`
`and 3,5-dinitrobenzoyl chloride reagent. The titra-
`tion was carried out potentiometrically beyond the
`second break (titration of the benzoate ester) using a
`glass-saturated calomel electrode system and one
`of several different bases as titrant. Following this,
`solid sodium or ammonium acetate was added to
`determine whether the potential of
`the solution
`could be reversed partially and made to stabilize
`at the beginning of the second break. A t this point,
`color development due to the ester is near a maxi-
`mum, while color due to the 3,5-dinitrobenzoic acid
`is negligible.
`In this manner solvent systems
`involving pyridine, acetonitrile, dimethylformamide,
`acetone and chloroform alone, and some selected
`mixtures, were tested to arrive at the solvent mixture
`used in the procedure. Tetrabutylammonium hy-
`droxide (1 M in methanol) and KOH (1 M in water)
`were tested as bases before turning to concentrated
`ammonia.
`It is obvious that any of the reactive compound
`types would constitute a positive interference in the
`analysis of any other. Similarly, reactive acyl
`functions (anhydrides or halides) will tend to show
`negative interference depending on the efficiency
`with which they compete with the 3,5-D reagent for
`the available hydroxyl groups. The fact that both
`hydrochloric and 3,5-dinitrobenzoic acids are
`formed during the esterification reaction indicates
`that these types of compounds would not constitute
`interferences if present in small amounts. Water
`will react preferentially with 3,5-D, but will not
`interfere seriously unless present in sufficient quan-
`tity to consume a large proportion of the available
`
`351
`reagent. Additional compounds which have been
`tested and shown not to interfere seriously in the
`determination of hydroxyl groups are summarized
`in the data of Table 11. Note also in Table I that
`the multifunctional steroids imply a lack of inter-
`ference by unsaturated moieties and by simple and
`conjugated ketones.
`
`SUMMARY
`A method has been presented and described for
`the colorimetric determination of organic alcohol,
`amine, and thiol groups. The method is free of
`interference from most common solvents and other
`functional groups. The procedure is rapid and the
`resulting products are adequately stable to provide
`ease of measurement.
`
`REFERENCES
`(1) Porter, C. C., Anal. Chem., 27, 805(1955).
`(2) Berezin, J. V., Doklady Akad. Nauk, S.S.S.R., 99,
`563(1954); through Chem. Absfr., 50,859g(1956).
`(3) Robinson, W. F., Jr., Cundiff, R. H., and Markunas,
`P. C., Anal. Chem., 33,1030(1961).
`,_ - -- (4) Dal Nogare, S., and Mitchell, J., Jr., dbid., 25, 1376
`( l Y 5 8 ) .
`(5) Schreiber, J., and Eschenmoser, A,, Helv. Chim. Ada,
`38, 1520(1955).
`(6) Critchfield, F. E., and Hutchinson, J. A,, Anal. Chem.,
`32,862(1960).
`(7) Baggett, B., Engel, L. L., and Fielding, L. L., J . Biol.
`CRem. 213 87(1955).
`(8)’ Gutkkov, G., and Schenk, G. H., Anal. Chem., 34,
`1316(1962).
`(9) Johnson, D. P., and Critchfield, F. E., ibid., 32,
`865(1960).
`(10) Fritz J. S. and Schenk G. H. ibid., 31, 1808(1959).
`(11) Schedk, G.’ H., and Sktiago: M., Jlicrochem. J.,
`6, 77(1962).
`(12) Sully, B. D.. Analyst, 87,940(1962).
`
`Behavior of Erythrocytes in Various
`Solvent Systems I11
`
`Water-Polyethylene Glycols
`
`By B. LESTINA SMITH* and DONALD E. CADWALLADER
`
`Hemolytic behavior of human and rabbit erythrocytes in aqueous solutions of poly-
`ethylene glycol (PEG) 200, 300, 400, and 600 was investigated. Complete he-
`molysis occurred in all PEG 200 and 300 solutions, with discoloration occurring in
`2. 2 5 per cent PEG 200 and 2 15 per cent PEG 300 solutions. Sodium chloride
`was effective in preventing hemolysis in 5 2 5 per cent PEG 200 or 5 40 per cent
`PEG 300 solutions. When possible, i values were calculated for sodium chloride
`in the various water-PEG 200 and 300 solutions. PEG 400 and 600 protected
`blood celIs from damage in > 10 per cent to < 40 per cent solutions, and i values
`were calculated for these PEG‘s. Solutions containing 2 40 per cent PEG 400 or
`600 (with and without NaC1) were damaging to red cells. The ability of liquid
`PEG‘S topenetraterabbit and human erythrocytes appeared to be 200 > 300 > 400
`> 600.
