None Dare Call It Relapse: Is Covid-19 a Chronic Infection in Acute Disguise?

For all the hopeful talk about “herd immunity”, and the increasingly confident consensus that the Covid has been overestimated and is only really a threat to the elderly and infirm (that is, those with one foot in the grave anyway), there has been a conspicuous lack of discussion of what may well be the most fundamental and momentous question about this illness. That is the question of what proportion of the infected fully recover—and for that matter, how many infected people truly get rid of the virus.

There has been a modest crop of news reports, particularly from the earlier stages of the pandemic, about the problem of relapses and reactivations. All show a worrying pattern of viral persistence, throwing into serious question the official statistics that ostensibly show mass “recovery” from the virus.

Many of these early reports of relapse came from China, the earliest battleground of the pandemic. One report from Guangdong Province, for example, revealed that 14% of those discharged as cured subsequently tested positive again for Covid, though whether this was reactivation of the virus or reinfection with the virus was not clear. Another report warned that in Wuhan, 10% of discharged, “recovered” patients subsequently tested positive again. The most shocking news report of all from Wuhan, in early April, indicated that 26 out of 44 Covid patients—an incredible 59 percent—had tested positive all over again after being released from the hospital.

More recently, some media attention has focused on ongoing problems aboard the Covid-stricken US aircraft carrier Theodore Roosevelt, which until recently was forced to remain stranded in Guam as a handful of sailors, most of them young and fit and having passed highly stringent military criteria for recovery from the virus, nonetheless reported a return of “influenza-like illness” and again tested positive. (Incidentally, it seems likely that any such problem in a military setting will be subject to much more intensive control over information, and therefore is likely be many times more severe than is reported.)

Multiple peer-reviewed scientific studies now reinforce these news reports, giving clinical confirmation that some measure of viral reactivation after apparent recovery is a commonplace with Covid. 

Even controlling for differing test brands, one group continued to find reactivation at the same rate, helping to rule out instrumental error. Another reported that about 9% of discharged Covid patients subsequently tested positive again for the virus, and specifically termed these cases “reactivations”. In particular, the authors of this study wrote,

“SARS-CoV-2 reactivation will be a vexing and persistent problem. Considering numerous patients infected or previously exposed to the virus, such a problem poses a major public health burden in terms of global morbidity and possibly mortality.”

A third study reported on a patient who, while still in convalescence, re-tested positive for Covid despite testing negative repeatedly.

It is worth noting that in most of these relapsing cases, these patients also showed symptoms, though as with the first-round Covid infection, these were for the most part relatively mild. It is also worth noting that the percentages for reactivation prevalence offered by these reports and studies are very possibly underestimates, since these studies were published in March and more reactivations may have occurred in the time since then.

In short, it appears more and more plausible that in at least a large portion of cases, Covid is not really an acute infection like influenza, as we have ceaselessly been told. It is instead something far trickier: a chronic infection that pretends to be an acute infection. It knows very well how and where to hide—and perhaps also, how to bide its time.

Yet so far, just as news coverage has been relatively scant, curiously few papers appear to have been published on this seemingly paramount subject. The silence, one might say, is deafening—both from media and academe.

* * *

For the past month and a half, one of the clearest windows into the Covid reactivation phenomenon, and one of the most helpful quantitative clues for how it might unfold, has been the South Korean Centers for Disease Control (KCDC) “reactivation” or “re-positive” numbers, included at the end of their daily press releases on the outbreak .

KCDC has almost daily posted these numbers, beginning in early April. They are simply a running count, broken down by age group, of Covid patients in Korea who repeatedly tested negative for the virus and seemed to be “recovered”—only to test positive again.

These numbers were of special significance for a number of reasons:

First, they are the only official published information known to this author that systematically documents reactivations in a nation-wide Covid outbreak.

Second, since South Korea was second only after China in developing a large-scale Covid outbreak, it could be expected to begin developing reactivation cases somewhat ahead of the rest of the world—reactivations being spotted only once substantial numbers of patients have made it through the acute phase of the illness and been declared “recovered”. The KCDC’s numbers therefore give a kind of “preview” of how reactivation might evolve in other nations.

