Nuclear Medicine Imaging in Diabetes Mellitus: Current Applications and Emerging Molecular Targets

Current Nuclear Medicine Applications in Diabetes

INuclear medicine plays a crucial role in the diagnosis and management of diabetes through various imaging techniques and therapeutic approaches. Nuclear imaging modalities such as positron emission tomography (PET) and single-photon emission computed tomography (SPECT) can be utilized to detect diabetes-related complications. These include identifying areas of inflammation or infection, which are common in diabetic patients, assessing pancreatic beta-cell function, evaluating cardiovascular risk, monitoring renal impairment, and diagnosing diabetic foot complications.

Radiopharmaceuticals can also be used for targeted therapy in diabetes management. For instance, targeted radionuclide therapy using radiolabeled somatostatin analogs may be applied in treating neuroendocrine tumors associated with conditions such as multiple endocrine neoplasia type 1, which may include insulinomas that contribute to insulin dysregulation in some diabetic patients.

Diabetic foot and infection imaging

Diabetic foot infection (DFI) is one of the most frequent and serious complications of DM, contributing significantly to lower-extremity amputation risk and hospital admissions. It often arises due to a combination of peripheral neuropathy, vascular insufficiency, and impaired cellular immunity. Accurate diagnosis is critical to distinguish between soft tissue infection, osteomyelitis, and neuroarthropathy (Charcot joint).

Diabetic foot

In diabetic foot, the University of Texas Diabetic Wound Classification is routinely used to determine the stage of diabetic foot ulcers. This classification assesses and stages ulcers based on their depth and the presence of infection or ischemia.

Nuclear medicine techniques for diagnosing infection in diabetic foot include three-phase bone scintigraphy and PET/CT with ¹⁸F-FDG.^[4] An illustrative example of SPECT imaging in Charcot joint is shown in Fig 1.^[5]

Close

Fig 1:

Three-phase bone scan in a patient with diabetic foot. Clinically, (a) The vascular phase shows increased vascularisation, and (b) The soft tissue phase image demonstrates increased uptake of methylene diphosphonate in the soft tissues of the right foot, particularly in the forefoot, indicative of cellulitis. The skeletal phase image shows mild increased MDP uptake in the midfoot of the right foot, suggesting possible early Charcot neuropathy^[5] MDP: Methylene diphosphonate, LT: Left; RT: Right

Export to PPT

Bone scintigraphy

Three-phase BS (TPBS), particularly with delayed imaging at 24 h, is a highly sensitive test, although not a specific imaging modality for diabetic foot osteomyelitis.^[6,7] TPBS using ^99mTc-MDP is a classic nuclear medicine technique for osteomyelitis detection. It evaluates perfusion, blood pool activity, and delayed bone uptake to assess hypervascularity and increased bone metabolism.

However, studies by Larcos et al. showed that although bone scan has high sensitivity (~93%), it suffers from low specificity (~43%) due to increased uptake in fractures, Charcot joints, or sterile inflammatory conditions.^[8] Similarly, Johnson et al. (1996) reported that TPBS alone cannot reliably differentiate infection from neuropathic complications.^[9] An illustrative example of SPECT imaging in Charcot joint is presented in Fig 2.^[10]

Close

Fig 2:

Dual-isotope SPECT/CT of the foot. (a) 99mTc-HDP planar: focal uptake in posterior right calcaneus (red arrow). (b) ¹¹¹In-WBC planar: corresponding uptake (red arrow), consistent with infection. (c) 99mTc-SC planar: no corresponding uptake, supporting osteomyelitis. (d) 99mTc-HDP SPECT: focal uptake in posterior right calcaneus (red arrow). (e) ¹¹¹In-WBC SPECT: concordant uptake (red arrow). (f) Fused SPECT/CT: localisation to posterior right calcaneus. Additional uptake in the left calcaneus and talus (green arrows in a–f) is consistent with Charcot joint without evidence of infection. DI SPECT/CT:Dual-Isotope single photon emission computed tomography /computed tomography, 9mTc-HDP: Technetium-99m hydroxymethylene diphosphonate, ¹¹¹In-WBC = Indium-111 white blood cells, 99mTc-SC: Technetium-99m sulfur colloid

Export to PPT

¹¹¹Indium-oxine labeled white blood cell

¹¹¹In-oxine labels all cell types indiscriminately, including platelets and red blood cells. Oxine forms a lipophilic complex with ¹¹¹In, which passively diffuses into leukocyte membranes. Once inside, 111In separates and binds cytoplasmic components, while the oxine is excreted.^[11]

