Persistent attenuation of lymphocyte subsets after mass SARS-CoV-2 infection

Zhengqi Jiang ∙ Tichao Shan∙ Yucan Li ∙ Heyu Ni ∙ et. al., DOI: 10.1016/j.ijid.2025.108287 External Link Also available on ScienceDirect External Link

Highlights

SARS-CoV-2 causes lasting immune dysregulation for over 20 months.

The impact of SARS-CoV-2 on lymphocytes was especially severe in patients with CVD.

Lymphocyte deficiency is related to long COVID pathogenesis.

Long-term immune dysregulation of long COVID demands tailored treatment.

Abstract

Objectives

Growing evidence suggests that lymphocyte subsets are declined in COVID-19 patients, but it is unclear if these alterations persist after widespread exposure to SARS-CoV-2 or how long they last.

Methods

We analyzed lymphocyte subset data from 40,537 patients across three phases: pre-COVID, mass infection, and post-COVID. The counts of lymphocyte subsets and CD4⁺/CD8⁺ ratios were compared using Mann–Whitney U test or Kruskal-Wallis H test. Monthly post-exposure data were compared with pre-exposure data to assess the persistence of impact on lymphocyte subsets by SARS-CoV-2, and subgroup analyses were performed in patients with cardiovascular disease.

Results

During mass infection, T cells, CD4⁺T cells, CD8⁺T cells, NK cells, and B cells dropped significantly. Even 20 months post-infection, CD8⁺ T cells remained 9.9% below baseline. Baseline lymphocyte subsets differed significantly by sex and age. Immune recovery varied by age and sex, with older adults and males showing prolonged lymphopenia. In cardiovascular disease patients, T lymphocytes remained 72.9% below baseline for 20 months post-infection.

Conclusion

Our findings redefine SARS-CoV-2 infection as a condition of long-lasting immune compromise. The sustained subnormal lymphocytes—particularly in cardiovascular disease cohorts—highlight a key immunologic feature of long COVID and underscore the need for personalized care.

Graphical abstract

Introduction

As of September 15, 2024, the COVID-19 pandemic had resulted in more than 776 million confirmed cases and seven million deaths worldwide [1]. While most patients recovered from the acute phase of COVID-19, a subset experiences a range of persistent symptoms collectively termed “long COVID” or “postacute COVID-19 syndrome (PACS)”. Long COVID, affecting an estimated 10-60% of survivors, [2] is a multisystem disorder with a wide array of clinical manifestations, including fatigue, myalgia, dyspnea, paresthesia, chest pain, or a sensation of a lump in the throat [3].

The pathogenesis of long COVID remains poorly elucidated. Hypotheses include viral persistence of SARS-CoV-2, triggering chronic inflammation and immune dysregulation [4–9]. SARS-CoV-2 causes immune system dysregulation, [3,5,10] which may further lead to the reactivation of latent pathogens, such as Epstein-Barr virus (EBV) and human herpesvirus 6 (HHV-6) [3,5,11–12]. Moreover, dysbiosis of the host microbiome and viral ecosystems, particularly gut microbiota, has also been implicated as a potential contributor to long COVID pathophysiology [3,13–15]. Additionally, SARS-CoV-2 may provoke sustained immune activation and autoantibody production through molecular mimicry mechanisms [3,16–18]. The infection may further induce microvascular thrombosis and endothelial dysfunction, which are known to play a role in several postviral syndromes [3,19–21]. Moreover, multiple studies have shown that patients with a history of cardiovascular diseases, cancer, diabetes, or pulmonary diseases experience worse clinical outcomes and have an elevated risk of developing post-COVID conditions [22–25].

SARS-CoV-2 infection exerts profound and persistent effects on the adaptive immune system, a cornerstone of pathogen clearance and immune homeostasis [26]. Emerging evidence implicates dysregulation of adaptive immunity—particularly in CD4⁺ T cells, CD8⁺ T cells, and B cells—as a key mechanism underlying long COVID [3,5,10]. These lymphocyte subsets, which coordinate antigen-specific defenses, exhibit prolonged alterations post-infection, potentially driving systemic inflammation and viral persistence [27]. Consequently, longitudinal monitoring of lymphocyte subsets provides critical insights for understanding immune dysregulation in long COVID and its clinical sequelae.

