Tip: Try author name, DOI (10.xxxx/…), or keywords.

ISSN (Online): 1694-4674
Annals of Medicine and Medical Sciences
Submit Manuscript
  1. Home
  2. Vol. 05, No. 09, (2026)
  3. Serum Adenosine Deaminase as a Biomarker of Glycaemic Control in Type …
Original Article Open Access

Serum Adenosine Deaminase as a Biomarker of Glycaemic Control in Type 2 Diabetes Mellitus: A Comparative Cross-Sectional Study with HbA1c

,,
Annals of Medicine and Medical SciencesVol. 05, No. 09, (2026) September 24, 2026pp. 2293 - 2299
466 238
Download PDF

Abstract

Background: Type 2 diabetes mellitus (T2DM) is accompanied by chronic low-grade inflammation and altered T-lymphocyte function. Adenosine deaminase (ADA), central to purine metabolism and lymphocyte maturation, has been proposed as an inexpensive marker of glycaemic burden, but published data are inconsistent. Objectives: To compare serum ADA between T2DM patients and non-diabetic controls, examine its correlation with HbA1c, and assess its diagnostic performance. Methods: In this cross-sectional study in Prayagraj, India, 50 adults with T2DM and 50 healthy controls were enrolled over one year. Those with hepatic or renal disease, infection, malignancy, steroid use or pregnancy were excluded. Glucose, HbA1c, lipids, renal and hepatic parameters, haemogram and serum ADA were measured using Student’s t-test, chi-square test, Mann–Whitney U test, Pearson correlation, ROC analysis and logistic regression. Results: Groups were age-matched (p = 0.953). Serum ADA was higher in T2DM (26.34 ± 6.38 versus 20.33 ± 11.52 U/L; p = 0.002). ADA correlated with HbA1c across the whole sample (r = 0.281; p = 0.005) but not within the diabetic group (r = −0.054; p = 0.711). ROC analysis gave an area under the curve of 0.759, with 92% sensitivity and 60% specificity at a cut-off of 20.68 U/L, and each 1 U/L rise in ADA raised the odds of diabetes by 8.7% (OR 1.087; p < 0.05). Conclusion: Serum ADA is elevated in T2DM and separates diabetic from non-diabetic subjects, but modest specificity and the absent within-group correlation confine it to a supportive role alongside HbA1c.

Keywords

adenosine deaminase type 2 diabetes mellitus glycated haemoglobin glycaemic control biomarker inflammation.

Introduction

Type 2 diabetes mellitus (T2DM) accounts for the overwhelming majority of diabetes worldwide, and the burden continues to shift towards low- and middle-income countries [1]. India carries one of the largest absolute caseloads, and projections suggest the figure will keep climbing over the coming decade [2]. The disorder arises from peripheral insulin resistance, inadequate compensatory insulin secretion and unrestrained hepatic glucose output, operating against a background of genetic susceptibility and adverse lifestyle exposure[3,4].

Glycated haemoglobin (HbA1c) remains the accepted standard for judging long-term glycaemic control, since it reflects average glucose exposure over the preceding eight to twelve weeks [5]. Its limitations are equally well recognised. Assay availability is uneven outside larger laboratories, repeated testing is costly, and conditions such as iron deficiency anaemia, haemoglobinopathy and altered erythrocyte survival distort the result [6]. These constraints sustain interest in cheaper biochemical markers that might track glycaemic burden.

Chronic hyperglycaemia is not merely a metabolic state. It maintains a low-grade inflammatory response in which T lymphocytes, macrophages and pro-inflammatory cytokines interfere with insulin receptor signalling [7][8]. Adenosine deaminase (ADA) sits where this immune activity meets purine metabolism. The enzyme irreversibly deaminates adenosine to inosine and 2′-deoxyadenosine to 2′-deoxyinosine, exists as two isoenzymes (ADA1 and ADA2), and shows its highest activity in T lymphocytes, where it is indispensable for maturation and proliferation [9]. Because ADA governs local adenosine concentration, and adenosine modulates insulin action in adipose tissue and myocardium, a change in enzyme activity carries consequences beyond immunity [10][11]. Experimental work in mice has also implicated ADA in the control of hepatic gluconeogenesis [12].

Raised serum ADA was reported in diabetic patients more than three decades ago [13], and several groups have since confirmed the observation in T2DM cohorts [14][15]. Absolute values differ considerably between studies, and the strength of association with HbA1c has ranged from negligible to strong [16][17]. Whether one isoenzyme fraction is chiefly responsible, and whether ADA adds anything to HbA1c in day-to-day practice, remains unsettled [18]. Treatment may complicate the picture further, since dipeptidyl peptidase-4 inhibitors alter measured ADA activity through the enzyme’s interaction with CD26 [19]. The inflammatory profile of poorly controlled diabetes offers a plausible mechanism for enzyme elevation, but the supporting evidence is heterogeneous and drawn largely from small single-centre series [20][21].

