Forensic Science International 266 (2016) 338–342

Contents lists available at ScienceDirect

Forensic Science International

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

Elevation of post mortem vitreous humour sodium and chloride levels

can be used as a reliable test in cases of suspected salt water drowning
when the immersion times are less than one hour
J. Garland a, R. Tse b,*, C. Oldmeadow a,c, J. Attia a,c, S. Anne d, A.D. Cala b
a
School of Medicine and Public Health, Faculty of Health and Medicine, University of Newcastle, Callaghan, New South Wales, Australia
b
Department of Forensic Medicine, Newcastle, Forensic & Analytical Science Service (FASS), NSW Health Pathology, John Hunter Hospital, Newcastle, New
South Wales, Australia
c
Public Health Stream, Hunter Medical Research Institute, Newcastle, New South Wales, Australia
d
Blacktown Hospital, New South Wales, Australia

A R T I C L E I N F O A B S T R A C T

Article history: Background: Previous studies in salt water drowning deaths (SWD) demonstrated an observable
Received 19 March 2016 elevation of post mortem vitreous sodium and chloride (PMVSC) levels. It remains unclear what the
Received in revised form 14 May 2016 underlying mechanism responsible for this change is: whether this is due to rapid electrolyte changes
Accepted 3 June 2016
from salt water inhalation/ingestion during drowning or from electrolyte diffusion and/or osmosis
Available online 16 June 2016
across the outer coats of the eyeballs during immersion. A recent animal study using sacrificed bovine
eyeballs immersed in salt water demonstrated no significant elevations in PMVSC when immersed for
Keywords:
less than one hour. Assuming similar physical properties between human and bovine, we extrapolate
Salt water
that an elevation in PMVSC in SWD with immersion times of less than one hour (SWD-1) would not be
Sea water
Drowning from immersion, but from drowning.
Post mortem vitreous humour Aim: Investigate whether there is an elevation in PMVSC in SWD-1.
Sodium Methods: Retrospective study comparing PMVSC in SWD-1 with controls from 2012 to 2015 inclusive.
chloride Results: PMVSC in SWD-1 was significantly elevated compared with controls. A PMVSC of 259 mmol/L
has a sensitivity, specificity and likelihood ratio of 0.9, 0.9 and 7.6, respectively.
Conclusion: The elevation in PMVSC in SWD-1 is due to drowning. A PMVSC of 259 mmol/L and above is a
reliable ancillary test in diagnosing drowning in bodies immersed in salt water for less than one hour.
Crown Copyright ß 2016 Published by Elsevier Ireland Ltd. All rights reserved.

drowning can be challenging due to the absence of specific post

1. Introduction
Drowning is the process of experiencing respiratory im-
pairment from submersion/immersion in a liquid [1]. During
drowning, large volumes of liquid may be inhaled and/or ingested,
and death can occur within minutes. The mechanism of death is
thought to be by asphyxiation/hypoxia [2]. Making a diagnosis of
Abbreviations: DNRD, immersion deaths not related to drowning but recovered
from salt water; DNRD-1, immersion deaths not related to drowning but recovered
from salt water; (DNRD), with immersion time of less than one hour; PMVSC, post
mortem vitreous sodium and chloride; SWD, salt water drowning death; SWD-1,
salt water drowning death (SWD) with immersion time of less than one hour.
* Corresponding author at: Newcastle Department of Forensic Medicine, John
Hunter Hospital, Lookout Road, New Lambton Heights, Newcastle, NSW 2305,
Australia. Tel.: +61 2 4922 3700; fax: +61 2 4922 3735.
E-mail address: rexson.tse@hnehealth.nsw.gov.au (R. Tse).
mortem findings and unreliable ancillary tests [2].
The authors have previously reported an elevation in post
mortem vitreous humour sodium and chloride levels (PMVSC) in salt
water drowning deaths (SWD) compared with immersion deaths
not related to drowning but recovered from water (DNRD) [3]. It is
hypothesized that the PMVSC elevation seen in SWD is related to
blood electrolyte changes secondary to large amounts of salt water
inhalation and/or ingestion which results in PMVSC elevation [3].
A major confounding factor causing PMVSC elevation is
electrolyte diffusion and/or osmosis across the outer coats of the
eyeball during salt water immersion. Although prolonged salt
water immersion causes PMVSC elevation, previous studies [4–6]
did not investigate PMVSC with shorter immersion times. Studying
PMVSC changes with shorter immersion times is critical as death
due to drowning can occur within minutes. A recent study carried
out by the authors using sacrificed bovine eyeballs to investigate

