|Microdialysis; Hydrofluoric acid diffusion; Fluoride
permeation; in situ calibration of fluoride ions; Fluoride ion recovery
|HF: Hydrofluoric acid; MD: Microdialysis; %
Recovery: Percent recovery by gain; F electrode: Fluoride ion selective
electrode; Ca electrode: Calcium ion selective electrode; % CV:
Coefficient of variation
|Hydrofluoric acid (HF) is one of the strongest inorganic acids,
used for manufacture of variety of chemicals and for industrial glassetching,
glass-polishing and metal cleaning. Accidental exposure to
this acid produces serious injury and extended disability .
|Microdialysis technique for monitoring drug availability in tissues such as brain, heart, lung, skin or eye is well known and established [2-
5]. This sampling technique is based upon passive diffusion of molecules
through a semi permeable membrane conforming to the Fick’s law
[6,7]. MD probe can be placed nearly in any tissue, organ or fluid and
the collected dialysate samples can be analyzed by a suitable analytical
technique such as ion selective fluoride and calcium electrodes in our
study [8,9]. The term recovery is often used in quantifying the MD
measurements and it denotes the ratio between the concentration of
analyte obtained in the microdialysate and its real concentration in the
extracellular fluid surrounding the probe [10-12].
|The transport of fluoride through tissues is predominantly by nonionic
diffusion. Though, due to very small sizes of both the HF and
the fluoride ions both of these entities are able to diffuse very rapidly
through the tissues [13-15]. However, there is not much data available for explaining the rate of diffusion. Therefore this arrangement of work
coupling the MD technique with ion selective electrodes is intended to
elucidate the diffusion of fluoride across full size sheep eye globe.
|The eyes were selected as the model for study as the HF burns can
cause serious corneal damage, ocular surface defects and a permanent
loss in vision. The fundamental goal of any therapy for treating HF
eye burns includes verification that effective concentrations of the
pharmacological agent are attained within significant intraocular
sections. Thus the study is relevant by suggesting the use of MD
technique for demonstrating intraocular HF acid disposition in aqueous
and vitreous humor segments, presenting crucial documentation of
optimal drug dose and schedule [16,17].
|To detect the fluoride ion, a combination fluoride electrode,
sensitive to fluoride ions and not to any of its complexed forms such
as HF or HF2- was used. Use of this electrode for the assay of HF
necessitates that all the fluoride is decomplexed prior to measurement
of electrode potentials. To free the complexed fluoride, the pH of
the solution must be adjusted from the weakly acidic to weakly basic
region before making the determination. The dilution of samples and
standards with a large excess of sodium acetate (10:1), buffers the pH
to above 5 and helps to adjust the total ionic strength of samples and
standards to the same level . Higuchi’s model was used to describe
the permeation pattern of fluoride ions, through the eyes. This equation
shows that the permeation of fluoride ions is proportional to square
root of time. Fick’s first law of diffusion was used to describe the
diffusive flux of fluoride ions at any time point .
|In this study, HF diffusion was measured in terms of the fluoride
content recovered in the microdialysate samples. The in vitro probe
calibration and integrity test for assessing the diffusion of acid was
performed. The palliative effect of calcium gluconate following
ophthalmic HF contamination was studied. The concentrations of
calcium gluconate solution for the treatment of HF toxicity ranges from
1%, 2.5%, 5% and 10% have been widely reported in the literature .
We selected the median concentrations of 2.5 and 5.0% to study their
neutralization capacity. Two different formulations, solution (2.5% and
5.0% w/v calcium gluconate) and the 3% HPMC gel containing 2.5%
w/v calcium gluconate were subjected to neutralization potential study.
