CLINICAL SCIENCES Aging Affects the Retrobulbar Circulation Differently in Women and Men

Alon Harris, PhD; Mira Harris, MD; Jose´ Biller, MD; Hanna Garzozi, MD; Drora Zarfty, MD; Thomas A. Ciulla, MD; Bruce Martin, PhD

Background: While aging clearly has protean biologi- Results: In both sexes, ophthalmic arterial end- cal effects on every organ system, the differential effects diastolic velocity decreased and the Pourcelot resis- of aging in women and men in the retrobulbar vascula- tance index rose with advancing age (each PϽ.001); peak ture, to our knowledge, have never been investigated. Be- systolic velocity in the ophthalmic vessel was age- cause and age-related are independent. In contrast, central retinal arterial flow ve- closely linked to advanced age, we performed a cross- locities were unaffected by age in both sexes. In the pos- sectional study using color Doppler imaging of 4 retro- terior ciliary arteries, in men, flow velocities and the bulbar vessels in both healthy women and men. Pourcelot resistance index were independent of age. How- ever, in women, end-diastolic velocity decreased with age Objective: To define the influence of aging per se on in both the nasal and temporal posterior ciliary vessel ocular . (each PϽ.05); peak systolic velocity was constant; the Pourcelot resistance index in each ciliary artery rose with Methods: Women (n=73) and men (n=55), aged from advancing age (each PϽ.05). 20 to 90 years, free of ocular and systemic disease, and with normal , were recruited for this Conclusion: In healthy women and men, aging- study. Postmenopausal women who were not receiving induced changes in retrobulbar hemodynamics are com- estrogen replacement were also recruited. Stud- parable to alterations seen in patients with glaucoma or ies involved color Doppler imaging analysis of the oph- age-related macular degeneration, suggesting that vas- thalmic, central retinal, and nasal and temporal poste- cular changes with senescence may contribute to in- rior ciliary arteries. Ophthalmic arterial peak systolic and creased risk for these diseases in older age. end-diastolic velocities and a Pourcelot resistance index were determined for each vessel. Arch Ophthalmol. 2000;118:1076-1080

GING INEVITABLY causes seventh and eighth decades of life. Nev- widespread physiological ertheless, the mechanisms accounting for declines that reduce func- either of these illnesses at any age remain tional capacities and in- incompletely described, and the age- crease susceptibility to dis- linked factors that increase disease risk in Aease.1 These declines necessarily include senescence remain unexplored. For ex- many alterations within the . For ex- ample, although elevated intraocular ample, aging reduces aqueous inflow and pressure is clearly a major risk factor outflow,2 lessens the mobility of the cili- for glaucoma, many patients develop glau- 3 From the Departments of ary muscle, decreases the optical perfor- comatous head damage in 4 (Drs A. Harris, mance of the eye, and slowly enlarges the normal or pharmacologically controlled in- M. Harris, Garzozi, Zarfty, and cup and thins the neuroretinal rim.5,6 The traocular pressure, and intraocular pres- Ciulla), and result of these multiple effects of aging on sure itself is largely independent of age.2 (Dr Biller), Indiana University the eye is to concentrate diseases such as Similarly, both the cause of AMD, and the School of , , glaucoma, and age-related macu- age-dependent factors that influence this Indianapolis, and the Medical lar degeneration (AMD) within older age cause, remain poorly detailed.8 While in- Sciences Program, Indiana groups.7 However, the specific link be- vestigators have traditionally believed that University, Bloomington (Dr Martin). The authors have no tween a given aging-induced decline and senescence of the retinal pigment epithe- commerical, proprietary, or disease initiation remains obscure for many lium leads to cellular engorgement and 9 financial interest in the ocular illnesses. drusen formation, others postulate an al- products or companies Both glaucoma and AMD are rare in ternative theory involving primary vascu- described in this article. younger persons and quite common in the lar changes in the .10

