Higher salivary nitrite is associated with improved glycemic control and insulin sensitivity in patients with type 2 diabetes Authors: Zahra Bahadoran, Reza Norouzirad, Sajad Jeddi, Asghar Ghasemi Published: 2026 ABSTRACT Background and aimSalivary nitrite (SNO2), a marker of oral nitrate-reducing capacity and nitric oxide (NO) bioavailability, often impaired in type 2 diabetes (T2DM). This study aimed to investigate the associations between SNO2 concentration, and metabolic, vascular, and renal parameters in patients with T2DM.MethodsSixty-four adults with T2DM (53.9 ± 7.2 years, 45.3% men, diabetes duration 8.3 ± 6.2 years) participated in this cross-sectional study. Salivary NO2 concentrations, glycemic parameters, blood pressure (BP) indices, renal function, lipid profiles, liver function, and systemic NO metabolites were measured. Participants with high- and low-SNO2 (< and ≥ 69.3 µmol/L) were compared for cardiometabolic parameters.ResultsCompared with the low-SNO2 group, participants with higher SNO2 had lower HbA1c (6.2% vs. 6.9%, P = 0.040), fasting insulin (6.1 vs. 8.2 µU/mL, P = 0.007), HOMA-IR (2.1 vs. 3.3, P = 0.003), and C-peptide (1.3 vs. 1.7 ng/mL, P = 0.023), with higher QUICKI (0.34 vs. 0.32, P = 0.003). Fasting serum glucose was marginally lower (142 vs. 167 mg/dL, P = 0.069) in the high- compared to the low-SNO2 group. Median serum nitric oxide metabolites (NOx) concentrations were higher in the high-SNO2 group (51.1 vs. 29.1 µmol/L, P = 0.015). Salivary nitrate (NO3) (694 vs. 463 µmol/L, P = 0.035) and NO2 (153 vs. 20.6, P = 0.001) were also higher in the high- compared to low-SNO2 group. No significant difference was observed in BP measures, lipid profile, and liver and renal function test across groups.ConclusionsHigh-SNO2 levels were associated with greater glycemic control, insulin sensitivity, and NO bioavailability in patients with T2DM. Introduction Nitric oxide (NO), a gaseous signaling molecule, has key roles in regulating glucose and insulin homeostasis [1]. Impaired NO metabolism is involved in the pathogenesis of type 2 diabetes mellitus (T2DM) [2, 3], and its complications, including vascular disorders [4], sarcopenia [5], neuropathy [6], adipose dysfunction [7], nephropathy [8], and diabetes-related wounds [9]. Beyond the classical L-arginine-NO synthase (NOS) pathway, NO can also be generated via the nitrate (NO3)-nitrite (NO2)-NO pathway, which relies on the enterosalivary circulation of NO3 [10, 11]. Approximately 25% of ingested NO3 is actively concentrated in the salivary glands and secreted into saliva, where commensal oral bacteria convert part of it to NO2 through NO3 reductase activity [12]. Once swallowed, salivary NO2 is further reduced to bioactive NO in the acidic gastric environment and various organs, supporting vascular and metabolic homeostasis [12, 13]. Salivary NO2 (SNO2) concentration serves as a surrogate marker of oral NO3-reducing capacity (ONRC) [14, 15]. A higher ONRC and salivary NO2 levels are associated with enhanced systemic NO bioavailability, lower blood pressure (BP), and improved insulin sensitivity [16, 17]. Higher NO bioavailability enhances microvascular perfusion and facilitating insulin-mediated glucose uptake in skeletal muscle, thereby improves postprandial glycemic control and insulin sensitivity [1]. In contrast, impaired ONRC reduces NO production and may contribute to endothelial dysfunction, hypertension, and metabolic disturbances [16, 17]. In T2DM, ONRC is often impaired, leading to reduced systemic NO availability compared with healthy individuals [18, 19]. Impaired ONRC in T2DM is associated with oral microbiota dysbiosis, reduced NO2 production, and diminished NO bioavailability, contributing to cardiometabolic dysfunction [19]. Evidence suggests that T2DM patients with higher ONRC exhibit greater systemic NO exposure, improved postprandial glycemic control, and larger reductions in BP following acute oral NO3 intake [18]. However, the association between salivary NO2 concentrations and cardiometabolic parameters in patients with T2DM remains to be elucidated. This study therefore aimed to investigate the association between SNO2 concentrations and key cardiometabolic parameters, including glucose and