ADA estimation is inexpensive, needs no specialised analytical platform, and is already available in most district-level laboratories in India, where it is used routinely in the workup of tuberculous effusion. If the enzyme tracks glycaemic control reliably, it could act as a supportive marker in settings where HbA1c testing is not easily obtained. We therefore measured serum ADA in patients with T2DM and in non-diabetic controls at a tertiary care hospital in northern India, examined its relationship with HbA1c and blood glucose, and assessed its diagnostic performance.

The study aimed to evaluate serum ADA as a biomarker for glycaemic control in type 2 diabetes mellitus, comparing it with HbA1c. It involved estimating serum ADA levels in both diabetic and non-diabetic subjects, assessing their correlation with HbA1c, and determining the utility of serum ADA in distinguishing between poor (HbA1c > 7%) and good (HbA1c < 7%) glycaemic control.

Methods

This was a hospital-based, cross-sectional analytical study conducted in the Department of General Medicine at Moti Lal Nehru Medical College and the attached Swaroop Rani Nehru Hospital, Prayagraj, a tertiary care teaching institution serving both urban and rural populations of eastern Uttar Pradesh. Recruitment ran over one year (2025–2026). Adults aged 18 years and above attending the outpatient and inpatient services were screened. The diabetic group comprised 50 patients with an established diagnosis of T2DM; the control group comprised 50 apparently healthy non-diabetic individuals drawn from the same hospital population. Participants with liver disease, renal disease, acute or chronic infection, malignancy, current corticosteroid therapy, or pregnancy were excluded, since each of these conditions independently alters serum ADA activity. The Institutional Ethics Committee approved the protocol before recruitment began, and written informed consent was obtained from every participant. Confidentiality was maintained throughout.

The sample size was calculated using the formula n = Z²pq/e², where Z = 1.96, p is the expected prevalence, q = 1 − p and e is the allowable error. Taking a local T2DM prevalence of 14.3% (p = 0.143, q = 0.857) and an allowable error of 7% (e = 0.07), the calculation yielded n = 96.08, rounded up to 100 participants, allocated equally to the two groups. The diabetic group was intended to be stratified by glycaemic control into a controlled subgroup (HbA1c < 7%) and an uncontrolled subgroup (HbA1c > 7%). In the event, every diabetic patient recruited had an HbA1c above 7%, so the planned controlled subgroup could not be constituted and the intended subgroup comparison was not possible.

Each participant underwent a structured clinical evaluation recording age, sex, smoking history, duration of diabetes, current medication and comorbid conditions, followed by general and systemic examination with measurement of body mass index, waist circumference, hip circumference and waist–hip ratio. Venous blood was drawn after an overnight fast of at least eight hours, with a second sample two hours after a standard meal for postprandial glucose. Investigations comprised complete blood count, erythrocyte sedimentation rate, C-reactive protein, fasting and postprandial blood glucose, HbA1c, fasting lipid profile, liver and kidney function tests, serum electrolytes, viral markers (HBsAg, anti-HCV, HIV) and serum ADA. Samples were processed in the institutional central laboratory on the day of collection. Serum ADA activity was estimated by the standard colorimetric method and expressed in U/L. Glucose was measured by the glucose oxidase–peroxidase method, HbA1c by high-performance liquid chromatography, and lipids by enzymatic colorimetric assays on an automated analyser.

Data were compiled in Microsoft Excel and analysed with SPSS 24th version. Continuous variables are presented as mean ± standard deviation and categorical variables as frequencies and percentages. Between-group comparison of continuous variables used the independent-samples Student’s t-test, and the Mann–Whitney U test was applied as a non-parametric check for variables with skewed distribution. Categorical variables were compared by the chi-square test with Yates’ continuity correction. The relationship between serum ADA and glycaemic indices was examined by Pearson correlation, both across the whole sample and within the diabetic group alone. Receiver operating characteristic (ROC) analysis was used to determine the discriminatory value of serum ADA for diabetic status, with the optimal cut-off identified by the Youden index. Binary logistic regression, with diabetic status as the dependent variable, was used to estimate the odds ratio associated with serum ADA. A two-tailed p value below 0.05 was taken as statistically significant.

Results

One hundred participants were studied, 50 with T2DM and 50 non-diabetic controls. The two groups were closely matched for age, with a mean of 55.52 ± 14.11 years in the diabetic group and 55.68 ± 12.66 years in controls (p = 0.953), which removes age as an explanation for any biochemical difference observed later. Men predominated in the diabetic group (76.0% versus 54.0%; p = 0.036), a difference that was statistically significant and reflects the referral pattern of the hospital as much as any biological effect. Smoking and hypertension were commoner among diabetic participants but neither difference reached significance, and no participant in either group was taking corticosteroids (Table 1).

