Dynamic immune response and its influencing factors in COVID-19 patients with non-alcoholic fatty liver disease: A cohort study

Abstract

BACKGROUND

Dynamic alterations in lymphocyte subsets demonstrate significant correlations with clinical disease severity in patients with coronavirus disease 2019 (COVID-19). As the most prevalent chronic liver disease globally, non-alcoholic fatty liver disease (NAFLD) exhibits distinct chronic inflammatory and immunometabolic disturbances that may substantially affect immune response patterns in COVID-19 patients. Nevertheless, the characteristics of lymphocyte subset dynamics and their clinical implications in COVID-19-NAFLD remain to be fully elucidated.

AIM

To characterize the dynamic changes in lymphocyte subsets among COVID-19 patients with NAFLD, in order to delineate their immunological profiles and inform clinical management strategies.

METHODS

The cohort study compared lymphocyte subpopulations in 858 COVID-19 patients and 670 COVID-19-NAFLD patients at admission, discharge, and 2-week/4-week post-discharge follow-ups.

RESULTS

Compared to COVID-19 patients without NAFLD, NAFLD-comorbid patients demonstrated persistently elevated CD3+CD4+ counts as well as lymphocyte counts and percentages at admission and at the 2-week and 4-week follow-ups post-discharge (all P < 0.05). Among COVID-19-NAFLD patients, those aged ≥ 60 years had significantly lower CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, and CD19+ counts and percentages at all assessed time points (all P < 0.05); significant liver fibrosis correlated with reduced CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts and percentages across all time points (all P < 0.05); multimorbidity (≥ 3 comorbidities) exacerbated immune imbalance, marked by elevated CD3+CD4+ percentages and CD56+ counts at admission, increased CD3+CD4+ counts, lymphocyte counts, and CD19+ counts and percentages at discharge, as well as sustained increases in CD3+CD4+ counts at the 2-week follow-up and higher CD3+CD4+ percentages at the 4-week post-discharge follow-up (all P < 0.05); and obesity and elevated liver enzymes were independently linked to higher CD3+CD4+ counts, CD19+ counts, and lymphocyte counts at all post-admission evaluations (from discharge through the 4-week follow-up) (all P < 0.05).

CONCLUSION

Age, liver fibrosis, comorbidities, obesity, liver enzyme abnormalities, vaccination status, low-density lipoprotein cholesterol, and hemoglobin A1c significantly modulate immune responses in COVID-19-NAFLD patients, warranting targeted clinical attention. Furthermore, patients with uncomplicated NAFLD (including lean NAFLD) also require particular clinical attention to mitigate risks of immune imbalance.

Key Words: COVID-19; Non-alcoholic fatty liver disease; Lymphocyte subsets; Dynamic immune response; Risk factors

Core Tip: This study reveals that coronavirus disease 2019 patients with non-alcoholic fatty liver disease (NAFLD) exhibit prolonged immune imbalance. Key risk factors—including age ≥ 60 years, significant liver fibrosis, multimorbidity (≥ 3 conditions), obesity, elevated liver enzymes, elevated low-density lipoprotein cholesterol, and elevated hemoglobin A1c—worsen immune imbalance. Notably, even lean NAFLD patients show persistent immune disturbances. These findings highlight the need for tailored clinical monitoring in coronavirus disease 2019-NAFLD patients, particularly for older adults, those with metabolic dysfunction, or those with advanced liver disease, to mitigate long-term immunological complications.

INTRODUCTION

The global pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has presented a substantial threat to public health. As of April 2023, there had been over 750 million confirmed cases of coronavirus disease 2019 (COVID-19) and more than 6.8 million reported deaths[1]. Among severe cases, the mortality rate ranged from 21% to 30%[2].

Coronaviruses and hepatic injury exhibit reciprocal pathophysiological interactions. A reciprocal mechanism involves SARS-CoV-2-triggered ectopic lipid accumulation in hepatocytes. COVID-19 induces significant dyslipidemia[3], and postmortem studies revealed a high prevalence of hepatic steatosis in affected patients[4,5]. These findings suggest that preexisting steatosis likely exacerbates COVID-19-related hepatic injury. Conversely, SARS-CoV-2 infection appears to potentiate hepatic fat deposition, creating a pathologic feedback loop[6]. Multiple studies demonstrated that obesity, which is strongly associated with non-alcoholic fatty liver disease (NAFLD), significantly correlates with more severe COVID-19 progression and increased mortality[7,8]. A clear association exists between excess adiposity and susceptibility to both autoimmune disorders and infections, supporting the hypothesis of impaired and altered immune responses in overweight populations. Viral respiratory infections are also consistently more prevalent in obese individuals compared to normal-weight cohorts[9,10].

Another critical mechanism involves SARS-CoV-2-induced dysregulation of both innate and adaptive immune responses[11]. In severe COVID-19, this immune dysregulation manifests as systemic inflammatory response syndrome, characterized by a fulminant cytokine storm driven by uncontrolled activation of T-cell-mediated pathways. This immunopathological cascade precipitates multi-organ injury, with notable hepatic involvement[12]. Mechanistically, this cytokine dyshomeostasis establishes a critical pathophysiological connection between COVID-19 and metabolic dysfunction, particularly in NAFLD/metabolic dysfunction-associated fatty liver disease[13]. The cytokine storm precipitates systemic hypotension and profound tissue hypoxia. In cases complicated by acute respiratory distress syndrome, this cascade significantly aggravates impaired pulmonary oxygenation, further compromising hepatic perfusion and exacerbating pre-existing circulatory insufficiency[14,15].

The resultant shock state and hypoxia trigger a marked increase in reactive oxygen species generation, which promotes extensive oxidative damage to cellular lipids, deoxyribonucleic acid, and proteins. Critically, these oxidative metabolites directly activate hepatic stellate cells, driving extracellular matrix deposition, while simultaneously stimulating resident liver immune cells to amplify pro-inflammatory cytokine production—collectively resulting in further hepatocellular injury[14,15].

In NAFLD/non-alcoholic steatohepatitis (NASH), inflamed hepatocytes exhibit mitochondrial dysfunction[16,17], which promotes upregulation of angiotensin-converting enzyme 2 receptors and consequently exacerbates viral infectivity[18]. Concurrently, COVID-19 also directly impairs mitochondrial function, and SARS-CoV-2 binding to angiotensin-converting enzyme 2 receptors enables viral replication in hepatocytes, leading to direct liver injury[19]. Additionally, NASH is characterized by chronic low-grade inflammation[20]. NAFLD-associated impairment of hepatic innate immunity potentiates COVID-19 pathogenesis through maladaptive immune responses[21]. These mutually amplifying effects contribute to more severe clinical progression in both diseases[6]. Patients with severe COVID-19 demonstrate elevated plasma levels of inflammatory markers, including ferritin, lactate dehydrogenase, D-dimer, C-reactive protein, interleukin (IL)-2, IL-10, and most notably IL-6[22]. Mechanistically, IL-6 plays a pivotal role in initiating and propagating the cytokine storm during COVID-19[12]. Notably, chronic elevation of IL-6 has been well-established in NAFLD patients[23], suggesting a potential positive feedback loop whereby COVID-19-mediated inflammation may exacerbate NAFLD progression through persistent activation of pro-inflammatory pathways[24]. A large-scale population-based cohort study (n = 6814; cardiovascular disease-free at baseline) revealed that NAFLD patients exhibit significantly elevated circulating IL-6 levels. Notably, IL-6 emerged as the only independent biomarker for predicting subclinical cardiovascular disease (quantified by non-contrast cardiac computed tomography) in this population[23].