http://dx.doi.org/10.1016/j.forsciint.2016.06.001
0379-0738/Crown Copyright ß 2016 Published by Elsevier Ireland Ltd. All rights reserved.
 J. Garland et al. / Forensic Science International 266 (2016) 338–342
the changes in PMVSC with shorter immersion times in salt water
did not show any significant elevation in PMVSC when the eyeballs
were immersed for less than one hour [7]. With comparable
permeability between human and bovine eyes, the authors
concluded that any PMVSC elevation in humans with an immersion
time of less than one hour would not be from immersion [7].
It is therefore reasonable to investigate whether there is an
elevation in PMVSC in SWD with immersion times of less than one
hour (SWD-1) in humans. An elevation in PMVSC in SWD-1 would
not be from immersion, but from drowning, and would support the
hypothesis that during salt water drowning, the inhaled and/or
ingested salt water causes electrolyte changes in blood which are
reflected in an elevation in PMVSC.
This study compared PMVSC in 24 cases of witnessed or
clinically confirmed SWD-1 with 96 controls (non-immersion
deaths) during 2012–2015 (inclusive).
2. Method and material
A retrospective study comparing the PMVSC in SWD-1 and
controls during 2012–2015 (inclusive) was performed at the
Department of Forensic Medicine, Newcastle, Forensic & Analytical
Science Service, New South Wales Health Pathology. The specified
study time period was chosen deliberately as routine PMVSC analysis
in bodies recovered from water was not performed prior to that time.
All witnessed or clinically confirmed SWD-1 deaths admitted to
the department during the study period were selected. Twenty-
four SWD-1 cases were identified by selecting all SWD deaths in
which there was a clear documentation of immersion time of less
than one hour in the referring police report. The age, gender,
location of death, circumstances relating to the death, time
between death and autopsy, and vitreous humour sodium (Na) and
chloride (Cl) levels were recorded.
A previous study by the authors used DNRD as controls
[3]. Since then, we have performed an experimental study
immersing sacrificed bovine eyeballs in salt water and shown
that PMVSC was unchanged within the one hour time window
[7]. Therefore, for this study DNRD data were no longer needed as
controls. Controls (non-immersion deaths) were selected sequen-
tially in cases for which vitreous humour Na and Cl were analyzed
at post mortem during the same study period in the same
department. Coronial cases such as sudden unexpected infant
deaths, suspicious deaths, homicides, and bodies which were
decomposed, incinerated or recovered from water were excluded.
Conditions which possibly have an impact on the vitreous humour
Na and Cl levels were also excluded (such as end stage liver disease,
hyponatraemia, and diabetic ketoacidosis or any other with
suspected metabolic derangement). The number of controls was
chosen to be four times the number of SWD-1 (96 cases) which
gives the study 80% power to detect a 6.5 mmol/L difference in
PMVSC at the 5% significance threshold (assuming a standard
deviation of 10 mmol/L [3]). The age, gender, cause of death, time
between death and autopsy, and vitreous humour Na and chloride
Cl levels were recorded.
All vitreous humour samples were analyzed for Na and Cl by a
local accredited biochemistry lab (Pathology North, John Hunter
Hospital, New Lambton Heights, New South Wales, Australia)
using an ion selective electrode analyzer (Abbott chemistry
analyzer C16000/C8000 or an Olympus 5400 Auto-analyzer). In
both SWD-1 and controls, the PMVSC was calculated by adding
vitreous humour Na and Cl levels.
2.1. Statistical methods
SAS 9.4 (SAS Institute, Cary, North Carolina, USA) was used for
analysis.
339
Continuous variables were described using means and standard
deviations. Medians with minima and maxima were presented.
Categorical variables were described using counts and percentages
for cases and controls. Differences between cases and controls in
the age, time between death and autopsy, vitreous humour Na,
vitreous humour Cl, and PMVSC were assessed using Students t-
test, and between-group differences in gender was assessed using
the Pearson Chi-square test.
Logistic regression models were used to assess the predictive
ability of the vitreous humour Na, vitreous humour Cl, and PMVSC
measures. We first created a baseline logistic regression model
including variables that differed between cases and non-cases. We
then assessed the incremental improvement in fit (according to the
likelihood ratio test) and discriminative ability (according to area
under the Receiver Operating Characteristic) by separately adding
vitreous humour Na, vitreous humour Cl, and PMVSC to the
baseline model. Optimal cut-points are presented for all measures,
where the value is chosen to minimize the Euclidean distance to
the perfect predictor (sensitivity of one and false positive rate of
zero). Sensitivity, specificity, positive and likelihood ratios are also
presented for the optimum cut-points.
3. Results
In the 24 cases of SWD-1 identified during the period, the
average age was 40 years (range: 7–70) with a male predominance
(male to female ratio of 19:5). The average time between death and
autopsy was 2.9 days (range: 1–6 days). The locations of the deaths
were from local beaches from our catchment area, being along the
east coast of Australia facing the Tasman Sea, a part of the Pacific
Ocean. The most common reason for entering the sea was
recreational swimming (14 cases), followed by accidentally falling
or being washed off rocks into the sea when fishing (4 cases),
diving (2 cases of SCUBA diving; 1 case of non-SCUBA diving), and
surfing (2 cases).
In the 96 sequential controls, we excluded certain known
conditions such as diabetic ketoacidosis, end stage liver disease,
insulin toxicity, pancreatitis, and hyponatraemia. The average age
was 51 years (range: 7–98) with a male predominance (male to
female ratio of 58:38). The average time between death and
autopsy was 3.3 days (range: 1–6 days).
The cases differed significantly to the controls in PMVSC and
vitreous Na and Cl (Table 1). Vitreous Na, vitreous Cl, and PMVSC
were very good discriminators of the outcome (cause of death was
drowning), with AUCs of 0.895, 0.905 and 0.908 (95% CIs: [0.83,
0.96], [0.84, 0.97], [0.84, 0.97]) for vitreous Na, vitreous Cl and
PMVSC, respectively (Table 2). There was substantial improvement
in model fit and discrimination from a baseline model including
only age, gender and time from drowning to autopsy (LR p-
value = <0.0001) (Table 3, Fig. 1). Of all the measures, PMVSC had
comparable sensitivity (0.8) and the greatest specificity (0.9) and
positive likelihood ratio (7.6).
4. Discussion
The definition of drowning has changed over the years [8], and
is now defined by the World Health Organization (WHO) as the
process of experiencing respiratory impairment from submersion/
immersion in liquid [1]. Drowning as a cause of death in Australia
has an incidence of approximately 1.5–2 deaths per 100,000 [9]
and accounts for approximately 19% of child injury deaths [10].
The pathophysiology of drowning is as a result of alveolar
collapse secondary to inhalation of liquid, leading to ventilation-
perfusion mismatch and eventually asphyxia/hypoxia [2]. Al-
though commonly encountered, diagnosing drowning in bodies
340 J. Garland et al. / Forensic Science International 266 (2016) 338–342

