The Associations of Floppy Eyelid

Syndrome: A Case Control Study

Daniel G. Ezra, MA, MRCOphth,1,2,3 Michèle Beaconsfield, FRCS, FRCOphth,2 Mano Sira, MRCOphth,4
Catey Bunce, DSc,2 Richard Wormald, FRCS, FRCOphth,2 Richard Collin, FRCS, FRCOphth2

Objective: To describe the demographic features of a large series of patients with floppy eyelid syndrome
(FES) and to investigate the associations of the condition with keratoconus, obstructive sleep apnea-hypopnea
syndrome (OSAHS), and a variety of upper and lower eyelid features.
Design: Case control study.
Participants: The test group comprised 102 patients with FES. A control group of 102 patients were
recruited from a diabetic retinopathy clinic and matched on a 1:1 basis on age, gender, and body mass index
(BMI).
Methods: A full medical and ophthalmic history was taken. Patients also underwent a full ocular examina-
tion, including an assessment of upper and lower lid laxity and upper lid levator function. Keratoconus grading
was made using the Oculus Instruments Pentacam imaging system (Oculus Optikgerate GmbH, Wetzlar,
Germany). Patients were screened for OSAHS using the Epworth daytime somnolence score. Matched statistical
analysis of dichotomous data was made using Mantel–Haenszel methods for odds ratios and McNemar’s test.
Analysis of continuous data was performed using a matched t test and tests for symmetry of larger tables were
made using the McNemar–Bowker test.
Main Outcome Measures: The significance of association of FES with keratoconus, OSAHS, smoking
history, medial and lateral canthal laxity of the upper and lower lids, levator function, lash ptosis, and
dermatochalasis.
Results: Significant associations were found between FES and OSAHS (P ⫽ 0.0008), keratoconus
(P⬍0.0001), lash ptosis (P⬍0.0001), dermatochalasis (P ⫽ 0.02), upper lid medial canthal laxity (P ⫽ 0.02), upper
lid distraction (P ⫽ 0.001), palpebral aperture (P ⫽ 0.004), and levator function (P ⫽ 0.005).
Conclusions: Floppy eyelid syndrome seems to be a condition strongly associated with OSAHS and
keratoconus. As well as providing a platform for an etiologic hypothesis for the condition, these findings should
also encourage clinicians to be aware of these associations and to direct further treatment.
Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed
in this article. Ophthalmology 2010;117:831– 838 © 2010 by the American Academy of Ophthalmology.

Floppy eyelid syndrome (FES) was first described by Cul-
bertson and Ostler1 in 1981. This condition was character-
ized by very elastic upper lids that became easily distorted
and everted with minimal lateral traction. The tarsal plate
was found to be pliant and “rubbery,” with a loss of
intrinsic rigidity allowing it to be folded over itself with
ease. A marked papillary conjunctivitis was also found to
be present.
There have been a growing number of reports of the con-
dition highlighting associations with an ever increasing range
of ocular and systemic conditions. Although some of these
reported associations have been sporadic, others have been
described more consistently. These associations have included
eyelid disorders, such as dermatochalasis,1–3 blepharitis and
tear film abnormalities,4 –11 blepharoptosis,6,12–16 and upper lid
lash ptosis.3,4,6,15 A range of corneal pathologies, including
superficial punctate keratopaty,9 scarring and neovascularisa-
tion,1,17–21 thinning,22,23 microbial keratitis,9,15,19,21,24,25 and
even corneal perforation,9,18,20,23 have also been reported.
Although all of these corneal associations seem to be the
result of chronic exposure keratopathy, one intriguing asso-
ciation of the condition is keratoconus. This was first noted
in a brief meeting communication from the Ocular Micro-
biology and Immunology Group meeting in 1981 in Atlanta,
Georgia (Am J Ophthalmol 1982;93:245–9), and there have
been several subsequent reports.26 –28
One of the most consistently reported associations is that
of obstructive sleep apnea-hypopnea syndrome (OSAHS).
This was first noted incidentally,5,29 but Woog,12 in a small
series of 3 patients, was the first to specifically identify this
as an important association. This proposed link was further
reinforced in a series by McNab,6 in which 26 of 27 patients
were found to have OSAHS.30 However, he also found that
when assessing a random cohort of 20 patients from a sleep
studies clinic, only 1 patient had FES. Many other series
have reported the association of FES with OSAHS.3,4,31
Other reports of the prevalence of FES in the OSAHS
population have varied from 2%32,33 to 32%.34
The assumption of a clinical association between FES
and OSAHS on the basis of numerous case reports must be
questioned. The independent association of both FES and
OSAHS with obesity and the male gender raises concern

