- Email: [email protected]
Intraocular Pressure-denpendent Retinal Vascular Change in Adult Chronic Open-angle Glaucoma Patients
Intraocular Pressuredependent Retinal Vascular Change in Adult Chronic Open-angle Glaucoma Patients DONG H. SHIN, MD, PhD,· CLARK S. TSAI, MD; KYLE A. PARROW, MD; CHAESIK KIM, BSEE; JIM Y. WAN, PhD/ DIAN X. SHI, MD·
Abstract: Intraocular pressure (IOP)-dependent retinal vascular changes were investigated in 33 eyes of 33 adult chronic open-angle glaucoma (COAG) patients by measuring major retinal vascular calibers at the optic disc border before and after mean (± standard deviation) lOP reduction from 35.3 (±7.2) to 16.5 (±4.7) mmHg for 11.2 (±13.5) weeks . Both mean arterial and venous calibers were significantly reduced (P < 0.0001 for each), from 87.8 (±14.2) and 128.3 (±20.8) JLm at high lOP to 82.0 (±13.8) and 121.4 (±1B.5) JLm at low lOP. Arterial and venous caliber decreases correlated positively with magnitude of lOP reduction (r = 0.503, P < 0.01 and r = 0.555, P < 0.001, respectively). While the lOP-dependent retinal arterial caliber change was highly significant in patients 55 years of age or younger (r = 0.636, P < 0.01) and in the overall study group, it was not significant in patients older than 55 years (r = 0,205, P > 0.1). Age seems to be more influential than magnitude of lOP reduction in patients older than 55. Thus, diminished lOP-dependent retinal arterial caliber change in the elderly may be one factor contributing to the higher incidence and prevalence of glaucoma in this population. Ophthalmology 1991; 98: 1087-1092
Autoregulation of blood flow is an intrinsic mechanism for maintaining constant flow despite changes in perfusion pressure,,- 13 Retinal autoregulation appears to be mediated, at least in part, by changes in arterial caliber: the Originally received: May 17, 1990. Revision accepted: March 5, 1991. 1
2
Kresge Eye Institute, Wayne State University, Detroit. Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor.
Presented in part as a poster at the American Academy of Ophthalmology Annual Meeting, New Orleans, Ocl/Nov 1989. Reprint requests to Dong H. Shin, MD, PhD, Kresge Eye Institute, 4717 St. Antoine Blvd, Detroit, MI 48201.
caliber increases as the perfusion pressure falls and decreases as the pressure rises. 2,3,6 Retinal perfusion pressure is defined here as the difference between retinal arterial blood pressure and intraocular pressure (lOP).7 Both retinal arterial caliber changes and age-associated decline of retinal arterial reactivity have been observed in the presence of systemic blood pressure changes. 2 However, there has been a paucity of information regarding the change in the retinal vascular caliber with respect to chronic prolonged alteration of the lOP in glaucoma patients. Therefore, we investigated whether retinal vascular calibers would decrease on lowering of the lOP and what factors might influence lOP-associated retinal vascular caliber changes in chronic open-angle glaucoma (COAG) patients. 1087
OPHTHALMOLOGY
•
JULY 1991
PATIENTS AND METHODS The study group consisted of 18 male and 15 female adult patients with early to moderate COAG. 14 Fourteen (42%) were black and 19 (58%) were white. The mean (± standard deviation) age of the patients was 53.0 (± 15.9) years (range, 23 to 81 years). The lOP, measured by Goldmann applanation tonometry, was reduced by a mean (± standard deviation) of 18.8 mmHg (±9.2 mmHg) from 35.3 mmHg (±7.2 mmHg) (range, 23 to 56 mmHg) to 16.5 mmHg (±4.7 mmHg) (range, 6 to 27 mmHg) for a duration of 11.2 (± 13.5) weeks. The lOP was reduced in 26 eyes with medication alone, in 5 eyes with combined laser trabeculoplasty, and in 2 eyes with filtering surgery. The 2 eyes that underwent filtering surgery had lOP reduced without chronic hypotony for at least 4 months. No known event or intervention was noted that would have caused a substantial change of blood pressure during the study period in any of the patients. Dipivefrin 0.1 % was used for lOP reduction in only 1 of 33 eyes in the study. Six eyes of 6 patients with a mean (± standard deviation) age of54.7 (±14.3) years and a mean (± standard deviation) lOP reduction of44.3% (±3.7%) were on ,a-blockers at only low lOP; 26 eyes of 26 patients with a mean age of 51.5 (± 15.7) years and a mean lOP reduction of 53.4% (±16.5%) were on ,a-blockers at both high and low lOPs; and 1 eye was without a ,a-blocker at both high and low lOPs. Optic disc photographs (15 0 color stereo slides, magnification X5) were taken with a telecentric Zeiss (Thornwood, NY) fundus camera in the 33 adult COAG patients both before and after lOP reduction. The eye with the higher magnitude of lOP reduction was chosen in each patient as long as the photographs were deemed of adequate quality to accurately assess the vascular caliber. Using a Topcon (Paramus, NJ) microfiche projector (X 10 magnification) and a vernier caliper of 0.02 mm calibration, the horizontal disc diameter and the two clearest major retinal arterial and venous calibers (one superior and one inferior) were measured at the optic disc border from fundus photographs taken before and after lOP reduction in a masked fashion. Measurements of the major retinal arterial and venous calibers at the optic disc border of the fundus photographs have been shown to be reliable and accurate by other investigators. 2, 15, 16 Images were taken with the Rodenstock Optic Nerve Head Analyzer on the same day as the optic disc photographs and were corrected for magnification with the axial length, based on Littmann's method. 17 The accuracy of the Rodenstock Analyzer was confirmed with a model eye l8 and in glaucomatous monkey eyes (Yamauchi et ai, unpublished data; presented at 1989 AR VO meeting). A correction factor was then derived by dividing the horizontal optic disc diameter obtained from the Rodenstock analyzer image by the horizontal optic disc diameter obtained from the fundus photograph. The correction factor was necessary to obtain the actual vascular calibers and to reduce possible errors resulting from any magnification difference between the initial and the subsequent photographs. 1088
•
VOLUME 98
•
NUMBER 7
Each vessel was measured five times and an average value was determined for each. Afterward, the mean arterial or venous calibers were computed by averaging the measurements of the two major arteries or veins and then multiplying the average values by the calculated correction factor. The following formula was used: MVC = (lh)(Rd/Pd){(SI
+ S2 + S3 + S4 + S5)/5 + (11 + 12 + 13 + 14 + 15)/5},
where MVC
=
Mean vessel caliber,
Rd
=
Disc diameter by Rodenstock Optic Nerve Head Analyzer,
Pd = Disc diameter by fundus photograph, Rd/Pd
=
Correction factor,
S = Superior vessel caliber, and I
=
Inferior vessel caliber.
Vascular caliber measurements of seven eyes of seven adult subjects also were obtained and corrected by the correction factor as described above from two consecutive fundus photographs and Rodenstock Analyzer images, taken without any intervention to alter lOP or blood pressure, for determination of total variabilities and their 95% confidence intervals. The total variability represents the sum of intraphotographic variability and interphotographic variability. The vascular calibers at the higher lOP were compared with those at the lower lOP using a two-tailed paired t test. Correlations were also sought among lOP reduction, the duration ofIOP reduction, and the patient's age, with the (absolute and percent) retinal arterial caliber changes using simple and multiple linear regression analyses in all 33 patients and in the two age groups (55 years and younger, and older than 55 years). All results of simple linear regression analysis were confirmed with Spearman correlation analysis.
