Longitudinal Evaluation of Metabolic Benefits of Inactivated COVID-19 Vaccination in Diabetic Patients in Tianjin, China

Introduction

Diabetes mellitus poses a significant and rising global health challenge, with its prevalence consistently increasing in both developed and developing countries. By 2045, it is projected that 783.2 million people worldwide will be affected [1,2]. This chronic metabolic disorder is intimately linked with a spectrum of severe complications, including cardiovascular diseases, renal impairment, neuropathies, retinopathies, and immune dysfunctions. These comorbidities collectively contribute to heightened morbidity and mortality rates among diabetic populations [3]. The intricate interplay between hyperglycemia and systemic complications underscores the imperative for comprehensive disease management strategies aimed at mitigating long-term health risks and improving patient outcomes.

The onset of the Coronavirus Disease 2019 (COVID-19) pandemic has further accentuated the vulnerability of individuals with diabetes. Epidemiological studies have consistently identified diabetic patients as a particularly susceptible group, not only exhibiting an increased risk of SARS-CoV-2 infection but also demonstrating a propensity for more severe clinical manifestations and higher mortality rates [4,5]. The compromised immune response associated with diabetes, coupled with chronic inflammation and coagulation abnormalities, exacerbates the severity of viral infections [6,7]. Several meta-analyses have demonstrated that diabetes is responsible for nearly 30% of COVID-19 deaths [8,9] and raises the risk of severe outcomes by 33% [10].This scenario has highlighted the critical necessity for effective glycemic control and immune system optimization in diabetic patients, especially in the face of pervasive viral exposures.

The advent of COVID-19 vaccines has been a pivotal milestone in efforts to control the spread of the virus, offering a promising avenue to reduce transmission rates, hospitalizations, and deaths. COVID-19 vaccines are developed using various platforms, including mRNA, inactivated, viral vector, and protein subunit vaccines. Among these, mRNA and inactivated vaccines are the most widely used globally. mRNA vaccines use synthetic genetic material for immune activation, while inactivated vaccines rely on killed virus particles. While mRNA vaccines offer strong immune responses, they may pose challenges for individuals with metabolic disorders like diabetes. In contrast, inactivated vaccines provide a safer and more stable option for such populations. This study focuses on the effects of inactivated vaccines in diabetic patients. These vaccines have been developed and deployed at an unprecedented pace, with extensive clinical trials validating their safety and efficacy in the general population [11,12]. However, concerns have emerged regarding their application in specific cohorts, particularly among individuals with diabetes [13–15]. Given the unique metabolic and immunological challenges inherent in diabetes mellitus, there is apprehension that vaccination could potentially disrupt glycemic control, trigger inflammatory responses, or exacerbate existing metabolic dysfunctions [16]. Such concerns are compounded by the lack of exhaustive data specifically addressing the vaccine’s impact on diabetic patients, leading to vaccine hesitancy within this group despite the clear public health benefits.

This hesitancy is not unfounded because the interplay between vaccines and metabolic diseases is complex and not yet fully understood. While existing research has provided assurances about the overall safety and efficacy of COVID-19 vaccines, the heterogeneity of diabetes necessitates a more nuanced understanding of how these factors may influence vaccine response and potential side effects [17]. Such research is essential not only for informing clinical guidelines and vaccination strategies but also for enhancing vaccine confidence among diabetic patients, thereby improving vaccination uptake rates and ultimately contributing to better health outcomes.

This study aims to provide a thorough evaluation of the effects of COVID-19 vaccination on patients with diabetes mellitus, addressing the unique concerns associated with this high-risk population. Recognizing that diabetic individuals face increased vulnerability to infections and may experience distinct physiological responses to vaccines, the research meticulously examines changes in key clinical parameters to assess both immediate and long-term impacts of vaccination. These parameters include glycemic control markers; immune function indicators; renal and liver function assessments and lipid profiles. By analyzing these parameters at baseline, 3 months, and 12 months after vaccination, and by comparing unvaccinated diabetic patients with those who received primary and booster vaccinations, the study aimed to elucidate the differential impacts of various immunization strategies on metabolic control, immune response and organ function in diabetic individuals. The objective was to assess both the immediate and sustained effects of vaccination across different immunization regimens, thereby providing a comprehensive understanding of how COVID-19 vaccines interact with the complex pathophysiology of diabetes.