Materials and Methods
|MD-2204 BR-4 Brain microdialysis probe (BAS) with 4 mm
membrane, BAS microinjection pump, sample collection apparatus,
sheep eyes from euthanized animals obtained from Musa’s poultry
farm, PermeGear Franz diffusion cell, anhydrous calcium gluconate
powder, Hydrofluoric acid (48-51% w/w), Methocel K100 M premium
grade, Spectra/Por molecular porous membrane (MWCO 3500
Standard calibration curves for fluoride and calcium ions
|From the stock solution of 1 M HF, different concentrations (10-5-
1 M) were prepared by serial dilution method. Calibration curve was
obtained for these concentrations using pH/mV meter (Corning pH/
mV meter 220). These calibration curves were used to determine the
concentration of fluoride ions in subsequent experiments by using
Thermo Scientific Fluoride ion selective electrode (F electrode) with
serial number 9409BN. The same was repeated in case of calcium
gluconate serial dilutions except the concentration (10-5-0.1 M) using
Thermo Scientific Calcium ion selective electrode (Ca electrode)
with serial number 9300BN. The Ag/AgCl electrode was used as the
Preparation of 2.5% calcium gluconate gel
|Methocel (6 g) was dispersed in 50 ml of warm water and 100 ml of
cold water. The mixture was homogenized for 1 min at 2000 rpm and
then the dispersion was left overnight. Calcium gluconate powder (5 g)
was dissolved in 50 ml of warm water. The resultant calcium gluconate
solution was slowly dispersed in the gel matrix and stirred to get a
In vitro probe calibration
|The efficiency of 4 mm BR microdialysis probes was tested by
using the BAS microinfusion pump. The dialysate sample volumes
were controlled by the syringe pump equipped with precision glass
syringe. Since temperature affects the probe recovery, the in vitro studies
were done at the temperature at which the probes were ultimately used
in ocular experiments which was 25°C.
|The property of the MD probe for dialyzing HF acid solution was
checked by performing the in vitro probe experiments. During the
experiments, it was observed that the fluoride ion recovery was not
reproducible with Phosphate Buffer Saline as the perfusion medium.
Therefore the probe efficiency testing was performed by perfusing
distilled water at a constant flow rate of 4 μl/ min in 10 min interval.
An equilibration time of 1 hour was given before the actual sampling
to stabilize the probe.
|To perform the probe efficiency test, various concentrations of HF
were prepared in the range of 1 M to 10-5 M by serial dilution of the
stock solution. The choice of HF concentration used for the testing was
critical as HF acid is extremely corrosive. The concentrated acid solution
could interact with probe material while the low concentration could
be far below the detection limit of the F electrode (10-6 M to saturated).
Hence, three different concentrations of HF were selected for in vitro
testing in order to balance between the loss in probe integrity due to
direct acid exposure and F electrode sensitivity.
|Three samples were collected from the microdialysis probe effluent
for each concentration in 300 μl polyethylene vials. The resultant 40 μl
of samples were diluted to 10 ml volume using distilled water. The ionic
strength of sample solution was adjusted with 9 parts of 15% sodium
acetate solution as per the manufacturer’s recommendations.
|The result was presented as Fluoride ion Percent recovery by gain
(% recovery) and was calculated as:
|The % recovery values were plotted against the known HF acid
|Since HF is extremely corrosive acid, the interactions of HF and
probe materials could add a bias to the results. It is appropriate, but
not completely essential, for probe recoveries to be the same in all
probes. The recovery studies were done on individual probes before
their intended ocular use.
|The probe membrane function of 4 mm BR microdialysis probes
was tested in vitro. The probes were soaked in distilled water overnight
after the recovery experiment. The following day the probes were
placed in HF solution with a known concentration of 0.05 M. The
check always started with a 30 min equilibration period with a flow rate of 4 μl/ min. The microdialysis probes were then tested with a flow
rate of 4 μl/ min and with a sampling time of 10 min. Three samples
were collected from each probe and the fluoride concentrations in
dialysate fractions were calculated. A coefficient of variation (% CV)
was calculated for each probe’s three measurements. The % recovery
for each probe was calculated for dialysate samples in triplicate. The %
CV was compared for all the ten probes used.
Flow rate selection
|The relation between recovery and flow rate was observed by
placing the microdialysis probe in HF solution of 0.01 M concentration.
