Professor Junwen Zeng, State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou 510060, CHINA, E-mail: firstname.lastname@example.org
Background: The aim of this retrospective study was to investigate the relationship between unilateral congenital ptosis in patients older than eight years and their refractive state and spherical equivalent refraction (SER).
Methods: The study involved a review of the clinical records of 85 patients admitted to the First Affiliated Hospital, Sun Yat-sen University between 1998 and 2010 with unilateral congenital ptosis. The average age was 16.83 years (nine to 27 years). The patients were classified into mild (27 cases), moderate (37 cases) or severe (21 cases) ptosis according to the degree of the droopy eyelid covering the cornea. The fellow eyes served as controls.
Results: In 85 eyes with unilateral ptosis, the frequency of myopia (SER of -0.50 D or more myopia) was significantly higher than in the fellow eye (47 versus 32, p = 0.031). The frequency of myopia in eyes with severe unilateral ptosis was significantly higher than in the fellow eyes (16 versus 7, p = 0.012), whereas there were no significant differences in patients with mild (15/27 versus 13/27, p = 0.79) or moderate (16/37 versus 12/37, p = 0.47) unilateral ptosis. Similarly, the SER was significantly more myopic in eyes with severe ptosis compared with the fellow eye (-1.37 D versus -0.85 D, p = 0.01), whereas no significant differences were found in patients with mild or moderate unilateral ptosis.
Conclusions: The results showed a higher frequency of myopia and more myopic SER in eyes with severe unilateral ptosis compared with the fellow eye. The myopia found in eyes with unilateral ptosis might be caused by a mechanism similar to that resulting in myopia among animals subjected to form deprivation. It is important to pay attention to possible refractive error in patients with unilateral ptosis. Surgical correction of unilateral ptosis at an early age is recommended.
Congenital ptosis is an eyelid disorder, often associated with amblyopia, refractive error, anisometropia and strabismus.1–3 With the establishment of models of form-deprivation myopia in primates,4 chickens,5 guinea pigs,6 tree shrews,7 fish8 and mice9 by suturing of the eyelids or covering the eyes with occluders, several authors have investigated whether a similar phenomenon might occur in humans.10–13 In 1979, O'Leary and Millodot10 reported that humans with unilateral ptosis were more myopic in the closed eye than in the normal eye. Hoyt and colleagues11 found axial myopia in eight out of 64 patients with complete neonatal eyelid closure. These studies suggest that neonatal eyelid closure may induce myopia in humans. Von Noorden and Lewis12 found that in two patients with complete unilateral ptosis through oculomotor palsy, the axial length of the affected eye was shorter than the fellow eye. Gusek-Schneider and Martus13 reported that the frequency of myopia was 15 per cent (spherical equivalent refraction (SER) negative) and the frequency of hyperopia 85 per cent (SER positive) in the ptotic eye of 68 patients with unilateral ptosis. A higher frequency of myopia was found in the ptotic eye than in the fellow eye (15 per cent versus 4.4 per cent). Most of their patients (57/68) were less than eight years old.
Few studies have focused on patients older than eight years with long-lasting congenital ptosis. In the current study, most patients with congenital ptosis were from a remote area of China and older than eight years. Moreover, earlier studies present only few data on SER. The purpose of the present study was to investigate the relationship between unilateral ptosis and refractive status in 85 patients older than eight years.
This study was approved by the Ethics Committee of Sun Yat-sen University in China and complied with the tenets of the Declaration of Helsinki for biomedical research involving human subjects. The pre-operative records of 176 consecutive inpatients with congenital ptosis, who had been surgically corrected between 1998 and 2010 at the First Affiliated Hospital of Sun Yat-sen University were reviewed for data such as age, gender, degree of ptosis, refractive error under cycloplegia, visual acuity, astigmatism and anisometropia. All patients had been examined and diagnosed by an experienced ophthalmologist. Medical records about clinical manifestations in ptotic patients were saved as a computer database.
The cases with bilateral ptosis, acquired ptosis or other ophthalmic or systemic disorders were excluded from the study. Inclusion criteria were:
1unilateral congenital ptosis
2no other ophthalmic or systemic disorders
3retinoscopic refraction under cycloplegia
4patients older than eight years.
Inclusion criteria were filled by 85 patients. The average age was 16.83 ± 4.66 years (SD), with a range nine to 27 years. The male/female ratio was 53/32.
Ptosis can be classified into three grades: mild, if the drooping upper lid is covering the cornea by one to two millimetres more than the normal eye; moderate, if the drooping upper lid is covering the cornea by three to four millimetres more than the normal eye and covering less than half of the pupil; and severe ptosis, if the drooping upper lid is covering the cornea by four millimetres more than the normal eye and covering more than half of the pupil and visual axis14 (Figure 1). Thus, the cases were classified into three groups (27 mild, 37 moderate and 21 severe ptosis) (Table 1). The fellow eyes of 85 unilateral ptosis cases served as a control.