© 2010 by the American Academy of Ophthalmology ISSN 0161-6420/10/$–see front matter 831
Published by Elsevier Inc. doi:10.1016/j.ophtha.2009.09.029
 Ophthalmology Volume 117, Number 4, April 2010
that obesity per se may be a confounding factor. The asso-
ciation between FES and OSAHS is particularly important
to determine because OSAHS has been suggested as a
possible cause for FES, and this hypothesis is strengthened
by a report of FES resolving simply with the use of a
continuous positive airway pressure mask.35 Furthermore,
elucidating this association may provide awareness for the
ophthalmologist to facilitate further management. This is
important because in addition to daytime somnolence, OSAHS
carries significant morbidity with a higher reported inci-
dence of cardiovascular and cerebral sequelae.36
A case control study was chosen to allow for investiga-
tion of associations of FES while controlling for possible
confounding factors such as body mass index (BMI), age,
and gender. The primary aims of this study are to describe
the characteristics of a large group of patients with FES and
to determine the relationships of FES with keratoconus and
OSAHS. The secondary aims are to identify whether a
variety of upper and lower lid features are associated with
FES.
Materials and Methods
Local and regional ethics committee approval was granted to this
study. All patients attending Moorfields Eye hospital with a diag-
nosis of FES made by an oculoplastics consultant over a 13-year
period since 1995 were identified from surgical and computer-
based log records and through the tracking of histopathology
samples. Patient tracing was performed through the National
Health Service strategic tracing service.37 The National Health
Service strategic tracing service provides a secured access national
database of patient tracing information, including the identity and
location of their current registered general practitioner. All patient
general practitioners were contacted to confirm the patients’ con-
tact details, and a letter of invitation and consent form for partic-
ipation were sent. All patients who returned a consent form were
invited to attend for examination. There were no exclusion criteria.
A control group was identified from a series of diabetic reti-
nopathy clinics at Moorfields Eye Hospital. All patients attending
these clinics were screened for BMI, age, and gender. Control
subjects were matched to the test group on the basis of these 3
characteristics. The inclusion criteria for the control group were
potentially all patients attending a variety of medical retina clinics
and who matched the appropriate demographic characteristics of
individuals in the test group. Any patients who underwent any
surgical procedures to the eyelid or who had suffered from any
ocular abnormality that may affect eyelid function (e.g., phthisis
bulbi) were excluded from the study, as were any control group
patients found to have FES.
All subjects in both groups underwent an identical assessment
consisting of a full medical and ophthalmic history that included
details on the patients’ preferred sleep side. Patients also under-
went a full ocular examination, and this included the following
parameters.
Assessment of the Lower Lid
The lower lid was assessed in terms of medial and lateral canthal
laxity and horizontal laxity. Medial canthal laxity was graded
according to a scale developed by Olver et al,38 measuring medial
canthal laxity from 0 to 6. The grading scores were dichotomized
for the purpose of analysis to those scoring ⬍2 and those scoring
ⱖ2. Lateral canthal laxity was defined clinically as either present
832
or absent. A standardized grading system for assessing lower lid
laxity39 was used and dichotomized for the purposes of analysis
into those scoring grade 0 and those scoring more than grade 0.
Assessment of the Upper Lid
Assessment of the upper lid medial and lateral canthal laxity was
made in accordance with the same methods used for the lower lid
outlined above. Horizontal distraction from the globe was assessed
as described by Iyengar and Khan40 and measured in millimeters.
Upper lid levator function and vertical palpebral aperture length
were measured in millimeters, a technique demonstrated to have
good reproducibility and repeatability.41 Only patients who had
not undergone upper lid surgery were entered into this analysis
because FES corrective surgery involves significant structural
changes to the upper lid and may bias the comparison.
Obstructive Sleep Apnea-Hypopnea Syndrome
Assessment
Different grading systems for OSAHS exist, and important mea-
sures of the severity of the condition are the respiratory distur-
bance index and apnea-hypopnea index. Because respiratory dis-
turbance index and apnea-hypopnea index information was not
available in most cases, the presence of OSAHS was defined in a
dichotomous fashion on the basis of whether or not the patient was
recommended continuous positive airway pressure respiratory sup-
port at night. Patients were screened for undetected OSAHS using
the Epworth daytime somnolence score.42 The Epworth daytime
somnolence score is widely used as a simple screening test for
OSAHS at a cutoff of 10 points.43,44 Any patient scoring more
than 10 was referred to a respiratory physician for sleep studies.
Keratoconus Assessment
Objective assessment of keratoconus was made using the Oculus
Instruments Pentacam Shiempflug photographic camera system
(Oculus Optikgerate GmbH, Wetzlar, Germany). This system mea-
sures several aspects of the corneal morphology and provides a
standardized score of keratoconus from absent to grades KC1 to
KC445 in accordance with the Amsler–Krumeich keratoconus clas-
sification system, which is widely used in the objective assessment
of keratoconus.46 For the purposes of analysis, the scale was
dichotomized with a demarcation at grade 1 and above.
Statistical Analysis
Matched analysis was performed on the basis of patient matching
rather than by matching eyes affected. This is as a result of the need
for consistency between the matching process (patients were
matched rather than eyes) and the subsequent analysis. Where
patients had bilateral disease, 1 eye was chosen at random for the
analysis. A computer-generated random sequence was created for
this purpose. In addition, where possible, a sensitivity analysis
excluding all cases of bilateral disease was performed to control
for the possible effect of FES being more severe in patients where
bilateral disease was present.
Dichotomous variables between the groups were compared
using Mantel–Haenszel methods for matched concordant and dis-
cordant pairs. Odds ratios and confidence intervals were calcu-
lated, and McNemar’s test was used as a test for significance.
Continuous variables were compared using a matched t test for
significance. Co-laterality of FES with keratoconus and of FES
with preferred sleeping position was explored using McNemar–
Bowker matching tests for larger 3⫻3 datasets.47
 Ezra et al 䡠 Floppy Eyelid Syndrome