Several studies demonstrate that acute COVID-19 is characterized by marked reductions of CD4⁺ and CD8⁺ T lymphocytes, a signature strongly associated with disease severity and prognosis [26,28–29]. Recently, studies reveal that CD4⁺ T cell counts remain below baseline even ten months post-infection [9–10,30–31]. Preliminary studies also report persistent lymphocyte subset alterations—including skewed CD4⁺/CD8⁺ T-cell ratios and B cell hyperactivation with elevated autoreactive antibodies—in patients with prolonged viral positivity, [3,32–33] yet robust epidemiological data linking these changes to long COVID remain scarce. Crucially, the tripartite relationship between viral persistence, chronic lymphocyte dysregulation, and multisystem manifestations of long COVID remains undefined, hindering mechanistic understanding and therapeutic development.

To address this gap, we conducted a multicenter, cross-sectional study with an ecological design to investigate the long-term effects of SARS-CoV-2 infection on lymphocyte subsets. Using data from 40,537 patients, we analyzed immune alterations in a large cohort of patients with SARS-CoV-2 exposure. This study aims to characterize the long-term dynamics of lymphocyte subsets following SARS-CoV-2 infection, and identify immunological signatures to better understand the disease’s mechanisms and provide insights for potential therapeutic interventions.

Methods

Study design

This multicenter, cross-sectional study, employing an ecological comparison design, was conducted at Qilu Hospital of Shandong University, the Second Hospital of Shandong University, and Shandong Provincial Hospital Affiliated to Shandong First Medical University in Jinan, China.

All relevant clinical and laboratory data were extracted from the Electronic Medical Record (EMR), including demographic information, initial diagnoses, and lymphocyte subset counts. This study was approved by the Medical Ethical Committee of Qilu Hospital, Shandong University. Informed consent was waived as the research involved the analysis of anonymized retrospective data. The research conformed to the Declaration of Helsinki.

Study population

We included all adult patients who underwent lymphocyte subset testing at these three hospitals from January 2021 to August 2024. Exclusion criteria were as follows: incomplete data (primarily pertaining to test results documented as “not performed” or those with missing results for one or more essential lymphocyte subsets), patients with known immune status influencers (ICU patients, hematological disease, pregnancy, infertility, documented recurrent pregnancy loss, cancer, and a history of organ transplant), and emergency outpatients (excluded due to short consultation times that may compromise the reliability of lymphocyte subset test results).

Procedure

To delineate the phases of exposure, patients’ data were divided into three groups: pre-COVID (January 2021 to November 2022), mass infection (December 2022 to February 2023), and post-COVID (March 2023 to August 2024). The SARS-CoV-2 Omicron wave, which began in December 2022, served as the key exposure event for this study. Under China’s long-term dynamic zero-COVID-19 strategy, prior to December 2022, COVID-19 in Shandong Province and Jinan remained limited, with only a small number of cases identified. In these instances, patients were isolated for treatment and recovery, preventing widespread transmission. From January 2020 to November 2022, the region maintained strict anti-COVID-19 measures, including widespread mask usage among residents. It is critical to note that these groups represent independent cohorts. The lymphocyte subset profiles from the pre-COVID group were used as the baseline reference to evaluate deviations observed during the mass infection and post-COVID groups. During the SARS-CoV-2 Omicron wave from December 2022 to February 2023, the positive rate of COVID-19 infections peaked successively across the country and then continuously declined. During this period, the majority of the population experienced the acute phase of COVID-19 infection. Therefore, patients of this period were defined as the mass infection group to assess the impact of the acute phase of COVID-19 infection. By February, the overall epidemic in the country had decreased to a relatively low level. The subsequent period had not experienced large-scale social outbreaks and was defined as the post-COVID group to assess the long-term effects of SARS-CoV-2 on the population.

To confirm the validity of this baseline, we compared lymphocyte subset measurements between 2021 and the first 11 months of 2022 and observed no significant differences across these pre-COVID time periods (Table S1).

Lymphocyte subset count

Lymphocyte subset count was analyzed using flow cytometry (FCM) (FACSCanto flow cytometers, BD, Franklin Lakes, USA), employing a single-platform method to determine absolute counts. The analysis included measurements of both relative counts (percentages) and absolute counts for the following lymphocyte subpopulations of CD3⁺ T cells (CD45⁺CD3⁺), CD8⁺ T cells (CD3⁺CD8⁺), CD4⁺ T cells (CD3⁺CD4⁺), NK cells (CD3^–CD16⁺CD56⁺), and B cells (CD3^–CD19⁺). The fluorescent monoclonal antibodies used (all from BD Biosciences Pharmingen, San Diego, CA, USA) were as follows: CD45-PerCP-Cy5.5 (clone 2D1), CD3-FITC (SK7), CD4-PC7 (SK3), CD8-APC-Cy7 (SK1), CD16-PE (B73.1), CD56-PE (NCAM16.2), and CD19-APC (SJ25C1). Standard side scatter (SSC) and singlet gating strategies were applied as part of the gating scheme (Figure S1). Cells were analyzed shortly after peripheral blood collection. Whole blood samples (50 μl) were stained with the relevant antibodies (20 μl) following erythrocyte lysis using BD FACS Lysing Solution. The stained tubes were then incubated at room temperature in the dark for approximately 15 minutes. This process was completed in a standardized laboratory that complied with ISO15189 standards, where the workflow was executed according to the standard operating procedures (SOP) and unified reference intervals, which were consistent across all participating centers, as verified through a consistency evaluation of SOPs.

