Связь между 25-hydroxyvitamin D в виде жидкой хроматографии-тандем масс-спектрометрии и ретинопатии по типу 2 Болит диабет.

1. Introduction

Diabetes mellitus is becoming increasingly prevalent globally. In 2021, approximately 537 million adults had diabetes, and this number is expected to increase to 783 million by 2045. [ 1 , 2 ] Incomplete statistical information suggests that approximately 34.6% of patients with diabetes worldwide have diabetic retinopathy (DR), 27% to 30% of which progresses to DR. [ 3 ] DR has also been reported in patients with prediabetes, [ 4 ] suggesting that early glucose metabolism disorders may trigger retinal damage. Diabetic retinopathy is the leading cause of irreversible blindness in young and middle-aged people, and the typical pathological manifestations include diabetic macular edema and proliferative DR, which may directly cause vision loss and ultimately blindness if left untreated. [ 5 ] DR is a microvascular complication and involves damage to the neurovascular unit (e.g., corneal neuropathy), pyroptosis, oxidative stress, and metabolic disorders induced by chronic hyperglycemia. [ 6 , 7 ] Early DR is often asymptomatic and the disease may be in advanced stages by the time of diagnosis; however, existing treatments have limited effect on advanced diabetic retinopathy. [ 8 ] Therefore, early screening (e.g., through retinal vessel density monitoring or metabolomic biomarkers) and targeted therapies are urgently required to reduce the risk of blindness in patients with diabetic retinopathy.

25(OH)D is the main circulating form of vitamin D hydroxylated in the liver via enzymes such as CYP2R1 and presents in 2 forms: 25(OH)D 3 (ergocalciferol) and 25(OH)D 2 (cholecalciferol). [ 9 ] It is the clinical gold standard for assessing the nutritional level of vitamin D owing to its long half-life (approximately 2–3 weeks) and stable serum concentrations. [ 10 ] The relationship between 25(OH)D and DR is somewhat controversial in current studies, with most evidence suggesting the presence of an relation. [ 11 , 12 ] Nevertheless, a few studies have shown that 25(OH)D levels in patients with diabetes, despite being lower than those in patients without diabetes, are not significantly correlated with the onset of DR. [ 13 ] Mendelian randomization studies have not confirmed the causal relationship between 25(OH)D and DR, suggesting that observational studies may be limited by confounding factors. [ 14 ]

Vitamin D testing presents unique challenges: low serum concentrations (at the nmol/L level), the presence of highly structurally similar isomers (e.g., vitamin D 2 and D 3 ), and multiple metabolites (e.g., 25(OH)D, 1,25(OH) 2 D), with stability being susceptible to light and temperature effects. LC-MS/MS has emerged as the gold standard in clinical and research settings for vitamin D determination due to its high specificity, sensitivity, and capability to analyze multiple metabolites simultaneously. [ 15 ] This method is particularly valuable for low-concentration samples, special populations, or metabolite detection, enabling comprehensive evaluation of vitamin D metabolic status and providing standardized reference and quality control for routine clinical testing. [ 16 ]

To provide new ideas for the prevention and treatment of DR, in this study, the relation between 25(OH)D and DR was investigated by observing 25(OH)D levels in patients with T2DM.

2. Research objects and methods

2.1. Research objects

Patients with T2DM who were admitted to the Department of Endocrinology and Metabolism II of Zhangzhou Hospital, Fujian Province, China between January 2022 and November 2022 were included in the study. The exclusion criteria were acute complications of diabetes mellitus or severe diabetic foot, acute, or chronic inflammatory diseases; other types of diabetes mellitus: type 1 diabetes mellitus; special types of diabetes mellitus; gestational diabetes mellitus; the presence of serious primary diseases in the respiratory, circulatory, or hematopoietic system etc, mental health disorders; and patients with incomplete data. In total, 206 patients were included in the study. This study was reviewed and approved by the institutional review board of the hospital (Number:2025LWB338).

2.2. Research methods

Determination of serum 25(OH)D levels: LC-MS/MS was performed to determine serum 25(OH)D levels. Specifically, 25(OH)D2-deuterium 3 and 25(OH)D3-deuterium 3 were used as internal standards, and the tests were performed in strict accordance with the instructions of the TRIPLE QUAD 4 500 MD (AB SCIEX Company) instrument and 25-hydroxyvitamin D2/D3 Assay Kit (Qingdao Healicon Biotechnology Co. Ltd.).

General data collection: The clinical data of the study participants, such as sex, age, height, weight, waist circumference, diabetes complications (including peripheral arterial disease (PAD), diabetic kidney disease (DKD), and diabetic autonomic neuropathy (DAN)), and history of hypertension, were collected. The body mass index (BMI) was calculated.

Testing of clinical indicators: Peripheral venous blood samples were collected by medical professionals during the resting and fasting state in the early morning of the following day to measure the levels of HbA1c, fasting C-peptide, fasting plasma glucose (FPG), 2-hour postprandial glucose (2hpG), total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol(HDL-C), low-density lipoprotein cholesterol(LDL-C), blood urea nitrogen (BUN), Scr, uric acid(UA), P, Ca, PTH, FT3, free thyroxine 4 (FT4), thyroid-stimulating hormone (TSH), UMA, and Ucr, etc.