`
`REXIOUS PAPERS in this series have reported water-glycerin and water-propylene glycol sys-
`behavior of erythrocytes in various
`tems (1, 2). Among other nonaqueous solvents
`that might be used in the preparation of paren-
`terals would be the liquid polyethylene glycOls.
`Polyethylene glycols (PEG’s) are products
`possessing a very low order of toxicity.
`spiegel
`and Noseworthy (3) have reviewed the physical
`properties, toxicities, and parenteral applications
`of
`these liquids. Skin penetration studies on
`
`Received September 19, 1966, from the School of Phar-
`macy, University of Georgia, Athens, GA 30601
`Accepted for publication December 15, 1966.
`Abstracted in part from a thesis submitted by B. Lestina
`Smith to the Graduate School, University of Georgia,
`Athens, in partial fnlfillment of Master of Science degree
`requirements.
`* Fellow of the American Foundation for Pharmaceutical
`Education.
`Previous paper: Cadwallader D. E Wickliffe, B. W., and
`Smith, B. L., J . Pharm. Sci.. 53.’927(1&4).
`
`MYLAN INST. EXHIBIT 1122 PAGE 1
`
`MYLAN INST. EXHIBIT 1122 PAGE 1
`
`
`
`352
`
`many of these products show that they do not
`penetrate the skin in harmful amounts (4).
`Meyer and Stunner (5) reported low oral and
`subcutaneous toxicities of PEG 200 and GOO in
`mice. Swanson and co-workers (6) found that
`sodium amobarbital in 60% PEG 200 and sodium
`secobarbital in 50% P E G 200 had approxi-
`mately the same potency and toxicity as aqueous
`solutions of these barbiturates. Lee and Ander-
`son (7) determined the toxicity of vancomycin in
`50% PEG 200 and of PEG 200 alone. Their
`results indicated that PEG 200 produced no
`apparent toxic effects when given to dogs at 1.0
`for 80 days intramuscularly, or
`ml./Kg./day
`0.5, 1.0, 2.5, and 5.0 ml./Kg. as single intra-
`venous doses.
`The purpose of this investigation was to con-
`duct experiments to study the behavior of red
`blood cells in aqueous polyethylene glycol solu-
`tions. Experiments were designed to determine
`the effect that aqueous solutions of PEG's
`200, 300, 400, and 600 have in preventing hemol-
`ysis alone, and in the presence of sodium chloride.
`
`EXPERIMENTAL
`
`glycols 200, 300, 400,
`Materials-Polyethylene
`and 600, supplied by Union Carbide Chemical Corp.,
`were used without further purscation. The sodium
`chloride used was reagent grade.
`Collection of Blood-Approximately
`10 ml. of
`blood was obtained from rabbits by heart puncture.
`An 18-gauge, 2-in. needle attached to a 10-ml.
`syringe was used to make entrance into the heart.
`The blood was placed in a 50-ml. round-bottom
`flask which contained 10-15 glass beads. After
`gently rotating the flask for approximately 5 min.,
`the ddbrinated blood was decanted into a 50-ml.
`conical flask. The blood was aerated by gently
`swirling the flask for about 5 min.
`The human blood used was obtained from the
`forearm veins of several 20-25-year-old healthy
`male Caucasians. The blood was treated in the
`same manner as the rabbit blood. Fresh blood
`samples were used in all experiments.
`Preparation of Solutions-All of the polyethylene
`glycol and sodium chloride solutions were weight-in-
`volume percentage preparations.
`Quantitative Determination of Per Cent Hemol-
`ysis-The method used to determine the degree
`of hemolysis in this investigation was dependent
`upon the fact that the amount of oxyhemoglobin
`liberated from the red corpuscles in hypotonic
`solutions is a direct function of the number of cells
`hemolyzed. A quantitative determination of par-
`tial hemolysis in any mixed solvent system was
`made by centrifuging solutions containing un-
`hemolyzed cells and determining the oxyhemoglobin
`in the supernatant solutions by means of a photo-
`electric colorimeter.
`The general method consisted of transferring 5 ml.
