805877

research-article2018
EJO0010.1177/1120672118805877European Journal of OphthalmologyEl Ameen et al.

EJO European
Journal of
Ophthalmology
Original Research Article

European Journal of Ophthalmology

Objective ocular surface tolerance in

1­–9
© The Author(s) 2018
Article reuse guidelines:
patients with glaucoma treated with sagepub.com/journals-permissions
https://doi.org/10.1177/1120672118805877
DOI: 10.1177/1120672118805877

topical preserved or unpreserved journals.sagepub.com/home/ejo

prostaglandin analogues

Ammar El Ameen, Guillaume Vandermeer,

Raoul K Khanna and Pierre-Jean Pisella

Abstract
Purpose: Preservatives in glaucoma medications have been associated with ocular toxicity. We compared ocular signs
and symptoms in patients with open-angle glaucoma or ocular hypertension treated in monotherapy with preserved or
preservative-free prostaglandin analogues.
Methods: Observational cross-sectional clinical study in real life. 82 patients treated for at least 6 months with
prostaglandin analogue were assessed for intraocular pressure, ocular symptoms and ocular signs including conjunctival
hyperaemia, tear break-up time and tear meniscus height measured using objective and non-invasive methods (OCULUS
Keratograph 5M). Patients presenting with symptoms of ocular toxicity with preserved prostaglandin analogues were
switched to preservative-free latanoprost, and a second assessment was processed 6 months after.
Results: At inclusion, 30 (36.6%) patients were treated with preservative-free latanoprost, 25 (30.5%) with preserved
latanoprost, 16 (19.5%) with preserved travoprost and 11 (13.4%) with preserved bimatoprost. Patients treated
with preservative-free latanoprost reported significantly less ocular symptoms upon instillation (mainly burning) and
between instillations than patients treated with preserved prostaglandin analogues. The mean conjunctival hyperaemia
(limbal + bulbar) was significantly lower with preservative-free latanoprost (2.08 ± 0.55) compared to preserved
latanoprost (2.50  ± 0.7, p = 
0.0085), preserved travoprost (2.67  ± 0.82, p = 0.0083) and preserved bimatoprost
(2.68 ± 0.67, p = 0.0041). There were no relevant between-group differences in mean tear meniscus height and break-up
time. Ocular symptoms and conjunctival hyperaemia improved when preserved prostaglandin analogues were switched
to preservative-free latanoprost for 6 months while intraocular pressure reduction was maintained.
Conclusion: Overall, this study suggests a better subjective and objective ocular tolerance when patients were treated
with preservative-free latanoprost than with other preserved prostaglandin analogues monotherapy. Switching to
preservative-free latanoprost maintained intraocular pressure at the same level as preservative prostaglandin analogue,
but improved ocular surface tolerance.

Keywords
Glaucoma, hyperaemia, meniscus, ocular hypertension, preservatives, prostaglandins
Date received: 1 May 2018; accepted: 19 September 2018