Diagnosis of DR: In this study, all subjects underwent fundus examination by ophthalmologists with intermediate qualifications at our hospital. Methods: ① Bilateral fundus color photography was performed using a non-mydriatic fundus camera (Kowa, VX-10α). ② To accurately assess macular structure and detect edema, all patients underwent frequency-domain optical coherence tomography (HEIDELBERG ENGINEERINE, Spectralis OCT) examination. The final diagnosis and severity grading of DR were strictly based on the “DR Preferred Practice Pattern (2019).” [ 17 ]

Group assignment: ① Patients were divided into 3 groups based on serum 25(OH)D levels: group A (normal group, serum 25(OH)D ≥ 30 ng/mL), with 38 patients; group B (insufficiency group, 30 ng/mL >serum 25(OH)D ≥ 20 ng/mL), with 103 patients; and group C (deficiency group, serum 25(OH)D <20 ng/mL), with 65 patients. ② DR was diagnosed via ophthalmology examination, and the 206 study patients were divided into 2 groups: NDR(144 patients) and DR (62 patients).

2.3. Statistical methods

SPSS 27.0 (Chicago) was used for statistical analysis. Normally distributed continuous data are expressed as mean ± standard deviation, and multiple groups were compared using analysis of variance. Continuous data with a skewed distribution are expressed as M (Q1,Q 3 ), and the rank sum test for multiple groups was performed. Categorical data are expressed as n (%), and the chi-square test or Fisher exact probability method was performed. A P -value  < .05 indicated that the difference was statistically significant.

3. Results

3.1. Intergroup analysis among the serum 25(OH)D subgroups

3.1.1. Comparison of clinical and laboratory data between groups with different serum 25(OH)D levels :

Comparison of general data for groups A, B, and C showed that the differences in P, Ca, AST, HbA1c, PTH, diabetic peripheral neuropathy (DPN), and DR among the groups were statistically significant ( P  <.05), while the differences in age, sex, height, weight, waist circumference, BMI, resting heart rate, TG, TC, HDL-C, LDL-C, BUN, UA, Scr, ALT, fasting C-peptide, UMA, Ucr, FT3, FT4, TSH, FPG, 2hPG, PAD, DKD, DAN, and hypertension were not statistically significant ( P  >.05) (Table 1 ).