RESULTS A statistically significant age-related decline of arterial caliber was seen at both high and low lOPs (P < 0.02 at initial high lOP and P < 0.03 at final low lOP) (Table 1, Fig 1), but not of venous caliber (P = 0.257 at initial high lOP and P = 0.418 at final low lOP). For the 33 eyes of 33 COAG patients, the mean (± standard deviation) arterial and venous calibers at low lOP were 82.0 ~m (± 13.8 ~m) and 121.4 ~m (± 18.5 ~m), respectively. These values were significantly lower than 87.8 ~m (±14.2 ~m) and 128.3 ~m (±20.8 ~m), respectively, at high lOP (P < 0.001 for both) (Table 2). Overall, the arterial and venous caliber decreases correlated positively with the magnitude ofIOP reduction (r = 0.503, P < 0.01 and r = 0.555, P < 0.001, respectively) (Table 1, Fig 2). The retinal venous caliber change was
SHIN et al
•
lOP· DEPENDENT RETINAL VASCULAR CHANGE
Table 1. Age, Intraocular Pressure Change, and Retinal Vascular Caliber Change
r·
p.
Rhot
Pt
Retinal arterial caliber Initial Final
-0.47 -0.385
0.0058 0.0270
-0.45 -0.397
0.011 0.025
Vessel caliber change Retinal artery Retinal vein
0.503 0.555
0.0029 0.0008
0.399 0.327
0.024 0.064
Retinal arterial caliber change Age s:55 yrs Age >55 yrs Retinal venous caliber change
0.636 0.205 0.646
0.006 0.447 0.0029
0.525 0.151 0.400
0.036 0.562 0.024
x Age (yrs) lOP reduction (mmHg) lOP reduction (mmHg) Retinal arterial caliber change (ILm)
y
• Simple linear regression analysis.
t Spearman correlation analysis.
directly proportional to the retinal arterial caliber change
1~~---------------------------------'
(r = 0.646, P = 0.0001) (Table 1).
The 95% confidence interval upper bounds of the total variabilities for the retinal arterial and venous caliber measurements were 3.4 ILm and 3.8 ILm, respectively. The mean arterial and venous caliber decreases observed in the present study were 5.8 ILm (±5.3 ILm) and 6.9 ILm (±6.9 ILm), with 95% confidence intervals of 3.9 to 7.7 ILm and 4.5 to 9.3 ILm, respectively (Table 2). Therefore, the mean retinal arterial and venous caliber decreases observed in the present study far exceeded the upper bound of their respective total variability's 95% confidence interval. Even the lower bounds of the 95% confidence intervals for the arterial and the venous caliber decreases were greater than the upper bound of their respective total variability's 95% confidence interval. Whereas the relationship between the retinal arterial caliber size and the age was suggestive of a continuum, the relationship between the retinal arterial caliber change and the magnitude of lOP reduction or the age was not. Therefore, we investigated the relationship of the retinal arterial caliber change to the magnitude ofIOP reduction or the age in younger and older groups separately. We chose the age of 55 years because it allowed us to divide our study population into two groups of similar sample size, and also because 55 years of age roughly represented the middle point of the age range in the study. Furthermore, after the division of the study population into the two different age groups, it was revealed that age demonstrated a significant correlation to the arterial caliber changes following lOP reduction in patients older than
•
• •
•
E
-=
1~~================================~ 120
•• •• • • ~f~·~--;·:-~--~·.;·--~~:~·~·~·~ • • • • 80 • •••
.
• •
~+----,----,----,----,---~~---.--~ 20 30 40 50 60 70 90 80
Atje (yaat1l)
Fig I. Age-related decline of retinal arterial c aliber. Top, before lOP reduction. BOllom. after lOP reduction.
55 years (multiple linear regression analysis, P < 0.05) (Table 3), but did not in younger patients (multiple linear regression analysis, P> 0.7) (Table 3). However, lOP re-
Table 2. lOP and Retinal Vessel Caliber Changes
lOP (mmHg) Retinal artery caliber (ILm) Retinal vein caliber (ILm)
Before lOP Reduction
After lOP Reduction
Change
p.
35.3 ± 7.2 87.8 ± 14.2 128.3 ± 20.8
16.5 ± 4.7 82.0 ± 13.8 121.4 ± 18.5
-18.8 ± 9.2 - 5.8 ± 5.3 - 6.9 ± 6.9
<0.0001 <0.0001 <0.0001
Values are mean ± standard deviation . • Two-tailed paired t test.
1089
OPHTHALMOLOGY
•
JULY 1991
& :i
"
!
•
•
20
<.)
10
~
<
~ a:
·i
0
• •
· 10 30
5
§o
2<.) j ~
<.)
•
•
E
~
•
NUMBER 7
20
DISCUSSION 10
.~
>
VOLUME 98
compared with 0.3 (±0.3) ~m/mmHg and 0.4 (±0.4) ~m/ mmHg, respectively, in the 26 eyes on ,B-blockers at both high and low lOPs. No statistically significant differences were seen in arterial or venous caliber change and mean age between the two groups (P > 0.4 for each by twotailed unpaired t test). The duration of lOP reduction ranged from 2 to 72 weeks, with a mean (± standard deviation) of 11.2 (± 13.5) weeks. Within this range, retinal arterial caliber change did not correlate significantly with duration of lOP reduction in either age group (P> 0.2) (Table 3). The results of Spearman correlation analysis agreed with those of simple linear regression (Table I).
30
E 5
•
0
:!
•
•
· 10 0
10
20
30
50
40
lOP RodUClion (mm Hg)
Fig 2. lOP-dependent retinal vessel caliber changes (expressed as decreases) of 33 COAG patients. Top, retinal artery. Ballam, retinal vein.
Table 3. Multiple Linear Regression of Retinal Artery Caliber Change or Percent Change on Age, lOP Reduction, and Duration Change of Retinal Artery (y) Age (yrs) Patients (eyes)
Age (Xl , yrs) lOP change (X2, mmHg) Durationt (X3, wk) Overall
Percent Change of Retinal Artery (y)
s 55 17
>55 16
s 55 17
>55 16
P*
P*
P*
P'
0.759 0.014 0.505 0.055
0.013 0.999 0.530 0.061
0.815 0.007 0.271 0.028
0.028 0.734 0.904 0.108
* P value from multiple regression analysis.
t Duration of lOP reduction.
