The Ocular Surface 18 (2020) 141–147

Contents lists available at ScienceDirect

The Ocular Surface

journal homepage: www.elsevier.com/locate/jtos

Original Research

Schirmer test results: are they associated with topical or systemic T

medication?
Ulrike Hampela,∗,1, Alexander K. Schustera,1, Stefan Nickelsa, Andreas Schulzb, Karl J. Lacknerc,
Thomas Münzeld,e, Philipp S. Wildb,e,f, Manfred Beutelg, Irene Schmidtmannh, Norbert Pfeiffera
a
Department of Ophthalmology, University Medical Center of the Johannes Gutenberg-University Mainz, Mainz, Germany
b
Preventive Cardiology and Preventive Medicine, Center for Cardiology, University Medical Center of the Johannes Gutenberg-University Mainz, Germany
c
Institute for Clinical Chemistry and Laboratory Medicine, University Medical Center of the Johannes Gutenberg-University Mainz, Mainz, Germany
d
Center for Cardiology I, University Medical Center of the Johannes Gutenberg-University Mainz, Mainz, Germany
e
DZHK (German Center for Cardiovascular Research), Partner Site Rhine-Main, Mainz, Germany
f
Center for Thrombosis and Hemostasis (CTH), University Medical Center of the Johannes Gutenberg-University Mainz, Germany
g
Department of Psychosomatic Medicine and Psychotherapy, University Medical Center of the Johannes Gutenberg-University Mainz, Germany
h
Institute for Medical Biostatistics, Epidemiology and Informatics, University Medical Center of the Johannes Gutenberg-University Mainz, Mainz, Germany

ABSTRACT

Purpose: To test whether Schirmer test (ST) results are associated with topical or systemic medication and to evaluate the distribution of tear fluid quantity in a 3-min
and 5-min ST.
Methods: The Gutenberg Health Study is a population-based, prospective, observational cohort study in Germany. ST was assessed in a sub-cohort of 1,999 parti-
cipants. ST was performed under topical anesthesia for 5 min (ST-5) or of 3 min (ST-3). Anthropometric factors, systemic diseases, use of systemic and eye medi-
cations were recorded. We used multivariable quantile regression analysis to assess the influence on ST measurements.
Results: The length of wetting of the Schirmer strips for ST-5 was 23.2 ± 9.31 mm for right and 22.9 ± 9.0 mm for left eyes. In ST-3, the measurements were
20.0 mm in right and 19.1 mm in left eyes. The clinical cut off of 10 mm for ST-5 corresponded with an 8 mm cut off for ST-3.
There was an association of smaller ST-5 measures with male sex, higher age, socioeconomic status and season (all p < 0.001), but not with diabetes or smoking.
The use of prostaglandin or beta-blocker eye drops or oral non-steroidal anti-inflammatory drugs, drugs for peptic ulcer and gastro-oesophageal reflux disease,
thyroid hormones, progesterone and estrogen combination drugs, and hypnotics and sedatives showed an association with smaller ST-5.
Conclusions: For the first time we describe the distribution of tear fluid quantity by ST in a very large cohort of the general population. Furthermore, we found
associations of ST measures with topical and systemic medication.

Background cataract extraction, LASIK, and glaucoma procedures [3]. Prevalence

The tear film has an important role for the maintenance of ocular
surface health [1]. In dry eye disease the lack of tear fluid is one of the
reasons leading to the loss of homeostasis [1].
Dry eye disease is defined by the Tear Film and Ocular Surface
Society Dry Eye Workshop II as “a multifactorial disease of the ocular
surface characterized by a loss of homeostasis of the tear film, and
accompanied by ocular symptoms, in which tear film instability and
hyperosmolarity, ocular surface inflammation and damage, and neu-
rosensory abnormalities play etiological roles” [2].
Dry eye disease is more common in higher age and associated with
variable vision and reduced quality of life [3,4]. It may lead to cessation
of contact lens use and can diminish results of eye surgeries, such as
data are not available for Germany, however the worldwide prevalence
for dry eye disease ranges from 8 to 60% [5]. Measuring the tear fluid
volume is one of several diagnostic methods for dry eye disease.
However, little is known about the physiological tear film volume. The
Schirmer test to measure tear fluid volumes was introduced in 1903 and
was named after Otto Wilhelm August Schirmer (1864–1917), a
German ophthalmologist [6,7]. So far no normative data for the general
population were available and only studies with less than 500 partici-
pants are published. The Tear Film and Ocular Surface Society Dry Eye
Workshop II report on iatrogenic dry eye summarized the current
known medication causing dry eye [8]. However, in the large epide-
miological studies that identified systemic drugs for risk of dry eye
disease the definition of dry eye disease were either self-reported

Corresponding author. Department of Ophthalmology, University Medical Center of the Johannes Gutenberg-University Mainz, Langenbeckstr. 1, 55131, Mainz,
∗

Germany.
E-mail address: ulrike.hampel@unimedizin-mainz.de (U. Hampel).
1
these authors contributed equally and share first authorship.