Introduction A large proportion of treated patients for open-angle

Glaucoma is the second leading cause of irreversible blind-
ness worldwide,1 and elevated intraocular pressure (IOP)
is the only known modifiable risk factor for onset and pro-
gression of the disease.2 In most patients, the IOP is effec-
tively reduced and controlled by daily administration of
topical medications, that is, beta-blockers, prostaglandin
analogues (PGAs) and carbonic anhydrase inhibitors, but
long-life treatment and follow-up are required.
glaucoma (OAG) or ocular hypertension (OHT) has ocular
symptoms upon and between instillations, ocular signs
Ophthalmology Department, CHRU Bretonneau, University of Tours,
Tours, France
Corresponding author:
Ammar El Ameen, Ophthalmology Department, CHRU Bretonneau,
University of Tours, 37000 Tours, France.
Email: ammar_elameen@yahoo.fr
2 European Journal of Ophthalmology 00(0)
including hyperaemia which overtime may impact quality
of life and adherence to treatment.3–5 Although ocular
adverse events are generally mild or moderate in severity,
adverse effects have been reported by physicians as the
second most common reason after lack of efficacy for
switching medication (19% vs 43%, respectively).5
However, poor adherence to treatment is believed to be
one major reason for treatment failure.2,6
Patients treated with preserved glaucoma eye drops,
including PGAs, have more frequently ocular signs and
symptoms when compared to preservative-free (PF) medi-
cations.7 Moreover, a significant improvement of ocular
symptoms and signs has been observed in patients in whom
the preserved eye drops were diminished in number or
switched into preservative-free eye drops.7 Benzalkonium
chloride (BAK) is the most commonly used preservative in
glaucoma eye drops and exerts an antimicrobial effect by a
powerful detergent action on bacterial walls and mem-
branes. This preservative has been proven to cause dose-
dependent toxic effects to the ocular surface.8,9 The
BAK-detergent effect in combination with a partial destruc-
tion of mucous gland cells is responsible for lacrimal film
instability, resulting in decreased tear break-up time (BUT)
and symptoms of irritation. BAK is believed to be largely
responsible for ocular toxicities and inflammation associ-
ated with the chronic use of many ophthalmic solutions.10
Clinical assessment of ocular intolerance can be chal-
lenging.11,12 Clinicians generally rely on the presence of
ocular redness, Schirmer’s test, fluorescein tear BUT and
ocular surface staining to evaluate the health of the ocular
surface. However, the methods used are often considered as
poorly accurate. Conjunctival hyperaemia can be evaluated
using image-based comparative grading scales, for exam-
ple, the photographic scale of Efron et al.13 or McMonnies
and Chapman-Davies.14 However, these methods rely on
the judgement of the observer, and gradations in terms of
severity may be inconsistent between different scales.15
Non-invasive techniques using keratography have been
recently considered as reliable objective measurements of
ocular signs including tear meniscus height (TMH), BUT
and conjunctival hyperaemia.16,17
In this study, we compared ocular symptoms and signs
as assessed by objective measurement with keratography
in patients treated with preserved or unpreserved PGAs.
Secondarily, we determined ocular surface tolerance
6 months after switching from preserved PGAs (P-PGAs)
to preservative-free latanoprost (PF-Latanoprost).
Methods
Study design
This observational cross-sectional study was conducted in
2015 and 2016 in a clinical ophthalmologic centre in France
(CHRU Bretonneau, Tours, France). Patients (males or
females aged ⩾ 18 years) were eligible if they were treated
since at least 6 months with a PGA monotherapy including
preserved latanoprost (P-Latanoprost) (Xalatan®; Pfizer PFE
France, Paris, France), preserved travoprost (P-Travoprost)
(Travatan®; Novartis Pharma, Rueil-Malmaison, France),
preserved bimatoprost (P-Bimatoprost) (Lumigan® 0.1%;
Laboratoires Allergan France SAS, Courbevoie, France) or
PF-Latanoprost (Monoprost®; Laboratoires Théa, Clermont-
Ferrand, France). At the end of the inclusion visit, the oph-
thalmologist decided to continue the current treatment or to
switch with PF-Latanoprost if the patient presented at least
two symptoms of ocular toxicity.
The study was conducted in compliance with the ethical
principles of the Declaration of Helsinki. Study informa-
tion was given to each patient before enrolment. Since the
study was carried out with currently approved IOP-lowering
treatments and without change in the current patient care,
the study protocol was not submitted for approval to an
independent Ethics Committee in agreement with the
French Law and French Good Epidemiological Practice.
All data were collected and processed anonymously.
Study assessments
At each visit, the patient was asked to report the pres-
ence or absence of ocular symptoms upon instillations
(none, pain or discomfort, burning or blurred vision) and
between instillations (none, stinging, itching, burning,
tearing, photophobia, foreign body sensation, eyelash
crust or ocular dryness sensation). One drop of oxybu-
procaine 0.4% and one of fluorescein 0.05% were
administered in each eye, and three IOP measurements
were performed in both eyes using a calibrated and vali-
dated Goldmann applanation tonometer (AT900, Haag-
Streit AG, Koeniz, Switzerland).
The OCULUS Keratograph 5M (K5M) Topographer
(OCULUS Optikgerate GmbH, Wetzlar, Germany) was
used to perform non-invasive measurements of tear menis-
cus height (NIK-TMH), conjunctival hyperaemia and tear
film break-up time (NIK-BUT). Lower tear film meniscus
images were first captured, and TMH was measured with
the integrated ruler. For conjunctival hyperaemia measure-
ments, the patient was asked to stare straight ahead and
focus on the fixation mark inside the camera so the entire
corneal area appeared in the centre of the image. The kera-
tograph image of 1156 × 873 pixels and with a resolution of
96 dpi was analysed, and the system displayed a redness
score (accurate to 0.1 unit) on the computer screen within
1 s. Both the bulbar and the limbal areas were scanned.
Redness scores (Oculus index) were calculated automati-
cally by the device’s software as the area percentage ratio
between the vessels and the rest of the analysed area. For
example, if the ratio is 12%, then the score is 1.2. The
El Ameen et al. 3