The volume of dialysate fractions was held constant at 40 μl. The flow
rates used for the experiments were 0.5, 1, 2 and 4 μl/ min. The constant
dialysate volume of 40 μl was selected because with lower volume the
F electrode was not sensitive enough and with the higher volumes, the
sampling period was much longer. Three samples were collected for
each flow rate, at 80, 40, 20 and 10 min respectively. The results were
presented as relative % recovery by gain and were calculated using
Selection of eyes
|Sheep eyes were selected for this study. Eyes of sheep with average
weight of 175-220 lb and age 3-4 years and of both the genders were
used for this study. All the eyes were used within a week of time of
sacrifice and 0.9% sodium chloride solution was used for preserving
the freshly plucked eyes. Sheep eye has bigger globe compared to
human and rabbit eye, convenient for probe insertion with accurate
positioning. Moreover, sheep eyes were easily accessible from butcher’s
|The solution was replaced every day to keep the eyes fresh during the
study. The eyes were inspected for any shrinkage or microbial growth
by observing the frozen cut section histopathology from random eyes.
In situ calibration
|Sheep eyes were carefully plucked out within 30 minutes of
euthanizing the animal and stored in 0.9% sodium chloride and kept
frozen until further use for the microdialysis experiment.
|To create a pinhole in the sheep eye for making the passage of
an MD probe, the eyes were incised using #14, 15, 16, 18, 20, 21 and
22- gauge needles. It was examined that the sheep eye has a very stiff
structure with the top layer being slippery. The finer needle gauges
could not give a sufficient insertion without altering the geometry
of the eye while with the thicker gauges; the insertion was too deep
causing the loss of aqueous humor or vitreous humor. Hence, 18-gauge
needle was selected to get the desired point insertion of eyes.
|An 18-gauge needle was inserted into the aqueous/vitreous humor
and then the MD probe was carefully inserted into the cavity and was
fixed with the adhesive so that there is no leakage of the fluids from the
cavity. The adhesive (ethyl cyanoacrylate) does not interact with the
|The donor cell for the eye experiments consisted of both the
aqueous and vitreous humors which represent a collective volume of
about 5-8 ml based on the samples drawn during the study. The volume
variance could be attributed to the change in the size, shape, age and
physiological conditions of the sheep’s eye.
|Full size sheep eye was used for the experimentation. The globe
was supported using the customized holder to keep the eye structure
secured in place. A volume of 0.15 ml of three HF acid concentrations was poured onto the surface of the eye to provide an illustration of the
maximum flux. The dialysate samples were withdrawn in 80 μl aliquots
at regular intervals of time, diluted with 15% sodium acetate solution
prior to their fluoride measurement using the combination electrode.
The probe was checked after the experiment, as described above.
|For the calcium gluconate neutralization studies, the contaminated
eyes were flushed with the excess of calcium gluconate solutions and
gel to provide the adequate contact of the antidote ions during the
whole experiment procedure.
|HF burnt eye ball and normal eye ball was placed in 1% Formalin
solution for 24 hours. Then, the eyes were placed in dry ice to make
the 20 micron sections. Thereafter, corneal sections were cut and each
section was placed on two separate slides, followed by 0.02% sodium
fluorescein solution dyeing for three times. Fluorescence microscopy
was used for visualization and image capturing from different positions.
|The eyes were frozen in 0.9% sodium chloride solution and the
solution was thawed whenever the eyes were required. The eyes were
tested for any shrinkage and microbial growth by examining the
histopathology of frozen cut section of randomly selected eyes under
the scanning microscope before experiments each day.
|For analyses of microdialysis samples, F electrode and Ca electrode
were used to measure their respective concentration. The instrument
was sensitive enough to detect these ions in the sampling range (10-6
M to saturated for fluoride ions). Microsoft Excel was used for analysis
with the one and two factor ANOVA without replication (p<0.05) was
used for each concentration for triplicate samples.
