Essential cookies by default. The browser ID, load-time metrics and error replays only run if you choose "Accept all".
We use essential cookies by default: sign-in, your choice in this banner (supstack-consent) and your language (NEXT_LOCALE). Vercel Web Analytics (cookieless page counts) and Sentry error reports (without replay) run either way. Only if you choose "Accept all" do we also create a browser ID in local storage (it ties your visits together in our site records), record error-replay diagnostics (Sentry: masked DOM snapshots, only when an error fires) and turn on load-time metrics (Vercel Speed Insights) so we can fix slow pages faster. Learn more about cookies
Clindamycin (topical) doesn't have enough studies with structured effect-size data yet to pool a verdict. Browse the studies below for individual findings.
Most evidence is from high-quality meta-analyses and randomised trials published 2010–2026 with a typical study size of 22,645 participants.
Based on 15 studies · 5 meta-analyses · 2 RCTs · 122,425 total participants
Study base
15 studies indexed
By outcome
Every outcome below: Too few graded studies.
Skin healthSubstantial reduction in inflammatory lesions when combined with a retinoid or benzoyl peroxide, and the best topical option for moderate-to-severe acne in that pairing. Slower than benzoyl peroxide for the first month, and of little measured benefit on lesion counts alone. · 4-12 weeks8 studies
8 studies
Muscle strength & power1 study
1 study
Women's health1 study
1 study
Fertility & reproductive1 study
1 study
Safety profile1 study
1 study
Active research area
12 studies in the last 5 years · Latest meta-analysis: 2026
201020182026
1Meta-Analysisn=22,645 · very large study2026
In plain English: Compared with placebo, only dapsone 5% and erythromycin 2% significantly reduced inflammatory lesion counts. … However, no treatment showed a significant improvement in total lesion counts compared to placebo.
Not stated in the abstract · Clin Drug Investig (2026)
32 studies comprising 34 randomised controlled trials, 22,645 patients, searched to October 2025.
Clindamycin was not among the agents that significantly reduced inflammatory lesion counts against placebo. Only dapsone 5% and erythromycin 2% were.
Nor was it among those reducing non-inflammatory lesions: azithromycin 2%, erythromycin 2%, metronidazole 2% and minocycline 4% were.
2Meta-AnalysisCited 7×n=33,472 · very large study2025
In plain English: For non-inflammatory acne, the addition of clindamycin in topical regimens is unnecessary and should be avoided.
Not stated in the abstract · J Eur Acad Dermatol Venereol (2025)
35 randomised clinical trials, 33,472 participants, nine topical agents, minimum 12 weeks of treatment.
Adapalene-BPO, clindamycin-BPO and clindamycin-tretinoin produced the greatest reductions in non-inflammatory, inflammatory and total lesion counts against placebo.
But clindamycin-tretinoin did not significantly outperform tretinoin alone (RoM 1.13, 95% CI 0.94 to 1.36) and clindamycin-BPO did not significantly outperform BPO alone (RoM 1.15, 95% CI 0.98 to 1.36) for non-inflammatory lesions.
3Systematic ReviewCited 6×n=40,910 · very large study2024
In plain English: The six reviews involved 40,910 people with acne from 275 trials and 1316 people with acne scars from 37 trials.
Not stated in the abstract · Cochrane Database Syst Rev (2024)
A Cochrane OVERVIEW of systematic reviews rather than of trials, searched to 2 December 2021, excluding reviews at high risk of bias using ROBIS.
Of 733 records retrieved, only six reviews met the inclusion criteria at low risk of bias. That is the headline about this literature.
Those six covered 40,910 people with acne from 275 trials, plus 1,316 with acne scars from 37 trials, aged 10 to 59.
4Systematic ReviewCited 316×2024
In plain English: Strong recommendations are made for benzoyl peroxide, topical retinoids, topical antibiotics, and oral doxycycline.
Reynolds RV, et al. · J Am Acad Dermatol (2024)
American Academy of Dermatology guideline built on a systematic review with GRADE assessment: 18 evidence-based recommendations and 5 good practice statements.