Table 1 The comparison of clinical and laboratory data among different subgroups of vitamin D levels. Variables Total (n = 206) Group A (n = 38) Group B (n = 103) Group C (n = 65) F / χ 2
P
Age(year) 57.92 ± 11.72 57.24 ± 11.74 57.23 ± 10.67 59.40 ± 13.26 0.76 .470 Sex (%)  F 88 (42.72%) 22 (57.89%) 40 (38.83%) 26 (40.00%) 4.41 .110  M 118 (57.28%) 16 (42.11%) 63 (61.17%) 39 (60.00%) Height (cm) 163.44 ± 7.72 162.84 ± 7.07 164.14 ± 8.02 162.68 ± 7.61 0.85 .430 Weight (kg) 62.70 ± 10.28 62.01 ± 10.17 62.72 ± 10.22 63.07 ± 10.57 0.13 .882 BMI (kg/m 2 ) 23.43 (21.48, 25.19) 23.07 (21.29, 24.72) 23.15(21.48, 24.90) 23.90 (22.20, 25.30) 1.76 .416 Waistline (cm) 85.00 (80.00, 92.00) 85.00 (80.00,87.75) 85.00 (79.00,92.00) 86.00 (80.00,92.00) 1.36 .506 RHR (sub/mi) 84.00 (78.00, 94.00) 85.00 (78.00, 91.00) 83.00 (78.00, 91.00) 86.00 (78.00, 97.00) 0.87 .649 TG (mmol/L) 1.42 (1.04, 2.03) 1.66 (1.26, 2.30) 1.29 (0.97, 1.92) 1.46 (1.10, 1.97) 5.32 .070 TC (mmol/L) 4.80 (3.96, 5.70) 4.84 (3.95,5.68) 4.81 (4.14,5.79) 4.62 (3.54,5.60) 0.99 .611 HDL-C (mmol/L) 1.04 (0.90, 1.25) 1.00 (0.89,1.12) 1.05 (0.91,1.27) 1.040.88,1.23) 3.24 .197 LDL-C (mmol/L) 3.00 (2.27, 3.73) 2.96 (2.34, 3.72) 3.04 (2.41, 3.64) 2.96 (2.03, 3.79) 0.54 .764 BUN (mmol/L) 5.50 (4.50, 6.70) 5.50 (4.50, 7.27) 5.40 (4.70, 6.50) 5.50 (4.50, 7.10) 0.22 .896 UA (mmol/L) 316.10 (251.30, 393.00) 328.40 (246.50, 406.80) 316.10 (251.72, 385.23) 306.40 (251.30, 380.40) 0.34 .844 Scr (mmol/L) 63.80 (51.30, 78.70) 63.75 (54.40,78.72) 65.40 (51.50, 78.00) 61.25 (51.20, 80.08) 0.30 .859 P (mmol/L) 1.30 (1.16, 1.44) 1.46 (1.38,1.58) 1.27 (1.15,1.47) 1.20 (1.11,1.33) 43.47 <.001 * Ca (mmol/L) 2.29 (2.20, 2.37) 2.39 (2.33, 2.45) 2.29 (2.19, 2.37) 2.25 (2.18, 2.30) 35.14 <.001 * ALT (U/L) 20.35 (14.90, 32.65) 25.65 (14.70,43.57) 21.90 (15.80,32.10) 18.00 (14.30,28.40) 4.31 .116 AST (U/L) 22.20 (17.52, 29.57) 26.50 (18.80, 44.10) 22.85 (18.28, 31.55) 21.00 (15.80, 25.10) 9.17 .010 * Fasting C-peptide (ng/mL) 0.86 (0.52, 1.27) 0.95 (0.57, 1.57) 0.83 (0.51, 1.29) 0.85 (0.52, 1.20) 0.93 .629 HbA1c (%) 9.46 (7.73, 11.52) 8.79 (7.44, 9.88) 9.49 (7.77, 11.21) 10.50 (7.81, 12.29) 7.54 .023 * MA (mg/L) 13.90 (11.00, 44.98) 18.10 (11.00,46.60) 11.00 (11.00,24.95) 15.10 (11.00,65.70) 5.11 .078 Ucr (µmol/L) 6937.15 (4283.82, 10093.30) 7845.15 (4103.23, 11196.25) 7155.05 (4368.22, 10442.72) 6555.05 (4127.20, 8477.92) 2.35 .308 PTH (pg/mL) 38.10 (30.55, 51.30) 37.05 (30.45, 47.98) 34.95 (27.08, 44.43) 49.90 (38.80, 60.80) 28.47 <.001 * FT3 (pmol/L) 4.87 (4.39, 5.29) 5.07 (4.47, 5.57) 4.87 (4.43, 5.30) 4.75 (4.29, 5.23) 3.71 .156 FT4 (pmol/L) 11.80 (10.40, 13.60) 11.70 (10.55, 13.15) 11.90 (10.80, 13.50) 11.60 (10.00, 13.90) 0.31 .856 TSH (pmol/L) 1.48 (0.92, 2.25) 1.74 (0.85, 2.82) 1.45 (0.84, 2.01) 1.43 (1.01, 2.50) 1.68 .432 FPG (mmol/L) 8.10 (6.30, 10.20) 8.60 (7.20, 11.40) 7.60 (6.30, 9.88) 7.90 (6.20, 10.30) 3.68 .159 2hPG (mmol/L) 11.20 (8.53, 15.17) 11.10 (8.70, 15.30) 11.85 (8.40, 15.62) 10.40 (8.70, 14.30) 1.07 .586 PAD (%) 57 (28.08%) 11 (28.95%) 29 (28.43%) 17 (26.98%) 0.06 .971 DKD (%) 53 (25.85%) 11 (28.95%) 23 (22.33%) 19 (29.69%) 1.35 .35 DR (%) 62 (30.10%) 6 (15.79%) 26 (25.24%) 30 (46.15%) 12.82 .002 * DPN (%) 108 (52.43%) 7 (18.42%) 47 (45.63%) 54 (83.08%) 44.01 <.001 * DAN (%) 89 (43.20%) 16 (42.11%) 40 (38.83%) 33 (50.77%) 2.34 .311 Hypertension (%) 97 (47.32%) 19 (50.00%) 44 (43.14%) 34 (52.31%) 1.47 .479 2hPG = 2-hour postprandial glucose, ALT = alanine aminotransferase, AST = aspartate aminotransferase, BMI = body mass index, BUN = blood urea nitrogen, Ca = calcium, DAN = diabetic autonomic neuropathy, DKD = diabetic kidney disease, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, FPG = fasting plasma glucose, FT3 = free triiodothyronine, FT4 = free thyroxine, HbA1c = hemoglobin A1c, HDL-C = high-density lipoprotein cholesterol, LDL-C = low-density lipoprotein cholesterol, MA = microalbumin, P = phosphorus, PAD = peripheral artery disease, PTH = parathyroid hormone, RHR = resting heart rate, Scr = serum creatinine, TC = total cholesterol, TG = triglycerides, TSH = thyroid-stimulating hormone, UA = uric acid, Ucr = urinary creatinine. * P  <.05, with a statistically significant difference.

The comparison of clinical and laboratory data among different subgroups of vitamin D levels.

2hPG = 2-hour postprandial glucose, ALT = alanine aminotransferase, AST = aspartate aminotransferase, BMI = body mass index, BUN = blood urea nitrogen, Ca = calcium, DAN = diabetic autonomic neuropathy, DKD = diabetic kidney disease, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, FPG = fasting plasma glucose, FT3 = free triiodothyronine, FT4 = free thyroxine, HbA1c = hemoglobin A1c, HDL-C = high-density lipoprotein cholesterol, LDL-C = low-density lipoprotein cholesterol, MA = microalbumin, P = phosphorus, PAD = peripheral artery disease, PTH = parathyroid hormone, RHR = resting heart rate, Scr = serum creatinine, TC = total cholesterol, TG = triglycerides, TSH = thyroid-stimulating hormone, UA = uric acid, Ucr = urinary creatinine.

P  <.05, with a statistically significant difference.

3.1.2. Ordinal logistic regression analysis of serum 25(OH)D and related factors:

Ordinal logistic regression analysis was performed with different serum 25(OH)D levels as the dependent variable, and factors significantly correlated with different serum 25(OH)D levels in the above analyses as the independent variables. The results showed that P, Ca, AST, HbA1c, PTH, and DPN were independently correlated with serum 25(OH)D levels (Table 2 ).

