- Email: [email protected]
Mechanisms of superficial micropunctate corneal staining with sodium fluorescein: The contribution of pooling
Contact Lens & Anterior Eye 35 (2012) 81–84
Contents lists available at SciVerse ScienceDirect
Contact Lens & Anterior Eye journal homepage: www.elsevier.com/locate/clae
Short communication
Mechanisms of superficial micropunctate corneal staining with sodium fluorescein: The contribution of pooling Kalika L. Bandamwar a,b,∗ , Qian Garrett a,b,c , Eric B. Papas a,b,c a b c
The Brien Holden Vision Institute, Sydney, Australia School of Optometry and Vision Science, Sydney, Australia The Vision Cooperative Research Centre, Sydney, Australia
a r t i c l e Keywords: Corneal staining Fluorescein Pooling
i n f o
a b s t r a c t Purpose: To establish if sodium fluorescein (SFL) dye accumulation within intercellular spaces on the ocular surface contributes to the appearance of superficial punctate corneal staining. Methods: Thirteen subjects bilaterally wore PureVisionTM lenses that had been pre-soaked in ReNu MultiPlus® multipurpose solution. After 1 h of lens wear, corneal staining with SFL was assessed using a standard slit-lamp technique. Participants who presented with bilateral, corneal staining were selected for further evaluation. A randomly selected eye was rinsed with saline three times. Fellow eyes (control) received no rinsing. After each rinse, the appearance of SFL staining was recorded without any further instillation of the dye. To eliminate any confounding effects of staining due to residual fluorescein in the tear menisci, corneal staining was induced in freshly excised, isolated, rabbit eyes by topical administration of 0.001% PHMB and staining, rinsing and grading were performed as above. Results: Nine out of 13 subjects presented with bilateral diffuse corneal staining (mean grade ± SD: 2.4 ± 0.7). The mean staining grades in test and control eyes respectively after each of the three rinses were (1) 2.41 ± 0.41, 2.25 ± 0.69 (p = 0.9); (2) 2.34 ± 0.79, 2.1 ± 0.83 (p = 0.8); and (3) 1.71 ± 0.65, 1.60 ± 0.79 (p = 0.6) there was no significant reduction in staining with rinsing (p > 0.05) and no difference was observed between test and control eyes at any sampling-point. Similar observations made in ex vivo rabbit eyes replicated these results. Conclusions: Pooling or accumulation of SFL solution within intercellular spaces does not appear to contribute to the appearance of superficial micropunctate corneal staining. © 2011 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
1. Introduction Sodium fluorescein (SFL) is an ophthalmic dye commonly used in clinical evaluation, especially for assessment of the corneal surface. Despite its long standing and regular use, the precise mechanism of SFL staining is not well understood. To accurately assess the consequences of a staining event requires an understanding of how sodium fluorescein solution interacts with ocular tissue to produce discrete regions of hyper-fluorescence. SFL staining of the cornea can have several different forms. One common manifestation that is encountered in everyday practice is diffuse, micropunctate, superficial staining of the corneal epithelium and much recent attention has been focused on a par-
∗ Corresponding author at: Brien Holden Vision Institute, School of Optometry & Vision Science, Level 3, North Wing, RMB, Gate 14, Barker Street, The University of New South Wales, Sydney NSW 2052 Australia. Tel.: +61 2 9385 6955; fax: +61 2 9385 7401. E-mail addresses: [email protected], [email protected] (K.L. Bandamwar).
