Hello! As always, quite a few news around COVID this week, but also some other interesting papers. Have a good reading!
1/ COVID-19
To abandon hydroxcychloroquine once and for all, The Lancet published a study based on the registries of more than 600 hospitals worldwide to compare mortality and the onset of arrythmias among patients with hydroxychloroquine (or chloroquine) alone or in association with azithromycin versus patients without any of those medications. Out of 96,000 included patients, around 15,000 had one of these medications. The authors found that, after adjusting on severity and other confounding factors, patients who were treated with hydroxychloroquine had a higher risk of death (18% vs 9.3%, NNH=12) and it was even worse with hydroxychlorquine + azithromycin (22% vs 9.3%, NNH = 8). We can also observe an increase in ventricular arrythmias for patients with those medications. Overall, it is not a randomized controlled trial, but those who treated their patients with compassionate use of hydroxychloroquine should have waited for more solid data. The major risk factors associated with an increase of mortality on top of these medications were smoking, cardiovascular diseases and COPD, whereas the use of ACEi seemed to be a protective factor.
Let us move on to a randomized controlled trial regarding remdesivir, that was published in the NEJM. Firstly, it is a study that was not funded by the pharmaceutical industry. Included patients were COVID+ with a lower respiratory tract infection (so not all COVID+ patients). The main endpoint was the time to recovery (i.e. the time to get out of the necessity of an hospitalization with active care). We can also see that they modified the primary endpoint during the study, but they do not hide it. The results are intermediate results of a still ongoing study, but the authors have reached the required number of events (recovery). There was not any adjustment on the number of tests, so we only look at the primary outcome. The statistical analysis plan was using the O'Brien and Fleming method for intermediate analyses: overall the alpha-risk is at 5%, but at the first intermediate analysis the significance threshold was 0.0001 for example, then 0.001 at the next one and the last 0.0489, so all in all the sum makes 5%. But we do not know what intermediate level the authors used for the final analysis of their study. In the end, over 1,000 patients were randomized (50% with HBP, 30% with diabetes, 30% with obesity). The recovery was faster with remdesivir: 11 days versus 15 days with placebo, p-value <0.001 (apparently, low enough to be considerated valid with the statistical conditions described above for the intermediate analysis). But there was not less mortality. Overall, this treatment seems to be efficient to reduce the length of hospital stay by 4 days (which is not that much) without a proven effect on mortality now, but maybe the results at the end of the follow-up will bring more precision on this topic.
The French Health Authority (link in French here) published guidance on rapid serology tests. Regarding the rapid screening tests in pathology lab, the indications are the same than the normal serology tests (see here). Regarding the point-of-care tests, the Health Authority has not found any published study about them. Even though these latter tests could be an option for diagnosis catching-up for patients who do not have access to a pathology lab or who stay symptomatic despite a negative PCR or who could not get a PCR test, the Health authority recommends that a normal serology test should be systematically performed to confirm the result of a positive POC test (and same for a negative POC tests, but it is only advised).
So, how many people have suffered from COVID? This JAMA study, performed in Los Angeles, found out that, among a representative sample of 1,700 individuals who accepted to be screened, 13% had fever with cough, 9% fever with shortness of breath and 6% dysosmia or dysgeusia. So, authors concluded that only 4.5% of the population suffered from COVID.
Another study in the JAMA about the olfactive impairment with COVID. It allows to discover lots of words to identify olfactive impairment: orthonasal, retronasal impairements, phantosmia, parosmia, parageusia. If the impairment is acute in a viral context, COVID can be an explanation. If the impairment is progressive with a fluctuating nasal obstruction, it is probably a naso-sinusal aetiology. If the impairment is progressive, non fluctuating, in older people with loss of memory, it is probably neuro-degenerative. At last, if the impairment is post-traumatic, it is usually brutal with a severe anosmia. Regarding the management of olfactive impairment due to COVID, the authors suggest to: 1/ put smoking detectors in the house and to verify expiry dates on food before eating it; 2/ try olfactive rehabilitation by sniffing lemon, rose, eucalyptus etc, 20 seconds each, twice daily for at least 3 months; 3/ try oral omega-3 and intranasal vitamin A.