Table 1
Summary of salt water drowning deaths with immersion time of less than one hour (SWD-1) and control.

Variable Level Deaths P

Case (n = 24) Control (n = 96)

Demographics Gender Female 5 (21%) 38 (40%) 0.0866

Male 19 (79%) 58 (60%)

Age Mean (SD) 40 (18) 51 (16) 0.0036

Days to P-M Mean (SD) 2.9 (1.2) 3.3 (1.5) 0.0994

Vitreous humour tests Na Mean (SD) 146 (9) 129 (10) <0.0001
Median (min, max) 148 (128, 159) 129 (110, 160)

Cl Mean (SD) 125 (8) 109 (10) <0.0001

Median (min, max) 125 (108, 143) 109 (89, 140)

PMVSC Mean (SD) 271 (16) 238 (18) <0.0001

Median (min, max) 275 (236, 298) 237 (201, 300)

Table 2
Optimal cut-points for vitreous humour Na, vitreous humour Cl and PMVSC derived from the unadjusted logistic regression models.

Measure Cut-point Sensitivity Specificity Positive LR Area under curve (Lower, Upper)

Sodium 136 0.83 0.78 3.8 0.90 (0.83, 0.96)

Chloride 119 0.83 0.83 5.0 0.90 (0.84, 0.97)
PMVSC 259 0.79 0.90 7.6 0.91 (0.84, 0.97)

Table 3
Improvement in predictive ability for vitreous Na, vitreous Cl and PMVSC derived from the logistic regression models.

