ARTICLE
The pathophysiological features of acne vulgaris are the overproduction
of sebum, the abnormal desquamation of the sebaceous follicle epithelium,
and the proliferation of Propionibacterium acnes (P. acnes) [1].
The colonization of P. acnes in the pilosebaceous duct gives rise
to the inflammatory lesions that characterize acne vulgaris, in the form
of red papules and pustules [2]. In Japan, systemic and topical antimicrobial
agents have been widely prescribed for the treatmen of inflammatory acne
lesions [3], and the oral tetracyclines, in particular, are the most common
systemic treatment for moderate acne. The topical antimicrobial drugs,
NDFX [4], CLDM, and lincomycin (LCM) [5] have also been used clinically
for inflammatory acne lesions. Apart from the bactericidal and bacteriostatic
effects of antibiotics, they can also inhibit neutrophil chemotactic factor
production by P. acnes, neutrophil phagocytosis, and oxygen metabolism
at below subminimal inhibitory concentrations (sub-MIC) [6, 7]. In Japan,
neither oral nor topical EM has been used for the treatment of acne. There
have been several reports of P. acnes highly resistant to EM, CLDM
and TC in the USA [8] and UK [9, 10], and we previously reported on resistant
P. acnes strains which were obtained at Kansai Medical University
in Japan. In this study, P. acnes strains were isolated from acne
lesions with the aim of determining the antimicrobial susceptibility of
the strains to 10 different antimicrobial agents.
Materials and methods
Materials
Bacteria: fifty strains of P. acnes isolated from acne lesions
were obtained from 31 female and 19 male acne patients treated at the
Department of Dermatology of Hyogo Prefectural Tsukaguchi Hospital and
the Kori Branch Hospital of Kansai Medical University between November
1994 and August 1995. The age of these patients ranged from 11 to 34 years
(mean: 21.3 years) and the history of acne ranged from 1 week to 15 years
(mean: 37.6 months). Of the 50 patients, 33 had not received any previous
oral or topical antimicrobial treatment for acne, and the remaining 17
had previously received antimicrobial treatment. Among the previously
treated patients, 10 had been treated with oral MINO (100 to 200 mg daily)
for a period ranging from 1 week to 7 years (mean: 2.4 months), and had
concomitantly used topical NDFX (6) or CLDM (3) or systemic metronidazole
(1). Two patients had been treated with oral roxithromycin (300 mg daily),
1 for one week and the other for 1.5 months, and had concomitantly used
topical CLDM and NDFX. Two patients had been treated with topical CLDM
alone, 1 for one week and the other for 2 weeks, and 2 had been treated
concomitantly with topical CLDM and an oral Chinese herb. One patient
had been treated with topical CLDM and NDFX for 1 week. None of the patients
had received topical benzoyl peroxide. In addition to the 50 strains obtained
from the above patients, the MIC for the 2 standard strains ATCC 6919
and ATCC 11827 were also determined.
Antimicrobial agents: 10 antimicrobial agents tested were nadifloxacin
(NDFX), ofloxacin (OFLX), erythromycin (EM), clindamycin (CLDM), tetracycline
(TC), minocycline (MINO), doxycycline (DOXY), ampicillin (ABPC), cephalexin
(CEX), and gentamycin (GM).
Methods
1) Identification of P. acnes strains. After cleansing
of the comedones (acne lesions) with 70% ethanol, the comedonal contents
were squeezed out and collected using a comedo extractor according to
Whiteside's method [11]. The contents were put into sterile, anaerobic
tubes, homogenized, and diluted, and then inoculated onto Brucella HK
agar medium (Kyokuto Seiyaku: Tokyo, Japan) supplemented with 5% defibrinated
horse blood, and anaerobically incubated for 4 days at 37° C in an
anaerobic gas pack (Becton Dickinson). The strains of P. acnes
were then identified using the Rap ID ANA II panel [12] (Innovative Diagnostic
Systems, Inc.). Identified strains of P. acnes were preserved in
semi-fluid GAM agar medium (Nissui, Tokyo, Japan).
2) Determination of MIC. The preserved P. acnes strains
were preincubated in GAM broth (Nissui, Tokyo, Japan) for 2 days at 37°
C. GAM agar (Nissui, Tokyo, Japan) plates containing each antimicrobial
agent at various concentrations were prepared, and 106 CFU/ml
of bacteria were immediately inoculated onto the plates and anaerobically
incubated for 2 days at 37° C. The MIC were then determined by the
two-fold agar dilution method according to the criteria of the Japan Society
of Chemotherapy [13].
