- Email: [email protected]
The effect of punctal occlusion on tear production, tear clearance, and ocular surface sensation in normal subjects
The Effect of Punctal Occlusion on Tear Production, Tear Clearance, and Ocular Surface Sensation in Normal Subjects MICHAEL T. YEN, MD, STEPHEN C. PFLUGFELDER, MD, AND WILLIAM J. FEUER, MS
● PURPOSE:
To evaluate the effect of temporary punctal occlusion on tear production, tear clearance, and ocular surface sensation in normal subjects. ● METHODS: Noncomparative interventional case series. Punctal occlusion with silicone punctal plugs was performed on nine normal subjects without complaints of ocular irritation and no known history of ocular surface disease. The lower punctum of both eyes was occluded in five subjects. The upper and lower puncta of only one eye were occluded in four subjects. Corneal and conjunctival sensations were measured with the Cochet-Bonnet anesthesiometer. Tear fluorescein clearance was evaluated with a CytoFluor II fluorophotometer by measuring the fluorescein concentration in minimally stimulated tear samples collected from the inferior tear meniscus 15 minutes after instillation of fluorescein. Schirmer test was performed without anesthesia. The tests were performed at days 0, 1, 3, 7, and 14 to 17 after punctal occlusion. Relationships were analyzed with linear regressions, and a quadratic term was used to model a return to preocclusion levels. Paired t test was used to study the change in tear fluorescein concentration. ● RESULTS: In subjects who had the lower puncta of both eyes occluded, conjunctival sensation decreased in both eyes (right eye, P ⴝ .008; left eye, P ⴝ .003), but there was no change in corneal sensation. Their tear fluorescein clearance did not show a significant change from baseline (P ⴝ .90). However, a decrease in Schirmer test scores approached statistical significance Accepted for publication Sep 19, 2000. From the Ocular Surface and Tear Center, Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami School of Medicine, Miami, Florida. This study was presented in part at the Association for Research in Vision and Ophthalmology Annual Meeting, Ft. Lauderdale, Florida, May 1999. This work was supported in part by Public Health Service Research Grant EY11915, Department of Health and Human Services, National Eye Institute, Bethesda, Maryland; an unrestricted Grant from Research to Prevent Blindness, New York, New York; and the Drs. David and Maureen Smith Ocular Surface and Tear Research Fund. Reprint requests to Stephen C. Pflugfelder, MD, Cullen Eye Institute, 6501 Fannin St, Houston, TX 77030; e-mail: [email protected]
314
©
2001 BY
(P ⴝ .056). In subjects with both puncta of only one eye occluded, we noted a decrease in corneal sensation (occluded eye, P ⴝ .042; nonoccluded eye, P ⴝ .036), conjunctival sensation (occluded, P ⴝ .001; nonoccluded, P ⴝ .060), and Schirmer scores (occluded, P ⴝ .022; nonoccluded, P ⴝ .011). Linear regression did not show a significant change in tear fluorescein clearance for either eye (occluded, P ⴝ .28; nonoccluded, P ⴝ .44). However, paired t test showed a significant worsening of tear clearance in the occluded eye from day 0 to day 3 (P ⴝ .001) followed by a subsequent improvement in tear clearance from day 3 to the end of the study period (P ⴝ .045). Paired t test did not reveal any significant changes in tear clearance in the nonoccluded eye. The quadratic term of the linear regression model demonstrated an increase toward preocclusion levels that approached statistical significance for corneal sensation (occluded, P ⴝ .053; nonoccluded, P ⴝ .099). It was statistically significant for conjunctival sensation (occluded, P ⴝ .001; nonoccluded, P ⴝ .045) and Schirmer scores (occluded, P ⴝ .047; nonoccluded, P ⴝ .044). ● CONCLUSIONS: Temporary punctal occlusion in normal subjects decreases tear production and ocular surface sensation. Our findings suggest that in addition to blocking tear drainage, punctal occlusion may affect the ocular surface/lacrimal gland interaction. These effects were more pronounced in subjects with both upper and lower puncta occluded. In normal subjects, there appears to be an autoregulatory mechanism to return tear production, tear clearance, and ocular surface sensation to preocclusion levels 14 to 17 days after punctal occlusion. (Am J Ophthalmol 2001;131:314 –323. © 2001 by Elsevier Science Inc. All rights reserved.)
O
CULAR IRRITATION IS A COMMON COMPLAINT
encountered by ophthalmologists. The mechanisms by which symptoms of irritation develop are still unclear; however, most patients with ocular irritation have been found to have a reduced tear break up time, indicative of tear film instability.1 This may be the result in part of an aqueous tear deficiency, a lipid tear
ELSEVIER SCIENCE INC. ALL
RIGHTS RESERVED.
0002-9394/01/$20.00 PII S0002-9394(00)00822-9
deficiency from meibomian gland disease, or other undefined causes.2,3 Recently, delayed tear clearance of fluorescein has been shown to correlate strongly with the severity of ocular irritation symptoms independent of aqueous tear production.1,2,4 Corneal and conjunctival sensitivity to touch were both found to decrease as tear clearance worsened. Furthermore, the concentration of the proinflammatory cytokine interleukin-1␣ in tear fluid has been shown to increase with progressive delay of tear clearance.4 These findings suggest that delayed tear clearance may lead to chronic ocular surface inflammation that affects ocular surface tactile sensation and causes irritation symptoms. These findings suggest that treatment of ocular irritation should therefore be directed toward modulating the ocular surface environment. Although the most commonly used therapy for ocular irritation is instillation of artificial tears, the improvement in symptoms is often short-lived, because the tears evaporate and drain through the lacrimal outflow system.5 Punctal occlusion is a simple procedure that can be used in an attempt to conserve naturally produced tears and also to prolong the contact time of artificial tears. The procedure has been shown to decrease elevated tear osmolarity and rose bengal staining of the ocular surface, consistent with increased tear volume from retention of aqueous tears.6 Punctal occlusion may have effects on tear physiology in addition to simple mechanical blockage of the lacrimal outflow tract. Decreased tear turnover after punctal occlusion, which could result from decreased drainage or reduced production of aqueous tears, has been previously reported.7 Paradoxically, complete occlusion of the lacrimal outflow tract often does not result in frank epiphora. These findings suggest that punctal occlusion may have an active, although still undefined, role in tear and ocular surface physiology. The purpose of our study was to evaluate the effect of temporary punctal occlusion on tear production, tear clearance, and ocular surface sensation in normal subjects.