Model LR test p-value Area under curve (Lower, Upper)

(1) Age + gender + time between death and autopsy Ref 0.72 (0.61, 0.82)
(2) Vitreous humour Na + model 1 <0.0001 0.91 (0.83, 0.98)
(3) Vitreous humour Cl + model 1 <0.0001 0.93 (0.86, 0.98)
(4) PMVSC + model 1 <0.0001 0.93 (0.86, 0.99)

[(Fig._1)TD$IG]

Fig. 1. Receiver Operating Characteristic (ROC) curves for vitreous humour Na (Na), vitreous humour Cl (Cl) and post mortem vitreous humour sodium and chloride PMVSC
(adjusted for age, gender, and time between death and autopsy (days-P)), overlaid with curve from comparison models.

recovered from water can be difficult due to the non-specific serum electrolytes, but they are unreliable due to post mortem
findings at autopsy, and often relies on eyewitnesses [2]. diffusion into organs and decomposition [11]. Another measure is
Attempts have been made devising a reliable ancillary test for the diatom test which is based upon the isolation of particular
the diagnosis of drowning. There are studies describing changes in algae in the body. This test is neither sensitive nor specific. It is also
 J. Garland et al. / Forensic Science International 266 (2016) 338–342
prone to contamination and requires the assistance of an algal
specialist [11].
A recent study demonstrated a significant elevation in vitreous
humour sodium concentrations in salt water/seawater drowning
compared to freshwater [5]. This raised the possibility of using
vitreous humour electrolytes in aiding the diagnosis of drowning
as vitreous humour is less prone to post mortem electrolyte
changes compared with blood.
Subsequently, in regards to salt water drowning, a study by the
authors demonstrated that an elevated PMVSC was useful in
differentiating SWD from immersion deaths not related to
drowning but recovered from salt water (DNRD) [3]. The proposed
mechanism for this was that the inhalation and ingestion of salt
water in SWD causes a rapid shift in plasma electrolyte
concentrations, a previously documented phenomenon [12], and
is reflected in an elevation of PMVSC.
A potential confounder that could affect the utility of PMVSC as
a test for SWD is the passive diffusion of electrolytes and osmosis
across the surface of the eyeball during immersion in salt water
[4]. A recent study conducted by the authors [7] investigated the
changes in PMVSC over time by immersing scarified bovine
eyeballs in salt water and measuring changes in PMVSC compared
to controls. The study demonstrated no statistically significant
elevation in bovine PMVSC when immersed in salt water for less
than one hour. The authors concluded that, assuming the physical
properties are similar in human and bovine, any elevation in
PMVSC in bodies immersed for less than one hour would not be
from immersion.
The current study compared PMVSC (and also vitreous Na and
Cl) between 24 cases of SWD-1 and 96 controls. There was no
significant difference between time of death and autopsy between
the groups and as such was not a confounding factor.
This study clearly demonstrated a significant PMVSC (and also
vitreous Na and Cl) elevation in SWD-1 compared with controls.
The elevation in PMVSC (and also vitreous Na and Cl) in SWD-1 was
due to drowning and not immersion. This finding supports the
authors’ hypothesis in which the inhalation and ingestion of salt
water during drowning causes a rapid shift in plasma electrolyte
concentrations which then results in a significant elevation of
PMVSC. This adds further understanding to vitreous humour
electrolyte changes in bodies immersed in water. Based on this
study and previous studies by the authors [3,7], in salt water
drowning, PMVSC appears to become elevated initially, plateaus
for a period of at least one hour and subsequently elevates due to
immersion.
Using the data generated in our study, optimal cut-points
with corresponding sensitivity, specificity and likelihood ratio
(LR) for vitreous humour Na, vitreous humour Cl and PMVSC
were 136 mmol/L (sensitivity = 0.8; specificity = 0.8; LR = 3.8),
119 mmol/L (sensitivity = 0.83; specificity = 0.83; LR = 5.0) and
259 mmol/L (sensitivity = 0.8; specificity = 0.9; LR = 7.6), respec-
tively.
5. Limitations
5.1. Bovine eyeballs as surrogate for human eyeballs in salt water
immersion studies
The present study extended previous studies by the authors
[3,7]. One of the assumptions was that immersion in salt water for
one hour would not increase PMVSC in humans, which was based