Results
The antimicrobial susceptibility of 50 strains of P. acnes isolated
from acne lesions and 2 standard strains is shown in Table
I. The cumulative curve for the MIC is shown in Figure
1. Overall, the 50 strains of P. acnes isolated from acne
lesions were very susceptible to the antimicrobial agents tested, with
the exception of GM, EM, ABPC, and CLDM were the most potent drugs, followed
by MINO, NDFX, CEX, DOXY, OFLX, and TC. NDFX, OFLX, MINO, ABPC, CEX and
GM showed MIC of 0.1 to 0.78, 0.39 to 3.13, 0.1 to 3.13, 0.025 to 0.2,
0.2 to 1.56, 1.56 to 6.25 µg/ml, respectively. For both EM and CLDM,
the strains were generally distributed into 2 groups, a large group of
very susceptible strains and a small group of highly resistant strains,
with 2 highly resistant strains being seen for each of the 2 drugs. The
P. acnes strains were also very susceptible to ABPC, with MIC ranging
from 0.025 to 0.2 µg/ml. No strains of P. acnes resistant
(MIC >= 12.5 µg/ml) to NDFX, OFLX, CEX, or GM were seen. GM, aminoglucoside
antibiotic, to which P. acnes is usually not susceptible, showed
a relatively low potency, with MIC ranging from 1.56 to 6.25 µg/ml.
In terms of the MIC80, EM and ABPC were the most potent, followed
by CLDM, NDFX, MINO, CEX, DOXY, OFLX, TC and GM.
Most of the P. acnes strains isolated from acne lesions in patients
who had been treated with topical or systemic antimicrobial therapy were
very sensitive to all of the antimicrobial agents tested. Therefore, no
significant differences in MIC were seen between treated and non-treated
patients.
Table II shows the background
of 3 cases in which strains resistant to EM, CLDM, and DOXY were observed.
In case 1, clinical isolates of P. acnes were highly resistant
to EM and CLDM, but no previous treatment had been performed. In case
2, a strain resistant to EM and CLDM was observed similar to case 1, and
this case had previously been treated with oral Keigairengyoto, a Chinese
herb, and topical CLDM for 1 month. In case 3, a strain resistant to TC
and DOXY was observed, and this patient had previously been treated with
oral MINO, systemic metronidazole, and topical CLDM and NDFX for 2 months.
In all 3 of these patients, the history of acne was relatively long (6
to 15 years).
Discussion
In Japan, because the use of benzoyl peroxide, an anti-androgen drug
isotretinoin, and azelaic acid is not permitted, thus greatly restricting
the available options for the treatment for acne vulgaris, the main treatment
methods are still oral admisnistration of antimicrobials agents such as
tetracyclines and topical application of antibiotics such as CLDM, NDFX,
and LCM.
In our study, clinical isolates of P. acnes obtained from acne
lesions were generally very sensitive to antimicrobial agents. A high
percentage (66%) of the patients had no history of previous antimicrobial
treatment, and even in those patients who had received previous antimicrobial
treatment, the mean duration of therapy had been relatively short (oral
MINO: 2.4 months, oral roxithromycin: 3.5 weeks, topical CLDM: 7.2 weeks,
topical NDFX: 4.4 weeks). There were no differences in MIC between treated
and non-treated patients, and the reason for this might have been attributable
to the relatively short term of the antimicrobial treatment in our study.
We have previously shown that the resistance of P. acnes is not
acquired after either topical or oral antimicrobial treatment for short
periods (1 to 8 weeks) [14]. In this study, 2 strains of P. acnes,
highly resistant to EM and CLDM were observed: one was obtained after
topical CLDM had been used for 1 month, and the other was obtained from
a patient who had not received any treatment. The number of resistant
strains seen in this study was too small to evaluate the relationship
between bacterial resistance and previous treatment. We believe further
investigation concerning the relationship between bacterial resistance
and previous long-term antimicrobial treatment is needed. Eady et al.
[9, 10] reported on P. acnes strains that were resistant to EM
and TC. They [9] suggested that treatment with oral EM induces the development
of bacterial resistance not to only EM but also to CLDM, due to cross-resistance,
and they also suggested that the resistance of P. acnes is brought
about by methylation of 23S ribosomal RNA [9]. Co-resistance to EM and
CLDM is caused by the 23S ribosomal methylase which may be coded for the
erm (erythromycin ribosomal methylase) gene located on the plasmids
or transporins. Very recently, resistance in cutaneous Propionibacteria
has been shown to be acquired within the target site (23S rRNA) [15].
