Keywords
Cutibacterium acnes
- Acne vulgaris - Antibigram - In vitro susceptibility -
Propionibacterium acnes
Introduction
Cutibacterium acnes (C. acnes), an anaerobic gram-positive bacillus formerly known as Propionibacterium acnes, is a significant emerging pathogen that has lately been incriminated in the pathogenesis
of acne vulgaris as well as life-threatening infections such as endocarditis, intravascular
infections, post-craniotomy infections, endophthalmitis, and septic arthritis.[1]
[2]
[3]
[4] The involvement of this bacterium is also being increasingly reported in the context
of orthopedic and cardiac prosthetic infections, infections of breast implants, intraocular
lenses, and ventriculoperitoneal shunts.[5]
It has been the most classical and consistent pathogen associated with acne vulgaris.[6] This organism is abundant in sebum-rich areas (which are prone to acne), causes
blockage of pilosebaceous glands by making biofilms, and also plays an important role
in initiation and propagation of inflammation.[7] Biofilm formation makes them highly resistant to antimicrobial agents and is capable
of causing chronic persistent infection that is difficult to treat.
Therapeutically, the most effective antibacterial agents in acne vulgaris target C. acnes.[6] Optimal antibiotic therapy, directed against C. acnes, is the key to the treatment of these infections. But the widespread and long-term
use of oral and topical Macrolides and Tetracyclines has resulted in significant dissemination
of cross-resistant strains of cutibacteria in many parts of the world. Resistant strains
have been reported to be widely prevalent throughout Europe, Japan, Korea, Singapore,
Australia, and USA.[8]
[9]
[10]
Information regarding the antimicrobial susceptibility profile and resistance mechanisms
of C. acnes is relatively scarce from India.[11] Being an anaerobic slow-growing organism, the isolation and antibiotic sensitivity
testing of this bacterium are technically challenging tasks that require rapid transport
of samples to the microbiology laboratory; use of special transport media; infrastructure
for performing anaerobic culture; skilled manpower; and relatively lengthy incubation
periods. Consequently, isolation and characterization of this organism is seldom undertaken
in clinical microbiology laboratories.
In view of the paucity of data in the Indian context and the need to tailor antibiotic
therapy according to local antibiogram patterns, the present study was conducted with
the objective of analyzing a repertoire of clinical isolates of C. acnes from patients afflicted with acne vulgaris, evaluating the outcome in these patients
following standard therapy with prescribed antibiotics, and delineating the association
between clinical response and in vitro susceptibility of the recovered isolates with
clinically relevant antibiotics.
Materials and Methods
Ethical Considerations
The study protocol was approved by the institutional human ethics committee, and informed
consent was obtained from each patient prior to recruitment.
Consecutive patients, attending the Dermatology OPD of a tertiary care teaching hospital
in Central India with acne vulgaris, were recruited into the study over a period of
2 months. Patients already receiving antibiotic treatment at the time of presentation
or within the last 4 weeks were excluded from the study. Relevant demographic and
clinical details of the recruited patients were recorded in a predesigned case study
form. The treatment given to the patients, following diagnostic workup, was also recorded.
The patients were called for follow- up at the end of 1 month during which the outcome
of treatment was recorded.
A single skin swab was collected from each recruited participant for anaerobic culture
of C. acnes. The sample from acne lesions was collected by applying firm pressure and rubbing
against the lesions with a transport swab. The swab was moistened in sodium phosphate
buffer (0.075 mol/L; pH 7.9) containing 0.1% Triton-X 100. After specimen collection,
the swabs were immediately transported to the microbiology laboratory where they were
cultured in 5% sheep blood agar under anaerobic conditions. The swabs were gently
rolled over the surface of a portion of the plate and then streaked to obtain isolated
colonies. The anaerobic incubation was done using the GasPak Anaerobic System of GasPaks
(HiMedia, Mumbai) at 37°C for 72 hrs. Colonies of C. acnes were identified among the recovered isolates using phenotypic features such as cultural
characteristics, microscopic morphology, and standard biochemical tests.[12] The primary isolate of C. acnes was then subcultured in 5% sheep blood agar to obtain a pure growth of C. acnes, which was then analyzed to determine the minimum inhibitory concentration (MIC)
of a range of clinically relevant antibiotics using commercially procured E-strips.
