Inhaled Carbon Monoxide Increases Retinal and Choroidal Blood Flow in Healthy Humans

Abstract

purpose. It has been hypothesized that carbon monoxide (CO) acts as an important vascular paracrine factor and plays a role in blood flow regulation in several tissues. The present study investigated the effect of inhaled CO on retinal and choroidal blood flow.

methods. Fifteen healthy male volunteers were studied in a randomized, double-masked, placebo-controlled design with washout periods of at least 1 week between study days. CO in a dose of 500 ppm or placebo (synthetic air without CO) was inhaled for 60 minutes. Ocular hemodynamics were measured at baseline and at 30 and 60 minutes after start of inhalation. Retinal vessel diameters were measured with a retinal vessel analyzer. RBC velocity was assessed using bidirectional laser Doppler velocimetry. Retinal blood flow was calculated based on retinal vessel diameters and RBC velocity. Fundus pulsation amplitude (FPA) was measured using laser interferometry, and submacular choroidal blood flow using laser Doppler flowmetry.

results. Breathing of CO significantly increased carboxyhemoglobine, from 1.2 ± 0.5% to 8.5 ± 0.9% and 9.4 ± 0.6% at the two time points, respectively (P < 0.01). The diameter of retinal arteries increased by +3.5 ± 3.8% and +4.2 ± 3.9% (P < 0.01) in response to CO inhalation. In retinal veins, CO also induced an increase in diameter of +4.3 ± 3.0% and +4.8 ± 5.0%, respectively (P < 0.01). By contrast, placebo did not influence retinal vessel diameter. RBC velocity tended to increase during CO inhalation (+8 ± 22%), but this effect did not reach the level of significance (P = 0.1). Calculated retinal blood flow increased significantly by +12 ± 5% (P < 0.02). FPA increased after breathing CO by +20 ± 20% and +26 ± 21% at the two time points, respectively (P < 0.01). Subfoveal choroidal blood flow increased by +14 ± 9% and +15 ± 9% during breathing of CO (P < 0.01).

conclusions. This experiment demonstrated that retinal and choroidal blood flow increase during inhalation of CO. Whether this increase is caused by tissue hypoxia or a yet unknown mechanism has to be clarified.

Carbon monoxide (CO) is one of the most important air pollutants in the industrialized nations and also the best-understood toxic component of cigarette smoke. Long-term exposure to moderate doses of CO is consistently associated with congestive heart failure ¹ and artheriosclerosis, whereas exposure to high concentrations may result in CO poisoning, with potentially lethal consequences. 

The discovery that CO is generated physiologically by vascular cells ² has recently changed our view of CO as only a toxic factor. There is increasing evidence that at low concentrations CO may act as an important vascular paracrine factor. ^{3 4 5} It has been shown that basal levels of carboxyhemoglobin (COHb) reach 1% to 3% depending on the contribution of environmental background CO. ⁶ Based on these findings, it has been hypothesized that CO acts as an endogenously produced vasoactive factor analogue to the l-arginine/nitric oxide system. ^{3 4 5} This hypothesis is compatible with a growing body of evidence indicating that CO acts as a strong vasodilator in several vascular beds. ^{7 8 9} However, the majority of these experiments are limited to the investigation of the effect of CO on isolated vessels or in animal models, and scant data are yet available about the effect of CO on blood flow in humans. ¹⁰  

Thus, the present study was performed to determine the effect of inhaled CO in different doses on the ocular hemodynamics in healthy volunteers. Because of the potential toxic effect of CO, two different studies were performed. As described in detail below, a dose-finding study was conducted in an open, noncontrolled design. Based on these findings, a double-masked two-way crossover study was performed to investigate the effect of inhaled CO on retinal and choroidal blood flow. 