Table 1.  Patients’ characteristics at inclusion.

Unpreserved Preserved Preserved Preserved

latanoprost latanoprost travoprost bimatoprost

  N = 30 N = 25 N = 16 N = 11

Age, years (mean ± SD) 72.5 ± 11.9 72.1 ± 10.8 66.1 ± 12.3 68.1 ± 12.4
Sex, n (%)
 Male 12 (40.0) 13 (52.0) 8 (50.0) 5 (45.5)
 Female 18 (60.0) 12 (48.0) 8 (50.0) 6 (54.5)
Comorbidities, n (%)
 Hypertension 6 (20.0) 4 (16.0) 5 (31.3) 0 (0.0)
 Diabetes 2 (6.7) 0 (0.0) 2 (12.5) 0 (0.0)
Ocular diseases, n (%)
  Open-angle glaucoma 28 (93.3) 25 (100.0) 15 (93.8) 11 (100.0)
  Cataract surgery 14 (46.7) 12 (48.0) 10 (62.5) 3 (27.3)
Intraocular pressure, mm Hg (mean ± SD)
  Right eye 16.0 ± 3.0 16.5 ± 3.5 18.2 ± 5.2 16.7 ± 3.1
  Left eye 16.1 ± 3.3 17.3 ± 3.3 17.1 ± 4.5 16.8 ± 3.8

SD: standard deviation.

