Retinal Hemodynamic Effects of Carbon Dioxide, Hyperoxia, and Mild Hypoxia

Transcription

1 Investigative Ophthalmology & Visual Science, Vol. 33, No. 6, May 1992 Copyright Association for Research in Vision and Ophthalmology Retinal Hemodynamic Effects of Carbon Dioxide, Hyperoxia, and Mild Hypoxia William E. Sponsel, Kathleen L. DePaul, and Sandy R. Zerlan Retinal leukocyte velocity and density were estimated using blue-field entoptic imaging techniques in a controlled double-masked study to determine the relative effects of oxygen and carbon dioxide on perimacular hemodynamics in single eyes of ten normal human subjects. Mild hypoxia (inspiration of 16% O 2 ) did not significantly alter leukocyte velocity or density from room-air baseline levels. Supplementing 16% oxygen with 5% CO 2 produced a tendency toward increased leukocyte velocity (+23%, P = 0.027) with no apparent effect on leukocyte density. Inspiration of pure oxygen was associated with significant reductions in both retinal leukocyte velocity (-20%, P < 0.007) and density (-23%, P = 0.013) relative to room-air baseline levels. Supplementation of pure oxygen with 5% CO 2 appeared to produce a dramatic change in perimacular hemodynamics, tending to increase leukocyte velocity (+26%, P = 0.018) with a limited density change (-11%, P = 0.049). These findings suggest that inspired 5% CO 2 can counteract the profound inhibitory effects of excess oxygen on retinal hemodynamics in the functionally important perimacular capillary bed. Invest Ophthalmol Vis Sci 33: ,1992 Excessive oxygenation can cause dramatic retinal arteriolar constriction with decreased perimacular capillary perfusion. 1 This effect may contribute to the development of retinopathy of prematurity in neonates. The general vasodilatory effect of carbon dioxide (CO 2 ) is well known, but it was found that this action did not extend to the visible retinal arteries and arterioles. 2 During breathing of oxygen-depleted isocapnic air, leukocyte velocity in the perimacular capillary loops of the retina increases. 3 The extent, if any, to which CO 2 can modulate the inhibitory effects of high levels of oxygen (O 2 ) on the retinal circulation is uncertain. If inhalation of pure O 2 reduces retinal blood velocity, and CO 2 has no visible vasodilatory effects, we might question the rationale of the clinically advocated administration of a high O 2 and high CO 2 mixture, such as carbogen, to patients with acute embolic retinal ischemia. In this report, we investigated the relative effects of CO 2 and O 2 on the retina and correlated hemodynamic and respiratory changes From the Department of Ophthalmology, University of Wisconsin-Madison, Madison, Wisconsin. Supported by a grant from Chibret International, Rahway, New Jersey. Submitted for publication: December 20, 1990; accepted November 27, Reprint requests: William Eric Sponsel, MD, Department of Ophthalmology, Indiana University, 702 Rotary Circle, Indianapolis, IN observed during breathing of pure O 2 and rarified oxygen mixtures, with and without supplemental CO 2. Materials and Methods Participants in this study included ten healthy adult subjects with no history of cardiovascular or respiratory dysfunction. The study population included five men and five women (age range, yr; mean age, 31 yr). Informed consent was obtained from each subject under an approved experimental protocol (UW HSC ). Measurements of macular leukocyte velocity and density were made during breathing each of the following gas mixtures: (1)100% O 2, (2) 5% CO 2 and 95% O 2 (carbogen), (3) 21% O 2 and 79% nitrogen (tanked air), (4) 16% O 2 and 84% nitrogen (low O 2 ), and (5) 16% O 2, 5% CO 2, and 79% nitrogen (hypoxycarb). All gases, including room air for baseline studies, were administered with nasal occlusion through a Rudolf 2700 two-way nonrebreathing valve (Hans Rudolph, Kansas City, MO) in accordance with standard safety specifications. All tanked gases were delivered through a partially expanded 5-1 reservoir bag at atmospheric pressure. The rarified oxygen mixtures were formulated to approximate the inspired oxygen levels that would be experienced under normal alpine conditions and constituted no hazard to healthy adult volunteers. Hypoxycarb contains approximately the 1864