Third, unlike China, South Korea is a technically advanced, democratic “open society” (more or less), and so its numbers seem more likely to be produced at a high technical standard and less likely to be at least blatantly politically manipulated (China, we’re looking at you).

Fourth, since South Korea has been almost uniquely successful in halting the spread of Covid—owing to a combination of contact tracing, strong social distancing measures, and high-volume mandatory testing of the population—the KCDC numbers present a sort of “best-case scenario” for the overall trajectory of reactivation.

Fifth, the KCDC did not just provide the total numbers of “recovered” cases that tested positive again for the virus; as mentioned, it also broke these numbers down by age group.

Here are some major takeaways from the daily KCDC reactivation numbers thus far released:

1. The number of reactivating Covid cases in South Korea has shown a remarkably consistent and disquieting upward march, which, up to the most recent data available, showed no signs of deceleration. This is true both of the total number of cases reactivating, as well as of the proportion of patients discharged as “recovered” who subsequently were found to reactivate—which has so far reached about 5%. (This is considerably lower than the other studies and reports already mentioned, which suggest a final percentage of at least 9-14% reactivation—meaning, in turn, that the South Korean reactivation situation probably still has a long way to play out.)

2. This rise in the quantity and proportion of “reactivating” cases has continued even as the rate of new infections has ground to a halt. This strongly argues against most of the reactivations being due to “reinfections”—patients being infected all over again—as overall transmission of the virus in the country had been nearly stopped during this timeframe.

3. Recovered patients ages 20-29 made up the largest single age-group of relapses, at 21.5% of all reactivations (they were also the largest group of infected). Furthermore, a substantial number of even younger patients also showed reactivations; patients under age 30 made up an astonishing 36.7% of all reactivations, and patients under 50 comprised 61.6%. This is in total contrast to the ubiquitous “only the old and infirm need worry” narrative about Covid (and more in line with the story about the US sailors who relapsed).

These ever-increasing quantities of reactivated cases in South Korea grabbed a few headlines, at least at first. After all, few things would seem more relevant than growing evidence that a virus currently ravaging the world might not be nearly so easy to recover from as we’d been told—or that those being sent home with a sense of relief might actually be in for long travails and a cruel disappointment.

But then a funny thing happened. In almost every quarter, media all but stopped mentioning anything about the reactivations. KCDC, too, began to distance itself from its earlier pronouncements. And then, just a few days ago, the KCDC decided that there was really “nothing to see here” and that, as a matter of fact, it would simply stop testing discharged patients for Covid reactivation altogether.

* * * 

The KCDC first began its strange 180-degree turn in its public statements on the significance of Covid patients re-testing positive in early May. It had started out in April by quite confidently calling these cases “reactivations” and frankly doubting that they were false positives due to test error. Then suddenly, in early May, KCDC director Jeong Eun-kyeong announced that the cases that retest positive for Covid were actually just “dead” virus and thus indeed “false positives“—and therefore, in fact, no problem at all.

Media reports immediately—and uncritically—lapped up this new interpretation, and from then on what little public attention had been given to the reactivations all but vanished. “Conservative” publications with a particular interest in “reopening” were often the most eager going so far as to assert that reinfection or relapse was quite simply “unfounded“.

So, about a week after dire warnings from KCDC and WHO about Covid reinfection, relapse, and doubtful immunity, the story had very suddenly become that reactivation cases were really no issue at all. The disease went away, the virus became undetectable—and then, for no reason anyone need to worry about, “dead virus” spontaneously was reappearing in many patents’ bodies. Who cares? Oh, and you’re probably immune happily ever after, maybe, even if it takes you several weeks to get there.

At exactly this same time, some inconsistencies appeared in the KCDC’s posted reactivation data. On May 4, no reactivation data was posted at all. Instead, KCDC substituted something quite different, though no less concerning—a count of “cases under long-term (43 days or more) isolation”.

This “long-term cases” count showed that as of May 1 about a tenth of S. Korea’s cases, or 1,035 Covid patients had taken extremely long to “recover” (or at least extremely long to test negative for Covid, which as we have seen, is not the same thing as true recovery). In effect, about a tenth of patients struggled to kick the virus and had to remain isolated for at least a month and a half. This picture was hardly consistent with Eun-kyeong’s and the media’s claims of “false positives” and “dead virus cells [sic]”.