The findings from this study offer important insights into the safety and efficacy of COVID-19 vaccines in diabetic patients, providing a solid evidence base to support vaccination in this high-risk population. In doing so, we hope to enhance vaccine confidence among diabetic individuals and contribute to the optimization of vaccination strategies, ultimately improving health outcomes for this vulnerable group.

Material and Methods

PARTICIPANTS AND GROUP CLASSIFICATION:

The study included a total of 548 individuals with a confirmed diagnosis of diabetes mellitus, who were under regular clinical care at Chu Hsien-I Memorial Hospital in Tianjin, China. Baseline data collection commenced in January 2021, and all participants were without any newly diagnosed complications or significant alterations in their therapeutic regimens. The inclusion criteria required participants to have stable glycemic control and no recent changes in medication that could affect metabolic parameters. Exclusion criteria encompassed acute illness, recent hospitalization, and any interventions that might influence the study outcomes. Participants’ demographics, including age, sex, and duration of diabetes were recorded to assess potential confounding variables. The participants were aged 20–84 years, with a male-to-female ratio of 325: 223.

Participants were systematically categorized into 3 distinct groups based on their COVID-19 vaccination status during the study timeframe. The vaccine is an inactivated type, with the only difference being the manufacturer: Sinopharm (Vero Cell) and Sinovac (Vero Cell). The unvaccinated group (Group A) comprised individuals who did not receive any doses of a COVID-19 vaccine throughout the study. The primary immunization group (Group B) included participants who received 2 doses of a COVID-19 vaccine between February 1, 2021, and November 30, 2021, constituting the complete primary vaccination series. The booster immunization group (Group C) consisted of individuals who completed the 2-dose primary vaccination series and subsequently received an additional booster dose within the same period. The choice of using Sinopharm and Sinovac vaccines was based on their widespread availability and approval by the Chinese regulatory authorities during the study period, ensuring a representative sample of the population at that time. During the follow-up phase from November 30, 2021, to December 30, 2022, no additional COVID-19 vaccine doses were administered to participants in the vaccinated groups, ensuring that the immunization status remained consistent for the duration of the study. Participants in the unvaccinated group remained vaccine-free throughout this period. This stratification allowed for a clear comparison of clinical outcomes based on vaccination status, minimizing confounding variables related to changes in immunization. We aimed for real-world representativeness in participant selection. Although there may have been demographic differences between groups, these were controlled for in the analysis through statistical adjustments, including factors such as age and occupation. The primary endpoints of this study were changes in glycemic control (as measured by HbA1c and fasting glucose). Secondary endpoints included the evaluation of inflammatory markers, liver function tests, renal function indicators and lipid profile. These endpoints were chosen based on their clinical relevance to diabetes management and their potential to provide insights into the broader effects of COVID-19 vaccination on metabolic health.

All participants provided informed consent prior to enrollment, and the study protocol was reviewed and approved by the Institutional Review Board of Chu Hsien-I Memorial Hospital on December 27^th, 2020. The research was conducted in accordance with the ethics principles outlined in the Declaration of Helsinki and adhered to all relevant guidelines for human subject research. Data confidentiality and participant anonymity were strictly maintained throughout the study, with all personal identifiers removed prior to analysis.

FOLLOW-UP TIME POINTS:

The study design incorporated 3 critical follow-up time points to systematically assess the effects of COVID-19 vaccination on individuals with diabetes mellitus. These time points were strategically selected to capture both immediate and long-term physiological changes associated with vaccination. Data collected at each time point included measurements of glycemic control (HbA1c and fasting glucose), inflammatory markers, liver function tests, renal function indicators, and lipid profiles.