insulin homeostasis, lipid metabolism, BP indices, and renal and liver function tests in patients with T2DM. We hypothesized that higher SNO2 concentrations, a potential surrogate of higher ONRC, are associated with an enhanced NO availability, improved glycemic control, and more favorable cardiometabolic and renal profiles in these patients. Materials and methods Study participants and design This study is a reanalysis of baseline data from a previously published randomized clinical trial [20]. Detailed descriptions of the study design, participant recruitment, inclusion and exclusion criteria, and intervention procedures have been published elsewhere [20]. In brief, patients with clinically diagnosed T2DM were recruited from the Iranian Diabetes Society between October 2018 and December 2019. All participants were non-smokers and had not used antibiotics or antibacterial mouthwash in the preceding month. Exclusion criteria were insulin use, poor glycemic control (HbA1c ≥ 11%), active cardiovascular, renal, or liver disease, systemic autoimmune or chronic inflammatory conditions, and use of medications that may affect nitric oxide metabolism or signaling [21, 22]. Participants with HbA1c ≥ 11% were excluded to reduce confounding from severely uncontrolled diabetes, as levels above 10% indicate substantial metabolic instability and heightened complication risk according to American Diabetes Association (ADA) Standards of Care [23]. The original trial was registered in the Iranian Registry of Clinical Trials (IRCT20180409039246N1) [20]. For the present study, we utilized baseline measurements (before intervention) from all trial participants (n = 64). The study was conducted in accordance with the principles of the original trial. All participants provided written informed consent prior to enrollment. The original trial was approved by the Ethics Committee of the Research Institute for Endocrine Sciences of Shahid Beheshti University of Medical Sciences (IR.SBMU.ENDOCRINE.REC.1395.322). Details of the demographic, anthropometric (including height, body weight, waist circumference), clinical (including systolic [SBP] and diastolic [DBP] blood pressures), and biochemical assessments, including collection of fasting blood, spot saliva, and spot and 24-hour urine samples, have been described previously [20]. Measurement of metabolic parameters Glycated hemoglobin (HbA1c) was measured using the Cera-Stat 2000 HbA1c analyzer (MH-200, LabonaCheck™, Ceragem Medisys Inc., South Korea). Fasting plasma glucose (FPG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), liver function tests [LFTs, including measurement of serum alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase (ALP), and γ-glutamyl transpeptidase (GGT)], and uric acid (UA) were measured via enzymatic colorimetric methods using Pars Azmoon kits (Pars Azmoon Inc., Tehran, Iran) on an auto-analyzer (Selectra E, Netherlands). Fasting serum insulin (FSI) and C-peptide levels were measured by enzyme-linked immunosorbent assay (ELISA) kits (Monobind Inc., USA) using a microplate reader (BioTek MQX2000R2, USA). Inter- and intra-assay coefficients of variation (CVs) for all assays were below 5%. Measurement of renal parameters Serum (SCr) and urine (UCr) creatinine were determined using the kinetic colorimetric Jaffe method. Urinary microalbumin was quantified using an ELISA kit (Padtan Elm Co., Tehran, Iran) and a microplate reader (BioTek MQX2000R2, USA), with inter- and intra-assay CVs of 9.3% and 9.8%, respectively. Measurement of nitric oxide metabolites NOx concentrations in serum, saliva, and urine were measured using a spectrophotometric method developed by Miranda et al. and validated in our laboratory [24, 25]. Intra-assay CVs for serum, saliva, and urine were 1.0%, 1.3%, and 2.9%, respectively, while inter-assay CVs were 1.4%, 1.5%, and 3.3%. The assay sensitivity was 2.0 µmol/L with a recovery rate of 93 ± 1.5%. Definitions and calculations According to the ADA, T2DM was defined as FSG ≥ 126 mg/dL, 2-hour post-load glucose ≥ 200 mg/dL, or the use of oral glucose-lowering medications [23]. Insulin resistance was estimated using the homeostatic model assessment of insulin resistance (HOMA-IR), calculated as FSI (µU/mL) × FSG (mmol/L)/22.5. The quantitative insulin sensitivity check index (QUICKI) was calculated as 1 ÷ [log FSI (µU/mL) + log FSG (mg/dL)]. Both HOMA-IR and QUICKI are among the most accurate surrogate indexes for determining insulin sensitivity in humans [26]. Fasting serum C-peptide and FSI concentrations were used to estimate the molar C-peptide-to-insulin ratio (C/I) and hepatic insulin extraction (HIE), i.e., calculated by 1–1 ÷ (C/ I) [27]. Creatinine clearance (CrCl, mL/min) was calculated using the following formula, CrCl= 24 h-UCr (mg/dL) ×24 h-UV (mL) ÷ SCr (mg/dL) ×1440. NOx clearance (NOxCl, mL/min) was calculated using the following formula, NOxCl= 24 h-UNOx (µmol/L) ×24 h-UV (mL) ÷ SNOx (µmol/L) ×1440. The fractional excretion of NOx (FENOx, %) was calculated using the following formula, FENOx=24 h-UNOx (µmol/L) × SCr (mg/dL) ÷ SNOx (µmol/L) × 24 h-UCr (mg/dL). The four BP index measures, including pulse pressure (PP), mean blood pressure (MBP), mean arterial pressure (MAP), and mean proportional arterial pressure (MPAP) were calculated by combining SBP and DBP using the following formula [28]: PP (mm Hg)= SBP–DBP ; MBP (mm Hg)= (SBP+DBP)÷2; MAP (mm Hg)= (SBP+2DBP)÷3; MPAP (mm Hg)= (SBP2+DBP2)÷(SBP+DBP). These BP indices are differentially associated with CVD incidence and all-cause mortality [28, 29]. The fatty liver index (FLI) was calculated using the following formula [30], ez ÷ (1 + ez), where z = 0.953 × ln (TG) + 0.139 × BMI + 0.718 × ln (GGT) + 0.053 × WC − 15.745. FLI ≥ 60 was considered as the probability of non-alcoholic fatty liver disease (NAFLD) [30]. To estimate glomerular filtration rate (eGFR), the Chronic Kidney Disease Epidemiology Collaboration equation was used [31]. Chronic kidney disease (CKD) was defined as eGFR < 60 mL/min/1.73 m2. Statistical analyses Statistical analyses were conducted using SPSS (version 20.0; Chicago, IL, USA). Two-sided P values < 0.05 were considered statistically significant. Normality of continuous variables was assessed using the Shapiro-Wilk test. Because no established clinical or population-based threshold for SNO2 concentration exists, we used its median in our study population (69.3 µmol/L) as the cut-off point. Classification of participants into high- or low-SNO2 groups was performed after all data were collected, based on measured SNO2 concentrations. Comparisons of continuous variables across median SNO2 concentrations (< 69.3 vs. ≥69.3 µmol/L) were performed using independent-samples t-tests for normally distributed data. The Mann-Whitney U test was used for non-normally distributed data, and the results were reported as median and interquartile range (IQR). As this study was an exploratory analysis, no sample size calculation was performed; the study was powered for the original study [20], and the current analyses were intended to generate hypotheses regarding associations between SNO2 and cardiometabolic parameters. Results Participant characteristics The participants had a mean age of 53.9 ± 7.2 years, and 45.3% were men. The mean duration of diabetes was 8.3 ± 6.2 years. Approximately 22% of participants used metformin alone, 6% were on Zipmet monotherapy, and 69% were treated with a combination of metformin and glibenclamide. Table 1 presents demographic, anthropometric, and clinical characteristics of the 64 study participants, compared across the median of SNO2 concentrations. Participants were generally comparable across SNO2 categories, with no significant differences in age, sex distribution, T2D duration, or anthropometric measures. Table 1Characteristics of the study participants across median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesAge, y53.9 ± 8.054.0 ± 6.40.973Men, % (number)53.1 (17)37.5 (12)0.315DM duration, y6.5 (3.2–10.0)7.5 (3.0–10.0)0.706Glucose-lowering medications, % (number)100 (32)97.0 (31)0.696 Metformin21.9 (7)21.9 (7) Glibenclamide0 (0)3.1 (1) Zipmet6.2 (2)6.2 (2) Multi-drugs68.8 (22)65.6 (21)BW, kg80.9 ± 12.378 ± 13.20.362WC, cm101 (92.2–106)97.0 (92.2–102)0.270BMI, kg/m229.4 (26.5–32.1)28.4 (26.1–30.9)0.546BW, body weight; BMI, body mass index; DM, diabetes mellitus; NO2, nitrite, NO3, nitrate; WC, waist circumference Table 1Characteristics of the study participants across