https://doi.org/10.1016/j.jtos.2019.11.003
Received 22 October 2018; Received in revised form 27 October 2019; Accepted 13 November 2019
1542-0124/ © 2019 Elsevier Inc. All rights reserved.
U. Hampel, et al.
symptoms or clinically diagnosed dry eye disease that was documented
as ICD code or reported by the patients [9–15]. For the Schirmer test, a
paper strip is placed laterally inside the lower eye lid for 5 min and the
wetting of the paper strip is a measure for tear production. The standard
test duration of 5 min is often associated with unpleasant feeling for the
patient. A shorter duration would be preferred also for the clinical
routines.
Therefore, we aimed to evaluate tear fluid quantity and for the first
time to provide normative data as well as to investigate the association
with factors so far described to be associated with dry eye disease,
namely age, sex and medication in a population-based study with over
1,300 participants. Furthermore, we aimed to compare Schirmer test
results of 3 and 5 min test duration.
Population, materials and methods
Study population
The Gutenberg Health Study (GHS) is a prospective, population-
based, single-center cohort study at the medical center of the Johannes
Gutenberg University Mainz in Germany [16]. The research followed
the tenets of the Declaration of Helsinki; informed consent was ob-
tained from the participants after explanation of the nature and possible
consequences of the study; and the research was approved by the in-
stitutional human experimentation committee. The study protocol and
study documents were approved by the local ethics committee of the
Medical Chamber of Rhineland-Palatinate, Germany (reference no.
837.020.07; original vote: 22.3.2007, latest update: 20.10.2015). Ac-
cording to the tenets of the Declaration of Helsinki, written informed
consent was obtained from all participants prior to entering the study.
The population sample was randomly drawn via local residents’ regis-
tration offices and equally stratified by sex for each decade of age. The
baseline examination included 15,010 participants aged 35–74 years
and was conducted from 2007 to 2012. From 2012 to 2017, the five-
year-follow-up was conducted. The examination consisted of an oph-
thalmological examination, general and cardiovascular examinations,
questionnaires and interviews, and biospecimen sampling. The medi-
cation of study participants is scanned and grouped into Anatomical
Therapeutic Chemical Classification (ATC)-codes. The ophthalmic
branch has been described in detail before [17]. In brief, we conducted
measurements of objective refraction and distance-corrected visual
acuity, intraocular pressure, visual field testing, Scheimpflug imaging of
the anterior segment and fundus photography. In addition after a short
resting pause of approximately 3 min, a tear fluid sampling with
Schirmer test stripes was performed and participants were questioned
for other eye diseases including dry eye.
Schirmer tests (ST) were performed with topical anesthesia, as since
ST with anesthesia may be more reliable in dry eye disease detection
compared to ST that is conducted without anesthesia [18]. Topical
anesthesia with Oxybuprocainhydrochlorid (Novesine 0.4% eye drops,
OmniVision GmbH, Puchheim, Germany, one drop per eye) were ap-
plied, excessive liquid was removed with paper tissue from the lid after
closing of the eyes after 1 min and the folded end of the Schirmer strip
paper was placed in the temporal one-third of the lower cul-de-sac. The
Schirmer strips were bought from Optitech® Eyecare, Allahabad, India.
In a sub-cohort, tear fluid quantity was determined as the measured
length of wetting. . The score is the measured length of wetting in mm.
ST was performed with a duration of 3 min (ST-3) in a subgroup,
while in another subset ST was carried out for 5 min (ST-5). ST was
performed in both eyes and statistical analysis integrated this fact. ST
with a fully soaked test strip was recorded as 35 mm.
Comorbidities
Diabetes mellitus was defined as diabetes mellitus diagnosed by a
physician, known therapy (oral medication or insulin), or HbA1c ≥ 6.5%.
142
The Ocular Surface 18 (2020) 141–147
Sociodemographic characteristics
The socioeconomic status (SES) was based on income, education
and occupation and was defined according to the SES-index as used
within the German Health Update 2009 (GEDA), ranging from 3
(lowest) to 21 (highest) [19].
Statistical analysis
The data management team performed quality controls and checked
for completeness and correctness by predefined algorithms, before the
data were incorporated in the database. The spherical equivalent of
each eye was calculated for each participant. Visual acuity measure-
ments were converted to logMAR [20].
Descriptive analysis was performed and absolute and relative fre-
quencies were calculated for categorical variables. Mean and standard
were computed for continuous variables, when approximately normal
distributed, otherwise median and interquartile range.
Correlations in ST measures between both eyes of one study parti-
cipant were evaluated computing Spearman Correlation Coefficients.
Proportion of ST for the cut-off of 10 mm (clinical cut-off [21]) was
computed for 5 min test duration and the corresponding percentiles.
The wetting distance in the 3 min ST for these percentiles were calcu-
lated.
Linear regression analysis was performed along with diagnostics to
determine whether the outcome is normally distributed. The plots of
residuals were not normally distributed further informing us of the
skewed distribution of the outcome. We, therefore, performed quanitle
regression analysis on the median to evaluate associations using mixed
model statistics and control for including both eyes from one study
participant. We also ran a generalized estimating equation (GEE)
model, keeping the outcome continuous for mean comparisons of the
population average effect (GEE results are included in the supplement).
As dependent variable ST II measurement was included, as independent
variables age, sex, SES, diabetes mellitus, smoking and season (winter:
December, January, February; spring: March, April, May; summer:
June, July, August; autumn: September, October, November). In addi-
tion, we conducted a model including eye drops (prostaglandin (ATC:
s01ee), beta blocking agents (ATC: s01ed), artificial tears (ATC:
s01xa20)) as well as systematic medication namely diuretics (ATC:
c03), beta-blockers (ATC: 07), thyroid hormones (ATC: h03aa), non-
steroidal anti-inflammatory drugs (ATC: m01a), contraceptives (ATC:
g03a), estrogen-hormones (ATC: g03c), combined hormone medication
(progesterones and estrogens; ATC: g03f), antiandrogen therapy (ATC:
g04c), vasodilators (ATC: c01d), drugs for peptic ulcer and gastro-oe-
sophageal reflux disease (ATC: a02b), anxiolytics (ATC: n05b), hyp-
notics and sedatives (ATC: n05c) and antidepressants (ATC: n06a). The
group progestogens and estrogens in combination (G03F) comprises
combined preparations used in the treatment of menopausal symptoms,
menstrual irregularities etc. There is a separate group for hormonal
contraceptives (G03A).
We performed an item-non-responder analysis with respect to age,
sex, SES, diabetes and smoking and compared our two subsets with the
general GHS population. The variables were chosen on the literature on
risk factors for dry eye disease as summarized in the TFOS DEWS II
epidemiology report [4] and iatrogenic report [8].
An explorative analysis was carried out using R version 3.5.1 (2018-
07-02; R Core Team (2018). R: A language and environment for sta-
tistical computing. R Foundation for Statistical Computing, Vienna,
Austria. URL https://www.R-project.org/) was used.
Results
Our sample consisted of 1999 GHS participants: 603 participants
had a 3 min sampling time and 1396 participants had a 5 min sampling
time. On average, our sample was 59 years old and 46% of the study
U. Hampel, et al. The Ocular Surface 18 (2020) 141–147