METHODS THIS STUDY WAS CONDUCTED ACCORDING TO A PROTO-
col approved by the University of Miami School of Medicine Institutional Review Board and in accordance with the tenets of the Declaration of Helsinki. An informed consent was obtained after the nature and possible consequences of the study were explained. Two groups of subjects were evaluated. Five normal subjects with no known history of ocular surface diseases had the lower punctum of both eyes occluded with silicone punctal plugs (Oasis Medical, Glendora, California). The appropriate sized plug was chosen for each subject by measuring the size of the puncta with a measuring probe. The other group included four normal subjects who had both the upper and lower puncta of the left eye occluded VOL. 131, NO. 3
EFFECT
OF
with similar punctal plugs. The puncta of the right eye were not occluded. All nine subjects underwent an identical protocol for assessment of ocular surface sensation, tear fluorescein clearance, and Schirmer 1 test (without anesthesia) scores. These parameters were assessed in all subjects at days 0, 1, 3, 7, and 14 to 17 after punctal occlusion. A Cochet-Bonnet anesthesiometer (Luneau, Paris, France) was used for measurement of both corneal and conjunctival sensation.8,9 The stimulus from the CochetBonnet consists of a nylon filament that can be varied in length from 0 to 6 cm. The procedure for measuring ocular surface sensitivity was as follows: under visual control, the nylon filament of the Cochet-Bonnet anesthesiometer was approached smoothly and perpendicularly toward the center of the cornea. Contact was detected by the slightest bend of the nylon; sensitivity was taken as the length of the filament that gave a 50% positive response from a minimum of four stimulus applications. Subject reliability was tested by bringing the filament close to the cornea without actually touching. The same procedure was used to test conjunctiva sensation with the stimulus applied to the middle of the exposed nasal-bulbar conjunctiva. Tear fluorescein clearance was then assessed by the method previously described by Afonso and associates.4 In summary, 5 l of 2% fluorescein (IOLAB Corporation, Claremont, California) was instilled into the inferior conjunctival sac. A sample of tear fluid was collected from the lower tear meniscus after 15 minutes using a preweighed polyester rod (Transorb Rods; American Filtrona, Richmond, Virginia).10 The volume of the collected tears was determined by the weight difference of the rod before and after collection of the sample. The fluid was then extracted from the rods and transferred to wells of a 96-well polycarbonate microtiter plate (Corning 96, Corning, New York). Fluorescence was measured within 30 minutes after collection of the tear fluid using a fluorescence multiplate reader (CytoFluor II; Perspective Biosystems, Framingham, Massachusetts). This technique has been found to be a valid and reproducible method of assessing tear volume and tear fluorescence.4 Tear production was measured by Schirmer 1 testing (without anesthesia). Bilateral Schirmer 1 testing was performed by placing a dry Schirmer test strip (Alcon Laboratories, Ft. Worth, Texas) over the lower lid margin into the tear lake at the junction of the middle and lateral one third for 5 minutes. The strips were then removed, and the amount of wetting in millimeters was recorded. If complete wetting of the strips occurred before 5 minutes, the Schirmer score was extrapolated linearly to the 5-minute end point. The relationship between each of the dependent variables and the day after occlusion was studied with linear regression in which in addition to the linear term, a quadratic term was also included to model a return to preocclusion levels. The subject was accounted for as a
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
315
TABLE 1. Results After Occlusion of Both Lower Puncta
Average corneal sensation (0–6) ⫾ SD Average conjunctival sensation (0–6) ⫾ SD Average tear fluorescein concentration (log units) ⫾ SD Average Schirmer score (mm) ⫾ SD
Baseline
Day 1
Day 3
Day 7
Days 14 to 17
Linear Regression
5.9 ⫾ 0.22 4.5 ⫾ 1.66 1.95 ⫾ 0.71
5.9 ⫾ 0.22 3.7 ⫾ 2.17 1.91 ⫾ 1.11
5.9 ⫾ 0.22 3.1 ⫾ 1.78 1.88 ⫾ 0.64
5.9 ⫾ 0.22 2.2 ⫾ 1.57 1.99 ⫾ 0.54
5.9 ⫾ 0.22 1.7 ⫾ 0.97 2.15 ⫾ 0.74
P ⫽ .008 P ⫽ .90
53.2 ⫾ 60
45.2 ⫾ 59
24.9 ⫾ 28
28.4 ⫾ 27
22.8 ⫾ 17
P ⫽ .056
TABLE 2. Results After Occlusion of Upper and Lower Puncta of One Eye
Baseline
Average corneal sensation, nonoccluded ⫾ SD/ occluded ⫾ SD Average conjunctival sensation, nonoccluded ⫾ SD/ occluded ⫾ SD Average tear fluorescein concentration, nonoccluded ⫾ SD/ occluded ⫾ SD Average Schirmer score, nonoccluded ⫾ SD/ occluded ⫾ SD
Day 1
Day 3
Days 14 to 17
Linear Regression
4.1 ⫾ 0.85/ 3.5 ⫾ 0.41
3.8 ⫾ 0.48/ 3.0 ⫾ 0.91
3.7 ⫾ 0.65/ 2.3 ⫾ 0.65
3.2 ⫾ 1.19/ 2.8 ⫾ 0.29
3.4 ⫾ 0.62/ 2.9 ⫾ 0.48
P ⫽ .036/.042
1.0 ⫾ 0.41/ 1.3 ⫾ 0.29
0.9 ⫾ 0.48/ 0.8 ⫾ 0.29
0.6 ⫾ 0.48/ 0.3 ⫾ 0.29
0.6 ⫾ 0.48/ 0.4 ⫾ 0.48
1.0 ⫾ 0.41/ 0.8 ⫾ 0.50
P ⫽ .060/.001
2.55 ⫾ 0.61/ 2.31 ⫾ 0.23
2.27 ⫾ 0.68/ 2.29 ⫾ 0.52
2.42 ⫾ 0.57/ 2.67 ⫾ 0.26
2.68 ⫾ 0.12/ 2.43 ⫾ 0.34
2.34 ⫾ 0.52/ 2.13 ⫾ 0.53
P ⫽ .44/.28
18.0 ⫾ 8.3/ 28.0 ⫾ 10.2
12.8 ⫾ 3.5/ 21.3 ⫾ 6.3
11.8 ⫾ 7.5/ 19.8 ⫾ 10.0
8.3 ⫾ 2.6/ 18.8 ⫾ 8.5
8.3 ⫾ 4.3/ 19.8 ⫾ 8.6
P ⫽ .011/.022
random effect in the regression models. The paired t test was also used to study the change in tear fluorescein concentration. Left and right eye data were analyzed separately for subjects with both lower puncta occluded, and the occluded and nonoccluded eye data were analyzed separately for subjects with the upper and lower puncta of one side occluded. All subjects were measured at all follow-up times.
subjects. In subjects with both lower puncta occluded, conjunctival sensation continued to decrease throughout the study period. Both eyes behaved in a similar fashion (right eye, P ⫽ .008; left eye, P ⫽ .003; Figure 2, top). The quadratic component of the regression model approached but was not statistically significant (right eye, P ⫽ .076; left eye, P ⫽ .059). In subjects with both puncta of one eye occluded, an initial decrease in conjunctival sensation was observed in the occluded eye (P ⫽ .001), and this decrease approached statistical significance in the nonoccluded eye (P ⫽ .060). Both eyes showed a statistically significant quadratic effect (occluded eye, P ⫽ .001; nonoccluded eye, P ⫽ .045) indicating that conjunctival sensation was increasing toward preocclusion levels by days 14 to 17 (Figure 2, bottom). Tear fluorescein clearance did not show a statistically significant change in subjects with both lower puncta occluded (P ⫽ .90; Figure 3, top). In subjects with both puncta of one eye occluded, linear regression did not reveal a significant change in the tear fluorescein clearance in either the occluded eye (P ⫽ .28) or the nonoccluded eye (P ⫽ .44; Figure 3, middle). Using the paired t test, however, the occluded eye demonstrated a significant decrease in tear fluorescein clearance (increased fluorescein concentration in the tear sample) from day 0 to day 3 (P ⫽ .001) followed by a significant increase in tear fluorescein clearance from day 3 to the end of the obser-
RESULTS TABLE 1 AND TABLE 2 SUMMARIZE THE RESULTS OBTAINED
for both groups of subjects. No measurable change in corneal sensation was noted in either eye of the subjects with only the lower puncta occluded. In subjects where both puncta of one eye were occluded, corneal sensation was found to decrease initially in both the occluded eye (P ⫽ .042) as well as the nonoccluded eye (P ⫽ .036). The sensation scores appeared to be returning to preocclusion levels by the end of the study period (Figure 1). The quadratic term in the regression models suggested that this increase in corneal sensation toward the end of the study period was of borderline significance in the nonoccluded eye (P ⫽ .099) and in the occluded eye (P ⫽ .053). Conjunctival sensation decreased in both groups of 316
Day 7
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2001
FIGURE 1. In subjects with both upper and lower puncta of only one eye occluded, corneal sensation decreased in both the occluded (P ⴝ .042) and nonoccluded (P ⴝ .036) eye. The quadratic term of the linear regression showed a near statistically significant return of corneal sensation toward preocclusion levels in the occluded (P ⴝ .053) and nonoccluded (P ⴝ .099) eye.
vation period (P ⫽ .045; Figure 3, bottom). The improvement in tear fluorescein clearance was not statistically significant in the nonoccluded eye (P ⫽ .20). Subjects with both lower puncta occluded had a decrease in Schirmer 1 test scores that approached statistical significance (P ⫽ .056; Figure 4, top). Subjects with both puncta of one eye occluded had a decrease in their Schirmer 1 test scores for both the occluded eye (P ⫽ .022) as well as the nonoccluded eye (P ⫽ .011). The quadratic term in the regression models demonstrates a statistically significant increase toward stabilization of Schirmer 1 test scores in both the occluded eye (P ⫽ .047) and the nonoccluded eye (P ⫽ .044; Figure 4, bottom). Schirmer test scores decreased in four of five subjects with both lower puncta occluded (Figure 5, top), and in the occluded and nonoccluded eye in all four subjects who had both puncta of one eye occluded (Figure 5, middle and bottom).