duction correlated most strongly with the absolute and percent changes of retinal arterial caliber in patients 55 years of age or younger (simple linear regression, P < 0.01) (Table 1) (multiple linear regression, P < 0.02) (Table 3) but not in older patients (simple linear regression analysis, P> 0.4) (Table I) (multiple linear regression analysis, P > 0.7) (Table 3). No statistically significant differences were found in the initial and the final lOPs and the magnitude and duration of lOP reduction between the two age groups (55 and younger versus older than 55) (P > 0.1 for each by two-tailed paired t test). The retinal arterial and venous caliber decreases were 0.2 (±0.2) ~m/mmHg and 0.3 (:to. I ) ~m/mmHg, respectively, in the 6 eyes on ,B-blockers at only low lOP 1090
The current study establishes the phenomenon ofIOPdependent retinal arterial caliber change. The lOP-dependent retinal arterial caliber change and the previously demonstrated blood pressure-dependent change2 are consistent with the theory of autoregulation 1-13: the vessels constrict as the retinal perfusion pressure rises (with elevation of blood pressure or reduction of lOP) and dilate as the perfusion pressure falls (with lowering of blood pressure or an increase in lOP). Thus, the measurement of the retinal arterial caliber change in response to change of lOP or blood pressure allows the study of the retinal circulation autoregulation. 2,3,6 Like previous investigators, 19 we observed a retinal venous caliber decrease after lOP reduction. In the present study, this venous caliber change was in direct proportion to the retinal arterial caliber change. Of greater importance is our observation of the significance of age and lOP in retinal arterial caliber change. In young COAG patients (55 years of age or younger), retinal arterial caliber decrease is in direct proportion to the magnitude of lOP reduction and is age-independent. In their elderly counterparts older than 55 years of age, however, retinal arterial caliber change is age-dependent and essentially unrelated to the magnitude ofIOP reduction. With increasing age, the retinal arterial lumen becomes smaller and, more importantly, less reactive. 2 This process of age-related decline in retinal arterial reactivity will result in impaired retinal autoregulation. The impaired retinal autoregulation will be detrimental to the health of retinal tissue, including the ganglion cells and their axons within the nerve fiber layer, in the presence of fluctuation of perfusion pressure. Thus, because of this age-related decline in retinal arterial reactivity and the possible longer duration of elevated lOP in older ocular hypertensive subjects, it is not surprising that old age is known to be an important risk factor in the development of glaucomatous optic nerve damage.20- 23 Studies on the effect of raised lOP on retinal blood flow have yielded somewhat conflicting and confusing results regarding retinal autoregulation because of differences in species and techniques used.3•5- 11 However, retinal autoregulation appears to be operative over a wide range of IOPin normalcircumstances.8- IO According to a Doppler velocimetry study in healthy human subjects with acute
SHIN et al
•
lOP-DEPENDENT RETINAL VASCULAR CHANGE
increase ofIOP using a suction cup without general anesthesia, the retinal autoregulation appears to be fully effective up to about 30 mmHg and at least partially operative up to about 42 mmHg OfIOP.ll Our observation of retinal autoregulation, as evidenced by retinal arterial caliber change in response to lOP reduction from the initial (high) lOP of 35.3 mmHg (±7.2 mmHg) with a range of 23 to 56 mmHg to the final (low) lOP of 16.5 mmHg (±4.7 mmHg) with a range of 6 to 27 mmHg, confirms that retinal autoregulation is effective over a relatively wide range of lOP at least in younger patients with glaucoma. Considering that 29 (90%) of 33 eyes in our study had their lOP change within the range of 6 to 42 mmHg, our observation is not incompatible with the range ofIOP for effective retinal autoregulation observed in the healthy human subjects. 11,12 Emest's24 observation that only a transient reduction of oxygen tension occurred in the prelaminar portion of the optic nerve head after a sudden reduction in blood pressure strongly suggests an autoregulation of blood flow to the prelaminar portion of the optic nerve head. Geijer and Bill8 and Sossi and Anderson 9 also observed that the blood flow to the prelaminar portion of the optic nerve is autoregulated within a wide range of perfusion pressures in a manner very similar to the retina. In addition, Pillunat and co-workers25 have observed that an artificial increase in lOP in control subjects causes an initial decrease in the visual-evoked response. However, this quickly stabilizes and does not decrease further despite further elevations in the lOP. They postulated that this finding is consistent with autoregulation of the circulation to the prelaminar portion of the optic nerve head. This phenomenon was found to be absent in primary open-angle glaucoma patients (Pillunat et aI, unpublished data; presented at 1989 AR VO meeting). The choroidal circulation, which is responsible for most of the blood supply to the prelaminar portion of the optic nerve head, lacks autoregulation. 5,7 Therefore, the apparent autoregulation of blood flow to the prelaminar portion of the optic nerve head could only be possible via a potential contribution from retinal circulation that is autoregulated. Such a mechanism may become operative to serve the important function of ensuring adequate nourishment to the optic nerve tissue in the critical situation of compromise in choroidal circulation as a result of a decrease in systemic blood pressure or an increase in lOP. Although there are conflicting reports,5,7-1O,26-28 blood supply from the retinal and the ciliary circulation to the prelaminar portion of the optic nerve head is consistent with available evidence of autoregulated blood flow to the region. 8- 1O,24,25 Thus, deficient optic nerve head autoregulation to the prelaminar portion of the optic nerve head and to the ganglion cells and their axons within the nerve fiber layer, secondary to a deficient retinal autoregulation seen in the elderly, may be a major reason for the increased incidence and prevalence of glaucoma in the group.20-23 Although we find retinal autoregulation to be deficient in elderly COAG patients, we do find it present in young adult glaucoma patients, in contrast to the report of Pillunat et al (unpublished data; presented at 1989 ARVO meeting). Thus, the loss or deficiency in the autoregulation
of blood flow to the prelaminar portion of the optic nerve head is only a risk factor when accompanied by another detrimental effect of elevated lOP such as blockade of axonal transport29- 32 : it may not be necessarily an essential or a sufficient factor by itself for the development of glaucomatous optic nerve damage. Individuals younger than 55 years of age may develop glaucoma with a marked increase ofIOP or even with a mild-to-moderate increase ofIOP in the presence of other risk factors such as vascular insufficiencl 3 or a weak lamina cribrosa. Ocular hypotensive medications such as nonspecific {3blockers and epinephrine or its derivative, dipivefrin, might cause retinal vascular constriction 34 or affect the capacity of the retina to autoregulate. 35 In our study, however, no statistically significant difference was found in the retinal vascular caliber decreases between the 6 eyes on {3-blockers at only low lOP and the 26 eyes on {3-blockers at both high and low lOPs. Because only 1 of the 33 patients was taking dipivefrin at only low lOP, dipivefrin could not have been an important factor influencing the retinal vascular caliber changes observed in our study. Therefore, our observations of retinal vascular caliber decreases most likely result from lOP reduction rather than from the use of dipivefrin or {3-blockers. This is not surprising in view of reports confirming that retinal arteries lack functional adrenergic {3-receptors,36-38 in stark contrast to the uveal vessels. 38 Furthermore, {3-blockers do not appear to affect either retinal arterial caliber38,39 or autoregulation of blood flow. 39 The mean retinal arterial and venous caliber changes observed in the current study far exceeds the upper bound of their respective total variability's 95% confidence interval. Even the lower bounds of the 95% confidence intervals for arterial and venous caliber changes were greater than the upper bound of their respective total variability's 95% confidence interval. Therefore, although minor changes of vessel diameter could occur during the cardiac cycle or from normal fluctuation of blood pressure, such changes must have been relatively small compared with changes caused by large magnitudes of lOP reduction. Because the retinal arterial caliber changes are in direct proportion with the lOP change, the vascular volume within the neuroglial tissue in the optic nerve head also must vary in direct proportion with lOP. Therefore, the increase or decrease of cupping of the optic nerve head after the elevation or reduction of lOPs may be underestimated using available techniques,40 since the vascular volume within the optic nerve head will increase at high lOP and decrease at low lOP. It has been suggested that the peripapillary vessel caliber can be used to estimate optic disc size. 16 However, it must be used cautiously, because retinal arterial caliber is now known to be influenced significantly by both age and lOP. The sample size of our study was not large enough to allow us to probe whether severity of glaucoma or presence of systemic vascular diseases such as hypertension and diabetes mellitus might influence the lOP-dependent changes of retinal vascular caliber. The sample size was, however, large enough to demonstrate the highly significant and strong correlation of retinal arterial caliber decrease to magnitude of lOP reduction and the significant 1091
OPHTHALMOLOGY
•
JULY 1991
influence of age on the phenomenon, the two main points of this report.
