In Vivo Antibacterial Effects of Tretinoin-Clindamycin and Clindamycin Aloneon Propionibacterium acnes With Varying Clindamycin Minimum InhibitoryConcentration Levels

December 2012 | Volume 11 | Issue 12 | Original Article | 1434 | Copyright © December 2012

James J. Leyden MD

Department of Dermatology, University of Pennsylvania, Philadelphia, PA


Objective: To quantify the antimicrobial effect of clindamycin phosphate 1.2% and tretinoin 0.025% gel and 1% clindamycin phosphate gel in patients with Propionibacterium acnes of varying sensitivity to clindamycin.
Design: Study 1 was an initial range-finding study that was neither blinded nor randomized. Study 2 was an open-label, randomized, splitface, single-center study. Both studies were conducted in Pennsylvania.
Patients: Study 1 (n=20) and study 2 (n=22) involved healthy patients aged 18 years or older with initial P acnes levels ≥104/cm2 and minimum inhibitory concentrations (MICs) ≥8 µg/mL for clindamycin.
Interventions: Study 1, clindamycin gel applied twice daily for 6 weeks. Study 2; once-daily application with the combination gel to one cheek and clindamycin gel to the other side for 6 weeks.
Main Outcome Measure: The comparative effectiveness of each product vs P acnes of varying sensitivity to clindamycin at 3 and 6 weeks posttreatment.
Results: For study 1, at 3 and 6 weeks, clindamycin-treated patients with MICs of ≤256 µg/mL showed greater reductions than those with MICs ≥512 µg/mL (P=.0001). Study 2 showed a significant reduction in P acnes for both products, with no differences found. Clindamycin alone was more effective in vivo in patients with MIC levels of ≤256 µg/mL than patients with higher MIC levels. The combination product produced a greater reduction than clindamycin alone after 6 weeks in patients with high MICs >512 µg/mL (P=.0047).
Conclusion: These studies suggest that 1% clindamycin alone produces a varying in vivo antimicrobial effect, with a breakpoint at ≤256 µg/mL. Use of clindamycin phosphate 1.2% and tretinoin 0.025% gel resulted in a significantly greater in vivo antimicrobial effect than clindamycin alone in patients carrying P acnes with MICs of ≥512 µg/mL (P=.0047).

J Drugs Dermatol. 2012;11(12):1434-1438.


Propionibacterium acnes is the most common anaerobic, grampositive organism recovered from normal skin, particularly in areas rich in sebaceous glands.1 It preferentially resides within sebaceous follicles as a symbiotic microbe. In the abnormal environment of corneocytes and sebum found in the microcomedones, P acnes proliferates and generates a variety of proinflammatory molecules responsible for the inflammation seen in acne vulgaris.2-6
Over the past 20 years, the prevalence of P acnes to various antibiotics has increased globally.7-8 The observed increase in resistance exhibits the following minimum inhibitory concentrations (MICs) rank order of insensitivity: erythromycin > clindamycin > tetracycline > doxycycline > minocycline.7 In unpublished results in 75 strains of P acnes with high erythromycin MIC levels (1,024 µg/mL or greater), we found a range of clindamycin MIC sensitivity levels that varied as follows: 8 strains, >1,024 µg/mL; 35 strains, 512 to 1,024 µg/mL; and 32 strains, <512 µg/mL. These findings are similar to other findings in patients around the world.7
Currently, follicular antibiotic concentrations cannot be measured and are unknown. As a result, microbiological breakpoints cannot be easily determined. Furthermore, MIC levels are determined in in vitro aqueous systems, which may not correlate with intrafollicular concentrations, since the latter is a lipid-rich environment. In vivo human studies in patients with varying MIC levels may be more helpful in determining MIC breakpoint sensitivity levels.9
The objectives of these in vivo studies were 2-fold: (1) to quantify the effectiveness of clindamycin in suppressing P acnes that exhibit varying sensitivities to clindamycin, and (2) to determine whether the presence of 0.025% tretinoin alters the in vivo antimicrobial effects of clindamycin.


Two studies in patients with high numbers of P acnes, high levels of erythromycin resistance, and varying MICs for clindamycin were conducted.

Study 1

The initial range-finding study population consisted of 20 healthy male and female patients aged 18 years or older whose facial (cheek) samples exhibited high P acnes MIC levels of at least ≥4 log10 colony-forming units (CFUs)/cm2 (≥10,000 colonies per cm2) and with MICs of ≥8 µg/mL for clindamycin. Each patient was treated with 1% clindamycin gel twice a day for 6 weeks.