pharmacol. 1996, 48: 1237-1242
`I. wived March 4, 1996
`AMpd Apd 19, 1996
`preservative Efficacy Tests
`Reproducibility and Factors
`
`0 1996 J. Pharm. Pharmacol.
`
`in Formulated Nasal Products:
`Affecting Preservative Activity
`
`G. W . H A N L O N A N D J . P. R E Y N O L D S
`N . A. H O D G E S , S. P. D E N Y E R ,
`Department of Pharmacy, University of Brighton, Moulsecoornb, Brighton, E. Sussex, BN2 4GJ, UK
`
`Abstract
`Preservative efficacy tests were performed in triplicate on each of three batches of three formulated nasal spray
`preparations to assess the inter- and intra-batch variation in preservative performance which typically results
`from these procedures, and to assess the relative importance of factors influencing preservative performance in
`nasal products.
`Tests were conducted using procedures conforming, as far as possible, to both the European and the US
`pharmacopoeias and the results interpreted using the performance criteria of both. Despite the adoption of
`practices designed to maximize reproducibility, a marked variation in the degree of microbial inactivation was
`observed, both within and between batches of product.
`A preservative system comprising benzalkonium chloride and phenylethyl alcohol was found to be far
`superior to combinations of either benzalkonium chloride plus disodium edetate or potassium sorbate plus
`disodium edetate, both of which failed to satisfy the EP performance criteria on a number of occasions.
`Proposals are made for the adoption of inactivation criteria which incorporate realistic error limits reflecting the
`inherent problems of reproducibility of the viable counting procedures involved.
`
`Over the course of the last 30 years very few new antimicrobial
`preservatives have been introduced for use in parenteral and
`ophthalmic products or in products applied to mucous mem-
`branes. Toxicity considerations have restricted the application
`of those preservatives which have been developed during this
`period to the protection of topical products, and the same
`considerations have caused a marked reduction in the use of
`some long-established agents, e.g. organomercurials, in par-
`enterals and ophthalmics. It is, therefore, becoming increas-
`ingly common for formulated medicines to be protected
`against microbial spoilage with a combination of preservatives
`rather than a single agent (Denyer & King 1988). Frequently
`the reason for this change is because a single agent cannot
`provide a sufficiently broad spectrum of antimicrobial activity.
`There is, however, also the need to achieve acceptable levels of
`product protection from the limited choice of available agents
`by capitalizing on potential synergistic interactions between
`Preservatives whilst at the same time minimizing the risk of
`adverse reactions by avoiding the use of unnecessarily high
`concentrations (Hodges & Denyer 1995).
`In striving to limit preservative concentrations without
`compromising product safety it is important to recognize, and
`investigate the many factors which influence preservative
`Performance (Denyer & Wallhauser 1990) and to maximize the
`reliability and repeatability of preservative efficacy tests in
`order to improve their ability to predict product vulnerability to
`spoilage and enhance the usefulness of the test results in the
`Pre-formulation process (Davison 1988; Baird 1995; Hodges &
`k y e r 1995).
`Because preservative efficacy tests are normally employed
`to provide a qualitative pass or fail result, it is important for the
`fornulator to have a good appreciation of the precision
`
`Correspondence: N. A. Hodges, Department of Pharmacy, Univer-
`sity of Brighton, Moulsecoomb, Brighton, E. Sussex, BN2 4GJ, UK.
`
`achievable, particularly when assessing results from a product
`in which preservative efficacy is at the borderline of accept-
`ability. Although pharmacopoeial tests have been in use since
`1973 and there are numerous publications which analyse and
`assess their virtues and shortcomings (Allwood 1986; Cooper
`1989; Spooner & Davison 1993), there appears to be very little
`published information about the level of inter- and intra-batch
`reproducibility that might be achieved using pharmacopoeia1
`methods; this work was intended to address this deficiency.
`It is proposed that nasal products are, for the purposes of
`preservative performance criteria, to be regarded as topicals
`(Anon 1993), but the selection of a preservative system in such
`products requires particular care because application to
`mucous membranes might carry a greater risk of sensitization
`or adverse reaction than application to the skin. Nasal pro-
`ducts, therefore, represent a product category where pre-
`servative performance needs to be properly optimized and
`assessed; this study evaluates the preservative activity in three
`such products each protected by a different system.