Table 1. Demographics of Patients Recruited Table 2. Frequency of Associated Medical Conditions

N Gender M:F Age, yrs BMI Medical Problems Frequency

FES group 102 85:17 50.4 (13.6) 37.8 (8.7) Hypertension 42 (41.2%)
Control group 102 85:17 53.6 (14.1) 37.0 (8.6) NIDDM 15 (14.7%)
Ischemic heart disease 15 (14.7%)
Hypercholesterolemia 11 (10.8%)
BMI ⫽ body mass index; FES ⫽ floppy eyelid syndrome.
Osteoarthritis 6 (5.9%)
Asthma 6 (5.9%)
Gastroesophageal reflux disease 4 (3.9%)
Results Chronic renal failure 4 (3.9%)
Schizophrenia 4 (3.9%)
Test Group Hypothyroid 4 (3.9%)
Epilepsy 3 (3.9%)
In total, 168 patients with FES were identified by the tracing methods Eczema 2 (2.0%)
outlined. At the time of recruitment, 14 patients were deceased and the IDDM 1 (1.0%)
remaining 154 patients were invited to participate in the study. Forty- Bowel carcinoma 1 (1.0%)
seven patients declined to participate or did not respond to an invita- Renal stones 1 (1.0%)
tion, and a further 5 patients were excluded from the analysis because Cerebral palsy 1 (1.0%)
their notes were unobtainable; therefore, the total number of patients Gout 1 (1.0%)
recruited was 102. Seventy-one patients were identified as having Depression 1 (1.0%)
undergone upper lid-tightening surgery. Thirty-one patients had not Stomach stapling 1 (1.0%)
Ankylosing spondylitis 1 (1.0%)
undergone surgery, and of these, 29 had declined surgery and 2 were
Acute lymphoblastic leukemia 1 (1.0%)
on the waiting list for surgical treatment.
Testicular teratoma 1 (1.0%)
Ewing sarcoma 1 (1.0%)
Control Group Rheumatoid arthritis 1 (1.0%)
Klinefelter’s 1 (1.0%)
An equal number of patients were selected as controls from a Sarcoidosis 1 (1.0%)
diabetic retinopathy clinic. Patients attending the clinic were Transient ischemic attacks 1 (1.0%)
screened for appropriate matching characteristics and invited to Spina bifida 1 (1.0%)
participate. The controls were matched to a ratio of 1:1 with Prostate carcinoma 1 (1.0%)
individuals in the test group on the basis of BMI, age, and gender. Spinal tuberculosis 1 (1.0%)
The demographic data of these groups are summarized in Table 1. Benign prostatic hypertrophy 1 (1.0%)
Patients had a mean age of approximately 50 years and a mean
BMI in the high 30s. This is consistent with the phenotype de- IDDM ⫽ insulin-dependent diabetes mellitus; NIDDM ⫽ non–insulin-
scribed by the previous studies that we have outlined. A scatter dependent diabetes mellitus.
plot demonstrating the distribution of age, gender, and weight is
demonstrated in Figure 1.
The median time from the initial presentation to an ophthal-
mologist until the correct diagnosis was made was 17 months A frequency table of all associated medical conditions in the
(range 0 –192 months). Several patients had waited more than 15 test group is included in Table 2. Table 2 demonstrates that the
years for a correct diagnosis, highlighting the nonspecific nature of associations of FES seem to be dominated by conditions that are
these symptoms and a poor awareness of the condition. associated with obesity, such as non–insulin-dependent diabetes
mellitus and hypertension.
All 102 patients of the test group were used in the matching
analysis for keratoconus, OSAHS, smoking history, and lower lid
factor parameters. The results of this analysis are summarized in
Table 3. No significant association between FES and lower lid
laxity was identified, indicating that this condition is confined to
changes of the upper lid. In addition, no association with an
increased history of smoking, which can lead to extracellular
matrix changes, was identified.
Despite correcting for weight, a strong association with
OSAHS was identified (odds ratio ⫽ 12.5). In total, 32 cases of
OSAHS were identified in the FES group; 28 patients had an
established diagnosis of OSAHS and 5 patients were referred to a
respiratory physician for sleep studies on the basis of a high
Epworth score screening test. Four patients were positive for
OSAHS, making 32 positive cases in total. Nine patients in the
control group were positive for OSAHS. Six patients had an estab-
lished diagnosis and 3 patients were referred to a sleep physician on
the basis of a high Epworth score, and all were found to be positive
for the condition, making a total of 9 cases. A statistically significant
Figure 1. Distribution of age and BMI for both men and women in the association between OSAHS and FES was also found in a sensitivity
FES group. This scatterplot demonstrates no demarcation of any subgroup analysis in which all bilateral FES cases were excluded and only
on the basis of age, BMI, or gender. BMI ⫽ body mass index. patients with unilateral disease were analyzed. The observation of

833
 Ophthalmology Volume 117, Number 4, April 2010

Table 3. Whole Group Comparisons of Obstructive Sleep Apnea-Hypopnea Syndrome, Keratoconus,