2 No. 6 RETINAL HEMODYNAMIC EFFECTS OF CO 2 AND O 2 / Sponsel er ol 1865 same concentrations of O 2 and CO 2 as would be inhaled during simple paper-bag rebreathing under normal atmospheric conditions. 4 Retinal leukocyte velocity and density in the perimacular capillary bed (a 14 zone surrounding fixation) were estimated subjectively using Riva and Petrig's 5 Oculix 1000 system of blue-field entoptic illumination (Berwyn, PA) with a concurrent computer simulation. Aided by a single trained operator, the subjects matched the pulse-synchronized computer simulation of retinal leukocytes with those of their own cells. The order of adjustment was the same in all subjects during each trial: density, velocity, and pulsatility. After each reading, the simulation density and velocity were randomized. Subjects were encouraged not to change the pulsatility setting unless it interfered with their ability to make velocity and density comparisons. To help determine intrasubject variability, five simulation-versus-simulation comparisons were completed on five subjects. These subjects compared a variable computer simulation of randomly moving leukocytes with a standardized computer simulation. The mean coefficient of variability for the five subjects was 2.6%. No significant differences were found between the baseline variabilities of these and the other five subjects during the gas-breathing studies. The blue-field entoptic leukocyte method has been used extensively in physiologic and pharmacologic studies elsewhere. 35 " 12 Retinal hemodynamic measurements were made on one eye only (randomized with five right and five left). A set of five baseline readings while breathing room air was followed immediately by a set of five readings while breathing one of the gas mixtures. Subjects received gases in a randomized sequence, each preceded by a new set of baseline readings. The gases were inspired for 2 min before initiation of the bluefield measurements, which required min each. A minimum of 15-min rest followed testing with each gas mixture. Right brachial artery blood pressures were measured twice during each gas administration using a digital sphygmomanometer, and the mean pressure was calculated as diastolic (systolic - diastolic)]. Exhaled tidal volume and respiratory rate were monitored continuously during the study by the Ohmeda 5410 volume monitor (Madison, WI). Pulse rate and oxygen saturation values were measured by POET noninvasive finger pulse oxymetry (Criticare, Milwaukee, WI). End-tidal CO 2 levels also were monitored using the POET system and the B&K 2800 digital multimeter (Dynascan, Chicago, IL). Mechanical failure prevented readings of end-tidal CO 2 for one of the ten subjects during part of one experimental session (three of the gas mixtures). The measurements wercrecorded at 2-min intervals. A double-masked randomized experimental protocol was used in which both the subject and the operator of the blue-field simulator were unaware of which tanked gas was being inspired. Gas tanks remained behind a curtain throughout the entire study. An assistant connected gas mixtures in accordance with a predetermined randomization code and recorded respiratory responses on a written flow sheet. All subjects were unaware of the putative ocular circulatory effects of CO 2 and O 2. Retinal leukocyte velocity (in millimeters per sec) and density (in cells per 14 radius of the central entoptic field) each were estimated by averaging a set of five readings. The mean velocity for the ten sets of five baseline readings was 0.91 mm/sec. The mean coefficient of variation for the velocity during baseline measurements (each baseline was the average offivereadings per individual) by individuals was 14% (range, %), and by previous exposure to a particular gas, it was 14% (range, %). The mean coefficient of variation for the velocity during breathing of tanked gases by individual was 14% (range, %), and by gas, it was 15% (range, %). The changes in retinal leukocyte velocity and density are presented in terms of the percent change from the corresponding baseline value (% A velocity and %Adensity, respectively). The percent change values were determined by dividing the change from baseline by the baseline value. Probability values were determined using a paired two-tailed student t-test comparing percent change for each parameter to zero change. End-tidal CO 2 measurements are presented as the difference between tanked gas and room air, or AETCO 2. For each of the three responses (% A velocity, %Adensity, and AETCO 2 ), a paired two-tailed student t-test was used to determine the significance of any observed differences between the following pairs of gases: (1) 100% O 2 and 16% O 2, (2) carbogen and 100% O 2, (3) hypoxycarb and 16% O 2, and (4) tanked air and room air. Because four comparisons were made for each variable, the Bonferroni correction for multiple comparisons was used to determine the significance level (alpha = 0.05/4 = 0.013). Results Table 1 summarizes the differences between baseline measurements and those observed while breathing the corresponding test gas for leukocyte velocity and density (in terms of percent change) and end-tidal CO 2 (in mm Hg). Table 2 summarizes the differences