`
`Materials and Methods
`Products examined
`Three nasal sprays used for the treatment of perennial or
`seasonal allergic rhinitis were studied. Flixonase (Allen &
`Hanbury) contains fluticasone propionate as the active agent,
`benzalkonium chloride 0.02% (w/v) with phenylethyl alcohol
`0.25% (w/v) as preservatives and dextrose, carmellose and
`polysorbate 80 as excipients; it is sold in a 25-mL brown glass
`bottle and the three batches examined had a mean measured
`pH of 5.9. Rhinocort Aqua (Astra) contains budesonide as the
`active agent, 0.12% (w/v) potassium sorbate with 0.01% (w/v)
`disodium edetate as preservative and glucose, carmellose
`sodium, polysorbate 80 and hydrochloric acid as excipients; it
`is sold in a 6-mL brown glass bottle and the three batches
`examined had a measured mean pH of 4.1. Rynacrom (Fisons)
`
`Opiant Exhibit 2043
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00694
`Page 1
`
`

`

`1238
`
`N. A. HODGES ET AL
`
`contains 4% (w/v) sodium cromoglycate as the active agent
`and 0.01 % (w/v) benzalkonium chloride with disodium edetate
`as preservatives; it is sold in a 22-mL white plastic bottle and
`the three batches examined had a mean measured pH of 5.2
`
`48-72 h for the bacteria and fungi, respectively. The A. niger
`plates were examined at intervals during the incubation period
`to avoid recording artificially low counts resulting from con-
`fluence of colonies.
`Controls confirmed that the addition of 9.00 mL letheen
`broth diluent to 1.00 mL product eliminated the activity of all
`three preservative systems when the mixture was inoculated
`with low concentrations of the most sensitive of the test
`organisms (P. aeruginosa).
`Colonies were enumerated, and the extent of inactivation
`calculated as the log,,
`reduction in viable count in the
`inoculated product using the control count of the inoculum
`suspension as the baseline. Any test in which the time zero
`count exceeded the expected value by more than 15% was
`repeated.
`All testing was performed by a single, experienced operator
`whose reproducibility of viable counting was confirmed, all the
`tests were performed within a three-month period using the
`same batches of media and diluents throughout and the three
`products were tested simultaneously not sequentially. All
`batches of the three products were within their expiry dates
`throughout the testing period and were stored according to the
`manufacturer's instructions both before, and during, testing.
`
`Preservative testing
`Three different batches of each of the three products were
`tested on each of three occasions. The testing procedure was
`designed to satisfy, as far as possible, the requirements of both
`the European Pharmacopoeia (1994) and the United States
`Pharmacopeia (1995). Thus, freshly prepared suspensions of
`the ATCC-recommended strains of Staphylococcus aureus,
`Pseudomonas aeruginosa, Escherichia coli, Aspergillus niger
`and Candida albicans in sodium chloride-peptone water were
`mL- ' using a turbidity calibration graph. Accurately weighed
`standardized to a concentration of approximately 1 x lo8 cells
`quantities of these suspensions were added to the products in
`their original containers in order to give the USP-specified
`200-fold dilutions and initial concentrations of approximately
`5 x lo5 mL-'. There is a requirement that E. coli should be
`used as a supplementary test organism when the EP pre-
`servative efficacy test is applied to oral products, and E. coli
`was used in this present study: to satisfy the USP requirement;
`to provide a fuller assessment of the antimicrobial spectra of
`the preservative systems; and because the potential hazards
`Results
`presented by enteric pathogens in oral products also exist to a
`It was possible to reduce the large volume of data generated in
`lesser, but significant extent, in nasal products.
`Samples (1 .00 mL) were removed at time zero (nominal),
`this work by eliminating from Tables 1-3 and Fig. 1 all data
`derived from Flixonase, because this product permitted no
`and at 2.7, 14, 21 and 28 days during storage at a temperature
`survivors of any bacterial species or C. albicans at 48 h, nor
`of 20-25°C protected from light. The samples were subjected
`to decimal dilutions in letheen broth (Difco), and 0.20-mL
`any survivors of A. niger at 14 days; all data obtained with P.