`of standard sodium chloride solutions (0.34, 0.36,
`. . ., 0.44, 0.46'%) into each of two test tubes.
`Iden-
`tical amounts of the mixed solvent systems were
`
`Journal of Pharmaceutical Sciences
`
`also transferred into each of two test tubes. Then
`0.05 ml. of blood was added to each test tube and
`the tubes inverted several times to obtain complete
`mixing. After 45 min. a t 37", the blood mixtures
`were centrifuged. Because of the viscosity of the
`PEG solutions, especially in high concentrations,
`it was necessary to use centrifuge speeds of 2000-
`3000 r.p.m. to bring about complete settling of in-
`tact cells. After centrifuging, the light absorbance
`of the supernatant liquid was measured using
`photoelectric
`colorimeter
`a Klett-Summerson
`equipped with a No. 54 filter. These absorbance
`readings were divided by the absorbance readings
`for 0.05 ml. of blood in 5 ml. of distilled water
`(standard for 100% hemolysis) and multiplied by
`100 to obtain per cent hemolysis occurring in each
`test solution. A blank, used to cancel any light
`absorbance inherent to the blood sample, was pre-
`pared by placing 0.05 ml. of blood in 0.9% sodium
`chloride solution, allowing to stand for 45 min.,
`and centrifuging in a like manner.
`Water-PEG solutions absorbed a small amount
`light, and this absorbance increased with an
`of
`increase in PEG content. This absorbance was
`determined for the various concentrations of PEG's
`used in the experiments, and these blank readings
`were subtracted from the colorimeter readings
`obtained at the end of the hemolysis experiments.
`A battery-operated model M Beckman pH meter
`was used in all pH measurements.
`i values (isotonic
`Calculation of i Values-The
`coefficients) for polyethylene glycols were calculated
`according to the equation of Grosicki and Husa
`(8) which was modified as follows.
`
`Gm. of NaCl
`
`(Gm.-mol. wt. of NaC1)
`
`- -
`
`Gm. of PEG
`
`(Gm.-mol. wt. of PEG)
`
`(Eq. 1)
`
`Concentrations of sodium chloride and poly-
`ethylene glycol causing the same degree of hemolysis
`(e.g., 25, 50, and 75%) were used in the above
`equation to calculate i values for the polyethylene
`glycols.
`Experiments were carried out to obtain data for
`calculating apparent i values for sodium chloride
`when the salt was present in a water-polyethylene
`glycol system in which the polyethylene glycol itself
`exhibited no protection to red blood cells, e.g.,
`In these calcula-
`sodium chloride in 10% PEG 200.
`tions, it was necessary to assume that polyethylene
`glycol did not contribute to the osmotic behavior
`of the solutions and that sodium chloride was solely
`responsible for the tonicity of the solutions. Since
`the calculations were concerned with i values of only
`sodium chloride in different solvents, the molecular
`weights in Eq. 1 are identical and the equation
`becomes
`
`( ivaluefor )( water )
`
`NaCl in water
`
`Gm. of NaCl
`in 100 ml. of =
`
`MYLAN INST. EXHIBIT 1122 PAGE 2
`
`MYLAN INST. EXHIBIT 1122 PAGE 2
`
`
`
`Vol. 56, No. 3, March 1967
`The value of i for sodium chloride was taken as
`1.86, which is the accepted value of i for 0.9%
`sodium chloride in water (8).
`Curves showing the degree of hemolysis in sodium
`chloride-water
`solutions and sodium chloride-
`water-PEG solutions were plotted on rectangular
`coordinate paper. From these curves the concen-
`trations of sodium chloride in Gm./100 ml. of water
`and PEG, causing 25, 50, and 75% hemolysis, were
`determined. These values were
`inserted
`into
`Eq. 2, and the values of i for sodium chloride in a
`particular water-PEG solution at concentrations
`giving 25, 50, and 75% hemolysis were determined.
`Preparation of Hemolysis Curves-Experiments
`employing human blood were carried out to deter-
`mine apparent i values for sodium chloride in
`various PEG 200 and 300 solutions. The average
`readings of these experiments were used to con-
`struct a standard hemolysis curve.
`In constructing
`the hemolysis curves of the various PEG solutions
`the grams of sodium chloride per 100 ml. of solution
`causing 25, 50, and 75% hemolysis were calculated
`with reference to the standard hemolysis curve.