ticular presentation known as solution induced corneal staining (SICS). This type of staining response is relatively asymptomatic, and occurs in association with specific lens and lens care solution combinations [1–4]. While a threshold of asymptomatic corneal staining in the non-contact lens wearing population with otherwise healthy corneas can be considered normal, the significance of SICS is the subject of debate. There appear to be several mechanisms that can lead to the appearance of a staining event in the cornea and these are set out briefly below. SFL dye might: (1) accumulate in putative inter-epithelial cellular spaces on the ocular surface (2) accumulate in ocular surface voids left by the removal of cells e.g. due to trauma, sloughing etc. (3) stain the contents of dead or damaged cells that possess compromised membranes (4) become associated with elements on the ocular surface such as tear compounds (proteins, lipids, or mucin) or glycocalyx, either
1367-0484/$ – see front matter © 2011 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.clae.2011.08.008
82
K.L. Bandamwar et al. / Contact Lens & Anterior Eye 35 (2012) 81–84
alone or in a complex with other molecules such as solution preservatives e.g. PHMB. Further, it is also possible that two or more of the above mechanisms may be active concurrently. The purpose of the current study was to focus on a specific type of staining event i.e. SICS, and evaluate whether pooling or accumulation of the dye in ocular surface spaces contributes to its appearance. We propose that simple accumulations of dye in such spaces (items 1 and 2 above) would be readily removed by a process of rinsing, whereas dye more strongly associated with cells or tear compounds would not (items 3 and 4 above). 2. Methods 2.1. Human eyes A randomized, examiner-masked, clinical trial was conducted at the Brien Holden Vision Institute (BHVI), Sydney, Australia. All procedures were conducted in accordance with the tenants of the Declaration of Helsinki. The study protocol was approved by the University of New South Wales, human research ethics committee, Sydney, Australia. All participants signed informed consent prior to the screening visit. Participants with history of ocular or systemic surgery, rigid contact lens wear, ocular injury, ocular or systemic allergies or current use of medication were excluded from the study. Based on previous work [2], SICS was induced using PureVisionTM (Bausch & Lomb, Rochester, NY, USA) lenses that had been presoaked in ReNu MultiPlus® (Bausch & Lomb, Rochester, NY, USA) overnight. These lenses were worn bilaterally for 1 h by 13 subjects aged between 22 and 34 years (Mean ± SD: 24 ± 6 years). Of these, 9 subjects who showed a bilateral staining response completed the study. Contact lens wearers were asked to discontinue their current lenses for at least 24 h prior to the study visit. Sodium fluorescein was instilled using strips (Fluorets, Chauvin Pharmaceuticals Ltd. England) wetted with a drop of sterile saline (sodium chloride 0.9%, AstraZenica Pty Ltd. NSW, Australia) and touched to the conjunctiva in the lower cul-de-sac. Staining was observed in both eyes, using a slit lamp bio-microscope at 16× magnification, with cobalt blue light and Wratten-12 Kodak filter. Observations were made at base line, before and after lens wear, and graded using the BHVI Grading Scale. [4] After 1 h of wear, subjects were examined to identify those presenting with bilateral SICS. These individuals were asked to continue for the remainder of the study while the rest were discharged. In those subjects with bilateral SICS, a test eye was assigned, at random, to receive 3 gentle rinses with 0.9% sodium chloride solution (AstraZenica Pty Ltd. NSW, Australia). Each rinse lasted 5 s and was followed by a 1 min pause, while corneal staining was recorded in both eyes. The fellow eye, which as described earlier also had staining, acted as control and was not rinsed at any stage. All the procedures were completed within 6–10 min following lens