`volumes of appropriate dilutions were surface-spread on tri-
`aeruginosa because this organism invariably showed no sur-
`vivors at 48 h; all data for S. aureus and E. coli after 7 days
`plicate tryptone soya agar plates incubated at 35°C for 48 h or
`because, invariably, no survivors were recorded; and part of
`on Sabouraud-dextrose agar plates incubated at 25°C for
`Table 1. Triplicate determinations of S. aureus and E. coli inactivation (log,,-, reduction in viable count m I - ' with time (days)) in three batches of
`rhinocort and rynacrom.
`
`Batch A
`S. aureus
`
`E. coli
`
`Batch B
`S. aureus
`
`E. coli
`
`Batch C
`S. aureus
`
`E. coli
`
`0
`
`2
`Rhinocort
`
`0.05
`0.04
`0.04
`0.16
`0.19
`0.08
`
`0.01
`- 0.02
`0.05
`0.08
`- 0.01
`0.06
`0.09
`- 0.05
`0.03
`0.01
`- 0.01
`0.29
`
`0.07
`0.49
`1.54
`2.44
`1.86
`1.97
`
`0.5 1
`0.34
`1 40
`1.07
`1.60
`1.17
`
`0.52
`0.15
`0.58
`0.92
`1.59
`1.19
`
`I
`
`> 4.06
`> 4.19
`> 4.26
`> 4.43
`> 4.60
`> 4.49
`
`> 4.19
`>4.16
`> 4.41
`> 4.49
`> 4.61
`> 4.46
`
`> 4.14
`> 4.20
`z 4.27
`4.20
`> 4.36
`> 4.60
`
`0
`
`2
`Rynacrom
`
`~~
`
`~
`
`~~
`
`0.10
`0.03
`0.03
`0.14
`0.18
`0.05
`
`-0.07
`0.00
`0.07
`0.15
`0.09
`0.09
`
`0.14
`-0.04
`0.06
`0.02
`-0.04
`0.09
`
`3.81
`> 4.20
`> 4.28
`2.17
`0.65
`0.94
`
`3.60
`3.03
`> 4.43
`1.31
`0.47
`1.52
`
`> 4.19
`3.23
`> 4.30
`1.08
`0.55
`1.71
`
`7
`
`>4.11
`> 4.20
`> 4.28
`> 4.43
`2.18
`4.50
`
`> 4.30
`> 4.21
`> 4.43
`2.95
`2.09
`> 4.49
`
`> 4.19
`> 4.31
`> 4.30
`3.42
`4.01
`> 4.66
`
`Opiant Exhibit 2043
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00694
`Page 2
`
`

`

`1239
`REPRODUCIBILITY IN PRESERVATIVE EFFICACY TESTS
`Table 2. Triplicate determinations of A. niger and C. albicans inactivation (loglo reduction in viable count mL-' with time) in three batches of
`rhinoCOrt and xynacrom.
`
`/
`
`Batch A
`A. niger
`
`C. albicms
`
`Batch B
`A. niger
`
`c. albicms
`
`Batch c
`A. niger
`
`C. albicans
`
`0
`
`0.19
`0.19
`0.09
`0.11
`0.13
`0.14
`- 0.02
`0.07
`0.03
`0.09
`0.07
`0.09
`
`0.04
`-
`0.19
`0.19
`0.09
`0.12
`- 0.04
`
`14
`
`Rhinocort
`
`21
`
`28
`
`0
`
`14
`
`Rynacrom
`
`21
`
`28
`
`1.81
`1.25
`1.54
`0.7 1
`0.29
`> 4.59
`1.33
`0.63
`1.71
`0.65
`0.36
`> 4.60
`
`1.40
`1.84
`1.34
`0.76
`1.84
`1.73
`
`1.89
`1.71
`1.75
`4.01
`1.56
`> 4.59
`1.67
`1.24
`2.42
`3.25
`> 4.46
`> 4.60
`
`2.29
`2.43
`1.74
`4.18
`> 4.35
`> 4.49
`
`2.89
`2.60
`4.27
`> 4.3 1
`3.98
`> 4.59
`2.47
`1.87
`3.68
`> 4.51
`> 4.46
`> 4.60
`
`2.64
`3.47
`2.55
`> 4.48
`> 4.35
`> 4.49
`
`- 0.03
`0.19
`- 0.01
`0.05
`0.09
`- 0.04
`0.06
`0.00
`0.04
`0.07
`0.03
`0.26
`
`0.04
`0.15
`0.23
`0.11
`- 0.06
`0.02
`
`2.50
`3.03
`3.53
`0.43
`0.39
`0.19
`
`2.32
`3.34
`4.39
`0.37
`0.23
`0.41
`
`2.90
`4.3 1
`2.92
`0.37
`0.33
`0.5 1
`
`3.48
`3.78
`3.63
`0.52
`0.65
`0.34
`
`2.90
`4.04
`3.27
`0.30
`0.46
`0.67
`
`3.30
`4.31
`3.62
`0.49
`0.67
`0.64
`
`>4.18
`4.00
`4.23
`0.73
`1.15
`0.72
`
`4.67
`3.86
`3.24
`0.76
`1.00
`0.90
`
`4.38
`>4.31
`3.20
`0.64
`1.06
`1 .04
`
`the data at 48 h and 7 days for C. albicans and A. niger because
`these time points are not considered in the EP or USP per-
`formance criteria.