REFERENCES 1. Lassen NA. Cerebral blood flow and oxygen consumption in man. Physiol Rev 1959; 39:183-238. 2. Tachibana H, Gotoh F, Ishikawa Y. Retinal vascular autoregulation in normal subjects. Stroke 1982; 13:149-55 3. Dobree JH. Calibre changes in retinal vessels occurring in raised ocular tension. Circulatory compen~ation in chronic glaucoma. Br J OphthalmoI1956; 40:1-13. 4. Russell RW. Evidence for autoregulation in human retinal circulation . Lancet 1973; 2:1048-50. 5. Aim A, Bill A. Ocular and optic nerve blood flow at normal and increased intraocular pressures in monkeys (Macaca irus): a study with radioactively labelled microspheres including flow determinations in brain and some other tissues. Exp Eye Res 1973; 15:15-29. 6. Ffytche TJ, Bulpitt CJ, Kohner EM, et al. Effect of changes in intraocular pressure on the retinal microcirculation. Br J Ophthalmol 1974; 58: 514-22. 7. Bill A. Blood circulation and fluid dynamics in the eye. Physiol Rev 1975; 55:383-417. 8. Geijer C , Bill A. Effects of raised intraocular pressure on retinal, prelaminar, laminar, and retrolaminar optic nerve flow in monkeys. Invest Ophtha/mol Vis Sci 1979; 18:1030-42. 9. Sossi N, Anderson DR. Effect of elevated intraocular pressure on blood flow. Occurrence in cat optic nerve head studied with iodoantipyrene I 125. Arch Ophthalmol 1983; 101 :98-1 01. 10. Sperber GO, Bill A. Blood flow and glucose consumption in the optic nerve, retina, and brain: effects of high intraocular pressure. Exp Eye Res 1985; 41 :639-53. 11. Riva CE, Grunwald JE, Petrig, BL. Autoregulation of human retinal blood flow. An investigation with laser Doppler velocimetry. Invest Ophthalmol Vis Sci 1986; 27:1706-12. 12. Grunwald JE, Sinclair SH, Riva CEo Autoregulation of the retinal circulation in response to decrease of intraocular pressure below normal. Invest Ophtha/mol Vis Sci 1982; 23:124-6. 13. Robinson F, Riva CE, Grunwald JE, et al. Retinal blood flow autoregulation in response to an acute increase in blood pressure. Invest Ophthalmol Vis Sci 1986; 27:722-6. 14. Aulhorn E. Visual field defects in chronic glaucoma. In: Heilmann K, Richardson KT, eds. Glaucoma: Conceptions of a Disease: PathogenesiS, Diagnosis, Therapy. Philadelphia: WB Saunders, 1978: 157-
68. 15. Jonas JB, Gusek GC, Guggenmoos-Holzmann I, Naumann GOH. Size of the optic disc nerve scleral canal and comparison to intravital determination of the disc dimensions. Graefes Arch Clin Exp Ophthalmol 1988; 226:213-5. 16. Jonas JB, Gusek GCH, Naumann GOH. OptiC disk morphometry in high myopia. Graefes Arch Clin Exp OphthalmoI1988; 226:587-90. 17. Littmann H. Zur Bestimmung der wahren GroBe eines Objektes auf dem Hintergrund des lebenden Auges. Klin Monatsbl Augenheilkd 1982; 180:286-9. 18. Shields MB, Tiedeman JS, Miller KN, et al. Accuracy of topographic measurements with the Optic Nerve Head Analyzer. Am J Ophthalmol 1989; 107:273-9.
1092·
•
VOLUME 98
•
NUMBER 7
19. Pederson JE, Herschler J. Reversal of glaucomatous cupping in adults. Arch Ophthalmol1982; 100:426-31 . 20. Armaly MF, Krueger DE, Maunder L, et al. Biostatistical analysiS of the collaborative glaucoma study. I. Summary report of the risk factors for glaucomatous visual field defects. Arch Ophthalmol1980; 98:216371. 21. Kass MA, Hart WM Jr, Gordon M, Miller JP. Risk factors favoring the development of glaucomatous visual field loss in ocular hypertension . Surv Ophthalmol1980; 25:155-62. 22. Kahn HA, Milton RC. Alternative definitions of open-angle glaucoma: effect on prevalence and associations in the Framingham eye study. Arch Ophtha/moI1980; 98:2172-7. 23. Anderson D. Glaucoma: the damage caused by pressure. XLVI Edward Jackson Memorial Lecture. Am J Ophthalmol1989; 108:485-95. 24. Emest JT. Optic disk oxygen tension . Exp Eye Res 1977; 24:271-8. 25. Pillunat LE, Stodtmeiser R, Wilmanns I, Christ TH. Autoregulation of ocular flow during changes of intraocular pressure: preliminary results. Graefes Arch Clin Exp Ophthalmol 1985; 223:219-23. 26. Hayreh SS. Structure and blood supply of the optic nerve. In: Heilmann K, Richardson KT, eds. Glaucoma: Conceptions of a Disease. Philadelphia: WB Saunders; 1978:78-96. 27. Anderson DR, Braverman S. Reevaluation of the optic disk vasculature. Am J Ophthalmol1976; 82:165-74. 28. Lieberman MF, Maumenee AE, Green WR. Histologic studies of the vasculature of the anterior optic nerve. Am J Ophthalmol 1976; 82: 405-23. 29. Levy NS. The effects of elevated intraocular pressure on slow axonal protein flow. Invest Ophthalmol Vis Sci 1974; 13:691-5. 30. Anderson DR, Hendrickson A. Effect of intraocular pressure on rapid axoplasmic transport in monkey optic nerve. Invest Ophthalmol Vis Sci 1974; 13:771-83. 31. Minckler DS, Tso MOM, Zimmerman LE. A light microscopic autoradiographic study ofaxoplasmic transport in the optic nerve head during ocular hypotony, increased intraocular pressure, and papilledema. Am J Ophthalmol1976; 82:741-57. 32. Quigley HA, Anderson DR. The dynamics and location of axonal transport block by acute intraocular pressure elevation in primate optic nerve. Invest Ophtha/mol Vis Sci 1976; 15:606-16. 33. Hayreh SS. Inter-individual variation in blood supply of the optic nervehead. Doc Ophthalmol1985; 59:217-46. 34. Martin XD, Rabineau PA. Vasoconstrictive effect of topical timolol on human retinal arteries. Graefes Arch Clin Exp Ophthalmol1989; 227: 526-30. 35. Grunwald JE. Effect of topical timolol on the human retinal circulation. Invest Ophthalmol Vis Sci 1986; 27:1713-9. 36. Nielsen PJ, Nyborg CB. Adrenergic responses in isolated bovine resistance arteries. Int Ophthalmol 1989; 13:103-7. 37. Hoste AM, Beels PJ, Brutsaert DL. Effect of alpha-1 and beta agonists on contraction of bovine retinal resistance arteries in vitro. Invest Ophthalmol Vis Sci 1989; 30:44-50. 38. Laties AM. Central retinal artery innervation: absence of adrenergic innervation to the intraocular branches. Arch Ophthalmol 1967; 77: 405-9. 39. Hoste AM, Boels PJ, Andries LJ, et al. Effects of j3-antagonists on contraction of bovine retinal microarteries in vitro. Invest Ophthalmol Vis Sci 1990; 31:1231-7. 40. Shin DH, Bielik M, Hong YJ, et al. Reversal of glaucomatous optic disc cupping in adult patients. Arch Ophthalmol1989; 107: 1599-603.