Table 1
Study characteristics of the Gutenberg Health Study with available data for Schirmer test with topical anesthesia. Abbreviations: OD: right eyes, OS: left eyes, ATC:
Anatomical Therapeutic Chemical Classification - Code.
Variable All (1999) Men (1078) Women (921)

Age [years] 59.0 ± 11.1 59.2 ± 11.1 58.9 ± 11.2

SES 13.46 ± 4.24 14.03 ± 4.23 12.79 ± 4.16
Diabetes (yes) 11.5% (229) 13.8% (148) 8.8% (81)
Smoking (yes) 14.9% (296) 16.5% (177) 13.0% (119)

Tear fluid quantity:

Tear fluid quantity [mm] (OD) (3min) (n = 603) 19.99 ± 10.61 19.52 ± 10.42 20.48 ± 10.81
Tear fluid quantity [mm] (OS) (3min) (n = 590) 19.09 ± 10.56 18.16 ± 10.27 20.07 ± 10.79
Tear fluid quantity [mm] (OD) (5min) (n = 1396) 23.15 ± 9.31 23.02 ± 9.32 23.31 ± 9.30
Tear fluid quantity [mm] (OS) (5min) (n = 1409) 22.93 ± 9.00 22.73 ± 8.92 23.18 ± 9.10

Ophthalmological parameters
Contact lenses (yes) 6.0% (119) 4.8% (52) 7.3% (67)
Contact lenses (hard) (yes) 1.2% (24) 0.7% (7) 1.9% (17)
Contact lenses (soft) (yes) 4.8% (95) 4.2% (45) 5.5% (50)
Spherical equivalent [dpt] (OD) 0 (−1.62/1.62) 0 (−1.62/1.62) 0 (−1.50/1.54)
Spherical equivalent [dpt] (OS) −0.75 (−1.25/-0.38) −0.75 (−1.25/-0.38) −0.62 (−1.25/-0.25)
Visual acuity in logMAR (OD) 0.10 (0/0.22) 0.10 (0/0.20) 0.10 (0/0.22)
Visual acuity in logMAR (OS) 0.10 (0/0.10) 0 (0/0.10) 0.10 (0/0.22)

Systemic medication:
Diuretics (ATC: c03) (yes) 4.5% (90) 4.9% (53) 4.0% (37)
Beta-blockers (ATC: c07) (yes) 18.3% (366) 18.5% (199) 18.1% (167)
Thyroid hormones (ATC: h03aa) (yes) 15.1% (301) 6.4% (69) 25.2% (232)
Non-steroidal anti-inflammatory drugs (ATC: m01a) (yes) 13.5% (270) 9.9% (107) 17.7% (163)
Drugs for peptic ulcer and gastro-oesophageal reflux disease (ATC: a02b) (yes) 13.4% (268) 13.7% (148) 13.0% (120)
Contraceptives (g03a) (yes) 1.8% (45) 0% (0) 3.8% (45)
Estrogens (g03c) (yes) 2.3% (59) 0% (0) 5.0% (59)
Progesterons and estrogens combination drugs (g03f) (yes) 1.3% (34) 0% (0) 2.9% (34)
Antiandrogen therapy (g04c) (yes) 3.5% (89) 6.6% (89) 0% (0)
Anxiolytics (n05b) (yes) 0.8% (20) 0.4% (6) 1.2% (14)
Hypnotics and sedatives (n05c) (yes) 1.5% (37) 1.0% (13) 2.0% (24)
Antidepressants (ATC: n06a) (yes) 6.0% (153) 2.8% (38) 9.7% (115)

