing and To determine whether the degree of arteriolar con- their fellow eyes without narrowing. Data were recorded us- striction influences severity of the visual field defect, we ing visual field tests at the time focal narrowing was apparent compared the subsets of patients with constrictions graded on the fundus photograph. When focal constrictions were pres- 1 (mild) and those with constrictions graded 2 to 4 (more ent in multiple photographs, we used visual field test results severe). No significant differences in MD or CPSD were concurrent with the most recent photograph. In most cases, noted between eyes with mild and severe narrowing visual field tests were done on the same date that fundus pho- (Table 2). In addition, no significant difference be- tographs displayed proximal narrowing. In all cases, visual field tween mild and more severe constrictions was detected tests were performed within 3 months of a fundus photograph when evaluating MD and CPSD in eyes with narrowing showing arteriolar constriction. To investigate whether the degree of constriction affected the visual field defect, the data for and their fellow eyes without narrowing (Table 3). eyes with focal constrictions graded 2 or worse (severe) were Between the groups of eyes with and without nar- compared with data from those with constrictions graded 1 rowing, there were no notable differences in ocular medi- (mild). Statistical analysis was conducted using both paired and cations. More specifically, 21 of the 31 patients received unpaired 2-tailed t tests. identical medications in both eyes. Of the remaining 10 patients, 5 were given 1 additional medication in the eye RESULTS with narrowing, and 5 received 1 additional medication in the unaffected eye. Similarly, there was no statisti- Eighty-seven patients were identified who had primary cally significant difference in intraocular pressure be- open-angle glaucoma with good-quality photographs dem- tween eyes with and without narrowing. The mean ± SD onstrating peripapillary focal arteriolar narrowing. Of these intraocular pressure from the last 3 patient visits in the 87 patients, 31 had focal narrowing in only one eye. Thir- group of eyes with narrowing was 15.8±4.8 mm Hg vs

(REPRINTED) ARCH OPHTHALMOL / VOL 120, NOV 2002 WWW.ARCHOPHTHALMOL.COM 1495

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 12 Corrected Pattern Standard Deviation Table 2. Comparison of Mean Deviation and Corrected Pattern Standard Deviation in Eyes With Severe Peripapillary Focal Arteriolar Narrowing and 10 Eyes With Mild Narrowing*

Eyes With Eyes With Mild Severe 8 Narrowing† Narrowing‡ P (n = 13) (n = 18) Value§ Mean deviation, dB −8.58 ± 8.0 −8.91 ± 8.7 .92 6 Corrected pattern standard 6.00 ± 3.4 4.29 ± 3.4 .18 deviation, dB

4 *Data are presented as mean ± SD unless otherwise indicated. †Indicates constrictions graded 1. ‡Indicates constrictions graded 2 to 4. Corrected Pattern Standard Deviation, dB §Determined using an independent 2-tailed t test. 2

Table 3. Mean Differences in MD and CPSD Between 0 Eyes With Peripapillary Focal Arteriolar Narrowing and Eyes With Narrowing Eyes Without Narrowing Fellow Eyes Without Narrowing: Comparison of Eyes With Severe vs Mild Narrowing* Figure 3. Box plot of corrected pattern standard deviation scores in eyes with and without peripapillary narrowing. The line within the box, the limits of the box, and the upper and lower bars indicate the mean, 25th to 75th Eyes With percentiles, and 10th to 90th percentiles, respectively. Eyes With Mild Severe Narrowing† Narrowing‡ P (n = 13) (n = 18) Value§ 15.8±4.2 mm Hg in the unaffected eyes (P=.97). Eight Mean difference in MD, dB 3.45 ± 6.10 4.85 ± 8.06 .61 eyes with arteriolar narrowing had phakic lenses sec- Mean difference in CPSD, dB 3.03 ± 4.00 1.23 ± 3.00 .16 ondary to cataract extraction, and 4 eyes without narrowing had phakic lenses. Five eyes with narrowing ex- *Data are presented as mean ± SD unless otherwise indicated. MD hibited some degree of current lens opacification, indicates mean deviation; CPSD, corrected pattern standard deviation. †Indicates constrictions graded 1. specifically cataracts or nodular sclerosis, compared with ‡Indicates constrictions graded 2 to 4. 6 eyes without narrowing. §Determined using an independent 2-tailed t test.