removal. There was no further installation of fluorescein to either eye subsequent to the rinsing procedure. Observations of staining were made in both eyes before lens insertion, after lens removal and after each rinse, by an examiner masked to the selection of the test and control eyes. Observations were always recorded in the right eye first. Repeated measures analysis of variance was used to compare staining scores of test and control eyes over the duration of the experiment. 2.2. Rabbit eyes During the course of the human trial it was observed that, even after several rinses, some fluorescein persisted in the tear menisci. This raised the possibility that, even without the instillation of fresh fluorescein, residual dye from the menisci could re-pool in potential cavities on the corneal surface during the action of normal blinking. Such an event would confound the interpretation of the results. Since instructing subjects to avoid blinking after vigorous rinsing was not practical, we conducted a supplementary experiment using eyes freshly excised from New Zealand white rabbits (n = 4). These animals had been used for other independent, non-ocular medical research and their eyes, which were excised immediately after euthanasia, were made available for our study. Unlike the humans in the study, these eyes were devoid of eyelids and so could be observed in isolation without the formation of lateral tear menisci. Corneal staining was induced using drop-by-drop topical administration of 1 ml 0.001% Polyhexamethylene biguanide (PHMB, SnowDrift® Farm, South Tucson, AZ, USA) (w/v in phosphate buffered saline) over a period of 5 min[5]. PHMB was selected as it is the same preservative used in several multipurpose solutions and has been suggested to be associated with occurrence of SICS [1–4]. Following administration of fluorescein, a small volume (500 l) of saline was instilled to assist in spreading the dye across the cornea, followed by three more rinses, as described above for the human trial. Staining, rinsing and grading occurred exactly as detailed above in the human clinical trial. All the experimental procedures were performed within 15–20 min of sacrifice. 3. Results 3.1. Human eyes All participants presented with staining at all study visits and with a SICS type response in at least one eye. As this form of corneal staining is superficial and micropunctate, neither the depth nor type of staining varied among the study sample. Thus, extent of staining was used for statistical analysis. Nine out of 13 presented with bilateral SICS (Mean extent ± SD: 2.4 ± 0.4) and the mean staining grades at each time point for test and control eyes are shown in Table 1. While the magnitude of evident staining reduced progressively with each rinsing, there were no significant differences between test and control eyes at any observation point (p ≥ 0.05, repeated measures ANOVA). Fig. 1 shows representative
Table 1 Extent of corneal staining in test and control eyes immediately after lens removal and after each rinse. p represents p-value for comparison between test and control eyes at each time point. p1 represents comparison between magnitude of reduction of staining grade in test and control eyes between each rinse.
Test eye Control eye p p1
Mean extent ± Standard deviaon of corneal staining (n=9) Aer lens removal Aer 1st rinse Aer 2nd rinse Aer 3rd rinse 2.4±0.4 2.4±0.4 2.3±0.8 1.7±0.7 2.3±0.5 2.3±0.7 2.1 ±0.8 1.6±0.8 0.8 0.9 0.8 0.6 0.1 0.5 0.9
K.L. Bandamwar et al. / Contact Lens & Anterior Eye 35 (2012) 81–84
83
Fig. 1. Representative images of slit lamp bio-microscopic appearance of corneal staining in test and control eyes of humans and rabbit eyes after inducing diffuse micropunctate damage, at base line and after each rinse in test eyes. Original magnification used for clinical pictures was 10×.
images of the corneal staining seen after lens removal in test and control eyes at base line and after each rinse.
grades at base line and after each rinse in rabbit eyes; no significant reduction in staining can be seen. 4. Discussion
3.2. Rabbit eyes All rabbit eyes exhibited diffuse corneal fluorescein staining after treatment with PHMB solution, mean grade ± standard deviation was 3.21 ± 0.7. There was no reduction in corneal staining grade until the 2nd rinse (p > 0.05, Mann–Whitney test). However, similar to the observation made in humans, staining reduced after the 3rd rinse but was still not significantly different from baseline (p > 0.05, Mann–Whitney test) (Fig. 1). Table 2 shows staining
As would be expected, the amount of staining visible in control eyes reduced with time after the instillation of fluorescein due to bleaching of the dye, although not by a statistically significant amount. As a result, there was still a substantial level of hyperfluorescence evident at the time of the final observation. Strikingly, this degree of reduction in staining was not different to that seen in fellow eyes which had received 3 saline rinses. Thus the rinsing process produced no additional reduction in staining compared to
Table 2 Mean extent of corneal staining for both test and control ex vivo rabbit eye immediately after lens removal and after each rinse. p represents comparison between staining grade after each rinse compared to before.