Table 2 Ordered logistic regression analysis of different vitamin D subgroups. Variables Single factor Multiple factor β S.E
t

P
OR (95% CI) β S.E
t

P
OR (95% CI) DR −1.05 0.30 −3.51 <.001 * 0.35 (0.19–0.63) 0. 04 0.41 0.10 .921 0.94 (0.46–2.35) DPN −2.00 0.32 −6.35 <.001 * 0.13 (0.07–0.25) −2.19 0.43 −5.08 <.001 * 0.11 (0.05–0.26) P 3.23 0.64 5.07 <.001 * 25.30 (7.26–88.12) 3.72 0.84 4.43 <.001 * 41.37 (7.96–215.08) Ca 4.17 1.06 3.95 <.001 * 64.96 (8.18–515.88) 2.25 1.29 1.74 .082 9.45 (0.75–118.33) AST 0.02 0.01 2.50 .013 * 1.02 (1.01–1.03) 0.03 0.01 2.35 .019 * 1.03 (1.01–1.05) HbA1c −0.18 0.06 −3.05 .002 * 0.84 (0.75–0.94) −0.16 0.08 −2.13 .033 * 0.85 (0.73–0.99) PTH −0.02 0.01 −2.60 .009 * 0.98 (0.96–0.99) −0.03 0.01 −2.94 .003 * 0.98 (0.96–0.99) Data are presented as regression coefficients (β), S.E, t -value, and P -value. OR and 95% CI are shown for the multiple factor analysis. * P  <.05 indicates a statistically significant difference. AST = aspartate aminotransferase, Ca = calcium, CI = confidence intervals, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, HbA1c = hemoglobin A1c, OR = odds ratios, P = phosphorus, PTH = parathyroid hormone, S.E. = standard error.

Ordered logistic regression analysis of different vitamin D subgroups.

Data are presented as regression coefficients (β), S.E, t -value, and P -value. OR and 95% CI are shown for the multiple factor analysis. * P  <.05 indicates a statistically significant difference.

AST = aspartate aminotransferase, Ca = calcium, CI = confidence intervals, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, HbA1c = hemoglobin A1c, OR = odds ratios, P = phosphorus, PTH = parathyroid hormone, S.E. = standard error.

3.2. Intergroup analysis between the NDR and DR groups

3.2.1. Comparison of clinical and laboratory data between the NDR and DR groups:

Comparison of general data for the DR and NDR groups showed statistically significant differences between disease time, TG, BUN, Scr, UMA, Ucr, FT3, 25(OH)D3, total 25(OH)D, DPN, DKD, and hypertension ( P  <.05). However, the differences in age, sex, height, weight, BMI, waist circumference, TC, HDL-C, LDL-C, UA, P, Ca, ALT, AST, fasting C-peptide, HbA1c, PTH, FT4, TSH, FPG, 2hPG, PAD, and DAN were not statistically significant ( P  >.05) (Table 3 , Fig. 1 ).

