Nociceptive Sensation and Sensitivity Evoked from Human Cornea and Conjunctiva Stimulated by CO₂

Abstract

purpose. To compare sensation and sensitivity evoked from human cornea and conjunctiva stimulated by CO_2.

methods. Twenty healthy participants were recruited for the study. Central corneal and temporal conjunctival chemical sensation and sensitivity of only one eye of each subject were evaluated. Air mixed with different concentrations of CO₂ was delivered by a modified Belmonte pneumatic esthesiometer. The ascending method of limits was used to determine the sensitivity and subjects were required to characterize the sensation at threshold.

results. The sensations evoked by CO₂ in the cornea and conjunctiva were stinging or burning. The sensation evoked by mechanical stimulation was that of irritation. The corneal and conjunctival chemical thresholds were 31% ± 2% and 54% ± 5% CO₂ (mean ± SE), respectively. The corneal and conjunctival mechanical thresholds were 80 ± 6 and 140 ± 10 mL/min (mean ± SE), respectively. The corneal sensitivity was significantly higher for both mechanical and chemical stimuli (P < 0.05).

conclusions. The results suggest that CO₂ stimulates similar corneal and conjunctival nociceptors in that the interpretations were the same (i.e., nociceptive). The central cornea had a higher sensitivity to CO₂ than the temporal conjunctiva, which may reflect a different peripheral innervation, such as different nerve density or different receptor characteristics. Sensations evoked by mechanical and chemical stimulation were different, which suggests that at the peripheral level, the two modalities stimulate two different kinds of molecular receptors or channels and that this information is somehow retained within the nociceptive system.

How do humans interpret external noxious stimulation on the ocular surface and what is the internal mechanism of the interpretation for a specific stimulus? This is a fundamental question that relates to psychophysics and neurophysiology. Compared with psychophysical somatosensory study of the skin, progress in understanding of ocular surface sensation has been relatively slow. 

One of the difficulties in this understanding of the ocular surface is in delivering a specific stimulus. The study of corneal sensation can be traced back to Von Frey in 1894, who measured corneal sensation with horse hairs. ¹ Based on Boberg-Ans’s (1955) device in which a single nylon filament was used to produce various forces by varying its length when applied to the cornea, Cochet and Bonnet (1960) developed the commercial form of this instrument. The Cochet-Bonnet esthesiometer is used widely in ophthalmic research. 

The basic idea of the esthesiometer is to use different forces to stimulate the cornea and thus test corneal mechanical sensitivity quantitatively. Because of the limitation of contact esthesiometers, which cause apprehension in the subject particularly when it is applied to the central cornea, noncontact pneumatic esthesiometers have been developed. ^{2 3 4} These instruments use air to deliver force to evaluate corneal and conjunctival mechanical sensation and sensitivity. 

Humans can sense not only mechanical stimulation, but chemical and thermal as well. A method of delivering chemical and thermal as well as mechanical stimuli to the ocular surface was described by Belmonte et al. ^{2 3 4 5} and Chen et al. ⁶ With this CO₂ pneumatic esthesiometer, which uses mixed air and CO₂ at different temperatures, chemical, thermal, and mechanical stimuli can be delivered to the ocular surface. 

Our understanding of the comparison of corneal and conjunctival chemical sensitivity is limited. There is only one study of human corneal and conjunctival sensitivity and sensation evoked by CO₂, ⁷ in which it was shown that the sensitivity of the human cornea and conjunctiva to CO₂ were nearly equal. In our study, we used more subjects and a different psychophysical method to compare the sensitivity of the human cornea and conjunctiva with mechanical (air) and chemical (CO₂) stimulation. In addition, the sensation evoked from the ocular surface with mechanical and chemical stimuli was characterized. 

Methods

Subjects

Twenty participants, 5 women and 15 men aged from 23 to 43 years, were enrolled in the study. None of them had a history of eye or systemic disease or surgery. None of them wore contact lenses. This study adhered to the tenets of the Declaration of Helsinki for research involving human subjects and received the clearance from the University of Waterloo Office of Research Ethics (Waterloo, Ontario, Canada). 

The Belmonte Esthesiometer

The Belmonte esthesiometer has been described in detail elsewhere. ² Briefly, an electronic control box regulates the mixture of CO₂ and air. There are two flow controllers to control the air and CO₂ flow rate. Two digital controllers set the total flow rate (air and CO₂) and the concentration of CO₂. A thermostat is built inside the nozzle assembly, allowing temperature control as well. The nozzle of the esthesiometer is housed inside a modified noncontact tonometer that allows accurate focus (essential, to keep the tip of the nozzle 5 mm from the ocular surface) and an X, Y, and rotational position control to ensure that the stimulus is delivered orthogonal to the ocular surface. 

Procedure

Measurements were made at the apical cornea and on the temporal conjunctiva, 5 mm away from the limbus, along the horizontal meridian. For apical measurements, subjects viewed the center of an annular light surrounding the esthesiometer’s nozzle. For conjunctival measurements, subjects viewed an eccentric target and the 5-mm distance between the limbus and temporal conjunctiva was estimated. A video camera, mounted on the eyepiece, was focused on the conjunctiva at the 5-mm distance and allowed us to verify this distance constantly during the experiments. All measurement on the left eye was performed in the afternoon. 

