The Effect of Caffeine on the Human Macular Circulation

Size: px

Start display at page:

Download "The Effect of Caffeine on the Human Macular Circulation"

Silvia Cross
5 years ago
Views:

1 Investigative Ophthalmology & Visual Science, Vol. 32, o. 12, ovember 1991 opyright Association for Research in Vision and Ophthalmology The Effect of affeine on the Human Macular irculation Karan Lotfi and Juan E. Grunwald The acute effect of caffeine on the retinal circulation was studied in 1 healthy volunteers using the blue field simulation technique, which provides measurements of the velocity of leukocytes flowing within the macular capillaries. Subjects adjusted the mean velocity (V m ) of computer-simulated leukocytes moving on a cathode ray tube screen to match that of their own entoptically perceived leukocytes before and 1 hr after a double-masked, randomized administration of 200 mg caffeine or placebo. affeine produced an average 13% ± 5% (SEM) decrease in V m { < 5) and a 9% ± 3% increase in diastolic blood pressure ( < 5). The decrease in V m and, presumably, blood flow occurring despite the increased diastolic blood pressure probably is attributable to retinal vasoconstriction. This effect may result from caffeine's known inhibitory effect on adenosine, a potent vasodilator of the retinal vasculature. Invest Ophthalmol Vis Sci 32: ,1991 affeine (1,3,7-trimethylxanthine) i s present in a wide variety of beverages, foods, and over-the-counter drugs. 1-2 Approximately 20-30% of the general population ingests more than mg of this drug daily, an amount that is roughly equivalent to 2-3 cups of regular coffee. 3 affeine-containing beverages and foods are served in most hospitals around the world, and patients of all age groups with a wide variety of diseases consume caffeine. affeine acts as a central nervous system stimulant. It produces diuresis and an increase in the basal metabolic rate. This alkaloid has been reported to increase blood pressure and decrease heart rate. 5 ' 6 Mathew and Wilson 7 showed that caffeine decreases cerebral blood flow by approximately 15%, and Higginbotham et al 8 found that it produces a small increase in intraocular pressure (IO) of about 1 mm Hg in glaucoma patients. The effect of caffeine on the human retinal circulation has not been studied previously. Because this is a widely used substance, any circulatory effect could be important, especially in people with compromised retinal circulation. In this study, we have used the blue field simulation (BFS) technique to determine whether caffeine taken in a dose comparable to normal daily consumption levels has any effect on retinal circulation. From the Scheie Eye Institute, Department of Ophthalmology, University of ennsylvania School of Medicine, hiladelphia, ennsylvania. resented at the ARVO Annual Meeting, Sarasota, Florida, April 22-May, Submitted for publication: January, 1991; accepted June 21, Reprint requests: Juan E. Grunwald, Scheie Eye Institute, 51 orth 39th Street, hiladelphia, A 19. Materials and Methods Fourteen normal volunteers with no history of hypertension, diabetes, or other systemic diseases participated in this study. Ages ranged from 20 to 1 yr (mean ± SEM, 2 ± 1 yr). There were 3 females and 11 males. The right eye was chosen as the study eye. All eyes studied had a best refracted visual acuity of 6/6 or better, normal IO, clear media, and normal fundi. one of the volunteers had taken topical ocular medication for a period of at least 1 month before the study. Their average daily caffeine consumption was the equivalent of one cup of coffee. ersons with a daily caffeine consumption of three or more cups of coffee were excluded from our study. In addition, all subjects abstained from the ingestion of any caffeine for at least 2 hr before the study. o cigarette smokers were included in this study. Informed consent was obtained from each volunteer after the nature of the study was explained. Subjects were seated in a darkened room in front of a blue field entoptoscope and a cathode ray tube (RT) screen. The entoptoscope provides diffuse and uniform maxwellian illumination of the retina at a wavelength of 30 nm. When this illumination is centered at the fovea, optimal entoptic visualization of the leukocytes flowing in one's own macular capillaries 9 is obtained. The diameter of the illumination field is approximately 2 degrees, and the intensity level needed to observe this phenomenon is well below the maximal permissible level of retinal irradiance. In the macular microcirculation, where capillaries measure 7- nm in diameter, the velocity of leukocytes is equal (within a few percent) to the mean velocity of whole blood. 11 This velocity can be assumed to represent flow because these capillaries probably maintain a constant diameter

o. 12 EFFET OF AFFEIE O HUMA MAULAR IRULATIO / Lorfi and Grunwald 3029 Quantification of the mean velocity (V m ) and number () of these leukocytes was made possible by a simulation technique in