`By utilizing Eq. 2, the grams of sodium chloride per
`100 ml. in a polyethylene glycol solution causing
`25% hemolysis was calculated as
`
`A X B
`
`x=- c
`where X = Gm. of sodium chloride in 100 ml. of
`polyethylene glycol solution causing 25% hemolysis,
`A = 1.86 as the i value for sodium chloride in water,
`B = Gm. of sodium chloride in 100 ml. of water
`causing 25y0 hemolysis (obtained from standard
`hemolysis curves), and C = previously calculated i
`value for sodium chloride in appropriate polyethylene
`glycol solution. Similar calculations were carried
`out to obtain i values a t 50 and 75% hemolysis.
`With these three i values at 25, 50, and 757, he-
`molysis, the hemolysis curves for the various poly-
`ethylene glycol solutions were constructed.
`
`RESULTS
`Polyethylene Glycol 200 and 3 O G T h e hemolysis
`of rabbit and human erythrocytes after 45 min. a t
`in various water-PEG 200 and 300 solu-
`37'
`tions are shown in Figs. 1 and 2. All of the PEG
`200 and 300 solutions void of sodium chloride
`caused complete hemolysis of rabbit and human
`erythrocytes; however, hemolysis of red blood cells
`in aqueous solutions containing more than 15-25%
`PEG 200 or 300 resulted in brown-green solutions
`instead of normal red solutions. The concentrations
`at which this discoloration occurred are shown in
`Table I.
`The inclusion of 0.9% sodium chloride in aqueous
`solutions containing 0.0 to 257, PEG 200 or 300
`afforded complete protection (no hemolysis) to
`rabbit and human erythrocytes. Howcvcr, the
`inclusion of sodium chloride in solutions containing
`25% and more of PEG 200 or 40y0 and more of PEG
`300 did not prevent damage of blood cells. At
`these critical concentrations (see Table I ) the red
`blood cells were destroyed resulting in brown-green
`solutions. The addition of 2, 3, or 5% sodium
`chloride to 25, 30, 40, and 50y0 PEG 200 and 300
`
`353
`
`80 .
`
`60.
`5
`5 4 0 -
`0
`
`1
`20 .
`
`0
`
`1 2 3 4 5 6 7 8 91015202530
`% PEG
`of rabbit erythrocytes after 45 min.
`Fig. 1-Henzolysis
`at 37" in various polyethylene glycol-ater
`solutions.
`Key: @, PEG 200 and 300; X , PEG 400; 0, PEG
`600; S, discoloration occurred.
`
`ER 20
`
`0
`
`1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 25 35 45
`% PEG
`Fig. 2-Hemolysis of human erythrocytes ufter 45 min.
`in various polyethylene glycol-ater
`solutions.
`at 37'
`Key: @, PEG 200 and 300; . XI PEG 400; 0, PEG
`600; S, discoloratzon occurred.
`
`TABLE I-CONCENTRATIONS (Gm./100 ml.) OF
`PEG IN AQUEOUS SOLUTIONS
`AT WHICH DISCOLORA-
`TION OF BLOOD OCCURRED"
`
`PEG
`2006
`300b
`40OC
`60OC
`
`7- N o N a C h ----O.,9y0 NaCl?
`Rabbit
`Human
`Rabbit
`Human
`Blood
`Blood
`Blood
`Blood
`25
`25
`30
`40
`25
`25
`15
`15
`15
`40
`40
`40
`20
`40
`40
`40
`
`Each value represents an average of at least four blood
`samples. *Discoloration in PEG 200 and 300 solutions
`resulted in brown-green solutions.
`Discoloration in PEG
`400 and 600 solutions resulted in brown-black precipitates.
`The solutions were not colored.
`
`solutions prevented hemolysis and discoloration of
`blood.
`The fragility of human erythrocytes in various
`water-PEG 200 and 300 solutions was modified or
`corrected by the addition of sodium chloride. It
`was possible to calculate i values for sodium chloride
`in various water-PEG 200 and 300 solutions. The
`average i values for sodium chloride in 5, 10, and
`20% PEG 200 and 300 solutions are shown in Table
`11. The i values for sodium chloride in 10 and 20T0
`PEG 200 solutions were less than 1.86 (the accepted
`value for 0.9% sodium chloride in water). The i
`values were greater than 1.86 for PEG 300 solutions.