(REPRINTED) ARCH OPHTHALMOL / VOL 118, AUG 2000 WWW.ARCHOPHTHALMOL.COM 1076

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 SUBJECTS AND METHODS linear probe was used. Samples of pulsed-Doppler signal from within a 0.2ϫ0.2-mm sample volume were ana- lyzed to calculate blood velocities. In each vessel, EXPERIMENTAL DESIGN peak systolic (PSV) and end-diastolic (EDV) velocities were determined, and a Pourcelot resistance index Healthy persons of both sexes, with no history of ocular [(PSV−EDV)/PSV] was calculated.11,12 The ophthalmic illness or ocular trauma, were recruited for this study. Sub- artery is localized approximately 25 mm behind the , jects did not have diabetes mellitus, heart disease, ocular while the central retinal vessel is detected within the ante- or systemic hypertension, or cancer. None were taking rior portion of the optic nerve about 3 mm behind the topical or systemic medications or receiving hormonal surface of the . Typically, neither the ophthalmic therapy for any of these illnesses. Postmenopausal women nor the central retinal arteries require angle corrections. were not receiving estrogen replacement therapy. Subjects The nasal and temporal short posterior ciliary arteries are signed informed consent to procedures that had been detected immediately nasal and temporal to the optic reviewed and approved by an institutional review board. nerve head shadow, at a depth similar to the central reti- All procedures conformed to the tenets of the Declaration nal artery. of Helsinki. Subjects were studied on a single occasion in Because of their small size, it is not always possible to the laboratory. resolve individual posterior ciliary vessels, although the ve- locity spectra from these arteries are easily distinguished EXPERIMENTAL MEASUREMENTS from the ophthalmic or central retinal artery.11-13,17,18 Angle corrections were performed in measurements from the pos- Heart rate was measured by palpation of the radial pulse; terior ciliary vessels. During testing, subjects reclined in a blood pressure was measured using sphygmomanom- chair as acoustic coupling gel was placed over a closed right etry. Flow velocities in the retrobulbar vasculature were eye and the probe gently positioned. After identification determined using color Doppler imaging. of the appropriate vessel, the sample volume was placed Color Doppler imaging is an ultrasound technique in the center of the vessel, the angle of incidence was se- that combines b-scan gray-scale imaging of tissue struc- lected, and several seconds of Doppler waveform were ture with colored representation of blood movement recorded. toward or away from the sensor based on Doppler shifted frequencies and pulsed-Doppler measurement of STATISTICAL ANALYSIS blood velocities. Blood velocities were measured in the ophthalmic, central retinal, and nasal and temporal pos- The dependence of flow velocities on age was determined terior ciliary arteries. by using least-squares linear regression, with PϽ.05 A color Doppler imaging system (Siemens Quantum regarded as statistically significant for the slope of the 2000; Siemens Quantum, Issaquah, Wash) with a 7.5-MHz best-fit line.

Both glaucoma and AMD are associated with evi- RESULTS dence for increased vascular resistance within the oph- thalmic, central retinal, or short posterior ciliary arte- EXPERIMENTAL SUBJECTS rial watersheds.11-13 Because aging is associated with gradual reductions in blood flow to many distal organs, The subject characteristics (mean age, age distribution, including the brain,14 it is possible that subclinical de- age range, and arterial blood pressure) are listed in clines in ocular perfusion increase susceptibility to these Table 1. Diastolic blood pressure did not correlate with diseases in senescence. Along this line, there is increas- age in either sex; systolic blood pressure correlated with ing evidence that estrogen replacement therapy in post- age only in women (r=0.35, PϽ.01). menopausal women exerts salutary vasodilatory effects in a number of vascular beds.15 Although the effects of COLOR DOPPLER IMAGING age and estrogen replacement therapy on ocular disease in women are not well defined, there is preliminary evi- Ophthalmic Artery dence that early menopause may be associated with in- creased risk for glaucoma.16 While several other studies The ophthalmic arterial PSV was uncorrelated with age in have examined the influence of age on retrobulbar vas- either sex (Table 2). In contrast, EDV decreased with age cular flow velocities,17-19 in these experiments we spe- in both sexes (each PϽ.001; Table 2). Estimates based on cifically compared the age-dependence of these flow the coefficient of determination (r2) suggest that 30% of velocities in healthy men and in healthy women not re- the population variation in ophthalmic arterial EDV arises ceiving estrogen replacement therapy. If the normal from aging effects in women, and 12% of that variation is hemodynamic changes with aging, indeed, facilitate glau- caused by age in men (Table 2). Because EDV decreased coma or AMD development in either women or men, then with age in both sexes at constant PSV, the derived Pource- healthy older persons should exhibit blood flow and vas- lot resistance index rose steadily with age in both sexes (each cular resistance changes similar to those seen in younger PϽ.001; Table 2 and Figure 1). Coefficient of determi- diseased individuals. nation (r2) analysis again found substantial contributions