3 1866 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / May 1992 Vol. 33 Table 1. Percent change (%A) in macular leukocyte velocity and density and absolute change (A) in end-tidal CO 2 from baseline* while breathing prepared gas mixtures Carbogen (5% CO 2, 95% OJ Pure O 2 (100% O 2 ) Tanked air (21% O 2 ) LowO 2 (16% O 2 ) Hypoxycarb (5% CO 2, 16% OJ %A Velocity %A Density AETCO 2 (mm Hg) ± 9.0 P = ±4.7 P = ± 1.5f P< ±5.7 P = ±7.5 P = ± 0.4f P< ± 2.6 P = ± 5.9 P = ± 1.2 P = ± 6.3 P = ±8.4 P = ±0.4f P = ± 8.5 P = ±6.8 P = ± 0.9 P< Average baseline values were 0.91 mm/sec for perimacular retinal leukocyte velocity, 187 cells/field for perimacular retinal leukocyte density, and 38 mm Hg for ETCO 2. * Values for means ± standard error of the mean for 10 subjects or fnine subjects. for the three responses between four pairs of gas mixtures. Comparing tanked air to room air indicates that there were no significant effects of the breathing apparatus on the measured responses. Table 3 presents the individual values of % A velocity and AETCO 2 by gas. Table 4 presents the group mean and standard deviation for leukocyte density, blood pressure, percent oxygen saturation, pulse, respiration, and minute volume during breathing of room air or tanked gas. Retinal leukocyte velocity was decreased significantly (P = 0.002) during inspiration of 100% O 2 compared with 16% O 2. The velocity was higher (P < 0.001) while breathing carbogen (95% O 2 and 5% CO 2 ) than while breathing 100% O 2. Retinal leukocyte velocity did not differ significantly when comparing hypoxycarb (16% O 2 and 5% CO 2 ) with 16% O 2. No significant differences in retinal leukocyte density were found between any of these pairs of gas mixtures. The end-tidal CO 2 was increased significantly (P < 0.001) during breathing of the high CO 2 mixtures compared with the corresponding mixture without CO 2 (hypoxycarb versus 16% CO 2, and carbogen versus 100% O 2 ). The end-tidal CO 2 was decreased (P = 0.002) on 100% O 2 compared with 16% O 2. To explore the possibility that the changes in retinal leukocyte velocity resulted from changes in end-tidal CO 2 levels, correlations were done between these two responses for all gas mixtures. During breathing of 100% O 2, a significant (r = 0.91, P < 0.001) correlation was found between AETCO 2 and % A velocity. The correlation between these responses was not significant during breathing of any of the other gas mixtures. Multiple regression was done to determine if other factors (ie, changes in blood pressure, oxygen saturation, pulse rate, respiratory rate, and minute volume) contributed to the observed decrease in velocity while the subject was receiving pure O 2. The raw data are presented in Table 4. The final regression model demonstrated that AETCO 2 (P < 0.001) and Arespiratory rate (P = 0.015) both were associated significantly with the % A velocity during pure oxygen breathing. Discussion The blue-field entoptic phenomenon has been ascribed to leukocyte motion in the retinal capillary loops, such as those identified in trypsin-digested preparations. 13 This recently was reaffirmed experimentally using video microscopy. 14 Oxygen, which promotes local vasoconstriction of large arteries and veins in the retina, 2 does not alter total ocular blood flow significantly (only 1-2% of which is accounted for by the retinal circulation). 15 In some instances, CO 2 does not appear to cause vasodilation of visible retinal arterioles, 216 although it increases total retinal bloodflow. 15 However, even small increases in vascular diameter precipitated by breathing 5% CO 2 in O 2 Table 2. Difference and significance level* for the percent change (%A) in macular leukocyte velocity and density and absolute change (A) in end-tidal CO 2 between 100% O 2 and 16% O 2, carbogen and 100% O 2, hypoxycarb and 16% O 2, and tanked air and room air 700% O 2-16% O 2 Carbogen-100% O 2 Hypoxycarb-16% O 2 Tanked air-room air %A Velocity %A Density AETCO , P = , P = , P = , P< , P = , P< , P = , P = , P< , P = , P = , P = * Calculated using paired two-tailed Student's t-test. Bonferroni correction for multiple comparisons limits significance to P values ^