COMMENT location of peripapillary atrophy and that of glaucoma- In subjects with peripapillary focal arteriolar narrowing tous visual field defects,8,9 so it was not surprising when of the retinal vessels in only one eye, we found that eyes Rankin and Drance found a similar correlation between with narrowing had significantly greater visual field de- the position of peripapillary narrowings and the pres- fects, characterized by worse MD and CPSD, compared ence of visual field defects in the corresponding hemi- with fellow eyes without narrowing. We have also shown field. Although proximal constriction has been related that the degree of constriction does not appear to influ- to the presence of visual field defects and optic nerve at- ence severity of the visual field defect. rophy, to our knowledge, our study is the first to relate This study extends the research that has been done peripapillary narrowing to the severity of disease. in the area of peripapillary retinal vasculature in pa- Papastathopoulos and Jonas10 proposed that peri- tients with glaucoma. In 1989, Jonas et al5 showed that papillary focal arteriolar narrowing could signal the pro- the caliber of peripapillary arterioles and veins was sig- gression of glaucomatous disease. In their study, the nar- nificantly smaller in patients with glaucoma compared rowness of constriction increased significantly in patients with healthy controls. More recently, Hall et al4 pub- with progressive glaucomatous optic neuropathy, whereas lished research showing that eyes with primary open- the narrowings remained stable in patients with nonpro- angle glaucoma display a correlation between decreased gressive glaucoma. Although we did not find that the de- peripapillary arteriole diameter and visual field defects gree of constriction correlated with worse visual field de- in the corresponding hemifield. fects, the presence of proximal narrowing may indeed be In addition, several correlations have been demon- a marker of disease progression. strated between reduced retinal vessel caliber and vari- It remains unclear whether peripapillary narrow- ous morphologic changes indicative of glaucomatous dam- ings occur secondary to glaucomatous damage or repre- age, including increased cup-disc ratio, diminution of the sent a primary pathologic finding that plays a causative neuroretinal rim area, and greater peripapillary atro- role in the disease. One theory proposes that arteriolar phy.6,7 In 1996, a study by Rankin and Drance3 found that narrowing occurs in response to axon loss from the nerve peripapillary focal arteriolar narrowing corresponded to fiber layer secondary to optic atrophy. Such narrowing an area of peripapillary atrophy in 72.2% of eyes. In pre- would be a natural autoregulatory response to the di- vious studies, a correlation had been shown between the minished need for blood flow.1,11-14

