OBJECTIVES:

To explore the association between Demodex infestation and rosacea and the pathogenesis of demodicosis rosacea by means of a meta-analysis.

DATA SOURCES:

Electronic searches of the ISI Web of Knowledge (Science Citation Index, ISTP [Index to Scientific & Technical Proceedings], Journal Citation Reports, BP [BIOSIS Preview], INSPEC [Ination Service in Physics, Electronics Technology, and Computer and Control], and DII [Derwent Innovation Index]), MEDLINE, and CNKI (China National Knowledge Infrastructure) databases (January 1, 1950, to December 31, 2009). We also performed hand searches of reference lists and conference proceedings.

STUDY SELECTION:

Predefined selection criteria were applied to all published case-control studies that analyzed the association between Demodex infestation and rosacea.

DATA EXTRACTION:

Two of us independently extracted data from the included studies. For disputed articles, a third party mediated whether to include the study.

DATA SYNTHESIS:

Forty-eight English- and Chinese-language articles, which covered 10 different countries and 28 527 participants, were eligible. The pooled odds ratio in random-effects models is 7.57 (95% confidence interval, 5.39-10.62). Stability is good according to sensitivity analysis. The fail-safe number is 18 456 in the quantitative analysis of publication bias.

CONCLUSIONS:

A significant association exists between Demodex infestation and the development of rosacea. Demodex infestation is a vital risk factor for rosacea according to the time-to-event relationship, and the degree of infestation played a more important role than did the mite infestation rate in the development of rosacea.