maximum ratio is 40%; therefore, the redness scores range
between 0.0 and 4.0.17 BUT was measured using the
Keratograph 5M by detecting localised breaks in the tear
film using infrared waves. After two blinks to reconstitute
the tear film, BUT is recorded until the subject blinks again.
The Keratograph 5M provides two scores for BUT: the time
taken for the first appearance of a break in the tear film
(NIK-BUT-First) and the average of the time taken to break-
up in all the regions monitored over the duration of the 25 s
(NIK-BUT-Average). The NIK-BUT-Average is, therefore,
higher than the NIK-BUT-First, and only the latter was used
in our analysis. In this study, conjunctival hyperaemia was
defined as the sum of the bulbar and limbal values. The pri-
mary efficacy outcome was the ocular signs and symptoms
at inclusion visit. For patients with ocular signs of intoler-
ance, the evolution of ocular signs and symptoms after
6 month of treatment switch to PF-Latanoprost was assessed
as a secondary study outcome.
Statistics
All data collected from patients files were entered anony-
mously into an Excel spreadsheet (Microsoft Corp.,
Redmond, WA, USA) and processed using the XLSTAT
statistical software (Addinsoft, NY, USA). At inclusion,
ocular symptoms were compared between the
PF-Latanoprost group and each other preserved prosta-
glandin groups using the Fisher exact test. Conjunctival
hyperaemia, NIK-BUT and NIK-TMH at inclusion were
compared between treatment groups using the one-way
analysis of variance (ANOVA) with post hoc Dunnett test
comparing the PF-Latanoprost group with other preserved
groups treatment groups. At the 6-month visit, the changes
from inclusion in ocular symptoms were compared using a
Fisher exact test and the changes in ocular hyperaemia,
NIK-TMH and NIK-BUT using the Wilcoxon test for
quantitative variables. p values below 0.05 were consid-
ered statistically significant.
Results
Patients’ characteristics at inclusion are shown in Table 1.
Overall, 82 patients (44 females and 38 males, mean age:
70.5 ± 8.2 years) with OAG or OHT were included: 30
(36.6%) patients (58 eyes; 2 patients were monophthal-
mic) were treated with PF-Latanoprost, 25 (30.5%) with
P-Latanoprost (50 eyes), 16 (19.5%) with P-Travoprost
(32 eyes) and 11 (13.4%) with P-Bimatoprost 0.1% (21
eyes; one patient was monophthalmic).
Ocular symptoms in PF-Latanoprost versus
P-PGAs
As shown in Figure 1(a), ocular symptoms upon instillations
were reported less frequently with PF-Latanoprost than with
each other P-PGAs: 11.5% with PF-Latanoprost versus
38.1%, 42.9% and 33.3%, with P-Latanoprost (p  = 
0.040), P-Travoprost (p = 0.044) and P-Bimatoprost (not sig-
nificant, p = 0.16), respectively. Statistically significant differ-
ences were obtained for burning upon instillation which was
lower with PF-Latanoprost (3.8%) compared to P-Latanoprost
(38.1%, p  = 0.006), P-Travoprost (42.9%, p  = 0.006) and
P-Bimatoprost (33.3%, p = 0.044). As shown in Figure 1(b),
the rate of patients reporting at least one ocular symptom
between instillations was also significantly lower with
PF-Latanoprost (42.3%) compared to P-Travoprost (85.7%,
4 European Journal of Ophthalmology 00(0)

Figure 1.  Ocular symptoms (PF-Latanoprost group vs each P-PGA): (a) ocular symptoms upon instillation and (b) ocular symptom
between instillation.
The percentages of patients reporting each ocular symptom and the percentage of patients with at least one symptom upon instillations and
between instillations are shown for each treatment group. The number of patients (N) in each treatment group is indicated in legend. *p < 0.05;
**p < 0.01; ***p < 0.001 compared to the PF-Latanoprost group.