(REPRINTED) ARCH OPHTHALMOL / VOL 118, AUG 2000 WWW.ARCHOPHTHALMOL.COM 1077

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 from age to the population variation in the ophthalmic ar- and temporal posterior ciliary arteries rose with increas- tery Pourcelot resistance index, the estimate proportion be- ing age (Table 2 and Figure 3). ing 26% in men and 50% in women (Table 2). COMMENT Central Retinal Artery In this cross-sectional study, we found that ophthalmic In contrast to results seen in the ophthalmic artery, in arterial EDV decreased with advancing age at constant neither sex were PSV, EDV, or Pourcelot resistance in- PSV, producing a rise in the Pourcelot resistance index dex in the central retinal artery dependent on age (Table in both sexes. In contrast, no effects of senescence were 2 and Figure 2). noted on flow velocities measured in the central retinal artery. Finally, in the nasal and temporal short posterior Nasal and Temporal Short Posterior Ciliary Arteries ciliary arteries, both sexes showed different responses with advancing age. In men, flow velocities in these small ar- A third pattern of response, distinct from results ob- teries were independent of age. In contrast, in women served in the ophthalmic and central retinal arteries, was not receiving estrogen replacement therapy, older age was seen in the nasal and temporal short posterior ciliary ar- teries. In these vessels, men showed no correlation of any Doppler index with age (Table 2 and Figure 3). In con- Table 2. Age Dependence of Retrobulbar Arterial Flow trast, in women, EDV in each posterior ciliary artery de- Velocities* creased with advancing age (each PϽ.05; Table 2). In ad- dition, the Pourcelot resistance indexes in both the nasal Men Women rPrP Ophthalmic Artery Table 1. Subjects’ Characteristics* Peak systolic velocity . . . NS . . . NS End-diastolic velocity −0.35 .008 −0.55 Ͻ.001 Men (n = 55) Women (n = 73) Pourcelot resistance index +0.51 Ͻ.001 +0.71 Ͻ.001 Age, y 44 ± 18 47 ± 20 Central Retinal Artery Age range, y 23-77 20-90 Peak systolic velocity . . . NS . . . NS Age distribution in years, End-diastolic velocity . . . NS . . . NS No. of subjects Pourcelot resistance index . . . NS . . . NS 20-29 16 21 Temporal Short Posterior Ciliary Artery 30-39 9 12 Peak systolic velocity . . . NS . . . NS 40-49 10 8 End-diastolic velocity . . . NS −0.22 .049 50-59 6 7 Pourcelot resistance index . . . NS +0.26 .03 60-69 7 12 Ն70 7 13 Nasal Short Posterior Ciliary Artery Blood pressure, mm Hg Peak systolic velocity . . . NS . . . NS Systolic 121 ± 11 116 ± 18 End-diastolic velocity . . . NS −0.35 .005 Diastolic 73 ± 10 70 ± 10 Pourcelot resistance index . . . NS +0.35 .005

*Values are expressed as mean (± SD), unless otherwise indicated. *Ellipses indicate not applicable; NS, not significant.