4 No. 6 RETINAL HEMODYNAMIC EFFECTS OF CO 2 AND O 2 / Sponsel er ol o 00 n o u 'o g So s 1 "1 S lit pi a: m ^ a; H ^ Is""? 1 SIS I m Ill 1 its o d ' d oor~-r--rnoooo IN * >o *^roooo^qpfnr-tnoso\r^ d o d o d o d d mr^vorn^ooo^oooofn^-oo r^mo^ootnr- vooovoo<n o d o d - d d o d d do p ; d <s -^ o o d d oq o\ oq r~ r- Oi~-O o *O O O O r <N oq r- q ; d o d o o d d : d d o ; ; i ; ; d o o d - d - o - d o r t O O o O P O O f (N O r~- u-i vo O u^ <^) u-i d <N ' r4 rn r- >ri r- od do d -d do do -d - d c> d ^ Osr t >o ^ oo m N \oo «iinoo'^if < io <N r- <N CN f> o r- o\ r-' d -* <i ^6 & r-' : d d c +1 may have proportionally larger effects on blood flow. 16 Although the retinal microvasculature has no classic bifurcatory precapillary sphincters, 17 it is under strong local autoregulatory control. 15 Trypsin-digest studies established that the arteriolar end of the retinal capillary bed has a uniquely uniform cytoarchitecture, with a 1:1 ratio between pericyte and endothelial nuclei. Subsequent studies indicate that these pericytes contain actin 20 and are capable of modulating capillary dimensions in response to experimentally induced distal venous obstruction. 21 Moreover, experiments by others showed that adenosine triphosphate can induce contraction in these retinal pericytes in vitro. 22 Recent primate studies suggest that pericytic density on retinal capillaries is substantially greater than that found on cerebral capillaries. 23 A vascular bed so endowed with contractile pericapillary elements might be expected to be exquisitely responsive to local metabolic stimuli, which might include ph, O 2, and/or CO 2 levels. Recently, it was found that leukocyte velocity did not differ significantly in leukemic, leukopenic, and normal control subjects, although density varied in accordance with the leukocyte count. 10 Such findings illustrate the independent nature of velocity and density, and these authors suggested microvascular autoregulation as a compensatory mechanism to maintain constant velocity under widely differing conditions of leukocyte density. Independent variations in density and velocity also were encountered repeatedly in our study under the experimental respiratory conditions we used. The responses to inspired O 2 we found were consistent with those reported elsewhere using a similar technique. 3 ' 24 The 20% decrease in velocity could arise from increased vascular resistance and/or reduced perfusion pressure. Laser Doppler 1 and photofluorescent studies 25 reveal significant reductions in blood velocity and vessel diameter in response to O 2, suggesting estimated decreases in blood flow of 64.3% and 40%, respectively. Our study was designed to determine the ability of CO 2 to modulate the retinal circulatory effects of O 2 excess. Our results show that in the presence of O 2 excess, CO 2 can exert a powerful stimulating effect on retinal capillary circulation, more than reversing the effect of O 2 on perimacular leukocyte velocity. Retinal leukocyte velocity changes seen during O 2 breathing alone may themselves in part be related to decreased CO 2 levels (or secondary changes in ph). 26>27 During inspiration of pure O 2, when both %Aleukocyte velocity and AETCO 2 are decreased in comparison with values obtained during breathing 16% O 2, there is a positive correlation between these two re-

5 1868 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / May 1992 Vol. 33 Table 4. values and standard deviations for macular leukocyte density, blood pressure, oxygen saturation, pulse rate, respiration rate, and minute volume during administration of each of the five gases. Density Density BP BP SaO 2 SaO 2 Pulse Pulse Resp. Resp. Minvol Minvol Carbogen 16% Oxygen 100% Oxygen Tank air Hypoxycarb sponse variables. Hypocarbia accompanying O 2 excess, as observed in our study, has been well documented elsewhere in the respiratory literature, including early studies relating retrolental fibroplasia to O 2 inspiration. 