Journal of Microbiological Methods 106 (2014) 104–109

Contents lists available at ScienceDirect

Journal of Microbiological Methods

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

An innovative challenge test for solid cosmetics using freeze-dried

microorganisms and electrical methods
M.R.S. Ferreira, F.R. Lourenço, M.T. Ohara, N.A. Bou-Chacra ⁎, T.J.A. Pinto
Department of Pharmacy, Faculty of Pharmaceutical Sciences, University of Sao Paulo, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Freeze-dried bacteria and fungi were used as inoculum in 28 days' PET. An electrical method was used in replace-
Received 30 May 2014 ment of the conventional plate count method. The use of freeze-dried microorganisms in association with the
Received in revised form 18 August 2014 electrical method can minimize the workload and the variability involved in PET for cosmetic powders.
Accepted 19 August 2014
© 2014 Elsevier B.V. All rights reserved.
Available online 28 August 2014

Keywords:
Preservative
Efficacy test
Solid cosmetic
Electrical methods

1. Introduction preservative system should not replace good manufacturing practices or

The cosmetic industry has been expanding and growing steadily in
recent decades. Sales in the beauty and personal care industry reached
US$68.7 billion in the United States and US$433.4 billion globally in
2012. The US industry will reach US$ 81.7 billion by 2017 according to
the Euromonitor International Forecast by Direct Selling News (Direct
Selling News, 2014). In Brazil, according to the 2012 Catalogue of the
Brazilian Association of Cosmetics, Toiletries, and Perfume Industry
(ABIHPEC), this sector earned R$ 29.4 billion in 2011 ex-factory
amounts, which represents 1.7% of Brazil's GDP. During this same peri-
od, the sector grew by 4.6%, more than Brazil's overall total GPD, which
grew by only 0.1%. Brazil ranks 3rd in the global makeup market. A joint
survey conducted by ABIHPEC and Booz & Company revealed that the
consumption of products in the sector is expected to grow at around
5% per year by 2015 (ABIHPEC, 2014).
The cosmetic industry is a very dynamic segment that readily em-
braces innovation to maintain a competitive edge and must offer high
quality products, involving esthetic considerations and microbiological
quality. Cosmetics have multiple uses, and it is not unusual that they
are stored under a variety of conditions, including warm and moist
bathrooms, women's handbags as well as forgotten in cars and then
used again. In this context, an adequate preservative system must be
used to assure product integrity during its shelf life (Russel, 2003) and
use (Orth, 1989; Magee et al., 1997; Russel, 2003). However, the
⁎ Corresponding author at: 580 Professor Lineu Prestes Avenue, Cidade Universitária,
Butantã, São Paulo, Brazil. Tel.: +55 11 38154418; fax: +55 11 30913628.
E-mail address: chacra@usp.br (N.A. Bou-Chacra).
reduce the viable microbial population of a nonsterile product (USP, 35).
The preservative efficacy test (PET) is performed to determine the
minimum effective concentration of antimicrobial preservatives re-
quired for adequate preservation of cosmetic and pharmaceutical prod-
ucts (Orth, 1991, 1997). This test assures the quality of the cosmetic
before it is marketed. Despite all efforts to continuously improve the
microbiological quality of cosmetics, microbial contamination of com-
mercially available products are still reported in the literature of devel-
oping countries (Abdelaziz et al., 1989; Okeke and Lamikanra, 2001;
Shaqra and Al-Groom, 2012; Tan et al., 2013), and even in Italy
(Campana et al., 2006).
Part of PET involves challenging a sample using different microor-
ganisms and determining survivors in specific time intervals by conven-
tional microbial counting (challenge test). A challenge test is performed
to verify if the preservative system (types of preservatives and their
concentrations) is capable of inhibiting microbial growth. This method
is applied only to liquid and semi-solid products, even though solid cos-
metic products can also suffer microbial contamination with repeated
use by consumers.
To evaluate samples, it is recommended to challenge them with mi-
crobial suspensions (Magee et al., 1997; British Pharmacopeia, 2010;
Unites States Pharmacopeia, 2012). However, for solid powders it is
difficult to guarantee microorganism homogeneity in the sample. This
difficulty leads to high variability in the results. A freeze-dried inoculum
could facilitate the sample homogeneity, since a mixture involving pow-
der is much easier to obtain. Considering this, the use of freeze-dried
microorganisms can be a practical alternative to PET.
Periodic evaluation of the survivors in preservative efficacy testing
using the pour plate technique involves extensive work in preparing