on data from immersing sacrificed bovine eyeballs in salt water [7].
Although having different physical properties, bovine, porcine
and rabbit eyes are commonly used as surrogates for human ocular
permeability studies [4,13–15]. Anatomically, the conjunctiva
overlies the cornea and parts of the sclera anteriorly, with only the
341
sclera covering the posterior aspect of the eye. Common across
species, the permeability varies amongst different components of
the eye [15,16]. The scleral permeability is about double that of the
cornea as well as the combined layer of conjunctiva plus sclera
covering the anterior eye [15]. The previous study used enucleated
bovine eyeballs, having large areas of exposed sclera, immersed in
salt water rather than orbits in situ which would overestimate the
overall permeability.
Another factor is that ocular permeability is inversely
proportional to molecular weight and size [15,17]. The difference
in permeability between human and bovine eyeballs decreases as
the molecular weight of the solute decreases [17]. Although there
is no single study examining the permeability to small elemental
molecules between these species, the permeability to Na, and most
probably Cl and water, are considered similar in human and bovine
[17]. Hence, it is reasonable that the enucleated bovine eyeball is a
robust surrogate for human in salt water immersion studies.
Based on the above factors, it is likely that human PMVSC
remains unaffected by immersion for at least one hour if not
longer.
5.2. Variable Na and Cl concentration in salt water
The results generated from our study were based on cases
retrieved from salt water from our catchment area. Salt water Na
and Cl concentration around the world is variable, but is generally
three to five times higher compared with plasma and vitreous
humour. Although the authors believe the impact of this variability
is minimal, extrapolation of our results with other salt water
environments should be done with a degree of caution. The authors
suggest that when interpreting PMVSC, salt water Na and Cl
concentrations in the vicinity of the body should also be analyzed.
5.3. Study population
Although the epidemiologic data in the sample (age, gender,
time between death and autopsy) had a minimal contribution to
the observed elevation of PMVSC in SWD-1, extrapolation of our
data beyond these parameters should be done cautiously.
5.4. Control selection
Our previous study used DNRD as a control to study SWD [3] as
it was previously unknown whether PMVSC would increase from
salt water drowning. Since then, the authors have performed an
experimental study using sacrificed bovine eyeballs immersed in
salt water and showed that PMVSC is unaffected by diffusion/
osmosis within the one hour time window [7]. Therefore, DNRD as
controls were not required in this study.
DNRD with immersion time of less than one hour (DNRD-1)
would be a good comparison with SWD-1. This was limited by the
number of actual cases. During the study period, nine cases of
DNRD were identified in which only four cases were DNRD-1. The
cause of death in these cases was from fatal trauma sustained in
salt water (male to female ratio of 2:2; age range: 11–50 years;
time between death and autopsy: 2–3 days). PMVSC in these cases
were 226, 244, 257, and 258 mmol/L, rendering an average of
246.25 mmol/L. The PMVSC were comparable to the controls and
lower than the SWD-1 group, which was in keeping with the study
results.
5.5. Unknown PMVSC prior to death
The baseline vitreous Na and Cl level prior to death is assumed
to be similar between SWD-1 and controls, although this cannot
ever be confirmed.
342 J. Garland et al. / Forensic Science International 266 (2016) 338–342
5.6. The less than one hour time period of immersion
There is a significant number of salt water drowning deaths in
which the immersion time is unknown. Our data is limited to
bodies with an immersion time of less than one hour. This chosen
time period is based on a previous animal study by the authors and
the reasonable assumption that bovine eyeballs are robust
surrogate for human for human. Further studies with more
frequent intervals beyond one hour may demonstrate that PMVSC
may be an effective test for drowning with immersion times of
greater than one hour.
5.7. Limitations as a clinical test
Similar to any investigation in medicine, PMVSC should be
interpreted in conjunction with the clinical history and autopsy