`For a product to be considered effectively preserved the USP
`requires that the concentration of viable bacteria should be
`reduced one thousand fold within 14 days of inoculation and
`the concentration of yeasts and moulds should remain at or
`below the initial level during the first 14 days.
`The survival of S. aureus and E. coli is shown in Table 1,
`and that of A. niger and C. albicans in Table 2 for all tests
`performed on rhinocort and rynacrom. It is evident from these
`data and the facts stated above that all batches of the three
`products satisfied these performance criteria on each occasion.
`
`The USP further requires that the concentration of each test
`organism remains at or below these designated levels during
`the remainder of the test period, and this requirement, also, was
`satisfied in every instance but one, where a small numerical
`increase was observed in the A. niger viable count in rynacrom
`between days 14 and 28.
`All three products invariably rendered the bacteria unde-
`tectable at 14 days, and bacterial survivors in Flixonase were
`rarely detectable even in the time zero sample. Three batches
`of Flixonase each examined on three occasions and inoculated
`with three bacterial species resulted in a total of 27 time-zero
`observations for this product, and survivors were detected in
`only three of these despite the control experiments demon-
`
`Table 3. Frequency of failure to achieve European Pharmacopoeia performance criteria for each batch in relation to the number of observations.
`
`Bacteria
`
`Yeast and mould
`
`Overall test fails
`
`NO of observations
`
`Rhinocort
`
`Mean % failure**
`Flixonase
`
`Mean 9% failure
`R yn a c r o m
`
`Mean % failure
`
`Batch
`
`A
`B
`C
`
`A
`B
`C
`
`A
`B
`C
`
`9*
`
`
`criteria criteria
`Performance
`
`B B
`A
`n
`0
`5
`
`0 0
`6
`
`0 0
`6
`63
`
`0 0
`
`0
`0
`0
`0
`
`2
`3
`3
`30
`
`0
`0
`0
`0
`
`0
`n
`0
`0
`
`6*
`
`
`A A
`5
`5
`
`5 5
`
`6 6
`
`89 89
`
`0
`0
`0
`0
`
`3
`3
`3
`50
`
`B
`2
`3
`1
`33
`
`0
`0
`0
`0
`
`3
`3
`3
`50
`
`3
`
`A
`3
`3
`3
`100
`
`0
`0
`0
`0
`
`3
`3
`3
`100
`
`B
`2
`3
`1
`67
`
`0
`0
`0
`0
`
`3
`3
`3
`100
`
`*Three species of bacteria and two yeastdmoulds with triplicate tests give nine and six observations, respectively. **No product ever failed
`against P. aeruginosa.
`
`Opiant Exhibit 2043
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00694
`Page 3
`
`

`

`1240
`
`N. A. HODGES ET AL
`
`Batch A
`
`Batch B
`
`Batch C
`
`1 -
`
`h r
`
`J
`
`0
`
`10
`
`20
`
`300
`
`10
`Time (days)
`
`20
`
`300
`
`10
`
`20
`
`30
`
`FIG. 1. Replicate survivor plots for C. dbicans in three batches of rhinocort. Symbols represent replicate tests
`
`strating that the letheen broth diluent effectively eliminated the
`antibacterial activity of the preservative in that product. This
`indicates that the preservative activity is such that even the
`transient exposure of the bacteria between inoculation and the
`removal of the (nominal) time zero sample was sufficient to
`produce irreversible fatal damage to the cells.