`The pH readings for PEG 200 and 300 solutions
`
`MYLAN INST. EXHIBIT 1122 PAGE 3
`
`MYLAN INST. EXHIBIT 1122 PAGE 3
`
`
`
`354
`TABLE 11-VALUES OF i FOR NaCl IN VARIOUS
`WATER-PEG SOLUTIONS, CALCULATED FROM CON-
`CENTRATIONS CAUSING 25, 50, AND 75% HEMOLYSIS
`OF HUMAN ERYTHROCYTES AT 37'0
`
`% w/v
`PEG 200
`5 b
`10
`20
`PEG 300
`5
`10
`20
`
`-----Hemolysis,
`50
`25
`
`yo-
`
`1.8
`1 . 7
`1.6
`
`4.0
`2 . 6
`2 . 8
`
`2.1
`1.7
`1.6
`
`5.2
`4.4
`3.1
`
`75
`
`2.2
`1.9
`1.7
`
`6.6
`6.9
`3.4
`
`Av.
`
`2.0
`1.8
`1 . 6
`
`5 . 3
`4 . 6
`3 . 1
`
`Unless otherwise indicated each i value represents an
`average of at least two blood samples.
`bAverage of four
`blood samples.
`
`before and after the addition of blood were within a
`range of 3.5 to 4.5.
`Hemolysis curves showing the amount of Iaking
`that occurred when human blood was added to
`various water-PEG 200 and PEG 300-sodium
`chloride solutions are shown in Figs. 3 and 4.
`These curves were constructed
`in the manner
`described under Experimental utilizing the data
`presented in Table 11. Unusual data were ob-
`tained for experiments using 5% PEG 200 solutions
`containing various amounts of sodium chloride and
`the results are shown in Fig. 5. Instead of the
`continual increase in hemolysis with decreasing
`sodium chloride concentration, hemolysis in 501,
`PEG 200 solutions decreased in those solutions con-
`taining 0.38 to 0.357, sodium chloride and then
`increased with further decrease in sodium chloride
`concentrations.
`Polyethylene Glycol 400 and 6 0 G T h e hemolysis
`of rabbit and human erythrocytes after 45 min. a t
`37" in various water-PEG 400 and 600 solutions is
`shown in Figs. 1 and 2. Hemolysis was prevented
`in various aqueous PEG 400 and 600 solutions up to
`critical concentrations where damage of red blood
`cells occurred resulting in brown-black precipitates
`(see Table I).
`It was possible to calculate i values for PEG
`400 and PEG 600. Average i values are shown in
`Table 111.
`"loride
`The addition Of O"%
`to
`'Odium
`tions containing less than 40% PEG 400 or 600
`
`100
`
`r
`
`.
`80
`?? fn
`5 6 0 -
`0 z
`W I 40
`E? 20
`
`.
`
`.
`
`Journal of Pharmaceutical Sciences
`
`.40
`
`.34.28
`
`.18
`.24
`'% NaCl
`of human erythrocytes after 45 min.
`Fig. 4-Hemolysis
`at 37" in various polyethylene glycol 300-saline solu-
`tions.
`
`.12
`
`.06 .02
`
`100 r
`
`80
`
`2
`
`60
`
`0 $ 40
`
`k?
`
`20
`
`of human erythrocytes after 45 ??tin.
`Fig. 5-Hemolysis
`at 37' in 5% polyethylene glycol 200-saline solutions.
`
`TABLE 111-VALUES OF i FOR PEG 400 AND 600
`AT CONCENTRATIONS (Gm./100 ml.) CAUSING 25,
`
`50, and 757, HEMOLYSIS OF RABBIT AND HUMAN
`ERYTHROCYTES AT 3 7 O "
`
`--Hemolysis.
`25
`50
`Rabbit Blood
`0 . 7
`0.8
`1 . 2
`1.4
`Human Blood
`0 . 5
`0.6
`0 . 9
`1 . 2
`
`%---
`
`75
`
`1.0
`1.6
`
`0 . 8
`1.6
`
`Av.
`
`0 . 8
`1 . 4
`
`0.6
`1.2
`
`PEG400
`PEG600
`
`PEG400
`PEG600
`
`a Each i value represents an average of two to seven blood
`samples.
