- Email: [email protected]
A Study of Aqueous Humor Dynamics in Keratoconus
Exp. Eye Res. (1996) 62, 95–99
A Study of Aqueous Humor Dynamics in Keratoconus W A R R E N T. G O O D M ANa ; W I L L I A M D. M A T H E R Sb*, P A U L M. M U N D E Nb, K. C. O S S O I N IGb THOMAS E. D A L E Yb a
School of Medicine and b Department of Ophthalmology, University of Iowa, Iowa City, IA, U.S.A. (Received Columbia 3 May 1995 and accepted in revised form 29 August 1995) Clinical observations suggest that patients with keratoconus have lower intraocular pressures, on average, than normal subjects. Our purpose was to determine whether differences in aqueous production and outflow facility could account for differences in intraocular pressure between a group of patients with keratoconus and a group of normal, age-matched control subjects. Aqueous humor dynamics were determined by the use of fluorophotometry in one eye of seven patients with keratoconus and ten agematched normal subjects. Intraocular pressure was measured by applanation tonometry. Keratoconus patients had a statistically significant lower mean intraocular pressure than normal control subjects (11±3³1±6 mmHg vs. 16±6³2±8 mmHg, P ¯ 0±0004). The difference in mean intraocular pressure remained significant even after correcting for possible errors in applanation tonometry due to thin corneal stroma. There was no difference in mean aqueous humor flow rates in the keratoconus patients as compared to controls (2±29³0±53 µl min−" vs. 2±21³0±48 µl min−", P ¯ 0±73). The mean apparent outflow facility was 0±21³0±07 µl min−" mmHg−" for keratoconus patients compared to 0±14³0±03 µl min−" mmHg−" for controls (P ¯ 0±02). Lower mean intraocular pressure in keratoconus patients appears to be due to increased outflow facility as compared to normal subjects. # 1996 Academic Press Limited Key words : keratoconus ; aqueous humor ; outflow facility ; trabecular meshwork ; uveoscleral pathway ; fluorophotometry.
1. Introduction Keratoconus is a non-inflammatory ocular disorder characterized by progressive corneal thinning, protrusion, and scarring resulting in distorted and diminished vision. The pathology of keratoconus has been documented and summarized (Krachmer, Feder and Belin, 1984), with focus placed on the role of keratocyte or collagen disorders in the pathogenesis of keratoconus (Bron et al., 1978 ; Caffi, 1966 ; Cannon and Foster, 1978 ; Newsome et al., 1981 ; Pouliquen et al., 1968 ; Pouliquen et al., 1970 ; Pouliquen et al., 1972 ; Robert et al., 1970). Clinical observations suggest that patients with keratoconus appear to have lower intraocular pressures (IOP) than average and few patients with keratoconus develop glaucoma. Goldmann (1951) established that the major determinants of IOP were flow of aqueous humor, resistance to outflow, and episcleral venous pressure. An additional determining factor, uveoscleral flow, was established by the work of Bill (1965). Steady-state IOP of the normal eye is maintained within physiologic limits by these four factors. This paper is the first to describe aqueous dynamics in keratoconus patients. The variables studied were IOP, aqueous humor flow rate (FR), and calculation of mean apparent outflow facility from the anterior chamber (C). We hypothesize that keratoconus * For correspondence at : University of Iowa Hospitals and Clinics, Department of Ophthalmology, 200 Hawkins Drive, Iowa City, IA 52242-1091, U.S.A.
0014-4835}96}01009505 $12.00}0
patients have lower IOPs as the result of higher outflow facility. The objective of our study was to determine whether differences in aqueous production, determined by fluorophotometry (Goldmann, 1950 ; Brubaker, 1982 ; Jones and Maurice, 1966) and outflow facility could account for differences in intraocular pressure between a group of patients with keratoconus and a group of normal, age-matched control subjects. 2. Materials and Methods This research followed the tenets of the Declaration of Helsinki, informed consent was obtained after the nature and possible consequences of the study was explained, and this project proceeded with the approval of the University of Iowa Human Subjects Committee. Our project included seven eyes corresponding to seven persons with clinically evident bilateral anterior keratoconus. The control group included ten age-matched eyes of ten volunteers with no previous ocular pathology and who were not receiving drug treatment. Three of the keratoconus subjects had received one corneal transplant (the study was carried out in their fellow eye), but none of the seven subjects were undergoing any drug treatment. The keratoconus patients were volunteers selected from the Cornea Clinic population of the University of Iowa Department of Ophthalmology. Our study employed a protocol similar to those of both Yablonski et al. (1978) and Brubaker (1986). In normals, beginning at 0800 hr, one drop of 0±25 % # 1996 Academic Press Limited
96
fluorescein with 0±4 % benoxinate was instilled in one eye every minute for 4 min. This was followed by a solution of 2 % fluorescein given in one-drop doses every minute for 8 min. At approximately 1300 hr, fluorophotometric measurements of the cornea and anterior chamber using the Fluorotron Master fluorophotometer (Coherent, Palo Alto, CA, U.S.A.) were commenced. These measurements were repeated at approximately 30 min intervals for a total of three to six measurements per subject. The procedure was repeated, on a different date, in subjects whose corneal fluorescein concentration values fell below 200 ng ml−" before three measurements could be taken. In a few cases, it was necessary to delay the experimental measurements until the corneal fluorescein concentrations fell below 1000 ng ml−", as values above that level have decreased accuracy due to inner filter effects (Brubacker, 1986). Delaying measurements, to achieve ideal corneal fluorescein concentrations between 200–1000 ng ml−", does not effect results because the equilibrium between the corneal fluorescein and the fluorescein in the aqueous humor has been established, and remains intact as fluorescein is gradually cleared from the eye. The protocol was identical between the two subject groups, except that in keratoconus subjects, no anesthetic (benoxinate) was administered because of increased penetration of fluorescein into the cornea and anterior chamber of these patients. In all subjects, the cornea appeared to be evenly stained when the fluorescein application was complete. The use of benoxinate in normal subjects has no effect on fluorophotometric measurements (5 hr after application) and its use has been reported by other investigators. The FR value was computed by software, based on the Yablonski equations, in the analysing program of the Fluorotron Master fluorophotometer. Both anterior chamber volume and cornea stromal volume were necessary to measure FR. Anterior chamber volume was established from the diameter of the anterior chamber (derived from the horizontal diameter of the cornea) and the depth of the anterior chamber. Horizontal corneal diameters (accurate to tenths of a millimeter) for each subject were measured with a one chip charged coupling device video camera coupled to a VIA-100 video caliper (Boekler, Tucson, AZ, U.S.A.). Corneal stromal volume was established from measurement of corneal thickness and horizontal corneal diameter using the geometric formula for the volume of a cylinder. Anterior chamber depth and corneal thickness were determined using Immersion A-scan Biometry. Measurements were made using the MiniA-Scan (Biophysic}Alcon, Fort Worth, TX, U.S.A.) with the beam objectively aligned with the axis of the eye. This technique has an accuracy of ³20 µm for anterior chamber depth and ³15 µm for corneal thickness. At any given time, most of the fluorescein is in the