Eye medication:
Prostaglandin eye drops (ATC: s01ee) (yes) 1.2% (24) 1.0% (11) 1.4% (13)
Beta blocking agents (ATC: s01ed) (yes) 1.6% (31) 1.7% (18) 1.4% (13)
Artificial tears (ATC: s01xa20) (yes) 6.8% (135) 3.6% (39) 10.4% (96)

participants were female (Table 1), 11.5% had diabetes mellitus and
14.9% smoked.
Median tear fluid quantity for ST-5 was 28.0 mm for right eyes and
27.0 mm for left eyes (Fig. 1). For ST-3, median tear fluid quantity was
13.0 mm in right eyes and 13.0 mm in left eyes (Fig. 2).
8.7% of the right eyes and 8.8% of the left eyes had a Schirmer test
of 10 mm or lower and met after 5 min therefore the clinical cut-off for
dry eye disease. 35.2% of the right eyes and 33.2% of the left eyes had a
measure over 35 mm, meaning that the strips of the ST-5 were com-
pletely watered. The 10 mm wetting length in a ST-5 corresponds to a
6 mm wetting length in ST-3 with anesthesia.
The distribution of Schirmer test score showed the highest fre-
quency between 10 and 15 mm in the 3 min test (ST-3), while in the
5 min (ST-5) test this was around 25 mm with one third of the examined

Fig. 1. Distribution of Schirmer test with topical anesthesia measurements with a duration of 5 min. Data from the population-based Gutenberg Health Study.

143
U. Hampel, et al. The Ocular Surface 18 (2020) 141–147

Fig. 2. Distribution of Schirmer test with topical anesthesia measurements with a duration of 3 min. Data from the population-based Gutenberg Health Study.

eyes reaching the upper limit of 35 mm (Figs. 1 and 2). Table 3

The correlation between right and left eyes was high for both ST-5 Investigation of associations between Schirmer test length (5 min) with topical
(rho = 0.64; p < 0.001) and ST-3 (rho = 0.72; p < 0.001). anesthesia and eye medication. Data from the population-based Gutenberg
There was an association of smaller tear fluid quantity (ST-5) with Health Study.
higher age, male sex, lower socioeconomic status and season, while Topical medication Estimate 95%-CI P-value
none with diabetes mellitus or smoking, as assessed by quantile re-
gression analysis on median (Table 2). Per 10 years of age, ST-5 was Prostaglandin eye drops (s01ee) −3.7 −5.9; −1.4 0.002
Beta blocking agents (s01ed) −2.4 −4.5; −0.39 0.020
1.1 mm shorter, male subjects had 1.8 mm shorter ST-5, and a higher
Artificial tears (s01xa20) −0.57 −1.5; 0.36 0.23
SES was associated to a 0.17 mm increase of ST-5 per SES-point (range
3–21). Compared to winter, ST-5 measure in spring was 2.7 mm longer, Quantile regression analysis for the median using a mixed model. Data are
in summer 3.7 mm longer and in autumn 1.5 mm longer. Similar find- adjusted for sex, age, SES, diabetes, smoking, season, and Schirmer-test ex-
ings were observed when performing linear regression analysis with aminer.
GEE (Supplemental Table 2). In addition, topical medication with
prostaglandins and beta-blocking agents showed an association with association with a reduced ST-5 measure while contraceptives, estrogen
lower tear fluid quantity (Table 3, all estimates are for the median): hormones, antiandrogen therapy, vasodilators and antidepressants did
subjects using topical medication with prostaglandins showed a 3.7 mm not (Table 4). When conducting linear regression with generalized es-
shorter ST-5 measure and those applying beta-blocking agents a 2.4 mm timating equations, thyroid hormones were associated with shorter ST-
shorter ST-5. This was not seen in linear regression analysis with GEE 5 (−0.87 mm) and anxiolytics with a longer ST-5 (3.1 mm) in multi-
(Supplemental Table 3). For systemic medications, oral non-steroidal variable analysis (Supplemental Table 3).
anti-inflammatory drugs, drugs for peptic ulcer and gastro-oesophageal
reflux disease, beta-blockers, contraceptives, antiandrogen therapy,
Discussion
anxiolytics and hypnotics and sedatives showed an association in uni-
variate analysis. In multivariable analysis with adjustment for anthro-
The diagnosis of dry eye disease is still complicated especially in
pometric and environmental factors, oral non-steroidal anti-in-
cases with severe symptoms, but hardly any signs of ocular surface
flammatory drugs (- 0.81 mm for intake of this medication), drugs for
damage. It is crucial to evaluate diagnostic tests even though they are
peptic ulcer and gastro-oesophageal reflux disease (−0.83 mm),
established in the diagnosis of dry eye disease. Surprisingly, distribu-
thyroid hormones (- 0.71 mm), progesterone and estrogen combination
tion of tear fluid quantity measured by a Schirmer test with in the
drugs (−2.6 mm) and hypnotics incl. sedatives (- 3.0 mm) showed an
general population is unknown. To the best of our knowledge, this is the
first paper that reports the distribution and associated factors of tear
Table 2 fluid quantity in the German population-based Gutenberg Health Study
Investigation of associations between Schirmer test length (5 min) with topical to contribute a further characterization of the ST. The Gutenberg Health
anesthesia and anthropometric parameters. Data from the population-based Study excels by a standardized study design and quality control with
Gutenberg Health Study. broad assessment of phenotype information and large sample size.
Estimate 95%-CI p-value Furthermore, the advantage of the Gutenberg health study is the po-
pulation-based sampling. We are able to provide the first normative
Sex (Women) 1.8 1.3; 2.3 < 0.0001 data for the Schirmer test with topical anesthesia in a large population.
Age [10y] −1.1 −1.3; −0.86 < 0.0001
Unexpectedly, there was a high percentage of Schirmer test results
SES 0.17 0.11; 0.23 < 0.0001
Diabetes 0.12 −0.66; 0.91 0.76 with over 35 mm wetting after 5 min so the test results did not show a
Smoking −0.53 −1.2; 0.14 0.12 normal distribution. In a smaller study with only 48 normal participants
Winter (Reference) normal distribution was observed for 5-min ST [22]. The instillation of
Spring 2.7 2.0; 3.5 < 0.0001 the used topical anesthesia might have caused an irritation of the eyes
Summer 3.7 2.8; 4.6 < 0.0001
Autumn 1.5 0.7; 2.2 < 0.0001
and tearing leading to a high number of participants with Schirmer test
results above 35 mm. The Schirmer test was performed over the course
Quantile regression model (for the median) is additional adjusted for Schirmer of one year. Climatic changes due to the different seasons may influence
test examiner. ST results. We found higher ST-5 measures in spring, summer and

