The frontalis suspension procedure creates a link between the frontalis muscle and the tarsus of the upper eyelid and is the general procedure of choice for congenital ptosis with poor levator function.1 Although the results of this procedure are more uniform and predictable than maximal levator resection results, the lagophthalmos and the eyelid lag on downward gaze is more pronounced, leading to superficial punctate corneal erosions of varying severity in the exposed area.2 Following ptosis correction, it is difficult for clinicians to judge the severity of symptoms in young children because they are not always able to clearly articulate symptoms other than, perhaps, tearing, photophobia, and frequent forced blinking of the eye.
Bell’s phenomenon involves the upward movement of the eye associated with eyelid closure and is a protective mechanism that may play an important role following ptosis correction. It does not occur during short blinks, and occurred in only approximately half of the subjects assessed during voluntary unrestrained prolonged eyelid closure.3 Cases of temporary inverted Bell’s phenomenon as an attempted protective mechanism have been reported following ptosis surgery.4,5 This phenomenon has been reported to be an important clinical marker for evaluation and follow-up of neuro-ophthalmic maturation in neonates.6 We frequently observe cases of patients with congenital ptosis who have poor Bell’s phenomenon or limited upward eye movement.
Little is known about the change in ocular surface and tear film characteristics associated with Bell’s phenomenon after ptosis surgery. The current study tested the hypothesis that the poor blinking associated with lagophthalmos after ptosis surgery leads to prolonged ocular surface exposure and increased evaporation of tear film.
The study examined whether prolonged drying of the ocular surface due to ptosis surgery-related lagophthalmos affected tear film stability, tear secretion, corneal sensitivity, and ocular surface alterations. However, a normally functioning Bell’s phenomenon mechanism may spread tears evenly over the ocular surface, reduce corneal exposure time, and lower the risk of ocular surface desiccation, even though eyelid blinking may be incomplete. Therefore, the study enrolled patients with both intact and poorly functioning Bell’s phenomenon. Patients were assessed using Schirmer tests, tear film breakup time tests, corneal sensitivity measurements, and fluorescein staining of the ocular surface. The patient tear parameter data were compared to data from normal subjects (control).
Patients and Methods
Of the patients with congenital ptosis who underwent frontalis suspension surgery in the Department of Ophthalmology of Yonsei University College of Medicine from November 2004 to December 2005, 23 eyes of 15 patients with poor Bell’s phenomenon (group 1) and 33 eyes of 21 patients with intact Bell’s phenomenon (group 2) were enrolled in the study. Bell’s phenomenon was defined as poor when no upward eye movement was observed during attempted forceful eyelid closure with the eyelids restrained in the open position.
The study involved pediatric patients with congenital ptosis only, so the data were not affected by any possible age-related changes to the eye and ocular adnexa that can affect tear parameters, such as conjunctivochalasis, eyelid laxity, and functional tear outflow obstruction. All patients had a levator function of 4 mm or less and underwent frontalis suspension repair using the autogenous fascia lata without reverse frost suturing performed by the same surgeon (SYL). The change in eyelid height was calculated as the difference in the interpalpebral fissure before surgery and at 6 months postoperatively. Control group data were derived from 30 eyes of 15 age-matched subjects without ptosis, entropion, allergic conjunctivitis, or other ocular surface conditions (group 3) (Table 1).
Table 1. Demographics of the Study Populations
The patients and control subjects underwent ocular surface examinations including corneal sensitivity measurement using a Cochet-Bonnet esthesiometer (Luneau Ophthalmologie, Chartres Cedex, France),7,8 Schirmer II testing (Schirmer test with anesthesia), break-up time measurement, and fluorescein staining scoring of the cornea and conjunctiva. Corneal sensitivity measurement began at the maximal filament length (60 mm). If no response was obtained, the length of the filament was reduced to 55 mm and then further reduced in 5-mm increments until a positive response was obtained. The break-up time was measured five times in succession for each eye. Staining was recorded for five regions of the cornea (superior, inferior, central, nasal, and temporal) and four regions of the conjunctiva (superior, inferior, nasal, and temporal).9 All regions were then summed to give a total staining score, and abnormal staining was classified as staining of grade 3 or greater.
Informed consent was obtained from all subjects and the study adhered to the tenets of the Declaration of Helsinki. Tests were performed prior to and at 1, 3, and 6 months after surgery. All measurements were performed by the same operator (JSY) to minimize bias. Data are presented as mean ± standard deviation.