Table 1 Baseline demographic and clinical characteristics of the study groups
T2DM (n = 50) Control (n = 50) Test statistic p value
Age, years (mean ± SD) 55.52 ± 14.11 55.68 ± 12.66 t = −0.060 0.953
Male, n (%) 38 (76.0) 27 (54.0) χ² = 4.396 0.036*
Female, n (%) 12 (24.0) 23 (46.0) — —
Smoking, n (%) 20 (40.0) 16 (32.0) χ² = 0.694 0.405
Hypertension, n (%) 10 (20.0) 5 (10.0) χ² = 1.961 0.161
Steroid use, n (%) 0 (0.0) 0 (0.0) — —

*Statistically significant at p < 0.05.

All three glycaemic measures separated the groups very clearly (Table 2). Fasting glucose averaged 160.42 ± 48.40 mg/dL in the diabetic group against 113.26 ± 11.09 mg/dL in controls, postprandial glucose 206.94 ± 75.86 versus 146.50 ± 24.45 mg/dL, and HbA1c 10.41 ± 2.64% versus 6.02 ± 1.34%, each difference significant at p < 0.001. Every diabetic participant had an HbA1c above 7%, so the cohort represents uncontrolled disease rather than the full spectrum of glycaemic control. Serum ADA was higher in the diabetic group at 26.34 ± 6.38 U/L compared with 20.33 ± 11.52 U/L in controls (t = 3.231; p = 0.002). The standard deviation in the control group was almost twice that of the diabetic group, indicating wider dispersion of ADA values among non-diabetic subjects and a substantial overlap between the two distributions despite the significant difference in means.

Table 2 Comparison of glycaemic parameters and serum adenosine deaminase between the groups
Parameter T2DM (n = 50) Control (n = 50) t value p value
Fasting blood glucose, mg/dL 160.42 ± 48.40 113.26 ± 11.09 6.716 < 0.001**
Postprandial blood glucose, mg/dL 206.94 ± 75.86 146.50 ± 24.45 5.362 < 0.001**
HbA1c, % 10.41 ± 2.64 6.02 ± 1.34 10.507 < 0.001**
Serum ADA, U/L 26.34 ± 6.38 20.33 ± 11.52 3.231 0.002**

**Significant at p < 0.01

Total cholesterol and triglycerides were both higher in the diabetic group, as expected in poorly controlled disease (Table 3). HDL cholesterol was also higher among diabetic participants, an unusual direction of effect for this comparison. LDL cholesterol did not differ. Blood urea and serum creatinine were higher in the diabetic group but neither reached significance, consistent with the exclusion of established renal disease at recruitment. Among liver enzymes, AST was significantly raised in the diabetic group while ALT was essentially identical between groups; the large standard deviations for both enzymes indicate a small number of participants with markedly abnormal values driving the group means. ESR was modestly but significantly higher in diabetic participants, in keeping with low-grade systemic inflammation, whereas haemoglobin showed only a non-significant difference.

Table 3 Comparison of lipid, renal, hepatic and haematological parameters between the groups
Parameter T2DM (n = 50) Control (n = 50) t value p value
Total cholesterol, mg/dL 163.24 ± 40.52 139.32 ± 26.79 3.482 < 0.001**
Triglycerides, mg/dL 173.92 ± 84.40 136.54 ± 72.40 2.377 0.019*
HDL cholesterol, mg/dL 61.04 ± 20.02 52.76 ± 14.72 2.356 0.021*
LDL cholesterol, mg/dL 80.52 ± 22.94 75.80 ± 15.15 1.214 0.228
Blood urea, mg/dL 63.33 ± 47.51 48.20 ± 32.87 1.850 0.067
Serum creatinine, mg/dL 1.48 ± 1.06 1.21 ± 0.83 1.366 0.175
AST (SGOT), IU/L 114.66 ± 113.49 65.80 ± 70.63 2.584 0.011*
ALT (SGPT), IU/L 55.47 ± 104.29 55.99 ± 64.75 −0.030 0.976
Haemoglobin, g/dL 12.09 ± 1.93 11.33 ± 2.30 1.806 0.074
ESR, mm/h 21.68 ± 4.08 19.64 ± 5.38 2.135 0.035*

*p < 0.05; **p < 0.01

Across the pooled sample of 100 participants, serum ADA showed a weak positive correlation with HbA1c (r = 0.281; p = 0.005) and a still weaker correlation with fasting glucose that only just reached significance (r = 0.198; p = 0.049). Neither association persisted when the analysis was restricted to the 50 diabetic participants, where the correlation with HbA1c was effectively nil and in the negative direction (r = −0.054; p = 0.711) and that with fasting glucose was non-significant (r = 0.140; p = 0.331). The pattern in Table 4 indicates that the overall correlation is generated by the separation between diabetic and non-diabetic subjects rather than by a graded relationship between enzyme activity and glycaemic burden within diabetic patients.