White blood cell (WBC) scintigraphy remains a cornerstone in the functional imaging of infection, particularly in cases of diabetic foot osteomyelitis. The technique involves labeling autologous leukocytes with radiotracers, commonly ¹¹¹Indium-oxine or 99mTechnetium-hexamethylpropylene amine oxime (^99mTc-HMPAO), and re-injecting them to track their migration toward infected sites. The utility of ¹¹¹In-labeled WBCs in DFI was demonstrated by Larcos et al., who reported a sensitivity of 79% and a specificity of 78% for the detection of osteomyelitis.^[8] In a larger prospective study, Johnson et al. found that combining ¹¹¹In-WBC scintigraphy with three-phase bone scanning improved sensitivity to 100%, though specificity ranged between 70% and 80%.^[9] This combination proved effective even in the presence of complicating factors such as chronic ulcers or ongoing antibiotic therapy, highlighting its robustness in complex clinical scenarios. However, ¹¹¹In-labeled leukocytes have low sensitivity for spinal infections, including osteomyelitis and discitis. More than half of spinal infections may appear as photopenic (cold) lesions, rather than hot spots.^[11] In addition, 111In-labeled WBCs accumulate in hematopoietically active bone marrow even without infection, reducing sensitivity for chronic osteomyelitis.^[13] An illustrative case of osteomyelitis in the sacroiliac joint is shown in Fig 3.^[11]

Close

Fig 3:

(a) Posterior view of the pelvis from a bone scan shows marked increased activity (arrow) in the central region of the right sacroiliac joint. (b) Indium-111 (¹¹¹In) leukocyte scan of the same patient shows relatively decreased activity (arrow) in the same region. This finding may be observed in leukocyte scans when osteomyelitis occurs in the central skeleton, particularly in the spine. WBC: White blood cell, DP: Methylene diphosphonate

Export to PPT

^99mTechnetium-hexamethylpropyleneamine oxime labeled white blood cell

The role of ^99mTc-HMPAO-labeled WBCs in inflammation imaging offers several advantages compared to ¹¹¹In-labeled leukocytes. 99mTc-HMPAO has become the preferred radiotracer for WBC labeling due to several advantages, including better image resolution, shorter imaging acquisition times, and lower radiation exposure.^[11] According to the meta-analysis by Lauri et al., ^99mTc-HMPAO-labeled WBC scintigraphy achieved a sensitivity of approximately 91% and a specificity of 92%, surpassing older methods and emerging as one of the most accurate nuclear techniques for diagnosing diabetic foot osteomyelitis.^[12] Its ability to distinguish active infection from sterile inflammation, such as Charcot neuroarthropathy or postsurgical changes, makes it particularly useful in diabetic patients, whose feet often exhibit overlapping structural abnormalities that may complicate interpretation with anatomic imaging alone.

Myocardial perfusion and silent ischemia

Silent coronary artery disease (CAD) is a frequent yet often underdiagnosed condition in patients with type 2 DM (T2DM), largely due to autonomic neuropathy that blunts anginal symptoms. The reported prevalence of silent CAD ranges between 2.5% and 11% in nondiabetic individuals, but increases dramatically to 12% to 57% among diabetic patients. Notably, in asymptomatic diabetic individuals, abnormal myocardial perfusion that induces ischemia has been observed in approximately 37% of case.^[14]

A seminal prospective study by Mohagheghie et al. assessed 83 asymptomatic T2DM patients with ≥5 years of disease duration using ^99mTc-sestamibi SPECT myocardial perfusion imaging (MPI).^[15] The study revealed that 30% of patients had perfusion defects, mostly reversible, indicative of silent ischemia. In patients who underwent further coronary angiography, 48% had significant CAD. Importantly, glycated hemoglobin (HbA1c) was an independent predictor of perfusion abnormality (odds ratio = 1.70; 95% confidence interval: 1.07–2.71; P = 0.02), suggesting that poor glycemic control may contribute to subclinical myocardial ischemia even in the absence of symptoms.^[15] In addition, longer diabetes duration (>10 years) has also been associated with a higher frequency of abnormal SPECT MPI findings.^[14]

