Despite Extensive Evaluations, Long COVID Causes Remain Unclear

Clinical and immunologic evaluations of people with post-acute sequelae of SARS-CoV-2 infection (PASC) — otherwise known as long COVID — showed a high burden of persistent symptoms but most cases pointed to no specific cause, NIH researchers reported.

Fatigue, labored breathing, chest discomfort, parosmia, headache, insomnia, memory impairment, anxiety, and impaired concentration were the most common persistent symptoms. Increased risk for PASC was seen in women and in people with a history of anxiety disorder, wrote Michael Sneller, MD, of the NIH in Bethesda, Maryland, and co-authors in Annals of Internal Medicine.

While people with PASC reported lower quality of life on standardized testing, abnormal findings on physical examination and diagnostic testing were uncommon, the researchers said. Exploratory studies showed no evidence of persistent viral infection, autoimmunity, or abnormal immune activation in long COVID patients.

This is the first report of an ongoing longitudinal study of COVID patients and controls being followed at the NIH Clinical Center, Sneller said. The biggest finding is the discrepancy between reported post-COVID symptoms and all the normal results on diagnostic testing, he told MedPage Today.

“Anyone who has not seen dozens, if not hundreds, of these patients really can’t appreciate this, especially when you see it over and over and over,” Sneller said. “These people are clearly suffering. They are clearly disabled by this.”

“We performed over 150 tests, including blood tests, lung function tests, and all the parameters that go with that,” he said. “It’s not like we didn’t look hard for evidence for organ damage, but by and large, we did not find that.”

Clinicians who see patients with long COVID symptoms need to thoroughly assess them and refer them to specialists or long COVID programs as needed, Sneller noted.

“The most important thing is that you need to spend more than 15 minutes with the patient,” he emphasized. “Listen carefully to them. Study the symptoms to see if there’s something there. In doing that, we picked up things that had nothing to do with COVID, like a congenital heart defect that one patient didn’t know he had.”

The findings point to a need to accelerate long COVID research, noted Aluko Hope, MD, MSCE, of Oregon Health & Science University in Portland, in an accompanying editorial.

“As we emerge from the hell of the COVID-19 pandemic, it is time to insist on painstaking study and care of survivors of COVID-19,” Hope wrote. “Without a fuller understanding of pathophysiology and disease course, we must not allow normal objective tests to negate our patients’ subjective experiences.”

“We need to explore whether real-world measurements during both restful and stressful conditions offer insight into the mechanisms of PASC,” he added. “While we gather evidence, our patients deserve personalized care pathways that acknowledge the many biopsychosocial factors involved in illness recovery.”

The study enrolled 189 patients who were at least 6 weeks out from laboratory-documented COVID-19 and 120 antibody-negative control patients in a 100-mile radius of the NIH Clinical Center in Bethesda from June 30, 2020 to July 1, 2021.

Median age of the COVID cohort was 50 and 55% were women. Median age of controls was 51 and 55.5% were women. Most COVID patients (88%) were not hospitalized during acute illness.

In the COVID group, 55% of previously infected patients experienced persistent symptoms. This does not reflect the prevalence of long COVID since people with lasting symptoms were likely more motivated to enroll in the study, Sneller pointed out.

The enrollment visit included neurocognitive assessments, pulmonary function testing, a 6-minute walk test, and echocardiography. Cognition was measured with NIH Toolbox tasks to evaluate processing speed, episodic memory, and executive functioning. Neurocognitive performance was corrected for age, sex, race, ethnicity, and education.

Participants also completed the Short Form-36 Health Survey to assess quality of life and the very brief Patient Health Questionnaire-2 and the Generalized Anxiety Disorder-2 surveys to assess depression and anxiety.

Abnormal findings on physical examination were less common than reported symptoms and did not correlate with the presence of specific symptoms in either group. They occurred with similar frequency in the COVID-19 and control groups, with the exception of abnormal musculoskeletal findings which were 8% in the COVID-19 group and 1% in the control group.

Performance scores for the three NIH Toolbox domains were not significantly associated with PASC or with persistent fatigue, neurologic symptoms, or cardiopulmonary symptoms.

In the COVID-19 cohort, 27% of those who were unvaccinated were negative for neutralizing antibody levels to spike protein, while all vaccinated patients had detectable levels of neutralizing antibodies. No evidence of persistent viral infection or damage to tissue and organs was seen in patients with persistent symptoms. However, people with persistent symptoms reported worsened physical and mental health and lower quality of life than either control participants or people who had COVID-19 without persistent symptoms.

Among potential pre-COVID-19 risk factors, only female sex (OR 2.34, 95% CI 1.25-4.41, adjusted P=0.033) and self-reported history of anxiety disorder (OR 2.78, 95% CI 1.35-5.98, adjusted P=0.027) were associated with increased risk for PASC.

This constellation of symptoms without test abnormalities resembles what’s been described with other illnesses including chronic fatigue syndrome, post-infection syndromes of certain viral and bacterial infections, and mental health disorders like depression or anxiety, Sneller noted.

The findings have several limitations, the researchers acknowledged. Most people with COVID had mild to moderate initial illness that did not require hospitalization and results do not reflect PASC experienced by persons with severe disease requiring hospitalization. The study also did not capture PASC that resolved before enrollment. Additionally, the sample may not be representative of all people with long COVID.