Inoculum for the susceptibility studies were prepared by suspending a culture grown
for 48 hours in reduced Brucella broth to achieve a final concentration equivalent
to a 0.5 McFarland standard (1.5 X 10 CFU/ ml). E-test strips of each of the test
antibiotics were then placed onto the Brucella blood agar supplemented with Vit.K1,
hemin and laked sheep blood, and the plates were incubated at 37°C under anaerobic
conditions[13]. MICs for each antibiotic were determined after 48 hours of incubation.
Antibiotic susceptibility and resistance breakpoints were defined as follows
|
Antibiotic agent
|
Clinical breakpoint
|
|
Clindamycin
|
≥ 8 μg/mL13
|
|
Tetracycline
|
≥ 16 μg/mL13
|
|
Ceftriaxone
|
≥ 64 μg/mL13
|
|
Ertapenem
|
≥ 16 μg/mL13
|
|
Meropenem
|
≥16 μg/mL13
|
|
Erythromycin
|
≥ 2 μg/mLl11
|
|
Doxycycline
|
≥ 4 μg/mL11
|
|
Minocycline
|
≥ 16 μg/mL11
|
Patients were evaluated during their follow-up visits after 1 month, and the clinical
response to the prescribed antibiotics was recorded.
Statistics
To understand the association between in vitro resistance and treatment outcome, we
compared the cure rates between patients yielding sensitive and resistant isolates
of C. acnes. Chi-square test was used to compare cure rates between patients with susceptible
and nonsusceptible isolates, and the p value was calculated with the help of the Epi Info online software.
Results
Skin swabs from a total of 52 consecutively recruited patients attending skin OPD
with acne vulgaris were collected for C. acnes culture. Growth of C. acnes was observed in 42 samples (80.76%). These 42 patients were included for further
analysis and follow-up.
The mean (± standard deviation [SD]) age of the patients was 21 (± 4) years and the
mean (± SD) duration of disease was 3 (± 2) years. Thirty-one patients presented with
acne only on their face, two presented with acne only on their trunk, and nine patients
had lesions both on face and trunk. Acne was mild in intensity in 8 patients, moderate
in 25 patients, and severe in 9 patients. The lesions were inflammatory in 9 patients,
noninflammatory in 7 patients, and of mixed type in 26 patients. Seven patients reported
the use of oil-based cosmetics on their face and three patients gave a history of
having facial massages. The patients received treatment according to their clinical
presentation from a single qualified dermatologist and were followed-up after 1 month
of treatment. They were categorized into nonresponder (n = 16) and responder (n = 26) groups, depending on improvement or deterioration with the prescribed treatment.
The two subgroups of patients were found to be comparable with respect to their age
and gender distribution, clinical characteristics, and nature of treatment administered
([Table 1]).