0.95 Women Men

0.90

0.85

0.80

0.75 Resistance Index

0.70 r = +0.71 r = +0.51 P < .001 P < .001

0.65

0.60 2030 40 50 60 7080 90 2030 40 5060 7080 90 Age, y Age, y

Figure 1. Ophthalmic arterial Pourcelot resistance index [(peak systolic velocity minus end-diastolic velocity) divided by peak systolic velocity) in women and men as a function of age. In both sexes, the Pourcelot resistance index rose significantly with advancing age (PϽ.001).

(REPRINTED) ARCH OPHTHALMOL / VOL 118, AUG 2000 WWW.ARCHOPHTHALMOL.COM 1078

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 1.0 Women Men

0.9

0.8

0.7 Pourcelot Resistance Index Pourcelot

0.6

0.5 2030 40 50 60 7080 90 2030 40 5060 7080 90 Age, y Age, y

Figure 2. Central retinal arterial Pourcelot resistance index in women and men as a function of age. In both sexes, the Pourcelot resistance index was independent of age.

0.9 Women Men

0.8

0.7

0.6

Pourcelot Resistance Index Pourcelot 0.5

r = +0.35 P = .005 0.4

0.3 2030 40 50 60 7080 90 2030 40 50 60 7080 90 Age, y Age, y

Figure 3. Nasal posterior ciliary arterial Pourcelot resistance index in women and men as a function of age. In women, the Pourcelot resistance index rose with advancing age ( P2=.005). In men, there was no correlation of the Pourcelot resistance index with age.

sociated with a reduced EDV, constant PSV, and an el- within the remaining vessels, and compromise of endo- evated Pourcelot resistance index in both short posterior thelium-dependent vascular relaxant mechanisms.23,24 Pre- ciliary vessels. vious whole-population studies consistently find age- Noninvasive color Doppler imaging measurements linked flow velocity reductions and Pourcelot resistance necessarily involve indirect signal interpretation.20 Never- index elevations in the short posterior ciliary arter- theless, a number of in vitro and in vivo comparisons of ies.17,18 Comparable retrobulbar hemodynamic changes Doppler flow velocities with direct measurements of total are found in patients with glaucoma and AMD.10,11 In this blood flow and vascular resistance have resulted in con- study, analogous results were restricted to women. Be- sensus interpretations of selected aspects of Doppler in- cause the women in this study were not receiving estro- formation.20-22 Specifically, reduced EDV at constant PSV, gen replacement therapy, it is possible that declining es- resulting in increased Pourcelot resistance index, has been trogen levels with age contributed to these results. shown in both anesthetized dogs and in in vitro cardio- Women, more than men, lose cerebral vasoreactivity with vascular models to result from elevated vascular resis- age25; estrogen replacement therapy blunts this loss of tance downstream from the arterial measurement site.20-22 responsivity25 and mitigates age-linked changes in arte- Aging, via reductions in growth hormone and in- rial structure and distensibility.26 However, the current sulinlike growth factors, increases vascular resistance by cross-sectional results are indirect and preliminary: di- an absolute loss of arterioles, reduction of distensibility rect comparisons of women receiving and not receiving