In a complementary study, Nasr et al. examined the association between HbA1c levels and both myocardial perfusion and left ventricular (LV) function using gated SPECT MPI in 200 patients (132 with diabetes).^[16] Although no significant relationship was found between HbA1c and Summed Stress Score, Summed Difference Score (SSS, SDS), patients with HbA1c >6.5% demonstrated significantly lower LV ejection fraction (LVEF), higher end-systolic volume, and more frequent wall motion abnormalities (all P < 0.05).^[3] These findings suggest that chronic hyperglycemia may exert a greater influence on LV functional remodeling than on perfusion itself, likely through pathways involving oxidative stress, endothelial dysfunction, and myocardial fibrosis.^[16]

Collectively, these studies underscore the importance of integrating MPI into cardiovascular risk assessment protocols for asymptomatic T2DM patients, particularly those with poor glycemic control or long disease duration. The evidence supports the dual role of MPI in detecting silent ischemia and evaluating early myocardial dysfunction, thereby providing a window of opportunity for early therapeutic intervention aimed at preventing major adverse cardiac events.

Nephropathy and renal function

Nuclear medicine renal imaging offers valuable insights for early detection and monitoring of diabetic nephropathy, particularly when structural changes or biochemical abnormalities have not yet emerged. Several techniques have been evaluated:

The study by Frieske et al. demonstrated that parametric renal scintigraphy using ^99mTc-EC could reveal subtle defects in renal parenchymal function and prolonged transit times, even when conventional scintigraphy and glomerular filtration rate (GFR) values remained within normal limits. The significant differences in mean transit time and clearance image patterns between diabetic patients and healthy controls support its role in early nephropathy detection before overt renal impairment occurs.^[17]

Meanwhile, Klinkhammer et al. highlighted the utility of PET/MRI molecular imaging to dynamically assess SGLT2 transporter activity and tubular reabsorption. Their findings suggest that functional renal alterations in T2DM can be visualized in vivo, and these changes are reversible with therapy, making this technique especially promising for monitoring therapeutic response and guiding precision interventions in diabetic nephropathy.^[18]

In contrast, Rodby et al. provided clinical validation for ⁹⁹mTc-DTPA renal scintigraphy as a practical tool to estimate GFR. With a high correlation to the gold-standard iothalamate clearance method (r = 0.80), this noninvasive approach allows for serial monitoring of renal function, reducing the need for cumbersome blood or urine sampling.^[19]

Taken together, these studies emphasize that renal scintigraphy is not limited to measuring GFR but can also reveal functional delays in tracer uptake and transit time, assisting in the early diagnosis, risk stratification, and treatment monitoring of diabetic nephropathy.

Peripheral Arterial Disease (PAD)

Peripheral arterial disease (PAD) is characterized by progressive occlusive atherosclerotic disease in the lower extremities. Its global prevalence ranges from 3% to 10% in the general population and increases to 15%–20% in individuals over 70 years of age. Patients with PAD have an approximately threefold increased risk of cardiovascular mortality, including myocardial infarction and stroke, compared to those without PAD.^[20] The risk is notably higher in patients with DM, where PAD often develops earlier and progresses more aggressively, frequently without the typical symptoms due to associated neuropathy.

Nuclear medicine imaging, including SPECT/CT and PET/CT, has emerged as an important noninvasive modality to assess both the extent and severity of PAD, offering unique insights into blood flow, tissue perfusion, and molecular markers of vascular pathology.

SPECT/CT using radiotracers such as ^99mTc-tetrofosmin is widely used to visualize peripheral limb perfusion, allowing clinicians to assess the degree of ischemia and monitor responses to interventions. Meanwhile, PET imaging with ¹⁸F-FDG and ¹⁸F-sodium fluoride (¹⁸F-NaF) enables molecular characterization of atherosclerosis, including arterial inflammation and early-stage calcification, particularly in the carotid, coronary, aortic, and peripheral limb arteries.^[21,22]

¹⁸F-FDG PET/CT detects arterial wall inflammation by targeting metabolically active macrophages within atherosclerotic plaques. In diabetic patients, arterial FDG uptake is significantly elevated even during early disease stages and correlates with arterial stiffness and central systolic blood pressure, indicating a state of subclinical systemic inflammation. This was demonstrated by de Boer et al. in individuals with early type 2 diabetes.^[23]

Complementing this, ¹⁸F-NaF PET/CT highlights microcalcification activity, a key feature of early atherosclerotic plaque development. Chou et al. reported that NaF uptake in ischemic lower extremity muscles significantly decreased after revascularization procedures, indicating improvement in tissue perfusion and confirming its potential as a marker of therapeutic response.^[24]