median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesAge, y53.9 ± 8.054.0 ± 6.40.973Men, % (number)53.1 (17)37.5 (12)0.315DM duration, y6.5 (3.2–10.0)7.5 (3.0–10.0)0.706Glucose-lowering medications, % (number)100 (32)97.0 (31)0.696 Metformin21.9 (7)21.9 (7) Glibenclamide0 (0)3.1 (1) Zipmet6.2 (2)6.2 (2) Multi-drugs68.8 (22)65.6 (21)BW, kg80.9 ± 12.378 ± 13.20.362WC, cm101 (92.2–106)97.0 (92.2–102)0.270BMI, kg/m229.4 (26.5–32.1)28.4 (26.1–30.9)0.546BW, body weight; BMI, body mass index; DM, diabetes mellitus; NO2, nitrite, NO3, nitrate; WC, waist circumference Characteristics of the study participants across median of salivary NO2 BW, body weight; BMI, body mass index; DM, diabetes mellitus; NO2, nitrite, NO3, nitrate; WC, waist circumference Glycemic and insulin sensitivity profiles As reported in Table 2, participants with SNO2 concentrations at or above the median (≥ 69.3 µmol/L) demonstrated a more favorable glycemic and insulin sensitivity profile compared with those below the median. HbA1c levels were significantly lower in the high-SNO2 group (6.2 ± 0.9% vs. 6.9 ± 1.4%, P = 0.040). Median FSI concentrations were significantly lower (6.1 vs. 8.2 µU/mL, P = 0.007), accompanied by lower HOMA-IR values (2.1 vs. 3.3, P = 0.003) and higher QUICKI scores (0.34 vs. 0.32, P = 0.003). C-peptide levels were also significantly lower in the high-SNO2 group (1.3 vs. 1.7 ng/mL, P = 0.023). Fasting serum glucose showed a marginally lower mean value among participants with higher SNO2 concentrations (142 ± 38.0 vs. 167 ± 66.9 mg/dL, P = 0.069). The prevalence of good glycemic control (HbA1c < 7.0%) did not differ significantly between groups. No significant differences were observed in lipid profile parameters, liver enzyme concentrations, fatty liver index, or the prevalence of NAFLD across SNO2 categories. Table 2Glycemic parameters, lipid profiles and liver function tests across the median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesHbA1c, %6.9 ± 1.46.2 ± 0.90.040Good glycemic control†, %56.271.90.149FSG, mg/dL167 ± 66.9142 ± 38.00.069FSI, µU/mL8.2 (6.0–12.6)6.1 (4.7–8.0)0.007HOMA-IR3.3 (2.1–5.1)2.1 (1.6–2.6)0.003QUICKI0.32 (0.30–0.34)0.34 (0.33–0.35)0.003C-peptide, ng/mL1.7 (1.4–2.0)1.3 (1.2–1.9)0.023FSI-to-C-peptide ratio, pmol/pmol10.3 (7.5–12.7)10.6 (8.4–14.2)0.245HIE0.90 (0.86–0.92)0.90 (0.88–0.93)0.243TG, mg/d1142 (106–190)128 (110–187)0.555TC, mg/dL162 ± 50.8168 ± 41.10.622LDL-C, mg/dL85.0 ± 29.886.5 ± 25.20.832HDL-C, mg/dL42.5 ± 10.646.0 ± 11.30.219TG-to-HDL-C3.92 ± 2.283.25 ± 1.440.168ALT, U/L16.0 (11.0–22.5)18.0 (11.2–24.0)0.448AST, U/L20.0 (16.2–24.0)23.0 (18.2–29.7)0.083ALT-to-AST ratio0.82 ± 0.280.79 ± 0.280.687ALP, U/L166 (144–198)168 (125–194)0.804GGT, U/L23.0 (16.2–28.0)23.0 (18.0–31.7)0.447FLI66.7 (47.4–82.7)55.0 (38.9–84.0)0.420NAFLD, %56.246.90.309Data are mean ± SD, median (interquartile range), or percent. Independent-samples t-test, Mann-Whitney U test, and chi-square test were used for normally distributed continuous, non-normally distributed continuous, and categorical variables, respectively†HbA1c < 7.0%ALT, alanine transaminase; AST, aspartate transaminase; ALP, alkaline phosphatase; FSG, fasting serum glucose; FLI, fatty liver index; FSI, fasting serum insulin, HbA1c, hemoglobin A1C; HIE, hepatic insulin extraction; HDL-C, high-density lipoprotein lipase; HOMA-IR, homeostatic model assessment of insulin resistance; GGT; γ-glutamyl transpeptidase NAFLD, non-alcoholic fatty liver disease; LDL-C, low-density lipoprotein lipase; QUICKI, quantitative insulin sensitivity check index; TG, triglycerides; TC, total cholesterol Table 2Glycemic parameters, lipid profiles and liver function tests across the median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesHbA1c, %6.9 ± 1.46.2 ± 0.90.040Good glycemic control†, %56.271.90.149FSG, mg/dL167 ± 66.9142 ± 38.00.069FSI, µU/mL8.2 (6.0–12.6)6.1 (4.7–8.0)0.007HOMA-IR3.3 (2.1–5.1)2.1 (1.6–2.6)0.003QUICKI0.32 (0.30–0.34)0.34 (0.33–0.35)0.003C-peptide, ng/mL1.7 (1.4–2.0)1.3 (1.2–1.9)0.023FSI-to-C-peptide ratio, pmol/pmol10.3 (7.5–12.7)10.6 (8.4–14.2)0.245HIE0.90 (0.86–0.92)0.90 (0.88–0.93)0.243TG, mg/d1142 (106–190)128 (110–187)0.555TC, mg/dL162 ± 50.8168 ± 41.10.622LDL-C, mg/dL85.0 ± 29.886.5 ± 25.20.832HDL-C, mg/dL42.5 ± 10.646.0 ± 11.30.219TG-to-HDL-C3.92 ± 2.283.25 ± 1.440.168ALT, U/L16.0 (11.0–22.5)18.0 (11.2–24.0)0.448AST, U/L20.0 (16.2–24.0)23.0 (18.2–29.7)0.083ALT-to-AST ratio0.82 ± 0.280.79 ± 0.280.687ALP, U/L166 (144–198)168 (125–194)0.804GGT, U/L23.0 (16.2–28.0)23.0 (18.0–31.7)0.447FLI66.7 (47.4–82.7)55.0 (38.9–84.0)0.420NAFLD, %56.246.90.309Data are mean ± SD, median (interquartile range), or percent. Independent-samples t-test, Mann-Whitney U test, and chi-square test were used for normally distributed continuous, non-normally distributed continuous, and categorical variables, respectively†HbA1c < 7.0%ALT, alanine transaminase; AST, aspartate transaminase; ALP, alkaline phosphatase; FSG, fasting serum glucose; FLI, fatty liver index; FSI, fasting serum insulin, HbA1c, hemoglobin A1C; HIE, hepatic insulin extraction; HDL-C, high-density lipoprotein lipase; HOMA-IR, homeostatic model assessment of insulin resistance; GGT; γ-glutamyl transpeptidase NAFLD, non-alcoholic fatty liver disease; LDL-C, low-density lipoprotein lipase; QUICKI, quantitative insulin sensitivity check index; TG, triglycerides; TC, total cholesterol Glycemic parameters, lipid profiles and liver function tests across the median of salivary NO2 Data are mean ± SD, median (interquartile range), or percent. Independent-samples t-test, Mann-Whitney U test, and chi-square test were used for normally distributed continuous, non-normally distributed continuous, and categorical variables, respectively †HbA1c < 7.0% ALT, alanine transaminase; AST, aspartate transaminase; ALP, alkaline phosphatase; FSG, fasting serum glucose; FLI, fatty liver index; FSI, fasting serum insulin, HbA1c, hemoglobin A1C; HIE, hepatic insulin extraction; HDL-C, high-density lipoprotein lipase; HOMA-IR, homeostatic model assessment of insulin resistance; GGT; γ-glutamyl transpeptidase NAFLD, non-alcoholic fatty liver disease; LDL-C, low-density lipoprotein lipase; QUICKI, quantitative insulin sensitivity check index; TG, triglycerides; TC, total cholesterol Blood pressure and renal function Table 3 compares BP and renal function parameters across the median of SNO2. Blood pressure indices and renal function parameters did not differ significantly across categories of SNO₂ concentration. Mean SBP was comparable between participants with low- and high-SNO2 (141 ± 20.6 vs. 136 ± 22.8 mm Hg, P = 0.373), and no significant differences were observed for DBP, HR, PP, MBP, MAP, or MPAP. Markers of renal function, including SCr, eGFR, CrCl, urine volume, and urinary albumin excretion, were also comparable between groups. Spot UCr tended to be lower in the high-SNO2 group (76.0 vs. 99.0 mg/mL, P = 0.055). The prevalence of CKD was lower in high-SNO2 group (37% vs. 53%), although this difference was not statistically significant (P = 0.158). Table 3Blood pressure and renal function parameters across the median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesSBP, mm Hg141 ± 20.6136 ± 22.80.373DBP, mm Hg88.0 (77.2–99.2)82.5 (75.2–70.7)0.214HR, bpm75.7 ± 8.773.9 ± 10.30.470PP, mm Hg52.6 ± 11.252.2 ± 15.10.903MBP, mm Hg186 ± 27.6179 ± 29.20.321MAP, mm Hg106 (93.8–119)101 (91.0–111)0.214MPAP, mm Hg284 (250–322)271 (245–296)0.248Serum Cr, mg/dL1.20 (1.00–1.30)1.10 (1.00–1.20)0.178Serum UA, mg/dL4.33 ± 1.204.65 ± 1.210.286sUCr, mg/mL99.0 (78.5–136)76.0 (42.0–111)0.05524 h-UCr, mg/mL53.0 (38.0–76.5)56.0 (30.5–89.5)0.861sUAlb, µg/mL5.6 (0.4–12.5)1.4 (0–11.6)0.11024 h-UAlb, µg/mL9.1 (0.0–22.1)5.4 (0.6–9.5)0.658CrCl, mL/min56.3 (45.0–75.1)51.5 (40.7–83.6)0.961eGFR, mL/min/1.73 m260.4 ± 9.561.2 ± 9.70.715Urine volume, mL1848 ± 5941828 ± 7320.658CKD, %53.137.50.158Data are mean ± SD or median (inter-quartile range). Independent-samples t-test, Mann-Whitney U test, and chi-square test were used for normally distributed continuous, non-normally distributed continuous, and categorical variables, respectivelyAlb, microalbumin; bpm, beats per minute; Cr, creatinine; CKD, chronic kidney disease; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; MBP, mean blood pressure; MAP, mean arterial pressure; MPAP, mean proportional arterial pressure; PP, pulse pressure; SBP, systolic blood pressure; sU, spot urine; UA, uric acid Table 3Blood pressure and renal function parameters across