Topical antibiotics carry a STRONG recommendation, alongside benzoyl peroxide, topical retinoids and oral doxycycline.
Oral isotretinoin is strongly recommended for acne that is severe, causes psychosocial burden or scarring, or fails standard therapy.
5Systematic ReviewCited 31×n=18,089 · very large study2021
In plain English: The combinations of BPO with adapalene (54% vs. 35%) or with clindamycin (49% vs. 35%) were ranked more effective than BPO alone.
Not stated in the abstract · Br J Dermatol (2021)
81 papers reporting 40 trials with 18,089 participants, searched to June 2020, with confidence assessed using CINeMA.
Self-reported improvement and trial withdrawal were CO-PRIMARY outcomes, which puts a patient-reported measure on equal footing with tolerability.
Benzoyl peroxide alone was effective against vehicle (35% vs 26%). Adding clindamycin took it to 49%.
6Meta-AnalysisCited 70×2022
In plain English: For moderate-to-severe acne, the most effective options for each treatment type were as follows: topical pharmacological - combined retinoid with lincosamide (clindamycin) (44·43%, 95% CrI 29·20-60·02%).
Not stated in the abstract · Br J Dermatol (2022)
179 randomised controlled trials and approximately 35,000 OBSERVATIONS (not unique participants) across 49 treatment classes, stratified by acne severity.
For MODERATE-TO-SEVERE acne, a retinoid combined with clindamycin was the most effective TOPICAL option: 44.43% (95% CrI 29.20-60.02%).
For MILD-TO-MODERATE acne the best topical was a retinoid with benzoyl peroxide (26.16%, 95% CrI 16.75-35.36%), with no clindamycin involved.
7Meta-AnalysisCited 47×n=7,309 · very large study2010
In plain English: At 10- to 12-week end points … weighted mean inflammatory lesion reduction: BPO = 43.7%, CL = 45.9%, BPO + SA = 51.8%, BPO/CL = 55.6%, placebo = 26.8%.
Seidler EM, Kimball AB · J Am Acad Dermatol (2010)
23 studies, 7,309 patients, conducted under Cochrane guidance and reported to PRISMA.
The abstract's own headline is a different drug: 5% benzoyl peroxide with salicylic acid had the greatest early reductions of any arm (55.2% inflammatory, 42.7% non-inflammatory at 2-4 weeks).
The time point changes the numbers. At 2-4 weeks clindamycin reads lowest of the actives (inflammatory 21.5% against benzoyl peroxide's 33.4%), and by 10-12 weeks the two read 45.9% against 43.7%.
8RCT2026
In plain English: Sequential treatment involving combined antimicrobial therapy followed by immunomodulation significantly improves clinical pregnancy rates in women with UI associated with BV and ASA.
Maindo Alongo MA et al. · BMC immunology (2026)
Among 123 participants, immunomodulators significantly increased clinical pregnancy rates compared with placebo (38.55% vs. 12.5%; aHR = 3.43 [1.34-8.81]; p = 0.010).
Adjusted analyses showed significantly higher pregnancy rates for both zinc acetate (40.48%; aHR 3.46 [1.27-9.40]; p = 0.012) and prednisolone (36.59%; aHR 2.75 [1.01-7.68]; p = 0.027) compared with placebo, with no significant difference between the two active agents (p = 0.623).
Combined antimicrobial therapy was superior to metronidazole alone in achieving therapeutic BV cure at visit 2 (58.06% vs. 37.7%; RR 1.52 [1.04-2.22]; p = 0.023) and, at the margins of the factorial design, was associated with a higher cumulative conception rate (aHR 2.20 [1.09-4.43]; p = 0.027).
In plain English: Furthermore, 70.0% and 72.0% of the patients agreed that the dermocosmetic facial cleanser and moisturizing cream, respectively, improved the tolerability of their acne treatment.ConclusionsThese findings suggest that the dermocosmetic regimen involving the use of a facial cleanser and moisturizing cream, when used in combination with topical prescription drugs is beneficial, well-tolerated, and acceptable among patients for the treatment of acne in real-world routine clinical practice in Japan.