Materials and Methods

Subjects

Six healthy male nonsmoking volunteers (mean age, 26 ± 4 years) were included for the dose-finding study, and 15 volunteers (mean age, 27 ± 4 years) participated in the main study. The nature of the study was explained and all subjects signed a written informed consent before participation. The study protocol was approved by the Ethics Committee of the Medical University of Vienna and followed the guidelines of Good Clinical Practice and the Declaration of Helsinki. Each subject passed a screening examination during the 4 weeks preceding the study, including medical history and physical examination; 12-lead electrocardiogram; complete blood count; activated partial thromboplastin time; thrombin time; fibrinogen; clinical chemistry; hepatitis B, C, and HIV serology; urine analysis; and a urine drug screening. 

Subjects were excluded if any abnormality was found as part of the pretreatment screening. Moreover, an ophthalmic examination, including slit-lamp biomicroscopy and indirect funduscopy, was performed. Inclusion criteria were normal ophthalmic findings, ametropia of <3 diopters (D) and anisometropia of <1 D. 

Carbon Monoxide

A mixture of CO and synthetic air (containing 80% N₂ and 20% O₂) was used for our experiments. For the dose-finding study, five mixtures of synthetic air and CO in five different concentrations (10, 50, 100, 250, and 500 ppm) were prepared. For the main study, a mixture of CO at a concentration of 500 ppm and synthetic air was used. All gases used in the study were prepared by a certified producer of medical gases (Linde Gas GmbH, Vienna, Austria). Synthetic air similar in composition, but not containing CO, was administered as placebo. All gases used in our experiments were administered by means of a breathing mask. 

Experimental Paradigm

For security reasons and to determine the optimal dose for the main study, two different protocols were performed. 

Study A.

Study A was performed as a dose escalation study. Accordingly, five different CO concentrations with increasing doses of 10, 50, 100, 250, and 500 ppm were administered in an open study in six volunteers on five different study days. Each concentration was administered for 60 minutes. A minimum washout period of 1 day was scheduled between each study day. During inhalation, blood COHb was measured at 10-minute intervals. Systemic hemodynamic measurements were performed throughout the whole study period, and retinal vessel diameter and fundus pulsation amplitude were measured at baseline conditions and in the last 15 minutes of the breathing periods. Based on the COHb concentrations measured in this study, the dose for study B was selected. 

Study B.

In the main study, 15 subjects were studied in a randomized, double-masked, placebo-controlled design with washout periods of at least 1 week between the two study days. On the trial days, 500 ppm CO or placebo was inhaled for 1 hour. 

All subjects were studied with the pupil dilated after instillation of tropicamide (Mydriatikum Agepha, Vienna, Austria). Baseline systemic and ocular hemodynamic parameters were recorded in a sitting position after an initial 20-minute resting period to ensure stable hemodynamic conditions. Afterward, CO or placebo inhalation was started over a minimum time period of 60 minutes. CO inhalation was stopped when the last blood flow measurements were completed. During inhalation, capillary blood samples were drawn from the earlobe every 15 minutes to evaluate blood COHb levels and blood-gas values. At 30 and 60 minutes after initiation of each inhalation period, measurement of systemic and ocular hemodynamic parameters was started again. Noninvasive hemodynamic measurements of FPA, retinal vessel diameters, choroidal blood flow, and red blood cell (RBC) velocity were started at each of the time points. Hemodynamic measurements were finished in <15 minutes. IOP was measured at the beginning and the end of the experiment. CO breathing was continued until all measurements were completed. 

Pulse rate (PR) and real-time electrocardiogram were monitored continuously, and blood pressure was measured at 5-minute intervals throughout the whole study period. The subjects went through the same study procedure at each of the study days, with a washout period of at least 1 week between study days. 

Retinal Vessel Analyzer (RVA)

The RVA (Imedos, Jena, Germany) is a commercially available system which comprises a fundus camera, a video camera, a high-resolution video recorder, a real-time monitor, and a personal computer with vessel diameter–analyzing software. The RVA allows the precise determination of retinal vessel diameter with a time resolution of 25 readings/second. The fundus was illuminated with light in the range of wavelengths between 567 and 587 nm. In this spectral range, the contrast between retinal vessels and the surrounding tissue is optimal. Retinal irradiance was approximately 220 μW/cm², which is approximately 50 times lower than the maximum level allowed for constant illumination of the retina at the wavelengths mentioned above. The system provides excellent reproducibility and sensitivity. ¹¹ In the present study, major temporal arteries and veins were studied. Measurements of retinal vessel diameters were taken within 1 and 2 disc diameters from the margin of the optic disc. 