the median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesSBP, mm Hg141 ± 20.6136 ± 22.80.373DBP, mm Hg88.0 (77.2–99.2)82.5 (75.2–70.7)0.214HR, bpm75.7 ± 8.773.9 ± 10.30.470PP, mm Hg52.6 ± 11.252.2 ± 15.10.903MBP, mm Hg186 ± 27.6179 ± 29.20.321MAP, mm Hg106 (93.8–119)101 (91.0–111)0.214MPAP, mm Hg284 (250–322)271 (245–296)0.248Serum Cr, mg/dL1.20 (1.00–1.30)1.10 (1.00–1.20)0.178Serum UA, mg/dL4.33 ± 1.204.65 ± 1.210.286sUCr, mg/mL99.0 (78.5–136)76.0 (42.0–111)0.05524 h-UCr, mg/mL53.0 (38.0–76.5)56.0 (30.5–89.5)0.861sUAlb, µg/mL5.6 (0.4–12.5)1.4 (0–11.6)0.11024 h-UAlb, µg/mL9.1 (0.0–22.1)5.4 (0.6–9.5)0.658CrCl, mL/min56.3 (45.0–75.1)51.5 (40.7–83.6)0.961eGFR, mL/min/1.73 m260.4 ± 9.561.2 ± 9.70.715Urine volume, mL1848 ± 5941828 ± 7320.658CKD, %53.137.50.158Data are mean ± SD or median (inter-quartile range). Independent-samples t-test, Mann-Whitney U test, and chi-square test were used for normally distributed continuous, non-normally distributed continuous, and categorical variables, respectivelyAlb, microalbumin; bpm, beats per minute; Cr, creatinine; CKD, chronic kidney disease; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; MBP, mean blood pressure; MAP, mean arterial pressure; MPAP, mean proportional arterial pressure; PP, pulse pressure; SBP, systolic blood pressure; sU, spot urine; UA, uric acid Blood pressure and renal function parameters across the median of salivary NO2 Data are mean ± SD or median (inter-quartile range). Independent-samples t-test, Mann-Whitney U test, and chi-square test were used for normally distributed continuous, non-normally distributed continuous, and categorical variables, respectively Alb, microalbumin; bpm, beats per minute; Cr, creatinine; CKD, chronic kidney disease; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; MBP, mean blood pressure; MAP, mean arterial pressure; MPAP, mean proportional arterial pressure; PP, pulse pressure; SBP, systolic blood pressure; sU, spot urine; UA, uric acid Nitric oxide metabolism Table 4 summarizes NO metabolism parameters across the median of salivary NO2 concentrations. Participants with SNO2 concentrations at or above the median (≥ 69.3 µmol/L) exhibited significantly higher systemic and salivary NOx compared with those below the median. Median serum NOx concentrations were significantly higher in the high-SNO2 group (51.1 vs. 29.1 µmol/L, P = 0.015). Salivary NOx levels were also significantly elevated among individuals with higher SNO2 concentration (986 vs. 482 µmol/L, P = 0.041), accompanied by a higher salivary NO2-to-NO3 ratio (0.29 vs. 0.05, P = 0.001). No significant differences were observed in 24-hour urinary NOx excretion, NOxCl, or FE-NOx between groups. Table 4Nitric oxide metabolism parameters across median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesSNOx, µmol/L29.1 (19.0–44.4)51.1 (27.3–92.5)0.015sUNOx, µmol/L964 (704–1475)1532 (1001–2447)0.15724 h-UNOx, µmol/L805 (467–1517)1025 (518–1762)0.282NOxCl, mL/min27.2 (14.3–45.8)15.9 (6.1–47.7)0.658FE-NOx, %0.51 (0.28–0.87)0.27 (0.13–0.92)0.558Salivary NO3, µmol/L463 (263–789)694 (372–2062)0.035Salivary NO2, µmol/L20.6 (9.1–39.2)153 (99.2–344)0.001Salivary NO2/NO30.05 (0.01–0.09)0.29 (0.14–0.55)0.001Data are median (inter-quartile range). Mann-Whitney U test was usedFE, fractional excretion rate; NO2, nitrite; SNOx, serum nitric oxide metabolites; sUNOx, spot urinary nitric oxide metabolites; 24 h-UNOx, 24 h urinary nitric oxide metabolites Table 4Nitric oxide metabolism parameters across median of salivary NO2Salivary NO2 concentration (µmol/L)< 69.3 ( n = 32) ≥ 69.3 ( n = 32) P valuesSNOx, µmol/L29.1 (19.0–44.4)51.1 (27.3–92.5)0.015sUNOx, µmol/L964 (704–1475)1532 (1001–2447)0.15724 h-UNOx, µmol/L805 (467–1517)1025 (518–1762)0.282NOxCl, mL/min27.2 (14.3–45.8)15.9 (6.1–47.7)0.658FE-NOx, %0.51 (0.28–0.87)0.27 (0.13–0.92)0.558Salivary NO3, µmol/L463 (263–789)694 (372–2062)0.035Salivary NO2, µmol/L20.6 (9.1–39.2)153 (99.2–344)0.001Salivary NO2/NO30.05 (0.01–0.09)0.29 (0.14–0.55)0.001Data are median (inter-quartile range). Mann-Whitney U test was usedFE, fractional excretion rate; NO2, nitrite; SNOx, serum nitric oxide metabolites; sUNOx, spot urinary nitric oxide metabolites; 24 h-UNOx, 24 h urinary nitric oxide metabolites Nitric