Postmortem hepatic biopsies from COVID-19 fatalities demonstrated marked T-cell hyperactivation and microvesicular steatosis, which strongly suggests an immune-mediated mechanism of liver injury[13]. Current research demonstrated that SARS-CoV-2 infection triggers robust adaptive immune responses in humans, including humoral immunity and antigen-specific CD4+ and CD8+ T cell responses. Peripheral lymphopenia has emerged as a consistent hallmark of SARS-CoV-2 infection and disease severity, with COVID-19 patients exhibiting significantly lower total blood lymphocyte counts—especially T cell subsets—compared to healthy controls[25].

Notably, symptomatic individuals showed more profound lymphopenia than asymptomatic carriers, while patients with radiologically confirmed pneumonia demonstrated greater lymphocyte depletion than those without pulmonary involvement[26-28]. In severe cases, both CD3+CD4+ and CD3+CD8+ counts were markedly further reduced relative to moderate cases[29-31]. In longitudinal studies, SARS-CoV-2-specific immune memory has been detected in both CD3+CD4+ counts and B cells for up to 5 months post-recovery (median approximately 3 months) among mild COVID-19 cases[32]. Importantly, lymphopenia was identified as a robust independent predictor of mortality in SARS-CoV-2 infection[33,34].

However, immunological studies investigating the interplay between COVID-19 and NAFLD remain limited. Liu et al[35] demonstrated that specific comorbidities—including cancer, chronic hepatitis B, NAFLD, and diabetes—significantly modulate immune responses in COVID-19 patients. Notably, a significant positive correlation was observed between NAFLD severity and peripheral CD3+ counts.

To elucidate the dynamic immunological characteristics and associated modulatory factors in COVID-19 patients with NAFLD, this cohort study analyzed lymphocyte subpopulations (both absolute counts and relative percentages) at multiple time points: Hospital admission, discharge, and 2-week and 4-week post-discharge follow-ups. Furthermore, we systematically evaluated the impact of these lymphocyte subsets on both the time to negative SARS-CoV-2 nucleic acid test result and clinical endpoints (clinical disease severity of COVID-19).

MATERIALS AND METHODS

Study population

This cohort study included 1528 patients admitted to the first and second hospital isolation wards of the Public Health Clinical Center of Chengdu from January 16, 2020 to September 30, 2021. The study received approval from the Ethics Committee of Public Health Clinical Center of Chengdu (No. PJ-K2020-26-01). Due to the urgent nature of evaluating an emerging infectious disease, the Ethics Commission waived the requirement for written informed consent.

Inclusion and exclusion criteria

Patients were included if they met the following inclusion criteria: (1) Age ≥ 18 years; (2) Hospitalization with COVID-19; (3) An initial positive test result for SARS-CoV-2; and (4) Evidence of liver steatosis based on abdominal ultrasound within the past 6 months, a prior clinical diagnosis of NAFLD, or a documented history of NAFLD[36]. The criteria for exclusion included: (1) Age < 18 years; (2) No COVID-19; (3) Liver conditions other than NAFLD[37]; (4) Liver decompensation; or (5) Immunosuppressive status (such as human immunodeficiency virus infection, cancer, organ transplantation, bone marrow transplantation, or use of immunosuppressive agents within the last 3 months)[37-39].

Groups

The study comprised 1528 patients. Of them, 670 patients (43.85%) with NAFLD were categorized into the COVID-19 with NAFLD group, 858 COVID-19 patients (56.15%) without NAFLD were classified into the COVID-19 group.

Liver fibrosis score

The fibrosis-4 scores were calculated using the following published formulas[40]:

The METAVIR liver histopathological scoring system was used as a reference: (1) Non-significant fibrosis (< F2): No fibrosis or portal tract expansion with the presence of cells with a stellate morphology, without fibrous septum formation; (2) Significant fibrosis (≥ F2): Portal area enlargement accompanied by either scattered or abundant fibrous septa; and (3) Cirrhosis (≥ F4)[41-43]. For COVID-19 patients with NAFLD, the optimal fibrosis-4 cutoff values for predicting significant fibrosis and cirrhosis were 1.3 and 2.67, respectively[44].

Diagnostic, clinical classification, and cure criteria

The diagnosis, clinical classification, and treatment criteria for COVID-19 were based on the seventh edition of the novel coronavirus pneumonia diagnosis and treatment guidelines[45]. NAFLD was diagnosed according to the 2018 updated guidelines for the Prevention and Treatment of NAFLD, defined as macrovesicular (or predominantly macrovesicular) steatosis involving > 5% of hepatocytes, which may be accompanied by mild nonspecific inflammation[46].

Data collection

Longitudinal clinical data were systematically collected and recorded at four predefined time points: Admission, discharge, and 2 weeks and 4 weeks post-discharge, including gender, age, comorbidities (including hypertension, hyperlipidemia, diabetes, and hyperuricemia), and lymphocyte subpopulations (both absolute counts and relative percentages) from 1528 subjects. The accuracy, completeness, and authenticity of the data were strictly maintained, and a comprehensive database was established.

Statistical analysis

Statistical analyses and plotting were conducted using SAS software (version9.4 for Windows). Quantitative data exhibiting a normal distribution are presented as the mean ± SD, and group comparisons were performed utilizing independent sample t-tests and analysis of variance (ANOVA). Qualitative data are expressed as percentages, with intergroup comparisons performed using the χ² test. One-way ANOVA was used to analyze the associations between lymphocyte subsets and the time to negative SARS-CoV-2 nucleic acid test result, while the χ² trend test was used to assess their relationships with clinical disease severity in COVID-19 patients. Pearson’s correlation analysis was utilized to evaluate the association between variables. Multivariable linear regression analysis was conducted to identify potential influencing factors of lymphocyte subsets and their relative percentages. A significance level of α = 0.05 was set for all analyses.

RESULTS

Baseline characteristics of study patients

Significant differences were observed in quantitative indices such as age, time to negative SARS-CoV-2 nucleic acid test result, body mass index (BMI), alanine aminotransferase, aspartate aminotransferase, gamma-glutamyl transferase, fasting plasma glucose (FPG); hemoglobin A1c (HbA1c); systolic blood pressure, diastolic blood pressure; triglycerides, total cholesterol, low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol between the two groups (all P < 0.05). Additionally, there were statistically significant differences in gender distribution, COVID-19 clinical stage, and the percentage of patients with liver function abnormalities between the two groups (all P < 0.05). Detailed baseline data are shown in Table 1.

Table 1 Patient characteristics, n (%)/mean ± SD.