There were other anomalies, too. On two days, April 22 and 24, KCDC reported exactly the same reactivation numbers, right down to the age breakdown—a total of 10 numbers exactly the same. This is statistically nigh-impossible, and also meant that the count of reactivations, between April 23 and 24, would have to have gone down instead of up, which in a cumulative count of cases is literally impossible. Obviously for some reason the reactivation numbers for April 22 had been exactly copied and substituted for the numbers from April 24. This may have been a mere clerical error, and did not appear to affect the overall trend; inquiries about this discrepancy, however, received no reply.

Possibly even more puzzling than these hiccups in the data were the shifts in terminology within the KCDC reports concerning patients who re-tested positive for Covid. At the outset, the reactivation numbers had rather straightforwardly been listed as “cases that retested positive”. But then, on May 6—the same day as Eun-keyong’s downplaying of the relapses—this title shifted to “cases tested positive after discharge”.

This appears to be a very minor change, but notice that the second description is actually much narrower: “tested positive after discharge” excludes any patients who may have retested positive before being discharged. Indeed, relapses before discharge are a very significant proportion of the total relapses, assuming we take seriously what we know from the case reports from China—not to mention KCDC’s own “long-term cases” count from a couple of days before.

Disconcertingly, it appeared that the KCDC had begun playing word-games to disguise or downplay the true number of Covid cases re-testing positive. An even bigger shift came on May 18, however, when KCDC announced that “based on experts’ recommendations” it would again change the name of these cases, this time to the “PCR re-detected after discharge from isolation”—a meaningless mouthful that manages to leave completely dangling the crucial question of what is being detected.

And so, all mention of South Korean Covid patients re-testing positive for Covid after supposed recovery was to be expunged from future accounts. Perhaps not since the WHO director’s declaration that “stigma” was a bigger danger than the Covid itself had Orwellian control of language seemed to take such precedence over reporting on a health issue of vital importance.

But this was not all. In addition to completely nullifying itself on the nature and seriousness of the reactivation cases, in this same May 18 posting KCDC announced that it even saw no point in testing for reactivations at all any longer: “Under the new protocols, no additional tests are required for cases that have been discharged from isolation”.

Nothing to see here, folks, KCDC seemed to be saying. Move right along now.

In fact, KCDC so urgently believed the Covid reactivations were now obviously, totally not a problem that it deemed it imperative, effective immediately, to stop even reporting on them any more—even as the number of such cases continued to increase unabated. As such, the count of reactivations for May 17—466 in total—stands as of this writing as the last reported figure for the number of Covid relapses in South Korea. Based on all prior evidence, though, one can only assume that the true number has continued to increase—out of sight, out of mind, the relapses that dare not speak the name.

* * *

In the same May 18 report in which future reporting of Covid reactivations was summarily ended, the KCDC also announced that findings from its ongoing “investigation and analysis” into the re-positive cases would be released shortly as a special appendix.

Given the massive downplaying of the reactivations thus far, these promised findings seemed to be teed up as a final reassurance—a way to put nagging concerns about reactivation and long-term Covid infection to rest once and for all. Instead, they ended up revealing that the reactivation situation in South Korea is in many ways far worse than even the previously posted reinfection numbers had implied, while doing nothing to dispel the fundamental concern that Covid might in fact be a chronic illness in acute clothes.

First things first: according to the report, how many of these reactivations are there likely to be? How prevalent is Covid relapse overall?

On this question, the KCDC’s report dropped an absolute bombshell: during the (unspecified) duration of the study, out of a cohort of 269 “recovered” patients monitored after release, 83 subsequently tested positive again for Covid.

That is an average rate of Covid reactivation of over 30 percent—more than six times the 5% rate that had been implied by the KCDC’s previously posted reactivation numbers.

This tells us KCDC must have been re-testing only a small fraction of the discharged patients when they were posting their earlier numbers, thus vastly under-reporting the number of relapses—while giving the impression that the posted numbers represented the relapses for the entire country. If the new figure is right, South Korea can be expected to have not 466, but over 3,000 reactivated Covid infections already: people walking around seeming normal, or with mild symptoms, but harboring the virus nonetheless.