DISCUSSION TEMPORARY PUNCTAL OCCLUSION WITH SILICONE PLUGS
is a simple procedure that can provide symptomatic relief to patients with ocular irritation symptoms, especially those with severe aqueous tear deficiency.11 It is generally believed that the therapeutic mechanism of punctal occlusion is to increase the aqueous component of the preocular tear film by blocking the lacrimal outflow tract.12 Our study clearly shows that punctal occlusion also has profound effects on ocular surface sensation and aqueous tear production. Indeed, the results of our study suggest that VOL. 131, NO. 3
EFFECT
OF
punctal occlusion influences the communication between the ocular surface and lacrimal glands. The results of our study indicate that one likely mechanism by which punctal occlusion affects tear secretion is by reducing ocular surface sensation. We found that ocular surface sensation decreased after punctal occlusion. When only the lower puncta were occluded, conjunctival sensation decreased significantly. However, no change in corneal sensation was noted in this group of subjects. One possible explanation for this finding may be that the Cochet-Bonnet anesthesiometer is not sensitive enough to measure subtle changes in corneal sensation. The cornea has a much greater sensory innervation compared with the conjunctiva.13 Because the anesthesiometer has a scale of only 0 to 6, it is plausible that any change in corneal sensation was not of sufficient magnitude to be measured with this instrument. Another possible explanation for the lack of any measurable change in corneal sensation could be that occlusion of a single punctum of an eye is not adequate to produce a measurable decrease of corneal sensation. When both puncta of one eye were occluded, conjunctival and corneal sensations were noted to decrease initially. In these normal subjects, however, the ocular surface sensation began to return to preocclusion levels by the end of the study. This finding suggests that an autoregulatory mechanism exists to normalize any changes in ocular surface sensation. This mechanism may be defective in patients with ocular irritation, because they have been noted to have decreased corneal sensitivity scores in multiple studies.1,4,9 Consistent with our findings of decreased ocular surface
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
317
FIGURE 2. (Top) In subjects with both lower puncta occluded, conjunctival sensation decreased through the course of the study. Both eyes behaved in a similar fashion (right eye, P ⴝ .008; left eye, P ⴝ .003). (Bottom) In subjects with both upper and lower puncta of only one eye occluded, conjunctival sensation decreased in both the occluded (P ⴝ .001) and nonoccluded (P ⴝ .060) eye. The quadratic term of the linear regression showed a statistically significant return of conjunctival sensation toward preocclusion levels in the occluded (P ⴝ .001) and nonoccluded (P ⴝ .045) eye.
sensation is a concomitant decrease in sensory stimulated tear production, measured by the Schirmer 1 test, after punctal occlusion. In the subjects with both puncta of one eye occluded, tear production began to stabilize toward the end of the observation period, similar to their ocular surface sensation. Several clinical reports have suggested that tear production and outflow of tears from the ocular surface are linked.14,15 Patients with acquired obstruction of the lacrimal drainage system rarely have symptoms of epiphora.16,17 Lack of significant epiphora has also been reported in patients with congenital absence of lacrimal puncta.18 Tomlinson and associates noted that a decrease 318
AMERICAN JOURNAL
in tear turnover correlated with a decrease in subjective symptoms of epiphora after punctal occlusion was performed.7 Aqueous tear production by the lacrimal glands is mainly driven by sensory neural stimulation from the trigeminal nerves innervating the ocular surface, adnexa, and nasal mucosa.19 –21 Our findings suggest that there may be receptors in the ocular surface, lacrimal outflow tract, or nasal mucosa that participate in a feedback mechanism controlling tear production. In contrast to our findings in normal subjects, some dry eye patients have been reported to have increased Schirmer 1 test scores after punctal occlusion.22 Perhaps the underlying cause of aqueous tear OF
OPHTHALMOLOGY
MARCH 2001
FIGURE 3. (Top) In subjects with both lower puncta occluded, tear fluorescein clearance did not show any significant change during the observation period (P ⴝ .90). (Middle) In subjects with both upper and lower puncta of only one eye occluded, linear regression did not reveal any significant changes in tear fluorescein clearance for either the occluded eye (P ⴝ .28) or the nonoccluded eye (P ⴝ .44). (Bottom) Paired t test showed a statistically significant decrease in tear fluorescein clearance (increasing tear fluorescein concentration) in the occluded eye from day 0 to day 3 (P ⴝ .001). This was followed by a statistically significant increase in tear fluorescein clearance (decreasing tear fluorescein concentration) from day 3 to the end of the observation period in the occluded eye (P ⴝ .045).
VOL. 131, NO. 3
EFFECT
OF
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
319
FIGURE 4. (Top) In subjects with both lower puncta occluded, Schirmer 1 test scores showed a near statistical significant decrease (P ⴝ .056). (Bottom) In subjects with both upper and lower puncta of only one eye occluded, Schirmer 1 test scores decreased in both the occluded (P ⴝ .022) and nonoccluded (P ⴝ .011) eye. The quadratic term of the linear regression showed a statistically significant stabilization of Schirmer 1 test scores in the occluded (P ⴝ .047) and nonoccluded (P ⴝ .044) eye.
tear fluorescein clearance was observed from day 0 to day 3 (Figure 3, bottom). This finding is expected, because obstruction of the lacrimal outflow tract should delay tear clearance. A decrease in aqueous tear production has also been noted to contribute to delayed tear clearance, and previous studies have correlated decreasing Schirmer 1 test scores with increasing tear fluorescein concentration.4 A significant improvement of tear clearance was also noted from day 3 to the end of the observation period in the occluded eye (Figure 3, bottom). This finding is similar to the return of ocular surface sensation and tear production, and again suggests that autoregulatory mechanisms exist to normalize any changes in the ocular surface environment.
deficiency in some dry eye patients is excessive negative feedback from the ocular surface or the tear drainage apparatus on lacrimal gland tear secretion. In these patients, punctal occlusion may reverse this process. In subjects with both puncta of one eye occluded, punctal occlusion was followed by an initial improvement in tear fluorescein clearance in both the occluded and nonoccluded eye (Figure 3, middle). This may be the reason why some patients with ocular irritation have improvement of their symptoms after punctal occlusion. This observation may be the result of simple accumulation of aqueous, resulting in a decreased tear fluorescein concentration. Using the paired t test, a significant decrease in 320
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2001
FIGURE 5. (Top) Schirmer 1 test scores decreased in four of five subjects with both lower puncta occluded. (Middle and bottom) In subjects with both upper and lower puncta of only one eye occluded, Schirmer 1 test scores decreased in all four subjects for the occluded (middle) and nonoccluded (bottom) eye.