`
`afforded complete protection (no hemolysis) to
`rabbit and human erythrocytes; however, at critical
`concentrations (see Table I) damage of red blood
`cells occurred as in PEG 400 and 600 solutions void
`of sodium chloride. This damage was not prevented
`by the addition of 2, 3, or 5Y0 sodium chloride to
`40, 50, and 60% solutions of PEG 400 and 600.
`The pH readings of all PEG 400 and 600 test
`solutions before and after the addition of blood was
`within a range of 3.5 to 4.5.
`
`DISCUSSION
`
`The concentrations of PEG 200, 300, 400, and
`600 in water that are iso-osmotic with 0.9% sodium
`chloride, according to calculations using the osmotic
`factor equation (9),
`
`. 8
`.30
`
`1
`
`.
`
`,
`
`I
`
`,
`
`,
`
`,
`
` . ' .
`.
`-50 .46
`-42 -38 .34
`'% NaCl
`Fig. 3-Hemolysis of human erythrocytes after 45 min.
`at 37' in various polyethylene glycol 200-saline solu-
`tions.
`
`MYLAN INST. EXHIBIT 1122 PAGE 4
`
`MYLAN INST. EXHIBIT 1122 PAGE 4
`
`
`
`Vol. 56, No. 3, Murch 1967
`osmotic factor =
`No. of particles
`from 1 molecule
`
`of solute )( soln.
`
`Gm. of solute
`in 100 ml. of
`
`)
`
`(Eq. 4)
`
`(Gm.-mol. wt. of solute)
`
`are 5.8, 8.7, 11.6, and 17.4Y,, respectively. Re-
`gardless of
`the polyethylene glycol content in
`aqueous solution, PEG 200 and 300 solutions failed
`to prevent hemolysis of rabbit and human erythro-
`cytes. These experimental data point out that
`when water-PEG 200 or water-PEG 300 are used as
`vehicles for intravenous solutions, the finished
`product should not be assumed hypertonic with
`respect to blood, even when there is a high concen-
`tration of polyethylene glycol present.
`In experi-
`mental studies, tissue reactions have been observed
`following parenteral doses of undiluted polyethylene
`glycols which are severe enough to warrant very
`thorough study of safety before any parenteral
`applications are made (10). Hemolysis has resulted
`from high concentrations injected into the blood
`stream. Viscosity and lack of diffusion have re-
`sulted in discomfort or pain after subcutaneous
`injections of undiluted material and ischemic
`necrosis has also been seen after intramuscular
`In a study of the polyethylene glycols
`injections.
`as vehicles for intramuscular and subcutaneous
`injections by Carpenter and Shaffer ( l l ) , tissue
`reactions at the site of subcutaneous and intra-
`muscular injections of undiluted PEG 300 in dosages
`2.5 to 10 times that anticipated for human use
`caused blanching of the skin and scab formation
`In a study by McCabe et al. (12), it was
`in 48 hr.
`found that daily administration of 240 mg. of nitro-
`furantoin in PEG 300 to 30 patients caused severe
`metabolic acidosis and nephropathy in seven patients
`resulting in two deaths. These damaging effects
`were attributed to polyethylene glycol rather than
`nitrofurantoin.
`Sodium chloride is effective in preventing hemol-
`ysis of human erythrocytes in aqueous polyethylene
`glycol solutions as long as the polyethylene glycol
`concentrations do not exceed 25 to 4oyO. At these
`higher polyethylene glycol concentrations, erythro-
`cytes are not protected from discoloration by the
`addition of 0.9% sodium chloride. Concentrations
`of polyethylene glycols causing discoloration of
`human and rabbit erythrocytes are summarized
`in Table I. Higher concentrations of
`sodium
`chloride prevented discoloration in PEG 200 and
`300 solutions but not in PEG 400 and 600 solutions.
`This damage does not appear to be hemolytic in
`character, but seems to be a chemical type of de-
`struction. It appears that pH was not a factor since
`the pH of all concentrations of polyethylene glycol
`solutions remained within a range of 3.5 to 4.5
`before and after the addition of blood.