The first time point, designated as the baseline, was established in January 2021. This baseline data set reflects the participants’ health status prior to any COVID-19 vaccination and serves as a foundational reference for subsequent comparisons. For vaccinated participants, the baseline measurements represent pre-vaccination health indicators, providing a clear benchmark against which post-vaccination changes could be evaluated. In the unvaccinated group, the baseline data offer comparable initial measurements, ensuring that analyses could accurately attribute any observed differences to factors other than vaccination status.

The second follow-up time point occurred 3 months after vaccination and was intended to capture the short-term effects of vaccination, particularly those associated with the acute immune response. For vaccinated participants, this time point included laboratory tests performed within 3 months following their final vaccine dose including inflammatory markers, and metabolic parameters. To maintain consistency in the timing of data collection across all groups, December 2021 was designated as the corresponding time point for the unvaccinated participants. This alignment minimized potential confounding variables in health indicators.

The third and final follow-up time point was set at 12 months after vaccination to assess long-term physiological changes and the sustained effects of vaccination over an extended period. For the vaccinated groups, data were collected 1 year after the administration of their final vaccine dose. For unvaccinated participants, the equivalent data were collected during the same period in December 2022, ensuring temporal consistency across all groups. Missing data or participant dropouts were handled carefully. Any participants who did not have at least 1 laboratory measurement at baseline were excluded from the analysis. This approach preserved the integrity of the analysis.

LABORATORY ASSESSMENTS:

The study assessed a range of clinical outcomes, including glycemic control, immune function, renal and hepatic function, and lipid profile using relevant biomarkers. Glycemic control, was evaluated using fasting plasma glucose (FPG) measured by the hexokinase endpoint method and glycated hemoglobin (HbA1c) assessed by high-performance liquid chromatography (HPLC). The dual assessment enables a robust evaluation of both acute and chronic glycemic status, important for preventing diabetes-related complications [18]. Hemoglobin concentration (Hb), measured by the SLS-hemoglobin method, was evaluated to assess oxygen-carrying capacity and detect anemia, commonly seen in diabetic patients due to factors like renal insufficiency and chronic inflammation [19,20]. The impact of hemoglobin content on HbA1c measurements was therefore considered negligible for this analysis.

Platelet levels (PLT) were measured by the hydraulic focal method to assess thrombotic risk, particularly relevant for diabetic COVID-19 patients [21]. Differential leukocyte counts (NEU, LYM, EOS) were assessed by semiconductor laser flow cytometry, as these markers reflect immune response and infection status [22,23].

Renal function was assessed by serum creatinine (Cr) using the enzyme endpoint method, and the estimated glomerular filtration rate (eGFR) was calculated using the CKD-EPI formula, taking into account factors such as serum creatinine, age, sex, and race to provide a more accurate estimate of kidney function [24]. Uric acid (UA) levels were measured by the uricase colorimetric method, as hyperuricemia can contribute to renal dysfunction through oxidative stress and endothelial dysfunction [25,26].

Hepatic function was assessed by liver enzymes AST and ALT using the IFCC rate method, which indicate hepatocellular integrity, with elevated levels suggesting liver injury or inflammation, which is common in diabetic patients [27]. ALP uses the IFCC rate method for cholestasis, and GGT as a marker of oxidative stress and hepatic injury [28]. Bilirubin levels (TBIL and DBIL) were assessed using the diazo method (DPD).

The lipid profile was assessed by measuring TG, TC, HDL-C, and LDL-C using appropriate enzymatic methods, including GPO-POD for TG and TC, and direct enzymatic methods for HDL-C and LDL-C [29].