Table 3 The comparison of clinical and laboratory data between NDR and DR. Variables Total (n = 206) NDR (n = 144) DR (n = 62) t/Z/χ 2
P
Age (year) 57.92 ± 11.72 57.34 ± 12.00 59.26 ± 11.04 −1.08 .283 Sex(%)  F 88 (42.72) 60 (41.67) 28 (45.16) 0.22 .642  M 118 (57.28) 84 (58.33) 34 (54.84) Disease time (year) 4.00 (0.83, 10.00) 2.00 (0.39, 7.00) 10.00 (4.25, 12.00) −5.56 <.001 * Height (cm) 163.0 (158.0, 170.0) 164.0 (158.0, 170.0) 161.0 (158.0, 170.0) −1.04 .298 Weight (kg) 62.70 ± 10.28 63.21 ± 10.42 61.52 ± 9.93 1.08 .279 BMI (kg/m 2 ) 23.43 (21.48, 25.19) 23.47 (21.51, 25.02) 23.41 (21.44, 25.23) −0.32 .752 Waistline (cm) 85.74 ± 9.33 85.53 ± 9.26 86.23 ± 9.57 −0.49 .621 RHR (sub/mi) 84.00 (78.00, 94.00) 83.00 (77.75, 91.00) 87.00 (78.25, 98.75) −1.89 .059 TG (mmol/L) 1.42 (1.04, 2.03) 1.48 (1.06, 2.13) 1.21 (1.03, 1.79) −1.83 .068 TC (mmol/L) 4.80 (3.96, 5.70) 4.88 (3.98, 5.62) 4.60(3.78, 5.85) −0.09 .929 HDL-C (mmol/L) 1.04 (0.90, 1.25) 1.04(0.89, 1.25) 1.05 (0.90, 1.25) −0.41 .683 LDL-C (mmol/L) 3.00 (2.27, 3.73) 3.00 (2.32, 3.66) 3.02 (2.26, 4.17) −0.90 .371 BUN (mmol/L) 5.50 (4.50, 6.70) 5.40 (4.30, 6.50) 5.50 (5.03, 7.97) −2.49 .013 * UA (µmol/L) 320.10 ± 104.61 315.19 ± 96.84 331.43 ± 120.75 −1.02 .308 Scr (mmol/L) 63.80 (51.30, 78.70) 63.60 (51.30, 75.12) 64.60(51.50, 99.60) −1.23 .218 P(mmol/L) 1.29 ± 0.23 1.31 ± 0.24 1.25 ± 0.21 1.74 .084 Ca (mmol/L) 2.29 (2.20, 2.37) 2.30 (2.21, 2.37) 2.27 (2.18, 2.37) −1.49 .137 ALT (U/L) 20.35 (14.90, 32.65) 23.00 (15.70, 34.10) 18.80 (14.10, 28.00) −1.94 .052 AST (U/L) 22.20 (17.52, 29.57) 22.90 (17.50, 32.30) 21.90 (17.90, 25.10) −1.38 .169 Fasting C-peptide (ng/mL) 0.86 (0.52, 1.27) 0.93(0.58, 1.31) 0.69 (0.48, 1.12) −1.80 .072 HbA1c (%) 9.46 (7.73, 11.52) 9.47 (7.70, 11.36) 9.45(7.83, 11.59) −0.81 .415 MA (mg/L) 13.90 (11.00, 44.98) 11.00 (11.00, 24.85) 39.55 (14.20, 232.75) −5.44 <.001 * Ucr (µmol/l) 6937.15 (4283.82, 10093.30) 7253.85 (5058.45, 10535.58) 4901.90 (3287.17, 8393.08) −2.73 .006 * PTH (pg/mL) 38.10 (30.55, 51.30) 38.00 (29.90, 50.10) 39.10 (31.35, 53.65) −0.64 .519 FT3 (pmol/L) 4.87 (4.39, 5.29) 4.96 (4.44, 5.37) 4.73 (4.27, 5.21) −2.03 .043 * FT4 (pmol/L) 11.80 (10.40, 13.60) 11.70 (10.43, 13.28) 11.90 (10.40, 14.00) −1.05 .292 TSH (pmol/L) 1.46 (0.92, 2.23) 1.44 (0.96, 2.15) 1.49 (0.84, 2.61) −0.33 .743 FPG (mmol/L) 8.10 (6.30, 10.20) 8.20(6.70, 9.80) 7.40 (5.80, 10.30) −0.94 .347 2hPG (mmol/L) 11.20 (8.53, 15.17) 12.10 (8.80, 15.70) 10.30 (8.40, 13.70) −1.49 .136 25 (OH) D 3 (ng/mL) 22.77 ± 7.60 24.17 ± 7.35 19.54 ± 7.21 4.17 <.001 * Total 25(OH)D (ng/mL) 23.42 ± 7.54 24.84 ± 7.32 20.12 ± 7.03 4.30 <.001 * 25 (OH) D 2 (ng/mL) 0.49 (0.17, 0.91) 0.48 (0.20, 0.84) 0.49 (0.09, 0.97) −0.17 .861 PAD (%) 146 (71.92%) 97 (68.31%) 49 (80.33%) 3.05 .081 DKD(%) 53 (25.85%) 18 (12.50%) 35 (57.38%) 45.02 <.001 * DPN(%) 108 (52.43%) 57 (39.58%) 51 (82.26%) 31.65 <.001 * DAN(%) 89 (43.20%) 63 (43.75%) 26 (41.94%) 0.06 .809 Hypertension(%) 97 (47.32%) 58 (40.56%) 39 (62.90%) 8.66 .003 * 2hPG = 2-hour postprandial glucose, ALT = alanine aminotransferase, AST = aspartate aminotransferase, BMI = body mass index, BUN = blood urea nitrogen, Ca = calcium, DAN = diabetic autonomic neuropathy, DKD = diabetic kidney disease, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, FPG = fasting plasma glucose, FT3 = free triiodothyronine, FT4 = free thyroxine, HbA1c = hemoglobin A1c, HDL-C = high-density lipoprotein cholesterol, LDL-C = low-density lipoprotein cholesterol, MA = microalbumin, NDR = non-diabetic retinopathy, P = phosphorus, PAD = peripheral artery disease, PTH = parathyroid hormone, RHR = resting heart rate, Scr = serum creatinine, TC = total cholesterol, TG = triglycerides, TSH = thyroid-stimulating hormone, UA = uric acid, Ucr = urinary creatinine. * P  < .05, with a statistically significant difference.

The comparison of clinical and laboratory data between NDR and DR.

2hPG = 2-hour postprandial glucose, ALT = alanine aminotransferase, AST = aspartate aminotransferase, BMI = body mass index, BUN = blood urea nitrogen, Ca = calcium, DAN = diabetic autonomic neuropathy, DKD = diabetic kidney disease, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, FPG = fasting plasma glucose, FT3 = free triiodothyronine, FT4 = free thyroxine, HbA1c = hemoglobin A1c, HDL-C = high-density lipoprotein cholesterol, LDL-C = low-density lipoprotein cholesterol, MA = microalbumin, NDR = non-diabetic retinopathy, P = phosphorus, PAD = peripheral artery disease, PTH = parathyroid hormone, RHR = resting heart rate, Scr = serum creatinine, TC = total cholesterol, TG = triglycerides, TSH = thyroid-stimulating hormone, UA = uric acid, Ucr = urinary creatinine.

P  < .05, with a statistically significant difference.

The comparison of 25 (OH) D between NDR and DR. DR = diabetic retinopathy, NDR = non-diabetic retinopathy.

3.2.2. Logistic regression analysis of DR and influencing factors:

Logistic regression analysis was performed with the onset of DR as the dependent variable and all other clinical indicators as independent variables. The results showed that 25(OH)D 3 , total 25(OH)D, and FT3 might be the risk factors for predicting DR ( P  <.05) (Table 4 ).