Characteristic	COVID-19 only (n = 858)	COVID-19 with NAFLD (n = 670)	P value
Age	35.22 ± 12.70	41.23 ± 11.48	< 0.001
Time to negative SARS-CoV-2 nucleic acid test result	11.96 ± 8.96	13.61 ± 11.15	0.002
BMI	22.03 ± 3.02	26.45 ± 3.16	< 0.001
ALT	22.62 ± 20.58	49.60 ± 37.84	< 0.001
AST	22.04 ± 8.58	30.56 ± 17.09	< 0.001
GGT	22.89 ± 19.17	45.10 ± 34.02	< 0.001
TBIL	8.70 ± 4.77	9.48 ± 5.10	0.002
FPG	5.05 ± 1.34	5.89 ± 2.61	< 0.001
HbA1c	5.40 ± 0.67	5.91 ± 1.22	< 0.001
SBP	124.66 ± 14.95	133.95 ± 16.26	< 0.001
DBP	83.60 ± 11.02	90.25 ± 12.01	< 0.001
TC	4.50 ± 0.97	4.83 ± 1.05	< 0.001
TG	1.31 ± 0.83	2.14 ± 1.50	< 0.001
HDL-C	1.27 ± 0.33	1.09 ± 0.26	< 0.001
LDL-C	2.79 ± 2.97	3.05 ± 0.88	< 0.001
Male	509 (59.3)	592 (88.36)	< 0.001
Female	349 (40.7)	78 (11.64)
Cardiopathy	7 (0.81)	15 (2.24)	< 0.001
Hypertension	46 (5.36)	134 (20.00)	< 0.001
Diabetes	10 (1.17)	78 (11.64)	< 0.001
Hyperlipidemia	71 (8.28)	204 (30.45)	< 0.001
Hyperuricemia	22 (2.56)	91 (13.58)	< 0.001
Asymptomatic cases	304 (35.43)	210 (31.34)	< 0.001
Mild cases	229 (26.69)	133 (19.85)
Moderate cases	312 (36.36)	310 (46.27)
Severe cases	4 (0.47)	12 (1.79)
Critical cases	4 (0.47)	4 (0.60)
Liver function abnormalities	247 (28.79)	420 (62.69)	< 0.001

COVID-19: Coronavirus disease 2019; NAFLD: Non-alcoholic fatty liver disease; BMI: Body mass index; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; GGT: Gamma-glutamyl transpeptidase; TBIL: Total bilirubin; FPG: Fasting plasma glucose; HbA1c: Hemoglobin A1c; SBP: Systolic blood pressure; DBP: Diastolic blood pressure; TC: Total cholesterol; TG: Triglycerides; LDL-C: Low-density lipoprotein cholesterol; HDL-C: High-density lipoprotein cholesterol.

Comparison of lymphocyte subsets and their percentages between obese patients without NAFLD and those with NAFLD

When comparing lymphocyte subsets between obese patients without NAFLD and those with NAFLD, a statistically significant difference was observed in CD19+ percentages (P < 0.05), as shown in Table 2.

Table 2 Comparison of lymphocyte subsets and their percentages between obese patients without non-alcoholic fatty liver disease and those with non-alcoholic fatty liver disease, mean ± SD.

Variable	Obese patients without NAFLD	Obese patients with NAFLD	P value
LY (cells/μL)	2023.96 ± 790.22	2306.34 ± 946.34	0.117
LY%	21.01 ± 7.14	21.67 ± 7.90	0.377
CD3+ (cells/μL)	1455.32 ± 584.53	1654.04 ± 693.65	0.097
CD3+CD4+ (cells/μL)	842.29 ± 385.28	977.5 ± 436.20	0.064
CD3+CD8+ (cells/μL)	515.00 ± 251.28	585.37 ± 312.55	0.475
CD3+%	72.38 ± 6.71	71.98 ± 7.15	0.939
CD3+CD4+%	41.56 ± 7.26	41.94 ± 7.37	0.550
CD3+CD8+%	27.12 ± 10.30	25.45 ± 7.43	0.067
CD4/CD8	1.81 ± 0.75	1.82 ± 0.74	0.759
CD19+ (cells/μL)	252.04 ± 238.43	361.28 ± 245.29	0.311
CD56+ (cells/μL)	276.14 ± 169.13	237.26 ± 170.36	0.788
CD19+%	10.61 ± 4.51	15.02 ± 6.40	0.019
CD56+%	13.77 ± 8.25	10.49 ± 6.79	0.249

NAFLD: Non-alcoholic fatty liver disease; LY: Lymphocyte counts.

Comparison of lymphocyte subsets and their percentages at admission, discharge, and 2 weeks and 4 weeks after discharge

Statistically significant differences in lymphocyte subsets were observed between the two groups across multiple time points (all P < 0.05). At admission, the COVID-19 with NAFLD group demonstrated significantly higher levels of CD3+ counts, CD3+CD4+ counts, and lymphocyte counts compared to the COVID-19 group. At discharge, group differences were noted in lymphocyte counts, CD56+ counts and percentages, CD3+CD4+ percentages, and the ratio of CD3+CD4+ to CD3+CD8+. Immunological differences persisted during the follow-up period: At 2 weeks post-discharge, CD3+CD4+ cell counts, lymphocyte counts, CD19+ cell counts, and CD56+ counts remained significantly elevated; by 4 weeks, these differences extended to CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts and percentages, as illustrated in Figure 1A and B. Additionally, longitudinal analysis exhibited statistically significant differences in CD3+ counts and percentages, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, CD19+ counts and percentages, CD56+ counts and percentages, and the CD3+CD4+/CD3+CD8+ ratio among COVID-19 patients with NAFLD across all assessment time points (admission, discharge, 2-week post-discharge, and 4-week post-discharge) (all P < 0.05; Figure 1C and D).

Figure 1 Comparison of lymphocyte subset counts and their percentages at admission, discharge, and 2 weeks and 4 weeks after discharge. A and B: Comparison between coronavirus disease 2019 (COVID-19) patients with non-alcoholic fatty liver disease and those without; C and D: Comparison across different time points in COVID-19 patients with non-alcoholic fatty liver disease. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001. COVID-19: Coronavirus disease 2019; LY: Lymphocyte counts.

Analysis of lymphocyte subsets and their percentages in COVID-19 patients with NAFLD at admission, discharge, and 2 weeks and 4 weeks post-discharge, after adjustment for comorbidities

After adjusting for confounding comorbidities (including overweight, obesity, diabetes, and dyslipidemia), COVID-19 patients with NAFLD demonstrated significantly higher CD3+ counts, CD3+CD4+ counts, and lymphocyte counts compared to those without NAFLD at admission and 4 weeks post-discharge (all P < 0.05) (Figure 2A and B). Notably, longitudinal analysis within the NAFLD-comorbid group revealed significant variations in CD3+ percentages, CD3+CD4+ counts and percentages, CD3+CD8+ percentages, lymphocyte counts and percentages, CD19+ counts and percentages, CD56+ counts and percentages, and the ratio of CD3+CD4+ to CD3+CD8+ across all evaluated time points (admission, discharge, and 2-week and 4-week post-discharge follow-ups) (all P < 0.05) (Figure 2C and D).

Figure 2 Analysis of lymphocyte subsets and their percentages in coronavirus disease 2019 patients with non-alcoholic fatty liver disease at admission, discharge, and 2 weeks and 4 weeks post-discharge after adjustment for comorbidities. A and B: Comparison between coronavirus disease 2019 (COVID-19) patients with non-alcoholic fatty liver disease and those without; C and D: Comparison across different time points in COVID-19 patients with non-alcoholic fatty liver disease. ^aP < 0.05; ^bP < 0.01; ^cP < 0.001. COVID-19: Coronavirus disease 2019; LY: Lymphocyte counts.