W. T. G O O D M A N E T A L.
corneal stroma, with much less in the anterior chamber. Therefore, the measurement of corneal fluorescence is critical. In most of the keratoconus patients in our study, corneal thicknesses were less than the 500 µm focal diamond of the fluorophotometer. However, the software in Fluorotron Master analysing program corrects for this to avoid underestimation of fluorescein concentration in the cornea. All subjects received an eye examination at which time their IOP values, using Goldmann applanation tonometry, were determined. The eye examination was performed after all FR measurements were completed and before the concluding ultrasound examination. Since the measurement of IOP in keratoconus may be influenced by corneal thickness and rigidity, IOP was also adjusted for variation in corneal stromal thickness (Whitacre, Stein and Hassanein, 1993). Whitacre developed equations to determine error in tonometry under manometrically controlled IOP in the 10 mmHg and also the 20 mmHg range. In our study, subjects with an IOP of 15 mmHg or less had their IOP adjusted according to the equation for the 10 mmHg controlled group, and those with an IOP greater than 15 mmHg had their IOP adjusted according to the equation for the 20 mmHg controlled group. Maximum adjustment in IOP in our study was 5 mmHg, which was the greatest adjustment made by Whitacre et al. (1993). While we are aware of the potential underestimation of TOP in patients with thin corneal stromas, we have, nevertheless, decided to use measured IOP instead of adjusted IOP in our reported calculations and results. However, the adjusted IOPs were run through all calculations to test for statistical significance. Our decision is based on the fact that the adjustment equations were specifically derived from structurally normal corneas. As the corneas in our study are not structurally normal, we do not believe that the adjusted values represent a more accurate appraisal of IOP. Apparent outflow facility (C) for each subject was calculated based on the equation C ¯ FR}IOP assuming that episcleral venous pressure (Pepi) was constant (9 mmHg), and assuming that outflow facility was primarily composed of both the reciprocal of resistance to outflow through the trabecular meshwork (Ctrab) and facility through the uveoscleral pathway (Bill, 1965 ; Bill and Phillips, 1971 ; Brubaker, 1986). We referred to this as apparent outflow facility to distinguish this measurement from the traditional tonographic outflow facility measurement. Results for FR, IOP and C were analysed using the two-sample independent groups t-test. 3. Results The demographics of the participants in the study are shown in Table I. In the normal subject group,
AQUEOUS HUMOR DYNAMICS
97
T I Age and sex characteristics for the two groups tested, normals and keratoconus patients
Age : mean range (yrs) Sex (M : F)
Normal
Keratoconus
25±7³3±6 23–33 8:2
33±6³8±2 22–46 3:4
eight were men and two were women, and the groups mean age was 26 years (range 23–33). In the keratoconus subject group, three were men and four were women, and the groups mean age was 34 years (range 22–46). Individual results for corneal thickness, FR, IOP,
and C are provided in Table II. The measured mean FR in the normal subjects was 2±21³0±48 µl min−" and in the keratoconus subjects was 2±29³0±53 µl min−" (Table III). The difference in FR values between these two groups was not statistically significant (P ¯ 0±73). The range of FR values was from 1±59 µl min−" to 2±97 µl min−" in the keratoconus group and from 1±41 µl min−" to 2±95 µl min−" in the normal group. For a P value equal to 0±05 and a sample size of 17, the β (type II error probability) is less than 0±10 yielding a power (1-β) of greater than 0±9. Keratoconus patients had a statistically significantly lower mean IOP than normal subjects (Table III). The mean IOP in keratoconus patients was 11±3³1±6 mmHg and was 16±6³2±8 mmHg in normals (P ¯ 0±0004). The range of IOPs was from 9 mmHg to 13 mmHg in the keratoconus group and
T II Individual results for corneal thickness, aqueous humor flow rate, intraocular pressure, adjusted intraocular pressure, and outflow facility in normal and keratoconus (KCN) subjects
Normal 1 2 3 4 5 6 7 8 9 10 KCN 1 2 3 4 5 6 7
Corneal thickness (µm)
Aqueous humor flow rate (µl min−")
IOP (mmHg)
Adjusted IOP (mmHg)
Outflow facility (µl min−" mmHg−")
530 580 580 580 490 580 580 550 580 550
2±20 2±95 2±81 1±58 2±01 1±41 2±31 2±27 2±39 2±21
16 17 19 16 11 16 19 21 17 14
16±6 16±7 18±7 15±7 10±6 15±7 18±7 21±2 16±7 13±7
0±14 0±17 0±15 0±10 0±18 0±09 0±12 0±11 0±14 0±15
390 490 420 290 360 550 520
2±60 1±59 1±92 2±97 1±83 2±80 2±33
9 12 13 10 10 12 13
13±3 11±6 15±9 15±0 15±0 11±7 13±4
0±29 0±13 0±15 0±30 0±18 0±23 0±18
T III Results for intraocular pressure, adjusted pressure, flow rate, and outflow facility for normals versus keratoconus patients
IOP (mmHg) mean³.. Range Adjusted IOP (mmHg) mean³.. Range Flow rate (µl min−") mean³.. Range Apparent outflow facility (µl min−" mmHg−") mean³.. Range
Normal
Keratoconus
P value
16±6³2±8 11–21 16±4³2±9 10±6–21±2 2±21³0±48 1±41–2±95 0±14³0±03 0±09–0±18
11±3³1±6 9–13 13±7³1±7 11±6–15±9 2±29³0±53 1±59–2±97 0±21³0±07 0±13–0±30
0±0004 0±04 0±73 0±02
98