Stacy emphasized that both PET and SPECT imaging can stratify PAD patients into high responders or nonrespondersto revascularization, offering valuable guidance for treatment planning and outcome prediction.^[22]

Importantly, preprocedural FDG uptake has been associated with higher rates of restenosis, as shown by Divakaran et al. and Chowdhury et al., suggesting that vascular inflammation identified through PET/CT may predict poor healing and long-term prognosis, particularly in high-risk diabetic populations.^[25,26]

Collectively, the integration of molecular imaging with PET/CT and SPECT/CT allows for a comprehensive assessment of perfusion status, inflammatory burden, and vascular remodeling, positioning nuclear medicine as a crucial tool in the management of PAD in diabetic patients.

β-cell imaging and glucose metabolism

Inflammation-driven β-cell dysfunction and pancreatic fat infiltration

Horii et al. further contributed to the understanding of diabetes-related pancreatic changes by reporting that pancreatic fat infiltration is associated with islet inflammation and hyperglycemia in T2DM. Chronic low-grade inflammation within the pancreatic islets has been linked to progressive β-cell dysfunction. These structural and inflammatory changes, which can be assessed using molecular imaging, offer a mechanistic explanation for metabolic failure in T2DM beyond insulin resistance (IR) alone.^[28]

VMAT2 imaging and direct quantification of β-cell mass

For more specific β-cell imaging, [¹⁸F]FP-(+)-DTBZ, a PET tracer targeting vesicular monoamine transporter 2 (VMAT2), has shown promising results. Normandin et al. (2018) successfully used this tracer to quantify endogenous β-cell mass in vivo. Their findings revealed significantly lower VMAT2 binding in individuals with type 1 diabetes (T1D) compared to healthy subjects, demonstrating the tracer’s ability to noninvasively assess β-cell loss. This approach holds potential for evaluating β-cell preservation in both type 1 and late-stage type 2 diabetes.^[29]

Future Directions in Nuclear Medicine and Diabetes

β-cell mass and function imaging 18F-FP-DTBZ (VMAT2-targeted imaging) Imaging of pancreatic β-cell mass plays a central role in understanding the pathophysiology and progression of T1D, especially in early detection and evaluation of therapeutic interventions. The vesicular monoamine transporter type 2 (VMAT2) is highly expressed in β-cells and has been utilized as a molecular target for PET imaging. The radiotracer 18F-FP-(+)-DTBZ binds selectively to VMAT2, allowing in vivo quantification of β-cell mass.

Veluthakal and Harris reported that ¹⁸F-FP-DTBZ PET imaging can effectively differentiate individuals with longstanding T1D from healthy subjects, based on the marked reduction in pancreatic uptake.^[32] This method enables noninvasive tracking of β-cell loss, offering potential applications in disease monitoring and evaluating islet transplantation. Furthermore, VMAT2 imaging could serve as an objective biomarker in clinical trials aimed at preserving or restoring β-cell function.^[32]

Freeby et al. conducted cross-sectional and test-retest evaluations of PET with 18F-FP-(+)-DTBZ and found consistent and reproducible binding in healthy controls, with significantly lower uptake in subjects with T1D. Their findings support the utility of this tracer in quantitatively assessing β-cell mass and distinguishing pathological states.^[33]

In parallel, Lin et al. performed whole-body biodistribution and dosimetric studies of 18F-FP-(+)-DTBZ and confirmed favourable kinetics and safety profiles, with pancreas-to-background ratios sufficient for diagnostic utility. However, one notable limitation is the non-specific uptake in exocrine pancreatic tissue and other organs such as the spleen and kidneys, which may reduce image contrast. In addition, interindividual variability in VMAT2 expression requires standardization of quantification protocols.^[34]

Despite these challenges, ¹⁸F-FP-DTBZ PET remains a promising imaging modality for non-invasive evaluation of β-cell mass in T1D and other β-cell disorders. Ongoing advances in tracer refinement, image processing, and hybrid quantification methods are expected to improve specificity and broaden its clinical applicability.

⁶⁸Ga-exendin-4 (glucagon-like peptide-1 receptor-targeted imaging)

The glucagon-like peptide-1 receptor (GLP-1R) is predominantly expressed on the surface of pancreatic β-cells and represents a highly specific target for imaging functional β-cell mass. Exendin-4, a GLP-1R agonist, can be radiolabeled with gallium-68 to create PET tracers such as ⁶⁸Ga-NOTA-exendin-4 and ⁶⁸Ga-HBED-CC-exendin-4. These tracers have shown excellent performance in visualizing insulin-secreting tissue.