Smoking History, and Lower Lid Parameters

Test Data Matched Data Statistic

All patients with FES 102 102
Unilateral cases 61 —
Bilateral cases 41 —
Total No. of eyes affected 144 144
Age/yrs 50.4 (13.6) 53.6 (14.1)
Gender/M:F 85:17 85:17
BMI 37.8 (8.7) 37.0 (8.6)
Median time until diagnosis/mos 17 (range 0–192) —
OSAHS
All patients 32 9 OR ⫽ 12.5* 95% CI, 3.1–108.9 P⬍0.0001†
Unilateral cases only 23 6 OR ⫽ 8.7* 95% CI, 2.7–44.7 P ⫽ 0.0008†
Keratoconus
All patients 61 6 OR ⫽ 19.3* 95% CI, 6.3–96.4 P⬍0.0001†
Unilateral cases only 36 2 OR ⫽ 35.0* 95% CI, 5.9–1421.4 P⬍0.0001
Smoking/pack yrs
Unilateral cases 12.5 (16.0) 9.2 (16.1) P ⫽ 0.72‡
All patients 12.5 (17.4) 12.2 (16.8) P ⫽ 0.98‡
Lower lid ectropion
All patients 7 4 OR ⫽ 2.0* 95% CI, 0.43–12.4 P ⫽ 0.5†
Unilateral cases 2 2 OR ⫽ 1.0* 95% CI, 0.07–13.8 P ⫽ 0.61†
Lower lid LC laxity
All patients 6 5 OR ⫽ 1.25* 95% CI, 0.27–6.3 P ⫽ 1.0†
Unilateral cases 3 2 OR ⫽ 1.5* 95% CI, 0.17–18.0 P ⫽ 1†
Lower lid MC laxity
All patients 6 6 OR ⫽ 1.0* 95% CI, 0.23–4.3 P ⫽ 0.75†
Unilateral cases 2 3 OR ⫽ 0.66* 95% CI, 0.06–5.8 P ⫽ 1.0†
Microbial keratitis
All patients 4 (1 PK) 0 —
Unilateral cases 1 0 —

BMI ⫽ body mass index; CI ⫽ confidence interval; FES ⫽ floppy eyelid syndrome; LC ⫽ lateral canthus; MC ⫽
medial canthus; OR ⫽ odds ratio; OSAHS ⫽ obstructive sleep apnea-hypopnea syndrome; PK ⫽ penetrating
keratoplasty.
*Odds ratios using Mantel–Haenszel methods for concordant/discordant pairs matched data.
†
McNemar’s test.
‡
Paired t test.

localization of FES to sleep laterality was further investigated using a
McNemar–Bowker test that demonstrated a strong association of FES
to the preferred sleeping side, which was statistically significant
(McNemar–Bowker, P ⫽ 0.005) (Table 4).
A strong association with keratoconus was also identified with
an odds ratio of 19.3. A sensitivity analysis excluding patients with
bilateral disease did not affect the significance of the association,
and the strength of the significance was greater for unilateral
disease only. This important relationship was further investigated
to explore the relationship between laterality of FES and kerato-
conus. These findings are demonstrated in Table 5. Although
co-laterality of keratoconus and FES is suggested by Table 5, this
relationship was not found to be statistically significant (McNemar–
Bowker, P ⫽ 0.849).

Table 4. Relationship between Sleep Laterality and Laterality

of Floppy Eyelid Syndrome Table 5. Relationship between Laterality of Floppy Eyelid
Syndrome and Keratoconus
Sleep Side
Keratoconus Laterality
Alternating Left Right
Bilateral Left Right Not Affected
FES laterality
Bilateral 37 1 3 FES laterality
Left 8 17 2 Bilateral 12 8 4 17
Right 14 1 19 Left 7 10 2 8
Right 6 1 11 16
FES ⫽ floppy eyelid syndrome.
Each cell refers to the number of patients with matching characteristics to FES ⫽ floppy eyelid syndrome.
the corresponding row and column category. This table demonstrates that No association between the laterality of keratoconus and FES was iden-
FES laterality strongly correlates to the sleep side. McNemar–Bowker test tified. McNemar–Bowker test P ⫽ 0.849. Patients not affected by kerato-
for symmetry along the diagonal of the matrix was calculated at P⫽0.005. conus were excluded from the analysis.

834
 Ezra et al 䡠 Floppy Eyelid Syndrome

Table 6. Upper Lid Comparisons

Test Data Matched Data Statistic

Patients who did not undergo surgery 31 31
Unilateral cases 16 —
Bilateral cases 15 —
Total No. of eyes affected 46 —
Age/yrs 51.0 (12.0) 51.7 (11.6)
Gender/M:F 25:6 25:6
BMI 35.6 (7.6) 35.1 (7.5)
Lash ptosis
Unilateral cases 12 0 OR ⫽ 0, P⬍0.002b
All cases 33 1 OR ⫽ 0, P⬍0.0001b
Dermatochalasis
Unilateral cases 11 4 OR ⫽ 3.3, CI, 0.86–18.9; P ⫽ 0.1b
All cases 20 9 OR ⫽ 3.75, CI, 1.2–15.5 P ⫽ 0.02b
Upper lid LC laxity
Unilateral cases 1 0 —
All cases 6 1 OR ⫽ 0, P ⫽ 0.07b
Upper lid MC laxity
Unilateral cases 2 0 OR ⫽ 0, P ⫽ 0.47b
All cases 7 0 OR ⫽ 0, P ⫽ 0.02b
Upper lid distraction/mm (SD)
Unilateral cases 18.1 (12) 5.6 (2.1) P⬍0.001a
All cases 13.7 (6.9) 5.8 (2.2) P⬍0.001a
Palpebral aperture /mm
Unilateral cases 8.7 (1.9) 10.1 (1.7) P ⫽ 0.003a
All cases 8.7 (1.8) 10.2 (1.5) P ⫽ 0.004a
Levator function/mm
Unilateral cases 16.9 (1.8) 15.0 (2.6) P ⫽ 0.001a
All cases 16.9 (1.4) 15.4 (2.2) P ⫽ 0.005a