(REPRINTED) ARCH OPHTHALMOL / VOL 118, AUG 2000 WWW.ARCHOPHTHALMOL.COM 1079

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021 estrogen replacement therapy will be required to fully de- 3. Croft MA, Oyen MJ, Gange SJ, Fisher MR, Kaufman PL. Aging effects on fine the role that estrogen may play in the regulation of and outflow facility responses to pilocarpine in humans. Arch Ophthalmol. 1996;114:586-592. ocular hemodynamics and in the prevention or mitiga- 4. Guirao A, Gonzalez C, Redondo M, Geraghty E, Norrby S, Artal P. Average opti- tion of ocular disease. cal performance of the human eye as a function of age in a normal population. Studies of both normal-tension and primary open- Invest Ophthalmol Vis Sci. 1999;40:203-213. angle glaucoma, and of AMD, characteristically find 5. Garway-Heath DF, Wollstein G, Hitchings RA. Aging changes of the optic nerve reduced flow velocities and increased ophthalmic arte- head in relation to open angle glaucoma. Br J Ophthalmol. 1997;81:840-845. 11-13,27 6. Healey PR, Mitchell P, Smith W, Wang JJ. The influence of age and intraocular rial Pourcelot resistance indexes. While previous pressure on the optic cup in a normal population. J Glaucoma. 1997;6:274-278. aging studies emphasize changes in PSV or EDV in the 7. Fafowora OF, Osuntokun OO. Age-related in the elderly members of ophthalmic artery, at an unchanged Pourcelot resistance rural African community. E African Med J. 1997;74:435-437. index,17,19 our results find reductions in ophthalmic arte- 8. Ciulla TA, Danis RP, Harris A. Age-related macular degeneration: a review of ex- rial EDV and increases in the Pourcelot resistance index perimental treatments. Surv Ophthalmol. 1998;43:134-146. 9. Young R. Pathophysiology of age-related macular degeneration. Surv Ophthal- with advancing age in both sexes. Analogous vascular mol. 1987;31:291-306. resistance rises in the ophthalmic arterial watershed are 10. Friedman E. A hemodynamic model of the pathogenesis of age-related macular found in aging rats.23 The similar effects of aging and degeneration. Am J Ophthalmol. 1997;124:677-682. disease, independent of sex, suggest that increased vas- 11. Rankin SJ, Walman BE, Buckley AR, Drance SM. Color Doppler imaging and spec- cular resistance distal to the ophthalmic artery occurring tral analysis of the optic nerve vasculature in glaucoma. Am J Ophthalmol. 1995; 119:685-693. as a normal consequence of aging could be one factor 12. Ciulla TA, Harris A, Chung HS, et al. Color Doppler imaging reveals reduced ocu- predisposing to glaucoma and/or AMD in older age lar blood flow velocities in nonexudative age-related macular degeneration. Am groups. J Ophthalmol. 1999;128:75-80. While we found evidence for rising resistance dis- 13. Friedman E, Krupsky S, Lane AM, et al. Ocular blood flow velocity in age-related tal to the ophthalmic artery, flow velocities in the cen- macular degeneration. Ophthalmology. 1995;102:640-646. 14. Krausz Y, Bonne O, Gorfine M, Karger H, Lerer B, Chisin R. Age-related changes tral retinal artery were independent of age in both sexes in brain perfusion of normal subjects detected by 99mTc-HMPAO SPECT. Neu- in our study. Other studies have found variously that ag- roradiology. 1998;40:428-434. ing is associated with reduced bulk retinal flow,28 con- 15. Penotti M, Farina M, Castiglione E, et al. Alteration in the pulsatility index values stant retinal flow despite elevated perfusion pressure,29 of the internal carotid and middle cerebral arteries after suspension of post- menopausal hormone replacement therapy: a randomized crossover study. Am or that aging and atherosclerosis render the monkey J Obstet Gynecol. 1996;175:606-611. 30 susceptible to ischemia. Other cross-sectional 16. Hulsman CA, Westendorp IC, Hofman A, de Jong PT. Early spontaneous meno- studies using color Doppler imaging have been equally pause is associated with primary open-angle glaucoma [abstract]. Invest Oph- conflicting, finding either an unchanged or increased thalmol Vis Sci. 1999;40:S384. central retinal arterial resistance index with advancing 17. Kaiser HJ, Schotzau A, Flammer J. Blood-flow velocities in the extraocular ves- 17-19 sels in normal volunteers. Am J Ophthalmol. 