To ensure consistency across the study, the same laboratory and equipment were used throughout, with strict calibration and quality control measures applied. All biomarkers were measured in standard units (eg, mmol/L, listed in Table 1) as per clinical practice guidelines. Relevant references for the methods used were included to ensure reproducibility and verification.

STATISTICAL ANALYSIS:

All statistical analyses were conducted using R software (version 4.2.2), with the following libraries: “lme4” and “lmerTest” for linear mixed-effects models, and “dplyr” for data manipulation. The normality of continuous variables was assessed using the Kolmogorov-Smirnov test. Continuous variables are summarized as mean±standard deviation (SD) for normally distributed variables and as median (interquartile range, IQR) for non-normally distributed variables. Categorical variables are presented as frequencies (n) and percentages (%). The choice of statistical tests was based on the distribution and scale of the data. Variables with normal distribution (WBC and UA) were compared across groups using one-way ANOVA, as this test is appropriate for comparing means across multiple groups. Non-normally distributed variables were compared using the Kruskal-Wallis test, which is a non-parametric method that compares median differences between groups without assuming normality [30]. For categorical variables, such as sex, differences between groups were assessed using the chi-squared test.

Boxplots were generated to show the distribution of key clinical markers across the 3 vaccination groups at baseline, shown in Figure 1, highlighting differences among the unvaccinated, primary immunization, and booster groups using.

To assess the effects of time (“Times”) and vaccine type (“VACTYPE”) on 22 clinical indicators, linear mixed-effects models (LMMs) were used for each indicator [31]. These models were adjusted for age and sex to minimize confounding effects. The interaction between VACTYPE and Times was included to test whether temporal changes varied across vaccination groups. The models accounted for both fixed effects (age, sex, VACTYPE, and Times) and random effects (individual variability), using Restricted Maximum Likelihood (REML) for more accurate variance estimates, especially in unbalanced or missing data. Degrees of freedom for fixed effects were estimated using Satterthwaite’s approximation, improving statistical precision.

The LMMs were specified as follows: ΔY~Age+Sex+VACTYPE*Times+(1|ID).

In this equation: ΔY represents the change in each clinical indicator from baseline. Age and Sex are fixed effects. VACTYPE denotes the vaccination type (unvaccinated, primary immunization, booster immunization). Times indicates the time points (baseline,3 months, 12 months). VACTYPE*Times is the main and interaction term. (1|ID) denotes random intercepts for each participant.

This model structure allows for examining both main and interaction effects, showing how vaccination status affects clinical indicators over time. The “emmeans” package in R was used for pairwise comparisons, considering the model’s fixed and random effects structure. Three comparisons were made between vaccination groups. All statistical analyses were two-sided, ensuring an unbiased test for both increases and decreases in clinical indicators.

SENSITIVITY ANALYSIS:

A sensitivity analysis was conducted to assess the robustness of the linear mixed-effects models (LMMs) applied to the 22 biochemical outcomes in this study.

Two distinct models were specified and fitted for each biochemical outcome to facilitate this evaluation. Model 1 was formulated using the following equation:

Model 2 extended Model 1 by incorporating additional interaction terms between age and time, and sex and time, to capture potential differential temporal effects of these demographic factors:

Both models were fitted using restricted maximum likelihood (REML) estimation, a method providing unbiased of variance estimates, particularly for complex random effects structures. REML improves the precision of fixed effect estimates by accounting for the loss of degrees of freedom due to random effects.

The comparison between Model 1 and Model 2 was conducted across all 22 biochemical outcomes using these statistical criteria: Akaike Information Criterion (AIC): Lower values indicate better model fit. Bayesian Information Criterion (BIC): Penalizes models with more parameters, favoring simpler models. Log-Likelihood Values: Higher log-likelihood values indicate a better model fit.