Table 1
Comparison of age and gender distribution, clinical characteristics, and nature of
treatment administered among nonresponders and responders
|
Parameters
|
Nonresponders (n = 16)
|
Responders (n = 26)
|
p-Value
|
|
Age (years) (mean ± SD)
|
22.12 ± 5.95
|
21 ± 2.92
|
0.00169
|
|
Gender (M/F)
|
13/03
|
20/06
|
0.7400
|
|
Duration of illness (months) (mean ± SD)
|
42 ± (31.58)
|
28.58 ± (22.68)
|
0.1386
|
|
Facial involvement (%)
|
16 (100)
|
24 (92.31)
|
0.3883
|
|
Extrafacial involvement (%)
|
06 (37.50)
|
05 (19.23)
|
0.3883
|
|
Grade of illness (mild:moderate:severe)
|
04:08:04
|
04:17:05
|
0.598
|
|
Inflammation present (%)
|
15 (93.75)
|
20 (76.92)
|
0.1554
|
|
Only topical antibiotic administered (%)
|
03 (18.75)
|
10 (38.46)
|
0.2701
|
|
Systemic antibiotic administered (%)
|
13 (81.25)
|
16 (61.54)
|
0.3393
|
|
History of previous treatment (%)
|
08 (50)
|
10 (38.46)
|
0.4631
|
Colonies of C. acnes were small (less than 0.5 mm in size), grey, and weakly β hemolytic ([Fig. 1a]). On microscopy after Gram staining, nonsporing Gram-positive bacilli were seen
(inset image of [Fig. 1a]). These colonies were confirmed to be C. acnes by performing standard biochemical tests.[11] Antibiotic susceptibility testing with clinically relevant antibiotics was performed;
minocycline (100% susceptible), doxycycline (97.6% susceptible), ceftriaxone (95.2%
susceptibility), and tetracycline (92.9% susceptibility) were found to be the most
effective antibiotics. Nonsusceptibility to clindamycin and erythromycin was observed
in 11.9% (n = 5) and 31% (n = 13) isolates, respectively, with 9.5% (n = 4) isolates being nonsusceptible to both. For all the antibiotics tested, a significantly
higher proportion of isolates was susceptible; Ertapenem was found to be the least
effective antibiotic (52.4% susceptibility) ([Table 2]).
Table 2
Antibiotic susceptibility pattern of C. acnes isolates (n = 42)
|
Name of Antibiotics
|
Susceptible
|
Nonsusceptible
|
p-Value
|
|
Clindamycin
|
37 (88.10%)
|
5 (11.90%)
|
< 0.001
|
|
Tetracycline
|
39 (92.86%)
|
3 (7.14%)
|
< 0.001
|
|
Ceftriaxone
|
40 (95.24%)
|
2 (4.76%)
|
< 0.001
|
|
Ertapenem
|
22 (52.38%)
|
20 (47.62%)
|
< 0.001
|
|
Meropenem
|
25 (59.52%)
|
17 (40.48%)
|
< 0.001
|
|
Erythromycin
|
29 (69.05%)
|
13 (30.95%)
|
< 0.001
|
|
Doxycycline
|
41 (97.62%)
|
1 (2.38%)
|
< 0.001
|
|
Minocycline
|
42 (100%)
|
0
|
< 0.001
|
Fig. 1 (a) C. acnes colonies on blood agar plate, inset image showing gram staining. (b) Clinical representation of acne on cheek and forehead area.
On examining the antibiotics prescribed for these patients, we found a total of five
regimens being administered in 35 patients. Antibiotics were not administered in 7
patients, while 19 patients received only topical antibiotics. Sixteen patients were
given erythromycin (n = 14) and doxycycline (n = 2) in addition to topical clindamycin. For none of the five regimens, significant
difference was observed between the proportion of nonresponders and responders ([Table 3]), implying that none of the regimens could claim superiority over others.
Table 3
Antibiotics regimens prescribed to patients (n = 35)
|
Antibiotics regimens
|
Total
|
Nonresponders
|
Responders
|
p-Value
|
|
Erythromycin + Clindamycin
|
14
|
5 (35.71%)
|
9 (65.29%)
|
0.6726
|
|
Doxycycline + Clindamycin
|
2
|
1 (50%)
|
1 (50%)
|
0.7662
|
|
Erythromycin
|
2
|
2
|
0
|
0.07447
|
|
Doxycycline
|
6
|
4 (66.67%)
|
2 (33.33%)
|
0.9569
|
|
Clindamycin
|
11
|
2 (18%)
|
9 (72%)
|
0.07447
|
We then tried to analyze if there was a difference in the proportion of susceptible
and nonsusceptible isolates between mild, moderate, and severe categories of patients.
For none of the antibiotics, we found significant difference in the proportion of
susceptible and nonsusceptible isolates among the three categories of disease severity.