BMI ⫽ body mass index; CI ⫽ confidence interval; LC ⫽ lateral canthus; MC ⫽ medial canthus; OR ⫽ odds ratio;
SD ⫽ standard deviation.
a
Paired t test.
b
McNemar’s test. Odds ratios based on Mantel–Haenszel methods of concordant/discordant pairs for matched data.

A separate subgroup analysis was performed to investigate
upper lid features of the condition. This analysis used only
patients with FES who had not undergone surgery for the
treatment of the condition. Thirty-one patients with FES were
analyzed with the corresponding 31 matched patients from the
control group. The demographics of these groups and the results of
comparisons are summarized in Table 6.
Comparisons between test and control groups were made on a
variety of upper lid features, including lash ptosis, dermatochala-
sis, lateral and medial canthal laxity, upper lid distraction, palpe-
bral aperture, and levator function. Significant associations with
lash ptosis, medial canthal laxity, and upper lid distraction were
identified. There was also an association with blepharoptosis and a
mild reduction in levator function in patients with FES. Lateral
canthal laxity was not observed to be significant, although the P
value did approach significance at 0.07. Finally, a sensitivity
analysis excluding all bilateral cases was also determined to ac-
count for the possible bias that may be a result of bilateral cases
being more severe. With the exception of dermatochalasis and
medial canthal laxity, bilaterality of the condition was not found to
alter the significance of any of the other factors investigated.
Discussion
Floppy eyelid syndrome has been associated with a wide
variety of ocular and systemic conditions. Two of the most
consistently described have been keratoconus and OSAHS.
In a comprehensive assessment of corneal disorders associ-
ated with FES, Culbertson and Tseng9 described a series of
60 patients, in whom keratoconus was identified in 11. They
also determined that keratoconus was always on the pre-
ferred sleeping side. However, the definitions of FES for
this study were not given and the definition of keratoconus
was ambiguous. They described their definition of kerato-
conus as being essentially clinical, or in their words, “clin-
ically evident by standard techniques such as slit-lamp
examination, retinoscopy and keratometry in 6 patients.”
Computerized videophotokeratoscopy was performed ran-
domly on 7 patients with no clinical evidence of keratoco-
nus. By using this technique, they defined keratoconus as
localized steepening of ⬎2.0 diopters. This led to the iden-
tification of a further 5 patients with keratoconus. The
identification of high proportions of patients with FES with
keratoconus in a small random sample size indicates that the
incidence of keratoconus may have been significantly un-
derestimated in this study.9
Similar limitations were observed in a series of 17 pa-
tients by Lee et al,48 who also attempted to characterize
corneal findings in FES. They reported that 5 of 17 patients
in the series had keratoconus. However, criteria for the
definition of keratoconus and the methods used to evaluate