To determine the height of lagophthalmos objectively, patients were photographed with gently closed eyes at 6 months postoperatively and the height was then measured using the Scion Map image program. An 8-mm diameter circular piece of tape was placed in the middle of the glabella as a guideline for measurement (Fig. 1). The correlation between the degree of lagophthalmos and break-up time and fluorescein staining scores at 6 months postoperatively was determined to assess the effect of lagophthalmos on tear film stability and ocular surface. The level of lagophthalmos and the eyelid elevation were compared between patients with abnormal and normal fluorescein staining scores in both groups 1 and 2.
Figure 1. Lagophthalmos Measured Using Photographic Analysis. (top) Photograph of a Patient 6 Months After Frontalis Suspension Surgery on the Right Eye. (bottom) Lagophthalmos in the Right Eye with Intact Bell’s Phenomenon.
SPSS software (version 13.0; SPSS Inc., Chicago, IL) was used for statistical analysis. Healthy controls and patients were compared in terms of age, levator function, and study parameters using t tests. The changes in Schirmer value, break-up time, and corneal sensitivity during postoperative follow-up were analyzed using repeated-measures analysis of variance. Pearson correlation analysis was used to determine the significance of the relationships between postoperative lagophthalmos and break-up time and fluorescein staining scores. P values of less than .05 were considered to indicate a significant difference.
Patient and control groups did not differ significantly in terms of age (one-way analysis of variance, P = .998). Groups 1 and 2 did not differ significantly in terms of mean levator function and postoperative lagophthalmos (independent two-sample t test, P = .541 and P = .735, respectively) (Table 1).
During the postoperative follow-up period, break-up time decreased in group 1 (repeated measures analysis of variance, P < .001) (Table 2) but remained stable in group 2. Furthermore, the mean break-up time was lower in group 1 than in group 2 at every follow-up time point (independent two-sample t test, P < .001).
Table 2. Test Results for Treated and Control Eyes
There were no changes in either Schirmer values or corneal sensitivity in either group 1 or 2 during the postoperative follow-up period (repeated measures analysis of variance, P > .05) (Table 2). Additionally, these parameters did not significantly differ between groups 1 and 2 at any follow-up time point (t test, P > .05).
Groups 1 (0.3 ± 0.5 mm) and 2 (0.4 ± 0.5 mm) did not differ significantly in terms of mean baseline fluorescein staining scores prior to surgery (independent two-sample t test, P = .345). Fluorescein staining scores increased at 6 months postoperatively in both groups 1 (3.3 ± 1.1 mm) and 2 (1.0 ± 0.9 mm) (paired t test, P < .001 and P = .002, respectively), with the increase being greater for group 1 (independent two-sample t test, P < .001). In addition, there were more abnormal fluorescein staining scores (grade 3 or more) at 6 months postoperatively in group 1 (17 eyes, 74%) than in group 2 (3 eyes, 9%) (Pearson chi-square, P < .001).
For group 1, the degree of postoperative lagophthalmos had a strong negative correlation with break-up time (Pearson correlation, r = −0.804, P < .001) (Fig. 2A) and a high positive correlation with fluorescein staining scores (Pearson correlation, r = 0.677, P < .001) (Fig. 2B) at 6 months postoperatively. In contrast, the degree of postoperative lagophthalmos did not correlate with break-up time (Pearson correlation, r = −0.027, P = .880) or fluorescein staining scores (Pearson correlation, r = 0.018, P = .920) in group 2.
Figure 2. Correlation of the Degree of Lagophthalmos with (A) Tear Break-Up Time (but) and (B) Corneal Fluorescein Staining Scores in Patients with Poor Bell’s Phenomenon. The Degree of Lagophthalmos Correlated Highly with (A) but (pearson Correlation, r = −0.804, p < .001) and (B) Fluorescein Staining Scores (pearson Correlation, r = 0.677, p < .001).
In group 1, the 17 eyes with abnormal fluorescein staining scores had a higher mean postoperative lagophthalmos and amount of eyelid elevation compared to the 6 eyes with normal fluorescein staining scores in that group (independent two-sample t test, P = .001 and P = .001, respectively) (Table 3). In contrast, significant differences were not observed between eyes with abnormal (3 eyes) and normal (30 eyes) fluorescein staining scores in group 2 (independent two-sample t test, P = .082 and P = .361, respectively) (Table 3). The amount of eyelid elevation was positively correlated with the level of postoperative lagophthalmos in both groups 1 (Pearson correlation, r = .956, P < .001) and 2 (Pearson correlation, r = .844, P < .001).