Laser Doppler Velocimetry (LDV)

The principle of RBC velocity measurement by LDV is based on the optical Doppler effect. Laser light, scattered by moving erythrocytes, is shifted in frequency. This frequency shift is proportional to the blood flow velocity in the retinal vessel. The maximum Doppler shift corresponds to the centerline erythrocyte velocity (V _max). ¹² With bidirectional LDV, the absolute velocity in the retinal vessels can be obtained. ^{12 13} In the present study, we used a fundus camera–based system with a single-mode laser diode at a center wavelength of 670 nm (Oculix 4000; Oculix Sarl, Arbaz, Switzerland). The Doppler shift power spectra were recorded simultaneously for two directions of the scattered light. The scattered light was detected in the image plane of the fundus camera. This scattering plane can be rotated and adjusted in alignment with the direction of V _max, which enables absolute velocity measurements. RBC velocity was measured at the same locations as retinal vessel diameters. 

Calculation of Retinal Blood Flow

Retinal blood flow in single retinal veins was calculated based on the measurements of V _max in retinal veins as assessed with LDV and retinal venous diameter (D) obtained with the RVA. From V _max, mean blood velocity in retinal vessels (V _mean) can be calculated as V _mean = V _max/2. Blood flow (Q) through a specific retinal vein was then calculated as Q = (V _max/2) * (π * D ²/4). 

Laser Doppler Flowmetry

Choroidal blood flow was assessed with a fundus camera–based laser Doppler flowmeter (Oculix 4000; Oculix Sarl, Arbas, Switzerland) introduced by Riva et al. ^{14 15} The principles of laser Doppler flowmetry have been described in detail elsewhere. ¹⁶ Briefly, the vascularized tissue is illuminated by coherent laser light. Scattering on moving RBCs leads to a frequency shift in the scattered light. In contrast, static scattering in tissue does not change light frequency but leads to a randomization of light directions impinging on the RBCs and consequently a broadening of the spectrum of scattered light. From this Doppler shift power spectrum, the mean RBC velocity, the blood volume, and the blood flow can be calculated in relative units. The system uses a laser beam at a wavelength of 670 nm. The laser beam was directed to the fovea to assess blood flow in the submacular fovea. 

Fundus Pulsation

Pulse-synchronous pulsations of the ocular fundus were assessed by a laser interferometric method, described in detail by Schmetterer et al. ¹⁷ The FPA, representing the maximum distance change between cornea and retina during the cardiac cycle, gives an estimate of pulsatile blood flow on the selected fundus location. ^{18 19} With this technique, the ocular fundus is illuminated by a high-coherence laser beam (λ = 783 nm) along the optical axis with a laser power of approximately 80 μW. The laser light is reflected at both the retina and the outer surface of the cornea, the latter serving as a reference wave. The relative change in distance between cornea and retina during a cardiac cycle may be evaluated by analyzing the interference fringes produced by the two reemitted waves. To assess pulsatile choroidal blood flow, measurements in the present study were performed in the macula. 

IOP Measurement

IOP was measured with a Goldmann applanation tonometer (Haag Streit, Vienna, Austria). Applanation tonometry measures the force required to flatten a small area of the central cornea. 

Systemic Hemodynamics

Systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP) were measured on the upper arm using an automated oscillometric device. Pulse rate was automatically recorded from a finger pulse-oxymeter. An electrocardiogram was monitored continuously using a standard four-lead device (HP-CMS patient monitor; Hewlett-Packard, Palo Alto, CA). 

Blood-Gas Analysis

Blood-gas values were determined from capillary blood samples drawn from the earlobe. After spreading a paste to induce capillary vasodilatation on the earlobe (Finalgon; Thomae, Biberach, Germany), a lancet incision was made. The arterialized blood was drawn into a thin glass capillary tube. COHb and O₂ saturation were determined by spectrophotometry, using an automatic blood-gas analysis system (AVL 912; CO-Oxilite, Graz, Austria). 