from 11 mmHg to 21 mmHg in the normal group. The statistically significantly lower mean IOP in keratoconus patients remained after the IOP was adjusted for differences in corneal thickness between the two groups (Table III). The mean adjusted IOP in keratoconus patients was 13±7³1±6 mmHg (range 11±6–15±9 mmHg), and in normal subjects was 16±4³2±9 mmHg (range 10±6–21±2 mmHg) with P ¯ 0±04. Keratoconus patients had a statistically significantly higher mean apparent outflow facility (C) than normal subjects (Table III). Calculated C was 0±21³0±07 µl min−" mmHg−" (range 0±13–0±30 µl min−" mmHg−") for keratoconus patients compared to 0±14³0±03 µl min−" mmHg−" (range 0±09–0±18 µl min−" mmHg−") for controls (P ¯ 0±02). 4. Discussion This study has shown that IOP is significantly lower in keratoconus patients compared to a group of normal subjects, even when correcting for possible errors in applanation tonometry due to thin corneal stroma. A lower mean pressure in keratoconus patients has been found by other investigators. The mean IOP value of 11±3³1±6 mmHg for keratoconus patients found in our study was very close to the values found in ‘ moderate ’ (11±7³2±0 mmHg) and ‘ advanced ’ (12±0³3±5 mmHg) levels of keratoconus as demonstrated in a study by Yamamoto, Nakayama and Kinoshita (1994) using 1085 eyes of 634 patients with keratoconus. Our results show no significant difference in FR, and a significant decrease in IOP between the keratoconus and normal groups. This suggests that the rate of flow of aqueous humor is constant and not affected by small shifts in IOP. Previous studies have shown that in the normal eye, the FR is independent of IOP (Goldmann, 1950 ; Carlson et al., 1987 ; Brubaker and McLaren, 1985) and that no difference in FR has been observed between men and women (Brubaker and McLaren, 1985). Our FR values were also within the range (between 1±5 and 2±8 µl min−") found by many other investigators (Goldmann, 1950 ; Brubaker, 1982 ; Jones and Maurice, 1966 ; Martin et al., 1992 ; Coakes and Brubaker, 1979 ; Brubaker et al., 1981 ; Bloom et al., 1976 ; Araie, 1980 ; Coulangeon, Menerath and Sole, 1987). Our results imply that facility of outflow of aqueous humor is increased in keratoconus patients. If the more specific equation for aqueous humor dynamics is considered [FR ¯ Ctrab (Pio®Pepi)Fuveo] with episcleral venous pressure assumed constant (9 mmHg), aqueous humor outflow facility through the trabecular meshwork or uveoscleral flow (Fuveo) must be markedly increased in patients with IOPs near 9 mmHg (i.e. some keratoconus patients), in order to maintain the same FR. Specifically, patients with an IOP below 10 mmHg, are thought to have a reduced or absent
W. T. G O O D M A N E T A L.
trabecular meshwork outflow, depending on episcleral venous pressure (Pederson, 1986). There is no reason to suspect that keratoconus patients have altered cardiovascular function with reduced episcleral venous pressure compared with normal patients. The reduction in trabecular meshwork episcleral venous pressure compared with normal patients. The reduction in trabecular meshwork outflow must be compensated for by an increase of outflow through an alternate pathway, most likely the uveoscleral pathway. Some of our patients with keratoconus may exhibit this outflow distribution. It has been convenient to assume that the difference in IOP between keratoconus and normal patients is due to error in applanation tonometry when measuring thin corneas. Our results indicate that increased apparent outflow facility in keratoconus patients may be at least part of the reason for decreased IOP found in the keratoconus patients. Further investigations following IOP, FR and outflow facility in keratoconus patients may answer this question. The potential for error in this project is significant. The keratoconus cornea presents many challenges to those attempting fluorophotometry. In the methods section, we have attempted to address as many possible sources of error as possible. First, we were forced to administer a slightly lower dose of fluorescein to keratoconus patients, by avoiding the use of the fluorescein–benoxinatesolution.Whenthefluorescein– benoxinate solution was used on keratoconus patients, fluorescein concentrations were greater than 2000 ng ml−" and were off the scale of the fluorophotometer. Fluorescein administered 5 hr prior to the measurement of aqueous fluorescence should not have affected aqueous production or its measurement. Additionally, if the fluorescein–benoxinate solution was not used in the normal patients, fluorescein concentrations were consistently low, and did not allow three measurements before the 200 ng ml−" level was reached. While no formula will perfectly compute corneal or anterior chamber volume (especially in keratoconus patients), and while uniform staining of the cornea in keratoconus patients may be difficult to achieve or guarantee, our effort has been to minimize these and other possible errors as much as possible. Our findings in this paper are of a preliminary nature, however we have independent verification that people with keratoconus seem to have lower intraocular pressures and we find evidence for higher apparent outflow facilities compared to normal which explains this. It is unclear why keratoconus patients should have such an increase in aqueous outflow facility. It is possible that the underlying biochemical defect in keratoconus may also affect the aqueous outflow pathways. We propose that new insights into the nature and function of the trabecular meshwork may be gained by further investigations with keratoconus patients.