Table 4 Correlation between serum ADA and glycaemic parameters
Correlation n Pearson r p value
ADA versus HbA1c, all subjects 100 0.281 0.005**
ADA versus HbA1c, T2DM only 50 −0.054 0.711
ADA versus fasting glucose, all subjects 100 0.198 0.049*
ADA versus fasting glucose, T2DM only 50 0.140 0.331

*p < 0.05; **p < 0.01.

ROC analysis gave an area under the curve of 0.759 (95% CI 0.661–0.858; p < 0.001), which falls in the range conventionally described as acceptable to good discrimination. At the cut-off identified by the Youden index, ADA above 20.68 U/L, sensitivity was 92.0% but specificity only 60.0%, meaning that two of every five non-diabetic participants were misclassified as test-positive. Binary logistic regression confirmed serum ADA as a significant predictor of diabetic status, with an odds ratio of 1.087 per 1 U/L increase (B = 0.083; p < 0.05), equivalent to an 8.7% rise in the odds of diabetes for each unit increment (Table 5). The regression model included ADA alone and was not adjusted for age, sex or other covariates.

Table 5 Diagnostic accuracy and predictive value of serum adenosine deaminase
Analysis Statistic Value
ROC Area under the curve 0.759
ROC Standard error 0.050
ROC 95% confidence interval 0.661–0.858
ROC p value < 0.001
ROC Optimal cut-off > 20.68 U/L
ROC Sensitivity 92.0%
ROC Specificity 60.0%
Logistic regression Odds ratio (per 1 U/L) 1.087
Logistic regression Coefficient (B) 0.083
Logistic regression p value < 0.05

ROC, receiver operating characteristic.

Because several variables were skewed, the principal comparisons were repeated with the Mann–Whitney U test (Table 6). Serum ADA, fasting glucose and HbA1c remained highly significant, and triglycerides retained significance. Blood urea remained non-significant, matching the parametric result. The agreement between the two approaches indicates that the findings are not artefacts of departures from normality.

Table 6 Mann–Whitney U test: non-parametric validation of key comparisons
Variable U statistic p value
Serum ADA 1898 < 0.001**
Fasting blood glucose 2182 < 0.001**
HbA1c 2412 < 0.001**
Triglycerides 1599 0.016*
Blood urea 1098 0.295

*p < 0.05; **p < 0.01.

Discussion

The central finding of this study is straightforward: serum ADA activity was significantly higher in patients with type 2 diabetes than in non-diabetic controls, and across the sample as a whole the enzyme tracked HbA1c. Beyond that, the picture is more qualified than the group means alone suggest. The correlation with HbA1c disappeared once the analysis was confined to diabetic patients; specificity at the optimal cut-off was only 60%, and the spread of ADA values in the control group was wide. These caveats are important for determining how the marker should be used in practice, and we will revisit them later in the discussion.

The two groups were well matched for age, which strengthens the biochemical comparisons. The excess of men in the diabetic group is consistent with hospital-based series from South Asia, where men present more often to tertiary care and where central obesity and tobacco use are more common among men [3]. Smoking and hypertension were more frequent in diabetic participants without reaching significance, and no participant was on corticosteroids, so the two commonest confounders of ADA activity in an outpatient population were reasonably controlled.

Glycemic indices were, unsurprisingly, markedly deranged in the diabetic group. What deserves emphasis is that every diabetic participant had an HbA1c above 7%, with a group mean of 10.41%. This is a cohort of uniformly uncontrolled disease. Comparable degrees of derangement have been reported in other hospital-based series [22][24], and studies that stratified patients by degree of control found the highest ADA values in the poorly controlled strata [25, 26]. A positive relationship between ADA activity and both fasting glucose and HbA1c has likewise been described [27].

Our mean serum ADA of 26.34 U/L in the diabetic group sits in the middle of the published range. Attention was first drawn to raised ADA in diabetes three decades ago [13], and the finding was subsequently confirmed in a dedicated T2DM cohort [14]. The largest of the comparable series reported 23.1 U/L in diabetic patients against 18.6 U/L in controls, values very close to ours [19]; another reported 32.06 ± 17.09 U/L against 19.28 ± 5.59 U/L [22], and a third found elevated ADA in both controlled and uncontrolled diabetic subgroups [28]. Far greater separation has been described elsewhere: 49.24 U/L in uncontrolled diabetics against 10.55 U/L in controls [16], 22.35 against 4.38 U/L [29], marked elevation confined largely to the poorly controlled group [30], and 38.97 against 20.05 U/L [31]. A three-group comparison reported 26.28 U/L in well-controlled and 35.46 U/L in poorly controlled patients against 12.03 U/L in controls [32]. This spread is too wide to be explained by biology alone. Different assay methods, notably the Giusti colorimetric technique versus kinetic enzymatic assays, yield different absolute values, and control populations have been defined variably; the earlier literature has already noted this problem of comparability [33]. It argues strongly against transferring any published cut-off from one laboratory to another.