Results

BASELINE CHARACTERISTICS AND COMPARATIVE ANALYSIS OF CLINICAL MARKERS AMONG DIFFERENT VACCINATION GROUPS:

In Table 1, baseline characteristics of the 548 participants were compared, stratified by COVID-19 vaccination status: unvaccinated (Group A), primary immunization (Group B), and booster immunization (Group C). Comparisons are based on available data for each variable, not strictly predetermined groups. No significant differences in age (median age: 60 years in Group A, 62 years in Group B, 60 years in Group C; p=0.603), sex (p=0.547), or glycemic control (FPG: 7 mmol/L in all groups, HbA1c: 6.9% in Group A, 6.7% in Groups B and C; p=0.391) were observed.

Hematological markers, including WBC, Hb, and PLT, were similar across all groups, and WBC and Hb were not significantly different in immune function or anemia prevalence.

As for renal function, eGFR, creatinine and UA levels were similar across the groups, with no statistically significant differences detected. This suggests that renal function was consistent among the participants, minimizing the potential impact of renal variability on the study’s findings.

Liver function markers were also assessed, given their importance in metabolism and detoxification processes. The enzymes levels demonstrated differences among the groups (AST p<0.01; ALT p<0.05). Figure 1 illustrates the differences in key clinical markers among every 2 groups at baseline, providing insights into data distribution and variability. We observed that in baseline AST levels among the 3 groups, Group C exhibited higher values than Group B; whereas for ALT, Group C elevated levels compared to Group A. However, since these assessments were made prior to vaccination, it is important to note that the observed differences are not indicative of the effects of vaccination itself.

In addition to the previously analyzed clinical parameters, lipid profiles were scrutinized to assess any potential variations associated with COVID-19 vaccination status among the diabetic patient cohort. The lipid parameters evaluated included TC, TG, HDL-C, and LDL-C. Statistical analyses revealed that there were no significant differences in them among the 3 groups(Table 1). The differences between each 2 groups are shown in Figure 1.

These findings suggest that key parameters showed less differences between the groups, providing a robust foundation for longitudinal analysis of the effects of COVID-19 vaccination in diabetic patients.

LONGITUDINAL ANALYSIS OF CLINICAL INDICATORS USING LINEAR MIXED-EFFECTS MODELS:

A comprehensive longitudinal analysis was conducted utilizing linear mixed-effects models to evaluate the temporal changes in 22 clinical indicators among the different COVID-19 vaccination groups. The detailed results of this analysis are presented in Table 2, providing critical insights into metabolic parameters, liver function tests, and lipid profiles, thereby elucidating the potential long-term effects of vaccination in patients with diabetes mellitus. The baseline characteristics analysis is a cross-sectional comparison, while the LMM evaluates changes over time and adjusts for both vaccination type and temporal factors.

The analysis of FPG and HbA1c revealed significant associations with vaccination type and time. Older age was correlated with lower levels of both FPG (Estimate=−0.019, p<0.001) and HbA1c (Estimate=−0.011, p<0.01). For VACTYPE B (group B), the interaction with time showed a significant decrease in FPG (Estimate=−0.118, p<0.001) and HbA1c (Estimate=−0.040, p<0.05). VACTYPE C (group C) exhibited significant reductions in both FPG (Estimate=−0.123, p<0.001) and HbA1c (Estimate=−0.049, p<0.01) over time. These results indicate that groups B and C are associated with notable decreases in blood glucose metrics.

In the analysis of immune indicators, we assessed the effects of different vaccination types (VACTYPE) and time (Times) on WBC, PLT, Hb, NEU, and LYM. The results indicated that age and sex may have an impact on them. Sex-related factors influence these indicators, for example, females generally exhibit lower hemoglobin levels physiologically [32]. Additionally, the impact of different vaccination types on immune indicators did not reach statistical significance. These findings suggest that the effects of different vaccination types on these indicators appear minimal.