Correlation between lymphocyte subsets and their percentages with the time to negative SARS-CoV-2 nucleic acid test result in patients with COVID-19 and NAFLD

The results of the ANOVA trend test indicated a significant linear trend between lymphocyte counts and percentages, CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, and CD19+ counts, and the time to negative SARS-CoV-2 nucleic acid test result (all P < 0.05), as shown in Table 3. Pearson correlation analysis revealed that CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, CD19+ counts, and CD56+ counts were inversely correlated with the time to negative SARS-CoV-2 nucleic acid test result (all P < 0.001), as shown in Table 4.

Table 3 Linear trend analysis of lymphocyte subsets and their percentages in association with the time to negative severe acute respiratory syndrome coronavirus 2 nucleic acid test result in patients with coronavirus disease 2019 and non-alcoholic fatty liver disease.

Variable	F	P value
LY (cells/μL)	53.817	< 0.001
LY%	27.065	< 0.001
CD3+ (cells/μL)	54.778	< 0.001
CD3+CD4+ (cells/μL)	50.838	< 0.001
CD3+CD8+ (cells/μL)	29.331	< 0.001
CD3+%	3.251	0.072
CD3+CD4+%	0.003	0.955
CD3+CD8+%	0.001	0.972
CD4/CD8	0.459	0.499
CD19+ (cells/μL)	8.599	0.008
CD56+ (cells/μL)	3.501	0.076
CD19+%	1.673	0.211
CD56+%	0.114	0.739

LY: Lymphocyte counts.

Table 4 Pearson correlation analysis of lymphocyte subsets and their percentages with the Time to negative severe acute respiratory syndrome coronavirus 2 nucleic acid test result in patients with coronavirus disease 2019 and non-alcoholic fatty liver disease.

Variable	Time to negative severe acute respiratory syndrome coronavirus 2 nucleic acid test result
	r	P value
LY (cells/μL)	-0.270	< 0.001
LY%	-0.193	< 0.001
CD3+ (cells/μL)	-0.274	< 0.001
CD3+CD4+ (cells/μL)	-0.264	< 0.001
CD3+CD8+ (cells/μL)	-0.204	< 0.001
CD3+%	-0.062	0.111
CD3+CD4+%	0.002	0.961
CD3+CD8+%	0.001	0.987
CD4/CD8	0.028	0.475
CD19+ (cells/μL)	-0.186	< 0.001
CD56+ (cells/μL)	-0.093	0.018
CD19+%	-0.014	0.726
CD56+%	0.081	0.039

LY: Lymphocyte counts.

Correlation of lymphocyte subsets and their percentages with clinical disease severity in COVID-19 among patients with NAFLD

For this analysis, severe and critical types of COVID-19 were combined into a single “severe” group, while all other cases were categorized as “non-severe”. A significant linear relationship was found between lymphocyte counts and percentages, CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, CD19+ counts, and CD56+ counts, and the incidence of severe COVID-19, with χ² trend values of 24.995, 24.038, 18.591, 17.564, 8.839, 5.444 and 22.745, respectively (all P < 0.001), as shown in Table 5. Pearson correlation analysis demonstrated an inverse association between CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, CD19+ counts, and CD56+ counts, and clinical disease severity of COVID-19 (all P < 0.001), as shown in Table 6.

Table 5 Linear trend analysis of lymphocyte subsets and their percentages in relation to clinical disease severity in coronavirus disease 2019 patients with non-alcoholic fatty liver disease.

Variable	χ2 value	P value
LY (cells/μL)	24.995	< 0.001
LY%	22.745	< 0.001
CD3+ (cells/μL)	24.038	< 0.001
CD3+CD4+ (cells/μL)	18.591	< 0.001
CD3+CD8+ (cells/μL)	17.564	< 0.001
CD3+%	0.545	0.461
CD3+CD4+%	0.031	0.861
CD3+CD8+%	0.051	0.822
CD4/CD8	0.108	0.742
CD19+ (cells/μL)	8.839	0.003
CD56+ (cells/μL)	5.444	0.02
CD19+%	0.165	0.685
CD56+%	1.404	0.236

LY: Lymphocyte counts.

Table 6 Pearson correlation analysis between lymphocyte subsets and their percentages and clinical disease severity in coronavirus disease 2019 patients with non-alcoholic fatty liver disease.

Variable	Clinical disease severity
	r	P value
LY (cells/μL)	-0.193	< 0.001
LY%	-0.185	< 0.001
CD3+ (cells/μL)	-0.190	< 0.001
CD3+CD4+ (cells/μL)	-0.167	< 0.001
CD3+CD8+ (cells/μL)	-0.162	< 0.001
CD3+%	-0.029	0.461
CD3+CD4+%	-0.007	0.861
CD3+CD8+%	-0.009	0.822
CD4/CD8	0.013	0.743
CD19+ (cells/μL)	-0.115	0.003
CD56+ (cells/μL)	-0.090	0.02
CD19+%	0.016	0.685
CD56+%	0.046	0.236

LY: Lymphocyte counts.

Multivariable linear regression analysis of factors influencing lymphocyte subsets in COVID-19 patients with NAFLD

Multivariable linear regression analysis indicated that factors including COVID-19 severity, BMI, impaired fasting glucose (IFG), and liver fibrosis demonstrated statistically significant influence on CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts (all P < 0.05). Notably, these factors demonstrated substantial modulatory effects on lymphocyte counts. A 1-unit increase in BMI was associated with a 33.191-unit increase in lymphocyte counts (β = 33.191, P < 0.001). Severe COVID-19 cases exhibited a 696.257-unit reduction in lymphocyte counts compared to non-severe cases (β = -696.257, P < 0.05).

Patients with IFG had a 233.764-unit decrease in lymphocyte counts than those with normal fasting glucose (β = -233.764,P < 0.05). And each 1-unit increase in liver fibrosis severity decreased lymphocyte counts by 390.922 units (b = -390.922,P < 0.001). Furthermore, age was associated with lymphocyte counts in the multivariable model, showing a 7.529-unit decrease in lymphocyte counts for each 1-year increase in age (β = -7.529, P < 0.05). In conclusion, the severity of COVID-19, IFG, liver fibrosis, and advanced age were independent risk factors for decreased lymphocyte counts, as illustrated in Table 7.

Table 7 Multivariable linear regression analysis of factors influencing lymphocyte subset counts in coronavirus disease 2019 patients with non-alcoholic fatty liver disease.