`The van’t Hoff factor (i value or isotonic co-
`efficient) is defined as the ratio of the colligative
`effect produced by a concentration (molal) of elec-
`trolyte divided by the effect observed for the same
`concentration of nonelectrolyte (13). In studying
`the effect of low concentrations of various substances
`on erythrocytes, previous workers (1, 2, 8) used
`molar concentrations in place of molal concentra-
`tions in their calculations. The use of molar
`instead of niolal concentration would introduce
`only a small error at electrolyte coricentrations of
`
`355
`0.1 M or less (8). However, in the present study,
`comparatively high concentrations of liquid poly-
`ethylene glycols (4-9%) were involved in calcula-
`tion of i valu’es. This meant that enough volume
`of the test solution was occupied by the polyethylene
`glycol to produce a substantial difference between
`molar and molal concentrations. Molal concentra-
`tions of the test solutions used in this study were
`calculated by direct proportion (no shrinkage was
`noticed when polyethylene glycols and water were
`mixed) and i values calculated using molal concen-
`trations. They were found to be 12 to 15% lower
`than the i values calculated on a molar basis. How-
`ever, it was decided to calculate hemolytic i values
`using molar concentrations since pharmaceutical
`calculations in the area of isotonic solutions are based
`on this concentration expression.
`The i values for sodium chloride in most aqueous
`PEG 200 solutions were less than 1.86. These low
`i values indicate that PEG 200 offers no protection
`to red blood cells against osmotic hemolysis.
`In
`fact, there probably is some deleterious effect since
`more sodium chloride is needed to protect the red
`blood cells against hemolysis in aqueous PEG 200
`solutions than in water. The higher values of i for
`sodium chloride in aqueous PEG 300 solutions in-
`dicate that PEG 300 contributes to the tonicity of
`aqueous solutions. The fact that i values could be
`calculated for PEG 400 and 600 shows that these
`polyethylene glycols have the ability to protect
`erythrocytes against osmotic hemolysis. Since i
`values for PEG 400 were less than i values for PEG
`600, it can be assumed that PEG 600 gives greater
`protection to red blood cells than PEG 400. There-
`fore, the order in which the liquid polyethylene
`glycols protect rabbit and human erythrocytes
`against hemolysis is 200 < 300 < 400 < 600. The
`ability of the liquid polyethylene glycols to con-
`tribute to the tonicity of aqueous solutions is de-
`pendent on the molecular weight. A possible
`explanation for this behavior might be that the
`lower molecular weight polyethylene glycols are
`able to penetrate the red blood cell membrane and
`therefore have little or no effective concentration in
`the extracellular solutions. It appears that in-
`creasing the molecular weight decreases the mem-
`brane penetrating properties of liquid polyethylene
`glycols.
`
`REFERENCES
`
`(1) Cadwallader, D. E., J . Pharm. Sci., 52,1175(1963).
`(2) Cadwalldder, D. E., Wickliffe, B. W., and Smith,
`B. L., ibid., 53, 027(1064).
`(3) Spiegel, A. J., and Noseworthy, M. M., ibid., 52,
`-..~.-”.,.
`91711963
`(4). “Carbowax, Polyethylene Glycols,” Union Carbide
`Chemical Co., New York, N. Y., 1960.
`(5) Meyer, G., and Sturmer, E., Arch. Intern. Phar-
`macodyn., 90,193(1952).
`(6) Swanson, E. E., Anderson, R. C., Harris, P. N., and
`Rose, C. L., J . A m . Pharm. Assoc., Sci. Ed., 42,571(1953).
`(7) Lee, C., and Anderson, R. C., Toxicol. A M . Phar-
`marol. 4 206(1962).
`(8)‘ drosicki, T. S.. and Husa, W. J., J . A m . Pharm.
`Assoc., Sci. E d . , 43, 632(1954).
`( 9 ) :‘Husa’s Pharmaceutical Dispensing,” 5th ed., Mack
`Publishing Co., Easton, Pa., 1959, p. 142.
`(10) “Carbowax. Polyethylene Glycols for Pharmaceu-
`ticals and Cosmetics.” Union Carbide Chemical Co.. New
`YgFk N. Y . ~ 1959.
`1 1 0 Cardenter, C. P., and Shaffer, C. B . , J . A m . Pharm.
`Sci. E d . ,
`A&.,
`41, 27(1952).
`, W. R., Jackson, G. G., and Griehel, H. G.,
`(12) McCabe
`Arch. Infernal dfed.. 104, 710(1959).
`(13) Maron, S. H., and Prutton, C. F., “Principles of
`Phvsical Chemistrv.” 3rd ed.. Macmillan Co.. New York.
`N.JY., 1958, p. 199:
`
`MYLAN INST. EXHIBIT 1122 PAGE 5
`
`MYLAN INST. EXHIBIT 1122 PAGE 5
`
`