Kurokawa I, Kobayashi M, Nomura Y, Abe M, Li X. · The Journal of international medical research (2026)
Tolerability for the dermocosmetics use was rated as "well-tolerated" or "tolerated" by 75.5% of physicians and 73.5% of patients.
Satisfaction with dermocosmetics use was high, with 87.5% of physicians and 92.0% of patients rating their satisfaction level as "very satisfied" or "satisfied." Most patients (86.0%) agreed or strongly agreed that they preferred using the study dermocosmetics over their previous cosmetics.
In plain English: Therefore, the aim of this review is to outline specific requirements for pre- and postoperative wound management in HS and to derive practical recommendations.
Ocker-Serger L et al. · Dermatologie (Heidelberg, Germany) (2026)
Hidradenitis suppurativa (HS) is a chronic, recurrent inflammatory systemic disease that often leads to extensive, exudative, and painful wound areas.
In addition to anti-inflammatory therapy, structured wound care represents a central component of multimodal treatment concepts.
Therefore, the aim of this review is to outline specific requirements for pre- and postoperative wound management in HS and to derive practical recommendations.
In plain English: Clindamycin/BPO seemed to act more quickly than adapalene. … Adapalene/BPO and clindamycin/BPO showed comparable TOA.
Not stated in the abstract · Br J Dermatol (2014)
Systematic review of head-to-head trials in mild-to-moderate papulopustular acne. Primary outcome was time until a 25% reduction in mean inflammatory lesions.
Clindamycin with benzoyl peroxide seemed to act more quickly than adapalene, and comparably to adapalene with benzoyl peroxide.
Against benzoyl peroxide alone the results were inconsistent, with only a slight indication of a shorter time to onset for the clindamycin combination.
12Review2026
In plain English: Poor adherence is a major barrier to treating acne, leading to negative impacts on patient outcomes.
Feldman SR et al. · Journal of drugs in dermatology : JDD (2026)
Poor adherence is a major barrier to treating acne, leading to negative impacts on patient outcomes.
To improve adherence, physicians should address barriers through careful treatment selection and patient education.  .
In plain English: Ozenoxacin demonstrated comparable efficacy and tolerability to clindamycin when used in combination with adapalene, in truncal acne.
Das A et al. · Clinical and experimental dermatology (2026)
Both groups demonstrated statistically significant improvement from baseline in all efficacy parameters at 12 weeks (p < 0.05).
Intergroup comparisons showed no statistically significant differences across clinical or quality-of-life outcomes.
Ozenoxacin demonstrated comparable efficacy and tolerability to clindamycin when used in combination with adapalene, in truncal acne.
In plain English: IDP-126 is therefore recommended to be applied as a thin layer to the affected area once daily.
Gupta AK, Mann A, Vincent K, Abramovits W. · Skinmed (2024)
CabtreoTM (1.2% clindamycin phosphate, 0.15% adapalene, and 3.1% benzoyl peroxide) or IDP-126 topical gel was approved by the US Food and Drug Administration (FDA) in October 2023 for the treatment of moderate to severe acne vulgaris in patients aged ≥12 years.
In Trial 1 (N = 183), treatment success was achieved in 49.6% (61/122) of subjects in the IDP-126 group versus 24.9% (15/61) of subjects in the vehicle group (P < 0.01).
In Trial 2 (N = 180), treatment success was achieved in 50.5% (61/120) of subjects in the IDP-126 group versus 20.5% (12/60) of subjects in the vehicle group (P < 0.01).
15Meta-AnalysisCited 6×2016
In plain English: An increase in bacterial resistance to topical erythromycin and clindamycin can be confirmed, thus the use of these antibiotics is recommended in selective cases for short periods, and in combination with benzoyl peroxide for the best clinical outcome.
Not stated in the abstract · Cir Cir (2016)
13 articles reporting antibiogram results from patients with acne over a ten-year window.
Mean odds ratio 1.24, described by the authors as a slight tendency toward resistance in Propionibacterium acnes (now Cutibacterium acnes). The abstract gives no confidence interval for it.
The effect is modest and the paper frames the topic as a controversy rather than a settled question. It is cited here for the direction of travel, not a precise risk.