VOL. 131, NO. 3
EFFECT
OF
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
321
The change in tear fluorescein clearance was not statistically significant in the nonoccluded eye, even though the changes in ocular surface sensation and tear production were significant. We noted a large degree of variability in the measurement of the tear fluorescein concentration. Many of our tear samples collected to measure tear clearance had a low volume (1 l or less), and it is possible that this induced greater variability in our measurement of tear fluorescein concentration. An interesting finding in our study was that a decrease in tear production and ocular surface sensation was also measured in the contralateral nonoccluded eye. Decreased tear production in the contralateral eye has been reported in patients with unilateral neurotrophic keratitis.21 Crossed sensory stimulation of tear production has been postulated, with decreased trigeminal stimulation of one eye decreasing sensory stimulated tear production bilaterally. Another possibility is that a central control of tear production exists. A third possibility is that decreased ocular surface sensation results in a decreased blink rate, which promotes an increase in tear film evaporation.1 It would seem unusual for unilateral punctal occlusion to affect the ocular surface sensation in the contralateral eye. The same patients with unilateral neurotrophic keratitis were reported to have near-normal corneal sensation scores in the unaffected eye.21 Perhaps the decreased aqueous tear production in the contralateral eye that was observed after punctal occlusion leads to this decreased sensation. This is consistent with a correlation between tear production and ocular surface sensation that was previously observed in patients with aqueous tear deficiency.1,4,9 Another possibility is that decreased tear clearance leads to accumulation of such factors as opioid peptides or inflammatory cytokines in the tear film that affect the threshold of the sensory nerves on the ocular surface. Yet another possibility may be that changes in tear osmolality could act on the sensory nerves of the ocular surface to decrease sensation. Similar to the occluded eye, ocular surface sensation and tear production returned toward preocclusion levels over time in the fellow eye. Again this suggests that regulation of tear production is a dynamic process and that autoregulatory processes appear to exist that function to maintain tear homeostasis. One implication of our study is that punctal occlusion may not be appropriate therapy for all patients with ocular irritation. Patients complaining of ocular irritation are often given a generic diagnosis of “dry eye.” However, ocular irritation may have other underlying causes, such as meibomian gland dysfunction.1 Prabhasawat and Tseng reported that delayed tear clearance strongly correlates with ocular irritation symptoms independent of aqueous tear production.2 The concentration of the proinflammatory cytokine interleukin-1␣ has also been found to increase in the tear fluid as tear clearance decreases.23 Ocular irritation may be caused by chronic ocular surface inflammation in some cases, and punctal occlusion in these 322
AMERICAN JOURNAL
patients may worsen their symptoms by further delaying their tear clearance and increasing the concentrations of pathogenic factors in the tear fluid. Anti-inflammatory therapy may be a better option for these patients.2,24 It is important to note that occlusion of only the inferior puncta in normal subjects did not significantly change their tear clearance. Therefore, additional studies will be required to assess the effects of therapeutic punctal occlusion in patients with dry eye disease. As greater knowledge is gained about the regulation of the ocular surface/lacrimal gland integrated unit, new paradigms may emerge regarding which patients with dry eye disease may benefit from punctal occlusion and which patients may have adverse consequences. We recognize that the small group of subjects in both parts of the study may be a limitation. However, despite the small sample size, statistical significance was obtained in several of the parameters examined. Also, this study was performed only on normal subjects with no complaints of ocular irritation, and therefore, our results may not be applicable to symptomatic patients with tear film deficiencies. Additional investigations into the effects of punctal occlusion in patients with ocular irritation are therefore warranted. Further studies are also needed to clarify the exact mechanism of decreased ocular surface sensation and tear production that develops after punctal occlusion. In summary, our study found that ocular surface sensation and tear production decreased after temporary punctal occlusion in normal subjects. These effects were more pronounced in subjects with both upper and lower puncta occluded. In normal subjects, there appears to be an autoregulatory mechanism to return tear production, tear clearance, and ocular surface sensation to preocclusion levels 14 to 17 days after punctal occlusion. ACKNOWLEDGMENT
We thank Oasis Medical for providing the punctal plugs used in the study.
REFERENCES 1. Pflugfelder SC, Tseng SCG, Sanabria O, et al. Evaluation of subjective assessments and objective diagnostic tests for diagnosing tear-film disorders know to cause ocular irritation. Cornea 1998;17:38 –56. 2. Prabhasawat P, Tseng SCG. Frequent association of delayed tear clearance in ocular irritation. Br J Ophthalmol 1998;82: 666 – 675. 3. Lee SH, Tseng SCG. Rose bengal staining and cytologic characteristics associated with lipid tear deficiency. Am J Ophthalmol 1997;124:736 –750. 4. Afonso AA, Monroy D, Stern ME, et al. Correlation of tear fluorescein clearance and Schirmer test scores with ocular irritation symptoms. Ophthalmology 1999;106:803– 810. 5. American Academy of Ophthalmology. Punctal occlusion for the dry eye: three-year revision. Ophthalmology 1997; 104:1521–1524. OF
OPHTHALMOLOGY
MARCH 2001
6. Gilbard JP, Rossi SR, Azar DT, Heyda KG. Effect of punctal occlusion by Freeman silicone plug insertion on tear osmolarity in dry eye disorders. CLAO J 1989;15:216 –218. 7. Tomlinson A, Craig JP, Lowther GE. The biophysical role in tear regulation. Adv Exp Med Biol 1996;438:371–379. 8. Lawrenson JG, Corbett MC, O’Brart DPS, Marshal J. Effect of beam variables on corneal sensitivity after excimer laser photorefractive keratectomy. Br J Ophthalmol 1997;81:686 – 690. 9. Xu K, Yagi Y, Tsubota K. Decrease in corneal sensitivity and change in tear function in dry eye. Cornea 1996;15:235–239. 10. Jones DT, Monroy D, Pflugfelder SC. A novel method of tear collection: comparison of glass capillary micropipettes with porous polyester rods. Cornea 1997;16:450 – 458. 11. Tuberville AW, Frederick WR, Wood TO. Punctal occlusion in tear deficiency syndromes. Ophthalmology 1982;89: 1170 –1172. 12. Cohen EJ. Punctal occlusion. Arch Ophthalmol 1999;117: 389 –390. 13. Mu¨ller LJ, Vrensen GFJM, Pels L, et al. Architecture of human corneal nerves. Invest Ophthalmol Vis Sci 1997;38: 985–994. 14. Norn MS. Tear secretion in diseased eyes. Acta Ophthalmol 1966;44:25–32. 15. Mishima S. Some physiological aspects of the precorneal tear film. Arch Ophthalmol 1965;73:233–241.
16. Dalgleish R. Incidence of idiopathic acquired obstructions in the lacrimal drainage apparatus. Br J Ophthalmol 1964;48: 373–376. 17. Franc¸ois J, Neetens A. Tear flow in man. Am J Ophthalmol 1973;76:351–358. 18. Allen JC. Congenital absence of the lacrimal punctum. J Pediatr Ophthalmol 1968;5:176 –178. 19. Tsubota K. The importance of the Schirmer test with nasal stimulation. Am J Ophthalmol 1991;111:106 –108. 20. Jordan A, Baum J. Basic tear flow. Does it exist? Ophthalmology 1980;87:920 –930. 21. Heigle TJ, Pflugfelder SC. Aqueous tear production in patients with neurotrophic keratitis. Cornea 1996;15:135– 138. 22. Willis RM, Folbert R, Krachmer JH, Holland EJ. The treatment of aqueous-deficient dry eye with removable punctal plugs. A clinical and impression-cytologic study. Ophthalmology 1987;95:514 –518. 23. Afonso AA, Sobrin L, Monroy DC, et al. Tear fluid gelatinase B activity correlates with Il-1a concentration and fluorescein clearance in ocular rosacea. Invest Ophthalmol Vis Sci 1999;40:2506 –2512. 24. Marsh P, Pflugfelder SC. Topical nonpreserved methylprednisolone therapy for keratoconjunctivitis sicca in Sjo¨gren syndrome. Ophthalmology 1999;106:811– 816.