144
U. Hampel, et al. The Ocular Surface 18 (2020) 141–147

Table 4
Associations between Schirmer test length (5 min with topical anesthesia) and systemic medication. Data from the population-based Gutenberg Health Study.
Systemic medication Univariate analysis Adjusted analysis

Estimate 95%-CI p-value Estimate 95%-CI p-value

Diuretics (c03) −2.1 −3.3; −0.98 0.00027 −0.81 −2.0; 0.37 0.18
Beta-blockers (c07) −2.2 −2.8; −1.5 < 0.0001 −0.32 −0.99; 0.34 0.34
Thyroid hormones (h03aa) −0.15 −0.83; 0.53 0.67 −0.71 −1.4; −0.01 0.047
NSAIDS (m01a) −0.83 −1.5; −0.13 0.019 −0.91 −1.6; 0.22 0.01
Contraceptives (g03a) 4.1 2.2; 6.0 < 0.0001 0.61 −1.3; 2.6 0.54
Estrogens (g03c) −0.72 −2.3; 0.88 0.38 −1.1 −2.7; 0.53 0.19
Progesterons + Estrogens (g03f) −0.77 −3.0; 1.5 0.50 −2.6 −4.9; −0.34 0.025
Antiandrogen therapy (g04c) −2.0 −3.3; −0.68 0.0030 0.65 −0.67; 2.0 0.33
Drugs for peptic ulcer and gastro-oesophageal reflux disease (a02b) −2.6 −3.3; −1.9 < 0.0001 −0.83 −1.6; −0.10 0.026
Anxiolytics (n05b) 4.2 1.6; 6.9 0.0017 2.4 −0.25; 5.0 0.075
Hypnotics and sedatives (n05c) −2.8 −5.0; −0.70 0.0093 −3.0 −5.1; −0.86 0.006
Antidepressants (n06a) 0.69 −0.33; 1.7 0.18 −0.079 −1.1; 0.96 0.88