2 o. 12 EFFET OF AFFEIE O HUMA MAULAR IRULATIO / Lorfi and Grunwald 3029 Quantification of the mean velocity (V m ) and number () of these leukocytes was made possible by a simulation technique in which subjects adjusted the number and the diastolic and systolic velocities of computer-simulated leukocytes moving on a RT screen (13B Hewlett-ackard o., alo Alto, A) to match those of their own entoptically perceived leukocytes. 13 After a demonstration of the blue field entoptic phenomenon, subjects were asked to describe the number, shape, movement pattern, and velocity of the particles observed. The concept of pulsatility was explained to them. A BFS trial consisted of matching the velocity, number, and pulsatility of a simulated set of particles on a RT to those of the entoptically observed leukocytes using three potentiometers. On the first baseline trial, as well as the first trial performed 1 hr after drug ingestion, the subjects adjusted velocity, pulsatility, and number by alternately looking at their entoptic phenomenon and the RT simulation. On all subsequent trials, the velocity was scrambled by the computer, whereas the number and pulsatility were left unchanged, and subjects adjusted only the velocity. Experimental rocedure Subjects performed five practice BFS trials followed by a -min relaxation period. In the next phase of the experiment, the subjects' brachial artery blood pressure was measured twice by standard sphygmomanometry, and the mean of these two determinations was documented. Heart rate (HR) also was measured twice manually using 15-sec counting intervals, and the average was recorded. Ten BFS trials were obtained and averaged to establish a baseline. Afterward, B and HR were measured as described while the subjects remained seated. Goldmann applanation tonometry was performed on the study eyes, and cc of blood was withdrawn for serum caffeine level determination (EMIT caffeine assay, Syva o., alo Alto, A). Subjects were given either a 200-mg caffeine capsule or a placebo capsule (200 mg lactose powder) in a randomized double-masked manner. After 1 hr, the entire protocol was repeated (without the five familiarization trials). Two or more days later, the experimental procedure was repeated with the other capsule. Mean blood pressure, B m, was calculated according to the formula B m = B d + 1/3(B. - B d ) for which B S and B d are the brachial artery systolic and diastolic pressures, respectively. erfusion pressure,, for the study eye was determined according to the formula = 2/3B m - IO Results were analyzed using paired Student's r-test, Wilcoxon signed rank test, and correlation analysis. Findings with an error probability of less than 5 were considered to be statistically significant. Because there was a large variability in baseline V m between subjects, we performed an analysis of the relative changes in V m from baseline (percentage changes). Both absolute and percentage changes in V m are reported. Results The results of the experiments are summarized in Tables 1 and 2. After caffeine ingestion, average V m Table 1. Measurements of mean leukocyte velocity (V m, cm/sec) and leukocyte number () at baseline and 1 hr after ingestion of placebo and caffeine lacebo affeine / hour 1 hour atient no. Age (yr) ±7 1.1 ± ± ±3 1. ± ± ± ± 1.30 ± ± ± 1.90 ± ± ± ± ± ().88 ± ±5 253 ().99 ± ± 5 13 ().88 ± ± ± ± ±5 5.6 ± ± ± ± 1. ± ±2 1.5 ± 0.71 ± ± 0.76 ± ± ± 0.80 ± ±6 1.5 ±3 0.8 ± ± ±2 1. ±5 1. ± ± ± ± 0.7 ± ± ± ± ± ± ± ±

3 3030 IVESTIGATIVE OHTHALMOLOGY 6 VISUAL SIEE / ovember 1991 Vol. 32 Table 2. ercentage changes from baseline in heart rate (HR), intraocular pressure (IO), diastolic blood pressure (Bd), systolic blood pressure (Bs), mean blood pressure (Bm), and perfusion pressure () 1 hr after placebo () and caffeine () ingestion atient no. HR IO c Bd Bs Bm decreased significantly by 0.1 ± mm/sec (SEM) from baseline ( < 5); this corresponds to an 8% ± 3% decrease (paired f-test and Wilcoxon signed rank test, < 5) (Fig. 1). After placebo ingestion, there was a nonsignificant average increase of 2 ± 5 mm/sec from baseline ( > 5), corresponding to a % ± 5% increase (> 5). When compared with the placebo effect, there was a significant average decrease in V m of 13% ± 5% 1 hr after caffeine ingestion ( < 5) (Fig. 2). The differences between the percentage change found after caffeine administration and that found after placebo administration for each subject were used in performing the last mentioned paired /-test and Wilcoxon signed rank test. A significant increase from baseline in average (number of leukocytes) of 5 ± 11 ( < 1) corresponding to a 28% ± 6% ( < 5) change was ob- served after caffeine ingestion. A nonsignificant average increase in from baseline of 20 ± 16 ( > 5) corresponding to a 17% ± 8% change was present after placebo ingestion ( > 5) (Table 1). The average percentage change in observed after caffeine ingestion was not significantly different from the average percentage change in after placebo ingestion ( > 5). Also, no significant changes in the ratio between systolic and diastolic velocities were observed after placebo or caffeine ingestion r - O I i " lacebo affeine Fig. 1. omparison of the percentage change from baseline in mean leukocyte velocity (V m ) 1 hr after ingestion of placebo or caffeine for the 1 subjects. 1 Hour -20?. 3 < Fig. 2. ercentage change from baseline in mean leukocyte (V m ) 1 hr after caffeine ingestion relative to placebo. In comparison with placebo, there was an average 13 ± 5% (SEM) decrease in V m after caffeine intake (paired t-test and Wilcoxon's signed rank test, < 5).