The full-text of AJO is now available online at www.ajo.com. Authors Interactive威, currently available in limited form, is undergoing an upgrade.
VOL. 131, NO. 3
EFFECT
OF
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
323
● PURPOSE:
To evaluate the effect of temporary punctal occlusion on tear production, tear clearance, and ocular surface sensation in normal subjects. ● METHODS: Noncomparative interventional case series. Punctal occlusion with silicone punctal plugs was performed on nine normal subjects without complaints of ocular irritation and no known history of ocular surface disease. The lower punctum of both eyes was occluded in five subjects. The upper and lower puncta of only one eye were occluded in four subjects. Corneal and conjunctival sensations were measured with the Cochet-Bonnet anesthesiometer. Tear fluorescein clearance was evaluated with a CytoFluor II fluorophotometer by measuring the fluorescein concentration in minimally stimulated tear samples collected from the inferior tear meniscus 15 minutes after instillation of fluorescein. Schirmer test was performed without anesthesia. The tests were performed at days 0, 1, 3, 7, and 14 to 17 after punctal occlusion. Relationships were analyzed with linear regressions, and a quadratic term was used to model a return to preocclusion levels. Paired t test was used to study the change in tear fluorescein concentration. ● RESULTS: In subjects who had the lower puncta of both eyes occluded, conjunctival sensation decreased in both eyes (right eye, P ⴝ .008; left eye, P ⴝ .003), but there was no change in corneal sensation. Their tear fluorescein clearance did not show a significant change from baseline (P ⴝ .90). However, a decrease in Schirmer test scores approached statistical significance Accepted for publication Sep 19, 2000. From the Ocular Surface and Tear Center, Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami School of Medicine, Miami, Florida. This study was presented in part at the Association for Research in Vision and Ophthalmology Annual Meeting, Ft. Lauderdale, Florida, May 1999. This work was supported in part by Public Health Service Research Grant EY11915, Department of Health and Human Services, National Eye Institute, Bethesda, Maryland; an unrestricted Grant from Research to Prevent Blindness, New York, New York; and the Drs. David and Maureen Smith Ocular Surface and Tear Research Fund. Reprint requests to Stephen C. Pflugfelder, MD, Cullen Eye Institute, 6501 Fannin St, Houston, TX 77030; e-mail: [email protected]
314
©
2001 BY
(P ⴝ .056). In subjects with both puncta of only one eye occluded, we noted a decrease in corneal sensation (occluded eye, P ⴝ .042; nonoccluded eye, P ⴝ .036), conjunctival sensation (occluded, P ⴝ .001; nonoccluded, P ⴝ .060), and Schirmer scores (occluded, P ⴝ .022; nonoccluded, P ⴝ .011). Linear regression did not show a significant change in tear fluorescein clearance for either eye (occluded, P ⴝ .28; nonoccluded, P ⴝ .44). However, paired t test showed a significant worsening of tear clearance in the occluded eye from day 0 to day 3 (P ⴝ .001) followed by a subsequent improvement in tear clearance from day 3 to the end of the study period (P ⴝ .045). Paired t test did not reveal any significant changes in tear clearance in the nonoccluded eye. The quadratic term of the linear regression model demonstrated an increase toward preocclusion levels that approached statistical significance for corneal sensation (occluded, P ⴝ .053; nonoccluded, P ⴝ .099). It was statistically significant for conjunctival sensation (occluded, P ⴝ .001; nonoccluded, P ⴝ .045) and Schirmer scores (occluded, P ⴝ .047; nonoccluded, P ⴝ .044). ● CONCLUSIONS: Temporary punctal occlusion in normal subjects decreases tear production and ocular surface sensation. Our findings suggest that in addition to blocking tear drainage, punctal occlusion may affect the ocular surface/lacrimal gland interaction. These effects were more pronounced in subjects with both upper and lower puncta occluded. In normal subjects, there appears to be an autoregulatory mechanism to return tear production, tear clearance, and ocular surface sensation to preocclusion levels 14 to 17 days after punctal occlusion. (Am J Ophthalmol 2001;131:314 –323. © 2001 by Elsevier Science Inc. All rights reserved.)
O
CULAR IRRITATION IS A COMMON COMPLAINT
encountered by ophthalmologists. The mechanisms by which symptoms of irritation develop are still unclear; however, most patients with ocular irritation have been found to have a reduced tear break up time, indicative of tear film instability.1 This may be the result in part of an aqueous tear deficiency, a lipid tear
ELSEVIER SCIENCE INC. ALL
RIGHTS RESERVED.
0002-9394/01/$20.00 PII S0002-9394(00)00822-9
deficiency from meibomian gland disease, or other undefined causes.2,3 Recently, delayed tear clearance of fluorescein has been shown to correlate strongly with the severity of ocular irritation symptoms independent of aqueous tear production.1,2,4 Corneal and conjunctival sensitivity to touch were both found to decrease as tear clearance worsened. Furthermore, the concentration of the proinflammatory cytokine interleukin-1␣ in tear fluid has been shown to increase with progressive delay of tear clearance.4 These findings suggest that delayed tear clearance may lead to chronic ocular surface inflammation that affects ocular surface tactile sensation and causes irritation symptoms. These findings suggest that treatment of ocular irritation should therefore be directed toward modulating the ocular surface environment. Although the most commonly used therapy for ocular irritation is instillation of artificial tears, the improvement in symptoms is often short-lived, because the tears evaporate and drain through the lacrimal outflow system.5 Punctal occlusion is a simple procedure that can be used in an attempt to conserve naturally produced tears and also to prolong the contact time of artificial tears. The procedure has been shown to decrease elevated tear osmolarity and rose bengal staining of the ocular surface, consistent with increased tear volume from retention of aqueous tears.6 Punctal occlusion may have effects on tear physiology in addition to simple mechanical blockage of the lacrimal outflow tract. Decreased tear turnover after punctal occlusion, which could result from decreased drainage or reduced production of aqueous tears, has been previously reported.7 Paradoxically, complete occlusion of the lacrimal outflow tract often does not result in frank epiphora. These findings suggest that punctal occlusion may have an active, although still undefined, role in tear and ocular surface physiology. The purpose of our study was to evaluate the effect of temporary punctal occlusion on tear production, tear clearance, and ocular surface sensation in normal subjects.