Thus, antibiotic susceptibility of C. acnes was not found to be associated with its virulence and pathogenic potential ([Table 4]).
Table 4
Antibiotics susceptibility pattern of C. acnes isolates in different categories of
disease severity
|
Name of Antibiotics
|
Mild (08)
|
Moderate (25)
|
Severe (09)
|
p-Value
|
|
Susceptible
|
Nonsusceptible
|
Susceptible
|
Nonsusceptible
|
Susceptible
|
Nonsusceptible
|
|
Clindamycin
|
7 (87.50%)
|
1 (12.50%)
|
22 (88.00%)
|
3 (12.00%)
|
8 (88.89%)
|
1 (11.11%)
|
0.9958
|
|
Tetracycline
|
7 (87.50%)
|
1 (12.50%)
|
24 (96.00%)
|
24 (96.00%)
|
8 (88.89%)
|
1 (11.11%)
|
0.6275
|
|
Ceftriaxone
|
7 (87.50%)
|
1 (12.50%)
|
24 (96.00%)
|
24 (96.00%)
|
8 (88.89%)
|
1 (11.11%)
|
0.6275
|
|
Ertapenem
|
3 (37.50%)
|
5 (62.50%)
|
11 (44.00%)
|
14 (56.00%)
|
5 (55.56%)
|
4 (44.44%)
|
0.7425
|
|
Meropenem
|
5 (62.50%)
|
3 (37.50%)
|
14 (56.00%)
|
11 (44.00%)
|
6 (66.67%)
|
3 (33.33%)
|
0.8399
|
|
Erythromycin
|
7 (87.50%)
|
1 (12.50%)
|
17 (68.00%)
|
8 (32.00%)
|
5 (55.56%)
|
4 (44.44%)
|
0.3581
|
|
Doxycycline
|
7 (87.50%)
|
1 (12.50%)
|
25 (100.00%)
|
0 (00.00%)
|
9 (100.00%)
|
0 (00.00%)
|
0.1134
|
|
Minocycline
|
8 (100%)
|
0(0.00%)
|
25 (100.00%)
|
0 (00.00%)
|
9 (100.00%)
|
0 (00.00%)
|
–
|
We next tried to find an association between the clinical response of the patients
and antibiotic susceptibility of the recovered isolates. Among the 35 patients who
received antibiotics, 27 patients received clindamycin, 16 patients received erythromycin
and 8 patients received doxycycline, either alone or in combination. It was observed
that of the 27 patients who received clindamycin, 16 of the 19 responders and 6 of
the 8 nonresponders yielded growth of clindamycin-susceptible isolates (p = 0.57).
Discussion
In the present study on the antibiogram of the emerging pathogen, C. acnes, it was observed that minocycline, doxycycline, ceftriaxone, and tetracycline were
the most effective antibiotics and carbapenems were the least effective. Nonsusceptibility
to clindamycin was observed in 11.9% of the isolates, and no association could be
drawn between clindamycin susceptibility and treatment response.
Similar to our findings, other authors from different parts of the world have also
reported consistently decreasing susceptibility to clindamycin and erythromycin among
C. acnes isolates. clindamycin susceptibility has been reported to vary from 7.5 to 91% and
erythromycin susceptibility from 10.4 to 98% in different studies.[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26] It is worth noting that development of antibiotic resistance in C. acnes was not encountered until the early 1980s despite years of systemic use of tetracycline
and erythromycin. But shortly after the introduction of topical formulation of these
drugs, the first strains of resistant C. acnes emerged in the USA, and by the late 1980s, very high MIC levels for these drugs became
commonplace among C. acnes isolates from UK and USA.[8] Several other studies have also demonstrated similar relationship between the increasing
resistance to these drugs and clinical prescription patterns. In a study from Singapore,
Tan et al[10] have shown resistance to erythromycin to be more significantly associated with increased
duration of antibiotic usage. Patients on no antibiotics or on short-term antibiotics
(6–18 weeks) were found to yield much lesser isolation of erythromycin-resistant strains,
and most of the erythromycin-resistant strains were also cross-resistant to clindamycin.