o. 12 EFFET OF AFFEIE O HUMA MAULAR IRULATIO / Lorfi and Grunwold 3031 After caffeine ingestion there were significant average increases of 12% ± 2% ( < 01) in B d, and % ± 1 % ( < 1) in B m (Fig. 3).

4 o. 12 EFFET OF AFFEIE O HUMA MAULAR IRULATIO / Lorfi and Grunwold 3031 After caffeine ingestion there were significant average increases of 12% ± 2% ( < 01) in B d, and % ± 1 % ( < 1) in B m (Fig. 3). There was a nonsignificant increase of 2% ± 1% ( > 5) in B S. o significant changes in B d, B m, and B S were observed after placebo ingestion. In comparison with the changes of placebo ingestion, a significant average increase of 9% ± 3% ( < 5) in B d was detected after caffeine ingestion. Heart rate decreased significantly by 1% ± 2% ( < 01) after caffeine ingestion and by 12% ± 2% ( < 01) after placebo ingestion (Table 2). However, the difference between the average percentage changes in HR after caffeine and placebo ingestion was not significant. Also, no significant changes in IO were detected (Table 2). There was a significant average % ± 2% ( < 5) increase from baseline in perfusion pressure () after caffeine ingestion. There was no significant change in average after placebo ingestion (0.2% ± 2%, > 5). In comparison with placebo ingestion, there was no significant change in average after caffeine intake. There were no significant correlations between the changes in V m and the changes in B m, IO, or m after caffeine ingestion. o significant correlation was observed between changes in V m and age. Also, no significant correlations were observed between baseline V m and B m, B S, or B d. The average blood caffeine level 1 hr after caffeine ingestion was. ± 0.2 mg/dl (range mg/dl). o significant correlation was seen between the changes in V m and the blood caffeine levels. Discussion Our results show that in comparison with the changes observed after placebo ingestion, there is a 13% reduction in V m and presumably retinal blood flow after caffeine intake ( < 5). Our study was designed to investigate the effects of caffeine on V m. Subjects were asked to determine V m.3 0 o D LAEBO M AFFEIE ESB AFF-LA Fig. 3. ercentage changes from baseline in diastolic, systolic, and mean blood pressure after placebo or caffeine ingestion. AFF- LA represents the percentage changes after caffeine minus the percentage changes after placebo. Error bars represent standard error. * < 5; ~ < 01. in ten successive trials at baseline and ten trials after the drug administration. The velocity of the simulated particles was randomly changed by the computer after each trial was completed, thus forcing subjects to make new velocity adjustments. However, adjustments of the number of entoptically observed leukocytes () were performed only on the first of the ten baseline trials and on the first of the ten trials after drug administration. During the following nine trials, was unchanged. Despite that values represent only one trial adjustment, we found a significant increase in of 28% ± 6% ( < 1) after caffeine intake and a nonsignificant increase of 17% ± 8% ( > 5) after placebo ingestion. However, the difference between the increase in after placebo and caffeine ingestion was not significant, so we cannot conclude that caffeine produced this effect. Average heart rate decreased significantly after both caffeine (-1% ± 2%, < 01) and placebo (-12% ± 2%, < 01) ingestion. Also, in this case the changes after caffeine intake were not significantly different from those seen after placebo ingestion, and we could not conclude that caffeine produced this effect. It is possible that the experimental conditions in which subjects may have been more relaxed and less nervous during the 2 hr after drug ingestion could have resulted in the decrease in heart rate seen after both placebo and caffeine intake. The purpose of performing placebo trials is to control for nondrug-related effects that may influence the measured variables during the experiments. affeine produced a significant increase in diastolic blood pressure. An increase in diastolic blood pressure would lead to an increase in V m if the vascular resistance would remain unchanged. Our results showing a decrease in V m after caffeine intake suggest the presence of retinal vasoconstriction. An interaction of caffeine with endogenous adenosine may be related to this vasoconstriction. Adenosine, an endogenous purine metabolic end product with a potent vasodilatory effect on multiple vascular beds, has been shown to produce an increase in retinal vessel diameter in the rabbit retina. 1 affeine, a xanthine derivative, is known to bind adenosine receptors, thus preventing the vasodilation mediated by adenosine. 15 hillis and DeLong, 16 for example, have shown that caffeine can prevent the adenosine-mediated increase in cerebral bloodflow that occurs during hypercapnia in the rat. The decrease in V m observed in our study after caffeine ingestion could result from blockage of the adenosine vasodilatory influence on the retinal circulation. Mathew and Wilson 7 have shown that caffeine decreases cerebral blood flow by approximately 15%. Because the retina is neural tissue, it is not surprising