		Β	SD	Beta	t	P value
CD3+ (cells/μL)	Age	-4.908	2.731	-0.088	-1.797	0.073
	COVID-19 severity	-450.591	175.134	-0.101	-2.573	0.01
	BMI	20.825	7.259	0.107	2.869	0.004
	IFG	-227.875	81.150	-0.152	-2.808	0.005
	Liver fibrosis	-324.154	71.100	-0.200	-4.599	< 0.001
CD3+CD4+ (cells/μL)	Age	0.507	1.778	0.014	0.285	0.776
	COVID-19 severity	-242.804	114.044	-0.085	-2.129	0.034
	BMI	11.363	4.690	0.091	2.423	0.016
	IFG	-138.203	52.844	-0.144	-2.615	0.009
	Liver fibrosis	-211.795	46.299	-0.204	-4.574	< 0.001
CD3+CD8+ (cells/μL)	Age	-4.062	1.201	-0.169	-3.381	< 0.001
	COVID-19 severity	-171.379	77.042	-0.089	-2.224	0.027
	BMI	9.546	3.197	0.113	2.986	0.003
	IFG	-80.441	35.698	-0.124	-2.253	0.025
	Liver fibrosis	-100.665	31.227	-0.144	-3.218	0.001
LY (cells/μL)	Age	-7.529	3.715	-0.100	-2.027	0.043
	COVID-19 severity	-696.257	238.256	-0.116	-2.922	0.004
	BMI	33.191	9.801	0.126	3.386	< 0.001
	IFG	-233.764	110.399	-0.116	-2.117	0.035
	Liver fibrosis	-390.922	128.213	-0.267	-2.570	< 0.001
CD19+ (cells/μL)	Age	-0.690	0.967	-0.037	-0.714	0.476
	COVID-19 severity	-98.248	62.003	-0.065	-1.585	0.114
	BMI	7.508	2.517	0.113	2.983	0.003
	IFG	-23.139	28.730	-0.046	-0.805	0.421
	Liver fibrosis	-46.090	25.172	-0.084	-1.831	0.068
CD56+ (cells/μL)	Age	-1.478	0.733	-0.106	-1.880	0.044
	COVID-19 severity	-88.335	46.978	-0.080	-1.880	0.061
	BMI	5.588	1.869	0.116	2.990	0.003
	IFG	-24.824	21.768	-0.066	-1.140	0.255
	Liver fibrosis	-3.042	19.072	-0.008	-0.159	0.873

Liver fibrosis: < F2: Non-significant liver fibrosis; ≥ F2: Significant liver fibrosis; ≥ F4: Cirrhosis. The number of comorbidities (0, 1, 2, 3 and more). LY: Lymphocyte counts; COVID-19: Coronavirus disease 2019; BMI: Body mass index; IFG: Impaired fasting glucose.

Multivariable linear regression analysis of association of plasma glucose and lipid profiles with lymphocytes subsets in COVID-19 patients with NAFLD

Multivariable linear regression analysis demonstrated that LDL-C and HbA1c were significantly associated with CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts (all P < 0.05). Specifically, a 1-unit increase in LDL-C and HbA1c was associated with a 348.534-unit and 136.512-unit increase in lymphocyte counts, respectively (β = 348.534 and 136.512, P < 0.001). Additionally, HbA1c and FPG were positively associated with CD19+ counts and CD56+ counts, respectively (β = 23.790 and 18.366, P < 0.05), as shown in Table 8.

Table 8 Multivariable linear regression analysis of association between plasma glucose and lipid profiles with lymphocyte subsets in coronavirus disease 2019 patients with non-alcoholic fatty liver disease.

		Β	SD	Beta	t	P value
CD3+ (cells/μL)	TC	16.645	97.254	0.029	0.171	0.864
	TG	-2.000	26.365	-0.005	-0.076	0.940
	HDL-C	390.391	359.852	0.154	1.085	0.278
	LDL-C	256.168	67.826	0.371	3.777	< 0.001
	FPG	46.102	27.497	0.194	1.677	0.094
	HbA1c	106.171	27.755	0.205	3.825	< 0.001
CD3+CD4+ (cells/μL)	TC	-0.291	62.833	-0.001	-0.005	0.996
	TG	-8.329	17.034	-0.035	-0.489	0.625
	HDL-C	191.088	232.492	0.118	0.822	0.411
	LDL-C	151.638	43.821	0.343	3.460	< 0.001
	FPG	20.552	17.765	0.135	1.157	0.248
	HbA1c	56.037	17.932	0.169	3.125	0.002
CD3+CD8+ (cells/μL)	TC	15.817	42.833	0.064	0.369	0.712
	TG	6.168	11.612	0.038	0.531	0.595
	HDL-C	184.052	158.486	0.169	1.161	0.246
	LDL-C	97.151	29.872	0.372	3.252	0.001
	FPG	24.161	12.110	0.236	1.995	0.046
	HbA1c	49.049	12.224	0.220	4.013	< 0.001
LY (cells/μL)	TC	-22.561	131.308	-0.029	-0.172	0.864
	TG	-0.236	35.598	0.000	-0.007	0.995
	HDL-C	569.767	485.859	0.167	1.173	0.241
	LDL-C	348.534	91.576	0.375	3.806	< 0.001
	FPG	43.050	37.125	0.134	1.160	0.247
	HbA1c	136.512	37.474	0.196	3.643	< 0.001
CD19+ (cells/μL)	TC	4.811	33.715	0.025	0.143	0.887
	TG	1.366	9.140	0.011	0.149	0.881
	HDL-C	81.301	124.749	0.095	0.652	0.515
	LDL-C	41.076	23.513	0.176	1.747	0.081
	FPG	7.725	9.532	0.096	0.810	0.418
	HbA1c	23.790	9.622	0.136	2.472	0.014
CD56+ (cells/μL)	TC	15.622	25.037	0.110	0.624	0.533
	TG	-5.151	6.787	-0.056	-0.759	0.448
	HDL-C	114.812	92.639	0.185	1.239	0.216
	LDL-C	18.185	17.461	0.107	1.041	0.298
	FPG	18.366	7.079	0.314	2.595	0.01
	HbA1c	-4.433	7.145	-0.035	-0.620	0.535

TC: Total cholesterol; TG: Triglycerides; HDL-C: High-density lipoprotein cholesterol; LDL-C: Low-density lipoprotein cholesterol; FPG: Fasting plasma glucose; HbA1c: Hemoglobin A1c; LY: Lymphocyte counts.

Impact of age on lymphocyte subsets and their percentages in COVID-19 patients with NAFLD: A longitudinal analysis from hospital admission to 4 weeks post-discharge

Stratification by age revealed statistically significant differences in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, and CD19+ counts and percentages at admission, discharge, and 2 weeks and 4 week post-discharge among different age groups (all P < 0.05). Notably, patients aged ≥ 60 years consistently demonstrated lower levels of these immune markers compared to the 18-44-year-old group at all measured time points (admission, discharge, 2-week follow-up, and 4-week follow-up) (all P < 0.05), as shown in Figure 3.

Figure 3 Impact of age on lymphocyte subsets and their percentages in coronavirus disease 2019 patients with non-alcoholic fatty liver disease: A longitudinal analysis from hospital admission to 4 weeks post-discharge. A: Lymphocyte subset counts (cells/μL); B: Lymphocyte subset percentages (%). ^aP < 0.05, 18-44 years vs 45-59 years; ^bP < 0.05, 18-44 years vs ≥ 60 years; ^cP < 0.05, 45-59 years vs ≥ 60 years; ^dP < 0.05; ^eP < 0.01; ^fP < 0.001. LY: Lymphocyte counts.

Impact of liver fibrosis on lymphocyte subsets and their percentages in COVID-19 patients with NAFLD: A longitudinal analysis from hospital admission to 4 weeks post-discharge

At admission, statistically significant differences were observed in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts, CD19+ counts, CD56+ counts, and the ratio of CD3+CD4+ to CD3+CD8+ between patients with varying levels of liver fibrosis (all P < 0.05). Furthermore, significant variations in lymphocyte counts and the ratio of CD3+CD4+ to CD3+CD8+ were consistently demonstrated at both the 2-week and 4-week post-discharge follow-ups (all P < 0.05). Moreover, patients with significant liver fibrosis (≥ F2) showed lower levels of CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts and percentages at admission, discharge, and 2-week and 4-week follow-ups compared to non-significant liver fibrosis (< F2) (all P < 0.05). Similarly, patients with cirrhosis (≥ F4) demonstrated reduced CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts and percentages at admission, along with persistently lower CD3+CD8+ counts and lymphocyte percentages at both 2-week and 4-week follow-up assessments (all P < 0.05), as shown in Figure 4.