p = 0.034) and P-Bimatoprost (88.9%, p = 0.022), but not
compared to P-Latanoprost (71.4%, p = 0.076).
Ocular signs in PF-Latanoprost versus each P-PGA
Mean values of conjunctival hyperaemia are shown in
Figure 2. Overall, mean conjunctival hyperaemia was
significantly different between treatment groups for bul-
bar hyperaemia (ANOVA, p = 0.019), limbal (p < 0.001)
and total (bulbar  + limbal) hyperaemia (p = 0.0011).
Patients treated with PF-Latanoprost had significantly
lower conjunctival hyperaemia (limbal  + bulbar:
2.08 ± 0.55) compared with P-Latanoprost (2.50 ± 0.70,
p = 0.009), P-Travoprost (2.67 ± 0.82, p = 0.008) and
P-Bimatoprost (2.68 ± 0.67, p = 0.004). The between-
group difference was mainly due to limbal hyperaemia
with a mean value of 0.84 ± 0.27 with PF-Latanoprost
versus 1.02  ± 0.33 with P-Latanoprost (p = 0.025),
1.20 ± 0.44 (p < 0.001) for P-Travoprost and 1.11 ± 0.32
(p = 0.007) for P-Bimatoprost. Bulbar hyperaemia was
El Ameen et al. 5
Figure 2.  Conjunctival hyperaemia (PF-Latanoprost vs each
P-PGA).
Mean values (±SD) of bulbar, limbal and bulbar + limbal conjunctival
hyperaemia are shown for each treatment group. The number of eyes
(NE) in each treatment group is indicated along the x-axis. One way
ANOVA: p = 0.001 for limbal + bulbar hyperaemia, p < 0.001 for limbal
hyperaemia, and p = 0.019 for bulbar hyperaemia. *p < 0.05; **p < 0.01;
***p < 0.001 compared with the PF-Latanoprost.
also lower in the PF-Latanoprost group (1.27 ± 0.35)
than in each other preserved treatment group, and the dif-
ference was significant with P-Bimatoprost (1.58 ± 0.40,
p = 0.013), but not with P-Latanoprost (1.43  ± 0.44,
p = 0.12) and P-Travoprost (1.47 ± 0.42, p = 0.098). There
were no relevant between-group differences in mean
TMH and NIK-BUT (Figure 3).
Ocular tolerance after treatment switch from
P-PGAs to PF-Latanoprost
At the end of the inclusion visit, 30/52 patients (57.7%) treated
with a preserved prostaglandin switched to PF-Latanoprost
because of ocular intolerance; 17/25 (68.0%) patients (34
eyes) previously treated with P-Latanoprost, 8/16 (50.0%)
patients (16 eyes) previously treated with P-Travoprost and
5/11 (45.5%) patients (9 eyes) previously treated with
P-Bimatoprost. For each prostaglandin used, the percentage of
patients with at least one symptom upon instillation was
reduced after 6 months (Figure 4(a)). The reduction was statis-
tically significant when all P-PGAs were analysed together
(12% vs 44%, p = 0.026), and this was mainly due to reduction
in burning (4% vs 44%, p = 0.0019). Similarly, the percentage
of patients with at least one symptom between instillations
was reduced for each preserved prostaglandin and overall
(40% vs 72%, p = 0.045) (Figure 4(b)).
After the switch, there was a small but significant
decrease in the mean conjunctival hyperaemia in patients
previously treated with P-Latanoprost (2.15  ± 0.60 vs
2.36 ± 0.81, p = 0.047), with P-Travoprost (2.20 ± 0.89 vs
2.56 ± 0.93, p = 0.009) and with P-Bimatoprost (2.31 ± 0.47
vs 2.79 ± 0.49, p = 0.030) (Figure 5). There was no relevant
change in NIK-TMH and NIK-BUT in each preserved
treatment group after the switch (data not shown). In addi-
tion, the mean IOP was maintained after the switch for each
preserved prostaglandin: 16.1 ± 3.3 mm Hg at inclusion
versus 15.9  ± 3.1 mm Hg after 6 months (p = 0.93) for
P-Latanoprost, 15.5 ± 1.3 versus 15.3 ± 2.2 (p = 0.86) for
P-Travoprost and 16.8 ± 2.8 versus 14.4 ± 0.5 (p = 0.12) for
P-Bimatoprost.
Discussion
The aim of this observational study in real life was to
determine the ocular tolerance in patients with glaucoma
and/or OHT treated in monotherapy with preserved prosta-
glandin analogues in comparison with PF-Latanoprost.
Overall, the study confirmed the higher prevalence of sub-
jective symptoms in patients treated with preserved prosta-
glandin eye drops compared to PF-Latanoprost. Beside a
lower prevalence of ocular symptoms, patients treated
with PF-Latanoprost monotherapy showed significantly
lower conjunctival hyperaemia, but similar BUT and
TMH, compared to patients treated with P-Travoprost or
P-Bimatoprost.
The prevalence of ocular symptoms in this study is
consistent with the rate of ocular surface diseases (38%)
reported in patients treated in monotherapy with glau-
coma eye drops.18 Ocular burning sensation upon instilla-
tions was the main ocular symptom and was significantly
less reported in patients treated with PF-Latanoprost
compared to each other preserved prostaglandin ana-
logues, including P-Latanoprost, P-Travoprost and
P-Bimatoprost. Ocular symptoms between instillations
including itching and burning, and eyelash crust were
also relatively more frequent in patients treated with
P-Travoprost and P-Bimatoprost compared to
PF-Latanoprost, although the differences were generally
not statistically significant, likely due to small sample
size in each treatment group.
Overall, the ocular signs determined using objective
non-invasive methods with the Keratograph 5M indicates
a mild ocular intolerance in all treatment groups with on
average low hyperaemic scores, normal TMH and BUT.
Nevertheless, using this method, conjunctival hyperaemia
was significantly lower in patients treated with
PF-Latanoprost compared to P-Latanoprost, P-Travoprost,
and P-Bimatoprost. These results are in agreement with a
randomised parallel group clinical trial, where conjuncti-
val hyperaemia was less frequent and severe in patients
treated with PF-Latanoprost compared to patients treated
with P-Latanoprost.19 Similarly, a meta-analysis of ran-
domised controlled clinical trials suggested a lower occur-
rence of conjunctival hyperaemia with PF-Latanoprost
6 European Journal of Ophthalmology 00(0)