http://dx.doi.org/10.1016/j.mimet.2014.08.009
0167-7012/© 2014 Elsevier B.V. All rights reserved.
 M.R.S. Ferreira et al. / Journal of Microbiological Methods 106 (2014) 104–109
the material, executing the test and counting plates. Several alternatives
to traditional colony-count techniques have been developed.
Electrical methods have been proposed as promising alternatives.
This method is based on the modifications of electrical properties in
culture medium due to metabolism of the microorganism (Jeffrey
et al., 1989; Connolly et al., 1993, 1994; Zhou and King, 1995; Pinto
et al., 1999; Basa and Flores, 2000; Chorianopoulos et al., 2008; Yang.
and Bashir, 2008). Non-charged molecules are turned into charged
molecules for microorganism growth, so these electrical changes may
detect microbial growth by the automatic system using signals such as
conductance, capacitance and impedance (Pinto et al., 1999; Szita
et al., 2007). Among the advantages, this method requires less material,
labor and time (Connolly et al., 1993; Connolly et al., 1994). Many
authors found promising results using this method for bacteria determi-
nation (Connolly et al., 1993; Chorianopoulos et al., 2008; Priego et al.,
2011).
In this context, the aim of this study is to evaluate the application of
freeze-dried microorganisms as inoculum in the preservative efficacy
test for a solid cosmetic and to verify the applicability of the impedance
method to determine survivor microorganisms instead of using the
plate count technique for bacteria.
2. Materials and methods
2.1. Freeze-dried microorganisms in preservative efficacy test
2.1.1. Microorganisms
The test organisms consisted of strains of freeze-dried Staphylococcus
aureus ATCC 6538, Pseudomonas aeruginosa ATCC 9027, Candida albicans
ATCC 10231 and Aspergillus niger ATCC 16404 (Souza and Ohara, 2003).
2.1.2. Cosmetic sample
The samples used in this study consisted of a powdered eye-shadow
containing 1.0% (p/p) of Glydant plus® (DMDM hydantoin and
iodopropynyl butylcarbamate).
2.1.3. Inactivation of the preservative system
The preservative system was inactivated by using the decimal dilu-
tion of three different diluents: 1) Peptone 1.0%, sodium thiosulfate
0.6%, sodium bisulfite 0.25%, soy lecithin 0.7%, and polysorbate 80 0.5%
in distilled water to the bacteria (Diluent 1). 2) Peptone 2.0%, sodium
thiosulfate 0.6%, sodium bisulfite 0.1%, soy lecithin 1.0%, and polysorbate
80 3.0% in distilled water was added to yeast (Diluent 2). 3) Casein soy
broth 3.0%, soy lecithin 0.5% and polysorbate 20 4.0% in distilled water
(Diluent 3) to the mold. All of the media were previously sterilized.
2.1.4. Challenge test
One-vial containing 107 CFU of the freeze-dried microorganisms
(Souza and Ohara, 2003) was mixed gradually with 15 g of the sample.
The inoculated samples were maintained in glass bottles at room tem-
perature and samples were aseptically removed after 0 h, 2 h, 4 h, 8 h,
24 h, 48 h, 7 days, 14 days, 21 days and 28 days for viable counting.
The sample was diluted as described in the inactivation of the preserva-
tive system, and after 30 min of contact, 10-fold serial dilutions were
made.
The pour plate technique was performed on a 1-mL aliquot taken
from the appropriate dilution using Tryptic Soy Agar for the bacteria
and Sabouraud Dextrose Agar for the fungi. The incubation time was
48 h at 32 ± 2.5 °C for bacteria and yeast; for the mold it was 72 h at
22.5 ± 2.5 °C.
At least six tests were performed for each microorganism and the re-
sults were compared with the specifications of the official compendia
and CTFA.
105
2.2. Challenge test using electrical methods to determine the survivors
2.2.1. Microorganisms
The test organisms consisted of strains of freeze-dried S. aureus ATCC
6538, P. aeruginosa ATCC 9027, A. niger ATCC 16404 and C. albicans ATCC
10231.
2.2.2. Electrical device
A Bactometer® model-128 microbial monitoring system (Biolab
Merieux®) was used. It has eight plastic modules in one unit. Each mod-
ule contains sixteen molded wells. Each well contains two stainless steel
electrodes holding up to 2 mL of culture medium. A total of 128 samples
can be monitored simultaneously in one unit. The instrument uses three
detection modes: impedance, capacitance and conductance.
2.2.3. Signal choice (impedance, capacitance and conductance)
Volumes of 1.5 mL of GPMplus Bactometer® medium were added to
the wells at least 24 h before use.
A suspension containing 10% of eye shadow was prepared using
Diluent 2 as described in the inactivation of the preservative system
for both bacteria. This mixture was held stationary for 30 min before
the test execution.
The freeze-dried microorganism was recovered in 10 mL of saline
solution 0.9% (w/v), and was diluted to give a bacterial concentration
range from 102 to 107 CFU/mL. A total of 0.1 mL of each dilution was
withdrawn and transferred to a group of six wells containing the culture
media in order to obtain 10 to 106 CFU/well. The amount of 0.1 mL of
the diluted sample described above was added to each well.
The detection time (DT) was determined using impedance, capaci-
tance and conductance signals for each two wells.
The incubation temperature was 35 °C for bacteria for 24 h and
28 °C for the fungi for 100 h.
Plate counting was performed in parallel using 1.0 mL of the same
microorganism suspension used above diluted to present around
102 CFU/mL. Tryptic Soy Agar was used as media and the incubation
time was 24–48 h at 32 ± 2.5 °C.
The choice of the better signal was based on the data obtained in this
test and also a calibration was conducted to correlate plate count with
the 3 signals used.
2.2.4. Challenge test and determination of the survivors by impedance
method
The test was conducted as described in the challenge test using
freeze-dried microorganisms in the preservative efficacy test. Samples
were aseptically removed after 0 h, 2 h, 4 h, 8 h, 24 h, 48 h, 7 days,
14 days, 21 days and 28 days to determine the detection time (DT).
Two tests were performed in duplicate for S. aureus, P. aeruginosa and
A. niger.
The wells were incubated at 35 °C for 24 h to obtain the detection
time (DT). The number of survivors was established using the calibra-
tion curve.
2.2.5. Statistical analysis
PET results obtained from electrical and pour plate methods were
compared using a linear least square regression analysis. We assumed
that both impedance and pour plate methods were equivalent if the
confidential intervals for slope and intercept include the values 1.0
and 0.0, respectively.
3. Results
3.1. Use of freeze-dried microorganisms in the challenge test
Table 1 shows the average number of survivors of S. aureus,
P. aeruginosa, C. albicans and A. niger. On the 7th day, the number of
106 M.R.S. Ferreira et al. / Journal of Microbiological Methods 106 (2014) 104–109