oxide metabolism parameters across median of salivary NO2 Data are median (inter-quartile range). Mann-Whitney U test was used FE, fractional excretion rate; NO2, nitrite; SNOx, serum nitric oxide metabolites; sUNOx, spot urinary nitric oxide metabolites; 24 h-UNOx, 24 h urinary nitric oxide metabolites Discussion In this study of patients with T2DM, we found that higher SNO2 concentration, a potential surrogate of higher ONRC, was associated with improved glycemic control, indicated by lower HbA1c, FSI, and HOMA-IR and higher insulin sensitivity index, measured by QUICKI. Participants with high-SNO2 also exhibited higher systemic NO metabolites, suggesting that efficient oral NO3-to-NO2 conversion may contribute to NO availability and metabolic health in patients with T2DM. These findings support the concept that ONRC, by improving systemic NO availability, may influence multiple metabolic pathways in T2DM and underscore the potential of SNO2 as a non-invasive marker of NO metabolism and metabolic health. A key finding of this study was the inverse association between SNO2 levels and surrogates of glycemic control in individuals with T2DM. In agreement with our study findings, a recent study on 144 participants from the San Juan Overweight Adult Longitudinal Study (SOALS) reported that higher ONRC activity was inversely associated with HOMA-IR under both aerobic (β = –0.48, 95% CI = –0.81, –0.15) and CO2-enriched (β = –0.42, 95% CI = –0.68, –0.17) conditions, as well as with FSG (β = –0.144, 95% CI = –0.268, –0.019, and β = –0.070, 95% CI = –0.130, –0.011, in aerobic and CO2-enriched conditions, respectively) and 2 h-SG (β = –0.21, 95% CI = –0.37, –0.04, in CO2-enriched condition) [17]. Measurement of ONRC under CO2-enriched conditions was used to mimic the microaerophilic environment of the posterior tongue, where NO3-reducing bacteria are most active [17]. A preclinical study also reported that among patients with T2DM, those with higher ONRC had improved postprandial glycemic responses following consumption of a NO3-rich meal [18]. Observational studies linking frequent antiseptic mouthwash use to increased risk of prediabetes and T2DM [32, 33], further emphasize on importance of functional role of ONRC on metabolic regulation. Reduced ONRC and impaired NO generation may contribute to impaired glucose and insulin homeostasis, since NO has central role in regulation of carbohydrate metabolism [1]. NO enhances skeletal muscle glucose uptake by increasing blood flow, upregulating GLUT4 expression and translocation, and induction of insulin transendothelial transport [34, 35]. In adipose tissue, NO promotes glucose uptake, browning of white fat, thermogenesis, and improve energy metabolism [36–39]. In the liver, NO suppresses gluconeogenesis [40], improves insulin sensitivity [41], and facilitates glucose clearance [42]. In the central nervous system, NO regulates hypothalamic glucose sensing [43, 44], and coordinates peripheral glucose metabolism via neural pathways linking the brain to the pancreas, liver, and skeletal muscle [45, 46]. Diminished NO therefore contributes to impaired insulin signaling and reduced glucose disposal, consistent with our observation of poorer glycemic parameters in individuals with lower SNO2. We did not find any significant association between SNO2 and BP regulation and renal function. The absence of a significant association between SNO2 and BP indices is consistent with prior evidence showing no association between higher levels of oral bacterial NO3 reductase genes or NO3-reducing bacteria and BP indices [15], and a recent meta-analysis reported no significant association between SNO2 and either SBP or DBP [47]. However, some previous observations showed that greater SNO2 levels or enhanced ONRC are associated with improved BP and endothelial function, whereas blunted oral NO3-reducing activity is associated with elevated BP and an increased risk of developing hypertension [16, 48, 49]. Salivary NO2-to-NO3 ratio was significantly reduced in high cardiometabolic risk compared to healthy individuals (1:4 vs.1:3) despite their higher salivary NO3 concentration (680 vs. 535 µmol/L) [47]. The associations observed between salivary NO2 and serum NOx concentrations