(REPRINTED) ARCH OPHTHALMOL / VOL 120, NOV 2002 WWW.ARCHOPHTHALMOL.COM 1496

©2002 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/29/2021 On the other hand, focal arteriolar narrowing of reti- secondary to glaucomatous damage. Only then can the signal vessels may play a causative role in the etiology of nificance of this clinical finding be fully understood. glaucomatous optic neuropathy by initiating ischemia. Optic nerve blood flow in glaucoma is an active area of Submitted for publication January 31, 2002; final revision research, and numerous studies using different measure- received May 20, 2002; accepted June 25, 2002. ment techniques have shown diminished blood flow in Supported by an unrestricted grant from Research to glaucomatous eyes.15-19 A recent study found that pri- Prevent Blindness, New York, NY. mary open-angle glaucoma suspects without visual field Corresponding author and reprints: Jody R. Piltz- defects showed similar decreases in optic nerve blood flow Seymour, MD, Scheie Eye Institute, 51 N 39th St, Philadel- to those in patients with advanced glaucoma, suggest- phia, PA 19104 (e-mail: [email protected]). ing that reduction in blood flow may play an etiologic role.20 Interestingly, Rankin and Drance3 postulated that REFERENCES the absence of tight junctions in areas of peripapillary atrophy could allow leakage of vasoactive substances that 1. Rader J, Feuer WJ, Anderson DR. Peripapillary vasoconstriction in the glauco- mas and the anterior ischemic optic neuropathies. Am J Ophthalmol. 1994;117: contribute to the development of proximal arteriolar con- 72-80. strictions. Areas of peripapillary atrophy have been as- 2. Papastathopoulos KI, Jonas JB. Fluorescein angiographic correlation of focal narrowing of retinal arterioles in glaucoma. Br J Ophthalmol. 1998;82:48-50. sociated with an increased prevalence of arteriolar con- 3. Rankin SJA, Drance SM. Peripapillary focal retinal arteriolar narrowing in open 21 strictions and the absence of tight junctions. Finally, it angle glaucoma. J Glaucoma. 1996;5:22-28. is also possible that vessel constriction and visual field 4. Hall JK, Andrews AP, Walker R, Piltz-Seymour JR. Association of retinal vessel caliber and visual field defects in glaucoma. Am J Ophthalmol. 2001;132:855- deterioration occur together as a result of another un- 859. known factor or process. 5. Jonas JB, Nguyen XN, Naumann GO. Parapapillary retinal vessel diameter in nor- Because peripapillary focal arteriolar narrowing is mal and glaucoma eyes: morphometric data. Invest Ophthalmol Vis Sci. 1989; 30:1599-1603. not specific for glaucoma, it has limited usefulness as a 6. Jonas JB, Nguyen XN, Naumann GO. Parapapillary retinal vessel diameter in nor- diagnostic tool; however, it may be useful as a marker of mal and glaucoma eyes: correlations. Invest Ophthalmol Vis Sci. 1989;30:1603- 1611. disease severity. In patients with early glaucoma, proxi- 7. Lee SB, Uhm KB, Hong C. Retinal vessel diameter in normal and primary open- mal narrowing may be an additional harbinger of ad- angle glaucoma. Korean J Ophthalmol. 1998;12:51-59. vancing glaucomatous disease. 8. Anderson DR. Correlation of the peripapillary anatomy with the disc damage and field abnormalities in glaucoma. Doc Ophthalmol. 1983;35:1-9. Our study has some limitations. As with any retro- 9. Heijl A, Samander C. Peripapillary atrophy and glaucomatous visual field de- spective study, accuracy of data collection and bias are fects. Doc Ophthalmol. 1985;42:403-407. potential sources of error. One potential confounding fac- 10. Papastathopoulos KI, Jonas JB. Follow up of focal narrowing of retinal arterioles in glaucoma. Br J Ophthalmol. 1999;83:285-289. tor is the possibility that a fundus photograph was taken 11. Petrig B, Werner EB, Riva CE, Grunwald J. Response of macular capillary blood during an episode of vasospasm and that this constric- flow to changes in intraocular pressure as measured by the blue field simulation 3 technique. Doc Ophthalmol. 1985;42:447. tion might be transient. Rankin and Drance proposed 12. Sinclair SH, Grunwald JE, Riva CE, Braunstein SN, Nichols CW, Schwartz SS. that focal narrowing could be caused by vasospasm and Retinal vascular autoregulation in diabetes mellitus. Ophthalmology. 1982;89: described 1 patient in whom narrowing actually re- 748-750. 13. Robinson F, Riva CE, Grunwald JE, Petrig BL, Sinclair SH. Retinal blood flow au- solved with time. However, these authors and others agree toregulation in response to an acute increase in blood pressure. Invest Ophthal- that the presence and location of arteriolar narrowing are mol Vis Sci. 1986;27:722-726. constant in most patients.10 All of our patients had con- 14. Riva CE, Grunwald JE, Petrig BL. Autoregulation of human retinal blood flow: an investigation with laser Doppler velocimetry. Invest Ophthalmol Vis Sci. 1986; sistent evidence of narrowing on multiple photographs 27:1706-1712. across time. 15. Grunwald JE, Piltz JR, Hariprasad SM, DuPont J. Optic nerve and choroidal cir- culation in glaucoma. Invest Ophthalmol Vis Sci. 1998;39:2329-2336. When comparing the eyes of patients who exhibit 16. Hamard P, Hamard H, Dufaux J, Quesnot S. Optic nerve head blood flow using focal arteriolar narrowing of the retinal vessels in only a laser Doppler velocimeter and haemorheology in primary open angle glau- one eye, we found that eyes with narrowing had signifi- coma and normal pressure glaucoma. Br J Ophthalmol. 1994;78:449-453. 17. Wolf S, Arend O, Sponsel WE, et al. Retinal hemodynamics using scanning laser cantly greater visual field defects. The degree of arterio- ophthalmoscopy and haemorheology in chronic open-angle glaucoma. Ophthal- lar narrowing did not correlate with the severity of vi- mology. 1993;100:1561-1566. sual field defects. By comparing the eyes of a single patient, 18. Nicolela MT, Hnik P, Drance SM. Scanning laser Doppler flowmeter study of retinal and optic disk blood flow in glaucomatous patients. Am J Ophthalmol. 1996; factors such as age, race, sex, systemic medications, and 122:775-783. other diseases are well controlled for. This study relates 19. Rojanapongpun P, Drance SM, Morrison BJ. Ophthalmic artery flow velocity in glaucomatous and normal subjects. Br J Ophthalmol. 1993;77:25-29. peripapillary narrowing of retinal arterioles to the sever- 20. Piltz JR, Grunwald JE, Hariprasad SM, Dupont J. Optic nerve blood flow is di- ity of glaucomatous damage and supports the idea that minished in eyes of primary open-angle glaucoma suspects. Am J Ophthalmol. proximal narrowing could be a marker for worse visual 2001;132:63-69. 21. Anderson DR. Relationship of peripapillary hales and crescents to glaucoma- field status. In the future, it will be important to elucidate tous cupping. In: Krieglstein GK, ed. Glaucoma Update III. Heidelberg, Ger- whether focal arteriolar narrowing occurs primarily or is many: Springer-Verlag; 1987:103-105.

(REPRINTED) ARCH OPHTHALMOL / VOL 120, NOV 2002 WWW.ARCHOPHTHALMOL.COM 1497