Abstract: Intraocular pressure (IOP)-dependent retinal vascular changes were investigated in 33 eyes of 33 adult chronic open-angle glaucoma (COAG) patients by measuring major retinal vascular calibers at the optic disc border before and after mean (± standard deviation) lOP reduction from 35.3 (±7.2) to 16.5 (±4.7) mmHg for 11.2 (±13.5) weeks . Both mean arterial and venous calibers were significantly reduced (P < 0.0001 for each), from 87.8 (±14.2) and 128.3 (±20.8) JLm at high lOP to 82.0 (±13.8) and 121.4 (±1B.5) JLm at low lOP. Arterial and venous caliber decreases correlated positively with magnitude of lOP reduction (r = 0.503, P < 0.01 and r = 0.555, P < 0.001, respectively). While the lOP-dependent retinal arterial caliber change was highly significant in patients 55 years of age or younger (r = 0.636, P < 0.01) and in the overall study group, it was not significant in patients older than 55 years (r = 0,205, P > 0.1). Age seems to be more influential than magnitude of lOP reduction in patients older than 55. Thus, diminished lOP-dependent retinal arterial caliber change in the elderly may be one factor contributing to the higher incidence and prevalence of glaucoma in this population. Ophthalmology 1991; 98: 1087-1092
Autoregulation of blood flow is an intrinsic mechanism for maintaining constant flow despite changes in perfusion pressure,,- 13 Retinal autoregulation appears to be mediated, at least in part, by changes in arterial caliber: the Originally received: May 17, 1990. Revision accepted: March 5, 1991. 1
2
Kresge Eye Institute, Wayne State University, Detroit. Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor.
Presented in part as a poster at the American Academy of Ophthalmology Annual Meeting, New Orleans, Ocl/Nov 1989. Reprint requests to Dong H. Shin, MD, PhD, Kresge Eye Institute, 4717 St. Antoine Blvd, Detroit, MI 48201.
caliber increases as the perfusion pressure falls and decreases as the pressure rises. 2,3,6 Retinal perfusion pressure is defined here as the difference between retinal arterial blood pressure and intraocular pressure (lOP).7 Both retinal arterial caliber changes and age-associated decline of retinal arterial reactivity have been observed in the presence of systemic blood pressure changes. 2 However, there has been a paucity of information regarding the change in the retinal vascular caliber with respect to chronic prolonged alteration of the lOP in glaucoma patients. Therefore, we investigated whether retinal vascular calibers would decrease on lowering of the lOP and what factors might influence lOP-associated retinal vascular caliber changes in chronic open-angle glaucoma (COAG) patients. 1087
OPHTHALMOLOGY
•
JULY 1991
PATIENTS AND METHODS The study group consisted of 18 male and 15 female adult patients with early to moderate COAG. 14 Fourteen (42%) were black and 19 (58%) were white. The mean (± standard deviation) age of the patients was 53.0 (± 15.9) years (range, 23 to 81 years). The lOP, measured by Goldmann applanation tonometry, was reduced by a mean (± standard deviation) of 18.8 mmHg (±9.2 mmHg) from 35.3 mmHg (±7.2 mmHg) (range, 23 to 56 mmHg) to 16.5 mmHg (±4.7 mmHg) (range, 6 to 27 mmHg) for a duration of 11.2 (± 13.5) weeks. The lOP was reduced in 26 eyes with medication alone, in 5 eyes with combined laser trabeculoplasty, and in 2 eyes with filtering surgery. The 2 eyes that underwent filtering surgery had lOP reduced without chronic hypotony for at least 4 months. No known event or intervention was noted that would have caused a substantial change of blood pressure during the study period in any of the patients. Dipivefrin 0.1 % was used for lOP reduction in only 1 of 33 eyes in the study. Six eyes of 6 patients with a mean (± standard deviation) age of54.7 (±14.3) years and a mean (± standard deviation) lOP reduction of44.3% (±3.7%) were on ,a-blockers at only low lOP; 26 eyes of 26 patients with a mean age of 51.5 (± 15.7) years and a mean lOP reduction of 53.4% (±16.5%) were on ,a-blockers at both high and low lOPs; and 1 eye was without a ,a-blocker at both high and low lOPs. Optic disc photographs (15 0 color stereo slides, magnification X5) were taken with a telecentric Zeiss (Thornwood, NY) fundus camera in the 33 adult COAG patients both before and after lOP reduction. The eye with the higher magnitude of lOP reduction was chosen in each patient as long as the photographs were deemed of adequate quality to accurately assess the vascular caliber. Using a Topcon (Paramus, NJ) microfiche projector (X 10 magnification) and a vernier caliper of 0.02 mm calibration, the horizontal disc diameter and the two clearest major retinal arterial and venous calibers (one superior and one inferior) were measured at the optic disc border from fundus photographs taken before and after lOP reduction in a masked fashion. Measurements of the major retinal arterial and venous calibers at the optic disc border of the fundus photographs have been shown to be reliable and accurate by other investigators. 2, 15, 16 Images were taken with the Rodenstock Optic Nerve Head Analyzer on the same day as the optic disc photographs and were corrected for magnification with the axial length, based on Littmann's method. 17 The accuracy of the Rodenstock Analyzer was confirmed with a model eye l8 and in glaucomatous monkey eyes (Yamauchi et ai, unpublished data; presented at 1989 AR VO meeting). A correction factor was then derived by dividing the horizontal optic disc diameter obtained from the Rodenstock analyzer image by the horizontal optic disc diameter obtained from the fundus photograph. The correction factor was necessary to obtain the actual vascular calibers and to reduce possible errors resulting from any magnification difference between the initial and the subsequent photographs. 1088
•
VOLUME 98
•
NUMBER 7
Each vessel was measured five times and an average value was determined for each. Afterward, the mean arterial or venous calibers were computed by averaging the measurements of the two major arteries or veins and then multiplying the average values by the calculated correction factor. The following formula was used: MVC = (lh)(Rd/Pd){(SI
+ S2 + S3 + S4 + S5)/5 + (11 + 12 + 13 + 14 + 15)/5},
where MVC
=
Mean vessel caliber,
Rd
=
Disc diameter by Rodenstock Optic Nerve Head Analyzer,
Pd = Disc diameter by fundus photograph, Rd/Pd
=
Correction factor,
S = Superior vessel caliber, and I
=
Inferior vessel caliber.
Vascular caliber measurements of seven eyes of seven adult subjects also were obtained and corrected by the correction factor as described above from two consecutive fundus photographs and Rodenstock Analyzer images, taken without any intervention to alter lOP or blood pressure, for determination of total variabilities and their 95% confidence intervals. The total variability represents the sum of intraphotographic variability and interphotographic variability. The vascular calibers at the higher lOP were compared with those at the lower lOP using a two-tailed paired t test. Correlations were also sought among lOP reduction, the duration ofIOP reduction, and the patient's age, with the (absolute and percent) retinal arterial caliber changes using simple and multiple linear regression analyses in all 33 patients and in the two age groups (55 years and younger, and older than 55 years). All results of simple linear regression analysis were confirmed with Spearman correlation analysis.
