Doctors at Tygerberg Hospital’s intensive care unit have explained in the South African Medical Journal how they’re treating Covid-19 patients. Photo: Ashraf Hendricks
12 May 2020
This is the third issue of Covid-19 Report. We point you to the latest quality science on the pandemic. If you come across unfamiliar terms, there is a glossary at the bottom of the article.
There are broadly three types of tests for an infection like Covid-19:
Ones that check for the genetic material of the virus. These are known as PCR tests. In South Africa when you get tested for Covid-19 this is currently what’s used. The advantages are that they are reliable and tell if you’re currently infected (though only a few days after being exposed). Disadvantages are that these are mostly time-consuming and expensive.
Antigen tests: These detect viral protein fragments. The FDA has given emergency authorization use (which is not the same as approval) to the first such test, made by Quidel Corporation. Advantages are it’s cheap (about $5 according to one source), gives results in 15 minutes (but as far as we can tell — we’re not sure — it has to be done in a lab), and Quidel can quickly start producing millions of them. If you test positive, you’re almost certainly currently infected. But the disadvantage is that it has a high false-negative rate. In other words, you may be infected but the test may say you’re not.
Antibody tests, on the other hand, detect if you have previously been infected with Covid-19. These should give a positive result even after you’ve recovered. They’re usually less expensive to make and usually give results quicker than PCR tests. This is vital for surveillance, to see what percentage of the population has been infected. From this we can learn how fast the virus spreads and how much immunity people have against reinfection. But they don’t tell you if you’re currently infected.
Up till now the accuracy of antibody tests has been too low and this has resulted in some poor quality surveillance studies that have got wide media coverage.
A study published in Nature found that every person infected with SARS-Cov-2 (the virus that causes Covid-19) develops a type of antibody called Immunoglobulin G, within 19 days of getting symptoms. The developed antibodies appear to be specific to this virus. So in theory we should be able to get close to perfect accuracy antibody tests.
Swiss-based pharmaceutical company Roche claims it has developed a test that detects SARS-Cov-2 antibodies with 100% accuracy if you have them (no false negatives), and with more than 99.8% accuracy when you don’t have them (very few false positives), with the condition that it is at least 14 days since a PCR test would have shown you are infected. It also claims to be able to produce over ten million tests from this month. The FDA has granted Roche’s test “Emergency Use Authorization” (again, this does not mean the agency has approved the test). Whether Roche has hyped its claims or made a breakthrough remains to be seen.
Last week we reported on the modest findings about the drug remdesivir. Now The Lancet has published the results of a clinical trial in Hong Kong comparing two drug regimens: interferon beta-1b, ribavirin and the anti-HIV drug lopinavir-ritonavir versus lopinavir-ritonavir alone. The study compared the time it took for the patients on each regimen, all with mild to moderate Covid-19, to test negative, in other words, be officially recovered. It took on average seven days for patients on the more intensive regimen versus 12 days for the lopinavir-ritonavir alone patients (from when they started treatment). None of the 127 trial participants died.
As one South African pharmacologist has pointed out: this trial had no placebo arm nor a standard-of-care arm (lopinavir/ritonavir is not known to treat Covid-19, though see below). Also the endpoint was virological (PCR test result) rather than clinical (how well the patients were doing).
On the same day the outcome of a trial of lopinavir/ritonavir was published in the NEJM, and found no benefit (or harm). Here the endpoint was clinical, rather than virological improvement.
Taken together these two studies do suggest a very modest benefit to the three drug regimen. Whether there’s sufficient benefit for this regimen (or remdesivir) to become part of standard practice remains to be seen. We’re not convinced.
An analysis published in the NEJM of 1,446 Covid-19 patients at a New York medical centre, 811 of whom received hydroxychloroquine, showed no benefit to using the drug.
It’s not a clinical trial — only an observational study — and this doesn’t necessarily mean that it’s the end of the road for hydroxychloroquine and Covid-19. Also, the patients who were given the drug tended to be sicker.
