This retrospective study was conducted at the Nephrology Department of the Air Force Specialized Medical Center. Hemodialysis patients diagnosed with COVID-19 infection during the Omicron variant pandemic were hospitalized and underwent HD at the facility. All adult patients on HD who were diagnosed with COVID-19 at the hospital between 10 December 2022 and 31 January 2023 were eligible for inclusion. The diagnostic criterion for COVID-19 was the isolation of SARS-CoV-2 from nasopharyngeal or oropharyngeal swabs using RT-PCR. Patients were excluded if they (1) died before the follow-up interview; (2) were difficult to follow up due to their mental disorders or dementia; and (3) could not be contacted. The Institutional Review Board of the Air Force Specialized Medical Center approved this study (2023–88-PJ01), and informed consent was waived by the committee because of the retrospective nature of the study.

Data on all variables were collected by electronic medical records; these included age; sex; duration of dialysis; body mass index (BMI); diabetes; coronary heart disease (CHD); COVID-19 vaccination status; COVID-19 course; COVID-19 severity; computed tomography (CT) examination; levels of white blood cells (WBC), lymphocytes (LYM), neutrophils (NEUT), monocytes (MONO), hemoglobin, platelets, high-sensitive C-reactive protein (hsCRP), interleukin-6 (IL-6), urea nitrogen before dialysis, potassium before dialysis, calcium before dialysis, phosphorus before dialysis, immunoreactive parathyroid hormone (iPTH), β2-microglobulin before dialysis, total cholesterol (TC), triglyceride (TG), ferritin, prealbumin, albumin (ALB), N-terminal pro-Brain Natriuretic Peptide (NT-pro-BNP), and 25(OH)D3; urea reduction ratio (URR); and, NEUT/LYM and SpKt/V ratios. Routine blood tests were performed using the electrical impedance method (Sysmex XN2800). Lymphocyte subpopulations were detected by flow cytometry (BD FACSlyric flow cytometer). IL-6 and NT-ProBNP levels were analyzed using an electrochemiluminescence immunoassay (Roche e801), and the 25(OH)D3 level was analyzed using a chemiluminescence immunoassay (MAGLUMIX8 fully automated chemiluminescence immunoassay analyzer). Renal and hepatic functions were tested using a Hitachi 008AS.

Long COVID-19 refers to a condition where individuals continue to experience symptoms related to the COVID-19 infection after the acute phase, persisting beyond 12 weeks, and these symptoms cannot be adequately explained by any alternative medical conditions [3]. If a patient still endures at least one of the following symptoms 90 days post-infection, they may be diagnosed with "Long COVID-19": persistent fatigue, incessant coughing, breathing difficulties, nasal congestion or runny nose, cardiac abnormalities, signs of anemia, recurrent headaches or dizziness, sleep disorders, anxiety, depression, cognitive dysfunction, significant hair loss, impaired sense of smell or taste, digestive issues (including nausea, vomiting, diarrhea, or constipation), muscle pain or joint aches, numbness in limbs, unusual skin rashes, or disruptions in reproductive system function [10,11,12]. The presence of long COVID symptoms in all participants were assessed through a questionnaire survey, including the Symptom Questionnaire, modified British Medical Research Council Dyspnea Scale, EuriQol Five-Dimension Five-level Questionnaire, Generalized Anxiety Disorder-7 Questionnaire, EuriQol Visual Analogue Scale, Patient Health Questionnaire, and PTSD Checklist-Civilian Version.

The characteristics of the patients with and without long COVID-19 were described using categorical and continuous data. Categorical data are shown as the number of events and percentages, and the differences between groups were compared using the Chi-square test or Fisher’s exact test. The mean and standard deviation or median (range) were used to describe continuous data, and differences between the groups were compared using the independent t test or Kruskal–Wallis test. The potential predictors for long COVID-19 were explored using a univariate logistic regression model, and the factors were subjected to the multivariate analysis, with α = 0.05, and β = 0.10. A receiver-operating characteristic (ROC) curve was constructed based on these factors to predict long COVID-19 risk, and the predictive value was assessed using area under the ROC curve (AUC) analysis. All reported P values were 2-sided, and P < 0.05 was considered as statistically significant. SPSS 26.0 (Chicago, IL, United States) was used to perform the statistical analysis.