survivors was lower than 10 (ten) Log CFU/g for the bacteria. C. albicans Table 2
presented a faster die-off with no recovery in 24 h. Correlation coefficient using S. aureus ATCC 6538, P. aeruginosa ATCC 9027, C. albicans
ATCC 10231 and A. niger ATCC 16404 for impedance, conductance e capacitance.
A. niger, presented the opposite behavior with a recovery rate of
4.7 CFU/g in 28 days. While the C. albicans was the least resistant micro- Microorganisms Correlation coefficient
organism, the A. niger was the most resistant. Impedance Conductance Capacitance

S. aureus ATCC 6538 0.9875 0.7699 0.9468

3.2. Challenge test using the electrical methods to determine survivors P. aeruginosa ATCC 9027 0.9748 0.8889 0.966
C. albicans ATCC 10321 – – 0.9914
A. niger ATCC 16404 – – 0.9512
The first aspect considered in this study was the choice of the best
signal for each microorganism.
The impedance and capacitance signals showed a better correlation
when the eye shadow and S. aureus were used (Table 2). 1997) present guidelines that describe the methodology used to deter-
As the impedance signal presented the best correlation coefficient, it mine the preservative efficacy for these products.
was decided that this signal would be chosen to construct the calibra- The use of freeze-dried microorganisms as inocula could be an inno-
tion curve for S. aureus. vative option to challenge the samples in the preservative efficacy test-
When P. aeruginosa was used as test microorganism, capacitance ing (Souza and Ohara, 2003). Entis (1984, 1989) used freeze-dried
presented faster results compared to conductance or impedance and microorganisms to validate alternative methods for counting viable mi-
also presented a good correlation coefficient of 0.966 (Table 2). croorganisms. Also Cremieux et al. (2005) used dried inocula prepared
For the fungi, the signal chosen was capacitance because it was the by membrane filtration followed by drying in an incubator as a proposal
only signal that presented any correlation at all. for the evaluation of the preservative efficacy in wet wipes. According to
The calibration curve for all microorganisms tested was constructed the Federation Internationale Pharmaceutique (FIP) (1984) the lyophi-
using the impedance signal for S. aureus and capacitance for the other lized inocula were specially developed to contaminate products such as
microorganisms and; an equation was obtained where Y represents oils. The main advantage of preserving lyophilized cultures is that they
Log CFU and X represents DT (Table 3). Using this equation it was pos- prevent loss of genetic resistance factors (Magee et al., 1997).
sible to determine the amount of each microorganism in determining In a previous work, Souza and Ohara (2003) verified that the micro-
the number of survivors in the preservative efficacy test. organisms typically used in the preservative efficacy test maintain good
Table 4 presents the detection time (DT) and Log of CFU/g calculated viability after the freeze-dried process and also the minimum inhibitory
using the equations obtained in the calibration curve (Table 3). It was concentration (MIC) using methylparaben, propylparaben and a combi-
possible to determine the CFU/g of S. aureus 7 days after the beginning nation of DMDM hydantoin and iodopropynyl butylcarbamate remains
of the evaluation. For P. aeruginosa and A. niger, this determination the same before the process.