Test eye p
Mean extent ± Standard deviaon of corneal staining (n=4) Aer PHMB exposure Aer 1st rinse Aer 2nd rinse Aer 3rd rinse 3.21±0.7 3.15±0.4 3.18±0.7 2.5±0.6 0.6
0.7
0.5
84
K.L. Bandamwar et al. / Contact Lens & Anterior Eye 35 (2012) 81–84
the lapse of time alone. Repeated rinsing should eliminate fluorescein solution from the ocular surface unless it is in some way tethered, adherent or adsorbed. Thus, dye pooling in inter-cellular spaces, the shells of ruptured cells or other voids would generally be removed or, at the very least, considerably diluted, commensurately reducing the intensity of any residual fluorescence. With this insight, any remaining fluorescence can reasonably be interpreted to represent areas on the ocular surface where fluorescein molecules have become immobilized. As mentioned earlier, the persistence of residual fluorescein in the tear menisci presented a potential confounding factor for this interpretation. It was thus important to confirm the results by observing excised rabbits eyes, in which there was no eyelid apparatus and thus no tear menisci in which residual fluorescein solution might accumulate. That the corneas from rabbits, as well as human test and control eyes, appeared essentially similar throughout the observation period strongly suggests that dye pooling is not a dominant mechanistic factor in the creation of punctuate epithelial fluorescein staining. Having eliminated pooling as a contributory factor, two main possibilities remain. The first is that fluorescein enters or stains the epithelial cells themselves. Both the observations of Wilson, who showed that the fluorescent appearance took on the shape of individual epithelial cells [6] and Tabery [7] using in vivo microscopy, point in this direction. The second proposal is that complexing of fluorescein molecules with care system components such as PHMB might also contribute if these became adherent to the ocular surface [8,9]. As the strength of such potential adhesions is not well known, it is not apparent how this system would respond to repeated irrigation such as was applied during the experiment. Thus, neither of these two mechanisms would necessarily be affected by rinsing of the ocular surface and so it is not possible to eliminate one or the other on the basis of the data collected during the current study. Further investigation is required to resolve this question and this will be the subject of a future report.
5. Conclusion Pooling or accumulation of sodium fluorescein solution within intercellular spaces does not appear to contribute significantly to the appearance of micropunctate corneal staining. This makes it likely that such clinical presentations are the result of fluorescein dye becoming more strongly associated with corneal epithelial cells or elements thereof. Conflict of interest None. References [1] Jones L, MacDougall N, Sorbara LG. Asymptomatic corneal staining associated with the use of balafilcon silicone-hydrogel contact lenses disinfected with a polyaminopropyl biguanide-preserved care regimen. Optom Vis Sci 2002;79(12):753–61. [2] Bandamwar KL, Garrett Q, Papas EB. Onset time course of solution induced corneal staining. Cont Lens Anterior Eye 2010;33(4):199–201. [3] Andrasko GJ, Ryen K. A series of evaluations of MPS and silicone hydrogel lens combinations. Rev Cornea Cont Lenses 2007;(March):7. [4] Carnt N, Willcox MDP, Evans DJ, Naduvilath T, Tilia D, Papas E, Sweeney DF, Holden BA. Corneal staining: the IER matrix study. Cont Lens Spectr 2007;(September). [5] Bandamwar, KL, Garrett, Q, Papas E. Sodium fluorescein staining of the corneal epithelium: what does it mean at a cellular level? In Association of Research in Vision and Ophthalmology, 2011: Forte Lauderdale, Florida. [6] Wilson G, Ren H, Laurent J. Corneal epithelial fluorescein staining. J Am Optom Assoc 1995;66(7):435–41. [7] Tabery HM. Corneal surface changes in Thygeson’s superficial punctate keratitis: a clinical and non-contact photomicrographic in vivo study in the human cornea. Eur J Ophthalmol 2004;14(2):85–93. [8] Bright FV, Maziarz P, Liu M, Zhang J, Merchea M. PHMB and PQ-1 impact on a liposome corneal surface membrane model, in Association of Research in Vision and Ophthalmology, IOVS, Ed., 2011: Forte Lauderdale, Florida. [9] Barrett RP, Mowery-McKee M, Hazlett LD. Punctate fluorescein corneal staining observed using polyhexametylene biguanide containing disinfecting solution not indicative of corneal surface damage. Invest Ophthalmol Vis Sci 2005;46 [E-Abstract 5732].