AQUEOUS HUMOR DYNAMICS
Acknowledgements This study was supported in part by an unrestricted grant from Research to Prevent Blindness, Inc NIH, and RO1EY10151-01. Proprietary interest : none.
References Araie, M., Sawa, M., Nagataki, S. and Mishima, S. (1980). Aqueous humor dynamics in man as studied by oral fluorescein. Jpn J. Ophthalmol. 24, 356–62. Bill, A. (1965). The aqueous humor draining mechanism in the cynomolgus monkey (Maccus irus) with evidence for unconventional routes. Invest. Ophthalmol. Vis. Sci. 4, 911. Bill, A. and Phillips, C. I. (1971). Uveoscleral drainage of aqueous humor in human eyes. Exp. Eye Res. 12, 275. Bloom, J. N., Levene, R. Z., Thomas, G. and Kimura, R. (1976). Fluorophotometry and the rate of aqueous flow in man : I. Instrumentation and normal values. Arch. Ophthalmol. 94, 435–43. Bron, A. J., Tripathi, R. C., Harding, J. J. and Crabbe, J. M. C. (1978). Stromal loss in keratoconus. Trans. Ophthalmol. Soc., UK 98, 393–6. Brubaker, R. F., Nagataki, S., Townsend, D. J., Burns, R. R., Higgins, R. G. and Wentworth, W. (1981). The effect of age on aqueous humor formation in man. Ophthalmology 88, 283–90. Brubaker, R. F. (1982). The flow of aqueous humor in the human eye. Trans. Am. Ophthalmol. Soc. 80, 391. Brubaker, R. F. and McLaren, J. W. (1985). Uses of fluorophotometry in glaucoma research. Ophthalmology 92, 884–90. Brubaker, R. F. (1986). Clinical evaluation of the circulation of aqueous humor. In Clinical Ophthalmology (Ed. Duane, T. D.). Vol. 3, Pp. 1–11. Harper & Row : Philadelphia, U.S.A. Caffi, M. (1966). Histopathology of keratoconus. Annali. Di Ottalmologia. E Clinica. Oculista. 92, 429–35. Cannon, D. J. and Foster, C. S. (1978). Collagen crosslinking in keratoconus. Invest. Ophthalmol. Vis. Sci. 17, 63–5. Carlson, K. J., McLaren J. W., Topper, J. E. and Brubaker, R. F.(1987).Effectofbodypositiononintraocularpressure and aqueous flow. Invest. Ophthalmol. Vis. Sci. 28, 1346–52. Coakes, R. L. and Brubaker, R. F. (1979). Method of measuring aqueous humor flow and corneal endothelial permeability using a fluorophotometry nomogram. Invest. Ophthalmol. Vis. Sci. 18, 288–302.
99
Coulangeon L. M., Menerath, J. M. and Sole, P. (1987). Fluorophotome! trie par instillation : I. Debit d’humeur aqueuse et permeabilitie endothe! liale. J. Fr. Ophthalmol. 10, 365–73. Goldmann, H. (1950). U> ber Fluorescein in der menschlichen Vorderkammer. Das Kammerwasser-Minutenvolumen des Menschen. Ophthalmologica 119, 65–95. Goldmann, H. (1951). Abflussdruck, Minutenvolumen und Widerstand der Kammerwasserstro$ mung des Menschen. Doc. Ophthalmol. 5–6, 278. Jones, R. F. and Maurice, D. M. (1966). New methods of measuring the rate of aqueous flow in man with fluorescein. Exp. Eye Res. 5, 208–220. Krachmer, J. H., Feder, R. S. and Belin, M. W. (1984). Keratoconus and related noninflammatory corneal thinning disorders. Surv. Ophthalmol. 28, 293–322. Martin, P. B., Fernandez-Vila, P. C., Martinez, P. and Perez, D. A. (1992). A fluorophotometric study on the aqueous humor dynamics in primary open angle glaucoma. Int. Ophthalmol. 16, 311–14. Newsome, D. A., Foidart, J. M., Hassell, J. R., et al. (1981). Detection of specific collagen types in normal and keratoconus corneas. Invest. Ophthalmol Vis. Sci. 20, 738–50. Pederson, J. E. (1986). Ocular hypotony. Trans. Ophthalmol. Soc. UK 220, 1986. Pouliquen, Y., Faure. J. P., Limon, S. and Bisson, J. (1968). Extracellular deposits in the corneal stroma of keratoconus : an electron microscopic study. Arch. Ophthalmol. (Paris) 28, 283–94. Pouliquen, Y., Graf, B., DeKozak, Y. D. et al. (1970). E; tude morphologique du ke! ratoco# ne (I). Arch. Ophthalmol. (Paris) 30, 497–532. Pouliquen, Y., Graf, B., Hamada, R. et al. (1972). Fibrocytes in keratoconus. Morphological appearance and changes in the extracellular spaces. Optical and electronmicroscopic study. Arch. Ophthalmol. (Paris) 32, 571–586. Robert, L., Schillinger, G., Moczar M. et al. (1970). E; tude biochimique du ke! ratoco# ne. Arch. Ophthalmol. (Paris) 30, 590–608. Whitacre, M. M., Stein, R. A. and Hassanein, K. (1993). The effect of corneal thickness on applanation tonometry. Am. J. Ophthalmol. 115, 592–6. Yablonski, M. E., Zimmerman, T. J., Waltman, S. R. et al. (1978). A fluorophotometric study of the effect of topical timolol on aqueous humor dynamics. Exp. Eye Res. 27, 135. Yamamoto, Y., Nakayama, C., Kinoshita, S. (1994). Intraocular pressure in keratoconus. ARVO Annual meeting, 1994, abstract No. 487.