This is momentous news even if the virus behaves itself happily ever after—but that is not the end of it. Given the unflinching increase in known reactivation cases at the time the reporting was stopped, it stands to reason that this 30+% figure is itself an underestimate of the ultimate number of reactivations; the latter could very well end up close to the nightmarish 59% observed in parts of Wuhan, if not higher.

The report also finds, as other studies did, that a great many reactivations include symptoms. Nearly half (44.7%) of those examined by KCDC were symptomatic—”coughs, sore throat, etc.” The presence of significant symptoms in so many patients who have re-tested positive bodes poorly for the argument that reactivation is just due to a testing error, or that the virus in these patients is really “dead”.

In fact, the virus seems to be surviving quite comfortably in these “recovered” patients. According to KCDC’s findings, of 76 re-positive patients tested for the specific quantity of virus, almost 90% had a Ct “above 30”, while the rest had a Ct between 25 and 30.

Ct is a measure of the number of amplification steps needed to get a signal from a viral RNA sample. It is a logarithmic measure, and lower Ct values indicate higher viral concentrations. A Ct of 25, for instance, indicates a viral concentration about 42 times higher than a Ct of 30.

A Ct value of 30, in particular, corresponds to about 26,000 viral copies per milliliter of sample. This is consistent with moderate Covid infection, and is higher than that found in many patients still in the acute phase of the disease. It is therefore not exactly a low value; saying someone tested positive for Covid with a Ct value of “above 30” is sort of like saying someone has a fever “below 102”.

Again something seems out of place. Why is KCDC is trying to reassure the world there is nothing to worry about when it has apparently found that a great many of these reactivated patients have viral copy numbers comparable to hospitalized patients?

In a myriad of ways, KCDC’s own report shows the Covid is also already proving to be far more of a chronic, intractable illness than is widely acknowledged. The median time from “recovery” just to retesting positive was over two weeks, with a sizable number taking far longer. Median time from first diagnosis to retesting positive was a month and a half—similar to the 43+ days in isolation mentioned in its update earlier in May.

Most pointedly, there is no mention in the report of resolution in any of these re-positive cases—we do not know how many of these re-positives have since turned negative again, or how many will stay that way, or what the prospects of a true cure might be. But this much is clear: where Covid is concerned, “cure” and “recovery” look more and more like quite disparate things. For many Covid patients, discharge from the hospital may only be “the end of the beginning”.

* * *

So there are likely a lot more reactivations than we’ve been led to believe. But could it be that KCDC, and for that matter most news outlets, still got it right about the actual danger of these reactivated cases? Could it be that there is simply no point in paying any close attention to this phenomenon?

To be blunt: almost certainly no. Indeed, as we have already seen, the reassurances given so far conspicuously sidestep many of the most concerning issues about reactivation.

The KCDC’s grounds for essentially abandoning the testing and reporting of Covid relapses rests mainly on two major claims in their report and other statements on the Covid relapses (oh, beg pardon—I mean the “PCR re-detecteds after discharge”): first, that contact tracing showed relapsing patients did not appear to be particularly contagious; and second, that culturable virus was not found in the relapse cases’ upper respiratory systems.

Both of these claims, however, are misleading.

On the matter of contagiousness of the reactivated cases, we have already seen that a very large proportion of these reported coughing and sore throat as symptoms. This is anything but reassuring, given that both of these are respiratory/airway symptoms and given the high airborne infectiousness of Covid.

In hopes of putting any worries to rest, KCDC assessed the infectiousness of the Covid re-positives by doing contact tracing on 285 of them. They found 790 contacts in total, 27 of whom turned out positive for Covid.

On its face, this seems like a fairly significant level of transmission. However, the report continues, 24 of these had come down Covid previously, so that only 3 were truly “new” cases.

Here it becomes frankly unclear just what KCDC actually did. According to the report, the 24 positive cases turned out to have been “previously confirmed” positives. Does this mean they were now positive again, so that eight out of nine of these cases were also reactivations? Or could they have even been reinfected by the reactivated cases they encountered? Either would be the very opposite of reassuring.

Questions keep coming to mind. How long were the 285 discharged patients actually monitored—a week, a month? The longer the study went on, the more transmissions one would expect to detect; if the tracing continued back only a few days, on the other hand, we could see an artificially low number of transmissions. Were the released patients diligently self-isolating, so that the chances of their passing the virus were minimized? Or did they interact normally?