As for renal function indicators, we primarily focused on creatinine (Cr), estimated glomerular filtration rate (eGFR), and uric acid (UA). The results revealed that both age and sex significantly influenced these parameters. Older age was linked to increased creatinine levels and decreased eGFR, while female sex was associated with lower creatinine and uric acid levels but slightly higher eGFR compared to males. Vaccination groups B(Estimate=−0.604, p<0.01) and C (Estimate=−0.481, p<0.05) exhibited a gradual decline in creatinine over time, though vaccine type showed no notable effect on eGFR. These results emphasize that age and sex are key factors influencing renal function indicators, while the short-term effects of different vaccine types appear minimal. While vaccines do affect the immune system, their direct impact on kidney function may be minimal.

In examining the liver function indicators, the results showed that the vaccine types did not have a significant effect on AST, ALT, GGT, and ALP, and the trend differences among these vaccine types for these indicators were minimal. Furthermore, the interaction between vaccine type and time also did not show significance, indicating that the impact of vaccine types on liver function indicators remained relatively stable over time. TBIL in group C demonstrated a significant upward trend (Estimate: 0.272, p<0.01), and both DBIL and IBIL also showed significant differences (DBIL compared to Group A: Estimate: 0.080, p<0.05; IBIL compared to Group A: Estimate: 0.205, p<0.001). This change is visually depicted in Figure 2, may be related to an enhanced immune response to the vaccine after booster immunization, leading to an increased burden on the liver for bilirubin metabolism.

In terms of lipid metabolism, such as TC, TG, HDL-C, LDL-C, vaccination did not show a significant impact. Age and sex remained important factor influencing lipid profile levels.

The linear mixed-effects model analysis results indicate that COVID-19 vaccination has significant long-term effects on key metabolic, liver function, and lipid profile indicators in patients with diabetes mellitus. Figures 2 and 3 effectively illustrate the longitudinal trends of these clinical markers across different vaccination groups, highlighting the potential modulatory effects of the vaccine on specific physiological pathways. Benefits in glucose suggest that booster vaccination may potentially enhancing glycemic control.

EVALUATION OF MODEL PERFORMANCE ACROSS BIOCHEMICAL OUTCOMES:

This study compared the performance of 2 statistical models(Model 1 and Model 2) in capturing the longitudinal changes of various clinical indicators. The results, summarized in Table 3, showed significant variability between the models for different clinical markers.

For most biochemical variables, including FPG, WBC, PLT, Hb, NEU, LYM, EOS, EGFR, UA, ALT, GGT, IBIL, TC, TG, HDL-C, LDL-C, Model 1 consistently demonstrated superior performance over Model 2 with lower values of the AIC and BIC, indicating a more optimal balance between model fit and complexity. This suggests that including fundamental covariates like Age, Sex, and the interaction term between vaccination type and time (VACTYPE*Times), provided a more parsimonious model without unnecessary complexity. The absence of additional interaction terms likely reduced the risk of overfitting and enhancing its generalizability.

In contrast, Model 2 showed significantly improved fit for specific biomarkers, including HbA1c and DBIL. The enhanced performance of Model 2 for these markers likely stems from its ability to capture complex demographic-time interactions. For other variables, differences between the 2 models were negligible, despite AIC and Log-Likelihood suggesting a slight advantage for Model 2, the simpler structure of Model 1, favored by BIC, made it the preferred choice.

Discussion

This study systematically evaluated the impact of COVID-19 vaccination on metabolic, immune, hepatic, and renal functions in diabetics patients, focusing on both short-term and long-term changes following vaccination. In the 3-month follow-up, there were decreasing changes in FPG and HbA1c levels across both primary and booster vaccinations, indicating that vaccination did not adversely affect short-term glycemic control. At 12-month, more pronounced differences emerged among the 3 groups. This period saw increased virus exposure, which allowed for assessment of the long-term effects of vaccination. Unvaccinated diabetic patients are at higher risk of severe symptoms and associated with severe infection can lead to decreased insulin sensitivity, impaired pancreatic β-cell function, and subsequent deterioration of glycemic control [33–35]. This supports the importance of vaccination for this group. Group C received booster vaccinations, demonstrated significant improvements in glycemic control, with decreased FPG and HbA1c levels. The booster may reduce systemic inflammation, viral load and metabolic stress, all of which can positively influence insulin sensitivity and glucose metabolism [36]. These findings highlight the benefit of boosters in maintaining long-term metabolic stability. Group B(primary vaccination) exhibited stable glucose levels over 12 months. However, primary vaccination still offered protection, stabilizing glucose levels after infection. This suggests that while primary vaccination helps, booster doses are necessary to maintain optimal metabolic control as immunity declines and new viral variants emerge.