findings. PMVSC is a reliable ancillary test to bring greater certainty
of a diagnosis of salt water drowning.
6. Conclusion
Significant elevations in post mortem vitreous humour sodium
and chloride (PMVSC) levels occur in salt water drowning deaths
with immersion times of less than one hour. This elevation is due to
drowning and not immersion.
PMVSC is a reliable ancillary test which can be used in
suspected salt water drowning deaths. A PMVSC of 259 mmol/L has
a sensitivity, specificity and likelihood ratio of 0.8, 0.9 and 7.6,
respectively, making it a good predictor for diagnosing salt water
drowning when the immersion time is less than one hour.
Ethics
The study was approved by the Hunter New England Human
Research Ethics Committee (Authorization number: AU201602-10).
Acknowledgement
The authors would like to thank Meghan Laver in proof-reading
the manuscript.
This study was funded by the Royal College of Pathologists of
Australasia (RCPA).
References
[1] E.F. Van Beeck, C.M. Branche, D. Szpilman, J.H. Modell, J.J. Bierens, A new
definition of drowning: towards documentation and prevention of a global
public health problem, Bull. World Health Organ. 83 (November (11)) (2005)
853–856.
[2] M.J. Shkrum, D.A. Ramsay, Bodies recovered from water, in: M.J. Shkrum, Ramsay
DA (Eds.), Forensic Pathology of Trauma: Common Problems for the Pathologist,
Humana Press Inc., Totowa, NJ, 2007, pp. 243–293.
[3] A.D. Cala, R. Vilain, R. Tse, Elevated postmortem vitreous sodium and chloride
levels distinguish saltwater drowning (SWD) deaths from immersion deaths not
related to drowning but recovered from saltwater (DNRD), Am. J. Forensic Med.
Pathol. 34 (June (2)) (2013) 133–138.
[4] W.Q. Sturner, A. Balko, J. Sullivan, Magnesium and other electrolytes in bovine
eyeballs immersed in sea water and other fluids, Forensic Sci. 8 (September–
October (2)) (1976) 139–150.
[5] R.W. Byard, G. Summersides, Vitreous humor sodium levels in immersion deaths,
J. Forensic Sci. 56 (May (3)) (2011) 643–644.
[6] J.G. Farmer, F. Benomran, A.A. Watson, W.A. Harland, Magnesium, potassium,
sodium and calcium in post-mortem vitreous humour from humans, Forensic Sci.
Int. 27 (January (1)) (1985) 1–13.
[7] S. Anne, R. Tse, C. Oldmeadow, J. Attia, A.D. Cala, Immersion of bovine eyeballs
after one hour in seawater does not result in elevation of post mortem vitreous
humor sodium and chloride levels, Am. J. Forensic Med. Pathol. 37 (Jun (2)) (2016)
108–111.
[8] J.H. Modell, Drown versus near-drown: a discussion of definitions, Crit. Care Med.
9 (April (4)) (1981) 351–352.
[9] Australian Bureau of Statistics, Australian Social Trends, cat. no. 4102.0, 2000
accessed 5 March 2016, <http://www.abs.gov.au/AUSSTATS/abs@.nsf/
2f762f95845417aeca25706c00834efa/
050bd63790a1c58bca2570ec000e2f5b!OpenDocument>..
[10] Australian Bureau of Statistics, Year Book Australia, cat. no. 1301.0, 2006 accessed
5 March 2016, <http://www.abs.gov.au/ausstats/abs@.nsf/featurearticlesbytitle/
052E93D885D061E7CA2570DE000DA2C1?OpenDocument>.
[11] M.H. Piette, E.A. De Letter, Drowning: still a difficult autopsy diagnosis, Forensic
Sci. Int. 163 (1–2) (2006) 1–9.
[12] R.J. Ellis, Severe hypernatremia from sea water ingestion during near-drowning in
a hurricane, West J. Med. 167 (December (6)) (1997) 430–433.
[13] P.W. Morrison, V.V. Khutoryanskiy, Enhancement in corneal permeability of
riboflavin using calcium sequestering compounds, Int. J. Pharm. 472 (September
(1–2)) (2014) 56–64.
[14] R.S. Kadam, U.B. Kompella, Cassette analysis of eight beta-blockers in bovine eye
sclera, choroid-RPE, retina, and vitreous by liquid chromatography-tandem mass
spectrometry, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 877 (January (3))
(2009) 253–260.
[15] C. Loch, S. Zakelj, A. Kristl, S. Nagel, R. Guthoff, W. Weitschies, A. Seidlitz,
Determination of permeability coefficients of ophthalmic drugs through different
layers of porcine, rabbit and bovine eyes, Eur. J. Pharm. Sci. 47 (August (1)) (2012)
131–138.
[16] K.M. Hämäläinen, K. Kananen, S. Auriola, K. Kontturi, A. Urtti, Characterization of
paracellular and aqueous penetration routes in cornea, conjunctiva, and sclera,
Invest. Ophthalmol. Vis. Sci. 38 (March (3)) (1997) 627–634.
[17] T.W. Olsen, H.F. Edelhauser, J.I. Lim, D.H. Geroski, Human scleral permeability.
Effects of age, cryotherapy, transscleral diode laser, and surgical thinning, Invest.
Ophthalmol. Vis. Sci. 36 (August (9)) (1995) 1893–1903.