Figure 3.  TMH and BUT (PF-Latanoprost vs each P-PGA): (a) tear meniscus height (mm) and (b) NIK break-up time (s).
Mean values (±SD) of NIK-TMH (Panel a) and NIK-BUT (Panel b) for each treatment groups. The number of eyes (NE) in each treatment group is
indicated along the x-axis.

compared to preserved latanoprost, bimatoprost and
travoprost.20
TMH and BUT measurements with the Keratograph
5M have been evaluated in patients with dry eye dis-
ease.21,22 Overall, for the glaucoma patients in this study,
the mean NIK-TMH ranged between 0.30 ± 0.09 and
0.41 ± 0.18 mm according to treatment groups, and BUT
between 7.18 ± 5.56 s and 9.44 ± 7.92 s. By comparison
using the Keratograph 5M, Tian et al.21 reported NIK-
TMH values of 0.27 ± 0.12 mm for normal eyes and
0.22 ± 0.07 mm for dry eyes and NIK-BUT-first values of
7.36 ± 3.99 s for normal eyes and 5.57 ± 3.31 s for dry
eyes. In another study, Ko et al. reported NIK-TMH val-
ues of 0.20 ± 0.05 mm for healthy eyes and 0.14 ± 0.03 mm
for dry eyes and NIK-BUT-first of 9.71 ± 6.68 s for
healthy eyes and 4.59 ± 1.25 s for dry eyes.16 In patients
under topical glaucoma treatment, the mean NIK-TMH
was 0.28 ± 0.10 mm and NIK-BUT value was 9.71 ± 5.8 s,
and these values were not significantly different in control
eyes of healthy age- and sex-matched volunteers.23 Thus,
patients of the present study had rather a normal TMH and
BUT, and no differences were shown in NIK-TMH and
NIK-BUT between patients treated with PF-Latanoprost
versus P-PGAs. Although previous studies showed a
reduced fluorescein BUT in patients treated with topical
glaucoma medication,24,25 the current findings did not
indicate overt dry eye disease when patients were treated
since at least 6 months with a PGA monotherapy.
El Ameen et al. 7

Figure 4.  Ocular symptoms after treatment switch from P-PGAs to PF-Latanoprost: (a) symptoms upon instillations and (b)
symptoms between instillations.
Percentages of patients reporting at least one symptom upon instillations (Panel a) and between instillations (Panel b) are shown at inclusion (before
the switch) and 6 months after the switch to PF-Latanoprost. *p < 0.05; **p < 0.01.

