New study aims to define long COVID through phenotypes of patients

smelling long COVID

Santiaga/iStock

A new study in the Journal of Infectious Diseases uses data from 1,988 SARS-CoV-2–positive US Military Health System beneficiaries to define the characteristics and clinical patterns observed in patients with long COVID, or post-COVID condition (PCC), grouping patients into three phenotypes based on clusters of symptoms. 

The authors of the study said they wanted to use machine learning to analyze data on clinical symptoms 6 months post-infection to identify new definitions for PCC. The longitudinal study took place from March 2020 through May 2022. Case-participants tested positive for COVID-19 during the study period and completed a survey approximately 1, 3, 6, 9, and 12 months after their first positive test.

At each time point, participants were asked about the duration and severity of their symptoms, the authors said. 

"To identify distinct PCC clusters based on differing symptomatology, we utilized a machine learning clustering algorithm of survey responses to identify patterns that differentiated participants based on their reported symptoms," they wrote. 

3 symptom clusters identified

In total, the authors said three symptom-based clusters were identified: a sensory cluster (loss of smell and/or taste), a fatigue/difficulty-thinking cluster, and a difficulty-breathing and exercise-intolerance cluster. 

"Cluster 1 (Sensory) was characterized by a higher frequency of sensory symptoms such as loss of smell and/or taste, Cluster 2 (Fatigue/Difficulty thinking) was characterized by a higher frequency of fatigue (including mental and physical fatigue) and difficulty thinking (e.g., brain fog), and Cluster 3 (Difficulty breathing/Exercise intolerance) was characterized by a higher frequency of difficulty breathing symptoms (e.g., shortness of breath) and exercise intolerance (e.g., difficulty exercising)," they said. 

Of the 1,988 participants included in the study, 60.4% were men, and 69.5% were 18 to 44 years old. Seventy-three percent had no significant comorbidities prior to SARS-CoV-2 infection, and 9.6% were hospitalized due to acute COVID-19 infection.

The authors also found that certain comorbidities were linked to certain PCC symptoms, for example, obesity during COVID-19 infection was linked to difficulty breathing in the PCC period. Everyone in the sensory cluster was treated as an outpatient during COVID infection. 

"The sensory cluster was less likely to be vaccinated (15.6%) compared to the fatigue/difficulty thinking (36.2%) cluster and the difficulty breathing /exercise intolerance (39.1%) cluster," the authors said. 

This study underscores that PCC is not a single condition but, rather, a multisystemic condition with distinct symptom-based phenotype

In an editorial on the study, researchers from the University of Cambridge write, "This study underscores that PCC is not a single condition but, rather, a multisystemic condition with distinct symptom-based phenotypes that have specific risk factors and perhaps unique early inflammatory profiles."

 


 

This week's top reads