Long COVID: the next public health crisis is already on its way

Conservative estimates suggest that over 200 million SARS-CoV-2 infected individuals worldwide will develop persistent symptoms of COVID-19, also referred to as "long COVID" (LC) or "post-COVID conditions".¹ LC symptoms remain largely unexplained and understudied, and it is unclear how long symptoms last.

In the current issue of The Lancet Regional Health - Europe, Wahlgren and colleagues report on a longitudinal follow-up of COVID-19 patients who were hospitalized during the first pandemic wave (between March and May 2020) in Sweden,² one of the longest follow-up studies of hospitalized COVID-19 patients to date. Using a 37-point questionnaire, patients reported that cognition, sensorimotor impairments and fatigue were their most common persistent symptoms; up to 84% of patients still had problems affecting their daily lives 24 months after acute infection. Approximately half of the cohort who were on sick leave associated with LC at 4 months after infection remained on sick leave at 24 months, with health risks such as hospitalization for cardiovascular disease.² Similarly, a UK study confirmed that COVID-19 infection, including LC, was associated with increased short- and long-term risk of cardiovascular disease and death, particularly in those who had severe COVID.³ Nevertheless, many questions remain regarding the reported symptoms of LC, the effectiveness of COVID-19 vaccines against LC, and the vulnerable population for LC.

To date, more than 200 LC symptoms have been reported. In Wahlgren's study cohort, up to 11% (19/165) of patients' mental health was affected by psychiatric disorders.² Importantly, indications of cognitive impairment, sensorimotor disorders, and fatigue are generally based on patient self-report, which is less objective than clinician diagnoses and might not be uniform and accurate across the study population. Persistent physical symptoms after the acute COVID-19 phase may not be due to SARS-CoV-2 infection, as some symptoms may have already been present before COVID-19.⁴ A Spanish group compared the differences in LC symptoms between hospitalized and non-hospitalized patients two years after acute infection. Their cross-sectional study showed that 59.7% of hospitalized patients and 67.5% of non-hospitalized patients had at least one LC symptom two years after infection.⁵ However, neither the Wahlgren nor Spanish studies included healthy controls, limiting their ability to separate effects independently with COVID-19 from those associated with matched control populations.^2,5 Indeed, few studies examining LC have used an uninfected control group, and the clinical presentation of LC patients, including symptoms, physical assessment, mental health and psychotherapy, and diagnostic investigation, remains incompletely characterized in most published studies. From a clinical viewpoint, LC patients require precise medical assessment to prevent pre-existing or unrelated symptoms from being erroneously attributed to COVID-19 infection, and to identify pathophysiologic mechanisms for targeted symptom relief. Currently, objective clinical measures are limited, and biomarkers are urgently needed to close the diagnostic gap for LC.

In Wahlgren's study, 91% of patients received more than two doses of COVID-19 vaccine, but the vaccine type was unknown.² Whether or not vaccination, particularly with inactivated virus, protects against LC symptoms, remains controversial. A recent national cohort study from Israel evaluated clinical sequelae in vaccinated patients with mild COVID-19 up to one year post-infection, compared to uninfected and unvaccinated SARS-CoV-2-infected patients as controls. This study concluded that vaccinated mild patients with a breakthrough SARS-CoV-2 infection had reduced risk of lingering dyspnea compared to unvaccinated individuals, while risks for other adverse health outcomes were comparable.⁶ A Hong Kong study found that those who received two doses of BNT162b2 (mRNA vaccine) had higher levels of protection against severe COVID-19 than those who received two doses of CoronaVac (inactive virus vaccine), especially in older age groups.⁷ Researchers in Malaysia compared the duration of protection using BNT162b2 and CoronaVac vaccines, and found protection diminished more rapidly after two doses of CoronaVac, especially against mild and moderate COVID-19.⁸ Turkish researchers reported that participants receiving a third dose of BNT162b2 had significantly increased neutralizing antibody titers compared to a third dose of CoronaVac.⁹ Surprisingly, a recent Chinese study showed that patients with breakthrough infection who had received three doses of inactivated virus COVID-19 vaccine had significantly lower titers of neutralizing antibodies to Omicron BA.2, BA.4 and BA.5 variants compared to patients with breakthrough infection who had received two doses of inactivated virus vaccine.¹⁰ These results suggest that repeated inactivated virus vaccination back-boosts prior memory and dampens immune responses to new antigenically-related but different viral strains,¹⁰ likely due to stronger immune imprinting against the ancestral strain and suppression of the neutralizing antibody response against new Omicron variants.

A recent study from the United States suggests current COVID-19 vaccines (including WA1/BA.5 bivalent mRNA vaccine) are ineffective against BQ and XBB Omicron due to the superiority of these subvariants in evading antibodies.¹¹ Emerging SARS-CoV-2 variants such as XBB 1.5 and evidence of waning vaccine efficacy due to immune escape pose significant potential barriers to control of the COVID-19 pandemic. According to a World Health Organization (WHO) report, only 30.61% of people worldwide have received booster vaccines.¹² Number of people at risk of developing LC might be greater than earlier predictions, especially in unvaccinated populations in lower-income countries, post-COVID-19 patients living in socioeconomically deprived areas, and seriously ill and vulnerable patients.^1,13 The increasing prevalence of LC threatens healthcare systems worldwide and seriously affects socioeconomic recovery. Strategies are urgently needed to address this developing public health crisis.

Contributors

CY and SJT conceived the manuscript. CY conducted the search and drafted the manuscript. CYand SJT revised the manuscript and accepted the final version for publication.

Declaration of interests

CY and SJT reported a grant from the Canadian Institutes of Health Research (grant number: 177747). No other conflicts were reported.