autumn compared to winter. We assume that heating in winter might
decrease the humidity causing higher evaporation. Air conditioning
mainly used in summer is not widely used in Germany which might also
reduce ST results. A study employing the Schirmer test (without an-
esthesia) showed no gradual seasonal changes due to climatic factors,
such as temperature, humidity, visibility, and barometric pressure [23].
About 9% of participants had a pathologically low Schirmer test
result with a wetting length ≤10 mm after 5 min. Whether this corre-
lates with a diagnosed dry eye disease cannot be determined with
certainty because dry eye disease was not tested by specific ques-
tionnaires or other tests. We found a correlation of tear fluid quantity
with sex and age. Sex is known to be a risk factor for dry eye disease
with women more likely to suffer from dry eye disease [24]. Male
participants and participants of higher age had shorter wetting of the
Schirmer strip. A shorter wetting of the Schirmer strip is not associated
with the use of artificial tears which may be interpreted as a sign for
successful supplementation of tear fluid. As a limitation, questionnaires
within the GHS only inquired about rheumatism but did not specify
Sjögren syndrome. Sjögren syndrome is an autoimmune-disease af-
fecting tear and saliva producing glands and leading to dry eyes and
mouth [25]. Therefore, a possible correlation between pathological
Schirmer test results and Sjögren syndrome in our population-based
study setup cannot be confirmed. The time between application of to-
pical anesthesia and insertion of the Schirmer paper strip might not
have been sufficient to avoid falsely-high wetting of the paper strip due
to soaking of residual anesthetic eye drop fluid. This might have caused
a ceiling effect in the ST-5 so that a quantile regression analysis had to
be used. To improve this issue in further studies the time between drop
application and insertion of the paper strip should be defined. The
Schirmer test with topical anesthesia should investigate the basal
tearing rate whereas the Schirmer test without topical anesthesia is
performed to measure reflex tearing. Our reported associations of
Schirmer test with topical anesthesia might be different when the test is
performed without topical anesthesia and therefore should be in-
vestigated in future studies. The use of various statistical tools might
lead to different results and have to be interpreted in its context. We
computed quantile regression analysis on the median of study partici-
pants incorporating the correlation structure between both eyes on
person-level, and computed population average effects using a GEE
model, which led to different results that are discussed.
Other studies with smaller sample sizes were able to show that a 5-
min Schirmer test correlates with Schirmer test measures with shorter
durations. A study with 162 healthy participants showed that normal
volunteers have a wetting length ≥10 mm in a 2-min Schirmer test
without topical anesthesia [26]. Another study with 30 DED patients
tested whether measures of a Schirmer test with topical anesthesia of
30 s and 1, 2, 3, 4 and 5 min duration are comparable [27]. They em-
phasized that measures of Schirmer test with topical anesthesia with
shorter duration correlate highly with those measures of a 5-min test
[27]. Similar results were reported by Kashkouli et al. showing a cor-
relation between the Schirmer test without topical anesthesia measures
of 1-min and 5-min test duration within 68 normal and dry eye study
participants [22]. The authors proposed an equation to calculate the 5-
min from 1-min Schirmer test and concluded that the 1-min Schirmer
test is reliable, faster and more comfortable [22]. We found that the cut
off of 10 mm wetting time for a 5-min Schirmer test with topical an-
esthesia corresponds to an 8 mm wetting time in a 3-min Schirmer test
with topical anesthesia. Shorter test durations are probably better tol-
erated by patients. We provide cut-off values for a 3-min Schirmer test
with local anesthesia from a large population based study that might be
a useful tool for clinicians.
We found a difference in Schirmer test results for topical medication
with prostaglandins and beta-blocking agents for the median in our
study sample. It is well-established that topical anti-glaucomatous
medication causes dry eye disease [8,28] and a reduction of tear pro-
duction [10]. This is in accordance with our results. Preservatives and
excipients can aggravate the effect on the ocular surface [8], never-
theless we are not able to distinguish between preserved and un-
preserved eye drops. In addition, using the statistical tool of a GEE
model to estimate the population average effect, we did not find a
statistically significant effect for topical medication.
For systemic medications, oral non-steroidal anti-inflammatory
drugs, thyroid hormones, progesterone and estrogen combination
drugs, drugs for peptic ulcer and gastro-oesophageal reflux disease as
well as hypnotics including sedatives showed an association with lower
ST-5 measures on the median, while statistics on population average
effect showed that thyroid hormones were associated with shorter ST-5
and anxiolytics with a longer ST-5. Schein et al. showed in an elderly
population that dry eye as defined with a questionnaire is associated
with non-steroidal anti-inflammatory drug use, antiulcer agents, an-
xiolytics and benzodiazepines [29]. With respect to the use of the es-
trogen and progesterone combination, the Tear Film and Ocular Surface
Society Dry Eye Workshop II report lists 11 studies showing improve-
ment or no effect of Schirmer test results [30]. In our study, women
under treatment with an estrogen and progesterone combination had
lower Schirmer test measures. This difference might be due to the fact
that the above mentioned studies were interventional studies and not
an observational population-based study, such as in our case
[12,14,31–39]. In addition, there might be a difference between re-
ported symptoms of dry eye disease and diagnostic test results in-
dicating the presence of dry eye disease. Until now, different studies
used different diagnostic criteria of dry eye disease. In our study, we
aim to report only on tear quantity but not on dry eye disease. Further
studies are required to study the reported associations.