The report provides maddeningly little information on any of these questions; it simply elides them. Nothing to see hereget back to work

Anyway, for the 3 “new” positive cases, all of them already had a confirmed case in the family, or contact with a religious group that had suffered many cases. The report’s authors use this to argue that these new cases could not be due to contagion from the re-positives, since “epidemiological studies cannot exclude the possibility of exposure to other infection sources”.

But really this point cuts both ways: if epidemiology cannot rule out infection by “other sources” for the new cases, it also cannot rule it in. Therefore the possibility that even the 3 new cases were in fact due to contact with the re-positive cases is not refuted by KCDC’s investigation, but remains a question-mark. From the unclear information KCDC has given us, the most we can say is that reactivated patients do not seem to be terribly contagious.

KCDC also claims, in line with earlier remarks that the virus in relapsing patients was “dead”, to have been unable to successfully isolate live virus—that is, virus that grow in cell culture—from any of 108 reactivation patients.

This seems, again, to provide reassurance. Unfortunately for the new “nothing to see here” storyline, logic continues to dictate that the only plausible source for persistent symptoms and sustained quantities of “dead virus” long after the initial infection is still… a living virus. For the Covid to reappear (dead or alive) in patients who had remained isolated, then, it must have been still present at undetectable but viable levels for a while, and then resumed replicating somewhere in the body.

KCDC itself seems to concede this indirectly in the same report, when it mentions two newly-confirmed Covid cases for which “Virus isolation cell culture result was negative”. Ergo: a negative cell culture does not preclude confirmed case or active virus.

In fact this has been well understood for some time. What the KCDC is failing to mention is that many Covid-positive patients do not show culturable virus, especially at later stages in their infection, and especially if the virus is looked for only in the upper respiratory system.

A lack of active virus in a throat or nasal swab, for example, is no sign that the virus is truly gone, as a German group reported in Nature over a month ago. These authors state: “Whereas virus was readily isolated during the first week of symptoms from a considerable fraction of samples (16.66% in swabs, 83.33% in sputum samples), no isolates were obtained from samples taken after day 8 in spite of ongoing high viral loads”.

Again there is a lack of details about what KCDC actually did. In particular, we do not know what methods were used to gather the virus. But assuming that the KCDC used one of the two standard methods, nasal or throat swabbing—both of which assess virus in the upper respiratory system—then the absence of “culturable” virus from these tests, which is the basis of many reports of “good news” about reactivations from KCDC and others, is mostly a red herring. While it may indicate that reactivations are less likely to be highly infectious, it in no way establishes the absence of even quite large amounts of live virus in these patients.

Once the red herring of “no culturable virus found” is properly put aside, the real question becomes: where and how is the live virus holding out in these patients? Most likely, the answer lies with the lower respiratory system. One particularly striking study found intact SARS-CoV-2 virus in a “recovered” or asymptomatic patient, hiding deep in the lungs—a place that normal diagnostic swabbing could not access. (The patient in question, unsettlingly, appeared to be fully recovered from Covid and was ready for discharge from the hospital, when she died suddenly of heart failure. Cardiovascular complications have been widely reported in Covid cases, so this may not have been a coincidence.)

If anything, the finding that Covid is capable of hiding in the lower respiratory tract is more concerning for the reactivation cases, since the most harmful effects of acute Covid infection typically only begin once it has reached the lower respiratory system. It also means that the number of “recovered” patients who actually still have active virus in their bodies is probably higher than the number who re-test positive, or test positive on discharge, since many who still harbor small deep-lung reservoirs of live Covid will test negative and be assumed recovered, merely because the standard swab fails to reach these reservoirs.

Some of the “good news” about reactivations has also mentioned “testing error”, as in tests falsely reporting re-positives where there is no virus. But all the foregoing suggests that if there is a problem with the testing, it is one of false negatives far more than of false positives.

* * *

Meanwhile, in the media, the mentions of the Covid’s persistence and recurrence (such few as can be found, that is) have grown eerier, subtler, but more disconcerting. The very understatement of some of these accounts seems to telegraph a growing unease, a dreadful realization not yet fully crystallized. This may just be the sensationalism of journalism at work, but the weight of the evidence increasingly suggests it is not. In light of the strange attempts to paper over or ignore the problem, as we see in the KCDC’s “what me worry” pronouncements that belie their own data, the sense instead is that something major is being hidden.

