Background Cryotherapy is used empirically as a symptomatic treatment in rheumatology, with well-known analgesic, vasoconstrictive, anti-inflammatory properties. It can be applied locally or to the whole body. This widespread use contrasts with a lack of standardization in techniques and a poor level of evidence.

Objectives The aim of this work is to quantify cryotherapy’s therapeutic effects in rheumatologic inflammatory diseases using available evidence.

Methods We performed a systematic review (pubmed, LILACS, Cochrane library). Keywords “cryotherapy”,“cold”,“inflammation”,“arthritis” were used alone and in combination. We also considered abstracts from rheumatology congresses (ACR, EULAR) and unpublished data. Selection criteria were inflammatory rheumatologic disease patients treated with local or whole-body cryotherapy, with endpoints evaluating pain, activity and inflammation parameters. Articles about post-operative joint cryotherapy and infectious diseases were excluded. Results were expressed as means +/- SD (continuous variables). A fixed effect model was used. Paired or unpaired t-tests (α=5%) were used to compare pooled means +/- SD (before/after cryotherapy and versus control when possible).

Results We identified 116 potentially relevant abstracts and excluded 96 articles (postoperative, non-inflammatory diseases, inadequate endpoints or outcomes, lack of accuracy (technique, results), duplicates). Among the 20 potentially appropriate articles for meta-analysis, 14 more were excluded (inappropriate design or data presentation, study populations). We finally selected 6 articles for quantitative analysis. There was no significant heterogeneity. We could only perform quantitative analysis for 2 endpoints (pain VAS and DAS28) in RA patients after chronic application (7 to 15 days).

-Local cryotherapy (cold packs, cold air, liquid nitrogen for 3 to 30 min (ice) – 10 to 20 applications on 1-5 joints) significantly decreased pain VAS (mm) in 68 RA patients (56,57 +/- 27,44 versus 32,5 +/- 19,99 mm ; p<10exp(-6)). However, there was no significant difference with 17 RA patients treated with “drug therapy”.

-Local cryotherapy also significantly decreased DAS28 in the same patients (5,47+/-1,53 versus 4,56 +/- 1,19 ; p<5X10exp(-5)) ; no usable control group.

-Whole-body cryotherapy (-110 or -160°C for 2 to 5 min ; 8 to 30 sessions) significantly decreased pain VAS in 124 RA patients (53,15+/-20,46 versus 35,64 +/- 26,69 mm ; p<10exp(-6)). There was again no statistical difference with 17 RA patients treated with low frequency magnetic field.

-Whole-body cryotherapy (-110°C for 2 to 3 min ; 8 to 20 sessions) decreased DAS28 significantly in 83 RA patients (4,86 +/- 0,83 versus 4,74 +/- 0,81 ; p<10exp(-3). There was no statistical difference with the same control group.

Physical therapy was associated to cryotherapy in 5/6 studies and drug therapy intake (NSAIDs, corticosteroids, DMARDs, biologics) was not assessed in 5/6 studies.

Conclusions Local and whole-body cryotherapy significantly decrease pain VAS and DAS28 in RA patients. Further RCTs with adequate control groups and methodology are required to calculate effect size properly.

Disclosure of Interest None Declared