Table 3. Comparison of Postoperative Lagophthalmos and Amount of Eyelid Elevation Between Patients with Abnormal and Normal Fluorescein Staining Scores in Groups 1 and 2
The current study found that Bell’s phenomenon was an important protective mechanism for maintaining tear function and ocular surface stability in eyes with lagophthalmos. Lagophthalmos in patients with a poorly functioning Bell’s phenomenon mechanism induces a dry eye state related to low break-up time with increased risk of ocular surface desiccation. Unprotected tear evaporation could cause tear film instability and subsequent ocular surface damage. The pathway may also travel in the reverse direction where a damaged ocular surface may result in a shortened break-up time, which may lead to exposure keratitis.
In normally functioning eyes, approximately 80% of blinks are complete (the descending upper eyelid covers more than two-thirds of the cornea), 18% are incomplete (the descending upper eyelid covers less than two-thirds of the cornea), and less than 2% are twitch blinks.10 The main function of blinking is to reform the precorneal tear film, and it is important for the release of meibomian gland lipid. A decrease in meibomian secretion due to incomplete blinking could result in a compromised lipid layer and increased evaporation of the aqueous subphase, with thinning of the tear film and increased contamination of the mucin layer.11 Most blinks are incomplete in eyes with higher levels of lagophthalmos, resulting in a lengthened tear drying time and tears not being spread evenly over the ocular surface by the eyelids, eventually leading to tear film thinning and rupture. However, in the current study, there were no significant changes in tear parameters after ptosis surgery in patients with intact Bell’s phenomenon. Thus, a normally functioning Bell’s mechanism appears to protect the ocular surface and maintain tear film stability even though the level of lagophthalmos is high, probably by reducing the ocular surface tension via compressing the tear layer and resurfacing the cornea, thereby shortening tear drying time.
Schirmer values did not change postoperatively in either of the groups with poor or intact Bell’s phenomenon and there was no difference in Schirmer values between these groups at any postoperative follow-up time. Only break-up time decreased postoperatively and was associated with the degree of lagophthalmos in patients with poor Bell’s phenomenon. This result suggests that evaporative tear loss in the unprotected eye without Bell’s phenomenon may be the main factor causing ocular surface desiccation, independent of the level of aqueous tear production. In the current study, any alterations in tear secretion may not have been observed due to the relatively short follow-up of 6 months.
Patients with dry eye syndrome have been previously noted to have reduced corneal sensitivity.12 A reduced blink rate in dry eye syndrome may reduce ocular surface sensation, which may create a vicious cycle of reduced aqueous tear production and hence worsening of dry eye symptoms. In the current study, corneal sensitivity did not change postoperatively in patients with either poor or intact Bell’s phenomenon. An intact ocular sensation could probably induce frequent reflex blinking, which is helpful in preventing corneal dryness. However, because most of the change in corneal sensitivity in dry eye syndrome is the result of long-term ocular surface damage, longer follow-up studies are likely required to observe such an alteration.
In patients with poor Bell’s phenomenon, the mean lagophthalmos and amount of eyelid elevation, which induces exposure keratitis, were 5.2 and 5.8 mm, respectively. In addition, correlation analysis indicated a higher eyelid elevation is associated with higher levels of lagophthalmos. Accordingly, to prevent exposure keratitis in patients with poor Bell’s phenomenon, ptosis should not be corrected more than 5.5 mm and patients with lagophthalmos of 5.0 mm or more should be examined and monitored carefully.
- Crawford JS. Repair of ptosis using frontalis muscle and fascia lata. Trans Am Acad Ophthalmol Otolaryngol1956;60:672–678.
- Beyer-Machule CK. Congenital ptosis and complications of ptosis surgery. Plast Reconstr Surg1988;81:789–799. doi:10.1097/00006534-198805000-00028 [CrossRef]
- Collewijn H, van der Steen J, Steinman RM. Human eye movements associated with blinks and prolonged eyelid closure. Neurophysiology1985;54:11–27.
- Gupta JS, Chatterjee A, Kumar K. Inverse Bell’s phenomenon as a protective mechanism. Am J Ophthalmol1965;59:931–933.