The correlation data require careful reading. Serum ADA correlated with HbA1c at r = 0.281 across all 100 participants, a weak association that nonetheless reached significance. Within the 50 diabetic patients, however, the correlation was −0.054 and clearly non-significant. Since the whole-sample correlation is calculated across two groups that differ in both variables by design, it largely reflects group membership rather than a dose–response relationship. Several earlier reports show a similar pattern of modest coefficients: r = 0.208 [19], a positive but limited association [21], r = 0.290 [22], and r = 0.03 in controlled patients versus 0.28 in uncontrolled patients [28]. Stronger correlations have been published—r = 0.585 [23], r = 0.64 [34], and r = 0.388 [24]—and strong coefficients in two further series [35, 36], but reported figures of r = 0.949 and r = 0.981 [32] are implausibly high for a biological association of this kind and should be treated with caution. Our data are better read as showing that ADA distinguishes diabetic from non-diabetic subjects than as showing that it grades severity within diabetes. The narrow HbA1c range in our diabetic group, all above 7%, restricts the variance available for correlation and may account for the null within-group result.

Mechanistically, the elevation is compatible with what is known about ADA. The enzyme is most active in T lymphocytes and is required for their maturation, and T-cell activation is a recognized feature of the chronic inflammatory state in T2DM [7, 8]. ADA also binds CD26 (dipeptidyl peptidase-4) on the lymphocyte surface, a co-stimulatory interaction linking the enzyme directly to immune activation and, incidentally, to a widely used class of antidiabetic drug [37]. By degrading adenosine, ADA reduces the local concentration of a nucleoside that inhibits lipolysis in adipocytes and modulates insulin sensitivity [11], and murine work suggests a role in hepatic gluconeogenesis [12]. A further dimension is added by the reported inverse relationship between serum ADA and islet beta-cell function in a series of 1,573 patients with T2DM [38]. Whether raised ADA is a consequence of hyperglycemia, a contributor to insulin resistance, or simply a bystander marker of inflammation cannot be settled by a cross-sectional design.

Total cholesterol and triglycerides were higher in the diabetic group, which matches the dyslipidemia expected in uncontrolled T2DM and the findings of earlier case–control series [23, 34]. ADA has been reported to correlate positively with cholesterol, triglycerides, LDL, and VLDL and negatively with HDL [39]. Our HDL result runs against the expected outcome: HDL was significantly higher in the diabetic group, the opposite of what is usually seen. The most likely explanation is treatment, since diabetic patients attending a tertiary center are frequently on statins or fibrates while controls are not. We did not record lipid-lowering therapy, so the hypothesis cannot be verified, and the HDL finding should be regarded as uninterpretable rather than as a genuine biological observation.

Renal and hepatic findings were largely as anticipated. Urea and creatinine trended higher in diabetic patients without reaching significance, which is consistent with the exclusion of overt renal disease. AST was significantly raised while ALT was not, and the enormous standard deviations for both enzymes point to a few outlying values rather than a general shift; hepatic steatosis associated with insulin resistance is a plausible contributor but was not assessed. These parameters are relevant to the ADA literature because organ dysfunction has been linked to enzyme levels: ADA has been found to correlate inversely with eGFR [40], to emerge as an independent correlate of diabetic kidney disease in a retrospective series of 5,485 patients [41], and to be associated with diabetic nephropathy in a prospective cohort with microvascular complications [42]. The significantly higher ESR in our diabetic group supports the presence of low-grade systemic inflammation, the pathophysiological context in which raised ADA is usually interpreted [20].

The diagnostic analysis is where the practical limits of the marker become clear. An area under the curve of 0.759 represents acceptable but not strong discrimination. A sensitivity of 92% at a cut-off of 20.68 U/L looks attractive, but the accompanying specificity of 60% means that 20 of the 50 control participants would have been labeled test-positive. A threshold this low also sits within the normal reference interval used by many laboratories, which limits its transferability. The cut-offs reported elsewhere differ substantially: a threshold of ≥14.2 U/L has been derived for diagnosing diabetic peripheral neuropathy, with 59.6% sensitivity and 75.5% specificity [43]. The diagnostic value of total ADA has also been shown to depend heavily on which isoenzyme fraction is measured [18], and a similar point has been made in demonstrating that ADA1 and ADA2 behave differently in T2DM [17]. Reporting isoenzyme fractions rather than total activity may be necessary before any cut-off can be regarded as usable. Our logistic regression, which gave an odds ratio of 1.087 per U/L, was unadjusted; because ADA was the only variable in the model, the estimate cannot be described as independent of age, sex, or adiposity, and we have avoided that word in interpreting it.