Data analysis

The reference ranges for lymphocyte subsets were based on the standardized values established for Shandong Province, China. Values below or above these ranges were defined as abnormal, and the abnormality rate was calculated as the proportion of patients with measurements outside the reference interval. Categorical variables were presented as frequencies and percentages, while continuous variables were expressed as median (interquartile range, IQR). Chi-squared (χ²) tests were used for categorical variables, and Mann–Whitney U test or analysis of Kruskal-Wallis H test were used for continuous data. To address multiple testing, false discovery rate (FDR) control (Benjamini-Hochberg procedure) was applied, and FDR-adjusted P-values were reported. To further validate the robustness of key findings given the imbalanced sample sizes across comparison groups, bootstrap resampling was performed to assess the stability of the observed differences. All statistical analyses were performed using R software (Version 4.2.2), with a P-value <0.05 considered statistically significant.

Results

Demographic characteristics of all patients

A total of 67,621 patients underwent lymphocyte subset testing. Of these, 27,084 patients were excluded for the following reasons: incomplete data (n=384), emergency outpatients (n=96), patients with known immune status influencers (n=26,604), like ICU patients, hematological disease, pregnancy, infertility, documented recurrent pregnancy loss, cancer, and a history of organ transplant. Ultimately, 40,537 patients were analyzed (Figure 1), with sample sizes for each group as follows: pre-COVID (n=15,874), mass infection (n=2844), and post-COVID (n=21,819). The demographic characteristics of the 40,537 patients were presented in Table S2. The median age was 54 years (38-65), with 52.2% of patients being male, and the highest proportion of patients belonging to the nephrology department, accounting for 24.2%.

Figure 1 dummy alt text — **Figure 1** Study flow diagram. Study flow diagram with inclusion and exclusion criteria for the patients.

Baseline characteristics

Baseline lymphocyte subsets were analyzed by age and sex. Age-based stratification (18-40, 41-60, 61-80, >80 years) revealed significant variability in most lymphocyte subsets. The percentage of NK cells and CD4⁺/CD8⁺ T-cell ratios increased with age (P<0.0001), while the count of B cells, the count of Th cells, percentage and count of CD8⁺ T cells, and percentage and count of total T cells declined with age (P<0.0001) (Figure S2). These trends highlight the influence of age on lymphocyte subset distributions (Table S3).

Sex-based analysis showed no significant differences in the percentage of Th cells (P=0.85), the absolute count of CD8⁺ T cells (P=0.82) or total T cells (P=0.26), between males and females. However, males exhibited higher CD4⁺/CD8⁺ T-cell ratios (P=0.0018), absolute counts of Th cells (P=0.035), and NK cell percentages and counts (P<0.0001), while females had higher B cell percentages and counts (P<0.0001). Females also showed higher percentages of CD8⁺ T cells and total T cells (P<0.0001), indicating distinct sex-related differences in lymphocyte subsets (Figure S3, Table S3).

These findings highlight the baseline variability of lymphocyte subsets by age and sex, emphasizing their importance in assessing immune responses to SARS-CoV-2 and long COVID, and providing a critical reference for studying lymphocyte dynamics.

Characteristics of lymphocyte subsets of mass infection group

To investigate the characteristics of lymphocyte subsets during the mass infection period, we compared lymphocyte subset levels from the mass infection group to those from the pre-COVID group, which served as the baseline. Results showed significant increases in the percentages of B cells (10.23% vs 11.39%, P<0.0001), CD4⁺/CD8⁺ T cells ratio (1.43 vs 1.50, P=0.0026), and NK cell percentages (13.33% vs 14.31%, P<0.0001) during the mass infection period. Meanwhile, the absolute counts and percentages of major T cell subsets (Th cells, CD8⁺ T cells, and total T cells), along with the absolute counts of B cells and NK cells, all showed significant decreases (Figure 2, Table S4). Given the sample size of the mass infection group, we repeatedly drew random sub-samples from the pre-COVID group that were matched in size to it. The consistency of results across these sub-samples confirmed the robustness of the observed differences. (Table S5)