`If nasal products are to be regarded as topicals (Anon 1993)
`the performance criteria recommended by the European
`Pharmacopoeia (the A criteria) are two decimal reductions in
`bacterial count within 48 h and three decimal reductions in
`seven days, combined with two decimal reductions in yeasts
`and moulds within 14 days. For both classes of organism there
`should be no subsequent increase in count above these levels
`during the remainder of the 28 day testing period. Flixonase
`invariably satisfied all the EP A performance criteria but rhi-
`nocort and rynacrom displayed certain weaknesses.
`All products showed satisfactory performance against P.
`aeruginosa, and inadequate antibacterial activity was seen with
`S. aureus and E. coli only. In no case were surviving bacteria
`detected in rhinocort after 7 days so the preservative defi-
`ciencies in this product were restricted to an inadequate bac-
`terial killing rate over the first 48 h. Although the rhinocort
`preservative system was more effective against E. coli than S.
`aureus, as shown by the consistently higher decimal reductions
`in viable count, it nevertheless failed to produce an adequate
`level of inactivation against these two species to satisfy the A
`criteria on every occasion except one (Table 1).
`In contrast, the rynacrom preservative was more active
`against S. aureus than E. coli and on every one of the nine
`occasions on which it was inoculated with S. aureus a satis-
`factory performance was observed at both 48 h and seven days.
`It failed the A criteria eight times out of nine and three times
`out of nine at the 48 h and seven day points respectively
`against E. coli (Table I), however.
`Seven days after inoculation the preservatives in both rhi-
`nocort and rynacrom had invariably achieved greater inacti-
`vation of A. niger than of C. albicans (data not shown) and this
`was also true for rynacrom at 14 days (Table 2), but not always
`for rhinocort at this time. Indeed, the activity of rynacrom
`against A. niger was sufficient to ensure that it never failed to
`satisfy the EP criteria against A. niger and its failures on nine
`occasions out of nine were always because of inadequate
`
`activity against C. albicans. In contrast, rhinocort failed the A
`criteria on almost every occasion against both organisms.
`Table 3 summarizes the performance of the various batches
`of the three products in terms of failures to meet the EP per-
`formance criteria in respect o f bacteria, yeasts and moulds,
`and overall. The EP describes a second set of criteria that must
`be satisfied in justified cases where the A criteria cannot be
`attained. The B criteria require the preservative to achieve
`three decimal reductions in the bacterial count and one decimal
`reduction in the yeast and mould count by the 14th day. It is
`clear that the preservative deficiencies in rhinocort and ryna-
`crom resulted in these products failing, on some occasions,
`even to achieve these minimal performance targets.
`The pattern of results for one of the three replicates of rhi-
`nocort Batch A inoculated with C. albicans was markedly
`different from the other two and this combination was, there-
`fore, examined on two further occasions. Similarly, rhinocort
`Batch B was examined a total of four times not three. It is clear
`from the results (Fig. 1) that this organism does not exhibit first
`order inactivation kinetics by this preservative but displays a
`marked shoulder on the survivor plot so that there is no
`appreciable reduction in viable count over the first seven days
`but there is rapid inactivation thereafter. Both the duration of
`the initial lag and the subsequent rate of decline varied from
`one replicate test to another, however.
`
`Discussion
`All three products contain two ingredients with antimicrobial
`activity, most probably because each of the four agents
`employed is considered to have gaps in its antimicrobial
`spectrum which necessitate
`its combination with others
`(Wallhauser 1984). Benzalkonium chloride is vulnerable to
`failure in the control of Pseudomonas species, and potassium
`sorbate is regarded primarily as an antifungal agent with
`deficiencies in its activity against bacteria, whereas the con-
`verse is true for phenylethyl alcohol. Disodium edetate is only
`weakly active when used alone and this activity is largely
`confined to Gram-negative species, thus it is better regarded as
`a potentiator rather than a preservative in its own right. Dis-
`odium edetate nevertheless has considerable value as a com-
`ponent of preservative systems, and Hart (1984) lists 32
`
`Opiant Exhibit 2043
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00694
`Page 4
`
`

`

`1241
`REPRODUCIBILITY IN PRESERVATIVE EFFICACY TESTS
`neither attainable nor reproducible, and he recorded a variation
`servatives, including benzalkonium chloride and potassium
`of up to three log cycles in the viable count reduction of
`ate, which it is known to potentiate. Synergy has also been
`@’& between phenylethyl alcohol and other preservatives,
`C. albicans and A. niger between duplicate tests performed
`on the same sample of a cetomacrogol cream. The data of
`knzalkonium chloride (Allwood & Baird 1994).