Table 4 Univariate binary logistic regression analysis of DR. Variables Single factor multiple factor β S.E Z
P
OR (95%CI) β S.E Z
P
OR (95% CI) 25 (OH) D 3 −0.09 0.02 −3.88 <.001 * 0.92 (0.88–0.96) −0.00 0.15 −0.00 0.997 1.00 (0.74–1.34) Total 25(OH)D −0.09 0.02 −3.97 <.001 * 0.91 (0.87–0.95) −0.10 0.15 −0.68 .495 0.90 (0.67–1.21) Disease time (year) 0.13 0.03 4.81 <.001 * 1.14 (1.08–1.20) 0.07 0.03 2.00 0.045 * 1.07 (1.01–1.15) BUN 0.19 0.07 2.86 .004 * 1.21 (1.06 ~ 1.38) 0.07 0.08 0.88 0.378 1.07 (0.92–1.26) MA 0.01 0.00 2.85 .004 * 1.01 (1.01 ~ 1.01) 0.00 0.00 0.52 0.606 1.00 (1.00–1.00) Ucr -0.01 0.00 -2.42 .015 * 0.99 (0.99–0.99) -0.00 0.00 -0.91 0.361 1.00 (1.00–1.00) FT3 -0.43 0.22 -1.98 .047 * 0.65 (0.43–0.99) 0.14 0.22 0.63 0.529 1.15 (0.75–1.75) DPN 1.96 0.37 5.24 <.001 * 7.08 (3.40–14.72) 1.17 0.59 1.99 0.046 * 3.23 (1.02–10.26) DKD 2.24 0.36 6.21 <.001 * 9.42 (4.64–19.13) 2.30 0.52 4.47 <.001 * 9.99 (3.64–27.40) Hypertension 0.91 0.31 2.91 .004 * 2.49 (1.34–4.59) 0.66 0.50 1.31 0.191 1.93 (0.72–5.16) Data are presented as regression coefficients (β), S.E, Z-value, and P -value. OR and 95% CI are shown. * P  < .05 indicates a statistically significant difference. 25(OH)D3 = 25-hydroxyvitamin D3, BUN = blood urea nitrogen, CI – confidence intervals, DKD = diabetic kidney disease, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, FT3 = free triiodothyronine, MA = microalbumin, OR = odds ratios, S.E. = standard error, Total 25(OH)D = total 25-hydroxyvitamin D, Ucr = urinary creatinine.

Univariate binary logistic regression analysis of DR.

Data are presented as regression coefficients (β), S.E, Z-value, and P -value. OR and 95% CI are shown. * P  < .05 indicates a statistically significant difference.

25(OH)D3 = 25-hydroxyvitamin D3, BUN = blood urea nitrogen, CI – confidence intervals, DKD = diabetic kidney disease, DPN = diabetic peripheral neuropathy, DR = diabetic retinopathy, FT3 = free triiodothyronine, MA = microalbumin, OR = odds ratios, S.E. = standard error, Total 25(OH)D = total 25-hydroxyvitamin D, Ucr = urinary creatinine.

3.2.3. Receiver operating characteristic curve analysis:

The receiver operating characteristic (ROC) curve analysis revealed that the optimal cutoff value of 25(OH)D 3 for predicting DR in T2DM was 23.15 ng/mL, with an area under the curve of 0.673, sensitivity of 56.3%, and specificity of 71%. Similarly, the optimal cutoff value of total 25(OH)D for predicting T2DM was 23.33 ng/mL, with an area under the curve of 0.674, a sensitivity of 59%, and a specificity of 69.4% (Table 5 , Fig. 2 ).

Table 5 ROC curve analysis of 25(OH)D3 and Total 25(OH)D in predicting DR. AUC
P
95% CI Cutoff value Sensitivity Specificity Yorden index 25(OH)D 3 0.673 .001 0.593 0.753 23.15 0.563 0.710 0.273 Total 25(OH)D 0.674 .001 0.594 0.754 23.33 0.590 0.694 0.284 25(OH)D3 = 25-hydroxyvitaminD3, AUC = area under the curve; CI = confidence interval, DR = diabetic retinopathy, ROC = receiver operating characteristic, Total25(OH)D = total 25-hydroxyvi -tamin D.

ROC curve analysis of 25(OH)D3 and Total 25(OH)D in predicting DR.

25(OH)D3 = 25-hydroxyvitaminD3, AUC = area under the curve; CI = confidence interval, DR = diabetic retinopathy, ROC = receiver operating characteristic, Total25(OH)D = total 25-hydroxyvi -tamin D.

ROC curves of 25(OH)D3 and total 25(OH)D predicting the occurrence of DR. 25(OH)D3 = 25-hydroxyvitamin D3, DR = diabetic retinopathy, ROC = receiver operating characteristic.