Figure 2 dummy alt text — **Figure 2** Lymphocyte subsets change in different period of SARS-CoV-2. Boxplots of various lymphocyte subsets stratified by pre-COVID, mass infection, and post-COVID groups. Central bars represent groups medians, with bottom and top bars representing 25th and 75th percentiles, respectively. The light blue area represents the reference range for the detection values of various lymphocyte subsets. Statistical significance determined by Mann–Whitney U test, and adjusted for multiple comparisons using the FDR method. P-values for pairwise comparisons between groups are indicated as follows: *: P<0.05, **: P<0.01, ***: P<0.001, ****: P<0.0001. Each dot represents an individual patient: baseline (red, n=15,874), mass infection (blue, n=2844), and post-COVID (orange, n=21,819). Detailed results are reported in Table S4.

To investigate the extent to which SARS-CoV-2 affects lymphocyte subsets, the abnormal rates of lymphocyte subsets during the mass infection period were also analyzed. The proportion of patients with lymphocyte subsets below the normal range demonstrated substantial increases from the baseline to mass infection group. This was particularly evident for the counts of B cells (36.4% vs 42.6%, P<0.0001), Th cells (23.3% vs 43.2%, P<0.0001), CD8⁺ T cells (4.2% vs 14.7%, P<0.0001), NK cells (18.4% vs 25.4%, P<0.0001), and total T cells (12.0% vs 30.9%, P<0.0001), and the percentage of Th cells (12.4% vs 17.7%, P<0.0001), CD8⁺ T cells (1.1% vs 2.7%, P<0.0001), and total T cells (2.7% vs 7.4%, P<0.0001). In contrast, the proportion of patients with some values above the normal range were significantly higher compared to pre-COVID, like the percentage of B cells (4.2% vs 7.8%, P<0.0001), NK cells (14.0% vs 18.6%, P<0.0001), and CD4⁺/CD8⁺ T cells ratio (13.7% vs 17.8%, P<0.0001) (Figure 2, Table S4). These findings indicate the substantial impact of SARS-CoV-2 exposure on lymphocyte subset distribution and functional status, reflecting alterations in adaptive immune responses during the mass infection period.

Characteristics of lymphocyte subsets in post-COVID group

To investigate lymphocyte subsets of the post-COVID group, we compared the levels with the baseline. Post-COVID, the percentage of B cells (10.23% vs 10.34%, P=0.0065), NK cells (13.33% vs 14.05%, P<0.0001), and CD4⁺/CD8⁺ T cells ratio (1.43 vs 1.45, P=0.00013) increased, while the percentage of Th cells (40.00% vs 39.41%, P<0.0001), CD8⁺ T cells (28.29% vs 27.07%, P<0.0001), T cells (73.48% vs 72.37%, P<0.0001), and counts of Th cells (672.43 vs 607.15 cells/µL, P<0.0001), CD8⁺ T cells (463.23 vs 406.24 cells/µL, P<0.0001), T cells (1227.00 vs 1112.78 cells/µL, P<0.0001), NK cells (217.82 vs 195.00 cells/µL, P<0.0001) were decreased. Despite some recovery in these levels during the post-COVID group compared to the mass infection group, they remained significantly different from the baseline (Figure 2, Table S4, Table S5). These decreases were more pronounced in males (Figure S4, Table S6) and in the age groups of 41-60 and 61-80 (Figure S5, Table S7). Abnormal rates of lymphocyte subsets showed that the proportion of patients with B cell counts (36.4% vs 43.3%, P<0.0001), the percentage of Th cells (12.4% vs 15.9%, P<0.0001), CD8⁺ T cells (1.1% vs 2.1%, P<0.0001), T cells (2.7% vs 6.0%, P<0.0001), the counts of Th cells (23.3% vs 32.8%, P<0.0001), CD8⁺ T cells (4.2% vs 10.6%, P<0.0001), T cells (12.0% vs 21.5%, P<0.0001), and NK cells (18.4% vs 22.0%, P<0.0001) below the normal range, and the proportion of patients with the percentage of B cells (4.2% vs 6.9%, P<0.0001) and NK cells (14.0% vs 17.6%, P<0.0001) above the normal range, were significantly increased post-COVID.