Taken together, these results place serum ADA as a supplementary rather than a substitute marker. It is inexpensive, needs no dedicated platform, and is already available in most secondary-level laboratories in India, which is a genuine advantage in settings where HbA1c is difficult to obtain. But it does not grade glycemic control within diabetic patients in our data, and its specificity is too low for use as a screening test in isolation. Its more intriguing potential may lie elsewhere. Raised ADA has been found in obese subjects with and without diabetes [44]; a four-group study reported the highest values in obese diabetic patients, with values already elevated in obese non-diabetics [45]; and ADA has been shown to correlate with fasting insulin and HOMA-IR [46]. If ADA reflects insulin resistance and immune activation rather than glycemia as such, then insulin resistance, and not glycemic control, may be the more appropriate outcome against which to test it.

This was a single-center study with a relatively small sample size, which may limit the generalizability of the findings to broader populations. Additionally, the absence of surgical or arthroscopic confirmation in all cases may have limited the ability to determine the true diagnostic accuracy of ultrasound and MRI for all musculotendinous shoulder pathologies.

Conclusion

Serum adenosine deaminase was significantly higher in patients with type 2 diabetes mellitus than in non-diabetic controls, and correlated weakly with HbA1c across the sample as a whole. The enzyme discriminated diabetic from non-diabetic subjects with acceptable accuracy and high sensitivity, but specificity was modest and no correlation with HbA1c was demonstrable within the diabetic group, whose members all had uncontrolled disease. On this evidence serum ADA is best regarded as an inexpensive supportive marker that may complement HbA1c in resource-constrained settings, not as a replacement for it. Larger multicentre studies that include patients across the full range of glycaemic control, that measure ADA isoenzymes separately, and that adjust for adiposity, disease duration and drug therapy are needed before the marker can be recommended for clinical use.

Declarations

Conflict of Interest

None

Author contributions

SK conceived the study, collected the data and drafted the manuscript. JS supervised the work and revised the manuscript critically. PG, AS and RKT contributed to study design, data interpretation and revision. All authors approved the final version and agree to be accountable for the work.

Funding Statement

None

Conflict of interest

The authors declare that they have no competing interests.

Ethical approval

Approved by the Institutional Ethics Committee, Moti Lal Nehru Medical College, Prayagraj [approval number and date to be inserted].