A Study of Aqueous Humor Dynamics in Keratoconus W A R R E N T. G O O D M ANa ; W I L L I A M D. M A T H E R Sb*, P A U L M. M U N D E Nb, K. C. O S S O I N IGb THOMAS E. D A L E Yb a
School of Medicine and b Department of Ophthalmology, University of Iowa, Iowa City, IA, U.S.A. (Received Columbia 3 May 1995 and accepted in revised form 29 August 1995) Clinical observations suggest that patients with keratoconus have lower intraocular pressures, on average, than normal subjects. Our purpose was to determine whether differences in aqueous production and outflow facility could account for differences in intraocular pressure between a group of patients with keratoconus and a group of normal, age-matched control subjects. Aqueous humor dynamics were determined by the use of fluorophotometry in one eye of seven patients with keratoconus and ten agematched normal subjects. Intraocular pressure was measured by applanation tonometry. Keratoconus patients had a statistically significant lower mean intraocular pressure than normal control subjects (11±3³1±6 mmHg vs. 16±6³2±8 mmHg, P ¯ 0±0004). The difference in mean intraocular pressure remained significant even after correcting for possible errors in applanation tonometry due to thin corneal stroma. There was no difference in mean aqueous humor flow rates in the keratoconus patients as compared to controls (2±29³0±53 µl min−" vs. 2±21³0±48 µl min−", P ¯ 0±73). The mean apparent outflow facility was 0±21³0±07 µl min−" mmHg−" for keratoconus patients compared to 0±14³0±03 µl min−" mmHg−" for controls (P ¯ 0±02). Lower mean intraocular pressure in keratoconus patients appears to be due to increased outflow facility as compared to normal subjects. # 1996 Academic Press Limited Key words : keratoconus ; aqueous humor ; outflow facility ; trabecular meshwork ; uveoscleral pathway ; fluorophotometry.
1. Introduction Keratoconus is a non-inflammatory ocular disorder characterized by progressive corneal thinning, protrusion, and scarring resulting in distorted and diminished vision. The pathology of keratoconus has been documented and summarized (Krachmer, Feder and Belin, 1984), with focus placed on the role of keratocyte or collagen disorders in the pathogenesis of keratoconus (Bron et al., 1978 ; Caffi, 1966 ; Cannon and Foster, 1978 ; Newsome et al., 1981 ; Pouliquen et al., 1968 ; Pouliquen et al., 1970 ; Pouliquen et al., 1972 ; Robert et al., 1970). Clinical observations suggest that patients with keratoconus appear to have lower intraocular pressures (IOP) than average and few patients with keratoconus develop glaucoma. Goldmann (1951) established that the major determinants of IOP were flow of aqueous humor, resistance to outflow, and episcleral venous pressure. An additional determining factor, uveoscleral flow, was established by the work of Bill (1965). Steady-state IOP of the normal eye is maintained within physiologic limits by these four factors. This paper is the first to describe aqueous dynamics in keratoconus patients. The variables studied were IOP, aqueous humor flow rate (FR), and calculation of mean apparent outflow facility from the anterior chamber (C). We hypothesize that keratoconus * For correspondence at : University of Iowa Hospitals and Clinics, Department of Ophthalmology, 200 Hawkins Drive, Iowa City, IA 52242-1091, U.S.A.
0014-4835}96}01009505 $12.00}0
patients have lower IOPs as the result of higher outflow facility. The objective of our study was to determine whether differences in aqueous production, determined by fluorophotometry (Goldmann, 1950 ; Brubaker, 1982 ; Jones and Maurice, 1966) and outflow facility could account for differences in intraocular pressure between a group of patients with keratoconus and a group of normal, age-matched control subjects. 2. Materials and Methods This research followed the tenets of the Declaration of Helsinki, informed consent was obtained after the nature and possible consequences of the study was explained, and this project proceeded with the approval of the University of Iowa Human Subjects Committee. Our project included seven eyes corresponding to seven persons with clinically evident bilateral anterior keratoconus. The control group included ten age-matched eyes of ten volunteers with no previous ocular pathology and who were not receiving drug treatment. Three of the keratoconus subjects had received one corneal transplant (the study was carried out in their fellow eye), but none of the seven subjects were undergoing any drug treatment. The keratoconus patients were volunteers selected from the Cornea Clinic population of the University of Iowa Department of Ophthalmology. Our study employed a protocol similar to those of both Yablonski et al. (1978) and Brubaker (1986). In normals, beginning at 0800 hr, one drop of 0±25 % # 1996 Academic Press Limited
96
fluorescein with 0±4 % benoxinate was instilled in one eye every minute for 4 min. This was followed by a solution of 2 % fluorescein given in one-drop doses every minute for 8 min. At approximately 1300 hr, fluorophotometric measurements of the cornea and anterior chamber using the Fluorotron Master fluorophotometer (Coherent, Palo Alto, CA, U.S.A.) were commenced. These measurements were repeated at approximately 30 min intervals for a total of three to six measurements per subject. The procedure was repeated, on a different date, in subjects whose corneal fluorescein concentration values fell below 200 ng ml−" before three measurements could be taken. In a few cases, it was necessary to delay the experimental measurements until the corneal fluorescein concentrations fell below 1000 ng ml−", as values above that level have decreased accuracy due to inner filter effects (Brubacker, 1986). Delaying measurements, to achieve ideal corneal fluorescein concentrations between 200–1000 ng ml−", does not effect results because the equilibrium between the corneal fluorescein and the fluorescein in the aqueous humor has been established, and remains intact as fluorescein is gradually cleared from the eye. The protocol was identical between the two subject groups, except that in keratoconus subjects, no anesthetic (benoxinate) was administered because of increased penetration of fluorescein into the cornea and anterior chamber of these patients. In all subjects, the cornea appeared to be evenly stained when the fluorescein application was complete. The use of benoxinate in normal subjects has no effect on fluorophotometric measurements (5 hr after application) and its use has been reported by other investigators. The FR value was computed by software, based on the Yablonski equations, in the analysing program of the Fluorotron Master fluorophotometer. Both anterior chamber volume and cornea stromal volume were necessary to measure FR. Anterior chamber volume was established from the diameter of the anterior chamber (derived from the horizontal diameter of the cornea) and the depth of the anterior chamber. Horizontal corneal diameters (accurate to tenths of a millimeter) for each subject were measured with a one chip charged coupling device video camera coupled to a VIA-100 video caliper (Boekler, Tucson, AZ, U.S.A.). Corneal stromal volume was established from measurement of corneal thickness and horizontal corneal diameter using the geometric formula for the volume of a cylinder. Anterior chamber depth and corneal thickness were determined using Immersion A-scan Biometry. Measurements were made using the MiniA-Scan (Biophysic}Alcon, Fort Worth, TX, U.S.A.) with the beam objectively aligned with the axis of the eye. This technique has an accuracy of ³20 µm for anterior chamber depth and ³15 µm for corneal thickness. At any given time, most of the fluorescein is in the