Comparing mass infection and post-COVID groups, no significant differences were observed in percentage of NK cells (P=0.16) and the CD4⁺T/CD8⁺T ratio (P=0.29), indicating lymphocyte subsets have partially recovered but have not reached pre-COVID levels. These findings suggest that SARS-CoV-2 exposure may have a long-term impact on the lymphocyte subsets of the population.

Persistent effect of lymphocyte subsets caused by SARS-CoV-2

To further investigate the persistence and duration of SARS-CoV-2’s effects on lymphocyte subsets, we compared monthly post-COVID lymphocyte subset data with baseline data. The results revealed significant changes in absolute lymphocyte counts, prompting a focus on these changes.

Following the Omicron BA.5.2 and BF.7 wave around December 2022, noticeable changes in the lymphocyte counts, particularly CD8⁺ T cells, Th cells, and total T cells, persisted through August 2024. Compared to baseline, CD8⁺ T cell counts showed significant decreases in all months except February 2023 (P=0.71), and this reduction persisted until August 2024 (463.23 vs 417.26 cells/µL, reduction: 9.9%, P<0.0001). Similarly, total T cell counts demonstrated consistent decreases throughout the observation period (1227.00 vs 1163.15 cells/µL, reduction: 5.2%, P<0.0001, by August 2024). Th cell counts also showed significant reductions in all months, maintaining this trend through August 2024 (672.43 vs 642.64 cells/µL, reduction: 4.4%, P<0.0001) (Figure 3, Table S8). Significant decreases in B cell counts were observed in all the months, except February (P=0.12) and March 2023 (P=0.19). NK cell counts did not significantly differ in February (P=0.96), March (P=0.45), May (P=0.053), and October 2023 (P=0.065), and April (P=0.10), May (P=0.11), and August 2024 (P=0.12), but decreased significantly in other months. This suggests a potential long-term impact of SARS-CoV-2 on the adaptive immune system, possibly contributing to long COVID.

Figure 3 dummy alt text — **Figure 3** Monthly comparison of post-COVID lymphocyte subsets’ counts to pre-COVID baseline data. (a) The trends of the counts of T cells, Th cells, and CD8⁺ T cells among monthly post-COVID data, with pre-COVID as the baseline. (b) The trends of the counts of B cells and NK cells among monthly post-COVID data, with pre-COVID as the baseline. The median values for each time period are represented by the data points. Abbreviations: CD3n, the absolute count of T cells; CD4n, the absolute count of Th cells; CD8n, the absolute count of CD8⁺T cells; NKn, the absolute of NK cells; CD19n, the absolute count of B cells; NK, natural killer. Detailed results are reported in Table S8.

Prolonged dysregulation of lymphocyte subsets in cardiovascular disease patients post-COVID

Based on prior evidence, we initially sought to analyze patient subgroups with specific comorbidities, including cancer, diabetes, cardiovascular diseases (CVD), and pulmonary diseases. However, patients with cancer were largely excluded due to their altered baseline immune status, and the sample size for patients with diabetes was too limited for meaningful analysis. We therefore focused on subgroups with pulmonary and cardiovascular diseases. Our analysis revealed that changes in lymphocyte subsets among patients with pulmonary diseases were generally consistent with the overall population, whereas patients with CVD exhibited more pronounced alterations in these subsets during the post-COVID period (Figure 4, Table S9).

Figure 4 dummy alt text — **Figure 4** Lymphocyte subsets of CVD patients in different periods of SARS-CoV-2. Boxplots of various lymphocyte subsets of cardiovascular disease patients stratified by pre-COVID, mass infection, and post-COVID groups. Central bars represent groups medians, with bottom and top bars representing 25th and 75th percentiles, respectively. The light blue area represents the reference range for the detection values of various lymphocyte subsets. Each dot represents an individual patient: baseline (red, n=174), mass infection (blue, n=195), and post-COVID (orange, n=2053). The median values for each time period are represented by the data points. Statistical significance determined by Mann–Whitney U test, and adjusted for multiple comparisons using the FDR method. P-values for pairwise comparisons between groups are indicated as follows: *: P<0.05, **: P<0.01, ***: P<0.001, ****: P<0.0001. Detailed results are reported in Table S9.

In the CVD subgroup, we observed a significant increase in the percentage of B cells (10.16% vs 19.03%, P=0.0065) and NK cells (14.73% vs 19.28%, P<0.0001). In contrast, the percentage of T cells (71.82% vs 57.44%, P<0.0001), including Th cells (41.97% vs 31.81%, P<0.0001) and CD8⁺ T cells (24.68% vs 20.66%, P=0.00072), was significantly decreased. This pattern was further reflected in absolute counts: marked reductions were seen in T cells (1087.14 vs 321.00 cells/µL, P<0.0001), Th cells (613.22 vs 177.00 cells/µL, P<0.0001), CD8⁺ T cells (360.31 vs 115.00 cells/µL, P<0.0001), NK cells (214.01 vs 109.00 cells/µL, P<0.0001), and B cells (144.78 vs 109.00 cells/µL, P=0.014). The CD4⁺/CD8⁺ T cell ratio also declined significantly (1.74 vs 1.56, P=0.029).

