Post-COVID syndrome. A case series and comprehensive review

Abstract

The existence of a variety of symptoms with a duration beyond the acute phase of COVID-19, is referred to as post-COVID syndrome (PCS). We aimed to report a series of patients with PCS attending a Post-COVID Unit and offer a comprehensive review on the topic. Adult patients with previously confirmed SARS-CoV-2 infection and PCS were systematically assessed through a semi-structured and validated survey. Total IgG, IgA and IgM serum antibodies to SARS-CoV-2 were evaluated by an electrochemiluminescence immunoassay. A systematic review of the literature and meta-analysis were conducted, following PRISMA guidelines. Univariate and multivariate methods were used to analyze data. Out of a total of 100 consecutive patients, 53 were women, the median of age was 49 years (IQR: 37.8-55.3), the median of post-COVID time after the first symptoms was 219 days (IQR: 143-258), and 65 patients were hospitalized during acute COVID-19. Musculoskeletal, digestive (i.e., diarrhea) and neurological symptoms including depression (by Zung scale) were the most frequent observed in PCS patients. A previous hospitalization was not associated with PCS manifestation. Arthralgia and diarrhea persisted in more than 40% of PCS patients. The median of anti-SARS-CoV-2 antibodies was 866.2 U/mL (IQR: 238.2-1681). Despite this variability, 98 patients were seropositive. Based on autonomic symptoms (by COMPASS 31) two clusters were obtained with different clinical characteristics. Levels of anti-SARS-CoV-2 antibodies were not different between clusters. A total of 40 articles (11,196 patients) were included in the meta-analysis. Fatigue/muscle weakness, dyspnea, pain and discomfort, anxiety/depression and impaired concentration were presented in more than 20% of patients reported. In conclusion, PCS is mainly characterized by musculoskeletal, pulmonary, digestive and neurological involvement including depression. PCS is independent of severity of acute illness and humoral response. Long-term antibody responses to SARS-CoV-2 infection and a high inter-individual variability were confirmed. Future studies should evaluate the mechanisms by which SARS-CoV-2 may cause PCS and the best therapeutic options.

Keywords: Long COVID; Post-COVID syndrome; Post-acute COVID-19; Prolonged COVID; anti-Sars-Cov-2 antibodies.

Fig. 1

Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) flow chart.

Fig. 2

Acute and post-COVID symptoms. A. Frequency bar plot for clinical manifestations on acute COVID-19. B. Frequency bar plot for post-COVID clinical manifestations. Frequency of depression was estimated by Zung scale. C. Venn diagram with the superposition of the main PCS symptoms. Analysis included 91 patients, because 9 out of 100 patients did not exhibit any of the four main symptoms. D. Mirrored bar plot for symptoms on acute COVID-19 and post-COVID syndrome.

Fig. 3

Anti RBD SARS-CoV-2 antibodies (see Supplementary Material 1 for details). A. Correlation between IgG anti-SARS-CoV-2 by ELISA (Euroimmun) and total anti-SARS-CoV-2 antibodies by ECLIA (Roche). B. Correlation between neutralizing anti-SARS-CoV-2 antibodies (PRNT₅₀) and total anti-SARS-CoV-2 antibodies by ECLIA (Roche). C. Linear model for neutralizing antibodies based on ECLIA. D. ROC curve for the estimated cut-off value. The AUC represents the discrimination capacity of this threshold. E. Scatter plot for total anti-SARS-CoV-2 antibodies (by ECLIA, Roche) and post-COVID time. Line represents the locally estimated scatterplot smoothing with 95% confidence intervals.

Fig. 4

Frequency bar plot for pooled prevalence of post-COVID manifestations in meta-analysis. Error bar represents the estimated 95% confidence interval. Only those clinical manifestations reported in more than 4 articles were included in this graph.

Fig. 5

Autonomic clusters in post-COVID syndrome. A. Principal components of K-means clustering for weighted COMPASS 31 domains. COMPASS 31: composite autonomic symptom score 31. B. Mirrored bar plot for acute COVID-19 symptoms on cluster 1 and 2. C. Mirrored bar plot for post-COVID clinical manifestations on cluster 1 and 2.