This study systematically evaluated the effects of COVID-19 vaccination in diabetic patients. However, some studies have highlighted potential risks in vulnerable populations. For instance, Pirzadeh et al reported that mRNA vaccination exacerbated minimal change disease (MCD) in a patient [37], while Shahrudin et al observed herpes zoster in diabetic patients after vaccination [38]. These findings underscore the need for careful monitoring in such groups.

In this study, several clinical parameters exhibited no significant changes among diabetic patients after COVID-19 vaccination, indicating a stable clinical profile. Immune function markers showed no significant differences between the vaccinated and unvaccinated groups, suggesting that vaccination does not adversely affect immune responses in diabetic patients. Hb levels were influenced by sex and age, with Hb levels in females and a slight decline with age, but no significant changes between different vaccination groups. This indicates that vaccination does not impact hematological parameters related to oxygen-carrying capacity in diabetic individuals. These results suggest that vaccination does not affect immune and hematological systems, confirming its safety. The absence of significant alterations in immune markers supports the idea. This aligns with previous studies indicating that diabetic patients can generate immune responses following COVID-19 vaccination, thereby reinforcing the vaccine’s immunogenic efficacy in diabetic populations [40]. Such findings are crucial as they provide evidence-based reassurance that vaccination does not exacerbate underlying immunological vulnerabilities in diabetic patients.

No significant differences in renal function markers eGFR were observed between the vaccinated and unvaccinated groups, while creatinine levels decreased. This suggests that COVID-19 vaccination exerts a minimal effect on renal function in diabetic patients and does not significantly increase the risk of renal impairment. Given the susceptibility of diabetic patients to nephropathy, post-vaccination Cr and eGFR values indicate that the vaccine does not cause renal stress or damage, supporting its safety. Age and sex significantly influence renal function markers. With advancing age, Cr levels increased while eGFR values declined, reflecting the natural age-related decline in renal function. Female patients demonstrated higher eGFR levels, suggesting a potential advantage in maintaining renal function, which may be influenced by hormonal differences, lower muscle mass, and varying exposure to nephrotoxic factors between males and females [41,42].

The study also evaluated hepatic function markers, including AST, ALT, GGT, ALP, and bilirubin, to investigate the impact of COVID-19 vaccination on diabetic patients’ liver function. AST and ALT levels showed no significant differences among the vaccine groups or over time, indicating no hepatic dysfunction. GGT and ALP levels also remained stable, supporting the safety of COVID-19 vaccines for liver health in diabetic patients. These results align with existing literature suggesting that vaccination does not compromise liver function, thereby reinforcing the vaccine’s safety profile in terms of hepatic health [43,44].

However, bilirubin levels revealed notable findings. Total bilirubin increased significantly in Group C. The interaction between vaccine type and time suggesting that booster vaccination may have a more substantial impact on bilirubin metabolism. Direct bilirubin and indirect bilirubin also increased in Group C with time, which may reflect mild effects on bile secretion or conjugation, further supporting the idea that vaccines can mildly affect bilirubin metabolism in diabetic patients. Although these changes did not reach pathological levels or result in clinical symptoms such as jaundice, they suggest the presence of a mild metabolic burden on the liver. In diabetic patients, even small metabolic stressors can elicit measurable changes. Nevertheless, the lack of significant liver enzymes changes and clinical manifestations suggest that fluctuations are benign. These findings highlight the importance of monitoring hepatic parameters in at-risk populations after vaccination, and further research is needed.