Because the sample size of pre-COVID group was small, we repeatedly drew random sub-samples of the post-COVID group that were matched in size to the pre-COVID group to test the robustness of our findings. Across 1000 resamplings, the counts of T cells, Th cells, CD8⁺ T cells, and NK cells, the percentage of T cells, Th cells, and B cells all remained significantly different (Table S10). The percentage of CD8⁺ T cells and NK cells remained consistently different across the vast majority of samples.

To determine the duration of these alterations, we serially compared absolute lymphocyte subset counts in post-COVID CVD patients against their own baseline values on a monthly basis. Due to limited testing volume from March to July 2023, data from these months were pooled for analysis. During this period (March-July 2023), the counts of T cells (P=0.51), Th cells (P=0.22), CD8⁺ T cells (P=0.91), and NK cells (P=0.54) showed no significant differences from baseline levels, suggesting recovery to pre-COVID states. However, in August 2023, dramatic reductions emerged in total T cells, Th cells, CD8⁺ T cells, and NK cells (P<0.0001)—plummeting to approximately one-third of baseline values. This suppressed state persisted through August 2024 (Figure 5, Table S11), with all four lymphocyte subsets maintaining significantly lower levels compared to the pre-COVID baseline: total T cells (1087.14 vs 294.00 cells/µl, reduction: 72.9%, P<0.0001), Th cells (613.22 vs 159.00 cells/µl, reduction: 74.1%, P<0.0001), CD8⁺ T cells (360.31 vs 113.00 cells/µl, reduction: 68.6%, P<0.0001), and NK cells (214.01 vs 115.00 cells/µl, reduction: 46.3%, P<0.0001). This biphasic pattern (recovery followed by abrupt depletion) raises the possibility that lymphocyte exhaustion underlies cell-mediated immune dysregulation in long COVID pathogenesis. And compared to baseline, the counts of B cells exhibited distinct pattern: significant decreases were only observed in August (P=0.014), November (P=0.0036), December 2023 (P=0.0030), and January (P=0.00021), February (P=0.040), March (P=0.011), April (P=0.017), June (P=0.039), and August 2024 (P=0.0023).

Figure 5 dummy alt text — **Figure 5** Monthly comparison of post-COVID lymphocyte subsets’ counts to pre-COVID baseline in CVD patients. The trends of the counts of T cells, Th cells, CD8⁺ T cells, the counts of B cells and NK cells among monthly post-COVID data, with pre-COVID as the baseline. The median values for each time period are represented by the data points. Detailed results are reported in Table S11. Abbreviations: CD3n, the absolute count of T cells; CD4n, the absolute count of Th cells; CD8n, the absolute count of CD8⁺T cells; NKn, the absolute of NK cells; CD19n, the absolute count of B cells; NK, natural killer.

Discussion

This study provides robust evidence that widespread exposure to SARS-CoV-2, particularly during the Omicron BA.5.2 and BF.7 wave, has a lasting impact on lymphocyte subsets. Our findings demonstrate that key immune populations, including CD4⁺ T cells, CD8⁺ T cells, NK cells, and total T cells, exhibit persistently unrecovered for up to 20 months post-exposure, underscoring the potential role of immune dysregulation in the development and persistence of long COVID symptoms. Regarding the phenomenon of long-term lymphopenia, previous studies demonstrating robust activation of CD8⁺ T cells alongside unaltered TREC (T cell Receptor excision circle) indicate that activation-induced apoptosis plays an important role [31].

Compared to baseline levels, significant decreases were observed in the absolute counts of CD4⁺ T cells, CD8⁺ T cells, NK cells, and total T cells, with sustained alterations in the CD4⁺/CD8⁺ T-cell ratio. Meanwhile, the proportion below the normal range was dramatically increased, particularly CD4⁺ and CD8⁺ T cells. Persistent reductions in CD8⁺ T cells, CD4⁺ T cells, and total T cells suggest chronic immune dysfunction, potentially predisposing individuals to immune exhaustion, reactivation of latent infections, and autoimmunity—all mechanisms linked to long COVID pathogenesis. Notably, increase in the percentage of a subset (e.g., B cells) despite a decrease in its absolute count, reflecting a severe reduction in the total lymphocyte and a disproportionately greater loss of T cells, which consequently reshaped the compositional landscape of circulating lymphocytes.