835
 Ophthalmology Volume 117, Number 4, April 2010
this were again unclear. It seems that a mixture of keratom-
etry and videophotokeratoscopy was used and that these
investigations were confined to the small number of patients
with clinically detectable keratoconus. This again suggests
that the incidence of the keratoconus, particularly in a
subclinical form, may be underestimated.
A wide variety of systemic associations with FES have
been proposed, with obesity being the most widely reported
and affecting the majority of patients. Other systemic con-
ditions have also been suggested as associations, although
these reports are sporadic. These include hyperglycin-
emia,49 schizophrenia,3 pachydermoperiostosis,10 eye rub-
bing,15 or conditions associated with eye rubbing, such as
psoriasis,24 cocaine use,25 severe learning difficulties,4,50 –52
epibulbar nodular fasciitis,53 and congenital cataracts facial
dysmorphism neuropathy syndrome.
Other associations, such as hypertension, diabetes,6,9 and
increased serum leptins,31 have all been described, but their
known close links with obesity and the failure to control for
this means that weight is likely to be a confounding factor.
The demographics of this series confirm that although
FES is more common in obese men in their sixth decade,
patients with a significant range of ages, BMI, and gender
are affected. Figure 1 demonstrates that there was no dis-
cernable subgroup or cluster on the basis of age, gender, or
BMI. This would not support the proposition of a distinct
entity known as “lax eyelid syndrome” based on similar
clinical findings, but with a normal BMI.7,54,55 Rather, there
is a continuum of different ages and weight distributions.
These findings confirm that OSAHS is significantly as-
sociated with FES even when confounding factors such as
BMI are controlled for. The definition for OSAHS used in
this study was based on continuous positive airway pressure
use, but this may well have underestimated the incidence of
OSAHS in the test and control groups. In addition, a strong
association between keratoconus and FES has also been
determined. Other upper lid abnormalities, such as lash
ptosis, blepharoptosis, upper lid distraction, and dermatoch-
alasis, have also been found to be associated with FES.
Although not typical of FES, concurrent lower lid laxity has
also been observed.2,6,16,22,29,54 However, this association
was not confirmed in this study and may warrant further
investigation.
The close relationship with OSAHS and the co-laterality
of the FES and sleep side would indicate that sleep itself
may be an important etiologic factor. Several mechanisms
have been proposed for this, including reperfusion isch-
emia9 and repeated mechanical trauma leading to matrix
metalloproteinase expression and connective tissue re-
modeling.21
The mechanical stress theory is supported by the close
association with keratoconus, a condition well described as
being related to eye rubbing, which can lead to extracellular
matrix ectasia.56 The hypothesis that eye rubbing may be
the causative factor in keratoconus and possibly also FES
has been highlighted.28,57–59 The strengths of these associ-
ations may lend support to this theory. In addition, the
interrelationship among sleep laterality, OSAHS, and FES
is strongly suggestive of an etiologic role for OSAHS. One
possible etiologic mechanism may lie in changes in CNS
836
arousability in OSAHS. A decrease in cortical arousability
in reflex responses to noxious stimuli, such as hypoxia, is
known to be a feature of OSAHS.60 These decreased cen-
tral reflexes may allow these patients to tolerate extreme
mechanical stress to the upper lid during sleep, which
would normally cause an unaffected individual to wake
and alter their sleep position. The exposure of the upper
eyelid to recurrent supraphysiologic mechanical stress
may be sufficient to induce subsequent extracellular ma-
trix remodeling.21
The strength of the association between FES and both
OSAHS and keratoconus is also of clinical importance. It is
essential that OSAHS is recognized due to its strong asso-
ciation with cerebrovascular morbidity.36 Other neurologic
conditions, such as epilepsy61 and psychiatric disorders, are
also strongly associated with OSAHS.62 The association
with keratoconus is also of clinical importance because the
presence of keratoconus should be considered as a refractive
cause for reduction in vision rather than ocular surface
disturbances commonly seen in FES.
This study also has limitations. This case control study
sample size was not based on a power calculation. This
means that findings of nonsignificance must be taken with
caution. Furthermore, the selection of the control group may
have lent unwanted bias to the comparison as there is
evidence of a weak independent association of diabetes with
OSAHS,63 although this is inconclusive.64 However, if this
association was warranted, it would have caused this study
to underestimate the association between FES and OSAHS.
In conclusion, floppy eyelid syndrome is a condition
strongly associated with OSAHS and keratoconus. As well
as providing a platform for an etiologic hypothesis for the
condition, these findings should encourage clinicians to be
aware of these associations and to direct further treatment.
Acknowledgment.We would like to thank Dr. Cathy Egan for
supporting patient recruitment from diabetic retinopathy clinics.
References
1. Culbertson WW, Ostler HB. The floppy eyelid syndrome.
Am J Ophthalmol 1981;92:568 –75.
2. Dutton JJ. Surgical management of floppy eyelid syndrome.
Am J Ophthalmol 1985;99:557– 60.
3. Langford JD, Linberg JV. A new physical finding in floppy
eyelid syndrome. Ophthalmology 1998;105:165–9.
4. Valenzuela AA, Sullivan TJ. Medial upper eyelid shortening
to correct medial eyelid laxity in floppy eyelid syndrome: a
new surgical approach. Ophthal Plast Reconstr Surg 2005;21:
259 – 63.
5. Gonnering RS, Sonneland PR. Meibomian gland dysfunction
in floppy eyelid syndrome. Ophthal Plast Reconstr Surg 1987;
3:99 –103.
6. McNab AA. Floppy eyelid syndrome and obstructive sleep
apnea. Ophthal Plast Reconstr Surg 1997;13:98 –114.
7. van den Bosch WA, Lemij HG. The lax eyelid syndrome. Br J
Ophthalmol 1994;78:666 –70.
8. van Nouhuys HM, van den Bosch WA, Lemij HG, Mooy CM.
Floppy eyelid syndrome associated with Demodex brevis.
Orbit 1994;13:125–9.
9. Culbertson WW, Tseng SC. Corneal disorders in floppy eyelid
syndrome. Cornea 1994;13:33– 42.
 Ezra et al 䡠 Floppy Eyelid Syndrome
10. Downes RN, Mininni F, Collin JR, et al. Floppy eyelid syn-
