COVID: negative hydroxychloroquine study

another observational (but better) study on hydroxychloroquine is in pre-print and pre-peer-reviewed form (see covid hydroxychloroquine not help france2020 in dropbox, or https://www.medrxiv.org/content/10.1101/2020.04.10.20060699v1.full.pdf )

Details:

--181 patients were from 4 French tertiary-care hospitals, all with documented SARS-CoV-2 pneumonia and needing at least 2L/min of oxygen.

--they compared  84 patients on hydroxychloroquine (HCQ) 600 mg/d vs 97 not on it, hospitalized between March 17-31

    --mean age 60, 71% men, oxygen sat 92%, CRP>40mg/L in 86%

--exclusion criteria included starting HCQ prior to hospitalization, or using other experimental drug within 48 hrs of admission (tocilizumab, lopinavir/ritonavir, remdesivir) or steroids

--mean of 7 days from symptom onset to hospitalization

--HCQ group received med within 48 hours of admission

--almost all had bilateral pneumonia and 75% moderate or severe lung infiltrates

--initial disease severity was well-balanced between groups, controlling for the usual suspects, as well as the use of ACE/ARBs, and CRP (>40mg/L vs less) [though, tough to balance the groups well when 86% had the higher value]

--composite endpoint: transfer to ICU within 7 days from inclusion and/or death from any cause

Results:

--ICU admission or death within 7 days: 16 patients (20.2%) on HCQ vs 21 (22.1%) if not; RR 0.91 (0.47-1.81)

--deaths within 7 days: 3 (2.8%) on HCQ vs 4 (4.6%), if not; RR 0.61 (0.13-2.90)

--ARDS within 7 days: 24 on HCQ vs 23 if not; RR 1.14 (0.65-2.00)

--overall results similar for all of these outcomes when limited to patients with better prognosis on admission (qSOFA <2)

--8 (9.5%) patients on HCQ developed EKG changes leading to HCQ discontinuation (7 had increases of >60ms, one developed a QTc >500ms)

    --1 patient also developed 1st degree AV block after 2 days of HCQ

--17 patients (20%) in the HCQ group got azithromycin and 64 (76%) got amoxicillin/clavulanic acid (unclear why)

Commentary:

--the % of patients in this French study needing oxygen on hospitalization or transferred to the ICU was similar to a Chinese study, as were the sex and comorbidities

--one issue here is that hydroxychloroquine may not be as effective as a solo agent. these patients mostly had high CRP levels (>40 mg/L), which reflects an increased risk of cytokine storm. Would adding steroids (as done frequently in the reports from China) have helped in preventing the cytokine storm, which is associated with such a high risk of morbidity/mortality?? or some of the other antiinflammatory options (tocilizumab is being tested, a specific IL-6 blocker, with IL-6 being an apparent bad actor; high IL-6 levels are associated with a worse prognosis)

    --and, given the multitude of potential benefits of HCQ, would giving the med earlier or to less sick group (eg with CRP<40) have worked better?? see http://gmodestmedblogs.blogspot.com/2020/04/covid-chloroquines-antiviral-effects.html

--also, they did not check PCR viral loads on therapy, which might have been useful given other studies suggesting that HCQ was beneficial in achieving a more rapid decrease (and, among other things, might protect health care workers from some SARS-CoV-2 exposure)

--the issue of severe adverse reactions to HCQ is real, though quite infrequent. they observed 2 adverse cardiac effects besides QTc prolongation. As noted, cardiac effects are not uncommon with Covid-19 itself (see http://gmodestmedblogs.blogspot.com/2020/03/covid-cardiac-injury-common.html) and QTc prolongation and torsades is probably more likely in older patients, especially if they have renal impairment (not so uncommon wth covid), or interact with other meds (not unlikely in sick patients)

    --8 patients did not continue HCQ because of EKG changes, but 17 were also on azithromycin (and the combo of HCQ and azithro additively increase QTc intervals, though no comment in the article if this combo was associated specifically with the increased QTc cases they found). And if these 8 patients did not have QTc prolongation (perhaps if the azithro were not added), this 10% of patients might have changed the results???

--this was not an RCT, and should not be confused for one: they retrospectively tried to match patients who seemed similar in 4 different hospitals in France. 

    --those who received hydroxychloroquine may have been sicker (or less sick) in the eyes of the treating clinical team, and this may not have been apparent in their attempt to match patients. 

    --And, as noted above, it is fraught with potential distortions when one tries to balance a factor like CRP, when only 14% had CRP<40 mg/dL.  I would also add that using a binary cutpoint could distort the results: a CRP of 40.01mg/L  may have a very different prognostic value than 98.3 mg/L. so, for such an important prognosticator, and especially one so prevalent in the overall group (and, by the way, more so in the HCQ group at 91% vs 82%), it might have given us more insight if it were a continuous variable, or at least had multiple cutpoints

so,

--overall a pretty good study (not an RCT, so we should be careful in overinterpreting), suggesting that hydroxychloroquine does not work so well.

--but as with many studies, it raises more questions than pprovides answers (and points to some new and modified studies that might be considered)

    --should we simultaneously give some anti-inflammatory agents (steroids, tocilizumab, others) to prevent decompensation from cytokine storm? in all patients? in those with high CRP levels? in those on HCQ and high CRP levels? in those with other high inflammatory markers (eg IL-6, also a marker of increased Covid-19 severity)?

    --their experience with prolonged QTc may undercut the not-so-substantial evidence that adding azithromycin is beneficial. on-going RCTs should answer this question, and hopefully comparing efficacy of HCQ vs HCQ/azithro. and perhaps HQC plus an effective anti-inflammatory in those with high inflammatory markers (CRP, IL-6)? or HCQ given earlier? or HCQ with ivermectin??? (see http://gmodestmedblogs.blogspot.com/2020/04/covid-potential-vaccine-and-med.html). lots of potential iterations of the HCQ theme. and given our current (dismal) reality with Covid-19, it seems that pretty much all comers deserve testing

geoff

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.orger "clinical stuff" for articles, or "powerpt presentations" for the powerpoint presentations

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list