METHODS THIS STUDY WAS CONDUCTED ACCORDING TO A PROTO-
col approved by the University of Miami School of Medicine Institutional Review Board and in accordance with the tenets of the Declaration of Helsinki. An informed consent was obtained after the nature and possible consequences of the study were explained. Two groups of subjects were evaluated. Five normal subjects with no known history of ocular surface diseases had the lower punctum of both eyes occluded with silicone punctal plugs (Oasis Medical, Glendora, California). The appropriate sized plug was chosen for each subject by measuring the size of the puncta with a measuring probe. The other group included four normal subjects who had both the upper and lower puncta of the left eye occluded VOL. 131, NO. 3
EFFECT
OF
with similar punctal plugs. The puncta of the right eye were not occluded. All nine subjects underwent an identical protocol for assessment of ocular surface sensation, tear fluorescein clearance, and Schirmer 1 test (without anesthesia) scores. These parameters were assessed in all subjects at days 0, 1, 3, 7, and 14 to 17 after punctal occlusion. A Cochet-Bonnet anesthesiometer (Luneau, Paris, France) was used for measurement of both corneal and conjunctival sensation.8,9 The stimulus from the CochetBonnet consists of a nylon filament that can be varied in length from 0 to 6 cm. The procedure for measuring ocular surface sensitivity was as follows: under visual control, the nylon filament of the Cochet-Bonnet anesthesiometer was approached smoothly and perpendicularly toward the center of the cornea. Contact was detected by the slightest bend of the nylon; sensitivity was taken as the length of the filament that gave a 50% positive response from a minimum of four stimulus applications. Subject reliability was tested by bringing the filament close to the cornea without actually touching. The same procedure was used to test conjunctiva sensation with the stimulus applied to the middle of the exposed nasal-bulbar conjunctiva. Tear fluorescein clearance was then assessed by the method previously described by Afonso and associates.4 In summary, 5 l of 2% fluorescein (IOLAB Corporation, Claremont, California) was instilled into the inferior conjunctival sac. A sample of tear fluid was collected from the lower tear meniscus after 15 minutes using a preweighed polyester rod (Transorb Rods; American Filtrona, Richmond, Virginia).10 The volume of the collected tears was determined by the weight difference of the rod before and after collection of the sample. The fluid was then extracted from the rods and transferred to wells of a 96-well polycarbonate microtiter plate (Corning 96, Corning, New York). Fluorescence was measured within 30 minutes after collection of the tear fluid using a fluorescence multiplate reader (CytoFluor II; Perspective Biosystems, Framingham, Massachusetts). This technique has been found to be a valid and reproducible method of assessing tear volume and tear fluorescence.4 Tear production was measured by Schirmer 1 testing (without anesthesia). Bilateral Schirmer 1 testing was performed by placing a dry Schirmer test strip (Alcon Laboratories, Ft. Worth, Texas) over the lower lid margin into the tear lake at the junction of the middle and lateral one third for 5 minutes. The strips were then removed, and the amount of wetting in millimeters was recorded. If complete wetting of the strips occurred before 5 minutes, the Schirmer score was extrapolated linearly to the 5-minute end point. The relationship between each of the dependent variables and the day after occlusion was studied with linear regression in which in addition to the linear term, a quadratic term was also included to model a return to preocclusion levels. The subject was accounted for as a
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
315
TABLE 1. Results After Occlusion of Both Lower Puncta
Average corneal sensation (0–6) ⫾ SD Average conjunctival sensation (0–6) ⫾ SD Average tear fluorescein concentration (log units) ⫾ SD Average Schirmer score (mm) ⫾ SD
Baseline
Day 1
Day 3
Day 7
Days 14 to 17
Linear Regression
5.9 ⫾ 0.22 4.5 ⫾ 1.66 1.95 ⫾ 0.71
5.9 ⫾ 0.22 3.7 ⫾ 2.17 1.91 ⫾ 1.11
5.9 ⫾ 0.22 3.1 ⫾ 1.78 1.88 ⫾ 0.64
5.9 ⫾ 0.22 2.2 ⫾ 1.57 1.99 ⫾ 0.54
5.9 ⫾ 0.22 1.7 ⫾ 0.97 2.15 ⫾ 0.74
P ⫽ .008 P ⫽ .90
53.2 ⫾ 60
45.2 ⫾ 59
24.9 ⫾ 28
28.4 ⫾ 27
22.8 ⫾ 17
P ⫽ .056
TABLE 2. Results After Occlusion of Upper and Lower Puncta of One Eye
Baseline
Average corneal sensation, nonoccluded ⫾ SD/ occluded ⫾ SD Average conjunctival sensation, nonoccluded ⫾ SD/ occluded ⫾ SD Average tear fluorescein concentration, nonoccluded ⫾ SD/ occluded ⫾ SD Average Schirmer score, nonoccluded ⫾ SD/ occluded ⫾ SD
Day 1
Day 3
Days 14 to 17
Linear Regression
4.1 ⫾ 0.85/ 3.5 ⫾ 0.41
3.8 ⫾ 0.48/ 3.0 ⫾ 0.91
3.7 ⫾ 0.65/ 2.3 ⫾ 0.65
3.2 ⫾ 1.19/ 2.8 ⫾ 0.29
3.4 ⫾ 0.62/ 2.9 ⫾ 0.48
P ⫽ .036/.042
1.0 ⫾ 0.41/ 1.3 ⫾ 0.29
0.9 ⫾ 0.48/ 0.8 ⫾ 0.29
0.6 ⫾ 0.48/ 0.3 ⫾ 0.29
0.6 ⫾ 0.48/ 0.4 ⫾ 0.48
1.0 ⫾ 0.41/ 0.8 ⫾ 0.50
P ⫽ .060/.001
2.55 ⫾ 0.61/ 2.31 ⫾ 0.23
2.27 ⫾ 0.68/ 2.29 ⫾ 0.52
2.42 ⫾ 0.57/ 2.67 ⫾ 0.26
2.68 ⫾ 0.12/ 2.43 ⫾ 0.34
2.34 ⫾ 0.52/ 2.13 ⫾ 0.53
P ⫽ .44/.28
18.0 ⫾ 8.3/ 28.0 ⫾ 10.2
12.8 ⫾ 3.5/ 21.3 ⫾ 6.3
11.8 ⫾ 7.5/ 19.8 ⫾ 10.0
8.3 ⫾ 2.6/ 18.8 ⫾ 8.5
8.3 ⫾ 4.3/ 19.8 ⫾ 8.6
P ⫽ .011/.022
random effect in the regression models. The paired t test was also used to study the change in tear fluorescein concentration. Left and right eye data were analyzed separately for subjects with both lower puncta occluded, and the occluded and nonoccluded eye data were analyzed separately for subjects with the upper and lower puncta of one side occluded. All subjects were measured at all follow-up times.
subjects. In subjects with both lower puncta occluded, conjunctival sensation continued to decrease throughout the study period. Both eyes behaved in a similar fashion (right eye, P ⫽ .008; left eye, P ⫽ .003; Figure 2, top). The quadratic component of the regression model approached but was not statistically significant (right eye, P ⫽ .076; left eye, P ⫽ .059). In subjects with both puncta of one eye occluded, an initial decrease in conjunctival sensation was observed in the occluded eye (P ⫽ .001), and this decrease approached statistical significance in the nonoccluded eye (P ⫽ .060). Both eyes showed a statistically significant quadratic effect (occluded eye, P ⫽ .001; nonoccluded eye, P ⫽ .045) indicating that conjunctival sensation was increasing toward preocclusion levels by days 14 to 17 (Figure 2, bottom). Tear fluorescein clearance did not show a statistically significant change in subjects with both lower puncta occluded (P ⫽ .90; Figure 3, top). In subjects with both puncta of one eye occluded, linear regression did not reveal a significant change in the tear fluorescein clearance in either the occluded eye (P ⫽ .28) or the nonoccluded eye (P ⫽ .44; Figure 3, middle). Using the paired t test, however, the occluded eye demonstrated a significant decrease in tear fluorescein clearance (increased fluorescein concentration in the tear sample) from day 0 to day 3 (P ⫽ .001) followed by a significant increase in tear fluorescein clearance from day 3 to the end of the obser-
RESULTS TABLE 1 AND TABLE 2 SUMMARIZE THE RESULTS OBTAINED
for both groups of subjects. No measurable change in corneal sensation was noted in either eye of the subjects with only the lower puncta occluded. In subjects where both puncta of one eye were occluded, corneal sensation was found to decrease initially in both the occluded eye (P ⫽ .042) as well as the nonoccluded eye (P ⫽ .036). The sensation scores appeared to be returning to preocclusion levels by the end of the study period (Figure 1). The quadratic term in the regression models suggested that this increase in corneal sensation toward the end of the study period was of borderline significance in the nonoccluded eye (P ⫽ .099) and in the occluded eye (P ⫽ .053). Conjunctival sensation decreased in both groups of 316
Day 7
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2001
FIGURE 1. In subjects with both upper and lower puncta of only one eye occluded, corneal sensation decreased in both the occluded (P ⴝ .042) and nonoccluded (P ⴝ .036) eye. The quadratic term of the linear regression showed a near statistically significant return of corneal sensation toward preocclusion levels in the occluded (P ⴝ .053) and nonoccluded (P ⴝ .099) eye.
vation period (P ⫽ .045; Figure 3, bottom). The improvement in tear fluorescein clearance was not statistically significant in the nonoccluded eye (P ⫽ .20). Subjects with both lower puncta occluded had a decrease in Schirmer 1 test scores that approached statistical significance (P ⫽ .056; Figure 4, top). Subjects with both puncta of one eye occluded had a decrease in their Schirmer 1 test scores for both the occluded eye (P ⫽ .022) as well as the nonoccluded eye (P ⫽ .011). The quadratic term in the regression models demonstrates a statistically significant increase toward stabilization of Schirmer 1 test scores in both the occluded eye (P ⫽ .047) and the nonoccluded eye (P ⫽ .044; Figure 4, bottom). Schirmer test scores decreased in four of five subjects with both lower puncta occluded (Figure 5, top), and in the occluded and nonoccluded eye in all four subjects who had both puncta of one eye occluded (Figure 5, middle and bottom).