145
U. Hampel, et al.
Conclusion
In summary, for the first time, we provide large normative data for
tear fluid quantity measured by a 3 and 5 min Schirmer test with topical
anesthesia. Participants with topical prostaglandins or beta-blockers
showed lower Schirmer test results when analyzing the median.
Furthermore, systemic non-steroidal anti-inflammatory drugs, thyroid
hormones, progesterone and estrogen combination drugs, drugs for
peptic ulcer and gastro-oesophageal reflux disease as well as hypnotics
incl. sedatives reduced Schirmer test measurements in univariate ana-
lysis. Other factors previously reported to be associated with dry eye
disease such as contraceptives, estrogen hormones, antiandrogen
therapy, vasodilators and antidepressants were not associated in mul-
tivariable analysis with adjustment for potential confounding factors.
Consent for publication
Not applicable.
Availability of data and material
The analysis presents clinical data of a large-scale population-based
cohort with ongoing follow-up examinations. This project constitutes a
major scientific effort with high methodological standards and detailed
guidelines for analysis and publication to ensure scientific analyses on
highest level. Therefore, data are not made available for the scientific
community outside the established and controlled workflows and al-
gorithms.
To meet the general idea of verification and reproducibility of sci-
entific findings, we offer access to data at the local database in ac-
cordance with the ethics vote upon request at any time. The GHS
steering committee, which comprises a member of each involved de-
partment and the head of the Gutenberg Health Study (PSW), convenes
once a month. The steering committee decides on internal and external
access of researchers and use of the data and biomaterials based on a
research proposal to be supplied by the researcher. Interested re-
searchers make their requests to the head of the Gutenberg Health
Study (Philipp S. Wild; philipp.wild@unimedizin-mainz.de). More de-
tailed contact information is available at the homepages of the GHS
(www.gutenberghealthstudy.org) or the ophthalmic branch of the GHS
(www.unimedizin-mainz.de/augenklinik/forschung/gutenberg-
gesundheitsstudie.html).
Funding
The Gutenberg Health Study is funded through the government of
Rhineland-Palatinate (“Stiftung Rheinland-Pfalz für Innovation”, contract
AZ 961-386261/733), the research programs “Wissen schafft Zukunft” and
“Center for Translational Vascular Biology (CTVB)” of the Johannes
Gutenberg-University of Mainz, and its contracts with Boehringer
Ingelheim and PHILIPS Medical Systems, including unrestricted grants for
the Gutenberg Health Study. Philipp S. Wild is funded by the Federal
Ministry of Education and Research (BMBF 01EO1503), and he is the PI of
the German Center for Cardiovascular Research (DZHK).
Schuster AK holds the professorship for ophthalmic healthcare re-
search endowed by „Stiftung Auge” and financed by „Deutsche
Ophthalmologische Gesellschaft” and „Berufsverband der Augenärzte
Deutschlands e.V.“ He received research funding from Bayer Vital,
Novartis and Heidelberg Engineering.
Authors' contributions
Conceived and designed the study: UH, AKS, SN, KJL, TM, PSW, MB,
IS, NP; analyzed and interpreted the data: AS, UH, AKS, NP; wrote the
paper: AKS, UH, SN, NP; all authors revised the manuscript critically
and approved the final version.
146
The Ocular Surface 18 (2020) 141–147
Declaration of competing interest
The authors declare that they have no competing interests.
Acknowledgements
We thank all study participants for their willingness to provide data
for this research project, and we are indebted to all coworkers for their
enthusiastic commitment.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.jtos.2019.11.003.
Quantile regression analysis for the median using a mixed model.
Crude associations without adjustment. Adjusted analysis incorporated
sex, age, SES, diabetes, smoking, season, Schirmer-test examiner and
eye drops (prostaglandin, beta blocking agents, artificial tears) as
confounding factors.
References
[1] Wolffsohn JS, Arita R, Chalmers R, Djalilian A, Dogru M, Dumbleton K, et al. TFOS
DEWS II diagnostic methodology report. Ocul Surf 2017;15:539–74.
[2] Craig JP, Nichols KK, Akpek EK, Caffery B, Dua HS, Joo CK, et al. TFOS DEWS II
definition and classification report. Ocul Surf 2017;15:276–83.
[3] Roy NS, Wei Y, Kuklinski E, Asbell PA. The growing need for validated biomarkers
and endpoints for dry eye clinical research. Investig Ophthalmol Vis Sci
2017;58:BIO1–19.
[4] Stapleton F, Alves M, Bunya VY, Jalbert I, Lekhanont K, Malet F, et al. TFOS DEWS
II epidemiology report. Ocul Surf 2017;15:334–65.
[5] Hampel U, Garreis F, Paulsen F. Epidemiologie. Bremen: UNI-MED; 2015.
[6] Bron AJ, de Paiva CS, Chauhan SK, Bonini S, Gabison EE, Jain S, et al. TFOS DEWS II
pathophysiology report. Ocul Surf 2017;15:438–510.
[7] Schirmer O. Studien zur Physiologie und Pathologie der Tränenabsonderung und
Tränenabfuhr. Albr Graefe's Arch Ophthalmol 1903;56:197–291.
[8] Gomes JAP, Azar DT, Baudouin C, Efron N, Hirayama M, Horwath-Winter J, et al.
TFOS DEWS II iatrogenic report. Ocul Surf 2017;15:511–38.
[9] Garcia-Feijoo J, Sampaolesi JR. A multicenter evaluation of ocular surface disease
prevalence in patients with glaucoma. Clin Ophthalmol 2012;6:441–6.
