Long COVID has exposed medicine's blind-spot

On Feb 23, 2021, the National Institutes of Health announced a new US$1·15 billion initiative to support research and resources for so-called long COVID.¹ This is the culmination of a year that has seen more scientific attention, public commentary, and media coverage of chronic unexplained medical symptoms (either post-infectious or not) than arguably the past decade combined. Indeed, one of the most concerning stories emerging out of the COVID-19 pandemic is the quandary of long COVID. Long COVID, or post-acute sequelae of SARS-CoV-2 infection, is being seen in a growing number of patients reporting a constellation of symptoms after SARS-CoV-2 infection that are persistent, debilitating, and have yet to be fully explained by known or measurable mechanisms. These symptoms include fatigue, cognitive difficulties, mood dysregulation, headaches, insomnia, dizziness, and a variety of other neurological, neuropsychiatric, autonomic, and systemic symptoms.² These symptoms are being reported by patients even with mild initial infection that did not require hospitalisation or medical attention. Many physicians have reported having such symptoms post-COVID-19, which has added support to initial pleas that long-COVID symptoms exist and can be debilitating.³ Media organisations across the world have highlighted the complexity of this topic with intrigue and concern. This attention and tone offers a relatively stark contrast to the cynicism that usually plagues chronic, unexplained symptoms and a history of patients who feel that they have been ignored by medicine.⁴

As many others have pointed out, the clusters of symptoms reported by patients post-COVID-19 are not unique or specific to long COVID. Patients with similar assortments of chronic symptoms are commonly encountered in neurology, rheumatology, infectious diseases, and other subspecialty clinics. Some patients will have similar post-infectious onsets, whereas others report other potential triggers, and, for some, there are no identifiable triggers at all. Unfortunately, for most of these symptoms, there are no validated objective biomarkers to aid in diagnosis or to quantifiably measure an abnormal structural state. Indeed, disruptions in brain and brain–body function that probably account for such symptoms cannot yet be reliably identified by conventional blood tests or brain scans. Thus, a common denominator in this field is medical consultations largely based on diagnostic exclusion, in which the absence of further answers or direction for recovery can leave patients feeling dismissed and dissatisfied.⁴

Two broad possibilities exist to explain where long COVID might fit in this complex and controversial field. First, COVID-19 could trigger post-infectious processes that generate persisting symptoms in a unique way that is distinct from previously encountered patients. Although this would traditionally defy guiding principles such as Occam's razor (favouring simple, unifying explanations), we cannot ignore SARS-CoV-2's many firsts, its use of the ACE2 receptor (similar to SARS-CoV), and the particularly aggressive interactions that have been observed with the brain, other organs, and blood vessels in some patients.⁵ Second, long COVID might exemplify the category of mysterious, unexplained, chronic symptoms (either post-infectious or not), and could operate via similar mechanisms to symptoms seen in other patients. The major problem in teasing this out is that the latter, despite a long history of high numbers of patients, has remained very poorly understood and constitutes one of medicine's largest blind-spots. The collective vacuum of uncertainty has opened the floodgates to many different disease notions, diagnostic labels, and conflicting explanations from purely so-called physiological to purely so-called psychological. Starting with physiological explanations, some models present inflammatory or immune-mediated cascades that might place primary importance on a given trigger (eg, the nature of an infection and its interaction with the host). Whereas traditional psychological theories heavily weight psychological factors and the potential overlap between these constellations of symptoms and bodily manifestations of stress responses and anxiety. Alternatively, contemporary neuropsychiatry models present this polarisation as a false dichotomy and highlight the potential importance of predisposing factors, including genetic and psychosocial factors, that might result in dysfunction of brain or brain–body circuits and networks that then interact with a potential triggering event.

Every proposed explanation on this topic has gaps and they are not necessarily mutually exclusive hypotheses. There might also be substantial heterogeneity between patients, with subpopulations at different points along this spectrum. How the brain connects within itself and to the body via complex neural, neurohormonal, and neuroimmune axes is one of the final frontiers of science and medicine.⁶ We still do not fully understand how these interactions occur under normal circumstances so how can we fully understand when they go wrong? Unfortunately, instead of humbly embracing the complexity of these interactions and encouraging collaboration, contrasting opinions are often ferociously defended, creating deep divisions.

For now, the most important thing is to study long COVID with no assumptions and to interrogate potential unique factors about COVID-19 that could explain why these symptoms seem to be triggered with particularly high propensity. The COVID-19 pandemic and the large (and growing) number of patients with long-term symptoms offers an unprecedented window to study these symptoms, their inter-relationships, and their puzzling pathogenesis. Attention is finally being paid to this important topic, and even if this line of research does not lead to definitive answers, we are confident that there will be valuable new insights for this field. We should not care about what ends up being right or more right, wrong or more wrong; we should care about getting closer to reliable, objective markers of these complex symptoms and easing the suffering of these oft-neglected patient populations.

We declare no competing interests. The content of this article is the opinion of the authors and does not necessarily represent the official views of the University of Toronto, Harvard Medical School, or Emory University School of Medicine (and their affiliated academic health-care centers). We thank Saadia Sediqzadah (Department of Psychiatry, University of Toronto) for her helpful comments and review of this manuscript.