Crawford et al. [16] reported that in 20% of their patients treated
with EM, strains acquired resistance to EM, but that this resistance was
subsequently lost 2 months after discontinuation of treatment. In addition,
they suggested that the cross-resistance of P. acnes to EM and
CLDM may result from an alteration of the 50S ribosomal subunit. Leyden
et al.[8] reported that the mean MIC for P. acnes isolated
from patients having received long-term oral EM, increased to 100 times
that in non-treated patients.
With regard to TC, oral TC such as MINO and DOXY have been widely used
for the treatment of acne in Japan for the past 20 years. Nevertheless,
only 1 strain resistant to TC and DOXY was seen in this study, and that
strain was susceptible to MINO in spite of previous treatment with oral
MINO. Thus, there was no evident relationship between bacterial resistance
and previous treatment. In general, TC may prevent P. acnes from
acquiring resistance, and, in particular, no strains resistant to MINO
were observed in this study. Eady et al. [10] also emphasized that
MINO is the only antibiotic to which P. acnes has not acquired
resistance, and pointed out that P. acnes has a higher susceptibility
to MINO than to TC or DOXY, and our results were in agreement with their
findings. Leyden et al. [8] reported on P. acnes strains
resistant to MINO with the mean MIC of MINO against P. acnes that
had been treated long-term with TC increasing to 4 to 5 times higher than
that in non-treated patients. In spite of the frequent use of topical
CLDM in Japan, very few CLDM-resistant strains of P. acnes were
seen in this study. We suppose that the reason for this was attributable
to the absence of treatment with oral EM which has been shown to induce
CLDM resistance in Japan. Kligman [17] pointed out that topical CLDM does
not affect the acquisition of resistance by P. acnes, but that
topical EM [16] does.
NDFX [4] has recently been used in Japan as
a topical treatment for acne. In our previous report [4], we found that
the MIC of NDFX used against P. acnes ranged from 0.1 to 0.2µ
g/ml, while in the present study, it ranged from 0.1 to 0.78 µg/ml.
Although no strains resistant to NDFX have yet been found, its frequent
use in the future may cause P. acnes to acquire resistance. Therefore,
it will be necessary to monitor the susceptibility of P. acnes
to NDFX to check for the development of resistance.
We previously reported on the susceptibility of P. acnes to various
antibiotics in 1969 [18], 1976 [19], 1982 [20] and 1985 [3]. We found
strains resistant (MIC >= 12.5µ g/ml) to EM, CLDM, TC and DOXY.
No strains of P. acnes resistant to MINO were observed in the present
study or in 1976, 1982 or 1985. P. acnes has not shown resistance
to MINO in Japan, in spite of the frequent use of oral MINO, enabling
the drug to still be of use. Thus, MINO may prevent P. acnes from
acquiring resistance.
The strains of P. acnes highly resistant to EM and CLDM that
have been reported in the UK [9] and USA [8, 16] may have been due to
the frequent use of oral and topical EM. Oral EM is given as a secondary
treatment in the UK [9]. Eady et al. [10, 21] suggested that first,
the use of oral EM should be limited to those patients with no previous
exposure to the drug, and it should be discontinued after 6 months to
allow any resistant organisms to become less virulent, and second, benzoyl
peroxide may be useful for eliminating the resistance of P. acnes.
Thus, it is necessary to avoid the concomitant oral and topical administration
of chemically dissimilar antibiotics so as to reduce the risk of developing
resistance to both. In Japan, it might be wise not to use oral EM at all
in order to avoid the development of P. acnes resistance.
From the results of the present study, it seems that P. acnes
has not developed appreciable resistance to MINO in Japan. We believed
MINO still remains the most effective antibiotic available for the treatment
of acne. Patients who do not respond to antibiotic therapy should be screened
for resistant strains. Further detailed studies are needed to evaluate
the resistance of P. acnes before and after long-term topical or
systemic antimicrobial treatment.
REFERENCES
1. Leyden JJ. New understandings of the pathogenesis of acne. J Am
Acad Dermatol 1995; 32: S15-25.
2. Strauss JS, Kligman AM. The pathologic dynamics of acne vulgaris.
Arch Dermatol 1960; 82: 779-90.
3. Kurokawa I, Nishijima S, Asada Y. The antibiotic susceptibility of
Propionibacterium acnes: a 15-year bacteriological study. J
Dermatol 1988; 15: 149-54.