was possible only after 48 h. C. albicans presented no detection time According to our results, the use of freeze-dried microorganisms
even just after the sample inoculation. improved the sample inoculation for the challenge test, which results
The linear least square regression between electrical and pour plate in the most homogeneous mixture between sample and inoculum.
methods considering the microorganisms used in this study is present- Microorganisms kept in culture media can also suffer modifications
ed in Fig. 1. in their characteristics (Curry et al., 1993), bear contamination and
lose their viability (Lapage et al., 2001). Considering this aspect, in its
24th edition, the United States Pharmacopeia predicts that the viable
4. Discussion microorganism used in tests must be removed from the original ATCC
culture no more than five passages in the antimicrobial effectiveness
4.1. Challenge test using freeze-dried microorganism in the challenge test testing (Unites States Pharmacopeia, 2000).
When the results were compared to the British and European Phar-
The preservative efficacy test is described only for pharmaceuticals macopeias (European Pharmacopeia, 2007; British Pharmacopeia,
in the official compendia (European Pharmacopeia, 2007; British 2010) using criteria A for topical preparation which implies a Log reduc-
Pharmacopeia, 2010; Unites States Pharmacopeia, 2012). In 1981, The tion of 2.0 in 2 days, 3.0 in seven days, and no increase until the end of
Personal Care Products Council adopted the name of Cosmetic Toiletry the test, all the tests passed for bacteria (Table 1).
and Fragrance Association (CTFA) and presented guidelines for a meth- Comparing to the specification of USP 35 (Unites States
odology used to determine the preservative efficacy for cosmetics. Also, Pharmacopeia, 2012) for optic products which demands a reduction of
the American Society for Testing and Materials (ASTM) (Magee et al., 1.0 log from the initial calculated count at 7 days and no increase from
the 14 day count at 28 days the samples also passed in the test, once
Table 1 more for both bacteria.
Average Log CFU/g of number of survivors of S. aureus ATCC 6538, P. aeruginosa ATCC 9027 If the criteria of CTFA (Magee et al., 1997) for water miscible person-
A. niger ATCC 16604 and C. albicans ATCC 10231 in the preservative efficacy test. al care is used, all samples also passed the test, which requires a reduc-
Time S. aureus P. aeruginosa C. albicans A. niger
tion of 99.9% in seven days and no increase to the end of the test period.
(Log CFU/g) (Log CFU/g) (Log CFU/g) (Log CFU/g) All tests performed presented a similar response compared to the
specifications indicating that freeze-dried organisms could be used as
T0 6.4 4.8 5.9 5.6
T2 6.3 4.5 4.7 – challenges in the preservative efficacy testing for both bacteria.
T4 6.2 4.3 3.4 5.6
T8 – – – 5.5
T24 4.7 2.9 b1 5.6 Table 3
T48 2.8 2.4 b1 5.2 Calibration curve equation for S. aureus (impedance), P. aeruginosa, C. albicans and A.
T7 b1 b1 b1 4.9 niger (capacitance).
T14 b1 b1 b1 4.9
Microorganisms Equation
T21 b1 b1 b1 4.6
T28 b1 b1 b1 4.7 S. aureus ATCC 6538 Y = −0.597X + 7.9865
P. aeruginosa ATCC 9027 Y = −0.5921X +8.2089
T0, T2, T4, T8, T24 and T48: enumeration immediately after contamination, 2, 4, 8, 24 and
C. albicans ATCC 10321 Y = −0.1736X + 7.3609
48 h, respectively.
A. niger ATCC 16404 Y = −0.1332X + 9.6186
T7, T14, T21 and T28: enumeration 7, 14, 21 and 28 days, respectively.
 M.R.S. Ferreira et al. / Journal of Microbiological Methods 106 (2014) 104–109 107