in our study are consistent with experimental evidence showing that suppression of oral bacteria with antiseptic mouthwash significantly reduces plasma NO2 levels and abolishes NO-dependent biological effects [16, 50, 51]. Reduced ONRC may contribute to the attenuated vascular and metabolic responsiveness to dietary NO3 reported in patients with T2DM [20, 52]. As reviewed elsewhere, reduced abundance of NO3-reducing species is hallmark of oral microbial dysbiosis in patients with T2DM [19], leading to lower NO2 production and reduced systemic NO availability [18]. In addition, germ-free animals show negligible NO2 and NO formation even after NO3 ingestion [53, 54], indicating that the overall composition of the oral microbiome, not NO3 exposure alone, determines the efficiency of systemic NO generation [13, 55]. On the other hand, higher ONRC, which depends on the abundance of key NO3-reducing bacteria [55–58], is associated with greater increases in circulating NO2 following dietary NO3 intake [59]. We previously reported that higher ONRC in patients with T2DM was associated with greater actual body NOx exposure following ingestion of an acute dose of NO3 (103 ± 31.4 vs. 58.9 ± 22.1 µmol×h/L) [18]. Our study findings align with growing evidence showing that the entero-salivary NO3-NO2-NO pathway represents an important alternative source of bioactive NO [10, 11], particularly in conditions of impaired endogenous NO synthesis such as T2DM [3]. Our findings should be interpreted in light of the study strengths and limitations. A major strength is the simultaneous assessment of salivary NO2 alongside multiple cardiometabolic outcomes, including glycemic parameters, BP indices, renal function, and systemic NO metabolites, which enables an integrated evaluation of the potential effects of SNO2 on cardiometabolic health in patients with T2DM. However, a single measurement may not capture long-term or diurnal variations in SNO2. SNO2 concentrations are influenced by recent intake of NO3-rich foods, such as leafy vegetables, which can elevate SNO2 independently of oral NO3-reducing capacity and may contribute to associations with glycemic and insulin sensitivity measures. In our study, background dietary NO3 intake, estimated using urinary NO3 excretion (24-h urinary NO3 = 0.55 × ingested NO3 + 0.22, accounting for endogenous production) [60], did not differ significantly between participants with high- and low-SNO2 (1866 vs. 2148 µmol/day, P = 0.471). These findings indicate that interindividual differences in SNO2 were not attributable solely to recent dietary NO3 intake, however, we acknowledge that diet was not controlled, which remains a limitation when interpreting salivary NO metabolites. In addition, SNO2 can be influenced by oral hygiene and oral microbiota composition, which we were not able to control. Another point is that ONRC we did not use direct measurement of ONRC (e.g., in vitro incubation of saliva with NO3, or NO3-reductase activity of oral resident bacteria [56, 61]) in this study, and SNO2 should be interpreted only as an indirect indicator of NO3-reducing capacity. Although the sex distribution in our study was more likely to be balanced, potential sex-specific associations between SNO2 and cardiometabolic profiles cannot be ruled out. Despite these limitations, the study provides valuable observational insights and generates hypotheses for longitudinal and interventional research on the role of SNO2 in metabolic and vascular health in patients with T2DM. In conclusion, higher SNO2 concentrations were associated with a more favorable metabolic profile in patients with T2DM, including lower HbA1c, fasting insulin, HOMA-IR, and C-peptide, as well as higher QUICKI values. These findings suggest that SNO2, as a non-invasive surrogate marker of ONRC, may help capture interindividual differences in NO metabolism and identify patients with more favorable insulin sensitivity and glycemic control. Longitudinal and randomized clinical trials are needed to expand on these hypothesis-generating findings and investigate whether improving ONRC through dietary NO3, oral microbiome modulation, or targeted probiotics can improve metabolic and vascular outcomes in patients with T2DM.