Results and Discussion
Flow rate selection
|The influence of flow rate on % recovery of fluoride ions was
observed, with 4 μl/min: 19.3% and 0.5 μl/min: 88.3% fluoride ions
recovered respectively (Figure 2). The decrease in flow rate (<2 μl/
min) resulted in increased sampling time to collect a constant sample
volume of 40 μl. It was observed by using ANOVA that that there is no
significant effect of changes in flow rate on sampling interval (p>0.05,
F<Fcritical). It was observed that due to the constraints in selecting high
flow rates in our study, the relative recovery of 15-20% was found to
be acceptable. An optimal flow rate of 4 μl/min was selected to balance
between the loss in % recovery and long sampling times.
|In vitro calibration
|The results from the % relative recovery for three different
concentrations of HF are summarized in Figure 3.
|Two factor ANOVA analysis showed that there was no significant
effect of the sample, or concentration changes on the % recovery
measurements for microdialysis probes at the p>0.05 (F=0.5218). It
was observed that the recovery of fluoride ions reduced with increased
concentration of HF to which the eye was exposed. The results showed
that the MD probe is capable of dialyzing the HF solution without
interacting with the probe, making the probe suitable for further in situ
|During the course of these recovery studies, it became obvious
that the in vitro concentration ranges were too low and hence the
ion selective electrodes may not be in the working range for use with microdialysis with in situ samples. As the samples collected from in
situ probes, would have to travel through the tortuous and complex
eye structure, the collected microdialysate levels will be really low in
fluoride concentration. A difference of 1 ppm may affect the % recovery
measurements and therefore, ocular changes could be missed or lost in
the random variations of the electrode potentials. It was decided that
the higher concentration of acid solutions would be poured for the
|The probe integrity test results for 0.05 M HF concentration
including mean recovery and % CV values for ten probes are
summarized in Table 1. The results show that 9 out of 10 probes give
recovery values without significant variance and one of the probes had
a high CV of 2.45% having possibility of experimental error. The single
factor ANOVA analysis was used for calculating the significant effects
of the % recovery values on each probe for the HF solution (p>0.05,
F<Fcritical). The results showed that the recoveries are the same,
within experimental error, for each probe. There was no interaction of
the acid with the probe material. This suggests that there should not be
any issues in using the probe for ocular testing.
In situ calibration
|The process of fluoride ion permeating in the eyes could be
diagrammatically elaborated in Figure 4. The recoveries for the MD
probe in three different concentrations of HF test solutions are shown.
Simple trend i.e., increase in dialysate fluoride ion concentration,
when the eyes were treated with higher concentration of HF was
observed for both aqueous and vitreous humor sampling. The lower
concentrations (0.1 and 1 M) did not show any appreciable difference in the fluoride ion permeation. As the samples collected from the in
situ probes were small in volume, their dilution could have a more
pronounced effect when the low concentration of acid was poured.
The ion selective electrodes may not be suitable, possibly require more
sensitive analytical technique. However, in the higher concentration,
fluoride ion slope is positive and plateau after the 180 min time sample.
This difference was because, as the HF acid permeates, it damages the
corneal membrane and the influx increases as the time progresses.
|In case of 0.1 M HF contamination, the lowest fluoride ion
concentration observed in the dialysate was found to be 8 ppm, which
is way higher than its safe permissible exposure limit of 3 ppm. So the
study is clinically important from the toxicological point of view .