27 In summary, our findings confirm that excess O 2 can reduce the velocity of leukocytes passing through the perimacular capillary network substantially, but supplemental CO 2 can counteract this O 2 -induced circulatory restriction effectively in healthy volunteers. Additional studies may clarify the role, if any, of CO 2 therapy or carbonic-anhydrase inhibitors in the management of retinal ischemic diseases. Key words: blue-field entoptic phenomenon, ocular circulation, macular hemodynamics, carbon dioxide, oxygen Acknowledgments The authors thank Paul Montague, David Off, and William Backes of the UW Department of Respiratory Therapy; Drs. Matthew D. Davis, Ingolf H. L. Wallow, and Paul L. Kaufman of the UW Department of Ophthalmology; Dr. James Skatrud of the UW Department of Pulmonary Medicine for their assistance; Dr. Juan Grunwald of the Scheie Eye Institute who provided protocol design assistance; and Alice Rogot, MS, of the UW Biostatistics Department who provided assistance with statistical analysis. References 1. Riva CE, Grunwald JE, and Sinclair SH: Laser Doppler velocimetry study of the effect of pure oxygen breathing on retinal blood flow. Invest Ophthalmol Vis Sci 24:47, Deutsch TA, Read JS, Ernest JT, and Goldstick TK: Effect of oxygen and carbon dioxide on the retinal circulation in man. Arch Ophthalmol 101:1278, Fallon TJ, Maxwell D, and Kohner EM: Retinal vascular autoregulation in conditions of hyperoxia and hypoxia using the blue field entoptic phenomenon. Ophthalmology 92:701, American Heart Association: A Manual for Instructors of Basic Cardiac Life Support. Dallas, American Heart Association, 1977, p Riva CE and Petrig B: Blue field entoptic phenomenon and blood velocity in the retinal capillaries. J Opt Soc Am [A] 70:1234, Davies EG, Hyer SL, and Kohner EM: Macular blood flow response to acute reduction of plasma glucose in diabetic patients measured by the blue light entoptic technique. Ophthalmology 97:160, Fallon TJ, Maxwell DL, and Kohner EM: Autoregulation of retinal blood flow in diabetic retinopathy measured by the blue-light entoptic technique. Ophthalmology 94:1410, Fallon TJ, Sleightholmm MA, Merrick C, Chahal P, and Kohner EM: The effect of acute hyperglycemia onflowvelocity in the macular capillaries. Invest Ophthalmol Vis Sci 28:1027, Fallon TJ, Chowiencyzk P, and Kohner EM: Measurement of retinal blood flow in diabetes by the blue-light entoptic phenomenon. Br J Ophthalmol 70:43, Rimmer T, Kohner EM, and Goldman JM: Retinal blood velocity in patients with leukocyte disorders. Arch Ophthalmol 106:1548, Rimmer T, Fallon TJ, and Kohner EM: Long-term follow-up of retinal blood flow in diabetes using the blue light entoptic phenomenon. Br J Ophthalmol 73:1, Robinson F, Petrig BL, and Riva CE: The acute effect of cigarette smoking on macular capillary blood flow in humans. Invest Ophthalmol Vis Sci 26:609, Cavender JC: Entoptic imagery and afterimages. In Biomedical Foundations of Ophthalmology, Duane TD, editor. Philadelphia, Harper and Row, 1987, pp Sinclair SH, Azar-Cavanagh M, Soper KA, Tuma RF, and Mayrovitz HN: Investigation of the source of the blue field entoptic phenomenon. Invest Ophthalmol Vis Sci 30:688, Henkind P, Hansen RI, and Fzalay J: Ocular circulation. In Biomedical Foundations of Ophthalmology, Duane TD, editor. Philadelphia, Harper and Row, 1987, pp Miles FP and Nuttall AL: In vivo capillary diameters in the stria vascularis and spiral ligament of the guinea pig cochlea. Hear Res 33:191, 1988.

6 No. 6 RETINAL HEMODYNAMIC EFFECTS OF CO 2 AND O 2 / Sponsel er ol Friedman E, Smith TR, and Kuwabara T: Retinal microcirculation in vivo. Invest Ophthalmol 3:217, Cogan DG and Kuwabara T: The mural cell in perspective. Arch Ophthalmol 78:133, Kuwabara T and Cogan DG: Studies of retinal vascular patterns: I. Normal architecture. Arch Ophthalmol 64:904, Wallow IHL and Burnside B: Actin filaments in retinal pericytes and endothelial cells. Invest Ophthalmol Vis Sci 19:1433, Danis RP and Wallow IHL: Microvascular changes in experimental branch retinal vein occlusion. Ophthalmology 94:1213, Das A, Frank RN, Weber ML, Kennedy A, Reidy CA, and Mancini MA: ATP causes retinal pericytes to contract in vitro. Exp Eye Res 46:349, Frank RN, Turczyn TJ, and Das A: Pericyte coverage of retinal and cerebral capillaries. Invest Ophthalmol Vis Sci 31:999, Petrig BL, Riva CE, Sinclair SH, and Grunwald JE: Quantification of changes in leukocyte velocity in retinal macular capillaries during oxygen breathing. ARVO Abstracts. Invest Ophthalmol Vis Sci 22(Suppl):194, Hickam JB and Frayser R: Studies of the retinal circulation in man: Observations on vessel diameter, arteriovenous oxygen difference, and mean circulation time. Circulation 33:302, Eperon G, Johnson M, and David NJ: The effect of arterial PO 2 on relative retinal blood flow in monkeys. Invest Ophthalmol Vis Sci 14:342, Patz A, Hoeck LE, and de la Cruz E: Studies of the effect of high oxygen administration in retrolental fibroplasia: I. Nursery observations. Am J Ophthalmol 35:1248, 1952.