4. Discussion

Diabetes epidemiology is characterized by a shift toward younger age groups, with a steadily rising prevalence globally. [ 18 ] It is a global leading cause of blindness, renal failure, cardiovascular disease, and lower-limb amputations. DR is one of the most common microvascular complications of diabetes mellitus and is the leading cause of visual impairment and blindness in the working age population worldwide. [ 19 ] The epidemiology of DR shows an increasing trend globally, particularly in areas with a high prevalence of diabetes mellitus. DR causes visual impairment and imposes a heavy socio-economic burden. DR is divided into a nonproliferative (mild to severe) and proliferative stage, and the latter may be accompanied by macular edema, which may lead to irreversible visual impairment. Approximately 100 million people are affected by DR globally, a figure expected to increase when the number of diabetes cases increases to 783 million in 2045. [ 20 ] Studies have shown that the prevalence of DR is significantly associated with risk factors such as duration of diabetes, poor glycemic control, hypertension, and dyslipidemia. [ 21 , 22 ] Low- and middle-income countries bear approximately 75% of the global burden of DR, but generally lack the resources for routine fundus screening, resulting in 95% of preventable blindness going undetected in time. [ 23 ] Therefore, early diagnosis and treatment of DR is of paramount importance.

The accurate determination of vitamin D is crucial for clinical diagnosis, epidemiological studies, and nutritional assessment. However, the detection of vitamin D faces dual challenges in specificity and sensitivity due to the low concentrations of vitamin D metabolites in serum, the presence of multiple structurally highly similar isomers (e.g., vitamin D 2 and D 3 ), and metabolites (e.g., 25-hydroxyvitamin D, 1,25-dihydroxyvitamin D). Current mainstream analytical methods primarily include immunoassays, chromatography, and LC-MS/MS, which exhibit significant differences in methodological performance. Compared to other methods, LC-MS/MS integrates the high-efficiency separation capability of chromatography with the precise identification capability of mass spectrometry, establishing itself as the “gold standard” in terms of sensitivity and specificity. [ 24 ] In terms of sensitivity, LC-MS/MS can achieve accurate quantification at the nanomolar (nmol/L) level or even lower, sufficient for precise determination of serum concentrations in various populations (including those with severe vitamin D deficiency). In terms of specificity, its advantages are more pronounced, as it can clearly distinguish and quantitatively separate key metabolites such as 25(OH)D 2 , 25(OH)D 3 , 3-dihydroxy-25(OH)D 3 , and 1,25(OH) 2 D, completely resolving cross-reactivity issues. [ 25 ] This study employs LC-MS/MS for vitamin D determination, which minimizes classification errors to the greatest extent, ensuring that the observed biomarker levels are truly and robustly associated with Dr The high-standard data generated can also serve as a basis for establishing localized reference intervals or for subsequent methodologicalcomparisons.

In addition to bone metabolism, serum vitamin D is significantly associated with the onset and development of various diseases, such as cardiovascular diseases, metabolic diseases, tumors, multiple sclerosis, biological infections, and autoimmune diseases, and is involved in immune regulation, such as the proliferation and differentiation of immune cells. [ 26 ] Vitamin D deficiency is prevalent globally. In Taiwan, the prevalence of vitamin D deficiency in the community population is 22.4%, and this number rises to 59% in critically ill patients. [ 27 ] In patients with T2DM, vitamin D deficiency is associated with elevated levels of PTH fragments. Moreover, the risk of liver disease is relatively high in individuals with both T2DM and vitamin D deficiency. [ 28 ] Female, overweightness and obesity, low birth weight, and inadequate exposure to sunlight are significant risk factors for vitamin D deficiency. [ 29 ] In this study, vitamin D was detected using LC-MS/MS, and statistical analysis showed that total 25(OH)D and 25(OH)D 3 were reduced in both DR and NDR groups; however, vitamin D levels were relatively low in the DR group. ROC curve analysis revealed that the optimal cutoff value of 25(OH)D 3 for predicting the onset of DR in patients with T2DM was 23.15 ng/mL, and that of total 25(OH)D for predicting the onset of DR was 23.33 ng/mL in patients with T2DM. These results were close to the cutoff values in most studies, [ 30 ] suggesting that maintaining vitamin D above the deficiency threshold may reduce the risk of DR onset.

Several studies have shown that vitamin D deficiency is significantly associated with an increased risk of diabetic microvascular complications such as DR, DKD and DPN. [ 31 ] A cross-sectional study involving 815 patients with T2DM revealed that patients with comparatively low serum 25(OH)D concentrations were relatively likely to develop retinopathy, nephropathy, and peripheral neuropathy. [ 32 ] Another study on 4284 Chinese patients with T2DM showed that vitamin D deficiency was associated with an increased incidence of DR, DKD, and diabetic foot ulcers. [ 33 ] A potential protective mechanism of vitamin D may be improving microvascular function by reducing oxidative stress and inflammatory responses (e.g., lowering angiotensin II levels). [ 34 ] In addition, animal studies have shown that vitamin D supplementation reverses microvascular dysfunction, improves vessel density, and reduces retinal damage in diabetic rats. [ 35 ] Although most studies supported the relation between vitamin D deficiency and microvascular disorder, certain findings were inconsistent. One study revealed no statistically significant correlation between vitamin D levels and retinopathy, but a significant correlation between vitamin D levels with neuropathy. [ 36 ] In addition, most of the existing studies have adopted a cross-sectional design, thereby making it difficult to determine causality. Our previous research showed that serum 25(OH)D may be used as a simple and effective screening tool to predict the onset of DPN, as a total 25(OH)D <19.94 ng/mL or a 25(OH)D 3  <18.85 ng/mL may indicate elevated risk for DPN. [ 37 ] Meanwhile, this study showed that the incidence of DPN and DR was significantly higher in the vitamin D deficiency group, thus supporting the argument that vitamin D deficiency is associated with an increased risk of diabetic microvascular complications and suggesting that 25(OH)D 3 and total 25(OH)D are protective factors against DPN and Dr Therefore, oral vitamin D supplementation has been proposed to improve glycemic control and reduce the risk of microvascular complications. [ 38 ] However, further investigations are required to determine the causal relationship between vitamin D and diabetic microvascular complications and the effectiveness of clinical interventions.

