Typical recording in a 44-year-old female subject showing measurements of relative choroidal blood velocity, volume, and flow. The highlighted area corresponds to the segment of the recording that was chosen for analysis.
Relationship between choroidal blood volume and subject's age. R=−0.52, P=.004; choroidal blood volume= −0.00523 age+0.6475.
Relationship between choroidal blood flow and subject's age. R=−0.54, P=.003; choroidal blood flow= −0.167 age+23.757.
Relationship between choroidal blood velocity and subject's age. R=0.08, P=.704; choroidal blood velocity =0.00044 age+0.4459.
Relative choroidal blood flow in younger (between 15 and 45 years) and older (between 46 and 76 years) subjects. Average choroidal blood flow was 29% lower in older subjects (2-tailed, unpaired Student t test, P=.007).
Grunwald JE, Hariprasad SM, DuPont J. Effect of Aging on Foveolar Choroidal Circulation. Arch Ophthalmol. 1998;116(2):150-154. doi:10.1001/archopht.116.2.150
Copyright 1998 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.1998
To determine the effect of aging on choroidal blood flow (ChBFlow) in the foveolar region of the normal ocular fundus.
Choroidal blood flow was determined using laser Doppler flowmetry. Twenty-nine eyes of 29 normal subjects whose ages ranged from 15 to 76 years (mean±SD, 42±18 years) were included in this study. Relative choroidal blood velocity (ChBVel), choroidal blood volume (ChBVol), and ChBFlow were determined in the foveolar region by asking the subjects to fixate on the probing laser beam.
Significant negative correlations were observed between ChBVol and the subject's age (R=−0.52, P=.004) and between ChBFlow and the subject's age (R=−0.54, P=.003). No significant correlation was detected between ChBVel and the subject's age (R=0.07, P=.70). Significant differences were observed in ChBVol and ChBFlow between the younger subjects aged 15 to 45 years (mean±SD, 0.48±0.20 arbitrary units [AU] and 18.9±5.8 AU, respectively) and the older ones aged 46 to 76 years (mean±SD, 0.34±0.11 AU and 13.3±3.3 AU, respectively; unpaired Student t test, P=.04 and P=.007, respectively).
In the subjects studied, foveolar ChBFlow decreases with age. This change is probably related to the decrease in density and diameter of the choriocapillaries that occurs with increasing age.
CHOROIDAL circulation provides nutrients and removes waste products from the outer half of the retina. An abnormal choroidal blood supply may disrupt the normal retinal function and lead to visual deterioration.
Abnormalities of the choroidal circulation have been hypothesized as part of the mechanism that leads to the development of age-related macular degeneration,1- 3 one of the most common causes of visual impairment in older individuals.4- 6 To determine what constitutes an abnormal choroidal blood flow (ChBFlow), it is important to determine the physiologic effect of aging on the choroidal circulation of the normal eye.
We have previously reported evidence suggesting that blood flow in the retinal macular microcirculation decreases with age.7 The purpose of this study was to determine if a similar decrease is also present in the foveolar choroidal circulation.
Twenty-nine volunteers whose ages ranged from 15 to 45 years (18 subjects) and 46 to 76 years (11 subjects) (42±18 years, mean±SD) were included in this study. All subjects had no pathologic findings on external, slitlamp, and funduscopic examinations. None of the subjects had large drusen (>63 µm), retinal pigment epithelial abnormalities, or any other abnormality consistent with age-related macular degeneration. The intraocular pressure (IOP) was less than 21 mm Hg, and the best-corrected visual acuity was 20/25 or better in all subjects. Mean brachial artery systolic pressure was 128±20 mm Hg (range, 100-181 mm Hg) and mean brachial artery diastolic pressure was 68±14 mm Hg (range, 51-107 mm Hg). Refractive errors ranged between −5.5 diopters (D) and +4.5 D of spherical equivalent. The average refractive error was −1.2±2.1 D.