But it does suggest that if hydroxychloroquine does help, its benefits will be modest. A well-run clinical trial that provides more definitive information on the drug’s safety and efficacy is still needed.
That trial may now have been done. A not yet peer-reviewed study of 150 patients in China found no difference in time to testing negative for people with mild and moderate Covid-19 taking hydroxychloroquine compared to not taking it. There were more adverse events in the hydroxychloroquine patients. It’s a small number of patients, and the paper needs more scrutiny, but it’s not looking promising for hydroxychloroquine.
A single paragraph from an article in the latest issue of Science is packed with useful information:
“The coronaviruses are single-stranded RNA viruses that infect vertebrates and move between different host species. With the emergence of SARS-CoV-2, there are now seven coronaviruses that are known to infect humans. Four of them (HCov-229E, HCoV-OC43, HCoV-NL63, and HCoV-HKU1) are responsible for ~30% of cases of the common cold in humans. Two of them caused recent epidemics that had considerable associated mortality: SARS-CoV-1, which emerged in 2002–2003 and causes ~10% mortality, and Middle East respiratory syndrome coronavirus (MERS-CoV), which emerged in 2012, is still active, and causes ~35% mortality. Both epidemics affected a relatively small number of patients compared with Covid-19, which is more transmissible for several reasons, including asymptomatic carriers, long latency period, and high infectivity. Before Covid-19, only SARS-CoV-1 and MERS-CoV caused severe disease. Therefore, coronaviral drug discovery has been a small effort relative to that for other viral diseases such as influenza.”
The MRC’s weekly update still shows that the number of registered deaths (7,135) for the latest week for which there is data, 22-28 April, is lower than the expected deaths (7,319 to 8,494). Natural deaths are as expected, but deaths due to homicides and vehicle accidents are down.
***
The NICD has published a very good article on who should get an influenza vaccine this year: “pregnant women, children aged between six months and five years, people older than 65 years, those with chronic medical conditions such as HIV, heart or lung problems, and healthcare workers”.
The jab doesn’t protect against Covid-19 at all. But it does reduce the risk of getting both Covid-19 and flu. If people get vaccinated it will also reduce the pressure on the health system.
***
The NICD has also published the second issue of its very useful Covid-19 weekly epidemiology brief. One of the many interesting tidbits in it is that in terms of tests per 100,000 people, the Western Cape (788) is still well ahead of Gauteng (560), KwaZulu-Natal (407) and the Eastern Cape (369).
***
A letter in the SAMJ by several Tygerberg Hospital clinicians describes how they are treating very ill Covid-19 patients. The first six patients to the intensive care unit were put on ventilators; all died. “Following international reports of improved survival rates with non-invasive ventilation, the Covid-19 ICU at Tygerberg … changed its standard operating procedure in mid-April 2020 from the initially promoted ‘early intubation and mechanical ventilation’ to High-flow nasal cannula oxygen therapy’, to avoid intubation.” One out of seven such patients died. This is a tiny number of patients and this wasn’t a clinical trial or study, but the results are promising.
****
Another letter in the SAMJ warns that placing people who live on the street in temporary shelters as part of the lockdown increases the risk of overdose among people who are dependent on heroin.
GlossaryFDA (Food and Drug Administration): US government organisation responsible for authorising or approving the use of medicines.
Hydroxychloroquine: Drug used to treat malaria, rheumatoid arthritis and other conditions.
MRC (Medical Research Council): South African state research institution.
Nature: Leading science journal
NEJM (New England Journal of Medicine): Leading medical journal based in the US.
NICD (National Institute for Communicable Diseases): South African state institution responsible for dealing with infectious diseases like TB and Covid-19.
Remdesivir: Antiviral drug manufactured by Gilead Sciences.
Science: Leading science journal
The Lancet: Leading medical journal published in the UK.
Read previous issues of Covid-19 Report.