Table 4
Detection Time (DT) and number of survivors (Log CFU/g) calculated using the calibration curve for S. aureus (impedance), P. aeruginosa and A. niger (capacitance).

Time S. aureus P. aeruginosa A. niger

DT (h) Log CFU/g DT (h) Log CFU/g DT (h) Log CFU/g

T0 6.50 6.13 8.30 5.29 37.58 6.61

5.80 6.55 8.20 5.35 37.90 6.57
T2 7.10 5.78 9.20 4.76 41.60 6.14
5.50 6.13 8.40 5.25 38.38 6.51
T4 7.30 5.66 10.00 4.29 – –
7.00 5.84 9.50 4.58 – –
T8 8.40 5.01 11.50 3.40 45.95 5.49
7.90 5.30 11.30 3.52 44.70 5.66
T24 10.50 3.76 16.7 1.67 49.05 5.09
11.70 3.05 14.62 2.44 49.03 4.65
T48 13.62 2.08 20.00 b1 49.23 5.06
14.82 2.79 19.4 b1 14.82 9.64
T7 29.6 b1 N48 b1 –
39.27 b1 N48 b1 –
T14 N48 b1 N48 b1 N100 b1
N48 b1 N48 b1 N100 b1
T21 N48 b1 N48 b1 N100 b1
N48 b1 N48 b1 N100 b1
T28 N48 b1 N48 b1 N100 b1
N48 b1 N48 b1 N100 b1

T0, T2, T4, T8, T24 and T48: enumeration immediately after contamination, 2, 4, 8, 24 and 48 h, respectively.
T7, T14, T21 and T28: enumeration 7, 14, 21 and 28 days, respectively.