[10] Leung EW, Medeiros FA, Weinreb RN. Prevalence of ocular surface disease in
glaucoma patients. J Glaucoma 2008;17:350–5.
[11] Schaumberg DA, Buring JE, Sullivan DA, Dana MR. Hormone replacement therapy
and dry eye syndrome. J Am Med Assoc 2001;286:2114–9.
[12] Affinito P, Di Spiezio Sardo A, Di Carlo C, Sammartino A, Tommaselli GA, Bifulco G,
et al. Effects of hormone replacement therapy on ocular function in postmenopause.
Menopause 2003;10:482–7.
[13] Schaumberg DA, Dana R, Buring JE, Sullivan DA. Prevalence of dry eye disease
among US men: estimates from the Physicians' Health Studies. Arch Ophthalmol
2009;127:763–8.
[14] Altintas O, Caglar Y, Yuksel N, Demirci A, Karabas L. The effects of menopause and
hormone replacement therapy on quality and quantity of tear, intraocular pressure
and ocular blood flow. Ophthalmologica Journal international d'ophtalmologie Int
J Ophthalmol Zeitschrift fur Augenheilkunde 2004;218:120–9.
[15] Mathers WD, Dolney AM, Kraemer D. The effect of hormone replacement therapy
on the symptoms and physiologic parameters of dry eye. Adv Exp Med Biol
2002;506:1017–22.
[16] Wild PS, Zeller T, Beutel M, Blettner M, Dugi KA, Lackner KJ, et al. The gutenberg
health study. Bundesgesundheitsblatt - Gesundheitsforsch - Gesundheitsschutz
2012;55:824–9.
[17] Hohn R, Kottler U, Peto T, Blettner M, Munzel T, Blankenberg S, et al. The oph-
thalmic branch of the gutenberg health study: study design, cohort profile and self-
reported diseases. PLoS One 2015;10. e0120476.
[18] Li N, Deng X-G, He M-F. Comparison of the Schirmer I test with and without topical
anesthesia for diagnosing dry eye. Int J Ophthalmol 2012;5:478–81.
[19] Lampert T, Kroll LE, Muters S, Stolzenberg H. [Measurement of the socioeconomic
status within the German health update 2009 (GEDA)]. Bundesgesundheitsblatt -
Gesundheitsforsch - Gesundheitsschutz 2013;56:131–43.
[20] Holladay JT. Proper method for calculating average visual acuity. J Refract Surg
1997;13:388–91.
[21] Bowling B. Kanski's clinical ophthalmology: a systematic approach. eighth ed.
Saunders Ltd./Elsevier; 2015.
[22] Kashkouli MB, Pakdel F, Amani A, Asefi M, Aghai GH, Falavarjani KG. A modified
Schirmer test in dry eye and normal subjects: open versus closed eye and 1-minute
versus 5-minute tests. Cornea 2010;29:384–7.
[23] Shapiro A, Merin S. Schirmer test and break-up time of tear film in normal subjects.
Am J Ophthalmol 1979;88:752–7.
[24] Pult H, Purslow C, Murphy PJ. The relationship between clinical signs and dry eye
symptoms. Eye 2011;25:502–10.
U. Hampel, et al.
[25] Vitali C, Bombardieri S, Jonsson R, Moutsopoulos HM, Alexander EL, Carsons SE,
et al. Classification criteria for Sjogren's syndrome: a revised version of the
European criteria proposed by the American-European Consensus Group. Ann
Rheum Dis 2002;61:554–8.
[26] Karampatakis V, Karamitsos A, Skriapa A, Pastiadis G. Comparison between normal
values of 2- and 5-minute Schirmer test without anesthesia. Cornea
2010;29:497–501.
[27] Bawazeer AM, Hodge WG. One-minute schirmer test with anesthesia. Cornea
2003;22:285–7.
[28] Pisella PJ, Pouliquen P, Baudouin C. Prevalence of ocular symptoms and signs with
preserved and preservative free glaucoma medication. Br J Ophthalmol
2002;86:418–23.
[29] Schein OD, Hochberg MC, Muñoz B, et al. Dry eye and dry mouth in the elderly: a
population-based assessment. Arch Intern Med 1999;159:1359–63.
[30] Sullivan DA, Rocha EM, Aragona P, Clayton JA, Ding J, Golebiowski B, et al. TFOS
DEWS II sex, gender, and hormones report. Ocul Surf 2017;15:284–333.
[31] Akramian J, Wedrich A, Nepp J, Sator M. Estrogen therapy in keratoconjunctivitis
sicca. Adv Exp Med Biol 1998;438:1005–9.
[32] Guaschino S, Grimaldi E, Sartore A, Mugittu R, Mangino F, Bortoli P, et al. Visual
function in menopause: the role of hormone replacement therapy. Menopause
2003;10:53–7.
[33] Coksuer H, Ozcura F, Oghan F, Haliloglu B, Coksuer C. Effects of estradiol-
147
The Ocular Surface 18 (2020) 141–147
drospirenone on ocular and nasal functions in postmenopausal women. Climacteric.
J Int Menopause Soc 2011;14:482–7.
[34] Moon JH, Jung JW, Shin KH, Paik HJ. Effect of hormone replacement therapy on
dry eye syndrome in postmenopausal women: a prospective study. J Korean
Ophthalmol Soc 2010;51:175–9.
[35] Pelit A, Bagis T, Kayaselcuk F, Dursun D, Akova Y, Aydin P. Tear function tests and
conjunctival impression cytology before and after hormone replacement therapy in
postmenopausal women. Investig Ophthalmol Vis Sci 2002;43:104.
[36] Kuscu NK, Toprak AB, Vatansever S, Koyuncu FM, Guler C. Tear function changes of
postmenopausal women in response to hormone replacement therapy. Maturitas
2003;44:63–8.
[37] Piwkumsribonruang N, Somboonporn W, Luanratanakorn P, Kaewrudee S,
Tharnprisan P, Soontrapa S. Effectiveness of hormone therapy for treating dry eye
syndrome in postmenopausal women: a randomized trial. J Med Assoc Thail =
Chotmaihet thangphaet 2010;93:647–52.
[38] Erdem U, Ozdegirmenci O, Sobaci E, Sobaci G, Göktolga U, Dagli S. Dry eye in post-
menopausal women using hormone replacement therapy. Maturitas
2007;56:257–62.
[39] Taner P, Akarsu C, Atasoy P, Bayram M, Ergin A. The effects of hormone replace-
ment therapy on ocular surface and tear function tests in postmenopausal women.
Ophthalmologica Journal international d'ophtalmologie Int J Ophthalmol
Zeitschrift fur Augenheilkunde 2004;218:257–9.