RESULTS A statistically significant age-related decline of arterial caliber was seen at both high and low lOPs (P < 0.02 at initial high lOP and P < 0.03 at final low lOP) (Table 1, Fig 1), but not of venous caliber (P = 0.257 at initial high lOP and P = 0.418 at final low lOP). For the 33 eyes of 33 COAG patients, the mean (± standard deviation) arterial and venous calibers at low lOP were 82.0 ~m (± 13.8 ~m) and 121.4 ~m (± 18.5 ~m), respectively. These values were significantly lower than 87.8 ~m (±14.2 ~m) and 128.3 ~m (±20.8 ~m), respectively, at high lOP (P < 0.001 for both) (Table 2). Overall, the arterial and venous caliber decreases correlated positively with the magnitude ofIOP reduction (r = 0.503, P < 0.01 and r = 0.555, P < 0.001, respectively) (Table 1, Fig 2). The retinal venous caliber change was
SHIN et al
•
lOP· DEPENDENT RETINAL VASCULAR CHANGE
Table 1. Age, Intraocular Pressure Change, and Retinal Vascular Caliber Change
r·
p.
Rhot
Pt
Retinal arterial caliber Initial Final
-0.47 -0.385
0.0058 0.0270
-0.45 -0.397
0.011 0.025
Vessel caliber change Retinal artery Retinal vein
0.503 0.555
0.0029 0.0008
0.399 0.327
0.024 0.064
Retinal arterial caliber change Age s:55 yrs Age >55 yrs Retinal venous caliber change
0.636 0.205 0.646
0.006 0.447 0.0029
0.525 0.151 0.400
0.036 0.562 0.024
x Age (yrs) lOP reduction (mmHg) lOP reduction (mmHg) Retinal arterial caliber change (ILm)
y
• Simple linear regression analysis.
t Spearman correlation analysis.
directly proportional to the retinal arterial caliber change
1~~---------------------------------'
(r = 0.646, P = 0.0001) (Table 1).
The 95% confidence interval upper bounds of the total variabilities for the retinal arterial and venous caliber measurements were 3.4 ILm and 3.8 ILm, respectively. The mean arterial and venous caliber decreases observed in the present study were 5.8 ILm (±5.3 ILm) and 6.9 ILm (±6.9 ILm), with 95% confidence intervals of 3.9 to 7.7 ILm and 4.5 to 9.3 ILm, respectively (Table 2). Therefore, the mean retinal arterial and venous caliber decreases observed in the present study far exceeded the upper bound of their respective total variability's 95% confidence interval. Even the lower bounds of the 95% confidence intervals for the arterial and the venous caliber decreases were greater than the upper bound of their respective total variability's 95% confidence interval. Whereas the relationship between the retinal arterial caliber size and the age was suggestive of a continuum, the relationship between the retinal arterial caliber change and the magnitude of lOP reduction or the age was not. Therefore, we investigated the relationship of the retinal arterial caliber change to the magnitude ofIOP reduction or the age in younger and older groups separately. We chose the age of 55 years because it allowed us to divide our study population into two groups of similar sample size, and also because 55 years of age roughly represented the middle point of the age range in the study. Furthermore, after the division of the study population into the two different age groups, it was revealed that age demonstrated a significant correlation to the arterial caliber changes following lOP reduction in patients older than
•
• •
•
E
-=
1~~================================~ 120
•• •• • • ~f~·~--;·:-~--~·.;·--~~:~·~·~·~ • • • • 80 • •••
.
• •
~+----,----,----,----,---~~---.--~ 20 30 40 50 60 70 90 80
Atje (yaat1l)
Fig I. Age-related decline of retinal arterial c aliber. Top, before lOP reduction. BOllom. after lOP reduction.
55 years (multiple linear regression analysis, P < 0.05) (Table 3), but did not in younger patients (multiple linear regression analysis, P> 0.7) (Table 3). However, lOP re-
Table 2. lOP and Retinal Vessel Caliber Changes
lOP (mmHg) Retinal artery caliber (ILm) Retinal vein caliber (ILm)
Before lOP Reduction
After lOP Reduction
Change
p.
35.3 ± 7.2 87.8 ± 14.2 128.3 ± 20.8
16.5 ± 4.7 82.0 ± 13.8 121.4 ± 18.5
-18.8 ± 9.2 - 5.8 ± 5.3 - 6.9 ± 6.9
<0.0001 <0.0001 <0.0001
Values are mean ± standard deviation . • Two-tailed paired t test.
1089
OPHTHALMOLOGY
•
JULY 1991
& :i
"
!
•
•
20
<.)
10
~
<
~ a:
·i
0
• •
· 10 30
5
§o
2<.) j ~
<.)
•
•
E
~
•
NUMBER 7
20
DISCUSSION 10
.~
>
VOLUME 98
compared with 0.3 (±0.3) ~m/mmHg and 0.4 (±0.4) ~m/ mmHg, respectively, in the 26 eyes on ,B-blockers at both high and low lOPs. No statistically significant differences were seen in arterial or venous caliber change and mean age between the two groups (P > 0.4 for each by twotailed unpaired t test). The duration of lOP reduction ranged from 2 to 72 weeks, with a mean (± standard deviation) of 11.2 (± 13.5) weeks. Within this range, retinal arterial caliber change did not correlate significantly with duration of lOP reduction in either age group (P> 0.2) (Table 3). The results of Spearman correlation analysis agreed with those of simple linear regression (Table I).
30
E 5
•
0
:!
•
•
· 10 0
10
20
30
50
40
lOP RodUClion (mm Hg)
Fig 2. lOP-dependent retinal vessel caliber changes (expressed as decreases) of 33 COAG patients. Top, retinal artery. Ballam, retinal vein.
Table 3. Multiple Linear Regression of Retinal Artery Caliber Change or Percent Change on Age, lOP Reduction, and Duration Change of Retinal Artery (y) Age (yrs) Patients (eyes)
Age (Xl , yrs) lOP change (X2, mmHg) Durationt (X3, wk) Overall
Percent Change of Retinal Artery (y)
s 55 17
>55 16
s 55 17
>55 16
P*
P*
P*
P'
0.759 0.014 0.505 0.055
0.013 0.999 0.530 0.061
0.815 0.007 0.271 0.028
0.028 0.734 0.904 0.108
* P value from multiple regression analysis.
t Duration of lOP reduction.
