Read on Twitter
Our paper about low-dose #Hydroxychloroquine use during the first wave of the pandemic in Belgium is now published. This work was based on the Belgian surveillance for hospitalized #COVID_19 performed by @sciensano since early March. 1/n
https://www.sciencedirect.com/science/article/abs/pii/S0924857920303423#!">https://www.sciencedirect.com/science/a...
https://www.sciencedirect.com/science/article/abs/pii/S0924857920303423#!">https://www.sciencedirect.com/science/a...
It includes data from 8075 patients : 4542 HCQ-treated and 3533 HCQ-untreated. Patients exposed to others therapies (TCZ, AZM, LPV/RTV) were excluded. In Belgium low-dose HCQ were recommended and administration was restricted to inpatients subjects after ECG w/o C-I 2/n
Patients treated by HCQ were younger (66 vs 77), more frequently male, less proportion of CV disease, hypertension, CKD, neurocognitive disorders and cancer but more obesity. Inflammatory markers were higher in HCQ group (LDH&CRP) as well as pneumonia,ARDS & ICU admission % 3/n
After adjustments, HCQ remained significantly associated with decreased mortality (0.684 95% CI 0.617-0.758). compared to no HCQ group. Other factors associated with mortality are those already described : Age, male sex, CV dis, diabetes, CKD, LDH, CRP, ARDS, pneumonia 4/n
Important points : statistical analyses were performed by an independent group not involved in COVID-19 treatment decisions. Calendar time of prescription was taken into account. Immortal time bias also. Corticosteroids prescriptions was low in both HCQ & no HCQ group 5/n
Our paper was published in IJAA but it was peer-reviewed in 2 majors journals and reviewed by 10 reviewers in total. Rejections were based on "editorial" choices. This is an observational study with its limitations. We are not promoting HCQ. DXM is SoC in Be. since Recovery. 6/n
Our study is also about risk factors of mortality. We found that non-survivors had shorter duration of symptoms before admission. This is in line with other studies suggesting different phenotypes of severe COVID-19 such as this one by @ElieAzoulay5 https://pubmed.ncbi.nlm.nih.gov/32780165/ ">https://pubmed.ncbi.nlm.nih.gov/32780165/... 7/n
HCQ has no antiviral activity in vivo. This has been shown both in NHP & human cells lines. The /possible/ benefits observed in our study and also in other obs. studies could be related to anti-inflammatory action of HCQ which inhibits IL-6, TNFa, MIP. 8/n https://doi.org/10.1093/cid/ciaa546">https://doi.org/10.1093/c...
This antiinflammatory effects is in line w/ DXM results in RECOVERY. RECOVERY has shown no effect of high dose HCQ (in Belgium : 2400 mg during 5 days vs 2400mg in 24h in RECOVERY). Such high HCQ doses were the reason why SOLIDARITY could not be implemented in Belgium. 9/n
This study was possible thanks to the collaboration of all hospitals in Belgium allowing the collection of data and also collaborators @ugent @sciensano @ITMantwerp @ULBRecherche @ChuSaintPierre @LeilaBelk 10/10