DISCUSSION TEMPORARY PUNCTAL OCCLUSION WITH SILICONE PLUGS
is a simple procedure that can provide symptomatic relief to patients with ocular irritation symptoms, especially those with severe aqueous tear deficiency.11 It is generally believed that the therapeutic mechanism of punctal occlusion is to increase the aqueous component of the preocular tear film by blocking the lacrimal outflow tract.12 Our study clearly shows that punctal occlusion also has profound effects on ocular surface sensation and aqueous tear production. Indeed, the results of our study suggest that VOL. 131, NO. 3
EFFECT
OF
punctal occlusion influences the communication between the ocular surface and lacrimal glands. The results of our study indicate that one likely mechanism by which punctal occlusion affects tear secretion is by reducing ocular surface sensation. We found that ocular surface sensation decreased after punctal occlusion. When only the lower puncta were occluded, conjunctival sensation decreased significantly. However, no change in corneal sensation was noted in this group of subjects. One possible explanation for this finding may be that the Cochet-Bonnet anesthesiometer is not sensitive enough to measure subtle changes in corneal sensation. The cornea has a much greater sensory innervation compared with the conjunctiva.13 Because the anesthesiometer has a scale of only 0 to 6, it is plausible that any change in corneal sensation was not of sufficient magnitude to be measured with this instrument. Another possible explanation for the lack of any measurable change in corneal sensation could be that occlusion of a single punctum of an eye is not adequate to produce a measurable decrease of corneal sensation. When both puncta of one eye were occluded, conjunctival and corneal sensations were noted to decrease initially. In these normal subjects, however, the ocular surface sensation began to return to preocclusion levels by the end of the study. This finding suggests that an autoregulatory mechanism exists to normalize any changes in ocular surface sensation. This mechanism may be defective in patients with ocular irritation, because they have been noted to have decreased corneal sensitivity scores in multiple studies.1,4,9 Consistent with our findings of decreased ocular surface
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
317
FIGURE 2. (Top) In subjects with both lower puncta occluded, conjunctival sensation decreased through the course of the study. Both eyes behaved in a similar fashion (right eye, P ⴝ .008; left eye, P ⴝ .003). (Bottom) In subjects with both upper and lower puncta of only one eye occluded, conjunctival sensation decreased in both the occluded (P ⴝ .001) and nonoccluded (P ⴝ .060) eye. The quadratic term of the linear regression showed a statistically significant return of conjunctival sensation toward preocclusion levels in the occluded (P ⴝ .001) and nonoccluded (P ⴝ .045) eye.
sensation is a concomitant decrease in sensory stimulated tear production, measured by the Schirmer 1 test, after punctal occlusion. In the subjects with both puncta of one eye occluded, tear production began to stabilize toward the end of the observation period, similar to their ocular surface sensation. Several clinical reports have suggested that tear production and outflow of tears from the ocular surface are linked.14,15 Patients with acquired obstruction of the lacrimal drainage system rarely have symptoms of epiphora.16,17 Lack of significant epiphora has also been reported in patients with congenital absence of lacrimal puncta.18 Tomlinson and associates noted that a decrease 318
AMERICAN JOURNAL
in tear turnover correlated with a decrease in subjective symptoms of epiphora after punctal occlusion was performed.7 Aqueous tear production by the lacrimal glands is mainly driven by sensory neural stimulation from the trigeminal nerves innervating the ocular surface, adnexa, and nasal mucosa.19 –21 Our findings suggest that there may be receptors in the ocular surface, lacrimal outflow tract, or nasal mucosa that participate in a feedback mechanism controlling tear production. In contrast to our findings in normal subjects, some dry eye patients have been reported to have increased Schirmer 1 test scores after punctal occlusion.22 Perhaps the underlying cause of aqueous tear OF
OPHTHALMOLOGY
MARCH 2001
FIGURE 3. (Top) In subjects with both lower puncta occluded, tear fluorescein clearance did not show any significant change during the observation period (P ⴝ .90). (Middle) In subjects with both upper and lower puncta of only one eye occluded, linear regression did not reveal any significant changes in tear fluorescein clearance for either the occluded eye (P ⴝ .28) or the nonoccluded eye (P ⴝ .44). (Bottom) Paired t test showed a statistically significant decrease in tear fluorescein clearance (increasing tear fluorescein concentration) in the occluded eye from day 0 to day 3 (P ⴝ .001). This was followed by a statistically significant increase in tear fluorescein clearance (decreasing tear fluorescein concentration) from day 3 to the end of the observation period in the occluded eye (P ⴝ .045).
VOL. 131, NO. 3
EFFECT
OF
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
319
FIGURE 4. (Top) In subjects with both lower puncta occluded, Schirmer 1 test scores showed a near statistical significant decrease (P ⴝ .056). (Bottom) In subjects with both upper and lower puncta of only one eye occluded, Schirmer 1 test scores decreased in both the occluded (P ⴝ .022) and nonoccluded (P ⴝ .011) eye. The quadratic term of the linear regression showed a statistically significant stabilization of Schirmer 1 test scores in the occluded (P ⴝ .047) and nonoccluded (P ⴝ .044) eye.
tear fluorescein clearance was observed from day 0 to day 3 (Figure 3, bottom). This finding is expected, because obstruction of the lacrimal outflow tract should delay tear clearance. A decrease in aqueous tear production has also been noted to contribute to delayed tear clearance, and previous studies have correlated decreasing Schirmer 1 test scores with increasing tear fluorescein concentration.4 A significant improvement of tear clearance was also noted from day 3 to the end of the observation period in the occluded eye (Figure 3, bottom). This finding is similar to the return of ocular surface sensation and tear production, and again suggests that autoregulatory mechanisms exist to normalize any changes in the ocular surface environment.
deficiency in some dry eye patients is excessive negative feedback from the ocular surface or the tear drainage apparatus on lacrimal gland tear secretion. In these patients, punctal occlusion may reverse this process. In subjects with both puncta of one eye occluded, punctal occlusion was followed by an initial improvement in tear fluorescein clearance in both the occluded and nonoccluded eye (Figure 3, middle). This may be the reason why some patients with ocular irritation have improvement of their symptoms after punctal occlusion. This observation may be the result of simple accumulation of aqueous, resulting in a decreased tear fluorescein concentration. Using the paired t test, a significant decrease in 320
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2001
FIGURE 5. (Top) Schirmer 1 test scores decreased in four of five subjects with both lower puncta occluded. (Middle and bottom) In subjects with both upper and lower puncta of only one eye occluded, Schirmer 1 test scores decreased in all four subjects for the occluded (middle) and nonoccluded (bottom) eye.