Contents lists available at SciVerse ScienceDirect
Contact Lens & Anterior Eye journal homepage: www.elsevier.com/locate/clae
Short communication
Mechanisms of superficial micropunctate corneal staining with sodium fluorescein: The contribution of pooling Kalika L. Bandamwar a,b,∗ , Qian Garrett a,b,c , Eric B. Papas a,b,c a b c
The Brien Holden Vision Institute, Sydney, Australia School of Optometry and Vision Science, Sydney, Australia The Vision Cooperative Research Centre, Sydney, Australia
a r t i c l e Keywords: Corneal staining Fluorescein Pooling
i n f o
a b s t r a c t Purpose: To establish if sodium fluorescein (SFL) dye accumulation within intercellular spaces on the ocular surface contributes to the appearance of superficial punctate corneal staining. Methods: Thirteen subjects bilaterally wore PureVisionTM lenses that had been pre-soaked in ReNu MultiPlus® multipurpose solution. After 1 h of lens wear, corneal staining with SFL was assessed using a standard slit-lamp technique. Participants who presented with bilateral, corneal staining were selected for further evaluation. A randomly selected eye was rinsed with saline three times. Fellow eyes (control) received no rinsing. After each rinse, the appearance of SFL staining was recorded without any further instillation of the dye. To eliminate any confounding effects of staining due to residual fluorescein in the tear menisci, corneal staining was induced in freshly excised, isolated, rabbit eyes by topical administration of 0.001% PHMB and staining, rinsing and grading were performed as above. Results: Nine out of 13 subjects presented with bilateral diffuse corneal staining (mean grade ± SD: 2.4 ± 0.7). The mean staining grades in test and control eyes respectively after each of the three rinses were (1) 2.41 ± 0.41, 2.25 ± 0.69 (p = 0.9); (2) 2.34 ± 0.79, 2.1 ± 0.83 (p = 0.8); and (3) 1.71 ± 0.65, 1.60 ± 0.79 (p = 0.6) there was no significant reduction in staining with rinsing (p > 0.05) and no difference was observed between test and control eyes at any sampling-point. Similar observations made in ex vivo rabbit eyes replicated these results. Conclusions: Pooling or accumulation of SFL solution within intercellular spaces does not appear to contribute to the appearance of superficial micropunctate corneal staining. © 2011 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved.
1. Introduction Sodium fluorescein (SFL) is an ophthalmic dye commonly used in clinical evaluation, especially for assessment of the corneal surface. Despite its long standing and regular use, the precise mechanism of SFL staining is not well understood. To accurately assess the consequences of a staining event requires an understanding of how sodium fluorescein solution interacts with ocular tissue to produce discrete regions of hyper-fluorescence. SFL staining of the cornea can have several different forms. One common manifestation that is encountered in everyday practice is diffuse, micropunctate, superficial staining of the corneal epithelium and much recent attention has been focused on a par-
∗ Corresponding author at: Brien Holden Vision Institute, School of Optometry & Vision Science, Level 3, North Wing, RMB, Gate 14, Barker Street, The University of New South Wales, Sydney NSW 2052 Australia. Tel.: +61 2 9385 6955; fax: +61 2 9385 7401. E-mail addresses: [email protected], [email protected] (K.L. Bandamwar).