Figure 4 Impact of liver fibrosis on lymphocyte subsets and their percentages in coronavirus disease 2019 patients with non-alcoholic fatty liver disease: A longitudinal analysis from hospital admission to 4 weeks post-discharge. < F2: Non-significant liver fibrosis; ≥ F2: Significant liver fibrosis; ≥ F4: Cirrhosis. A: Lymphocyte subset counts (cells/μL); B: Lymphocyte subset percentages (%). ^aP < 0.05, < F2 vs ≥ F2; ^bP < 0.05, ≥ F2 vs ≥ F4; ^cP < 0.05, < F2 vs ≥ F4; ^dP < 0.05; ^eP < 0.01; ^fP < 0.001. LY: Lymphocyte counts.

Impact of comorbidities on lymphocyte subsets and their percentages in COVID-19 patients with NAFLD: A longitudinal analysis from hospital admission to 4 weeks post-discharge

A comparative analysis of lymphocyte subsets in COVID-19 patients with NAFLD demonstrated statistically significant variations in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, CD19+ counts, CD56+ percentages, and the ratio of CD3+CD4+ to CD3+CD8+ at hospital admission, stratified by the number of comorbidities (all P < 0.05). Longitudinal analysis revealed that CD3+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, and the ratio of CD3+CD4+ to CD3+CD8+ maintained statistical significance throughout the entire post-discharge period, with persistent differences observed at both 2-week and 4-week follow-ups (all P < 0.05).

Compared to NAFLD patients without comorbidities, those with two comorbidities demonstrated significantly higher levels of CD3+CD4+ counts and the ratio of CD3+CD4+ to CD3+CD8+ at admission, an elevated CD3+CD4+ to CD3+CD8+ ratio at discharge, and persistently higher CD3+CD4+ percentages and CD3+CD4+/CD3+CD8+ ratio at both 2-week and 4-week post-discharge follow-ups (all P < 0.05).

Moreover, patients with three or more comorbidities displayed further immunological alterations, including higher CD3+CD4+ percentages and CD56+ counts at admission, elevated CD3+CD4+ counts, lymphocyte counts, and CD19+ counts and percentages at discharge, as well as sustained increases in CD3+CD4+ counts at the 2-week follow-up and higher CD3+CD4+ percentages at the 4-week post-discharge follow-up (all P < 0.05), as illustrated in Figure 5.

Figure 5 Impact of comorbidities on lymphocyte subsets and their percentages in coronavirus disease 2019 patients with non-alcoholic fatty liver disease: A longitudinal analysis from hospital admission to 4 weeks post-discharge. A: Lymphocyte subset counts (cells/μL); B: Lymphocyte subset percentages (%). ^aP < 0.05, 0 vs 1; ^bP < 0.05, 0 vs 2; ^cP < 0.05, 0 vs 3; ^dP < 0.05, 1 vs 2; ^eP < 0.05, 1 vs 3; ^fP < 0.05, 2 vs 3; ^gP < 0.001; ^hP < 0.05; ⁱP < 0.01. LY: Lymphocyte counts; 0: No comorbidities; 1: One comorbidity; 2: Two comorbidities; ≥ 3: Three or more comorbidities.

Impact of BMI on lymphocyte subsets and their percentages in COVID-19 patients with NAFLD: A longitudinal analysis from hospital admission to 4 weeks post-discharge

Stratified analysis by BMI revealed significant differences in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts, and CD19+ counts and percentages at admission among COVID-19 patients with NAFLD (all P < 0.05). Comparative analyses demonstrated that obese individuals maintained consistently higher CD3+CD4+ counts, CD19+ counts, and lymphocyte counts throughout all clinical phases—from hospitalization through discharge and during the 2-week to 4-week recovery period—when compared with normal-weight controls (all P < 0.05). Similarly, lymphocyte counts and CD19+ counts significantly differed between overweight and obese groups at admission, discharge, and 2 weeks post-discharge (all P < 0.05). CD3+ percentages also showed significant intergroup differences at the 2-week and 4-week follow-ups (all P < 0.05), as demonstrated in Figure 6.

Figure 6 Impact of body mass index on lymphocyte subsets and their percentages in coronavirus disease 2019 patients with non-alcoholic fatty liver disease: A longitudinal analysis from hospital admission to 4 weeks post-discharge. A: Lymphocyte subset counts (cells/μL); B: Lymphocyte subset percentages (%). ^aP < 0.05, normal weight vs overweight; ^bP < 0.05, normal weight vs obesity; ^cP < 0.05, overweight vs obesity; ^dP < 0.05; ^eP < 0.01; ^fP < 0.001. Normal weight: 18.5 ≤ body mass index (BMI) < 24; Overweight: 24 ≤ BMI < 28; Obesity: BMI ≥ 28; LY: Lymphocyte counts.

Impact of liver enzymes on lymphocyte subsets and their percentages in COVID-19 patients with NAFLD: A longitudinal analysis from hospital admission to 4 weeks post-discharge

Patients with elevated liver enzymes exhibited persistently higher CD3+ counts, CD3+CD4+ counts, lymphocyte counts, and CD19+ counts at all post-admission evaluations (discharge through the 4-week follow-up), compared to those with normal liver enzymes (all P < 0.05). Additionally, CD3+CD8+ counts and lymphocyte percentages remained significantly different at 4 weeks post-discharge (all P < 0.05), as illustrated in Figure 7.

Figure 7 Impact of liver enzymes on lymphocyte subsets and their percentages in coronavirus disease 2019 patients with non-alcoholic fatty liver disease: A longitudinal analysis from hospital admission to 4 weeks post-discharge. A: Lymphocyte subset counts (cells/μL); B: Lymphocyte subset percentages (%). ^aP < 0.05; ^bP < 0.01; ^cP < 0.001. LY: Lymphocyte counts.

Impact of vaccination on lymphocyte subsets and their percentages in COVID-19 patients with NAFLD: A longitudinal analysis from hospital admission to 4 weeks post-discharge

Vaccinated patients demonstrated significantly lower CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts and percentages at admission compared to unvaccinated patients (all P < 0.05). However, from discharge through the 4-week follow-up, vaccinated patients demonstrated higher CD3+ counts, CD3+CD4+ counts, lymphocyte counts and percentages, and CD19+ counts (all P < 0.05). Additionally, CD3+CD8+ counts also remained significantly elevated at both discharge and the 4-week post-discharge time points (all P < 0.05), as shown in Figure 8.

Figure 8 Impact of vaccination on lymphocyte subsets and their percentages in coronavirus disease 2019 with non-alcoholic fatty liver disease: A longitudinal analysis from hospital admission to 4 weeks post-discharge. A: Lymphocyte subset counts (cells/μL); B: Lymphocyte subset percentages (%). ^aP < 0.05; ^bP < 0.01; ^cP < 0.001. LY: Lymphocyte counts.