drome in pachydermoperiostosis. Orbit 1989;8:93–9.
11. Liu DT, Di Pascuale MA, Sawai J, et al. Tear film dynamics
in floppy eyelid syndrome. Invest Ophthalmol Vis Sci 2005;
46:1188 –94.
12. Woog JJ. Obstructive sleep apnea and the floppy eyelid syn-
drome. Am J Ophthalmol 1990;110:314 –5.
13. Schwartz LK, Gelender H, Forster RK. Chronic conjunctivitis
associated with ‘floppy eyelids’. Arch Ophthalmol 1983;101:
1884 – 8.
14. Da Pra A, Brunzini M, Croxatto JO, Nano H. Floppy eyelid
syndrome: 5 cases [in French]. Ophtalmologie 1989;3:314.
15. Netland PA, Sugrue SP, Albert DM, Shore JW. Histopatho-
logic features of the floppy eyelid syndrome: involvement of
tarsal elastin. Ophthalmology 1994;101:174 – 81.
16. Karesh JW, Nirankari VS, Hameroff SB. Eyelid imbrication:
an unrecognized cause of chronic ocular irritation. Ophthal-
mology 1993;100:883–9.
17. Goldberg R, Seiff S, McFarland J, et al. Floppy eyelid syn-
drome and blepharochalasis. Am J Ophthalmol 1986;102:
376 – 81.
18. Imbert P, Williamson W, Leger F, et al. Bilateral corneal
neovascularization and floppy eyelid syndrome: a case report
[in French]. J Fr Ophtalmol 1990;13:223–5.
19. Bouchard CS. Lateral tarsorrhaphy for a noncompliant patient
with floppy eyelid syndrome [letter]. Am J Ophthalmol 1992;
114:367–9.
20. Mortemousque B, Williamson W, Poirier L, et al. Floppy
eyelid syndrome: apropos of 4 cases and review of the liter-
ature [in French]. J Fr Ophtalmol 1995;18:542–7.
21. Schlotzer-Schrehardt U, Stojkovic M, Hofmann-Rummelt C,
et al. The pathogenesis of floppy eyelid syndrome: involve-
ment of matrix metalloproteinases in elastic fiber degradation.
Ophthalmology 2005;112:694 –704.
22. Moore MB, Harrington J, McCulley JP. Floppy eyelid
syndrome: management including surgery. Ophthalmology
1986;93:184 – 8.
23. Rossiter JD, Ellingham R, Hakin KN, Twomey JM. Corneal
melt and perforation secondary to floppy eyelid syndrome in
the presence of rheumatoid arthritis [letter]. Br J Ophthalmol
2002;86:483.
24. Fleishman JA, Hoffman RO. An unusual case of floppy eyelid
syndrome. Ann Ophthalmol 1988;20:468 –9.
25. Reifler DM. Floppy eyelids and crack eye syndrome [letter].
Ophthalmology 1993;100:975.
26. Parunovic A. Floppy eyelid syndrome. Br J Ophthalmol 1983;
67:264 – 6.
27. Parunovic A, Ilic B. Floppy eyelid syndrome associated with
keratotorus. Br J Ophthalmol 1988;72:634 –5.
28. Donnenfeld ED, Perry HD, Gibralter RP, et al. Keratoconus
associated with floppy eyelid syndrome. Ophthalmology
1991;98:1674 – 8.
29. Goldberg RA, Coden DJ, Hornblass A, Mitchell JP. Floppy
eyelid syndrome associated with marked lower eyelid ectro-
pion. Am J Ophthalmol 1989;108:610 –2.
30. McNab AA. The eye and sleep. Clin Experiment Ophthalmol
2005;33:117–25.
31. Taban M, Taban M, Perry JD. Plasma leptin levels in patients
with floppy eyelid syndrome. Ophthal Plast Reconstr Surg
2006;22:375–7.
32. Robert PY, Adenis JP, Tapie P, Melloni B. Eyelid hyperlaxity
and obstructive sleep apnea (O.S.A.) syndrome. Eur J Oph-
thalmol 1997;7:211–5.
33. Karger RA, White WA, Park WC, et al. Prevalence of floppy
eyelid syndrome in obstructive sleep apnea-hypopnea syn-
drome. Ophthalmology 2006;113:1669 –74.
34. Mojon DS, Goldblum D, Fleischhauer J, et al. Eyelid, con-
junctival, and corneal findings in sleep apnea syndrome. Oph-
thalmology 1999;106:1182–5.
35. McNab AA. Reversal of floppy eyelid syndrome with treat-
ment of obstructive sleep apnoea. Clin Experiment Ophthal-
mol 2000;28:125– 6.
36. Ali LK, Avidan AY. Sleep-disordered breathing and stroke.
Rev Neurol Dis 2008;5:191– 8.
37. NHS Strategic Tracing Service (NSTS) [database online].
Leeds, England; National Health Service; 2009. Available at:
http://www.connectingforhealth.nhs.uk/systemsandservices/
nsts. Accessed September 4, 2009.
38. Olver JM, Sathia PJ, Wright M. Lower eyelid medial canthal
tendon laxity grading: an interobserver study of normal sub-
jects. Ophthalmology 2001;108:2321–5.
39. Moe KS, Linder T. The lateral transorbital canthopexy for correction
and prevention of ectropion: report of a procedure, grading system,
and outcome study. Arch Facial Plast Surg 2000;2:9–15.
40. Iyengar SS, Khan JA. Quantifying upper eyelid laxity in symptom-
atic floppy eyelid syndrome by measurement of anterior eyelid
distraction [letter]. Ophthal Plast Reconstr Surg 2007;23:255.
41. Bobridis K, Assi A, Indar A, et al. Repeatability and repro-
ducibility of upper eyelid measurements. Br J Ophthalmol
2001;85:99 –101.
42. Johns MW. A new method for measuring daytime sleepiness:
the Epworth sleepiness scale. Sleep 1991;14:540 –5.
43. Johns MW. Sensitivity and specificity of the multiple sleep
latency test (MSLT), the maintenance of wakefulness test and
the Epworth sleepiness scale: failure of the MSLT as a gold
standard. J Sleep Res 2000;9:5–11.
44. Rosenthal LD, Dolan DC. The Epworth sleepiness scale in the
identification of obstructive sleep apnea. J Nerv Ment Dis
2008;196:429 –31.
45. Bürki E. New possibilities of corneal diagnostics using video-
keratometry (parts 1&2). Wetzlar, Germany: Oculus; 2000:1–
21. Available at: http://www.oculus-optikgeraete.de/de/sites/
detail_ger.php?page⫽478. Accessed September 4, 2009.
46. Ertan A, Kamburoglu G, Bahadir M. Intacs insertion with the
femtosecond laser for the management of keratoconus: one-
year results. J Cataract Refract Surg 2006;32:2039 – 42.
47. Bowker A. A test for symmetry in contingency tables. J Am
Stat Assoc 1948;43:572– 4.
48. Lee WJ, Kim JC, Shyn KH. Clinical evaluation of corneal
diseases associated with floppy eyelid syndrome. Korean J
Ophthalmol 1996;10:116 –21.
49. Gerner EW, Hughes SM. Floppy eyelid with hyperglycinemia.
Am J Ophthalmol 1984;98:614 – 6.
50. Rao LG, Bhandary SV, Devi AR, Gangadharan S. Floppy
eyelid syndrome in an infant [letter]. Indian J Ophthalmol
2006;54:217– 8.
51. Boulton JE, Sullivan TJ. Floppy eyelid syndrome and mental
retardation. Ophthalmology 2000;107:1989 –91.
52. Mack WP. Floppy eyelid syndrome and mental retardation
[letter]. Ophthalmology 2001;108:2159 – 60.
53. Stone DU, Chodosh J. Epibulbar nodular fasciitis associated
with floppy eyelids. Cornea 2005;24:361–2.
54. Burkat CN, Lemke BN. Acquired lax eyelid syndrome: an
unrecognized cause of the chronically irritated eye. Ophthal
Plast Reconstr Surg 2005;21:52– 8.
55. Shah-Desai S, Sandy C, Collin R. Lax eyelid syndrome or
‘progeria’ of eyelid tissues. Orbit 2004;23:3–12.
56. McMonnies CW. Abnormal rubbing and keratectasia. Eye
Contact Lens 2007;33:265–71.
57. Negris R. Floppy eyelid syndrome associated with keratoco-
nus. J Am Optom Assoc 1992;63:316 –9.
837
 Ophthalmology Volume 117, Number 4, April 2010