Faster reduction was observed when C. albicans was used as a chal-
lenge organism (Table 1) attending all specifications used in this study
(Magee et al., 1997; European Pharmacopeia, 2007; British Pharmaco-
peia, 2010; Unites States Pharmacopeia, 2000.
A. niger (Table 1) was the most resistant microorganisms evaluated
in the preservative system. Nonetheless this mold was within the guide-
lines for optic products stated by the United States pharmacopeia prod-
ucts having no increase from the initial calculation for the 28 day test
considering that it is necessary to evaluate the burden on the 7th,
14th and 28th day. On the other hand, the test was not consistent
with other guidelines (Magee et al., 1997; European Pharmacopeia,
2007; British Pharmacopeia, 2010).
Once more, the use of freeze-dried inoculum presented good repeat-
ability among the tests performed for fungi.
It is possible to conclude from the results that a freeze-dried inocu-
lum could be used in the preservative efficacy test to challenge solid
samples. The main goal of this paper was to provide an innovative chal-
lenge test using freeze-dried microorganisms and an electrical method
to evaluate microbial growth. It was not our purpose to optimize the
preservative system, considering different types of preservatives and
their concentration in the formulation. Further studies using other pre-
servative systems are needed to provide a better understanding of using
this method in PET, since only one preservative was evaluated in this
study.
Fig. 1. Linear least square regression analysis for electrical and pour plate results using
S. aureus, P. aeruginosa and A. niger.
It is also important to consider that this inoculum could be used in
any sort of sample including liquid or semi-solid.
4.2. Challenge test using the electrical methods to determine survivors
The traditional method using the plate count technique has been
conventionally used in the preservative efficacy test to verify the micro-
bial reduction (Curry et al., 1993; British Pharmacopeia, 2010; Unites
States Pharmacopeia, 2012). However methods considered more rapid
have been recommended to determine the number of microorganisms,
including among them, electrical methods (DePasquale et al., 1985;
Jeffrey et al., 1989; Connolly et al., 1993; Muscatiello, 1993; Connolly
et al., 1994; Pinto et al., 1999).
These methods are based on the detection time (DT), which is the
time that the microorganism takes to reach the threshold level neces-
sary for its detection (approximately 107 CFU/mL). Nevertheless this
time depends on the generation time of each microorganism presenting
different curves (Jeffrey et al., 1989).
On the other hand, if this method is used to determine the number of
survivors in the preservative efficacy test that consists of using pure cul-
tures of microorganism, the interference of the generation time would
be eliminated and calibration curves could be constructed. Connolly
et al. (1993) concluded that the impedance method is the only one
that presented satisfactory results to be used in the challenge test
compared to the direct epifluorescence technique (DEFT) and ATP
bioluminescence (ATP –B).
Despite that the major advantage of this method consists of reducing
analysis time, this is not a determinant factor, since in the preservative
efficacy, the last evaluation of the sample burden is performed 28 days
after its contamination. The major advantage claimed for these methods
is that they are more economical in the use of materials, labor and the
results are obtained within a shorter time period than the 18–60 h typ-
ical for colony count methods (Connolly et al., 1993, 1994). Its use also
spares the need for multiple dilutions and exhaustive colony counting
in plates using the conventional method (DePasquale et al., 1985).
Comparing the results obtained using impedance for determining
the burden and the traditional method, S. aureus presented similar be-
havior as it was possible to detect this microorganism in a 48-hour
test and no recovery for the remaining 28 days. When these findings
are compared with the specifications adopted in this study (Magee
108 M.R.S. Ferreira et al. / Journal of Microbiological Methods 106 (2014) 104–109
et al., 1997; European Pharmacopeia, 2007; British Pharmacopeia, 2010;
Unites States Pharmacopeia, 2012) the response is similar to those ob-
tained using plate counting to determine the amount of microorganisms
(Table 1).
In this study impedance and capacitance signals showed better
correlation when using S. aureus (Table 2). Pinto et al., 1999 observed
that conductance provided the best results when evaluating eye shad-
ow. These results showed that an adequate signal depends not only on
the type of the sample (if liquid or solid for example) tested, but also