DISCUSSION

In this cohort study, we enrolled 1528 patients comprising two distinct subgroups: COVID-19 patients with NAFLD and those without. Longitudinal analyses were conducted to assess potential determinants of lymphocyte subset distributions and their relative proportions at four predefined time points: Admission, discharge, and 2 weeks and 4 weeks post-discharge. Additionally, we examined the potential associations between lymphocyte profiles and two key clinical outcomes: The time to negative SARS-CoV-2 nucleic acid test result and progression to severe COVID-19.

Pearson correlation analysis identified statistically significant inverse associations between key lymphocyte populations (CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, CD19+ counts, and CD56+ counts) and both the time to negative SARS-CoV-2 nucleic acid test result and clinical disease severity in COVID-19. Compared to COVID-19 patients, NAFLD-comorbid patients demonstrated persistently elevated CD3+CD4+ counts and lymphocyte counts and percentages at admission, and at 2-week and 4-week follow-ups post-discharge.

Multivariate analysis demonstrated that age, COVID-19 severity, BMI, IFG, liver fibrosis, vaccination status, LDL-C, and HbA1c significantly influenced CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts, with particularly pronounced effects on lymphocyte counts. Furthermore, elevated HbA1c and FPG specifically impacted both CD19+ counts and CD56+ counts.

Stratification by comorbidity burden uncovered statistically significant intergroup differences in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, CD19+ counts, CD56+ percentages, CD19+ counts, CD56+ percentages, and the ratio of CD3+CD4+ to CD3+CD8+ at hospital admission.

Most remarkably, these patients demonstrated persistent immunological disturbances —particularly in CD3+ counts, CD3+CD8+ counts, and lymphocyte counts—that remained detectable throughout the post-discharge recovery phase (2/4 weeks). Compared to NAFLD patients without comorbidities, patients with two comorbidities showed statistically significantly elevated CD3+CD4+ counts and the ratio of CD3+CD4+ to CD3+CD8+ at admission, a higher CD3+CD4+/CD3+CD8+ ratio at discharge, and persistently higher CD3+CD4+ percentages and CD3+CD4+ to CD3+CD8+ ratio at both 2-week and 4-week post-discharge follow-ups.

Moreover, patients with three or more comorbidities displayed further immunological alterations, including higher CD3+CD4+ percentages and CD56+ counts at admission, higher CD3+CD4+ counts, lymphocyte counts, and CD19+ counts and percentages at discharge, as well as sustained increases in CD3+CD4+ counts at the 2-week follow-up and increased CD3+CD4+ percentages at the 4-week post-discharge follow-up.

After adjusting for confounding comorbidities, COVID-19 patients with NAFLD exhibited significant alterations in multiple lymphocyte parameters—including CD3+ percentages, CD3+CD4+ counts and percentages, CD3+CD8+ percentages, lymphocyte counts and percentages, CD19+ counts and percentages, CD56+ counts and percentages, and the CD3+CD4+/CD3+CD8+ ratio—across all clinical stages, from hospitalization through the 4-week post-discharge recovery. Additionally, COVID-19 patients with NAFLD demonstrated significantly higher CD3+ counts, CD3+CD4+ counts, and lymphocyte counts at admission and 4 weeks post-discharge, compared to those without NAFLD. These persistent differences indicate that NAFLD is an independent factor of lymphocyte subset distribution and phenotypic profiles. Patients with uncomplicated NAFLD (including lean NAFLD) also require particular clinical attention to dynamic lymphocyte subset alterations to mitigate risks of immune imbalance.

Analyses stratified by liver fibrosis stage uncovered significant intergroup differences in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts, CD19+ counts, CD56+ counts, and the CD3+CD4+/CD3+CD8+ ratio at admission. Furthermore, patients with significant liver fibrosis (≥ F2) exhibited lower CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages at admission, discharge, and 2-week and 4-week follow-ups compared to those with non-significant liver fibrosis (< F2). Similarly, patients with cirrhosis (≥ F4) demonstrated reduced CD3+CD4+ counts, CD3+CD8+ counts, and lymphocyte counts and percentages at admission, along with persistently lower CD3+CD8+ counts and lymphocyte percentages at both 2-week and 4-week follow-up assessments. Furthermore, patients with elevated liver enzymes demonstrated persistently higher CD3+ counts, CD3+CD4+ counts, lymphocyte counts, and CD19+ counts at all post-admission assessments (discharge through the 4-week follow-up), compared to those with normal liver enzymes.

During NAFLD pathogenesis, resident Kupffer cells and monocyte-derived macrophages function as hepatic sentinel cells that demonstrate enhanced sensitivity to locally enriched exogenous antigens, metabolic byproducts, and damage-associated molecular patterns (DAMPs). This induces a fundamental transition from immune tolerance to an activated state, evidenced by: (1) Suppressed secretion of anti-inflammatory mediators IL-10 and transforming growth factor-β; and (2) Significant upregulation of pro-inflammatory cytokines (tumor necrosis factor-α, IL-1β, and IL-6)[47,48]. T cell-specific effects of IL-6 in COVID-19 might include the facilitation of T helper (Th17) cell responses and indirectly the recruitment of Th17 cells to the infected epithelium. Within the tissue, Th17 cells are responsible for the attraction of neutrophils and amplification of immunopathology[49]. A subset of tissue-resident memory-like Th17 cells has been reported in the lungs of COVID-19 patients, which might contribute to hyperinflammation in severe cases by the production of IL-17A and granulocyte-macrophage colony-stimulating factor, as well as interactions with lung macrophages and cytotoxic CD8+ T cells[50].

Notably, in COVID-19 patients, a proinflammatory milieu is driven by factors including the high expression of CD302, a C-type lectin receptor induced in vitro upon lipopolysaccharide stimulation, alarmins such as S100 calcium binding protein A8 and S100 calcium binding protein A12, calcium-binding proteins, and other DAMPs. The expression of these molecules is regulated by proinflammatory mediators such as interferon-γ and tumor necrosis factor-α and that can lead to the secretion of IL-6 and IL-8[50]. In NAFLD, this inflammatory milieu is perpetuated through complex crosstalk between hepatocytes and both adaptive and innate immune cells, driving a state of chronic low-grade inflammation that underlies disease progression[47,48].

Diffuse lymphocytic lobular infiltration was consistently identified as a characteristic histological feature in NAFLD[51,52]. Immunohistochemical analysis revealed focal lymphoid aggregates containing both B and T cell populations in approximately 60% of NASH cases[53]. Liver infiltration of B cells and CD4+ and CD8+ T cells was evident in different NASH models, which exacerbates liver parenchymal damage and lobular inflammation. These T cells express the memory or effector markers (CD25, CD44, and CD69) and exhibit enhanced production of cytokines, indicating their functional activation[52,54,55].

Mechanistically, natural killer cells isolated from NASH livers secrete higher levels of pro-inflammatory cytokines (interferon-γ, IL-1β, IL-12, C-C motif ligand 4, C-C motif ligand 5, and granulocyte-macrophage colony-stimulating factor), which could activate hepatic Janus kinase-signal transducer and activator of transcription (JAK/STAT) 1/3 and nuclear factor kappa B signaling and induce hepatocyte damage evidenced by increased reactive oxygen species and apoptosis rate[56]. Moreover, SARS-CoV-2 infection triggers inflammatory via the JAK/STAT pathway, leading to the recruitment of pneumocytes, endothelial cells, macrophages, monocytes, lymphocytes, natural killer cells, and dendritic cells, thereby driving the progression towards a cytokine storm. These produces various inflammatory markers in the host that dictate disease severity. The JAK/STAT signaling pathway is also pivotal for immune responses, exerting its effects through the modulation of B cell function and T cell differentiation[57].