58. Brown MD, Potter JW. Floppy eyelid syndrome: a case report
and clinical review. J Am Optom Assoc 1992;63:309 –14.
59. Pham TT, Perry JD. Floppy eyelid syndrome. Curr Opin
Ophthalmol 2007;18:430 –3.
60. Broderick M, Guilleminault C. Neurological aspects of obstruc-
tive sleep apnea. Ann N Y Acad Sci 2008;1142:44 –57.
61. Malow BA, Levy K, Maturen K, Bowes R. Obstructive sleep
apnea is common in medically refractory epilepsy patients.
Neurology 2000;55:1002–7.
62. Krahn LE. Psychiatric disorders associated with disturbed
sleep. Semin Neurol 2005;25:90 – 6.
63. Punjabi NM, Shahar E, Redline S, et al. Sleep Heart Health
Study Investigators. Sleep-disordered breathing, glucose intol-
erance, and insulin resistance: the Sleep Heart Health Study.
Am J Epidemiol 2004;160:521–30.
64. Colagiuri S. Obstructive sleep apnoea—another important diabe-
tes-associated problem. Diabetes Res Clin Pract 2008;
81:1.

Footnotes and Financial Disclosures

Originally received: May 11, 2009. Funded by the special trustees of Moorfields Eye Hospital. The authors
Final revision: September 12, 2009. acknowledge a proportion of their financial support from the Department of
Accepted: September 18, 2009. Health through the award made by the National Institute for Health Research
Available online: January 25, 2010. Manuscript no. 2009-630. to Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of
1
Ophthalmology for a Specialist Biomedical Research Centre for Ophthalmol-
National Institute for Health Research Biomedical Research Centre for ogy. The views expressed in this publication are those of the authors and not
Ophthalmology, Moorfields Eye Hospital and UCL Institute of Ophthalmol- necessarily those of the Department of Health.
ogy, Moorfields Eye Hospital, NHS Foundation Trust, London, United Kingdom.
Financial Disclosure(s):
2
Adnexal Department, Moorfields Eye Hospital NHS Foundation Trust, The author(s) have no proprietary or commercial interest in any materials
London, United Kingdom. discussed in this article.
3
Department of Cell Biology, UCL Institute of Ophthalmology, London, Correspondence:
United Kingdom. Daniel G. Ezra, MA, MRCOphth, National Institute for Health Research
4 Biomedical Research Centre for Ophthalmology, 2nd Floor Richard Des-
Western Eye Hospital, London, United Kingdom.
mond Childrens Eye Centre, Moorfields Eye Hospital NHS Trust and UCL
This article forms part of a doctoral thesis submitted to Cambridge Institute of Ophthalmology, City Road, London EC1V 2PD. E-mail:
University. d.ezra@ucl.ac.uk.

838