Statistical Analysis

Data from the pilot study were analyzed using a one-way ANOVA model. In the main study, effects of CO on hemodynamic parameters were assessed by ANOVA for repeated measurements. The planned comparison test was used for post-hoc testing. The Shapiro-Wilk W test was used to prove normal distribution of the data. Values of P < 0.05 were considered significant. Calculations were performed using a statistical software package (Statistica; StatSoft, Tulsa, OK). 

Results

Inhalation of CO was well tolerated by all subjects in both studies. None of the participating volunteers experienced headache or discomfort. 

Study A

As shown in Figure 1 , breathing of 10 and 50 ppm CO did not induce a measurable increase in blood COHb. Breathing of 100, 250, and 500 ppm significantly increased blood COHb to as much as 8.7 ± 1.7% (P < 0.01). Breathing of 10, 50, and 100 ppm did not induce a significant change in retinal vessel diameter. At a concentration of 250 ppm, retinal arterial diameter tended to increase by +1.4 ± 2.7% (P = 0.4); retinal vein diameter significantly increased by +2.8 ± 1.8% (P < 0.05). At a concentration of 500 ppm, retinal arterial diameter increased by +3.2 ± 3.6% (P < 0.05) and venous diameter by +3.1 ± 3.5% (P < 0.05). Breathing of 10, 50, and 100 ppm did not affect FPA. Breathing of 250 ppm induced an increase of FPA of 5.9 ± 10% (P < 0.05). At 500 ppm, FPA increased by 18 ± 5% (P < 0.05). No significant changes in systemic hemodynamic parameters were observed in response to the five concentrations of CO (data not shown). 

Study B

Systemic Hemodynamic Parameters and IOP.

PR, MAP, SBP, DBP, and IOP were comparable on all study days at baseline (Table 1) . None of the study drugs influenced the hemodynamic parameters or IOP. There was no change in O₂ saturation in arterialized blood from the earlobe on either of the two study days (Table 2) . 

Retinal Blood Flow.

As shown in Figure 2 , breathing of CO significantly increased COHb, from 1.2 ± 0.5% at baseline to 5.6 ± 0.6% after 30 minutes and to 9.4 ± 0.6% after 60 minutes of inhalation, compared to placebo (P < 0.01). Baseline values of all parameters are shown in Table 1 . Retinal arterial diameter increased by +3.5 ± 3.8% and +4.2 ± 3.9% at the two time points, respectively (P < 0.01). Likewise, in retinal veins, CO induced an increase in diameter of +4.3 ± 3.0% and +4.8 ± 5.0% at the two time points, respectively (P < 0.01). By contrast, placebo did not influence retinal vessel diameter. RBC velocity tended to increase during CO inhalation (+8 ± 22% and +4 ± 20%), but this effect did not reach the level of significance (P = 0.1). Calculated retinal blood flow, however, increased significantly, by +18 ± 30% and +12 ± 25% at the two time points, respectively (P < 0.02). 

Choroidal Blood Flow.

Breathing of CO increased FPA by +20 ± 20% and +26 ± 21% after 30 and 60 minutes, respectively (placebo versus CO, P < 0.01). Subfoveal choroidal blood flow increased by +14 ± 9% and +15 ± 9% at the two time points, respectively (P < 0.01). This effect was mainly attributable to an increase in the laser Doppler flowmetry volume parameter (velocity, −8 ± 30% and −8 ± 22%; volume, +22 ± 45% and +20±44% after 30 and 60 minutes, respectively; P < 0.6). Placebo had no effect on choroidal blood flow parameters. 