Previous studies have reported T-cell exhaustion and a decline in effector memory CD4⁺ and CD8⁺ T cells lasting at least 13 months after mild COVID-19 [3,32,34]. Our findings extend this observation, demonstrating that even a short, widespread SARS-CoV-2 outbreak can lead to prolonged lymphocyte dysregulation, which is potentially driven by impaired T-cell maturation, increased apoptosis, or prolonged immune activation. This is consistent with reports that SARS-CoV-2 persistence in tissues contributes to long-term immune remodeling, similar to what is seen in chronic viral infections [9,35]. Such alterations may impair the immune system’s ability to mount an effective response against latent infections, thereby increasing susceptibility to opportunistic pathogens. The observed decrease in NK cells further supports this hypothesis, which aligns with studies suggesting a reactivation of latent pathogens, such as Epstein-Barr virus (EBV) and human herpesvirus 6 (HHV-6), in patients with long COVID [3,5,11–12].

Changes in the CD4⁺/CD8⁺ T-cell ratios provide further insight. While a reduced CD4⁺/CD8⁺ T-cell ratio has been described in chronic viral infections such as HIV, [36] our study found that the ratio was increased during the mass infection phase before declining in the post-COVID phase. This suggests that CD8⁺ T cells depletion occurs early, likely due to viral-induced apoptosis or T-cell exhaustion, followed by a later decline of CD4⁺ T cells may also be affected, possibly driven by chronic immune activation or reduced thymic output. These findings highlight the differential effects of SARS-CoV-2 on distinct T-cell subsets across different phases, reflecting a dynamic immune response.

The high infectivity and variability of SARS-CoV-2 have made subsequent infections inevitable, with accumulating evidence suggesting that repeat exposure increases the severity and incidence of long COVID [2]. Additionally, studies have shown that anti-RBD antibody levels significantly decline in 80% of patients 10 months after initial infection, indicating a significantly higher risk of reinfection [31]. National surveillance data revealed surges in COVID-19 positive rates were observed in May 2023, which coincided with periods of heightened viral transmission, were entirely attributable to omicron variants, predominantly the XBB series (including XBB.1.9, XBB.1.16, and their sublineages). These additional exposures could further amplify immune exhaustion, chronic inflammation, and impaired lymphocyte recovery (Figure S6). Our findings support immune dysregulation is a central driver of long COVID pathogenesis, likely mediated by interconnected mechanisms. For example, immunosuppression during acute infection may permit SARS-CoV-2 persistence, while endothelial dysfunction could precipitate microvascular thrombosis, and SARS-CoV-2 spike protein receptor-binding domain (RBD) -induced platelet activation/clearance via αIIbβ3 integrin binding—processes that may act synergistically or sequentially to sustain symptoms [37–40]. Prolonged CD4⁺ and CD8⁺ T-cell depletion resembles immune patterns seen in post-viral fatigue syndromes (e.g., myalgic encephalomyelitis), which share hallmark symptoms such as fatigue and cognitive impairment. These prolonged lymphocyte subset perturbations, now robustly documented in our cohort, suggest that chronic immune depletion or exhaustion may directly contribute to the persistence of long COVID symptoms.

Current studies indicate that SARS-CoV, which is closely related to SARS-CoV-2, appears not to exhibit such long-term effects. Although lymphocyte subsets in SARS patients showed significant reduction within the first two weeks post-infection, they recovered to normal levels after five weeks [41].

Age and sex differences further shaped immune recovery. Older adults showed more pronounced T cell declines, consistent with “immunosenescence” and the higher burden of long COVID symptoms reported in this group [42–43]. Men exhibited deeper and more persistent lymphopenia than women, supporting previous evidence that males exhibit stronger inflammatory responses, delayed viral clearance, and more sensitive to immune exhaustion [42–43]. Sex differences in immune responses have been attributed to hormonal influences and X-linked immune-related genes, as studies have shown [44].

Moreover, the impact was especially severe in patients with CVD—a finding with critical clinical implications. In this subgroup, transient recovery was followed by a sharp and sustained decline in T-cell subsets, remaining markedly reduced for over a year. We hypothesize that this pronounced and persistent T-cell depletion may reflect a state of immune exhaustion, which can fuel chronic inflammation and impair immunosurveillance of latent viruses, thereby promoting vascular inflammation. Concurrently, the loss of effective T-cell-mediated control may destabilize atherosclerotic plaques and increase the risk of thromboembolic and other acute cardiovascular events. These observations provide a mechanistic basis for epidemiological evidence of long-term cardiovascular morbidity after COVID-19 [22–23].