In addition, this study revealed that FT3 is a protective factor for both DR and DPN. [ 27 ] Several studies have shown that reduced FT3 level is the most common independent factor for diabetic microvascular and macrovascular complications, and that this correlation is independent of other variables. In addition, restoration of normal thyroid function via levothyroxine supplementation may play a potential therapeutic role in DKD. [ 39 ] Although previous studies have not elaborated on the specific mechanisms, the central role of oxidative stress in diabetic microvascular complications has been mentioned. FT3 may be involved in the progress of microvascular injury by affecting the level of oxidative stress through regulation of the NADPH oxidative enzymes (Nox family). [ 40 ]

5. Limitations

This study has the following limitations. First, this was a retrospective study and a nondiabetic population was not included as a control group. Second, staging analysis was not performed for the DR group.

6. Conclusion

Decreased levels of 25(OH)D in patients with T2DM are strongly associated with the development of DR, and 25(OH)D is a protective factor for Dr Given that vitamin D deficiency may increase the risk of diabetic microvascular complications, routine testing of vitamin D, particularly vitamin D 3 , is recommended for patients with T2DM.

Acknowledgments

We are grateful to all the participants who took part in this study.

Author contributions

Conceptualization: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Xiuwen Zhou, Nuoqi Chen, Jinfeng Chen.

Data curation: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Xiuwen Zhou, Jinfeng Chen.

Formal analysis: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Xiuwen Zhou, Jinfeng Chen.

Funding acquisition: Senzhen Chen, Nuoqi Chen, Jinfeng Chen.

Investigation: Senzhen Chen, Shujing Zheng, Jinfeng Chen.

Methodology: Senzhen Chen, Shujing Zheng, Yiling Yan, Xiuwen Zhou, Jinfeng Chen.

Project administration: Senzhen Chen, Shujing Zheng, Yiling Yan, Xiuwen Zhou, Jinfeng Chen.

Resources: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Jinfeng Chen.

Software: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Xiuwen Zhou, Jinfeng Chen.

Supervision: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Jinfeng Chen.

Validation: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Jinfeng Chen.

Visualization: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Jinfeng Chen.

Writing – original draft: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Xiuwen Zhou, Jinfeng Chen.

Writing – review & editing: Senzhen Chen, Shujing Zheng, Yiling Yan, Lili Lin, Nuoqi Chen, Jinfeng Chen.

Abbreviations: 25(OH)D 25-hydroxyvitamin D BMI body mass index BUN blood urea nitrogen DAN diabetic autonomic neuropathy DKD diabetic kidney disease DPN diabetic peripheral neuropathy DR diabetic retinopathy FPG fasting plasma glucose HDL-C high-density lipoprotein cholesterol LC-MS/MS liquid chromatography-tandem mass spectrometry LDL-C low-density lipoprotein cholesterol PAD peripheral arterial disease RHR resting heart rate Scr serum creatinine T2DM type 2 diabetes mellitus TC total cholesterol TG triglyceride UA uric acid Ucr urinary creatinine

Abbreviations:

25-hydroxyvitamin D

body mass index

blood urea nitrogen

diabetic autonomic neuropathy

diabetic kidney disease

diabetic peripheral neuropathy

diabetic retinopathy

fasting plasma glucose

high-density lipoprotein cholesterol

liquid chromatography-tandem mass spectrometry

low-density lipoprotein cholesterol

peripheral arterial disease

resting heart rate

serum creatinine

type 2 diabetes mellitus

total cholesterol

triglyceride

uric acid

urinary creatinine

The author(s) declare that financial support was received for the research and/or publication of this article. This study was supported by the Startep Fund for Scientific Research, Fujian Medical University (grant number: 2020QH1281).

The Ethics Committee of Zhangzhou Affiliated Hospital of Fujian Medical University approved and the study was carried out in compliance with the Declaration of Helsinki for the parts that involved human subjects, human materials, or human data. The human samples used in this study were acquired from primarily isolated as part of our previous study for which ethical approval was obtained. Written informed consent for participation was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and institutional requirements.

The authors have no conflicts of interest to disclose.

All data generated or analyzed during this study are included in this published article [and its supplementary information files].

How to cite this article: Chen S, Zheng S, Yan Y, Lin L, Zhou X, Chen N, Chen J. Relationship between 25-hydroxyvitamin D by liquid chromatography-tandem mass spectrometry and retinopathy in type 2 diabetes mellitus. Medicine 2026;105:24(e49091).

SC, SZ, and YY contributed to this article equally.

Contributor Information

Senzhen Chen, Email: 498972923@qq.com.

Shujing Zheng, Email: 531874104@qq.com.

Yiling Yan, Email: 1056060181@qq.com.

Lili Lin, Email: linmoli@126.com.

Xiuwen Zhou, Email: 492533252@qq.com.

Nuoqi Chen, Email: 498972923@qq.com.

References