Discussion

Previous experiments evaluating the role of inhaled CO have concentrated mainly on the toxic potential of the gas and the associated systemic hemodynamic effects. Inhalation of high toxic doses of CO has been shown to induce tachycardia, ²⁰ whereas in moderate exposures with a COHb up to 20%, no change in heart rate ^{10 21 22} or blood pressure ²³ is seen. Furthermore, prolonged exposure to CO for eight days with COHb levels up to 12% does not cause changes in blood pressure. ²⁴  

However, more recent experiments indicate that even lower doses of CO may influence the cardiovascular system. Data from animal studies show that CO plays a role in vasomotor control. ^{25 26 27} Furthermore, in vitro as well as animal studies indicate that CO is a potent vasodilator in several tissues. ^{7 8 9 26 27} On the other hand, it has been shown that acute exposure to moderate COHb levels does not affect forearm blood flow, blood pressure, or heart rate, indicating that the vasodilator effects of CO are only seen in some vascular beds. ^{10 28}  

Since CO is a major component of cigarette smoke, it is of interest to compare our present data with previously published results on the effect of acute smoking on ocular hemodynamics. Cigarette smoking does not alter optic nerve head blood velocity in light smokers, but increases this parameter in habitual smokers, as evidenced from laser speckle measurements. ^{29 30} This is in keeping with an earlier blue-field entoptic study showing an increase in perimacular white blood cell velocity in habitual smokers. ³¹ The data from the present study indicate that the increase in COHb may well contribute to smoking-induced vasodilatation. Long-term smoking, however, appears to be associated with additional effects, including an abnormal capacity for increased blood flow velocity in the central retinal artery and abnormal blood flow regulation in response to hyperoxia, hypercapnia, and isometric exercise. ^{32 33 34}  

Our data indicate that COHb levels of 8% induce an increase of retinal blood flow. Exposure to CO for 60 minutes produced an increase in COHb level similar to that achieved by successive smoking of approximately five to nine cigarettes. ^{35 36} The observed retinal blood flow increase can be mainly attributed to an increase in retinal vessel diameter and not to an increase in RBC velocity. Since the microvasculature is the major source of resistance to flow, this, in turn, indicates that CO affects mainly the major retinal branch vessels and may have less effect on the retinal microvessels. However, to finally confirm this hypothesis, direct measurement of the diameter of the resistance vessels would be required, which is currently impossible because of the limited resolution of the instruments available. 

The mechanism for the CO-induced increase in tissue blood flow is still a matter of controversy. The classical theory introduced hypoxia as an explanation. Since the affinity of hemoglobin for CO is 200-fold greater than that for oxygen, the presence of COHb reduces the number of oxygen carriers. COHb levels as low as 5% have been shown to cause a considerable loss in the oxygen-carrying capacity of blood. ³⁷ High COHb levels may lead to impaired visual discrimination and motor coordination, ^{38 39} especially affecting organs with high oxygen demand, such as the eye or the brain. Given that the oxygenation of the photoreceptors is barely adequate under normal conditions ⁴⁰ one could hypothesize that the vasodilating effect of CO can be mainly attributed to local tissue hypoxia, which in turn leads to retinal vasodilatation. 

Whether CO induces hypoxia was recently investigated in rats, where intramitochondrial NADH redox levels were used to determine the redox state of the tissue. ⁴¹ Exposure to 1000 ppm CO in air resulted in an increased cerebral blood flow without any concomitant changes in any of the other metabolic or ionic parameters measured. ⁴¹ The unchanged NADH levels were interpreted by the authors as evidence that tissue hypoxia does not necessarily occur during CO exposure. However, since the authors do not report blood COHb levels, the comparison of the latter data to other experiments investigating the effect of CO is difficult. In the present study, CO breathing did not alter oxygen saturation in arterialized blood drawn from the earlobe, which is in keeping with previous results. ⁴² This does not, however, necessarily exclude hypoxia at the level of the eye, because increased COHb may well alter retinal oxygen extraction even though arterial blood oxygen content is not affected. In the choroid, however, hypoxia most likely does not account for the increase in blood flow, because choroidal blood flow is insensitive to a reduction in pO₂. ⁴³  