The rate of fluoride permeation decreases with the decrease in amount
of HF exposure (29 M>1 M>0.1 M).
|In case of calcium studies, the influx decreases as the time
progresses (Figure 5). With the higher concentration, permeation is
high and the gel preparation shows the lowest influx due to slower
release from formulation.
|The study suggests that the MD technique can be employed for
demonstrating the intraocular HF disposition in both the aqueous
and vitreous humor segments. The fluoride and calcium control
experiments give a point of comparison for target neutralization with
calcium gluconate solutions and gel.
|Applying single factor ANOVA, it could be concluded that there
is no significant effect of concentration on fluoride diffusion for 29 M
HF solution (p>0.05, F<Fcritical) for both aqueous and vitreous humor
control samples. But there is a significant effect of concentration
on fluoride diffusion when the eye was exposed to 0.1 and 1 M HF
solutions for both aqueous and vitreous humor control samples. This
means that the variations of recoveries in samples for low acid exposure
could occur from the variability in assay method. This could be due to
less sensitive analytical technique.
|There is no significant difference in calcium ion concentrations for
samples in the calcium control studies. It suggests that the assay method
could measure the changes in calcium flux during in situ tests with the
calcium gluconate samples. It confirms the Ca electrode usability for
the neutralization studies.
Calcium gluconate treatment
|The in situ probe calibration performed in the preceding section
provides the baseline fluoride concentrations required for the calcium
gluconate therapy. When the acid burnt eyes were treated with calcium
gluconate solution, the fluoride ions formed the calcium fluoride
complex with the calcium ions. This results in less available free
fluoride ions and thus, decreased flux with low dialysate concentrations
(Figures 6-8). Within different calcium gluconate formulations, 2.5%
calcium gluconate gel is more effective than solution of the same
concentration. Whereas, the 5% calcium gluconate solution has almost
similar extent of complexation as compared to 2.5% calcium gluconate
gel. It can be inferred that the gel formulation is more effective for the
treatment as compared to all the solution formulations. The efficiency
of gel formulation is attributed to its extended ocular residence.
|A comparison study of HF neutralization with the calcium
gluconate gel vs. Methocel control formulation was performed to
eliminate any contribution from the gel component. Hence, sheep
eyes were contaminated with the 29 M HF solution, and flushed with
excess of Methocel gel to provide the adequate gel contact. The drawn
aqueous humor samples showed that the fluoride penetration followed the similar pattern in the presence of Methocel gel as the fluoride
control studies in the absence of any gel (Figure 4). It concluded that
the Methocel gel was not involved in neutralizing the fluoride ions.
|Applying single factor ANOVA, it could be concluded that there
was no significant effect of the sample fluoride concentrations on the recovery results for 29 M HF solution neutralization studies with all the
three formulations of calcium gluconate (p>0.05, F<Fcritical).
|Likewise, there was no significant effect of the sample fluoride
concentrations on the recovery results for the 2.5% calcium gluconate
gel neutralization studies for all the HF concentrations; while there
was a significant effect for the other neutralization studies for 0.1 and
1 M HF concentrations (p<0.05, F>Fcritical). This could be due to less
sensitive analytical technique.
|The neutralization of fluoride with the calcium ions is a time bound
process. It was easier to monitor the fluoride level changes for sampling
period where the amount of acid exposed was high; resulting in more
complexation. Therefore, more time could be required before the
fluoride concentration levels could return to the basal levels. The data
comparison for the control and neutralization studies show that the
MD technique can be used for preventing serious burns and systemic
toxicity or mitigating them by observing the time bound penetration
|The degree of changes in aqueous humor fluoride concentration
levels were very less obvious when exposed to the low acid
concentrations, probably could not be monitored efficiently.
|The technique of microdialysis could be used successfully to
ascertain the relative movement of fluoride ions permeating in the
HF-burnt eyes. Probes were found to be stable during the course of
investigation. In this study, at low concentrations of acid no appreciable
differences were found in permeation of fluoride ions as compared to
high concentrations. Complete neutralization by calcium ions takes up
to, or more than, 1-2 hours with a lag time of about 10-15 minutes so
the patient should be observed vigilantly for this period. By increasing
the amount of calcium ions available on eye surface, increases the
neutralization as shown in the results; the gel preparation is better than
the solution of same concentration. The intraocular levels of calcium
ion following application of calcium gluconate solution or gel were not
correlated to intraocular fluoride ions. It can be concluded that only
gel therapy provides necessary counter ion to neutralize the destructive
power of fluoride ions peri- and intra- ocularly presumably by offering
the necessary “sink” conditions that can draw fluoride from the affected
eye and because of reduced clearance from ocular surface.
|The research described above was performed by Ms. Navpreet Pandher,
a student at St. John’s University, New York. The credit goes to the valuable
guidance of Emilio Squillante III and Murali Mohan Bommana. The research and
ideas of Tom Needham and Chandra S. Chaurasia, on microdialysis played a
significant role in the research work.