`::ding
`me combination of benzalkonium chloride and EDTA is
`Fig. 1 clearly illustrate the potential for such variations to
`occur in other products, particularly when the inactivation
`extensively employed as a preservative system in products
`kinetics comprise an initial lag followed by a period of rapid
`to sensitive tissues; e.g., of the 58 proprietary oph-
`::mic
`viability loss.
`products currently available on the UK market 47
`contain benzalkonium chloride, and of these, 17 also contain
`Although the variability recorded between the replicate
`determinations on the three batches of rhinocort inoculated
`&sodium edetate (Anon 1996). There has been some con-
`with C. albicans appears to be substantial, such marked var-
`troversy in recent years about the potential toxicity of ben-
`iation is not confined to this product-organism combination.
`fionium chloride. Concern has been expressed that it might
`be responsible for adverse effects on mucociliary function after
`For example, the three replicate determinations of A. niger
`survival in rynacrom batch B show log reductions at 14 days
`to decongestant products (Berg et a1 1994; Graf et a1
`19951, although studies using the frog-palate preparation,
`which vary from 2.32 to 4.39, and it is possible that variations
`of similar magnitude could have been recorded elsewhere if
`which resembles the nasal mucosa in retaining an intact mucus
`layer, have shown no effect on ciliary structure or mucus
`samples had been taken at more frequent time intervals. The
`extent of variation is less apparent if the organisms are all
`in concentrations up to 0.05% w/v (Braga et a1 1992).
`N~ adverse effects on nasal ciliated epithelium were, further-
`killed quickly, as in this study, when all the bacteria were
`more, seen in monkeys or rats which received steroid-con-
`killed by day 7; this leaves only the day-2 data by which to
`assess variability. If these for rhinocort are examined with the
`taining nasal sprays preserved with benzalkonium chloride for
`rynacrom data for A. niger and C. albicans at day 21 (selected
`28 days (Ainge et a1 1994). When assessing such conflicting
`because the organisms still had survivors in every instance so
`indications of potential toxicity, it is, however, essential not to
`there was not an artificial limit on the extent of the log
`overlook the obligate requirement for adequate preservation of
`reduction imposed by complete inactivation), marked varia-
`the product, and a reduction in preservative concentration
`tions between replicates are apparent. Indeed, the extent of this
`prompted by toxicity considerations is only of value if it does
`is such that analyses of variance show no evidence (at P =
`not create an even greater risk to the consumer as a result of
`0.05) that the apparent differences between batches are real,
`product spoilage by pathogenic organisms.
`except in the case of the three rhinocort batches inoculated
`Of the many factors known to influence preservative activity
`in pharmaceutical products (Denyer & Wallhauser 1990) the
`with E. coii, where the mean loglo reductions of 2.09, 1.28 and
`presence of non-ionic surfactants, sorption on to plastic con-
`1.23 were different (P = 0.025).
`Possible explanations for the observed variation in results
`tainers and sensitivity to product pH are the three most perti-
`include inadequate mixing of the inoculum and variation in fill
`nent to the present study. Benzalkonium chloride, phenylethyl
`volume of the product containers. Inadequate mixing of the
`alcohol and potassium sorbate are all known to be inactivated
`to varying extents by non-ionic surfactants (Wade & Weller
`inoculum was, however, considered improbable because the
`1994) and this might be considered as a possible contributory
`low product viscosities enabled easy mixing, and variation in
`factor to the failure of rhinocort, which contains polysorbate
`fill volume was unlikely to be of sufficient magnitude to
`80, to meet the EP preservative efficacy test performance cri-
`account for the observed differences. The most likely possi-
`teria. Both components of the Flixonase preservative system
`bility was considered to be variation in the temperature of
`are, however, also vulnerable to non-ionic surfactants. Sorption
`storage of the inoculated products. The storage temperature
`on to plastics, particularly polypropylene and polyethylene, has
`used conformed to the USP and EP requirements (between 20
`been shown to cause significant losses of benzalkonium
`and 25°C) but was not constant within this range. It is known
`chloride and other preservatives from solution (Autian 1968).