ticular presentation known as solution induced corneal staining (SICS). This type of staining response is relatively asymptomatic, and occurs in association with specific lens and lens care solution combinations [1–4]. While a threshold of asymptomatic corneal staining in the non-contact lens wearing population with otherwise healthy corneas can be considered normal, the significance of SICS is the subject of debate. There appear to be several mechanisms that can lead to the appearance of a staining event in the cornea and these are set out briefly below. SFL dye might: (1) accumulate in putative inter-epithelial cellular spaces on the ocular surface (2) accumulate in ocular surface voids left by the removal of cells e.g. due to trauma, sloughing etc. (3) stain the contents of dead or damaged cells that possess compromised membranes (4) become associated with elements on the ocular surface such as tear compounds (proteins, lipids, or mucin) or glycocalyx, either
1367-0484/$ – see front matter © 2011 British Contact Lens Association. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.clae.2011.08.008
82
K.L. Bandamwar et al. / Contact Lens & Anterior Eye 35 (2012) 81–84
alone or in a complex with other molecules such as solution preservatives e.g. PHMB. Further, it is also possible that two or more of the above mechanisms may be active concurrently. The purpose of the current study was to focus on a specific type of staining event i.e. SICS, and evaluate whether pooling or accumulation of the dye in ocular surface spaces contributes to its appearance. We propose that simple accumulations of dye in such spaces (items 1 and 2 above) would be readily removed by a process of rinsing, whereas dye more strongly associated with cells or tear compounds would not (items 3 and 4 above). 2. Methods 2.1. Human eyes A randomized, examiner-masked, clinical trial was conducted at the Brien Holden Vision Institute (BHVI), Sydney, Australia. All procedures were conducted in accordance with the tenants of the Declaration of Helsinki. The study protocol was approved by the University of New South Wales, human research ethics committee, Sydney, Australia. All participants signed informed consent prior to the screening visit. Participants with history of ocular or systemic surgery, rigid contact lens wear, ocular injury, ocular or systemic allergies or current use of medication were excluded from the study. Based on previous work [2], SICS was induced using PureVisionTM (Bausch & Lomb, Rochester, NY, USA) lenses that had been presoaked in ReNu MultiPlus® (Bausch & Lomb, Rochester, NY, USA) overnight. These lenses were worn bilaterally for 1 h by 13 subjects aged between 22 and 34 years (Mean ± SD: 24 ± 6 years). Of these, 9 subjects who showed a bilateral staining response completed the study. Contact lens wearers were asked to discontinue their current lenses for at least 24 h prior to the study visit. Sodium fluorescein was instilled using strips (Fluorets, Chauvin Pharmaceuticals Ltd. England) wetted with a drop of sterile saline (sodium chloride 0.9%, AstraZenica Pty Ltd. NSW, Australia) and touched to the conjunctiva in the lower cul-de-sac. Staining was observed in both eyes, using a slit lamp bio-microscope at 16× magnification, with cobalt blue light and Wratten-12 Kodak filter. Observations were made at base line, before and after lens wear, and graded using the BHVI Grading Scale. [4] After 1 h of wear, subjects were examined to identify those presenting with bilateral SICS. These individuals were asked to continue for the remainder of the study while the rest were discharged. In those subjects with bilateral SICS, a test eye was assigned, at random, to receive 3 gentle rinses with 0.9% sodium chloride solution (AstraZenica Pty Ltd. NSW, Australia). Each rinse lasted 5 s and was followed by a 1 min pause, while corneal staining was recorded in both eyes. The fellow eye, which as described earlier also had staining, acted as control and was not rinsed at any stage. All the procedures were completed within 6–10 min following lens