Table 1. Age of patients in mild, moderate and severe unilateral congenital ptosis (nine to 27 years old)
18.81 ± 4.16
16.22 ± 5.01
15.33 ± 3.89
16.83 ± 4.66
Refractive error examination
Cycloplegic retinoscopy was performed using streak retinoscopy by two experienced refractionists following administration of 1% atropine sulphate ointment (on patients up to 10 years, once a day for three days) or administration of combined 0.5% phenylephrine hydrochloride and 0.5% tropicamide eye drops (on patients older than 10 years, administered three times with five-minute intervals). The droopy eyelid was lifted and fixed on the upper orbital rim to expose the pupil, avoid pressure on the eyeball and cornea and maintain the tear film. The eyelid was raised and lowered to wet the cornea from time to time.
Criteria and statistics
The refraction was calculated as SER. Myopia was defined as SER of at least -0.50 D and hyperopia as +2.00 D or more. Eyes with no myopia or hyperopia were classified as emmetropic (SER between -0.5 D and 2.00 D). Astigmatism was defined as 0.75 DC. Differences between paired eyes of 1.00 D in SER or more were designated as spherical anisometropia. In pairs of eyes with mixed refraction, the anisometropia was classified in the eye with the larger refractive error.15 Amblyopia was defined as visual acuity (VA) of 6/9 or less and greater than two Snellen lines of difference between the two eyes in the absence of any other ocular pathology.16
Data were expressed as the mean and standard deviation. Paired sample t-tests were used to compare SER between the ptotic and the fellow eyes. Chi-squared tests were used to compare the frequency of myopia, hyperopia, amblyopia and astigmatism between the ptotic and the fellow eyes. Statistical analysis of SER was performed using SPSS version 13.0 (Chicago, IL, USA). A p-value less than 0.05 was considered statistically significant.
Frequency of myopia and hyperopia
Forty-seven of the 85 unilateral ptotic eyes were found to be myopic. Among these, 41 (87.2 per cent) had low myopia (SER between -3.0 D and -0.5 D) and six (12.8 per cent) medium myopia (SER between -6.0 D and -3.0 D). Compared with the control fellow eyes, frequency of myopia was higher in the ptotic eyes (47/85 versus 32/85, χ2= 5.32, p = 0.031) (Figure 2D). When taking the severity of the ptosis into account, the frequency of myopia was found to be higher in eyes with severe unilateral ptosis than in the control fellow eyes (16/21 versus 7/21, χ2= 7.79, p = 0.012) (Figure 2C), whereas no significant differences were found in cases with mild (15/27 versus 13/27, χ2= 0.30, p = 0.79) (Figure 2A) or moderate ptosis (16/37 versus 12/37, χ2= 0.92, p = 0.47) (Figure 2B).
Among the 85 unilaterally ptotic eyes, two hyperopic eyes were found among patients at the ages of nine and 14 years. The frequency of hyperopia in ptotic eyes did not differ significantly from that of the fellow control eyes (two versus zero, p = 0.50, Fisher's exact test). Both of the hyperopic eyes with unilateral ptosis were amblyopic (Figure 2D).
Spherical equivalent refraction
There was no significant difference in SER between the 85 ptotic eyes and the control fellow eyes (-0.96 ± 1.44 D versus -0.78 ± 1.28 D, p = 0.051); however, the SER in eyes with severe ptosis was more myopic than in the control fellow eyes (-1.37 ± 1.47 D versus -0.85 ± 1.37 D, p = 0.001). In contrast, no significant difference between the ptotic and the fellow eyes occurred in cases with mild (-1.25 ± 1.54 D versus -1.2 ± 1.49 D, p = 0.75) or moderate unilateral ptosis (-0.51 ± 1.24 D versus -0.43 ± 0.95 D, p = 0.65) (Figure 3A).
The difference in SER between the ptotic eye and the fellow eye was also analysed. There was no difference in 41 of the 85 cases (48.2 per cent). The ptotic eyes were more myopic than the fellow eyes in 33 (38.8 per cent) cases and the ptotic eyes were more hyperopic than the fellow eyes in 11 (12.9 per cent) cases (Figure 3B).
In the 85 ptotic individuals, 15 were amblyopic. In 13 of these, the amblyopic eye was the ptotic eye and in two the amblyopic was the non-ptotic eye (13 versus 2, χ2= 9.12, p = 0.003, Chi-squared statistic). Of the 13 eyes that were both ptotic and amblyopic, ptosis was mild in two (7.4 per cent), moderate in four (10.8 per cent) and severe in seven (33.3 per cent). All of these 13 eyes were ametropic, with anisometropia or astigmatism (1.5 to 2.0 D) in some cases (Figure 4A).