This study highlights the urgent need for long-term immune monitoring after SARS-CoV-2 exposure. Persistent T-cell reductions, particularly in CD4⁺ and CD8⁺ T cells, raise concerns about chronic immune dysfunction even in nonhospitalized individuals. Routine lymphocyte subset analysis, particularly in high-risk groups such as older adults and CVD patients, may help in risk stratification and guide personalized therapeutic interventions.

This study provides a unique perspective to evaluate SARS-CoV-2’s immune impact, as local pandemic policies ensured that COVID-19 infections were absent until late 2022. The short, well-defined infection window (December 2022 to January 2023) and rapid herd immunity created a near-ideal model for studying viral-induced immune alterations without prior infections. The large sample size, multicenter design, and longitudinal approach further enhance the robustness of these findings.

This study’s ecological design allowed us to assess macro-level trends in lymphocyte subsets, several limitations should not be ignored. First, the use of cross-sectional data from different individuals prevents tracking immune changes over time and limits individual-level causal inference. Second, the “mass infection period” was defined based on national policy and epidemiologic trends rather than laboratory-confirmed infections, and the cohort—predominantly Han Chinese from Eastern China—may limit generalizability. We were also unable to determine reinfection rates or viral variants, and comorbidities may not have been fully matched across groups. Although stratified by age and sex, residual confounding remains possible. Finally, because this study focused on population-level immune trends, we did not include ICU patients or detailed symptom data, and thus did not aim to assess the relationship between disease severity and immune changes. Future studies integrating immune and clinical data across diverse, repeatedly exposed populations are needed to clarify how viral persistence shapes immune remodeling and long-term outcomes.

In summary, this large multicenter study provides compelling evidence that SARS-CoV-2 exposure leads to long-term alterations in lymphocyte subsets, particular CD4⁺ T cells, CD8⁺ T cells, NK cells and total T cells. The persistence of these changes for up to 20 months highlights the need for monitoring of lymphocyte subsets in patients with long COVID. Addressing lymphocyte dysregulation may provide a novel approach for mitigating the long-term consequences of SARS-CoV-2 infection.

Ethical approval

The study was approved by the Medical Ethics Committee of Qilu Hospital, Shandong University (KYLL-202411-028). This study was registered with chictr.org.cn (ChiCTR2500098013). The informed consent was waived because this study didn’t involve individual patients.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Funding

This study was supported by National Key R&D Program of China (grant number 2023YFC2507800); National Natural Science Foundation of China (grant number 81900124); Young Taishan Scholar Foundation of Shandong Province (grant number tsqn201812133); Norman Bethune Boundless Medicine Hematological Oncology Research Fund (grant number HSKY-015), and Canadian Institutes of Health Research Foundation (grant number 389035).

Author contributions

MX, JP, TCS, ZQJ, and YCL conceived and designed the study. ZQJ and TCS drafted the paper. JP, MX, HYN, JM, ZQJ, TCS, and YCL revised the manuscript. ZQJ and YCL did the analysis. ZQJ, TCS, YCL, FJH, BBF, XHZ, and JM collected and verified the data. JP, MX, HYN, JM supervised the study and provided funding support. All authors had full access to the data in the study, critically revised the manuscript for important intellectual content, and had final responsibility for the decision to submit for publication. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We are grateful to the Qilu Hospital of Shandong University, for providing administrative and technical support.

Appendix Supplementary materials (1)

Document (1.68 MB)

References

WHO COVID-19 dashboard. [Available from: https://data.who.int/dashboards/covid19/cases?n=c; 2024. Accessed September 15, 2024.

Highlights

Abstract

Objectives

Methods

Results

Conclusion

Graphical abstract

Introduction

Methods

Study design

Study population

Procedure

Lymphocyte subset count

Data analysis

Results

Demographic characteristics of all patients

Baseline characteristics

Characteristics of lymphocyte subsets of mass infection group

Characteristics of lymphocyte subsets in post-COVID group

Persistent effect of lymphocyte subsets caused by SARS-CoV-2

Prolonged dysregulation of lymphocyte subsets in cardiovascular disease patients post-COVID

Discussion

Ethical approval

Data availability

Funding

Author contributions

Declaration of competing interest

Acknowledgments

Appendix Supplementary materials (1)

References

Leave a Reply Cancel reply