Information about medical history and current systemic and ocular therapy was obtained from each subject. Four subjects had elevated cholesterol levels. Of these 4 subjects, 2 were receiving simvastatin and 2 were receiving lovastatin. Two of the 4 subjects with elevated cholesterol levels also had a history of systemic hypertension. Of these, one was receiving therapy with enalapril and the other therapy with hydrochlorothiazide. The other 25 subjects had no history of chronic systemic disease and were not receiving systemic therapy. No significant difference in any of the circulatory parameters were observed between the subjects who received medication for systemic hypertension or elevated cholesterol levels and those who did not. Three of the 29 subjects were smokers. No significant difference in any of the circulatory parameters was observed between smokers and nonsmokers.
After a detailed explanation of the procedures, all subjects were asked to sign an appropriate consent form approved by the human experimental committee of our institution. The tenets of the Declaration of Helsinki were followed.
Before the measurements, pupils were dilated with 1% tropicamide (Alcon, Fort Worth, Tex) and 10% phenylephrine hydrochloride (Sanofi Winthrop, New York, NY). Blood flow measurements were obtained in one eye of each subject. The eye to be included in this study was determined by tossing a coin.
Relative choroidal blood velocity (ChBVel), choroidal blood volume (ChBVol), and ChBFlow were obtained using a method based on the laser Doppler flowmetry technique. Choroidal blood velocity corresponded to the speed of moving blood cells, and ChBVol represented the amount of blood present at the measurement site. Choroidal blood flow was calculated by the instrument from these 2 parameters. Detailed descriptions of the method have been previously published.8- 11 A diode laser beam (670 nm) with an intensity of 20 µW was delivered through a fundus camera (model TRC, Topcon, Tokyo, Japan). The diameter of the probing laser beam was approximately 200 µm.
During blood flow measurements, an area of the posterior retina (30° in diameter) was illuminated by the fundus camera at a wavelength of 570 µm. This illumination enabled the observation of the position of the probing laser during the measurements. Subjects were asked to fixate on the probing laser beam to determine foveolar ChBFlow. Proper fixation during the measurements was ascertained by direct observation of the foveola through the fundus camera. All measurements were performed with the subjects seated in a darkened room.
In each subject, a single continuous measurement of the choroidal circulation was obtained for about 20 to 30 seconds. An observer masked with regard to the age and all other attributes of each subject performed the analysis of the data using a NeXT computer (NeXT Computer, Inc, Redwood City, Calif) with software (NeXT Software, Inc, Redwood City) specifically developed for the analysis of Doppler signals from ocular tissues.12
The masked observer selected for analysis parts of the recording that showed stable circulatory parameters. This selection was necessary because disturbances such as eye motion, head motion, blinks, and poor fixation produced unstable blood flow readings. Therefore, only stable portions of the measurements were included in the analysis. The average recording time that was included in the analysis of the data was about 3 seconds. Figure 1 shows a typical recording in a 44-year-old subject. The highlighted section (toward the end of the recording) depicts a segment of about 3 seconds selected for analysis.
In 16 subjects, the above-described procedure was repeated and 3 separate determinations were conducted on the same experimental session to assess the reproducibility of the measurements. Coefficient of variability (CV) was defined as CV=(mean/SD)×100.
The average pulsatile component of ChBFlow was determined over the cardiac cycle that was recorded with an infrared pulse monitor. The pulsatile component of ChBFlow was calculated as (1−ChBFlowsystolic)/ChBFlowdiastolic.
Brachial artery systolic and diastolic blood pressures (BPs and BPd, respectively) and heart rate were determined after blood flow measurements. Intraocular pressure (IOP) was measured by applanation tonometry. The mean brachial artery pressure (BPm) was calculated according to the following formula: BPm=BPd+23(BPs−BPd).
Perfusion pressure (PP) for the study eye was determined according to the following formula: PP=(23 BPm−IOP).
Nonpaired Student t tests, linear regressions, and correlations were used in the statistical analysis of the results. The assumption of normality for ChBFlow was assessed with the Shapiro-Wilk test and by the examination of the residuals of the regression analysis. Probability values less than .05 were considered statistically significant.