The temperature of the air was set at 50°C and decreased to 33.4°C at the ocular surface at room temperature of 22°C. This was calibrated using a custom electronic thermometer positioned 5 mm from the probe tip (which corresponds to the position of the ocular surface in the experiments). The stimulus duration was 3 seconds. The subject blinked freely between trials. Only after the sensation caused by the last stimulus had disappeared completely was the next stimulus triggered. The stimulus was delivered to the ocular surface immediately after a blink. 

Using the ascending methods of limits, the lowest air flow rate (with CO₂ set at 0%) that the subject could detect was determined first. The flow-rate steps were set at 10 mL/min, and the threshold was the average of three readings when the subject first reported the stimulus. The flow rate was then set at half this threshold, and CO₂ was added to the air. The increment size of CO₂ was set at 10%. The chemical threshold was obtained in the same way as the air-flow threshold, using ascending methods of limits. At threshold, subjects were asked to report the characteristics of the sensation. These psychophysical procedures were specifically chosen to minimize any suprathreshold adaptation effects and allow us to capture the attributes of the sensations evoked by the stimuli at threshold. 

Results

Mechanical stimulation caused irritation, whereas CO₂ caused stinging or burning. Corneal and conjunctival mechanical thresholds were 80 ± 6 and 140 ± 10 mL/min (mean ± SE), respectively. The corneal sensitivity was significantly higher (paired t-test, P < 0.001). The responses evoked by CO₂ in the cornea and conjunctiva were similar: stinging or burning pain. Corneal and conjunctival chemical thresholds were 31% ± 2% and 55% ± 5% CO₂ (mean ± SE), respectively. The corneal chemical sensitivity was significantly higher (paired t-test, P < 0.001). These data are shown in Figure 1 for the mechanical and chemical stimuli. 

Discussion

One of the anatomic features of the eyes that is different from skin is that there is a tear film on the ocular surface. CO₂ dissolves in the tears to produce hydrogen ions that can mimic the tissue acidosis that characterizes infection, ischemia, and inflammation and can stimulate the C fibers, possibly mediated by vanilloid receptor-1 (VR-1) and/or acid-sensing ion channels. ^{8 9 10} Studies have shown that when the human cornea was stimulated by CO₂, the interpretation was stinging irritation. ^{2 6 11} In this study, the same nociceptive sensation was evoked by CO₂ in both the cornea and conjunctiva. This suggests that CO₂ stimulates nociceptors in both ocular tissues and that the central processing may be the same. The sensation evoked by CO₂ was dissimilar from that caused by simple mechanical stimulation. The different sensations experienced were presumably due to the different modes of stimulation at the neural level. Specifically, though, why these two stimuli caused different sensations is speculative and rather difficult to reconcile with the principle that ocular nociception is essentially unidimensional, in that all nociceptive stimuli should be sensed along the same sensory continuum. 

At a molecular receptor/channel level, VR-1, acid-sensitive ion channels (ASICs), and epithelial Na⁺ channels (ENaC) have been demonstrated to be expressed by the trigeminal ganglion. ^{8 9 10 12 13 14} Both VR-1 and ASICs mediate sensitivity to acid stimulation, and ENaC mediates mechanical sensitivity. Although there are differences in the expression of these receptors in myelinated and unmyelinated fibers, there is no clear way of accounting for the mechanical and chemically induced sensations on the ocular surface, based on the analysis of receptor type and their expression on Aδ and C terminals. 

The human cornea is innervated by Aδ and C fibers. ^{15 16} Perhaps these two attributes (mechanical irritation and chemical stinging) are related to differential activation of Aδ and C neurons (shown to be associated with a sharp pricking pain and dull or burning pain, respectively; see, for example, Ref. ¹⁷ ). Unfortunately, both Aδ and C fibers have been shown to be sensitive to all nociceptive stimuli (polymodal) ^{5 6 11 18} and so the straightforward association of sensation with neuron diameter appears problematic. ^{19 20} As pointed out by a reviewer, perhaps the differences can be explained by assuming that mechanical stimuli activate all populations of fibers (Aδ mechanosensory, Aδ polymodal, and C polymodal fibers), whereas CO₂ activates only polymodal fibers. Alternatively, it is possible that the short-lasting mechanical stimulus activates predominantly the faster Aδ fibers, whereas the CO₂ recruits the slower C polymodal fibers. 

The feeling evoked by the pneumatic stimulus was transient, occurred during the stimulus presentation, and then disappeared. The feeling evoked by CO₂, however, was delayed and persisted after the presentation of the stimulus. This delay decreased as the concentration of CO₂ increased. The exact mechanism is not clear, but we postulate two possibilities: The first one is that mechanical stimuli may excite more Aδ fibers, whereas the chemical stimuli may excite the more slowly conducting C fibers. Second, the delay may be caused by the time taken for CO₂ to dissolve in the tears and reside in dissolved form after the stimulus is delivered. The protracted sensation is possibly a combination of these two. 