W. T. G O O D M A N E T A L.
corneal stroma, with much less in the anterior chamber. Therefore, the measurement of corneal fluorescence is critical. In most of the keratoconus patients in our study, corneal thicknesses were less than the 500 µm focal diamond of the fluorophotometer. However, the software in Fluorotron Master analysing program corrects for this to avoid underestimation of fluorescein concentration in the cornea. All subjects received an eye examination at which time their IOP values, using Goldmann applanation tonometry, were determined. The eye examination was performed after all FR measurements were completed and before the concluding ultrasound examination. Since the measurement of IOP in keratoconus may be influenced by corneal thickness and rigidity, IOP was also adjusted for variation in corneal stromal thickness (Whitacre, Stein and Hassanein, 1993). Whitacre developed equations to determine error in tonometry under manometrically controlled IOP in the 10 mmHg and also the 20 mmHg range. In our study, subjects with an IOP of 15 mmHg or less had their IOP adjusted according to the equation for the 10 mmHg controlled group, and those with an IOP greater than 15 mmHg had their IOP adjusted according to the equation for the 20 mmHg controlled group. Maximum adjustment in IOP in our study was 5 mmHg, which was the greatest adjustment made by Whitacre et al. (1993). While we are aware of the potential underestimation of TOP in patients with thin corneal stromas, we have, nevertheless, decided to use measured IOP instead of adjusted IOP in our reported calculations and results. However, the adjusted IOPs were run through all calculations to test for statistical significance. Our decision is based on the fact that the adjustment equations were specifically derived from structurally normal corneas. As the corneas in our study are not structurally normal, we do not believe that the adjusted values represent a more accurate appraisal of IOP. Apparent outflow facility (C) for each subject was calculated based on the equation C ¯ FR}IOP assuming that episcleral venous pressure (Pepi) was constant (9 mmHg), and assuming that outflow facility was primarily composed of both the reciprocal of resistance to outflow through the trabecular meshwork (Ctrab) and facility through the uveoscleral pathway (Bill, 1965 ; Bill and Phillips, 1971 ; Brubaker, 1986). We referred to this as apparent outflow facility to distinguish this measurement from the traditional tonographic outflow facility measurement. Results for FR, IOP and C were analysed using the two-sample independent groups t-test. 3. Results The demographics of the participants in the study are shown in Table I. In the normal subject group,
AQUEOUS HUMOR DYNAMICS
97
T I Age and sex characteristics for the two groups tested, normals and keratoconus patients
Age : mean range (yrs) Sex (M : F)
Normal
Keratoconus
25±7³3±6 23–33 8:2
33±6³8±2 22–46 3:4
eight were men and two were women, and the groups mean age was 26 years (range 23–33). In the keratoconus subject group, three were men and four were women, and the groups mean age was 34 years (range 22–46). Individual results for corneal thickness, FR, IOP,
and C are provided in Table II. The measured mean FR in the normal subjects was 2±21³0±48 µl min−" and in the keratoconus subjects was 2±29³0±53 µl min−" (Table III). The difference in FR values between these two groups was not statistically significant (P ¯ 0±73). The range of FR values was from 1±59 µl min−" to 2±97 µl min−" in the keratoconus group and from 1±41 µl min−" to 2±95 µl min−" in the normal group. For a P value equal to 0±05 and a sample size of 17, the β (type II error probability) is less than 0±10 yielding a power (1-β) of greater than 0±9. Keratoconus patients had a statistically significantly lower mean IOP than normal subjects (Table III). The mean IOP in keratoconus patients was 11±3³1±6 mmHg and was 16±6³2±8 mmHg in normals (P ¯ 0±0004). The range of IOPs was from 9 mmHg to 13 mmHg in the keratoconus group and
T II Individual results for corneal thickness, aqueous humor flow rate, intraocular pressure, adjusted intraocular pressure, and outflow facility in normal and keratoconus (KCN) subjects
Normal 1 2 3 4 5 6 7 8 9 10 KCN 1 2 3 4 5 6 7
Corneal thickness (µm)
Aqueous humor flow rate (µl min−")
IOP (mmHg)
Adjusted IOP (mmHg)
Outflow facility (µl min−" mmHg−")
530 580 580 580 490 580 580 550 580 550
2±20 2±95 2±81 1±58 2±01 1±41 2±31 2±27 2±39 2±21
16 17 19 16 11 16 19 21 17 14
16±6 16±7 18±7 15±7 10±6 15±7 18±7 21±2 16±7 13±7
0±14 0±17 0±15 0±10 0±18 0±09 0±12 0±11 0±14 0±15
390 490 420 290 360 550 520
2±60 1±59 1±92 2±97 1±83 2±80 2±33
9 12 13 10 10 12 13
13±3 11±6 15±9 15±0 15±0 11±7 13±4
0±29 0±13 0±15 0±30 0±18 0±23 0±18
T III Results for intraocular pressure, adjusted pressure, flow rate, and outflow facility for normals versus keratoconus patients
IOP (mmHg) mean³.. Range Adjusted IOP (mmHg) mean³.. Range Flow rate (µl min−") mean³.. Range Apparent outflow facility (µl min−" mmHg−") mean³.. Range
Normal
Keratoconus
P value
16±6³2±8 11–21 16±4³2±9 10±6–21±2 2±21³0±48 1±41–2±95 0±14³0±03 0±09–0±18
11±3³1±6 9–13 13±7³1±7 11±6–15±9 2±29³0±53 1±59–2±97 0±21³0±07 0±13–0±30
0±0004 0±04 0±73 0±02
98