- Betharia SM, Sharma V. Inverse Bell’s phenomenon observed following levator resection for blepharoptosis. Graefes Arch Clin Exp Ophthalmol2006;244:868–870. doi:10.1007/s00417-005-0122-4 [CrossRef]
- Snir M, Kremer I, Kuperman A, Merlob P, Yassur Y. Bell’s phenomenon in newborns and premature babies. Br J Ophthalmol1996;80:553–555. doi:10.1136/bjo.80.6.553 [CrossRef]
- Lawrenson JG, Corbett MC, O’Brat DP, Marshall J. Effect of beam variables on corneal sensitivity after excimer laser photorefractive keratectomy. Br J Ophthalmol1997;81:686–690. doi:10.1136/bjo.81.8.686 [CrossRef]
- Macri A, Pflugfelder SC. Correlation of the Schirmer 1 and fluorescein clearance tests with the severity of corneal epithelial and eyelid disease. Arch Ophthalmol2000;118:1632–1638.
- Lemp MA. Report of the National Eye Institute/Industry workshop on Clinical Trials in Dry Eyes. CLAO J1995;21:221–232.
- Wolkoff P, Nojgaard JK, Troiano P, Piccoli B. Eye complaints in the office environment: precorneal tear film integrity influenced by eye blinking efficiency. Occup Environ Med2005;62:4–12. doi:10.1136/oem.2004.016030 [CrossRef]
- Holly FJ. Physical chemistry of the normal and disordered tear film. Trans Ophthalmol Soc U K1985;104:374–380.
- Pflugfelder SC, Tseng SC, Sanabria O, et al. Evaluation of subjective assessments and objective diagnostic tests for diagnosing tear-film disorders known to cause ocular irritation. Cornea1998;17:38–56. doi:10.1097/00003226-199801000-00007 [CrossRef]
Demographics of the Study Populations
|Characteristic||Group 1 (Poor BP)||Group 2 (Intact BP)||Group 3 (Control)|
|No. of patients (eyes)||15 (23)||21 (33)||15 (30)|
|Age (y)||11.5 ± 3.4||10.2 ± 3.6||13.1 ± 3.9|
|Type of surgery (bilateral:unilateral)||8:7||12:9|
|Levator function (mm)||2.8 ± 1.1||2.7 ± 0.9||12.5 ± 2.1|
|Postoperative lagophthalmos (mm)||4.7 ± 1.4||4.8 ± 1.2|
Test Results for Treated and Control Eyes
|1 Month||3 Months||6 Months|
|Tear break-up time (sec)|
| Group 1 (poor BP)||8.8 ± 1.4||5.4 ± 1.8||5.0 ± 1.8||4.7 ± 1.9||< .001|
| Group 2 (intact BP)||8.8 ± 0.9||8.3 ± 1.4||8.3 ± 1.3||8.6 ± 1.1||.678|
| Group 3 (control)||8.9 ± 1.1|
| Pb||.901||< .001||< .001||< .001|
|Schirmer test (mm)|
| Group 1 (poor BP)||9.5 ± 2.2||9.4 ± 1.9||9.2 ± 2.7||8.7 ± 2.9||.597|
| Group 2 (intact BP)||9.7 ± 2.2||9.6 ± 2.2||9.2 ± 2.6||9.2 ± 3.0||.791|
| Group 3 (control)||9.8 ± 2.1|
|Corneal sensitivity (mm/g/s)|
| Group 1 (poor BP)||5.7 ± 0.4||5.6 ± .0.4||5.6 ± 0.4||5.6 ± 0.4||.917|
| Group 2 (intact BP)||5.7 ± 0.4||5.7 ± 0.4||5.6 ± 0.4||5.7 ± 0.4||.860|
| Group 3 (control)||5.8 ± 0.3|
Comparison of Postoperative Lagophthalmos and Amount of Eyelid Elevation Between Patients with Abnormal and Normal Fluorescein Staining Scores in Groups 1 and 2
|Group||FSS||No. of Eyes (%)||Postoperative Lagophthalmos (mm)||P||Amount of Eyelid Elevation (mm)||P|
|1||Normal||6 (26%)||3.25 ± 0.42||.001a||3.92 ± 0.38||.001a|
|Abnormal||17 (74%)||5.24 ± 1.17||5.82 ± 3.91|
|2||Normal||30 (91%)||4.75 ± 1.19||.082||5.38 ± 1.14||.361|
|Abnormal||3 (9%)||5.67 ± 0.58||6.17 ± 1.15|