Alternatively, evidence gained from animal experiments indicates that tissue hypoxia is not the only trigger of CO-induced vasodilatation. ⁴¹ In addition to uptake of exogenous gas, cells and tissues produce significant amounts of CO as a byproduct of heme degradation. In particular, CO is generated under physiological conditions predominately by vascular cells as a byproduct of heme catabolism, in which heme oxygenases (HOs) catalyze the degradation of heme to CO, free iron, and biliverdin. ² The activity of those HOs can be increased by several physiological stimuli as different as hypoxia, ⁴⁴ hypertension, ⁴⁵ and shear stress. ⁴⁶ The vasodilatory property of CO is independent of the presence of the intact endothelium ⁴⁷ and thought to act via the activation of soluble guanylate cyclase, enhancing the synthesis of cGMP. ⁴⁷ Furthermore, it was observed that this endogenous carbon monoxide production from the catabolism of hemoglobin may be a component of normal blood flow regulation. ^{45 48}  

In conclusion, the data from the present study provide evidence that COHb levels comparable to those measured in moderate smokers induce an increase in retinal and choroidal blood flow. Whether this increase is caused by tissue hypoxia or by yet unknown mechanisms remains to be clarified. 

 Submitted for publication April 1, 2005; revised June 8 and July 15, 2005; accepted September 13, 2005.

 Disclosure: H. Resch, None; C. Zawinka, None; G. Weigert, None; L. Schmetterer, None; G. Garhöfer, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Dose-finding study: blood COHb values after inhalation of CO at the different concentrations. Values are means ± SD (n = 6); *P < 0.01.

Table 1.

 

Hemodynamic Parameters and IOP as Recorded at Baseline Conditions

Parameter	Placebo Day	CO Day
MAP (mm Hg)	79 ± 7	80 ± 9
SBP (mm Hg)	119 ± 9	114 ± 9
DBP (mm Hg)	61 ± 8	61 ± 8
IOP (mm Hg)	13 ± 2	13 ± 2
PR (bpm)	63 ± 6	63 ± 7
O2 saturation (%)	96 ± 2	97 ± 1
COHb (%)	1.4 ± 0.6	1.3 ± 0.5
Arterial diameter (μm)	120 ± 18	121 ± 16
Venous diameter (μm)	157 ± 19	155 ± 19
FPA (μm)	3.7 ± 1.3	3.6 ± 1.4
Subfoveal choroidal blood flow (AU)	9.2 ± 2.5	9.0 ± 2.2
Mean velocity (cm/s)	1.2 ± 0.2	1.1 ± 0.3
Retinal blood flow (μL/min)	13.8 ± 3.0	12.1 ± 2.7

 Mean velocity and retinal blood denote blood flow through one specific vein, not total retinal blood flow. Values are means ± SD (n = 15).

Table 2.

 

Hemodynamic Parameters during Placebo and CO Administration

Parameter	Placebo					CO
		Baseline	30 min	60 min	Baseline	30 min	60 min
SBP (mm Hg)	114 ± 8	112 ± 6	111 ± 6	119 ± 9	114 ± 7	116 ± 7
DBP (mm Hg)	61 ± 8	60 ± 7	59 ± 8	61 ± 8	60 ± 7	58 ± 9
MAP (mm Hg)	79 ± 7	78 ± 6	77 ± 7	80 ± 9	78 ± 9	76 ± 11
PPA (mm Hg)	53 ± 11	52 ± 8	52 ± 7	59 ± 11	54 ± 9	59 ± 7
PR (bpm)	64 ± 7	61 ± 6	61 ± 6	63 ± 6	62 ± 6	63 ± 6
O2 saturation (%)	96.3 ± 2.1	95.8 ± 1.4	96.2 ± 2.2	96.7 ± 1.3	96.6 ± 1.3	95.8 ± 2.2

Figure 2.

 

Blood COHb concentration (top left), percent change of retinal arterial (top right) and venous vessel diameter (middle left), retinal blood flow (middle right), FPA (bottom left), and subfoveolar choroidal blood flow (bottom right) as recorded at the three different time points. Measurements were performed at baseline conditions and at 30 and 60 minutes after beginning inhalation of CO (▴) or placebo (•), respectively. Values are means ± SD (n = 15); *P < 0.05.