duction correlated most strongly with the absolute and percent changes of retinal arterial caliber in patients 55 years of age or younger (simple linear regression, P < 0.01) (Table 1) (multiple linear regression, P < 0.02) (Table 3) but not in older patients (simple linear regression analysis, P> 0.4) (Table I) (multiple linear regression analysis, P > 0.7) (Table 3). No statistically significant differences were found in the initial and the final lOPs and the magnitude and duration of lOP reduction between the two age groups (55 and younger versus older than 55) (P > 0.1 for each by two-tailed paired t test). The retinal arterial and venous caliber decreases were 0.2 (±0.2) ~m/mmHg and 0.3 (:to. I ) ~m/mmHg, respectively, in the 6 eyes on ,B-blockers at only low lOP 1090
The current study establishes the phenomenon ofIOPdependent retinal arterial caliber change. The lOP-dependent retinal arterial caliber change and the previously demonstrated blood pressure-dependent change2 are consistent with the theory of autoregulation 1-13: the vessels constrict as the retinal perfusion pressure rises (with elevation of blood pressure or reduction of lOP) and dilate as the perfusion pressure falls (with lowering of blood pressure or an increase in lOP). Thus, the measurement of the retinal arterial caliber change in response to change of lOP or blood pressure allows the study of the retinal circulation autoregulation. 2,3,6 Like previous investigators, 19 we observed a retinal venous caliber decrease after lOP reduction. In the present study, this venous caliber change was in direct proportion to the retinal arterial caliber change. Of greater importance is our observation of the significance of age and lOP in retinal arterial caliber change. In young COAG patients (55 years of age or younger), retinal arterial caliber decrease is in direct proportion to the magnitude of lOP reduction and is age-independent. In their elderly counterparts older than 55 years of age, however, retinal arterial caliber change is age-dependent and essentially unrelated to the magnitude ofIOP reduction. With increasing age, the retinal arterial lumen becomes smaller and, more importantly, less reactive. 2 This process of age-related decline in retinal arterial reactivity will result in impaired retinal autoregulation. The impaired retinal autoregulation will be detrimental to the health of retinal tissue, including the ganglion cells and their axons within the nerve fiber layer, in the presence of fluctuation of perfusion pressure. Thus, because of this age-related decline in retinal arterial reactivity and the possible longer duration of elevated lOP in older ocular hypertensive subjects, it is not surprising that old age is known to be an important risk factor in the development of glaucomatous optic nerve damage.20- 23 Studies on the effect of raised lOP on retinal blood flow have yielded somewhat conflicting and confusing results regarding retinal autoregulation because of differences in species and techniques used.3•5- 11 However, retinal autoregulation appears to be operative over a wide range of IOPin normalcircumstances.8- IO According to a Doppler velocimetry study in healthy human subjects with acute
SHIN et al
•
lOP-DEPENDENT RETINAL VASCULAR CHANGE
increase ofIOP using a suction cup without general anesthesia, the retinal autoregulation appears to be fully effective up to about 30 mmHg and at least partially operative up to about 42 mmHg OfIOP.ll Our observation of retinal autoregulation, as evidenced by retinal arterial caliber change in response to lOP reduction from the initial (high) lOP of 35.3 mmHg (±7.2 mmHg) with a range of 23 to 56 mmHg to the final (low) lOP of 16.5 mmHg (±4.7 mmHg) with a range of 6 to 27 mmHg, confirms that retinal autoregulation is effective over a relatively wide range of lOP at least in younger patients with glaucoma. Considering that 29 (90%) of 33 eyes in our study had their lOP change within the range of 6 to 42 mmHg, our observation is not incompatible with the range ofIOP for effective retinal autoregulation observed in the healthy human subjects. 11,12 Emest's24 observation that only a transient reduction of oxygen tension occurred in the prelaminar portion of the optic nerve head after a sudden reduction in blood pressure strongly suggests an autoregulation of blood flow to the prelaminar portion of the optic nerve head. Geijer and Bill8 and Sossi and Anderson 9 also observed that the blood flow to the prelaminar portion of the optic nerve is autoregulated within a wide range of perfusion pressures in a manner very similar to the retina. In addition, Pillunat and co-workers25 have observed that an artificial increase in lOP in control subjects causes an initial decrease in the visual-evoked response. However, this quickly stabilizes and does not decrease further despite further elevations in the lOP. They postulated that this finding is consistent with autoregulation of the circulation to the prelaminar portion of the optic nerve head. This phenomenon was found to be absent in primary open-angle glaucoma patients (Pillunat et aI, unpublished data; presented at 1989 AR VO meeting). The choroidal circulation, which is responsible for most of the blood supply to the prelaminar portion of the optic nerve head, lacks autoregulation. 5,7 Therefore, the apparent autoregulation of blood flow to the prelaminar portion of the optic nerve head could only be possible via a potential contribution from retinal circulation that is autoregulated. Such a mechanism may become operative to serve the important function of ensuring adequate nourishment to the optic nerve tissue in the critical situation of compromise in choroidal circulation as a result of a decrease in systemic blood pressure or an increase in lOP. Although there are conflicting reports,5,7-1O,26-28 blood supply from the retinal and the ciliary circulation to the prelaminar portion of the optic nerve head is consistent with available evidence of autoregulated blood flow to the region. 8- 1O,24,25 Thus, deficient optic nerve head autoregulation to the prelaminar portion of the optic nerve head and to the ganglion cells and their axons within the nerve fiber layer, secondary to a deficient retinal autoregulation seen in the elderly, may be a major reason for the increased incidence and prevalence of glaucoma in the group.20-23 Although we find retinal autoregulation to be deficient in elderly COAG patients, we do find it present in young adult glaucoma patients, in contrast to the report of Pillunat et al (unpublished data; presented at 1989 ARVO meeting). Thus, the loss or deficiency in the autoregulation
of blood flow to the prelaminar portion of the optic nerve head is only a risk factor when accompanied by another detrimental effect of elevated lOP such as blockade of axonal transport29- 32 : it may not be necessarily an essential or a sufficient factor by itself for the development of glaucomatous optic nerve damage. Individuals younger than 55 years of age may develop glaucoma with a marked increase ofIOP or even with a mild-to-moderate increase ofIOP in the presence of other risk factors such as vascular insufficiencl 3 or a weak lamina cribrosa. Ocular hypotensive medications such as nonspecific {3blockers and epinephrine or its derivative, dipivefrin, might cause retinal vascular constriction 34 or affect the capacity of the retina to autoregulate. 35 In our study, however, no statistically significant difference was found in the retinal vascular caliber decreases between the 6 eyes on {3-blockers at only low lOP and the 26 eyes on {3-blockers at both high and low lOPs. Because only 1 of the 33 patients was taking dipivefrin at only low lOP, dipivefrin could not have been an important factor influencing the retinal vascular caliber changes observed in our study. Therefore, our observations of retinal vascular caliber decreases most likely result from lOP reduction rather than from the use of dipivefrin or {3-blockers. This is not surprising in view of reports confirming that retinal arteries lack functional adrenergic {3-receptors,36-38 in stark contrast to the uveal vessels. 38 Furthermore, {3-blockers do not appear to affect either retinal arterial caliber38,39 or autoregulation of blood flow. 39 The mean retinal arterial and venous caliber changes observed in the current study far exceeds the upper bound of their respective total variability's 95% confidence interval. Even the lower bounds of the 95% confidence intervals for arterial and venous caliber changes were greater than the upper bound of their respective total variability's 95% confidence interval. Therefore, although minor changes of vessel diameter could occur during the cardiac cycle or from normal fluctuation of blood pressure, such changes must have been relatively small compared with changes caused by large magnitudes of lOP reduction. Because the retinal arterial caliber changes are in direct proportion with the lOP change, the vascular volume within the neuroglial tissue in the optic nerve head also must vary in direct proportion with lOP. Therefore, the increase or decrease of cupping of the optic nerve head after the elevation or reduction of lOPs may be underestimated using available techniques,40 since the vascular volume within the optic nerve head will increase at high lOP and decrease at low lOP. It has been suggested that the peripapillary vessel caliber can be used to estimate optic disc size. 16 However, it must be used cautiously, because retinal arterial caliber is now known to be influenced significantly by both age and lOP. The sample size of our study was not large enough to allow us to probe whether severity of glaucoma or presence of systemic vascular diseases such as hypertension and diabetes mellitus might influence the lOP-dependent changes of retinal vascular caliber. The sample size was, however, large enough to demonstrate the highly significant and strong correlation of retinal arterial caliber decrease to magnitude of lOP reduction and the significant 1091
OPHTHALMOLOGY
•
JULY 1991
influence of age on the phenomenon, the two main points of this report.