from 11 mmHg to 21 mmHg in the normal group. The statistically significantly lower mean IOP in keratoconus patients remained after the IOP was adjusted for differences in corneal thickness between the two groups (Table III). The mean adjusted IOP in keratoconus patients was 13±7³1±6 mmHg (range 11±6–15±9 mmHg), and in normal subjects was 16±4³2±9 mmHg (range 10±6–21±2 mmHg) with P ¯ 0±04. Keratoconus patients had a statistically significantly higher mean apparent outflow facility (C) than normal subjects (Table III). Calculated C was 0±21³0±07 µl min−" mmHg−" (range 0±13–0±30 µl min−" mmHg−") for keratoconus patients compared to 0±14³0±03 µl min−" mmHg−" (range 0±09–0±18 µl min−" mmHg−") for controls (P ¯ 0±02). 4. Discussion This study has shown that IOP is significantly lower in keratoconus patients compared to a group of normal subjects, even when correcting for possible errors in applanation tonometry due to thin corneal stroma. A lower mean pressure in keratoconus patients has been found by other investigators. The mean IOP value of 11±3³1±6 mmHg for keratoconus patients found in our study was very close to the values found in ‘ moderate ’ (11±7³2±0 mmHg) and ‘ advanced ’ (12±0³3±5 mmHg) levels of keratoconus as demonstrated in a study by Yamamoto, Nakayama and Kinoshita (1994) using 1085 eyes of 634 patients with keratoconus. Our results show no significant difference in FR, and a significant decrease in IOP between the keratoconus and normal groups. This suggests that the rate of flow of aqueous humor is constant and not affected by small shifts in IOP. Previous studies have shown that in the normal eye, the FR is independent of IOP (Goldmann, 1950 ; Carlson et al., 1987 ; Brubaker and McLaren, 1985) and that no difference in FR has been observed between men and women (Brubaker and McLaren, 1985). Our FR values were also within the range (between 1±5 and 2±8 µl min−") found by many other investigators (Goldmann, 1950 ; Brubaker, 1982 ; Jones and Maurice, 1966 ; Martin et al., 1992 ; Coakes and Brubaker, 1979 ; Brubaker et al., 1981 ; Bloom et al., 1976 ; Araie, 1980 ; Coulangeon, Menerath and Sole, 1987). Our results imply that facility of outflow of aqueous humor is increased in keratoconus patients. If the more specific equation for aqueous humor dynamics is considered [FR ¯ Ctrab (Pio®Pepi)Fuveo] with episcleral venous pressure assumed constant (9 mmHg), aqueous humor outflow facility through the trabecular meshwork or uveoscleral flow (Fuveo) must be markedly increased in patients with IOPs near 9 mmHg (i.e. some keratoconus patients), in order to maintain the same FR. Specifically, patients with an IOP below 10 mmHg, are thought to have a reduced or absent
W. T. G O O D M A N E T A L.
trabecular meshwork outflow, depending on episcleral venous pressure (Pederson, 1986). There is no reason to suspect that keratoconus patients have altered cardiovascular function with reduced episcleral venous pressure compared with normal patients. The reduction in trabecular meshwork episcleral venous pressure compared with normal patients. The reduction in trabecular meshwork outflow must be compensated for by an increase of outflow through an alternate pathway, most likely the uveoscleral pathway. Some of our patients with keratoconus may exhibit this outflow distribution. It has been convenient to assume that the difference in IOP between keratoconus and normal patients is due to error in applanation tonometry when measuring thin corneas. Our results indicate that increased apparent outflow facility in keratoconus patients may be at least part of the reason for decreased IOP found in the keratoconus patients. Further investigations following IOP, FR and outflow facility in keratoconus patients may answer this question. The potential for error in this project is significant. The keratoconus cornea presents many challenges to those attempting fluorophotometry. In the methods section, we have attempted to address as many possible sources of error as possible. First, we were forced to administer a slightly lower dose of fluorescein to keratoconus patients, by avoiding the use of the fluorescein–benoxinatesolution.Whenthefluorescein– benoxinate solution was used on keratoconus patients, fluorescein concentrations were greater than 2000 ng ml−" and were off the scale of the fluorophotometer. Fluorescein administered 5 hr prior to the measurement of aqueous fluorescence should not have affected aqueous production or its measurement. Additionally, if the fluorescein–benoxinate solution was not used in the normal patients, fluorescein concentrations were consistently low, and did not allow three measurements before the 200 ng ml−" level was reached. While no formula will perfectly compute corneal or anterior chamber volume (especially in keratoconus patients), and while uniform staining of the cornea in keratoconus patients may be difficult to achieve or guarantee, our effort has been to minimize these and other possible errors as much as possible. Our findings in this paper are of a preliminary nature, however we have independent verification that people with keratoconus seem to have lower intraocular pressures and we find evidence for higher apparent outflow facilities compared to normal which explains this. It is unclear why keratoconus patients should have such an increase in aqueous outflow facility. It is possible that the underlying biochemical defect in keratoconus may also affect the aqueous outflow pathways. We propose that new insights into the nature and function of the trabecular meshwork may be gained by further investigations with keratoconus patients.