The ChBVel ranged from 0.29 to 0.82 arbitrary units (AU) and the average (±SD) ChBVel for the entire group was 0.46±0.10 AU. The ChBVol ranged from 0.19 to 0.85 AU and the average ChBVol was 0.43±0.18 AU. The ChBFlow ranged from 7.5 to 32.1 AU and the average ChBFlow was 16.8±5.6 AU.
Statistically significant, negative linear correlations were observed between ChBVol and the subject's age (R=−0.52, P=.004, Figure 2) and between ChBFlow and the subject's age (R=−0.54, P=<.01, Figure 3). No significant correlation was detected between ChBVel and the subject's age (R=0.07, P=.70, Figure 4). No significant correlation was observed between the pulsatile component of ChBFlow and age (R=0.26, P=.17).
The age span of the study subjects was divided arbitrarily into 2 groups with similar age spans of 30 years each. The group of younger subjects included those between 15 and 45 years of age (18 subjects) and the group of older subjects included those between ages 46 and 76 years (11 subjects).
Significant differences were observed in ChBVol and ChBFlow between the younger group, aged 15 to 45 years (0.48±0.20 AU and 18.9±5.8 AU, respectively), and the older group, aged 46 to 76 years (0.34±0.11 AU and 13.3±3.3 AU, respectively; unpaired Student t test, P=.04 and P=.007, respectively, Figure 5). Both ChBVol and ChBFlow were, on the average, about 29% lower in the older group. No significant differences in ChBVel and the pulsatile component of ChBFlow were observed between the younger group (0.47±0.12 AU and 0.59±0.17, respectively) and the older one (0.45±0.06 AU and 0.63±0.12, respectively; P=.64 and P=.51, respectively).
Significantly higher values for IOP, BPs, BPd, BPm, and PP were observed in the older group (Table 1). No significant differences in any of the measured parameters were observed between males and females or between right and left eyes.
Significant negative correlations were observed between BPm and ChBVol (R=−0.49, P=.007) or ChBFlow (R=−0.46, P=.012), and between PP and ChBVol (R=−0.44, P=.017) or ChBFlow (R=−0.43, P=.019). No other significant correlations were detected between heart rate, BPm, IOP, or PP and our hemodynamic parameters.
Multiple regression models were used to determine if the relationship between age, ChBVol, and ChBFlow could be attributed to refraction, BPm, IOP, PP, or sex. None of these covariates had a statistically significant effect on ChBVol or ChBFlow when considered simultaneously with age.
Our results show a decrease in foveolar ChBVol and ChBFlow in normal subjects with increasing age. The correlation between age and ChBVol has a regression coefficient of R=−0.52 and the correlation between age and ChBFlow has a regression coefficient of R=−0.54, suggesting that approximately 27% (R2) of the change in ChBVol and 29% (R2) of the change in ChBFlow can be predicted by age.
We chose to perform our circulatory measurements in the foveola for several reasons: (1) the importance of the foveola in terms of visual function, (2) the fact that at this site there is no retinal circulation, (3) technically, the measurements are easy to obtain because there is no need for a fixation device (subjects are asked to fixate on the measuring laser beam), and (4) it is a site that is affected by pathologic conditions that may have a choroidal vascular origin (ie, age-related macular degeneration and macular dystrophies).
Riva et al,8 have previously demonstrated the validity of the laser Doppler flowmetry technique for the noninvasive assessment of relative ChBFlow in the human foveola. These ChBFlow measurements derive mainly from erythrocytes moving in the choriocapillaries rather than the large vessels behind them. As discussed by Riva et al,8 this is supported by the relatively low Doppler frequency shifts obtained, which represent slower blood velocities, and the exponential shape of the Doppler spectrum, which typically corresponds to a signal deriving from the microcirculation. The relatively small pulsatile component observed in our ChBFlow measurements provides further evidence that the method measures blood flow in the choriocapillaries.