removal. There was no further installation of fluorescein to either eye subsequent to the rinsing procedure. Observations of staining were made in both eyes before lens insertion, after lens removal and after each rinse, by an examiner masked to the selection of the test and control eyes. Observations were always recorded in the right eye first. Repeated measures analysis of variance was used to compare staining scores of test and control eyes over the duration of the experiment. 2.2. Rabbit eyes During the course of the human trial it was observed that, even after several rinses, some fluorescein persisted in the tear menisci. This raised the possibility that, even without the instillation of fresh fluorescein, residual dye from the menisci could re-pool in potential cavities on the corneal surface during the action of normal blinking. Such an event would confound the interpretation of the results. Since instructing subjects to avoid blinking after vigorous rinsing was not practical, we conducted a supplementary experiment using eyes freshly excised from New Zealand white rabbits (n = 4). These animals had been used for other independent, non-ocular medical research and their eyes, which were excised immediately after euthanasia, were made available for our study. Unlike the humans in the study, these eyes were devoid of eyelids and so could be observed in isolation without the formation of lateral tear menisci. Corneal staining was induced using drop-by-drop topical administration of 1 ml 0.001% Polyhexamethylene biguanide (PHMB, SnowDrift® Farm, South Tucson, AZ, USA) (w/v in phosphate buffered saline) over a period of 5 min[5]. PHMB was selected as it is the same preservative used in several multipurpose solutions and has been suggested to be associated with occurrence of SICS [1–4]. Following administration of fluorescein, a small volume (500 l) of saline was instilled to assist in spreading the dye across the cornea, followed by three more rinses, as described above for the human trial. Staining, rinsing and grading occurred exactly as detailed above in the human clinical trial. All the experimental procedures were performed within 15–20 min of sacrifice. 3. Results 3.1. Human eyes All participants presented with staining at all study visits and with a SICS type response in at least one eye. As this form of corneal staining is superficial and micropunctate, neither the depth nor type of staining varied among the study sample. Thus, extent of staining was used for statistical analysis. Nine out of 13 presented with bilateral SICS (Mean extent ± SD: 2.4 ± 0.4) and the mean staining grades at each time point for test and control eyes are shown in Table 1. While the magnitude of evident staining reduced progressively with each rinsing, there were no significant differences between test and control eyes at any observation point (p ≥ 0.05, repeated measures ANOVA). Fig. 1 shows representative
Table 1 Extent of corneal staining in test and control eyes immediately after lens removal and after each rinse. p represents p-value for comparison between test and control eyes at each time point. p1 represents comparison between magnitude of reduction of staining grade in test and control eyes between each rinse.
Test eye Control eye p p1
Mean extent ± Standard deviaon of corneal staining (n=9) Aer lens removal Aer 1st rinse Aer 2nd rinse Aer 3rd rinse 2.4±0.4 2.4±0.4 2.3±0.8 1.7±0.7 2.3±0.5 2.3±0.7 2.1 ±0.8 1.6±0.8 0.8 0.9 0.8 0.6 0.1 0.5 0.9
K.L. Bandamwar et al. / Contact Lens & Anterior Eye 35 (2012) 81–84
83
Fig. 1. Representative images of slit lamp bio-microscopic appearance of corneal staining in test and control eyes of humans and rabbit eyes after inducing diffuse micropunctate damage, at base line and after each rinse in test eyes. Original magnification used for clinical pictures was 10×.
images of the corneal staining seen after lens removal in test and control eyes at base line and after each rinse.
grades at base line and after each rinse in rabbit eyes; no significant reduction in staining can be seen. 4. Discussion
3.2. Rabbit eyes All rabbit eyes exhibited diffuse corneal fluorescein staining after treatment with PHMB solution, mean grade ± standard deviation was 3.21 ± 0.7. There was no reduction in corneal staining grade until the 2nd rinse (p > 0.05, Mann–Whitney test). However, similar to the observation made in humans, staining reduced after the 3rd rinse but was still not significantly different from baseline (p > 0.05, Mann–Whitney test) (Fig. 1). Table 2 shows staining
As would be expected, the amount of staining visible in control eyes reduced with time after the instillation of fluorescein due to bleaching of the dye, although not by a statistically significant amount. As a result, there was still a substantial level of hyperfluorescence evident at the time of the final observation. Strikingly, this degree of reduction in staining was not different to that seen in fellow eyes which had received 3 saline rinses. Thus the rinsing process produced no additional reduction in staining compared to
Table 2 Mean extent of corneal staining for both test and control ex vivo rabbit eye immediately after lens removal and after each rinse. p represents comparison between staining grade after each rinse compared to before.