T-cell activation GTPase activating protein plays a role in immune cell regulation. Predominantly distributed in CD4+/CD8+ T cells, T-cell activation GTPase activating protein influences CD4+ T cell differentiation and function through the JAK/STAT pathway, promoting immune infiltration and cytotoxicity[58]. Additionally, B cells not only directly mediate the chemotaxis of other immune cells enriched in immune damage and inflammation, but also act as antigen-presenting cells to activate adaptive immunity[59]. Both innate and adaptive immune mechanisms are actively involved in the development of inflammation in NAFLD and fibrogenesis[60]. The innate immune system triggers initial inflammatory responses via DAMP recognition, subsequently driving fibrotic progression. Accumulating evidence demonstrated that T cell-mediated adaptive immunity also contributes to NASH pathogenesis via multiple mechanisms, including direct cytotoxicity, cytokine production, and secretion of other pro-inflammatory and pro-fibrotic mediators[61].

Moreover, regarding the impact of comorbidities on lymphocyte subsets, prior studies have demonstrated that in COVID-19 patients, the NAFLD-diabetes group, the NAFLD with hyperlipidemia group, and the NAFLD with hyperuricemia/gout group exhibited significantly increased levels of various immunological parameters compared to the NAFLD-only group: Lymphocyte counts and percentages, CD3+ counts and percentages, CD19+ counts and percentages, CD56+ counts and percentages, CD3+CD4+ counts and percentages, CD3+CD8+ counts and percentages, and the CD3+CD4+/CD3+CD8+ ratio[35].

Stratification by BMI revealed distinct immunological profiles among NAFLD-COVID-19 patients at admission, showing significant intergroup variations in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts, and CD19+ counts. Comparative studies demonstrated that obese individuals maintained elevated CD3+CD4+ counts, CD19+ counts, and lymphocyte counts throughout all clinical course—from hospitalization through discharge and during the 2-week to 4-week recovery period—when compared with patients of normal weight.

Excess body fat was demonstrated to impair immune function, with population-based data revealing an elevated incidence of autoimmune and immune-mediated disorders in obese individuals[62]. Obesity-associated immune dysfunction, characterized by impaired T lymphocyte effector function and macrophage polarization defects, was demonstrated to weaken immune defenses, thereby increasing susceptibility to infections and chronic respiratory diseases[63,64].

Notably, severe obesity was identified as a significant risk factor for viral infectious diseases[10,65]. Specifically, excess adiposity was found to adversely affect immune function and host defense mechanisms[65], while the response to viral and bacterial challenges was shown to be impaired in obese hosts[65-67]. Obese individuals with diabetes exhibited marked polyclonal B-cell activation and elevated antibody secretion, while demonstrated impaired responsiveness to neo-antigens like seasonal influenza vaccines. This heightened B-cell activation was significantly associated with the progression of type 2 diabetes in obese subjects[68]. An analysis demonstrated significantly reduced natural killer cell activity in patients with type 2 diabetes mellitus compared to normoglycemic or prediabetic controls[69].

Stratification by age revealed significant differences in CD3+ counts, CD3+CD4+ counts, CD3+CD8+ counts, lymphocyte counts and percentages, and CD19+ counts and percentages at admission, discharge, and 2-week and 4-week post-discharge follow-ups. Most strikingly, patients aged ≥ 60 years exhibited significantly lower levels of these parameters compared to those aged 18-44 years at all measured time points: Admission, discharge, 2-week follow-up, and 4-week follow-up. Immune function was demonstrated to decline progressively with advancing age[70]. This age-associated deterioration of immune competence, termed immunosenescence, was identified as a key contributor to the increased incidence and severity of infectious diseases, as well as reduced vaccine efficacy in elderly populations. Immunosenescence was characterized by both cellular immune dysfunction (particularly in T cell-mediated responses) and impaired humoral immunity. Age-dependent functional deficits in T and B lymphocyte responses were consistently observed to coincide with progressive alterations in innate immune system components[70].

Considering all contributing factors, in the clinical management of COVID-19 patients with NAFLD, special attention should be given to age, liver fibrosis, comorbidity burden, obesity, and hepatic function due to their significant immunomodulatory effects. High-risk populations—particularly those with obesity (BMI ≥ 28 kg/m²), advanced age (≥ 60 years), multiple comorbidities (≥ 2 concurrent conditions), hepatic dysfunction, or significant liver fibrosis/cirrhosis—require enhanced clinical surveillance.

A comprehensive management approach should include concurrent monitoring of LDL-C, HbA1c, and FPG, with immunomodulatory therapies being integrated with targeted control of these metabolically driven immunosuppressive risk factors. Furthermore, discharge criteria should encompass not only resolution of COVID-19 symptoms and SARS-CoV-2 viral clearance but also the stabilization of both chronic disease parameters and lymphocyte subset profiles. Furthermore, patients with uncomplicated NAFLD (including lean NAFLD) also require particular clinical attention to dynamic changes in lymphocyte subsets to mitigate risks of immune imbalance.

Our study has several limitations, such as insufficient liver biopsy data due to the avoidance of additional infection risks for COVID-19-NAFLD patients. This consequently led to an insufficient assessment of the correlation between hepatic steatosis severity and dynamic changes in lymphocyte profiles. However, through rigorous multivariate adjustment, we demonstrated that several metabolic parameters were independently associated with distinct immune imbalance patterns in COVID-19 patients with NAFLD. Notably, after adjusting for potential confounders, we identified some new associations with altered immune responses in COVID-19 patients with NAFLD: (1) Advanced age (≥ 60 years); (2) Obesity (BMI ≥ 28 kg/m²); (3) IFG; (4) Liver fibrosis; (5) Elevated liver enzymes; (6) Elevated LDL-C levels; and (7) Increased HbA1c. Importantly, patients with uncomplicated NAFLD (including lean NAFLD) were also identified as requiring particular clinical attention to mitigate risks of immune imbalance.

CONCLUSION

Dynamic changes in lymphocyte subsets significantly influence disease progression and prognosis in COVID-19 patients with NAFLD. Therefore, enhanced clinical management should be prioritized for high-risk groups, particularly individuals with obesity (BMI ≥ 28 kg/m²), advanced age (≥ 60 years), multiple comorbidities (≥ 2 concurrent conditions), hepatic insufficiency, or significant liver fibrosis/cirrhosis. Comprehensive management strategies must also target metabolic markers including LDL-C, HbA1c, and FPG.

Notably, patients with uncomplicated NAFLD (including lean NAFLD) also require clinical attention to mitigate the risk of immune dysregulation. Furthermore, discharge criteria should incorporate the stabilization of chronic disease parameters and lymphocyte subset profiles. Integrating immunomodulatory therapies with targeted control of metabolic immunosuppressive risk factors will optimize comprehensive clinical management.

ACKNOWLEDGEMENTS

We thank Xiu Li (Rehabilitation Division, the Public and Health Clinic Center of Chengdu) for providing assistance during the course of the study.