References

  1. Ong KL, Stafford LK, McLaughlin SA, Boyko EJ, Vollset SE, Smith AE, et al. Global, regional, and national burden of diabetes from 1990 to 2021, with projections of prevalence to 2050: a systematic analysis for the Global Burden of Disease Study 2021. Lancet. 2023;402(10397):203-34. DOI ↗ Google Scholar ↗
  2. Magliano DJ, Boyko EJ. IDF Diabetes Atlas. 10th ed. Brussels: International Diabetes Federation; 2021. DOI ↗ Google Scholar ↗
  3. Chatterjee S, Khunti K, Davies MJ. Type 2 diabetes. Lancet. 2017;389(10085):2239-51. DOI ↗ Google Scholar ↗
  4. Galicia-Garcia U, Benito-Vicente A, Jebari S, Larrea-Sebal A, Siddiqi H, Uribe KB, et al. Pathophysiology of type 2 diabetes mellitus. Int J Mol Sci. 2020;21(17):6275. DOI ↗ Google Scholar ↗
  5. ElSayed NA, Aleppo G, Aroda VR, Bannuru RR, Brown FM, Bruemmer D, et al. 2. Classification and diagnosis of diabetes: Standards of Care in Diabetes—2023. Diabetes Care. 2023;46(Suppl 1):S19-40. Google Scholar ↗
  6. Radin MS. Pitfalls in hemoglobin A1c measurement: when results may be misleading. J Gen Intern Med. 2014;29(2):388-94. Google Scholar ↗
  7. Xia C, Rao X, Zhong J. Role of T lymphocytes in type 2 diabetes and diabetes-associated inflammation. J Diabetes Res. 2017;2017:6494795. Google Scholar ↗
  8. Berbudi A, Rahmadika N, Tjahjadi AI, Ruslami R. Type 2 diabetes and its impact on the immune system. Curr Diabetes Rev. 2020;16(5):442-9. DOI ↗ Google Scholar ↗
  9. Gao ZW, Yang L, Liu C, Wang X, Guo WT, Zhang HZ, et al. Distinct roles of adenosine deaminase isoenzymes ADA1 and ADA2: a pan-cancer analysis. Front Immunol. 2022;13:903461. Google Scholar ↗
  10. Manjunath S, Sakhare PM. Adenosine and adenosine receptors: newer therapeutic perspective. Indian J Pharmacol. 2009;41(3):97-105. Google Scholar ↗
  11. Ohisalo JJ. Effects of adenosine on lipolysis in human subcutaneous fat cells. J Clin Endocrinol Metab. 1981;52(2):359-63. Google Scholar ↗
  12. Ding Z, Ge W, Xu X, Xu X, Sun Q, Xu X, et al. A crucial role of adenosine deaminase in regulating gluconeogenesis in mice. J Biol Chem. 2024;300(7):107425. Google Scholar ↗
  13. Hoshino T, Yamada K, Masuoka K, Tsuboi I, Itoh K, Nonaka K, et al. Elevated adenosine deaminase activity in the serum of patients with diabetes mellitus. Diabetes Res Clin Pract. 1994;25(2):97-102. Google Scholar ↗
  14. Kurtul N, Pence S, Akarsu E, Kocoglu H, Aksoy Y, Aksoy H. Adenosine deaminase activity in the serum of type 2 diabetic patients. Acta Medica (Hradec Kralove). 2004;47(1):33-5. Google Scholar ↗
  15. Khemka VK, Bagchi D, Ghosh A, Sen O, Bir A, Chakrabarti S, et al. Raised serum adenosine deaminase level in nonobese type 2 diabetes mellitus. Scientific World Journal. 2013;2013:404320. Google Scholar ↗
  16. Niraula A, Thapa S, Kunwar S, Lamsal M, Baral N, Maskey R. Adenosine deaminase activity in type 2 diabetes mellitus: does it have any role? BMC Endocr Disord. 2018;18(1):58. Google Scholar ↗
  17. Sapkota LB, Thapa S, Subedi N. Correlation study of adenosine deaminase and its isoenzymes in type 2 diabetes mellitus. BMJ Open Diabetes Res Care. 2017;5(1):e000357. Google Scholar ↗
  18. Larijani B, Heshmat R, Ebrahimi-Rad M, Khatami S, Valadbeigi S, Saghiri R. Diagnostic value of adenosine deaminase and its isoforms in type II diabetes mellitus. Enzyme Res. 2016;2016:9526593. Google Scholar ↗
  19. Lee JG, Kang DG, Yu JR, Kim Y, Kim J, Koh G, et al. Changes in adenosine deaminase activity in patients with type 2 diabetes mellitus and effect of DPP-4 inhibitor treatment on ADA activity. Diabetes Metab J. 2011;35(2):149-58. Google Scholar ↗
  20. Tsalamandris S, Antonopoulos AS, Oikonomou E, Papamikroulis GA, Vogiatzi G, Papaioannou S, et al. The role of inflammation in diabetes: current concepts and future perspectives. Eur Cardiol. 2019;14(1):50-9. Google Scholar ↗
  21. Kaur A, Kukreja S, Malhotra N. Serum adenosine deaminase activity and its correlation with glycated haemoglobin levels in patients of type 2 diabetes mellitus. J Clin Diagn Res. 2012;6(2):252-6. Google Scholar ↗
  22. Ramani NS, Krishnamurthy N, Raghavendra Prasad BN, Ashakiran S, Sumathi ME, Harish R. Role of adenosine deaminase to predict glycemic status in type 2 diabetes mellitus. J Clin Biomed Sci. 2012;2(3):126-33. Google Scholar ↗
  23. Pinnelli VB, Jayashree CA, Mohanty S, Anjaneyulu G, Mathai MM, Rao DS. Elevated levels of serum adenosine deaminase in type 2 diabetes mellitus patients. Int J Res Med Sci. 2016;4(1):131-4. Google Scholar ↗
  24. Singh S, Suri A, Sinha M, Fiza B. A study on the correlation of adenosine deaminase and glycated haemoglobin in the patients of type 2 diabetes mellitus. Int J Med Biomed Stud. 2019;3(12):43-50. Google Scholar ↗