its formula can interfere with the results.
The findings of Connolly et al. (1993) were also considered in this
study . P. aeruginosa presented detectable DT only 24 h after the test
(Table 4). The outcome differed when the plate count technique is
used. In this case burdens of around 2.0 Logs were still detected during
the first 48-hour test (Table 1). According to DePasquale (1985)
P. aeruginosa is a very sensitive bacteria and 10 to 103/g die before
detection time. However this point is questionable since in this study
it was possible to recover amounts below 10 for this microorganism to
construct the calibration curve.
Of course two tests for each microorganism are too few to predict if
the electrical methods could be used as a substitute to determine the
amount of microorganisms in the challenged sample, but they were
helpful to provide an idea of its feasibility.
It is also important to notice that the linear least square regression
analysis showed a correlation (r = 0.9039 and lack-of-fit p-value =
0.94) between electrical and pour plate methods considering all the mi-
croorganisms used in this study (Fig. 1). According to statistical analysis,
electrical methods provided results equivalent to those obtained by the
pour plate method. It is shown by the confidence intervals for slope
(IC95% = 1.101 ± 0.367) and intercept (IC95% = − 0.647 ± 1.869)
in the equation Y = -0.647 + 1.101X (where Y is electrical results
and X is pour plate results). After the 7th day, all results were found to
be below detection limits of the pour plate method. Electrical methods
provide similar results, which corroborate the equivalence of these
methods.
In addition, it was possible to construct a calibration curve for
C. albicans using the capacitance as signal (Table 3) and determine the
curve equation (Table 4), but it was not possible to conduct the test
due its rapid rate of killing. Just 2 h after the sample contamination,
the DT was above 100 h.
When the test was performed using A. niger, after 14 days of assay,
the detection time was also above 100 h. This result made it impossible
to determine the microbial load by using the curve equation at this
point.
5. Conclusion
It can be concluded that freeze-dried microorganisms could be used
to inoculate the sample in the preservative efficacy test for solid cos-
metics, but further studies using other preservative systems should be
evaluated. The results using the electrical methods to evaluate the bur-
den of the contaminated samples also demonstrated that these methods
provide results equivalent to those obtained from the pour plate meth-
od, except for C. albicans. Further studies would be necessary to better
understand the use of the electrical methods, particularly for C. albicans.
6. Conflict of interest
The authors report no conflict of interest in this work.
Acknowledgements
Faculty of Pharmaceutical Sciences of the University of São Paulo,
Natura Cosmetics, bioMérieux and CAPES for the financial support. Jim
Hesson of AcademicEnglishSolutions.com revised the English.
References
Abdelaziz, A.A., Ashour, M.S.E., Hefni, H., El-Tayebt, O.M., 1989. Microbial contamina-
tion of cosmetics and personal care items in Egypt—shaving creams and sham-
poos. J. Clin. Pharm. Ther. 14, 29–34. http://dx.doi.org/10.1111/j.1365-2710.
1989.tb00218.x.
AHBIPEC (The Brazilian's Toiletries, Fragrance and Cosmetics Association), 2014. http://
www.abihpec.org.br/wp-content/uploads/2012/12/ABIHPEC_2012_internet.pdf
((accessed April, 2014) (12:25, 04/12/2014)).
Basa, E.R., Flores, E., 2000. Preservative efficacy testing for stability using impedance.
Cosmet. Toilet 155, 61–66.
British Pharmacopeia, 2010. BP. Her majesty's Stationery Office, London.
Campana, R., Scesa, C., Patrone, V., Vittoria, E., Baffone, W., 2006. Microbiological study of
cosmetic products during their use by consumers: health risk and efficacy of preser-
vative. Lett. App. Micro. 43, 301–306. http://dx.doi.org/10.1111/j.1472-765X.2006.
01952.x.
Chorianopoulos, N.G., Giaouris, E.D., Skandamis, P.N., Haroutounian, S.A., Nychas, G.J.
E., 2008. Disinfectant test against monoculture and mixed-culture biofilms com-
posed of technological, spoilage and pathogenic bacteria: bactericidal effect of
essential oil and hydrosol of Satureja thymbra and comparison with standard
acid-base sanitizers. J. App. Microbiol. 44, 1586–1596. http://dx.doi.org/10.1111/j.
1365-2672.2007.03694.x.
Connolly, P., Bloomfield, S.F., Denyer, S.P., 1993. A study of the use of rapid methods for
preservative efficacy testing of pharmaceuticals and cosmetics. J. Appl. Bacteriol. 75,
456–462. http://dx.doi.org/10.1111/j.1365-2672.1993.tb02802.x.
Connolly, P., Bloomfield, S.F., Denyer, S.P., 1994. The use of impedance for preservative