Test eye p
Mean extent ± Standard deviaon of corneal staining (n=4) Aer PHMB exposure Aer 1st rinse Aer 2nd rinse Aer 3rd rinse 3.21±0.7 3.15±0.4 3.18±0.7 2.5±0.6 0.6
0.7
0.5
84
K.L. Bandamwar et al. / Contact Lens & Anterior Eye 35 (2012) 81–84
the lapse of time alone. Repeated rinsing should eliminate fluorescein solution from the ocular surface unless it is in some way tethered, adherent or adsorbed. Thus, dye pooling in inter-cellular spaces, the shells of ruptured cells or other voids would generally be removed or, at the very least, considerably diluted, commensurately reducing the intensity of any residual fluorescence. With this insight, any remaining fluorescence can reasonably be interpreted to represent areas on the ocular surface where fluorescein molecules have become immobilized. As mentioned earlier, the persistence of residual fluorescein in the tear menisci presented a potential confounding factor for this interpretation. It was thus important to confirm the results by observing excised rabbits eyes, in which there was no eyelid apparatus and thus no tear menisci in which residual fluorescein solution might accumulate. That the corneas from rabbits, as well as human test and control eyes, appeared essentially similar throughout the observation period strongly suggests that dye pooling is not a dominant mechanistic factor in the creation of punctuate epithelial fluorescein staining. Having eliminated pooling as a contributory factor, two main possibilities remain. The first is that fluorescein enters or stains the epithelial cells themselves. Both the observations of Wilson, who showed that the fluorescent appearance took on the shape of individual epithelial cells [6] and Tabery [7] using in vivo microscopy, point in this direction. The second proposal is that complexing of fluorescein molecules with care system components such as PHMB might also contribute if these became adherent to the ocular surface [8,9]. As the strength of such potential adhesions is not well known, it is not apparent how this system would respond to repeated irrigation such as was applied during the experiment. Thus, neither of these two mechanisms would necessarily be affected by rinsing of the ocular surface and so it is not possible to eliminate one or the other on the basis of the data collected during the current study. Further investigation is required to resolve this question and this will be the subject of a future report.
5. Conclusion Pooling or accumulation of sodium fluorescein solution within intercellular spaces does not appear to contribute significantly to the appearance of micropunctate corneal staining. This makes it likely that such clinical presentations are the result of fluorescein dye becoming more strongly associated with corneal epithelial cells or elements thereof. Conflict of interest None. References [1] Jones L, MacDougall N, Sorbara LG. Asymptomatic corneal staining associated with the use of balafilcon silicone-hydrogel contact lenses disinfected with a polyaminopropyl biguanide-preserved care regimen. Optom Vis Sci 2002;79(12):753–61. [2] Bandamwar KL, Garrett Q, Papas EB. Onset time course of solution induced corneal staining. Cont Lens Anterior Eye 2010;33(4):199–201. [3] Andrasko GJ, Ryen K. A series of evaluations of MPS and silicone hydrogel lens combinations. Rev Cornea Cont Lenses 2007;(March):7. [4] Carnt N, Willcox MDP, Evans DJ, Naduvilath T, Tilia D, Papas E, Sweeney DF, Holden BA. Corneal staining: the IER matrix study. Cont Lens Spectr 2007;(September). [5] Bandamwar, KL, Garrett, Q, Papas E. Sodium fluorescein staining of the corneal epithelium: what does it mean at a cellular level? In Association of Research in Vision and Ophthalmology, 2011: Forte Lauderdale, Florida. [6] Wilson G, Ren H, Laurent J. Corneal epithelial fluorescein staining. J Am Optom Assoc 1995;66(7):435–41. [7] Tabery HM. Corneal surface changes in Thygeson’s superficial punctate keratitis: a clinical and non-contact photomicrographic in vivo study in the human cornea. Eur J Ophthalmol 2004;14(2):85–93. [8] Bright FV, Maziarz P, Liu M, Zhang J, Merchea M. PHMB and PQ-1 impact on a liposome corneal surface membrane model, in Association of Research in Vision and Ophthalmology, IOVS, Ed., 2011: Forte Lauderdale, Florida. [9] Barrett RP, Mowery-McKee M, Hazlett LD. Punctate fluorescein corneal staining observed using polyhexametylene biguanide containing disinfecting solution not indicative of corneal surface damage. Invest Ophthalmol Vis Sci 2005;46 [E-Abstract 5732].