  25. Tonde D, Mahajan A. Study of serum ADA levels in type 2 diabetes mellitus patients. Int J Sci Res. 2020;9(6):1-4. Google Scholar ↗
  26. Bhowmik M, Dey PK, Akhter QS. Correlation of serum ADA with glycated haemoglobin (HbA1c) in type 2 diabetes mellitus patients: a single centre study in Bangladesh. Ann Int Med Dent Res. 2022;8(3):250-5. Google Scholar ↗
  27. Talebi Mehrdar M, Ebadi G. Increased levels of serum ADA enzyme and increased risk of T cell activation markers in type 2 diabetes. Cell Mol Biomed Rep. 2024;4(3):159-67. Google Scholar ↗
  28. Singh P, Khan S, Mittal RK. Adenosine deaminase activity and its relation with glycated hemoglobin and uric acid in type 2 diabetic patients. Iran J Diabetes Obes. 2013;5(1):1-6. DOI ↗ Google Scholar ↗
  29. Muthiah B, Anuradha LS, Peddi NK, Menon VB. Adenosine deaminase as marker of insulin resistance. Int J Res Med Sci. 2016;4(7):2972-9. Google Scholar ↗
  30. Kiranmai P. Serum adenosine deaminase levels in patients with type 2 diabetes mellitus and its relation with fasting blood glucose and glycated hemoglobin levels. MedPulse Int J Biochem. 2018;5(3):100-4. Google Scholar ↗
  31. Yengkhom US, Noarem S, Devi PJ, Kshetrimayum V, Marbaniang RS, Debbarma U. Serum levels of adenosine deaminase and insulin in type 2 diabetes mellitus. Int J Biochem Res Rev. 2020;29(5):18-23. Google Scholar ↗
  32. Hassan FK, Genga S. Serum adenosine deaminase and its association with type 2 diabetes mellitus. Int J Acad Med Pharm. 2024;6(2):251-5. Google Scholar ↗
  33. Warrier AC, Rao NY, Kulpati DS, Mishra TK, Kabi BC. Evaluation of adenosine deaminase activity and lipid peroxidation levels in diabetes mellitus. Indian J Clin Biochem. 1995;10(1):9-13. Google Scholar ↗
  34. Vijayalakshmi P, Supriya. Altered levels of serum adenosine deaminase in type 2 diabetes mellitus. Natl J Lab Med. 2019;8(3):BO09-11. Google Scholar ↗
  35. Hemalatha D, Manjula Devi J, Kalaiselvi PR. Study of serum adenosine deaminase level in type 2 diabetes mellitus and its correlation with glycemic control. J Pharm Res Int. 2021;33(23A):39-49. Google Scholar ↗
  36. Nayani SSS, Rao KS, Prasad BV. Estimation of serum adenosine deaminase levels and correlation with glycemic status in diabetes mellitus patients. Eur J Cardiovasc Med. 2024;14:827-30. Google Scholar ↗
  37. Richard E, Arredondo-Vega FX, Santisteban I, Kelly SJ, Patel DD, Hershfield MS. The binding site of human adenosine deaminase for CD26/dipeptidyl peptidase IV: the Arg142Gln mutation impairs binding to CD26 but does not cause immune deficiency. J Exp Med. 2000;192(9):1223-36. Google Scholar ↗
  38. Cao J, Wang H, Su JB, Wang XQ, Zhang DM, Wang XH, et al. Inverse relationship between serum adenosine deaminase levels and islet beta cell function in patients with type 2 diabetes. Diabetol Metab Syndr. 2021;13(1):54. DOI ↗ Google Scholar ↗
  39. Mahagaonkar-Badarkhe P, Badarkhe P, Sonune S, Sonune S. Adenosine deaminase in complicated and uncomplicated type II diabetes mellitus and its association with glycaemic status, lipid profile and serum uric acid. Int J Curr Res. 2025;17(5):33011-5. Google Scholar ↗
  40. Usman M, Shah F. Association of serum adenosine deaminase levels with acute kidney injury in diabetic patients. Biol Clin Sci Res J. 2024;2024(1):1236. Google Scholar ↗
  41. Zhang PR, Peng GL, Zhang YL, Zhang XY, Sun LJ, Long M. Correlation between serum adenosine deaminase levels and the risk of diabetic kidney disease in patients with type 2 diabetes mellitus. Chin J Clin Res. 2025;38(9):1319-23. Google Scholar ↗
  42. Raut SS, Wanjari A, Deolikar V, Kumar S, Acharya S, Rathod N, et al. A prospective cross-sectional study on the correlation of adenosine deaminase and HbA1c with microvascular complications in type 2 diabetes mellitus at a tertiary care hospital in Central India. Cureus. 2024;16(10):e70732. Google Scholar ↗
  43. Yu C, Zhuang L, Xu F, Zhao LH, Wang XH, Wang CH, et al. Increased levels of serum adenosine deaminase and increased risk of diabetic peripheral neuropathy in type 2 diabetes. Front Endocrinol (Lausanne). 2022; 13:997672. Google Scholar ↗
  44. Jain AR, Kumar A, Saxena T, Arya A, Jadhav AA. Is adenosine deaminase level a predictive biochemical marker of type II diabetes mellitus in obese Indian subjects? Int J Clin Biochem Res. 2019;6(2):165-9. Google Scholar ↗
  45. Srinivas R, Meghana BS, Balachandra G. Assessment of serum ADA levels as a predictive marker for diabetes mellitus in obese individuals: a case–control study at a tertiary care centre. Int Med. 2025;11(1):1-5. Google Scholar ↗
  46. Medikonda R, Bindra M. Association of adenosine deaminase with insulin resistance and pancreatic enzymes in type 2 diabetes mellitus. Front Health Inform. 2024;13(8):1973-81. Google Scholar ↗