A critical point related to the preceding discussion is that mechanical and chemical stimuli evoke different nociceptive sensations: The chemical stimulus stings, and the mechanical stimulus feels scratchy. This implies that these two irritative attributes are retained within the nociceptive system. 

The stimulus duration may seem protracted, with the possibility of physically drying the cornea. This duration of application of the stimulus was chosen primarily because of the long latencies of the response to the CO₂, which must dissolve in the tears and then stimulate chemical receptors within the cornea and conjunctiva. This latency has been shown to be quite long (1.9 ± 0.2 seconds) in the reaction of the corneal sensory nerves to CO₂ in the feline cornea. ⁶ Although this may not be exactly the same in humans, we had to consider it when we designed the study. 

In this study, the subjects exhibited greater mechanical sensitivity in the cornea than in the conjunctiva, similar to that observed in other psychophysical studies. ^{1 3} It is also in accord with the neurohistologic and electrophysiological data, which show that the cornea has the highest nerve density and the polymodal and mechanical nociceptors in the cornea have lower mechanical thresholds than the conjunctiva. ^{5 21}  

The subjects were also more sensitive to CO₂ when it was delivered to the cornea than to the conjunctiva. There are no electrophysiological data that show how conjunctival nerves respond to CO₂. Our result may reflect greater spatial summation and recruitment due to the high innervation density of the cornea, the sensitivity of the nerves themselves, for example, due to greater receptor density or some other unidentified physiological differences. It is also possible that tear film differences between cornea and conjunctiva plays a role (as discussed later). 

In a recent study in which a similar esthesiometer was used, ⁷ the characteristics of the feelings evoked by CO₂ were same as we found (i.e., nociceptive). In that study, however, there were no differences between the corneal and conjunctival mechanical and chemical sensitivities. This may be because of the different methods for determining the thresholds and perhaps also reflects sample size (and therefore power) differences. 

There are two physiological and/or structural features of the tear film that should not be neglected. The first is the thickness of the tear film. Theoretically, the greater the aqueous component, the more CO₂ dissolves to produce more hydrogen ions. At present, we do not know whether there is a difference between the thickness of the tear film on the corneal and conjunctival surfaces. The second feature is the buffering capacity of the tear film. Normal human tears are slightly alkaline compared with serum. Tears have substantial buffering capacity, especially in the acid range, mainly because of a bicarbonate buffering system and partly from protein buffer and other components. The dissolved CO₂ changes the pH, and this change is buffered by the tears. Theoretically, the buffering capacity on the corneal and the conjunctival surfaces should be similar. But because the conjunctiva is vascular, it is possible that it has a relatively higher buffering capacity, and this leads to what appears to be less sensitive to CO₂. This should be explored further. 

In comparison with the Cochet-Bonnet esthesiometer, the Belmonte esthesiometer stimulates specific nerves on the ocular surface, because the mixed gas can deliver force, temperature, and CO₂ stimuli simultaneously. The psychophysical difficulty, however, is to attempt to isolate each mechanism, without cross talk or masking from the other, which would affect the thresholds. To do this for the mechanical stimulus, we simply did not add CO₂ and attempted to minimize the thermal effects by presenting stimuli at ocular surface temperature. For the chemical stimulus, we excluded thermal and mechanical stimuli. To ensure that the sensation was evoked primarily by the chemical component of the air-CO₂ mixture, the air temperature was similar to that of the ocular surface, and the flow rate was at half the mechanical threshold. When this stimulus was sensed by subjects, stinging developed slowly and then declined and, perhaps most important, there was no other mechanical or thermal sensation. If this control of stimulus intensity does not occur, it is unclear when thresholds are measured that a specific mechanism is isolated. For example, if CO₂ is added and the flow rate is the same as the mechanical threshold, how can one be sure that the detection is not due to the flow rather than the CO₂? 

In summary, the Belmonte esthesiometer enabled us to investigate ocular surface sensation. CO₂ dissolved in tears appeared to stimulate the corneal and conjunctival polymodal nociceptors or chemonociceptors. The sensation evoked by CO₂ on both cornea and conjunctiva was nociceptive, with higher sensitivity in the central cornea than in the temporal conjunctiva. If the central processes are the same, the different sensitivity may reflect different peripheral innervation of the cornea and conjunctiva and the effects of the tear film. The different sensations evoked by mechanical and chemical stimulation may be attributed to the two modalities stimulating different peripheral molecular receptors and/or nerves. 

 Supported by National Sciences and Engineering Research Council of Canada.

 Submitted for publication January 3, 2002; revised June 10 and October 4, 2002; accepted October 7, 2002.

 Commercial relationships policy: N.

 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.

 

Scatterplots showing relationships between corneal and conjunctival mechanical (A) and chemical (B) thresholds.

 

The authors thank the subjects who participated in the study and the anonymous reviewers for constructive help.