Long COVID appears to be driven by ‘long infection’. Is that what the science says?
The data we have does not substantiates this. It is important to keep an unbiased view.
The article in The Conversation (linked here) suggests that persistent infection might be a factor in Long COVID. However, it is important to note two key points:
The title does not claim that persistent infection causes Long COVID—only that it might play a role.
The article does not specify which infection(s) could potentially drive the disorder, leaving the matter open to interpretation.
These nuances are crucial to understanding the ongoing scientific debate surrounding the mechanisms of Long COVID. It asks the question what the science has revealed so far. This appears an open question, however, the answer to it seems more forceful than the data warrants.
The article claims, “Around 5–10% of people with COVID infections go on to experience long COVID, with symptoms lasting three months or more.” However, this figure references an Australian Government website rather than peer-reviewed scientific literature, leaving its derivation unclear. There is significant uncertainty about the true proportion of individuals who developped Long COVID following SARS-CoV-2 infection, partly due to the lack of a precise definition. Many studies rely on the presence of at least one symptom, often very general ones, such as fatigue or poor sleep, which occur frequently in the general population and are not unique to Long COVID. The incidence of symptoms caused by SARS-CoV-2 is also likely to have declined over time with the establishment of widespread immune memory. Australia's effective measures to keep SARS-CoV-2 at bay delayed mass exposure until 2022, after widespread vaccination campaigns. Thus, the 5–10% estimate likely reflects a specific context, time frame, and initial vaccinated population. It should not be interpreted as SARS-CoV-2-specific and a background incidence due to other environmental factors, including other infections, should be kept in mind. Needless to write, this does not make Long COVID any less worse, it highlights the absence of collective knowledge regarding this condition and the long-term post-infection conditions in general.
Importantly, the cited webpage acknowledges limitations regarding time, definitions, and the presence of a high baseline prevalence of such symptoms in the general population—factors critical to interpreting these figures.
The authors of The Conversation article discuss several potential biological mechanisms that might explain the development of Long COVID, but their central argument is that Long COVID is primarily or solely driven by the persistent presence of SARS-CoV-2.
What is commonly observed in the scientific literature—both in Long COVID patients and those recovered from SARS-CoV-2 infection—are remnants of the virus, such as viral proteins or RNA fragments. The article refers to this phenomenon as "viral persistence." However, this terminology may create confusion, as it can imply that the virus remains active, capable of transcribing RNA, translating proteins, or infecting cells. To date, there is no convincing evidence of active or infectious SARS-CoV-2 in such cases. The authors clarify that there is a distinction between remnants of the virus and infectious material, emphasizing that while traces of the virus might linger, they do not necessarily indicate an ongoing infection. This distinction is crucial in accurately interpreting claims of "viral persistence" in the context of Long COVID.
The authors identify two significant implications of persistent infection. First, as seen at least with the Alpha and Omicron variants, persistent infections—typically observed in immunocompromised individuals—can drive intra-host selection, leading to the emergence of new variants. Second, both infectious material and viral debris may prolong inflammatory signaling, potentially contributing to chronic symptoms. This is in agreement with much data of heightened immune activity found in Long COVID patients, which sets them apart from those recovered from COVID-19.
However, whether persistent infection is causal for Long COVID (or for conditions like ME/CFS after other infections) remains uncertain. It could plausibly be a consequence or a side issue of Long COVID rather than its cause. To date, no infectious agent has been definitively identified as the root cause of ME/CFS, despite extensive research.
What is the current evidence regarding SARS-CoV-2 and its role in Long COVID? While viral remnants have been detected in some studies, no direct, causal link between SARS-CoV-2 persistence and the development of Long COVID has been established. This question remains a critical area for future investigation.
What does the research say?
There is no study that shows that Long COVID is caused by persistent SARS-CoV-2 infection. Immunocompromised people are not suffering from Long COVID symptoms while infected for many months. In addition, there is no study showing Long COVID is caused by viral remnants. This is frustrating for anyone wishing to reduce Long COVID symptoms or to prevent them. It is, however, the current state.
The authors in The Conversation piece suggest that persistent infection is a primary driver of Long COVID, citing several key studies. However, a closer examination of the referenced studies reveals significant nuances and limitations, which complicate the argument for persistent SARS-CoV-2 infection as the sole or primary cause of Long COVID.
Persistent Shedding of Viral RNA
The article states:
"In February, a study in Nature found a high number of people with mild COVID symptoms had extended periods of shedding the genetic material of the virus, so-called viral RNA, from their respiratory tract."
Persistent infections are indeed documented, especially in individuals with some level of immune compromise. However, viral shedding is not unique to SARS-CoV-2. Other respiratory viruses, including influenza, have shown prolonged shedding under certain conditions, as evidenced by studies such as:
The referenced paper in Nature shows persistent shedding in approximately 0.1–0.5% of individuals, typically those with underlying conditions. While this may double the risk of Long COVID in this subgroup, it is misleading to describe this small fraction as a "high number of people."
Detection of Viral RNA and Proteins
The authors state:
"Other key papers detected replicating viral RNA and proteins in blood fluid of patients years after their initial infection, a sign that the virus is likely replicating for long periods in some hidden reservoirs in the body, perhaps including blood cells."