Because measurements are performed in the center of the foveola, most of the signal obtained is derived from the choroidal and not the retinal circulation. This is supported by experiments showing that hyperoxia does not cause any significant change in foveolar ChBFlow measurements.8
The choroidal circulation is insensitive to hyperoxia,13- 15 whereas the retinal circulation is known to decrease by 30% to 60% under conditions of 100% oxygen breathing.16- 21 The lack of response of ChBFlow to hyperoxia suggests that the measurements are derived from the choroidal circulation and not the retinal circulation.
Significantly higher BPm and PP are present in our study in older individuals (Table 1), and significant negative correlations are observed between these parameters and ChBFlow. This is somewhat surprising since increased PP would theoretically be associated with an increase in ChBFlow and not a decrease. An autoregulatory response, if present, would probably modulate such an increase in ChBFlow but would not result in decreased ChBFlow, unless the metabolic needs of the choroid would considerably change.
Our results suggest that the effect of aging on ChBFlow is probably more important than the effects of increases in BPm or PP. This is indeed supported by the multiple regression analyses showing that age is the only significant variable associated with ChBFlow changes. The relationship between age and ChBFlow cannot be explained by any of the other variables previously mentioned.
A recent report by Ravalico et al,22 has shown that pulse amplitude, measured with the Langham OBF System (Langham Ophthalmic Technologies, Timonium, Md), decreases with age. Although there is a controversy as to what exactly these measurements represent hemodynamically, these results suggest that the pulsatile component of the ocular blood flow may decrease with age because a decreased pulsatility probably results in a lower pulse volume and perhaps a lower pulsatile blood flow.
These measurements, however, represent the effect of an overall intraocular blood flow since they are inferred from the change in IOP produced by the heart beat. Our measurements, in contrast, are obtained directly from the choroidal circulation at the foveola.
Our results are somewhat different from those of Ravalico et al22 since we do not find a decrease in the pulsatile component of ChBFlow with age. The discrepancy between the 2 studies may be due to the marked differences in the techniques used and the parameters measured.
The decrease in ChBFlow with age observed in our study is mainly due to a decrease in ChBVol since there is no statistically significant correlation between ChBVel and age. This decrease in ChBVol can be explained by the findings of Ramrattan et al23 showing that between the first and the 10th decades, there is a 45% decrease in the density and a 34% decrease in the diameter of the lumen of the choriocapillaries, in the normal human macula.
In addition, there are a number of age-related structural changes that occur in the retina that could lead to a decrease in ChBFlow. Such age-related changes include decreases in photoreceptor density24 and retinal pigment epithelial cell cytoplasm volume.25 A decrease in the number of viable cells in the retina and choroid could result in a decreased metabolic demand and, consequently, a decrease in both choroidal and retinal blood flows.
Age-related changes in the retinal circulation have been previously reported. Grunwald et al,7 using the Blue Field Simulation (Oculix, Inc, Berwin, Pa), described decreases in leukocyte velocity and density that suggested a decrease in blood flow in the retinal macular capillaries of older individuals. Groh et al,26 using pulsed Doppler sonography, reported a decrease in blood velocity in the central retinal artery.
Although speculative at this stage, it is possible that a concurrent decrease of both the choroidal and the retinal blood flow in older individuals may have a role in the development of age-related macular degeneration. Further assessment of the circulation of these vascular beds in patients with this disease are needed to support or refute this speculation.
Accepted for publication October 20, 1997.
This investigation was supported by the Vivian Simkins Lasko Research Fund Philadelphia, Pa, the Nina C. Mackall Trust New York, NY, an unrestricted grant from Research to Prevent Blindness, New York, The Pennsylvania Lions Sight Conservation and Eye Research Foundation, Philadelphia, and by grant NEI R21 EY10964.4 from the National Institutes of Health, Bethesda, Md.
We thank Maureen Maguire, PhD, for her statistical consultation which was supported through grant NEI R21 EY10964 from the National Institutes of Health.
Reprints: Juan E. Grunwald, MD, Scheie Eye Institute, 51 N 39th St, Philadelphia, PA 19104.