REFERENCES 1. Lassen NA. Cerebral blood flow and oxygen consumption in man. Physiol Rev 1959; 39:183-238. 2. Tachibana H, Gotoh F, Ishikawa Y. Retinal vascular autoregulation in normal subjects. Stroke 1982; 13:149-55 3. Dobree JH. Calibre changes in retinal vessels occurring in raised ocular tension. Circulatory compen~ation in chronic glaucoma. Br J OphthalmoI1956; 40:1-13. 4. Russell RW. Evidence for autoregulation in human retinal circulation . Lancet 1973; 2:1048-50. 5. Aim A, Bill A. Ocular and optic nerve blood flow at normal and increased intraocular pressures in monkeys (Macaca irus): a study with radioactively labelled microspheres including flow determinations in brain and some other tissues. Exp Eye Res 1973; 15:15-29. 6. Ffytche TJ, Bulpitt CJ, Kohner EM, et al. Effect of changes in intraocular pressure on the retinal microcirculation. Br J Ophthalmol 1974; 58: 514-22. 7. Bill A. Blood circulation and fluid dynamics in the eye. Physiol Rev 1975; 55:383-417. 8. Geijer C , Bill A. Effects of raised intraocular pressure on retinal, prelaminar, laminar, and retrolaminar optic nerve flow in monkeys. Invest Ophtha/mol Vis Sci 1979; 18:1030-42. 9. Sossi N, Anderson DR. Effect of elevated intraocular pressure on blood flow. Occurrence in cat optic nerve head studied with iodoantipyrene I 125. Arch Ophthalmol 1983; 101 :98-1 01. 10. Sperber GO, Bill A. Blood flow and glucose consumption in the optic nerve, retina, and brain: effects of high intraocular pressure. Exp Eye Res 1985; 41 :639-53. 11. Riva CE, Grunwald JE, Petrig, BL. Autoregulation of human retinal blood flow. An investigation with laser Doppler velocimetry. Invest Ophthalmol Vis Sci 1986; 27:1706-12. 12. Grunwald JE, Sinclair SH, Riva CEo Autoregulation of the retinal circulation in response to decrease of intraocular pressure below normal. Invest Ophtha/mol Vis Sci 1982; 23:124-6. 13. Robinson F, Riva CE, Grunwald JE, et al. Retinal blood flow autoregulation in response to an acute increase in blood pressure. Invest Ophthalmol Vis Sci 1986; 27:722-6. 14. Aulhorn E. Visual field defects in chronic glaucoma. In: Heilmann K, Richardson KT, eds. Glaucoma: Conceptions of a Disease: PathogenesiS, Diagnosis, Therapy. Philadelphia: WB Saunders, 1978: 157-
68. 15. Jonas JB, Gusek GC, Guggenmoos-Holzmann I, Naumann GOH. Size of the optic disc nerve scleral canal and comparison to intravital determination of the disc dimensions. Graefes Arch Clin Exp Ophthalmol 1988; 226:213-5. 16. Jonas JB, Gusek GCH, Naumann GOH. OptiC disk morphometry in high myopia. Graefes Arch Clin Exp OphthalmoI1988; 226:587-90. 17. Littmann H. Zur Bestimmung der wahren GroBe eines Objektes auf dem Hintergrund des lebenden Auges. Klin Monatsbl Augenheilkd 1982; 180:286-9. 18. Shields MB, Tiedeman JS, Miller KN, et al. Accuracy of topographic measurements with the Optic Nerve Head Analyzer. Am J Ophthalmol 1989; 107:273-9.
1092·
•
VOLUME 98
•
NUMBER 7
19. Pederson JE, Herschler J. Reversal of glaucomatous cupping in adults. Arch Ophthalmol1982; 100:426-31 . 20. Armaly MF, Krueger DE, Maunder L, et al. Biostatistical analysiS of the collaborative glaucoma study. I. Summary report of the risk factors for glaucomatous visual field defects. Arch Ophthalmol1980; 98:216371. 21. Kass MA, Hart WM Jr, Gordon M, Miller JP. Risk factors favoring the development of glaucomatous visual field loss in ocular hypertension . Surv Ophthalmol1980; 25:155-62. 22. Kahn HA, Milton RC. Alternative definitions of open-angle glaucoma: effect on prevalence and associations in the Framingham eye study. Arch Ophtha/moI1980; 98:2172-7. 23. Anderson D. Glaucoma: the damage caused by pressure. XLVI Edward Jackson Memorial Lecture. Am J Ophthalmol1989; 108:485-95. 24. Emest JT. Optic disk oxygen tension . Exp Eye Res 1977; 24:271-8. 25. Pillunat LE, Stodtmeiser R, Wilmanns I, Christ TH. Autoregulation of ocular flow during changes of intraocular pressure: preliminary results. Graefes Arch Clin Exp Ophthalmol 1985; 223:219-23. 26. Hayreh SS. Structure and blood supply of the optic nerve. In: Heilmann K, Richardson KT, eds. Glaucoma: Conceptions of a Disease. Philadelphia: WB Saunders; 1978:78-96. 27. Anderson DR, Braverman S. Reevaluation of the optic disk vasculature. Am J Ophthalmol1976; 82:165-74. 28. Lieberman MF, Maumenee AE, Green WR. Histologic studies of the vasculature of the anterior optic nerve. Am J Ophthalmol 1976; 82: 405-23. 29. Levy NS. The effects of elevated intraocular pressure on slow axonal protein flow. Invest Ophthalmol Vis Sci 1974; 13:691-5. 30. Anderson DR, Hendrickson A. Effect of intraocular pressure on rapid axoplasmic transport in monkey optic nerve. Invest Ophthalmol Vis Sci 1974; 13:771-83. 31. Minckler DS, Tso MOM, Zimmerman LE. A light microscopic autoradiographic study ofaxoplasmic transport in the optic nerve head during ocular hypotony, increased intraocular pressure, and papilledema. Am J Ophthalmol1976; 82:741-57. 32. Quigley HA, Anderson DR. The dynamics and location of axonal transport block by acute intraocular pressure elevation in primate optic nerve. Invest Ophtha/mol Vis Sci 1976; 15:606-16. 33. Hayreh SS. Inter-individual variation in blood supply of the optic nervehead. Doc Ophthalmol1985; 59:217-46. 34. Martin XD, Rabineau PA. Vasoconstrictive effect of topical timolol on human retinal arteries. Graefes Arch Clin Exp Ophthalmol1989; 227: 526-30. 35. Grunwald JE. Effect of topical timolol on the human retinal circulation. Invest Ophthalmol Vis Sci 1986; 27:1713-9. 36. Nielsen PJ, Nyborg CB. Adrenergic responses in isolated bovine resistance arteries. Int Ophthalmol 1989; 13:103-7. 37. Hoste AM, Beels PJ, Brutsaert DL. Effect of alpha-1 and beta agonists on contraction of bovine retinal resistance arteries in vitro. Invest Ophthalmol Vis Sci 1989; 30:44-50. 38. Laties AM. Central retinal artery innervation: absence of adrenergic innervation to the intraocular branches. Arch Ophthalmol 1967; 77: 405-9. 39. Hoste AM, Boels PJ, Andries LJ, et al. Effects of j3-antagonists on contraction of bovine retinal microarteries in vitro. Invest Ophthalmol Vis Sci 1990; 31:1231-7. 40. Shin DH, Bielik M, Hong YJ, et al. Reversal of glaucomatous optic disc cupping in adult patients. Arch Ophthalmol1989; 107: 1599-603.