1996;122:364-370. age. Our result, obtained from healthy individuals 18. Greenfield DS, Heggerick PA, Hedges TR. Color Doppler imaging of normal or- free from systemic or , suggests that bital vasculature. Ophthalmology. 1995;102:1598-1605. the effect of aging per se on central retinal arterial 19. Williamson TH, Lowe GDO, Baxter GM. Influence of age, systemic blood pres- hemodynamics is small. It is unclear why aging-induced sure, smoking, and blood viscosity on orbital blood velocities. Br J Ophthalmol. increases in the Pourcelot resistance index, so apparent 1995;79:17-22. 20. Miles RD, Menke JA, Boshirer M, Colliver JA. Relationships of five Doppler mea- in the ophthalmic artery, are absent in the central reti- sures with flow in an in vitro model and clinical findings in newborn infants. nal vessel. J Ultrasound Med. 1987;6:397-404. In summary, because our results were obtained from 21. Spencer JAD, Guissani DA, Moore PJ, Hanson MA. In vitro validation of Doppler a healthy cohort characterized by normal intraocular and indices using blood and water. J Ultrasound Med. 1991;10:305-308. systemic arterial pressure, they capture the effects of un- 22. Evans DH, Barrie WW, Asher MJ, Bentley S, Bell PRF. The relationship between ultrasonic pulsatility index and proximal arterial stenosis in an animal model. Circ complicated aging on retrobulbar hemodynamics in both Res. 1980;46:470-475. sexes. Our findings imply that vascular factors may play 23. Salter JM, Cassone VM, Wilkerson MK, Delp MD. Ocular and regional cerebral a role in the increased susceptibility to glaucoma and AMD blood flow in aging Fischer-344 rats. J Appl Physiol. 1998;85:1024-1029. seen in both sexes in senescence. 24. Sonntag WE, Lynch CD, Cooney PT, Hutchins PM. Decreases in cerebral micro- vasculature with age are associated with the decline in growth hormone and insulin- like growth factor. . 1997;138:3515-3520. Accepted for publication February 12, 2000. 25. Kastrup A, Dichgans J, Niemeier M, Schabet M. Changes of cerebrovascular CO2 Dr Ciulla is a recipient of a Career Development Award reactivity during normal aging. Stroke. 1998;29:1311-1314. from Research to Prevent Blindness Inc, New York, NY. 26. McGrath BP, Liang YL, Teede H, Shiel LM, Cameron JD, Dart A. Age-related de- Corresponding author: Alon Harris, PhD, Depart- terioration in arterial structure and function in postmenopausal women: impact of hormone replacement therapy. Arterioscler Thromb Vasc Biol. 1998;18:1149- ment of Ophthalmology, 702 Rotary Cir, Indiana Univer- 1156. sity, School of Medicine, Indianapolis, IN 46202 (e-mail: 27. Hosal BM, Karakoc G, Gursel E, Camur M. Color Doppler imaging of the retro- [email protected]). bulbar circulation in age-related macular degeneration. Eur J Ophthalmol. 1998; 8:234-238. 28. Groh MJ, Michelson G, Langhans MJ, Harazny J. Influence of age on retinal and REFERENCES optic nerve head blood circulation. Ophthalmology. 1996;103:529-534. 29. Yoshida A, Feke GT, Ogasawara H, Goger DG, McMeel JW. Retinal hemodynam- 1. Miller RA. When will the biology of aging become useful? future landmarks in ics in middle- aged normal subjects. Ophthal Res. 1996;28:343-350. biomedical gerontology. J Am Geriatr Soc. 1997;45:1258-1267. 30. Hayreh SS, Bill A, Sperber GO. Effects of high intraocular pressure on the glu- 2. Toris CB, Yablonski ME, Wang YL, Camras CB. Aqueous humor dynamics in the cose metabolism in the retina and optic nerve in old atherosclerotic monkeys. aging human eye. Am J Ophthalmol. 1999;127:407-412. Graefes Arch Clin Exp Ophthalmol. 1994;232:745-752.

(REPRINTED) ARCH OPHTHALMOL / VOL 118, AUG 2000 WWW.ARCHOPHTHALMOL.COM 1080

©2000 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/23/2021