This claim needs more context. Studies indeed detect viral remnants, such as RNA or proteins, in both Long COVID patients and controls, though often at higher levels in Long COVID cohorts. However, remnants do not equate to active infection. Several points emerge from these analyses:
Remnants vs. Infectious Virus: No infectious SARS-CoV-2 has been isolated from these patients. Detecting remnants does not confirm active viral replication.
Confounding Factors: Observed viral RNA loads often correlate with pre-existing conditions, immune activation states, and vaccination status. Notably, vaccines reduce the likelihood of high viral RNA loads (odds ratio [OR] 0.36), while comorbidities increase it (OR 1.61).
Study Timing: Many studies sampled patients during the early pandemic waves before widespread immunity, which skews findings toward slower viral clearance in those with severe acute infections or immune vulnerabilities.
Tissue and Blood Reservoirs
Another study cited claims to have detected viral RNA in multiple tissue sites months after infection, correlating this with Long COVID symptoms. While intriguing, the study has important caveats:
The cohort largely consisted of immunocompromised individuals or those undergoing surgery or chemotherapy—populations inherently predisposed to slower viral clearance.
By four months, detection rates dropped markedly, indicating effective clearance mechanisms in most individuals.
The presence of viral RNA correlated with Long COVID symptoms (OR 5.17), but this association does not imply causation. Viral remnants may reflect a higher viral burden or impaired clearance mechanisms rather than causing Long COVID.
Antigen Presence and the RECOVER Study
The Conversation piece also references findings from the RECOVER initiative, stating:
"∼65% of those individuals diagnosed with PASC were antigen-positive during at least one time point."
This statement requires clarification:
Antigen positivity was detected early in some PASC (Long COVID) patients but was not sustained. Most PASC patients cleared antigens while symptoms persisted, indicating that antigen presence is not necessarily causal or at least not required to sustain (some) symptoms defined as Long COVID.
Among those without PASC, 21% were antigen-positive. This suggests antigen detection may correlate with PASC but does not necessarily result in PASC or confirm a causal relationship. Persistent antigens might result from slower immune clearance or inflammation rather than driving Long COVID.
Threads of the studies:
https://x.com/Marc_Veld/status/1760393258029424882
https://x.com/Marc_Veld/status/1783856453062635992
https://x.com/Marc_Veld/status/1795391034882465918
https://x.com/Marc_Veld/status/1782830646085189640
https://x.com/Marc_Veld/status/1845751447570333759
Antivirals
If Long COVID symptoms are caused by persistence infectious virus, their use should show a substantial reduction in these symptoms. Unfortunately, the first, albeit small, studies have not provided evidence for this. Improving immunity or use of antivirals, prior to infection or during the acute phase does reduce the risk of Long COVID.
https://x.com/Marc_Veld/status/1746509393342943282
https://x.com/Marc_Veld/status/1632068752706789379
https://x.com/Marc_Veld/status/1737503452442267989
Metformin does show hopeful results, but its mechanism of actions remains unclear and may not be directly against the virus, but control aspects of the immune response.
Paxlovid had mixed results https://www.researchsquare.com/article/rs-3359429/v1
Where we stand
The evidence gathered to date does not substantiate the claim that persistent SARS-CoV-2 infection is the primary cause of Long COVID. Viral remnants and antigens have been detected in both Long COVID patients and controls, often correlating with immune function, pre-existing comorbidities, or the severity of the initial infection.
While the hypothesis that persistent infection or viral debris may contribute to Long COVID cannot be completely dismissed and warrants further investigation, interpreting the current studies as definitive proof oversimplifies what is clearly a complex, multifactorial condition. Ongoing clinical trials and future studies will be crucial in providing more clarity and robust evidence on this matter.
“The intensity of a conviction that a hypothesis is true has no bearing over whether it is true or not.“
With the data currently available, it would be premature to proclaim widespread viral persistence as the primary or sole cause of Long COVID. Such assertions risk creating confusion and diverting valuable time and resources away from exploring alternative hypotheses. Focusing on a single hypothesis prematurely may overshadow other plausible explanations that are equally deserving of investigation.
Notably, some findings challenge the persistent infection hypothesis. The presence of viral remnants could be a contributing factor or merely incidental, reflecting delayed clearance or altered immune responses rather than a direct causative role. Other strong hypotheses, such as the role of autoantibodies and their association with specific symptoms, are supported by more robust experimental evidence and merit significant attention.
Critical questions remain unanswered: Is there a (systemic) issue with clearing viral debris or remnants? Which cells are involved in these processes? Is there any causal link to viral reactivation? How many symptoms are explained by auto-reactive antibodies? Will suppressing general immune activity for a limit time allow symptom elevation, or is the heightened activity a consequence and not a cause of Long COVID?
While the field has made strides in mapping these phenomena, it is imperative to move forward with research that tests these hypotheses more rigorously and in an unbiased open way. The future direction of the field remains uncertain, but broadening the scope beyond viral persistence is essential for a more comprehensive understanding and eventual treatment of Long COVID.