`that variation in temperature might influence the activity of
`antimicrobial chemicals (Denyer & Wallhauser 1990) and
`The poorer performance of the benzalkonium chloride-EDTA
`combination compared with the benzalkonium chloride-phe-
`greater reproducibility might be achieved by specifying a
`nylethyl alcohol system is, therefore, likely to be primarily
`single storage temperature rather than a range for the perfor-
`because of the higher benzalkonium chloride concentration in
`mance of the test.
`mixonase, but sorption by the plastic of the rynacrom container
`The data obtained in this study were produced using a single
`is a possible contributing factor. Preservative activity of weak
`experienced operator whose counting technique was shown to
`acids resides primarily, but not exclusively, in the undisso-
`be good. Throughout the work triplicate Petri dishes were used
`ciated fraction of the molecules and this fraction is directly
`for viable counts rather than the duplicates recommended in
`dependent upon pH. The pK, of sorbic acid is 4.76, so products
`the EP, so these, together with the use of the same batches of
`with pH less than this will contain the majority of their sorbate
`culture media and diluents, were steps which would be
`molecules in the undissociated state and thus have maximum
`expected to enhance reproducibility. It is, nevertheless, clear
`Preservative activity. The recorded pH of rhinocort was 4.1, at
`from Tables 1-3 that in certain situations marked differences
`which value approximately 80% of the sorbate molecules are
`arose between replicate determinations, and in the experience
`not ionized, and the data of Eklund (1983) suggest that the
`of the authors, such variation is not unusual in preservative
`growth-inhibitory action of the sorbate would not be appreci-
`efficacy tests. There appears to be very little published infor-
`ably enhanced by a further reduction in pH.
`mation to indicate the magnitude of the variability that might
`Davison (1988) commented that the recommendations of the
`be expected in test data, but there are statements in official
`1980 BP for preservative performance in topical products were
`compendia which indicate how different two viable counts
`
`Opiant Exhibit 2043
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00694
`Page 5
`
`

`

`~
`
`N. A. HODGES ET AL
`1242
`industries. In: Brown, M. R. W., Gilbert, P. (eds) Microbiological
`must be before the difference is to be considered significant. In
`Quality Assurance: a Guide Towards Relevance and Reproducibilib
`the context of validation of counting methods, the EP states
`of Inocula, CRC Press, New York, pp 149-162
`that if a limit of 100 organisms is specified in a monograph the
`Berg, PI., Lie, K., Steinsvig, S. K. (1994) The effect of decongestive
`maximum limit of acceptance should be 500 organisms. Fur-
`nose drops on human respiratory mucosa in vitro. Laryngosco&
`104: 1153-1 I58
`thermore, in this context of preservative efficacy testing, it was
`
`. ~ ~ . ~~.
`Braga, P. C., Piatti, G., Del Sasso, M., Bemini, A. J. (1992) The eff
`of calcitonin nasal preparations and their excipients on mucocillary ?
`proposed in Pharmeuropa (Anon 1993) that “no increase in
`viable count” should be taken to mean not more than half a log
`clearance in an ex-vivo frog palate test. J. Pharm. Pharmacol. 4:
`scale higher than the last value for which a criterion is given.
`938-940
`In view of these statements and the variability observed in
`Cooper, M. S. (1989) Preservative efficacy: compendial and regulatory
`issues. J. Parent. Sci. Technol. 43: 187-190
`some aspects of this study it is clear that a measure of repro-
`Davison, A. L. (1988) Preservative efficacy testing of non-skde
`ducibility is likely to be of major benefit in the interpretation of
`pharmaceuticals, cosmetics and toiletries and its limitations.
`preservative efficacy data and the authors suggest that this
`Bloomfield, S. F., Baird, R., Leak, R. E., Leech, R. (eds) Microbial
`should become an integral part of the test.