In the analysis of lipid profiles, most parameters showed no significant differences among the 3 vaccine groups. Specifically, TC, TG, HDL-C, and LDL-C levels exhibited similar trends across all groups, with no significant differences detected. This indicates that vaccination status had no significant impact on these lipid parameters, and COVID-19 vaccines did not adversely affect lipid metabolism in diabetic patients. Maintaining stable lipid levels is crucial in this population due to their increased risk of cardiovascular disease.

This study possesses several strengths. It had a large sample size, including 3 vaccination groups – Group A (unvaccinated), Group B (primary vaccination), and Group C (booster vaccination) – with follow-up at 3 and 12 months. This strategy enhances data reliability and broadens the findings applicability. By integrating both short-term and long-term follow-ups, the study elucidates the temporal effects of COVID-19 vaccination on glycemic control in diabetic patients. Furthermore, it evaluated a wide array of clinical parameters including blood glucose levels, renal function, hepatic function, lipid profiles, and immune function markers, offering a comprehensive overview of the vaccination’s effects on diabetic patients. This approach yields valuable data for clinical applications and diabetes management, focusing on immune responses and metabolic alteration. In addition, the comparative analysis of the 3 groups, especially concerning blood glucose and bilirubin, shows the potential impacts of booster vaccinations. These findings support optimizing vaccination strategies, which could inform public health policies for diabetic patients. The study’s relevance is further enhanced by its timing, assessing vaccine efficacy and glycemic control during a critical phase of the pandemic.

However, several limitations exist. As an observational study, we could not entirely eliminate potential confounding factors. Although sex and age were adjusted for, unmeasured confounders may still have affected the results. The study was confined to diabetic patients from a specific geographic region, which may restrict the generalizability of the findings. Additionally, the focus on specific COVID-19 vaccines limits the broader applicability to other vaccine formulations. Moreover, changes in participants’ behaviors such as diet, physical activity, and medication adherence were not accounted for, which may have influenced metabolic outcomes. The evolving nature of the pandemic and varying public health measures add complexity to interpreting the findings. While the 12-month follow-up provides valuable data, longer-term observation is needed to assess the ongoing safety and efficacy of vaccines in diabetic patients and detect delayed effects. Finally, although most participants contracted COVID-19 during the follow-up, the study did not thoroughly evaluate the severity of infections or the specific impact on the results, limiting the analysis of the interaction between vaccination and infection outcomes.

Based on the findings of this study, we offer several suggestions for future research. Multi-center studies are essential to validate the effects of COVID-19 vaccines across different ethnicities, age groups, and demographic profiles. Additionally, long-term follow-up studies are necessary to assess the sustained safety and efficacy of vaccines in individuals with diabetes mellitus. These studies should aim to identify any potential delayed complications or benefits that may not be apparent within a 12-month timeframe. Finally, future research should thoroughly evaluate the severity of COVID-19 infections among study participants and their impact on metabolic parameters to enhance the understanding of the interplay between vaccination, infection severity, and metabolic control in diabetic patients.

Conclusions

This study systematically evaluated the short-term and long-term effects of COVID-19 vaccination on metabolic, immune, hepatic, and renal functions in diabetic patients. The findings demonstrated significant decreases in HbA1c and FPG levels over time within the vaccinated groups, suggesting potential protective effects on glycemic control. Additionally, creatinine and bilirubin levels exhibited changes within physiological ranges over time following vaccination. These results underscore the safety and potential metabolic benefits of COVID-19 booster vaccination in diabetic patients, providing valuable insights for clinical management and public health strategies. Future research should focus on larger, multi-center studies to validate these findings and explore the underlying mechanisms behind the observed metabolic improvements following vaccination.