Among the 85 patients with unilateral ptosis, the frequency of astigmatism was significantly higher among the ptotic eyes than the controls (24 versus 11, χ2= 5.69, p = 0.02, Chi-squared statistic) (Figure 4B). The astigmatic eyes included 20 (83.3 per cent) with myopic astigmatism and four (16.7 per cent) with hyperopic astigmatism. Among the 24 astigmatic eyes, four had mild ptosis, five had moderate ptosis and 15 had severe ptosis.
Myopic and hyperopic anisometropia
Of the 85 patients with unilateral ptosis, 19 (22.4 per cent) had myopic anisometropia and three (3.5 per cent) had hyperopic anisometropia. Among the 19 cases of myopic anisometropia, there were three cases of mild ptosis, nine with moderate ptosis and seven with severe ptosis. Among the three cases of hyperopic anisometropia, two had moderate ptosis and one had severe ptosis.
In the present study, we found that in eyes with unilateral ptosis, the frequency of myopia is higher and the refraction more myopic than in the fellow eye; however, when the cases of unilateral ptosis were divided into mild, moderate and severe, the differences were only significant in cases with severe ptosis. The myopia found in ptotic eyes was of mild or medium degree and no cases of high myopia were found. Amblyopia and astigmatism occurred with a higher frequency in ptotic eyes compared with fellow eyes and the frequency of amblyopia increased with increasing severity of ptosis.
The present study supports existing data showing a higher frequency of myopia in the ptotic eye than that in the fellow eye. The frequency of myopia (47/85, 55.3 per cent) was higher than that of hyperopia (2/85, 2.4 per cent) in the ptotic eye. This result differs from that reported by Gusek-Schneider and Martus,13 who found a lower frequency of myopia (10/68, 15 per cent) than hyperopia (58/68, 85 per cent) in the unilaterally ptotic eyes. The different criteria of hyperopia (SER above zero in their study but 2.00 D or more in the present study) and myopia (SER less than zero in their study and -0.5 D or more myopia in the present study) would define more of their cases as hyperopes.
Another confounding factor could be the ethnicity of our patients, as the prevalence of myopia in Chinese is known to be higher than in Caucasians.17 The increase in myopia might also be related to age and the incidence of developmental myopia, because the prevalence of myopia in teenagers is much higher than that in pre-school children.16,18 Nevertheless, the big difference in the frequency of myopia between the two studies might be related to the higher age of the patients in the present study (16.83 years compared with 4.7 years). Thus, the patients in the present study had been subjected to partial or total covering of the pupil for a longer time. An increasing prevalence of myopia with increasing age would be consistent with the hypothesis that long-standing congenital ptosis produces myopia by interfering with image formation on the retina. All cases of myopia were mild or medium. This is different from several earlier and some smaller studies9,10 showing that high myopia occurs in patients with congenital ptosis or neonatal eyelid closure; however, congenital ptosis is rarely complete and a narrow palpebral fissure is most often preserved, even in severe cases. A narrow palpebral fissure allows sufficient light stimulation of the retina and the formation of a clear retinal image. Therefore, the visual consequences of unilateral congenital ptosis differ from those found in complete ptosis or in animal models of myopia using suturing of the eyelid or translucent occluders, where no clear image on the retina is formed.
The amount of light that reaches the retina will depend on the severity of the ptosis, with less light reaching the retina in severe cases. The present study showed that the frequency of myopia was higher in severe ptosis and that SER was more myopic in eyes with severe unilateral ptosis compared with the fellow eye, whereas the differences were not significant in cases with mild of moderate unilateral ptosis.
The prevalence of amblyopia (VA of 6/12 or worse in the better eye) is 0.46 per cent among children aged between 13 and 17 years in the south of China,19 whereas among patients with congenital ptosis the incidence of amblyopia can be as high as 6.9 per cent (VA of 6/12 or worse),20 17 per cent (VA of 6/12 or worse),2 20 per cent (VA of 6/9 or worse)16 and 70 per cent (VA of 6/6 or worse).21 In the current study, the frequency of amblyopia in the ptotic eye was 15.3 per cent (VA of 6/9 or less). The different criteria of amblyopia might explain the differences. According to our data, in mild or moderate ptosis, uncorrected refractive error from an early age and anisometropia were the only reasons resulting in amblyopia, because retinal images were not impeded by the droopy eyelid. While in severe ptosis, the reasons for amblyopia were complicated, including uncorrected refractive error from an early age, anisometropia and stimulus deprivation because of the droopy eyelid interfering with vision.
The present study is retrospective and therefore might be biased regarding recorded information. There were no data on axial length or corneal topography. A prospective, multi-centre study including different age groups is being planned. The findings of the present study suggest that more attention should be paid to early surgical correction of unilateral ptosis to avoid the development of myopia and amblyopia.
GRANTS AND FINANCIAL ASSISTANCE
This study was supported by grant 201102A211005 from Key Projects in Medicine Health Science and Technology from Guangzhou, China and the Fundamental Research Funds of the State Key Laboratory, China.