4. Kurokawa I, Akamatsu H, Nishijima S, Asada Y, Kawabata S. Clinical
and bacteriologic evaluation of OPC-7251 in patients with acne : a double-blind
group comparison study versus cream base. J Am Acad Dermatol
1991; 25: 674-81.
5. Kurokawa I, Tanaka M, Ikeda T. Topical aqua Kummelfeld containing
1% lincomycin for acne vulgaris. J Dermatol Treat 1992; 3: 209-11.
6. Akamatsu H, Nishijima S, Takahashi M, Ushijima T, Asada Y. Effects
of subminimal inhibitory concentrations of EM, TC, CLDM, MINO on the neutrophil
chemotactic factor in Propionibacterium acnes biotype 1-5. J
Dermatol 1991; 18: 247-51.
7. Akamatsu H, Niwa Y, Kurokawa I, Masuda R, Nishijima S, Asada Y. Effects
of subminimal inhibitory concentrations of minocycline on neutrophil chemotactic
factor production in comedonal bacteria, neutrophil phagocytosis and oxygen
metabolism. Arch Dermatol Res 1991; 283: 524-8.
8. Leyden JJ, McGinley KJ, Cavalieri S, Webster GF, Mills OH, Kligman
AM. Propionibacterium acnes resistance to antibiotics in acne patients.
J Am Acad Dermatol 1983; 8: 41-5.
9. Eady EA, Cove JH, Holland KT, Cunliffe WJ. Erythromycin resistant
Propionibacteria in antibiotic treated acne patients: association
with therapeutic failure. Br J Dermatol 1989; 121: 51-7.
10. Eady EA, Jones CE, Gardner KJ, Taylor JP, Cove JH, Cunliffe WJ.
Tetracycline-resistant Propionibacteria for acne patients are cross-resistant
to doxycycline, but sensitive to minocycline. Br J Dermatol 1993;
128: 556-60.
11. Whiteside JA, Voss JG. Incidence and lipolytic activity of Propionibacterium
acnes (Corynebacterium acnes group I) and Propionibacterium
granulosum (C. acnes, group II) in acne and in normal skin.
J Invest Dermatol 1973; 60: 94-7.
12. Celling DM, Schreckenberger PC. Clinical evaluation of the Rap ID-ANA
II panel for identification of anaerobic bacteria. J Clin Microbiol
1991; 29: 457-62.
13. Japan Society of Chemotherapy. Revised standardized method for determination
of the minimum inhibitory concentrations in bacteria. Chemotherapy
1981; 29: 76-9.
14. Nishijima S, Kurokawa I, Kawabata S. Sensitivity of Propionibacterium
acnes isolated from acne patients: comparative study of antimicrobial
agents. J Int Med Res 1996; 24: 473-7.
15. Ross JI, Eady EA, Cove JH, Jones CE, Ratyal AH, Miller YW, Vyakrnam
S, Cunliffe WJ. Clinical resistance to erythromycin and clindamycin in
cutaneous Propionibacterium isolated from acne patients is associated
with mutations in 23S rRNA. Antimicrobiob. Agents Chemother 1997;
41: 1162-5.
16. Crawford WW, Crawford IP, Stoughton RB, Cornell RC. Laboratory induction
and clinical occurrence of combined clindamycin and erythromycin resistance
in Corynebacterium acnes. J Invest Dermatol 1979; 7: 187-90.
17. Kligman AM. The treatment of acne. Part II: Antibacterials. Cutis
1995; 56: 315-6.
18. Asada Y, Nakai Y. Antibiotic susceptibility of Corynebacterium
acnes. Acta Dermatol (Kyoto) 1969; 64: 1-7 (in Japanese).
19. Ushioda T, Ikegami T, Asada Y. Antibiotic susceptibility of Propionibacterium
acnes. Hifu (Skin research) 1976; 18: 273-7 (in Japanese).
20. Nishijima S. Studies on Propionibacterium acnes isolated
from acne lesions, antibiotic susceptibility of biotyped strains and repression
of lipase activity of Propionibacterium acnes by tetracycline.
Hifu (Skin research) 1982; 24: 27-37 (in Japanese).
21. Eady EA, Bojar RA, Jones CE, Holland KT, Cunliffe WJ. The effects
of acne treatment with a combination of benzoyl peroxide and erythromycin
on skin carriage of erythromycin-resistant Propionibacteria. Br J Dermatol
1996; 134: 107-13.
|
Version imprimable