One report in Bloomberg at the beginning of May briefly mentioned so-called “false-dawn recoveries”, essentially relapses. These come complete with symptoms of Covid and positive tests for viral RNA. The author of the article was careful to hew to the “dead virus” narrative, however, barely mentioning the possibility that these could be bona fide chronic Covid infections.

More recent mentions of the phenomenon have been stranger and darker. A May 10 piece appearing in the New York Times announces that “Surviving Covid-19 May Not Feel Like Recovery for Some”, and details the growing number of survivors of the Covid onslaught in Italy who report ongoing symptoms long after the acute infection ended, delaying full recovery—”if it ever arrives”. “It leaves something inside you,” one of these chronic patients ominously explains, “and you never go back the way you were before.”

Another piece in the Guardian alludes to a vast registry (over 200,000) of largely ignored persistent or “long tail” infections in the UK and USA, which feature “symptoms for months”. These symptoms are described as “weird as hell”, and include neurological problems. Rupert Jones, a professor from King’s College declares, “I’ve studied 100 diseases. Covid is the strangest one I have seen in my medical career”.

Perhaps most revealingly of all, the Guardian piece hints that “long-tail” Covid infections are causing “lots of immunological changes in the body”, and may call for patient support strategies similar to those used in the HIV/AIDS epidemic—hardly an association that instills confidence about the virus’s long-term harmlessness.

Finally, not to be outdone, the UK Telegraph ran its own article on long-term effects of Covid, which include fatigue and mental/memory issues, even in the young. One physician admits there is “emerging evidence that people are shedding the virus for months afterwards, so the virus isn’t going away”—another score against the “dead virus” narrative.

The same Professor Jones from the Guardian piece here mentions his concern that we are currently “underestimating” the virus: “The Government is telling people that this is just like the flu and only checking on a few symptoms, but it’s not at all like the flu […] For many people it can linger on; many people are saying they’ve had it for over three months now.”

Perhaps the most interesting aspect of the Telegraph piece, though, is its description of the way that persistent and/or reactivated cases seem to be getting systematically ignored. “There are people emailing me every day saying that no one is interested,” says Jones. Much of this correspondence seems to include patients fearing for their sanity: “they are going through hell in their heads. They think they’re crazy”. The haunting question of whether this sense of madness comes from simply feeling ignored, or from actual neurological issues caused by the Covid, is left unaddressed.

So this is the difficult, nasty pill that we are now slowly, slowly being fed: many of you may never really get better. It is a hard pill to swallow; even so, it is still quite possibly a sugar-coated version.

It might be the apparent blindness to the existence and enormous possible repercussions of Covid persistence and reactivation has something to do with the preconceptions of the medical and epidemiological community, which is trained on the idea that once a viral infection is cleared and a good antibody to it is produced, it stays cleared. As the Bloomberg article notes, “Such incidents [reactivations] don’t align with the generally accepted understanding of how virus infections work and spread.”

Or perhaps, we are seeing something of what control of information in modern liberal-democracies really looks like when truly hard, no-win choices rear up. The case of the Covid relapses seems a prime test case of this. If it is true that “recovered” are failing to eliminate the virus and are coming down with symptoms again in huge and increasing numbers, and if young adults, far from being impervious are actually the most common age group in which this happens, widespread knowledge of such facts could inspire some, shall we say, unwieldy responses from the general public.

The possibility of relapse is, frankly, a nightmare that the world cannot deal with right now—not while so many hands are full dealing with the acute form of Covid. And so the strategy may well be to suppress the information and suppress the concern.

But the hints are still getting out.

It may yet turn out that the reactivations are mostly harmless, non-contagious, and even eventually do resolve once and for all. But the way that reliable information about this one way or the other has become so scarce (if not suppressed) does not inspire confidence; the fact that governments have every incentive in downplaying the risk does not help matters either. Very likely, we will not learn the truth about the mysterious Covid reactivations until long after the pandemic has resolved—and at a time when most peoples’ attentions have long since shifted to other matters.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s