2/ Respiratory medicine
The NICE has published guidelines about venous thromoboembolic diseases. Nothing new compared to the European and French guidelines we spoke about here (link in French). The authors recommend the use of the PERC criteria to rule out a pulmonary embolism in case of low-risk and the use of D-dimers age-adjusted threshold for patients over 50. They recommend against cancer screening for idiopathic events without any other symptoms. The advised first-line treatments are apixaban and rivaroxaban, for a duration of 3 months for the events with an identified cause ; for an unknown duration, depending on the risk-benefits balance for events without. For patients refusing long-term anticoagulation, the authors suggest using low-dose aspirin.
3/ Gastro-enterology
After the discussion about bowel cancer screening (link in French here), here comes a BMJ article discussing the colonoscopy surveillance and when to stop it. So, if a colonoscopy was indicated and polyps were found, what should the surveillance be?
- For 1 or 2 sessile polyps or adenomas < 10 mm, return to non-invasive screening tests is advised.
- From 3 of them (English guidelines' threshold is at 5), a 3-year colonoscopy is recommended (if it is only sessile polyps, returning to non-invasive screening can be considered).
Going back to non-invasive testing means waiting for 10 years after the colonoscopy before restarting any test! Only American guidelines recommend a colonoscopy surveillance every 3 or 10 years, depending on the situation. The former is also what I usually see for my patients in France, but the return back to non-invasive testing is motivated by the low risk (<0.5%) of severe complications during colonoscopies that can add up with repeated colonoscopies and starts to be significant.
- In the other situations (adenomas over 10 mm or with high-grade dysplasia; or sessile polyps > 10 mm or with any grade of dysplasia): a 3-year colonoscopy surveillance is recommended.
- From 3 of them (English guidelines' threshold is at 5), a 3-year colonoscopy is recommended (if it is only sessile polyps, returning to non-invasive screening can be considered).
Going back to non-invasive testing means waiting for 10 years after the colonoscopy before restarting any test! Only American guidelines recommend a colonoscopy surveillance every 3 or 10 years, depending on the situation. The former is also what I usually see for my patients in France, but the return back to non-invasive testing is motivated by the low risk (<0.5%) of severe complications during colonoscopies that can add up with repeated colonoscopies and starts to be significant.
- In the other situations (adenomas over 10 mm or with high-grade dysplasia; or sessile polyps > 10 mm or with any grade of dysplasia): a 3-year colonoscopy surveillance is recommended.
4/ Infective diseases
In the PrEP context (link in French), a new preventive treatment of the HIV infection is currently under development: cabetogravir, which has a 8-week lasting effect. A randomized controlled trial among men (and transsexuals) who have sex with men, compared IV cabotegravir to oral Truvada, which is the current advised drug for PrEP. After around 4 years of treatment, in an intermediate analysis, 0.38% patients with cabotegravir had a seroconversion versus 1.21% with Truvada. But 80% of the patients with cabotegravir had side effects versus 30% with Truvada. Let us wait for the final publication of the results now.
5/ Diabetology
Can a biologist tell me how OGTT tubes during pregnancy are centrifugated? It is because this study finds that glucose levels are higher in case of an early centrifugation compared to a late one: early centrifugation could lead to increase the number of gestational diabetes diagnoses by 11.6%!
A JAMA article speaks about the role of the insulin in uncontrolled type 2 diabetes. The authors recommend not to delay insulin initiation when initial HbA1c is over 9% = 75 mmol/mol (and not 10% = 86 mmol/mol as guidelines say), because non-insulin treatments only lower HbA1c by 1.6% = 5.5 mmol/mol on average. This is insufficient to go under the 8% = 64 mmol/mol threshold, that seems to be associated with less mortality and less complications (please note that the authors do not call for a strict control (link in French), which has no evident benefit).
Thanks for your loyalty, let us meet again next week for the next Dragi Webdo!
@Dr_Agibus (free translation by @carttom)
No comments:
Post a Comment