In a larger study covering six European centers, widespread use of topical erythromycin
and clindamycin was found to cause significant dissemination of cross-resistant strains
of Cutibacteria both in patients and in their untreated contacts. This study also
revealed remarkable variations in the proportion of resistant Cutibacteria between
the European countries, which was reflective of the usage pattern in these countries.
Interestingly, a significant proportion of dermatologists specializing in acne treatment
were found to be colonized with resistant Cutibacteria in this study, while none of
the nondermatologist physicians were similarly colonized.[27] In a recent study from Korea too, increasing resistance to clindamycin and erythromycin
has been reported among C. acnes isolates from acne patients, compared to previous studies from the same region.[28] In sync with these studies, our finding of a relatively high proportion of clindamycin
nonsusceptibility could be explained by virtue of its being the most commonly prescribed
antibiotic among our patients.
In the event of the rising trend of resistance to clindamycin and erythromycin, alternative
therapeutic choices against this emerging pathogen need to be explored. In contrast
to the obvious advantages of topical formulation for skin lesions, systemic administration
of antibiotics is imperative in C. acnes isolates recovered from deep-seated infections such as endocarditis, septic arthritis,
endophthalmitis, etc. and diverse device- and implant-related infections. Among the
systemic choices, our finding regarding minocycline, doxycycline, ceftrioxone and
tetracyclines being the most effective antibiotics is also corroborated by other authors.
Universal susceptibility to ceftrioxone has been reported in independent studies from
USA and Switzerland.[3]
[26]
[29] In a study from Singapore, susceptibility of C. acnes to different tetracyclines was compared and the average MIC to tetracyclines was
found to be higher than that of doxycycline which, in turn, was higher than that for
minocycline.[10] In another study from Korea, Song et al[9] observed a consumption-related decline in susceptibility toward doxycycline with
patients with a history of treatment with topical and systemic antibiotics showing
higher MIC to doxycycline. Like other antibiotics, geographical variation has also
been noted in doxycycline susceptibility pattern of C. acnes, with resistance rates ranging from 7 to 63% being reported in literature. We observed
considerably higher proportion of resistance among the recovered C. acnes isolates toward carbapenems. Contrary to our findings, Shame et al[3] and Crane et al[26] have reported all the isolates to be within the susceptible range of MIC for ertapenem
and meropenem in their studies conducted in 2006 and 2013, respectively. This difference
with our findings could be explained in view of geographical and temporal variation
and also by relatively unrestricted use of carbapenems in Indian clinical practice.
We observed the lack of association between clindamycin susceptibility and treatment
outcome in our study. This difference in in vitro result and in vivo response could
partially be explained by the known propensity of this bacterium to produce biofilms
at the site of clinical lesions.[29] This ability to produce biofilm could be responsible for offering protection to
the bacteria from the inhibitory effect of the administered antibiotics. Pharmacokinetic
factors like subinhibitory concentration of the antibiotics at the site of infection
could also be responsible for inadequate treatment outcome. Moreover, the requirement
of alkaline pH for antibacterial action of clindamycin and erythromycin could also
have remained unmet at the site of clinical lesions.[30]
The present study has two major limitations. First, the sample size in this study
was relatively small and as a result, adequate representation of the different treatment
regimens could not be achieved. Second, the molecular mechanism of resistance to the
different antibiotics could not be included in the scope of the study. Building on
the findings of this pilot study, we are planning to conduct a broader and more elaborate
study on clinico-microbiological and molecular characteristics of antibiotic susceptibility
in C. acnes isolates recovered at our institute.
Conclusion:
Our pilot study brings out the antibiogram of clinical isolates of C. acnes in the Indian context and records a lack of association between the antibiotic susceptibility
of these isolates with the clinical response to treatment. More elaborate studies
need to be undertaken to validate our findings and identify the determinants of treatment
outcome targeting this emerging pathogen.