AQUEOUS HUMOR DYNAMICS
Acknowledgements This study was supported in part by an unrestricted grant from Research to Prevent Blindness, Inc NIH, and RO1EY10151-01. Proprietary interest : none.
References Araie, M., Sawa, M., Nagataki, S. and Mishima, S. (1980). Aqueous humor dynamics in man as studied by oral fluorescein. Jpn J. Ophthalmol. 24, 356–62. Bill, A. (1965). The aqueous humor draining mechanism in the cynomolgus monkey (Maccus irus) with evidence for unconventional routes. Invest. Ophthalmol. Vis. Sci. 4, 911. Bill, A. and Phillips, C. I. (1971). Uveoscleral drainage of aqueous humor in human eyes. Exp. Eye Res. 12, 275. Bloom, J. N., Levene, R. Z., Thomas, G. and Kimura, R. (1976). Fluorophotometry and the rate of aqueous flow in man : I. Instrumentation and normal values. Arch. Ophthalmol. 94, 435–43. Bron, A. J., Tripathi, R. C., Harding, J. J. and Crabbe, J. M. C. (1978). Stromal loss in keratoconus. Trans. Ophthalmol. Soc., UK 98, 393–6. Brubaker, R. F., Nagataki, S., Townsend, D. J., Burns, R. R., Higgins, R. G. and Wentworth, W. (1981). The effect of age on aqueous humor formation in man. Ophthalmology 88, 283–90. Brubaker, R. F. (1982). The flow of aqueous humor in the human eye. Trans. Am. Ophthalmol. Soc. 80, 391. Brubaker, R. F. and McLaren, J. W. (1985). Uses of fluorophotometry in glaucoma research. Ophthalmology 92, 884–90. Brubaker, R. F. (1986). Clinical evaluation of the circulation of aqueous humor. In Clinical Ophthalmology (Ed. Duane, T. D.). Vol. 3, Pp. 1–11. Harper & Row : Philadelphia, U.S.A. Caffi, M. (1966). Histopathology of keratoconus. Annali. Di Ottalmologia. E Clinica. Oculista. 92, 429–35. Cannon, D. J. and Foster, C. S. (1978). Collagen crosslinking in keratoconus. Invest. Ophthalmol. Vis. Sci. 17, 63–5. Carlson, K. J., McLaren J. W., Topper, J. E. and Brubaker, R. F.(1987).Effectofbodypositiononintraocularpressure and aqueous flow. Invest. Ophthalmol. Vis. Sci. 28, 1346–52. Coakes, R. L. and Brubaker, R. F. (1979). Method of measuring aqueous humor flow and corneal endothelial permeability using a fluorophotometry nomogram. Invest. Ophthalmol. Vis. Sci. 18, 288–302.
99
Coulangeon L. M., Menerath, J. M. and Sole, P. (1987). Fluorophotome! trie par instillation : I. Debit d’humeur aqueuse et permeabilitie endothe! liale. J. Fr. Ophthalmol. 10, 365–73. Goldmann, H. (1950). U> ber Fluorescein in der menschlichen Vorderkammer. Das Kammerwasser-Minutenvolumen des Menschen. Ophthalmologica 119, 65–95. Goldmann, H. (1951). Abflussdruck, Minutenvolumen und Widerstand der Kammerwasserstro$ mung des Menschen. Doc. Ophthalmol. 5–6, 278. Jones, R. F. and Maurice, D. M. (1966). New methods of measuring the rate of aqueous flow in man with fluorescein. Exp. Eye Res. 5, 208–220. Krachmer, J. H., Feder, R. S. and Belin, M. W. (1984). Keratoconus and related noninflammatory corneal thinning disorders. Surv. Ophthalmol. 28, 293–322. Martin, P. B., Fernandez-Vila, P. C., Martinez, P. and Perez, D. A. (1992). A fluorophotometric study on the aqueous humor dynamics in primary open angle glaucoma. Int. Ophthalmol. 16, 311–14. Newsome, D. A., Foidart, J. M., Hassell, J. R., et al. (1981). Detection of specific collagen types in normal and keratoconus corneas. Invest. Ophthalmol Vis. Sci. 20, 738–50. Pederson, J. E. (1986). Ocular hypotony. Trans. Ophthalmol. Soc. UK 220, 1986. Pouliquen, Y., Faure. J. P., Limon, S. and Bisson, J. (1968). Extracellular deposits in the corneal stroma of keratoconus : an electron microscopic study. Arch. Ophthalmol. (Paris) 28, 283–94. Pouliquen, Y., Graf, B., DeKozak, Y. D. et al. (1970). E; tude morphologique du ke! ratoco# ne (I). Arch. Ophthalmol. (Paris) 30, 497–532. Pouliquen, Y., Graf, B., Hamada, R. et al. (1972). Fibrocytes in keratoconus. Morphological appearance and changes in the extracellular spaces. Optical and electronmicroscopic study. Arch. Ophthalmol. (Paris) 32, 571–586. Robert, L., Schillinger, G., Moczar M. et al. (1970). E; tude biochimique du ke! ratoco# ne. Arch. Ophthalmol. (Paris) 30, 590–608. Whitacre, M. M., Stein, R. A. and Hassanein, K. (1993). The effect of corneal thickness on applanation tonometry. Am. J. Ophthalmol. 115, 592–6. Yablonski, M. E., Zimmerman, T. J., Waltman, S. R. et al. (1978). A fluorophotometric study of the effect of topical timolol on aqueous humor dynamics. Exp. Eye Res. 27, 135. Yamamoto, Y., Nakayama, C., Kinoshita, S. (1994). Intraocular pressure in keratoconus. ARVO Annual meeting, 1994, abstract No. 487.