When a patient treated with glaucoma medication
shows signs which could evoke ocular toxicity, the issue
is to determine which of the active substance or the pre-
servative is more responsible for side effects. BAK used
as a preservative in glaucoma eye drops may increase
ocular surface disease by disrupting the tear film and
increasing conjunctival inflammation.8 In this study,
ocular symptoms are significantly less frequent when
preserved eye drops were switched to PF-Latanoprost
consistent with a role for the preservative, as suggested
in a previous switching study.26 There was also a small
but statistically significant decrease in bulbar and/or
limbal conjunctival hyperaemia when preserved prosta-
glandins were switched with PF-Latanoprost for
6 months. Conjunctival hyperaemia decreased for all
classes of P-PGAs, including P-Latanoprost, suggesting
a role of the preservative.
The main limitation of this study is the small size and
the lack of information regarding the duration of the pre-
served and unpreserved treatments before inclusion. Thus,
further studies are needed to understand the part of the
active ingredient and the preservative in ocular intolerance
during topical glaucoma therapy.
In conclusion, this study in real life suggests a better
subjective and objective ocular tolerance when patients are
treated with PF-Latanoprost compared to other PGAs.
Further studies are required to determine the benefit to
switch from P-PGAs to PF-Latanoprost.
8 European Journal of Ophthalmology 00(0)

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with
respect to the research, authorship and/or publication of this
article.

Funding
The author(s) received no financial support for the research,
authorship and/or publication of this article.