Figures at a glance
- Dibbell DG, Iverson RE, Jones W, Laub DR, Madison MS (1970) Hydrofluoric acid burns of the hand. J Bone Joint Surg Am 52: 931-936.
- Rittenhouse KD, Pollack GM (2000) Microdialysis and drug delivery to the eye. Adv Drug Deliv Rev 45: 229-241.
- Waga J, Ohta A, Ehinger B (1991) Intraocular microdialysis with permanently implanted probes in rabbit. Acta Ophthalmol (Copenh) 69: 618-624.
- Waga J, Ehinger B (1995) Passage of drugs through different intraocular microdialysis membranes. Graefes Arch Clin Exp Ophthalmol 233: 31-37.
- Waga J, Ehinger B (2000) NGF administered by microdialysis into rabbit vitreous. Acta Ophthalmol Scand 78: 154-155.
- Benveniste H, Hüttemeier PC (1990) Microdialysis--theory and application. Prog Neurobiol 35: 195-215.
- Kehr J (1993) A survey on quantitative microdialysis: theoretical models and practical implications. J Neurosci Methods 48: 251-261.
- Elmquist WF, Sawchuk RJ (1997) Application of microdialysis in pharmacokinetic studies. Pharm Res 14: 267-288.
- Garrison KE, Pasas SA, Cooper JD, Davies MI (2002) A review of membrane sampling from biological tissues with applications in pharmacokinetics, metabolism and pharmacodynamics. Eur J Pharm Sci 17: 1-12.
- Chaurasia CS (1999) In vivo microdialysis sampling: theory and applications. Biomed Chromatogr 13: 317-332.
- Stahl M, Bouw R, Jackson A, Pay V (2002) Human microdialysis. Curr Pharm Biotechnol 3: 165-178.
- Ettinger SN, Poellmann CC, Wisniewski NA, Gaskin AA, Shoemaker JS, et al. (2001) Urea as a recovery marker for quantitative assessment of tumor interstitial solutes with microdialysis. Cancer Res 61: 7964-7970.
- Kirkpatrick JJ, Enion DS, Burd DA (1995) Hydrofluoric acid burns: a review. Burns 21: 483-493.
- MacKinnon MA (1988) Hydrofluoric acid burns. Dermatol Clin 6: 67-74.
- McCulley JP (1990) Ocular hydrofluoric acid burns: animal model, mechanism of injury and therapy. Trans Am Ophthalmol Soc 88: 649-684.
- Rittenhouse KD, Pollack GM (2000) Microdialysis and drug delivery to the eye. Adv Drug Deliv Rev 45: 229-241.
- Burstein NL (1995) Ophthalmic drug formulations. In Clinical Ocular Pharmacology. Bartlet JD, Jaanus SD (Eds) Butterworth-Heinemann Medical, Woburn, Massachusetts. pp: 21-45.
- Thermo Scientific Fluoride ion selective electrode manual.
- Higuchi WI (1967) Diffusional models useful in Biopharmaceutics drug release rate processes. J Pharm Sci 56: 315-324.
- Segal EB (2000) First aid for a unique acid HF: A sequel. JCHAS 7: 18-23.
- Barbosa VT, Thiesen R, Soares EG, Machado MRF, Laus JL (2009) Effects of ethyl- cyanoacrylate and octyl-cyanoacrylate on experimental corneal lesions in rabbits. Arq Bras Med Vet Zootec 61: 1067-1076.
- The Code of Federal Regulations of the United States of America(1993) Title 29 CFR, Part 1910, Section 1000, Table Z-2.