VOL. 131, NO. 3
EFFECT
OF
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
321
The change in tear fluorescein clearance was not statistically significant in the nonoccluded eye, even though the changes in ocular surface sensation and tear production were significant. We noted a large degree of variability in the measurement of the tear fluorescein concentration. Many of our tear samples collected to measure tear clearance had a low volume (1 l or less), and it is possible that this induced greater variability in our measurement of tear fluorescein concentration. An interesting finding in our study was that a decrease in tear production and ocular surface sensation was also measured in the contralateral nonoccluded eye. Decreased tear production in the contralateral eye has been reported in patients with unilateral neurotrophic keratitis.21 Crossed sensory stimulation of tear production has been postulated, with decreased trigeminal stimulation of one eye decreasing sensory stimulated tear production bilaterally. Another possibility is that a central control of tear production exists. A third possibility is that decreased ocular surface sensation results in a decreased blink rate, which promotes an increase in tear film evaporation.1 It would seem unusual for unilateral punctal occlusion to affect the ocular surface sensation in the contralateral eye. The same patients with unilateral neurotrophic keratitis were reported to have near-normal corneal sensation scores in the unaffected eye.21 Perhaps the decreased aqueous tear production in the contralateral eye that was observed after punctal occlusion leads to this decreased sensation. This is consistent with a correlation between tear production and ocular surface sensation that was previously observed in patients with aqueous tear deficiency.1,4,9 Another possibility is that decreased tear clearance leads to accumulation of such factors as opioid peptides or inflammatory cytokines in the tear film that affect the threshold of the sensory nerves on the ocular surface. Yet another possibility may be that changes in tear osmolality could act on the sensory nerves of the ocular surface to decrease sensation. Similar to the occluded eye, ocular surface sensation and tear production returned toward preocclusion levels over time in the fellow eye. Again this suggests that regulation of tear production is a dynamic process and that autoregulatory processes appear to exist that function to maintain tear homeostasis. One implication of our study is that punctal occlusion may not be appropriate therapy for all patients with ocular irritation. Patients complaining of ocular irritation are often given a generic diagnosis of “dry eye.” However, ocular irritation may have other underlying causes, such as meibomian gland dysfunction.1 Prabhasawat and Tseng reported that delayed tear clearance strongly correlates with ocular irritation symptoms independent of aqueous tear production.2 The concentration of the proinflammatory cytokine interleukin-1␣ has also been found to increase in the tear fluid as tear clearance decreases.23 Ocular irritation may be caused by chronic ocular surface inflammation in some cases, and punctal occlusion in these 322
AMERICAN JOURNAL
patients may worsen their symptoms by further delaying their tear clearance and increasing the concentrations of pathogenic factors in the tear fluid. Anti-inflammatory therapy may be a better option for these patients.2,24 It is important to note that occlusion of only the inferior puncta in normal subjects did not significantly change their tear clearance. Therefore, additional studies will be required to assess the effects of therapeutic punctal occlusion in patients with dry eye disease. As greater knowledge is gained about the regulation of the ocular surface/lacrimal gland integrated unit, new paradigms may emerge regarding which patients with dry eye disease may benefit from punctal occlusion and which patients may have adverse consequences. We recognize that the small group of subjects in both parts of the study may be a limitation. However, despite the small sample size, statistical significance was obtained in several of the parameters examined. Also, this study was performed only on normal subjects with no complaints of ocular irritation, and therefore, our results may not be applicable to symptomatic patients with tear film deficiencies. Additional investigations into the effects of punctal occlusion in patients with ocular irritation are therefore warranted. Further studies are also needed to clarify the exact mechanism of decreased ocular surface sensation and tear production that develops after punctal occlusion. In summary, our study found that ocular surface sensation and tear production decreased after temporary punctal occlusion in normal subjects. These effects were more pronounced in subjects with both upper and lower puncta occluded. In normal subjects, there appears to be an autoregulatory mechanism to return tear production, tear clearance, and ocular surface sensation to preocclusion levels 14 to 17 days after punctal occlusion. ACKNOWLEDGMENT
We thank Oasis Medical for providing the punctal plugs used in the study.
REFERENCES 1. Pflugfelder SC, Tseng SCG, Sanabria O, et al. Evaluation of subjective assessments and objective diagnostic tests for diagnosing tear-film disorders know to cause ocular irritation. Cornea 1998;17:38 –56. 2. Prabhasawat P, Tseng SCG. Frequent association of delayed tear clearance in ocular irritation. Br J Ophthalmol 1998;82: 666 – 675. 3. Lee SH, Tseng SCG. Rose bengal staining and cytologic characteristics associated with lipid tear deficiency. Am J Ophthalmol 1997;124:736 –750. 4. Afonso AA, Monroy D, Stern ME, et al. Correlation of tear fluorescein clearance and Schirmer test scores with ocular irritation symptoms. Ophthalmology 1999;106:803– 810. 5. American Academy of Ophthalmology. Punctal occlusion for the dry eye: three-year revision. Ophthalmology 1997; 104:1521–1524. OF
OPHTHALMOLOGY
MARCH 2001
6. Gilbard JP, Rossi SR, Azar DT, Heyda KG. Effect of punctal occlusion by Freeman silicone plug insertion on tear osmolarity in dry eye disorders. CLAO J 1989;15:216 –218. 7. Tomlinson A, Craig JP, Lowther GE. The biophysical role in tear regulation. Adv Exp Med Biol 1996;438:371–379. 8. Lawrenson JG, Corbett MC, O’Brart DPS, Marshal J. Effect of beam variables on corneal sensitivity after excimer laser photorefractive keratectomy. Br J Ophthalmol 1997;81:686 – 690. 9. Xu K, Yagi Y, Tsubota K. Decrease in corneal sensitivity and change in tear function in dry eye. Cornea 1996;15:235–239. 10. Jones DT, Monroy D, Pflugfelder SC. A novel method of tear collection: comparison of glass capillary micropipettes with porous polyester rods. Cornea 1997;16:450 – 458. 11. Tuberville AW, Frederick WR, Wood TO. Punctal occlusion in tear deficiency syndromes. Ophthalmology 1982;89: 1170 –1172. 12. Cohen EJ. Punctal occlusion. Arch Ophthalmol 1999;117: 389 –390. 13. Mu¨ller LJ, Vrensen GFJM, Pels L, et al. Architecture of human corneal nerves. Invest Ophthalmol Vis Sci 1997;38: 985–994. 14. Norn MS. Tear secretion in diseased eyes. Acta Ophthalmol 1966;44:25–32. 15. Mishima S. Some physiological aspects of the precorneal tear film. Arch Ophthalmol 1965;73:233–241.
16. Dalgleish R. Incidence of idiopathic acquired obstructions in the lacrimal drainage apparatus. Br J Ophthalmol 1964;48: 373–376. 17. Franc¸ois J, Neetens A. Tear flow in man. Am J Ophthalmol 1973;76:351–358. 18. Allen JC. Congenital absence of the lacrimal punctum. J Pediatr Ophthalmol 1968;5:176 –178. 19. Tsubota K. The importance of the Schirmer test with nasal stimulation. Am J Ophthalmol 1991;111:106 –108. 20. Jordan A, Baum J. Basic tear flow. Does it exist? Ophthalmology 1980;87:920 –930. 21. Heigle TJ, Pflugfelder SC. Aqueous tear production in patients with neurotrophic keratitis. Cornea 1996;15:135– 138. 22. Willis RM, Folbert R, Krachmer JH, Holland EJ. The treatment of aqueous-deficient dry eye with removable punctal plugs. A clinical and impression-cytologic study. Ophthalmology 1987;95:514 –518. 23. Afonso AA, Sobrin L, Monroy DC, et al. Tear fluid gelatinase B activity correlates with Il-1a concentration and fluorescein clearance in ocular rosacea. Invest Ophthalmol Vis Sci 1999;40:2506 –2512. 24. Marsh P, Pflugfelder SC. Topical nonpreserved methylprednisolone therapy for keratoconjunctivitis sicca in Sjo¨gren syndrome. Ophthalmology 1999;106:811– 816.
The full-text of AJO is now available online at www.ajo.com. Authors Interactive威, currently available in limited form, is undergoing an upgrade.
VOL. 131, NO. 3
EFFECT
OF
PUNCTAL OCCLUSION
IN
NORMAL SUBJECTS
323