Ivermectin Plus Metronidazole Most Effective for Reducing Demodex

A review evaluates the most effective current method for blepharitis-causing mite control.

Systemic and topical ivermectin plus metronidazole were the most effective in reducing mite count in patients with Demodex-induced blepharitis, according to research results published in Contact Lens and Anterior Eye. 

 Because of the high prevalence of disease, quality of life repercussions, and lack of integrative studies that analyze therapeutic choices, researchers conducted a systematic review and meta-analysis evaluating the effectiveness of treatments for Demodex blepharitis. 

Both randomized and nonrandomized controlled studies that compared the efficacy and adverse reactions of 2 medications, or medication vs placebo, in patients older than 16 years. The primary outcome included the mites per eyelash count at treatment conclusion; secondary outcomes included total eradication rates, tear break-up time, Ocular Surface Disease Index (OSDI), Schirmer test, osmolarity, cylindrical dandruff score, adverse reaction rate, and a stratified subanalysis comparing either tea tree oil (TTO)-derived treatments or pharmacologic vs nonpharmacologic interventions. 

Overall, 3426 references were found, 18 of which met the inclusion criteria. These randomized clinical trials included 1574 eyes from 1195 participants who were positive for Demodex infestation. Participants differed from each other in terms of follow-up length — ranging from 28 to 90 days — and the type of evaluated intervention. A total of 4 studies compared therapeutic interventions vs placebo, 12 compared 2 interventions, and 2 compared an intervention with the absence of treatment. Only 1 study specifically evaluated adverse reactions. 

Six studies were industry-funded and 4 included fewer than 30 participants. In terms of bias risk, low risk was identified in 30.6% of studies; for performance, detection, attrition, and reporting biases, low risk was established in 27.8%, 55.6%, 88.9%, and 5.6% of studies, respectively. 

The analysis included 8 studies of Demodex count. Overall change in Demodex per eyelash count was -2.07 (95% CI, -3.99 to -0.15), with highest effects noted for systemic and topical ivermectin-metronidazole (-5.10 and -7.50 Demodex per eyelash, respectively). Other interventions that had a significant reduction on Demodex per eyelash count included artificial tears plus topical steroid drops plus eye shampoos with 5% TTO (-1.71 Demodex per eyelash) and eyelid scrubs with 50% TTO weekly plus 10% TTO daily (-1.00 Demodex per eyelash). 

A stratified meta-analysis did not indicate significant subgroup differences between TTO-derived and non-TTO-derived treatments, but an exclusive analysis of this variable in TTO-derived treatments indicated a statistically significant effect of -0.90 Demodex pereyelash (95% CI, -1.39 to -0.40) vs a nonsignificant statistical effect in non-TTO-derived treatments (-2.76 Demodex per eyelash; 95% CI, -1.39 to -0.40). 

When pharmacologic and nonpharmacologic interventions were compared, researchers noted a significant between-subgroup difference in favor of pharmacologic interventions (-2.88 Demodex per eyelash; 95% CI, -5.74 to -0.02). 

The analysis also included 13 studies of total eradication of mites. An overall risk ratio (RR) of 1.84 (95% CI, 1.27-2.66) was noted, “favoring total eradication in the intervention groups.” Highest significant RRs were seen for 50% TTO lid scrubs, manuka honey, and topical ivermectin plus metronidazole. 

Results of a stratified meta-analysis did not demonstrate significant subgroup differences between TTO- and non-TTO-derived treatments, although a separate evaluation indicated a non-statistically significant RR for TTO- vs a statistically significant RR in non-TTO-derived treatments (1.35 vs 1.68). 

Another 12 studies were included to analyze  OSDI, tear break-up time, Schirmer test, osmolarity, and clinical improvement. None showed statistically significant results, and clinical improvement was associated with an overall RR of 0.96 (95% CI, 0.90-1.04). 

Additionally, 3 studies looked at  cylindrical dandruff. Investigators found an overall decrease with statistical significance (-0.76; 95% CI, -1.09 to -0.44). No significant subgroup difference was seen between TTO- and non-TTO-derived treatments. 

The team also included 9 studies that evaluated adverse drug reactions. Intervention groups had a tendency to present with adverse reactions, with an overall effect according to risk difference of 0.24 (95% CI, 0.08-0.41). The biggest individual effects favoring an adverse reaction were in 2 studies, one evaluating an eyelid scrub with 50% TTO weekly plus 10% TTO daily (0.05; 95% CI, 0.02-0.08) and the other evaluating TTO and terpinene 4-ol (1.00; 95% CI, 0.93-1.07). All reported data included only mild adverse reactions. 

Study limitations include a lack of standardization and comparability among existing studies, the availability of few studies with adequate methodologies to evaluate adverse reactions, the need to obtain information from studies with incomplete or unavailable data, and the “total absence” of information in current medical literature about rates of reinfestation. 


Martínez-Pulgarín D, Ávila MY, Rodríguez-Morales AJ. Interventions for Demodex blepharitis and their effectiveness: A systematic review and meta-analysis. Cont Lens Anterior Eye. Published online May 7, 2021. doi:10.1016/j.clae.2021.101453