3032 IVESTIGATIVE OHTHALMOLOGY & VISUAL SIEE / ovember 1991 Vol. 32 that the circulatory changes produced by caffeine in the retina are similar to those observed in cerebral tissue.

5 3032 IVESTIGATIVE OHTHALMOLOGY & VISUAL SIEE / ovember 1991 Vol. 32 that the circulatory changes produced by caffeine in the retina are similar to those observed in cerebral tissue. In conclusion, our study shows results suggesting that caffeine produces a significant decrease in V m and, presumably, blood flow. Such an effect may be of importance in patients with compromised retinal circulation. Key words: retinal blood flow, caffeine, blue field simulation technique, normal humans, leukocyte velocity Acknowledgments The authors thank Joan Baine and andace Furubayashi for their help with the protocol of this study and Dolly Scott for her preparation of the manuscript. References 1. Gilbert RM: affeine as a drug of abuse. In Research Advances in Alcohol and Drug roblems, Gibbins RJ, Israel Y, Kalant H, and opham RD, editors. ew York, John Wiley and Sons, 1976, pp Bunker M and McWilliams M: affeine content of common beverages. J Am Diet Assoc 7:28, Gilliland K and Andress D: Adlib caffeine consumption, symptoms of caffeinism and academic performance. Am J sychol 138:512, Ritche JM: entral nervous system stimulus: The xanthines. In The harmacological Basis of Therapeutics, 5th ed., Goodman L and Gilman A, editors.,ew York, MacMillan ublishing o., Inc., 1975, pp Whitsett TL, Manion V, and hristensen HD: ardiovascular effects of coffee and caffeine. Am J ardiol 53:918, Smits, Thien TH, and Van'tlaar A: The cardiovascular effects of regular and decaffeinated coffee. Br J harmacol 19:852, Mathew RJ and Wilson WH: affeine induced changes in cerebral circulation. Stroke 5:81, Higginbotham EJ, Kilimanjaro HA, Wilensky JT, Batenhorst RL, and Herman D: The effect of caffeine on IO in glaucoma patients. Ophthalmology 5:62, Sinclair SH, Loebl M, and Riva E: Blue field entoptic phenomenon in cataract patients. Arch Ophthalmol 97:92, Riva E, Kelley JJ, Sinclair SH, and Loebl M: Optical transmission of cataractous lens at 30 nm and blue field entoptoscopy. Vision Res 19:1181, Schmid-Schonbein GW, Skalak R, Usami S, and hein S: ell distribution in capillary networks. Microvasc Res 19:18, Friedman E, Smith TR, and Kuwabara T: Retinal microcirculation in vivo. Investigative Ophthalmology 3:217, Riva E and etrig BL: Blue field entoptic phenomenon and blood velocity in the retinal capillaries. J Opt Soc Am 70:1, ampochiaro A and Sen H: Adenosine and its agonists cause vasodilatation and hemorrhages. Implications for ischemic retinopathies. Arch Ophthalmol 7:12, Fredholm BB and ersson G: Xanthine derivatives as adenosine receptor antagonists. Eur J harmacol 81:673, hillis JW and DeLong RE: An involvement of adenosine in cerebral blood flow regulation during hypercapnia. Gen harmacol 2:133, 1987.

Effect of Timolol Maleate on the Retinal Circulation of Human Eyes with Ocular Hypertension

Effect of Timolol Maleate on the Retinal Circulation of Human Eyes with Ocular Hypertension Investigative Ophthalmology & Visual Science, Vol. 31, No. 3, March 1990 Copyright Association for Research in Vision and Ophthalmology Effect of Maleate on the Retinal Circulation of Human Eyes with Ocular