efficacy testing of pharmaceuticals and cosmetic products. J. Appl. Bacteriol. 76,
68–74. http://dx.doi.org/10.1111/j.1365-2672.1994.tb04417.x.
Cremieux, A., Cupferman, S., Lens, C., 2005. Method for evaluation of the efficacy of anti-
microbial preservatives in cosmetic wet wipes. Int. J. Cosmet. Sci. 27, 223–236. http://
dx.doi.org/10.1111/j.1467-2494.2005.00267.
Curry, A.S., Graf, J.G., McEwen, G.N., 1993. Microbiology Guidelines. CFTA, Washington.
DePasquale, D., Kahn, P., Firstenberg-Eden, R., 1985. Impedimetric determination of anti-
microbial preservative efficacy: an automated challenge test. Las Vegas: 1985 Annual
Meeting of the American Society of Microbiology.
Direct Selling News, 2014. http://directsellingnews.com/index.php/view/ibd_reports_
beauty_industry_growth_and_trends#.UlqSstLkvBM ((accessed April, 2014) (12:20,
04/12/2014)).
Entis, P., 1984. Enumeration of total coliforms, and Escherichia coli in foods by
hydrophobic grid membrane filter: collaborative study. J. Ass. Off. Anal. Chem.
67, 812–823.
Entis, P., 1989. Hydrophobic grid membrane filter/MUG method for total coliform and
Escherichia coli enumeration in foods: collaborative study. J. Ass. Off. Anal. Chem.
72, 936–950.
European Pharmacopeia (2007) EP 3. Paris: Ed Supplement.
Federation Internationale Pharmaceutique (FIP), 1984. The test for effectiveness of anti-
microbial preservation of pharmaceuticals. Cosmetic and Drug Preservation: Princi-
ples and Practice. 1. Marcel Dekker, New York, pp. 423–440.
Jeffrey, M., Kaiserman, J.M., Wolf, B.A., 1989. A rapid impedimetric procedure to
determine bacterial content in cosmetic formulations. J. Soc. Cosmet. Chem. 40,
21–31.
Lapage, S.P., Redway, K.F.,Rudge R. Preservation of microorganism. In: Laskin, A.I. and
Lechevalier, H.A. Chemical rubber company handbook of microbiology. Florida: Chemi-
cal Rubber, 743-758.
Magee, M.A., Sutton, S.V., Brannan, D.K., 1997. Cosmetic Microbiology — A Practical Hand-
book. CRC Press, Boca Raton, pp. 94–140.
Muscatiello, M.J., 1993. CTFA's preservation guidelines: a historical perspective and
review. Cosmet. Toiletries 108, 53–59.
Okeke, I.N., Lamikanra, A., 2001. Bacteriological quality of skin-moisturizing creams and
lotions distributed in a tropical developing country. J. App. Microbiol. 91, 922–928.
http://dx.doi.org/10.1046/j.1365-2672.2001.01456.x.
Orth, D.S., 1989. Microbiological considerations in cosmetic formula development
and evaluation. I. Microbiological quality of a product. Cosmet. Toiletries 104,
52–64.
Orth, D.S., 1991. Standardizing preservative efficacy test data. Cosmet. Toiletries 106,
45–51.
Orth, D.S., 1997. Preservative efficacy testing: a review of various testing methods and
their reliability. Cosmet. Toiletries 112, 59–62.
Pinto, T.J.A., Ohara, M.T., Kaneko, T.M., Souza, M.R.S.E.L., Bou Chacra, N.A., Flauto, S.,
Schereinger, G.M., 1999. Validation Study for Preservative Efficacy Testing of
Cosmetics by the Impedance Method. XIV Congreso Latinoamericano e Ibérico de
Químicos e Cosmeticos. Santiago, Sociedad Chilena de Químicos Cosméticos.
Priego, R., Medina, L.M., Jordano, R., 2011. Bactometer system versus traditional methods
for monitoring bacteria populations in salsichón during its ripening process. J. Food
Prot. 74, 145–148.
Russel, A.D., 2003. Challenge testing: principles and practice. Int. J. Cosmet. Sci. 25,
147–153. http://dx.doi.org/10.1046/j.1467-2494.2003.00179.x.
Shaqra, Q.M.A., Al-Groom, R.M., 2012. Microbiological quality of hair and skin cosmetics
manufactured in Jordan. Int. Biodeterior. Biodegrad. 69, 69–72. http://dx.doi.org/10.
1016/j.ibiod.2011.12.009.
Souza, M.R.S.E.L., Ohara, M.T., 2003. The preservative efficacy testing method for
powdered eye shadows. J. Cosmet. Sci. 54, 411–420.
Szita, G., Gyenes, M., Soós, L., Rétfalvi, T., Békési, G., Csikó, G., Bernárth, S., 2007. Detection
of Pseudomonas aeruginosa in water samples using a novel synthetic medium and
impedimetric technology. App. Microbiol. 45, 42–46. http://dx.doi.org/10.1111/j.
1472-765X.2007.02145.x.
 M.R.S. Ferreira et al. / Journal of Microbiological Methods 106 (2014) 104–109 109

Tan, A.S.B., Tuysuz, M., Otuk, G., 2013. Investigation of preservative efficacy and microbiolog-
ical content of some cosmetics found on the market. Pakistan J. Pharm. Sci. 26, 153–157.
Unites States Pharmacopeia, 2000. USP 24. Unites States Pharmacopoeial Convention,
Rockville.
Unites States Pharmacopeia, 2012. USP 35. Unites States Pharmacopoeial Convention,
Rockville.
Yang, L., Bashir, R., 2008. Electrical/electrochemical impedance for rapid detection of
foodborne pathogenic bacteria. Biotechnol. Adv. 26, 135–150. http://dx.doi.org/10.
1016/j.biotechadv.2007.10.003.
Zhou, X., King, V.M., 1995. An impedimetric method for rapid screening of cosmetic
preservatives. J. Ind. Microbiol. 15, 103–107.