As hundreds of thousands of people continue to become infected with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), leading to coronavirus disease 2019 (COVID-19), a significant minority will develop severe disease, and about 1% will die. However, many more who recover from the acute infection will have persistent symptoms and debility. This sequalae, encompassing a range of different post-viral complications, has come to be known as “long covid.”
A new study, released as a preprint on the medRxiv* server, discusses the typical symptoms and signs of this syndrome, with the intention of improving the specificity of detection and reporting.
Possible post-acute sequelae
Many symptoms have been recorded in the post-acute stage (post-acute sequelae of COVID-19, PASC), including tiredness, breathing difficulty, chest pain and cough; anxiety, depression and post-traumatic stress disorder, among emotional and mental disorders; neurocognitive symptoms such as poor memory and concentration.
These may last for weeks or months; however, all may not be the direct result of COVID-19.
Since many earlier studies on this syndrome have been based on small numbers of confirmed and often hospitalized-discharged COVID-19 patients, and since they rely on self-reported symptoms and outcomes, they may be subject to some potential biases.
The current study aimed at analyzing newly diagnosed symptoms and conditions in a large number of patients, irrespective of their COVID-19 history, in order to help disentangle those conditions that are truly associated with the acute infection.
The study was retrospective in design, with over 57,000 patients, all of whom had undergone the gold standard reverse transcriptase-polymerase chain reaction (RT PCR) test for the virus.
The analysis was performed with the help of a computational model called MLHO that was designed to reveal associations using electronic health records (EHR), and that has been shown to be useful in the evaluation of this condition.
All these symptoms were recorded in the EHR at two or more months from the performance of a COVID-19 test in patients not hospitalized with the infection at any time, and who were still alive, and had two entries in their EHR, six months apart or more, within three years before the test. This was to make sure these patients would probably be captured by the study center.
Of the 57,000 patients, about a fifth of them were positive for the virus (around 11,400).
The researchers found new symptoms reported between 3-6 months, and 6-9 months, respectively, following the COVID-19 test, in both patients who tested negative and those with positive results. This yielded 28 positive associations with a history of SARS-CoV-2 infection, in groups varying by age and sex and by time.
Seven of the phenotypes had over 97% confidence for a history of COVID-19.
Between 3-6 months, alopecia was 3.5 times more likely overall, as well as in women under 65 years. In women over 65 years, the odds of alopecia in this period were over four-fold higher. The risk was not increased between 6-9 months.
Non-specific chest pain was also present at a higher frequency between 3-6 months overall, with 35% higher odds but more than 2.5-fold higher odds in the group of women under 65.
Men who were 65 years or more had 1.8-fold higher odds of edema, while those under 65 showed 2.6-fold higher odds of proteinuria, within 3-6 months.
In women aged 65 and over, within 3-6 months, gallstones (three-fold higher odds) and dementias (2.5-fold increase) were observed to be associated with a history of COVID-19, as well as twice the odds of Paget’s disease or other bone conditions, more in men.
In the later time period, nail disease was found to be associated with a history of prior COVID-19 in men 65 years or more.
Similarly, in the later time period, women of 65 or over had twice as high odds of anxiety disorders and dizziness. Those younger than this had two-fold higher odds of menstrual disturbances.
A less strong association was seen with acute-angle glaucoma in men under 65 at 6-9 months after diagnosis, at four-fold the odds.
Both time periods
Altered taste and smell were found to be twice as likely in both time periods, with women under 65 showing 2.5-fold higher odds.
Other less robust associations with previous COVID-19 were seen, such as chronic fatigue syndrome, especially in women and those under 65 years, in both time periods. Patients under 65 also had over two-fold higher odds of conjunctival disorders.
What are the conclusions?
These findings suggest that some of the previously identified post sequelae of COVID-19 may not be accurate and that most of the PASC are observed in patients under 65 years of age.”
Earlier studies have shown that PASC includes alopecia, anosmia and dysgeusia, and non-specific chest pain. The current study confirms them, and shows these to be early sequelae.
The findings also show Paget’s disease as an association with prior COVID-19. This will require further work for validation. Chronic fatigue syndrome is also an important association in those with a history of this infection.
Nail disease was strongly associated with a history of COVID-19, at 3.5-fold higher risk. This includes leukonychia, onycholysis, Beau’s lines and mee’s lines. These are markers of general ill-health, being associated with kidney and liver disease as well as infections. However, Beau’s lines seem to show a specific association with COVID-19.
The researchers consider this to be a possibly common PASC, and probably the result of the widespread inflammation and, in many cases, kidney damage caused by COVID-19.
Similarly, proteinuria in young patients associated with a positive history of this infection seems to indicate that the renal damage is not so short-lived as previously thought. Acute glaucoma should also be considered as a potential PASC, according to this study.
The approach used in this study was designed to prevent a plethora of false associations along with more flexible criteria for identification of PASC than the standard phenome-wide association study (PheWAS).
In future work, hospitalized patients should also be included. EHR data is abundantly available, “providing exceptional opportunities for instrumenting healthcare systems to study evolving pandemic byproducts.”
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.