`Quality Assurance in Pharmaceuticals, Cosmetics and Toiletria,
`In conclusion, therefore, it is evident that marked differences
`Ellis Honvood, Chichester, pp 119-128
`Denyer, S. P., King, R. 0. (1988) Development of preservative
`in the performance of different preservative systems can arise
`systems. In: Bloomfield, S. F., Baird, R., Leak, R. E., Leech, R.
`even when the preservatives are employed at concentrations
`(eds) Microbial Quality Assurance in Pharmaceuticals, Cosmetics
`within the normal ranges (Wade & Weller 1994), and that
`and Toiletries, Ellis Honvood, Chichester, pp 156-170
`Denyer, S. P., Wallhauser, K.-H. (1990) Antimicrobial preservatives
`products which are on the borderline of acceptability in terms
`and their properties. In: Denyer, S. P., Baird, R. (eds) Guide
`of preservation might perform satisfactorily on one occasion
`Microbiological Control in Pharmaceuticals, Ellis Horwood, Chi-
`and fail on another. Such a situation would dictate that single
`Chester, pp 251-273
`point acceptance criteria are inadequate. In recognition of this,
`Eklund, T. J. (1983) The antimicrobial effect of dissociated and
`it is suggested that performance criteria are adopted together
`undissociated sorbic acid at different pH levels. J. Appl. Bacteriol.
`with realistic limits of error which adequately reflect the pre-
`54: 383-390
`European Pharmacopoeia (1994) 2nd edn, Maisonneuve SA, 57-Sainte
`cision of the viable counting procedures involved.
`Ruffine, France, VIII. 14
`Graf, P., Hallen, H., Juto, J.-E. (1995) Benzalkonium chloride in a
`decongestant nasal spray aggravates rhinitis medicamentosa in
`healthy volunteers. Clin. Exper. Allergy 25: 395-400
`Hart, J. R. (1984) Chelating agents as preservative potentiators. In:
`Kabara, J. J. (ed.) Cosmetic and Drug Preservation, Marcel Dekker,
`New York, pp 323-337
`Hodges, N. A,, Denyer, S. P. (1995) Preservative testing. In: Swar-
`brick, J., Boylan, A. (eds) Encyclopedia of Pharmaceutical Tech-
`nology, Marcel Dekker, New York, pp 21-37
`Spooner, D. F., Davison, A. L. (1993) The validity of the criteria for
`pharmacopoeia1 antimicrobial preservative efficacy tests. Pharm. J.
`251: 602-605
`US Pharmacopoeia (1995) 23rd rev., US Pharmacopeial Convention,
`Rockville, MD, p. 1681
`Wade, A,, Weller, P. J. (1994) Handbook of Pharmaceutical Excipi-
`ents, 2nd edn, The Pharmaceutical Press, London
`Wallhauser, K.-H. (1984) Antimicrobial preservatives used by the
`cosmetic industry. In: Kabara, J. 3. (ed.) Cosmetic and Drug
`Preservation. Marcel Dekker, New York, pp 605-745
`
`References
`Ainge, G., Bowles, J. A. K., McCormick, S. G., Richards, D. H.,
`Scales, M. D. C. (1994) Lack of deleterious effects of corticosteroid
`sprays containing benzakonium chloride on nasal ciliated epithe-
`lium. Drug Invest. 8: 127-133
`Allwood, M. C. (1986) Preservative efficacy testing of pharmaceuti-
`cals. Pharm. Int. 7: 172-175
`Allwood, M. C., Baird, R. (1994) Phenylethyl alcohol. In: Wade, A,,
`Weller, P. J. (eds) Handbook of Pharmaceutical Excipients, 2nd edn,
`The Pharmaceutical Press, London, p. 340
`Anonymous. (1993) Efficacy of antimicrobial preservation: rationale
`for the harmonization proposals. Pharmeuropa 5 : 355-358
`Anonymous. (1996) Monthly Index of Medical Specialities, Haymar-
`ket Publishing Services, London, February: p. 243
`Autian, J. (1968) Interrelationships of the properties and uses of
`plastics for parenterals. Bull. Parent. Drug Assoc. 22: 276-288
`Baird, R.