References
10. Guenoun J-M, Baudouin C, Rat P, et al. In vitro study of
Figure 5.  Conjunctival hyperaemia after treatment switch
from P-PGAs to PF-Latanoprost: (a) bulbar + limbal
hyperaemia, (b) bulbar hyperaemia and (c) limbal hyperaemia.
Mean (±SD) of total (Panel a), bulbar (Panel b) and limbal (Panel c)
conjunctival hyperaemia are indicated for each treatment group at in-
clusion (before the switch) and 6 months after the switch to PF-Latano-
prost. The number of eyes (NE) in each group is indicated along with
the x-axis. *p < 0.05 and **p < 0.01 (Wilcoxon test, before vs after the
switch to PF-Latanoprost).
Acknowledgements
The authors thank Thierry Radeau for medical writing support.
A.E.A., R.K.K., G.V. and P.-J.P. enrolled patients and collected
the data; A.E.A. made the statistical analyses and wrote the man-
uscript; and R.K.K., G.V. and P.-J.P. reviewed the manuscript.
1. Tham Y-C, Li X, Wong TY, et al. Global prevalence of glau-
coma and projections of glaucoma burden through 2040: a
systematic review and meta-analysis. Ophthalmology 2014;
121: 2081–2090.
2. Harasymowycz P, Birt C, Gooi P, et al. Medical manage-
ment of glaucoma in the 21st century from a Canadian per-
spective. J Ophthalmol 2016; 2016: 6509809.
3. Lemij HG, Hoevenaars JG, van der Windt C, et al. Patient
satisfaction with glaucoma therapy: reality or myth? Clin
Ophthalmol Auckl NZ 2015; 9: 785–793.
4. Rossi GCM, Pasinetti GM, Scudeller L, et al. Ocular surface
disease and glaucoma: how to evaluate impact on quality of
life. J Ocul Pharmacol Ther Off J Assoc Ocul Pharmacol
Ther 2013; 29: 390–394.
5. Zimmerman TJ, Hahn SR, Gelb L, et al. The impact of ocular
adverse effects in patients treated with topical prostaglandin
analogs: changes in prescription patterns and patient persis-
tence. J Ocul Pharmacol Ther Off J Assoc Ocul Pharmacol
Ther 2009; 25: 145–152.
6. Arias A, Schargel K, Ussa F, et al. Patient persistence with
first-line antiglaucomatous monotherapy. Clin Ophthalmol
Auckl NZ 2010; 4: 261–267.
7. Jaenen N, Baudouin C, Pouliquen P, et al. Ocular symptoms
and signs with preserved and preservative-free glaucoma
medications. Eur J Ophthalmol 2007; 17: 341–349.
8. Baudouin C, Labbé A, Liang H, et al. Preservatives in eye-
drops: the good, the bad and the ugly. Prog Retin Eye Res
2010; 29: 312–334.
9. Rosin LM and Bell NP. Preservative toxicity in glaucoma
medication: clinical evaluation of benzalkonium chloride-
free 0.5% timolol eye drops. Clin Ophthalmol Auckl NZ
2013; 7: 2131–2135.
inflammatory potential and toxicity profile of latanoprost,
travoprost, and bimatoprost in conjunctiva-derived epi-
thelial cells. Invest Ophthalmol Vis Sci 2005; 46: 2444–
2450.
11. Pflugfelder SC and Baudouin C. Challenges in the clinical
measurement of ocular surface disease in glaucoma patients.
Clin Ophthalmol Auckl NZ 2011; 5: 1575–1583.
12. Savini G, Prabhawasat P, Kojima T, et al. The challenge
of dry eye diagnosis. Clin Ophthalmol Auckl NZ 2008; 2:
31–55.
13. Efron N, Morgan PB and Katsara SS. Validation of grading
scales for contact lens complications. Ophthalmic Physiol
Opt J Br Coll Ophthalmic Opt Optom 2001; 21: 17–29.
14. McMonnies CW and Chapman-Davies A. Assessment of
conjunctival hyperemia in contact lens wearers: part I. Am J
Optom Physiol Opt 1987; 64: 246–250.
El Ameen et al. 9
15. Schulze MM, Hutchings N and Simpson TL. The perceived
bulbar redness of clinical grading scales. Optom Vis Sci
2009; 86: E1250–E1258.
16. Koh S, Ikeda C, Watanabe S, et al. Effect of non-invasive
tear stability assessment on tear meniscus height. Acta
Ophthalmol (Copenh) 2015; 93: e135–e139.
17. Wu S, Hong J, Tian L, et al. Assessment of bulbar redness
with a newly developed keratograph. Optom Vis Sci Off
Publ Am Acad Optom 2015; 92: 892–899.
18. Baudouin C, Renard J-P, Nordmann J-P, et al. Prevalence
and risk factors for ocular surface disease among patients
treated over the long term for glaucoma or ocular hyperten-
sion. Eur J Ophthalmol 2012; 23: 47–54.
19. Rouland J-F, Traverso CE, Stalmans I, et al. Efficacy and
safety of preservative-free latanoprost eyedrops, compared
with BAK-preserved latanoprost in patients with ocular
hypertension or glaucoma. Br J Ophthalmol 2013; 97: 196–
200.
20. Cucherat M, Stalmans I and Rouland J-F. Relative efficacy
and safety of preservative-free latanoprost (T2345) for the
treatment of open-angle glaucoma and ocular hypertension:
an adjusted indirect comparison meta-analysis of rand-
omized clinical trials. J Glaucoma 2014; 23: e69–e75.
21. Tian L, Qu J, Zhang X, et al. Repeatability and reproduc-
ibility of noninvasive Keratograph 5M measurements in
patients with dry eye disease. J Ophthalmol 2016; 2016:
8013621.
22. Craig JP, Nichols KK, Akpek EK, et al. TFOS DEWS II
definition and classification report. Ocul Surf 2017; 15:
276–283.
23. Pérez-Bartolomé F, Martínez-de-la-Casa JM, Arriola-
Villalobos P, et al. Ocular surface disease in patients under
topical treatment for glaucoma. Eur J Ophthalmol 2017; 27:
694–704.
24. Cvenkel B, Štunf S, Srebotnik Kirbiš, et al. Symptoms and
signs of ocular surface disease related to topical medication
in patients with glaucoma. Clin Ophthalmol Auckl NZ 2015;
9: 625–631.
25. Tomić M, Kaštelan S, Soldo KM, et al. Influence of BAK-
preserved prostaglandin analog treatment on the ocular
surface health in patients with newly diagnosed primary
open-angle glaucoma. BioMed Res Int 2013; 2013: 603782.
26. Walimbe T, Chelerkar V, Bhagat P, et al. Effect of benza-
lkonium chloride-free latanoprost ophthalmic solution on
ocular surface in patients with glaucoma. Clin Ophthalmol
Auckl NZ 2016; 10: 821–827.