In a prospective cohort study, Luo et al. demonstrated that ⁶⁸Ga-NOTA-exendin-4 PET/CT achieved a sensitivity of 97.7% and specificity of 100% in detecting localized insulinomas.^[35] Beyond neuroendocrine tumors, GLP-1R imaging has been applied in T2DM. Eriksson et al. reported reduced pancreatic tracer uptake in T2DM patients, which correlated with C-peptide levels and indicated reduced β-cell function.^[36]

Tracer optimization remains a subject of active investigation. A comparative study by Li et al. showed that both ⁶⁸Ga-NOTA and ⁶⁸Ga-HBED-CC labeled exendin-4 had similar sensitivity, although the HBED-CC variant yielded slightly better tumor-to-background ratios.^[37]

However, the widespread application of GLP-1R imaging is limited by high renal uptake and low central nervous system penetration. These factors can hinder visualisation of the pancreas and necessitate mitigation strategies, such as the use of amino acid co-infusion.^[38]

Despite these limitations, GLP-1R-targeted imaging holds strong potential for evaluating residual β-cell mass in T1D, assessing the success of islet transplantation, and monitoring therapies designed to preserve or regenerate β-cells. While its use in central nervous system imaging remains limited due to poor blood–brain barrier penetration-as demonstrated in the study by Deden et al. using ⁶⁸Ga-NODAGA-exendin-4 PET in obese subjects, its high receptor specificity makes it well-suited for imaging pancreatic tissue.^[39] With further improvements in tracer pharmacokinetics and quantification protocols, ⁶⁸Ga-Exendin-4 PET may become a key tool in diabetes management and research.

Imaging inflammation and insulin resistance

Imaging of adipose tissue and metabolic activity

Imaging adipose tissue using ¹⁸F-FDG PET has provided novel insights into the metabolic roles of fat depots and the pathophysiology of IR. Cohade demonstrated that BAT activation can lead to markedly increased FDG uptake, with its distribution influenced by insulin effects and metabolic status.^[45] This is critical as BAT, unlike white adipose tissue, facilitates increased glucose utilisation through adaptive thermogenesis.

Lundström et al. employed PET/MRI to assess BAT perfusion, lipid content, and glucose metabolic rate in humans, highlighting BAT’s complex metabolic response to thermogenic stimuli and its close association with insulin sensitivity.^[46] Meanwhile, Maliszewska et al. showed that dietary protein sources can enhance FDG uptake in BAT, pointing to nutritional modulation as a tool for adipose metabolism intervention.^[47]

Boss et al. observed that under controlled hypoglycemia, BAT metabolism increases, indicated by elevated FDG uptake, potentially reflecting a metabolic compensation mechanism to maintain glucose homeostasis.^[48] In addition, Ng et al. demonstrated that regional adipose tissue depots such as visceral and subcutaneous fat play distinct roles in systemic glucose metabolism, which can be visualized using PET imaging.^[49]

However, Reijrink et al. cautioned that increased FDG uptake in adipose tissue among patients with type 2 diabetes does not necessarily indicate inflammation, but may reflect metabolic activity, necessitating a broader metabolic context in image interpretation.^[50]

Collectively, these studies support the use of PET, particularly ¹⁸F-FDG, to evaluate metabolically active fat depots. Future development of more specific PET tracers may allow for better differentiation between harmful visceral fat and metabolically protective fat like BAT, facilitating targeted interventions in IR and obesity-related diabetes.

Hybrid imaging and novel radiopharmaceuticals for glucagon-like peptide 1 receptor and pancreatic β-cell visualisation

Recent advances have highlighted the promise of GLP-1R imaging as a key molecular target in type 2 diabetes and metabolic diseases. Brand et al. developed a novel bimodal PET/fluorescence imaging agent targeting GLP-1R, enabling simultaneous in vivo molecular and optical visualisation. This hybrid tracer is particularly valuable for translational research from preclinical models to human application.^[51]

Wei et al. reviewed current strategies for β-cell molecular imaging, covering GLP-1R, VMAT2, and somatostatin receptor-based tracers. The review underscores the potential of GLP-1R-targeted imaging in detecting functional β-cell mass, although interindividual variability and limitations in binding affinity remain barriers to clinical adoption.^[30]

Fernandes et al. compared PET imaging and whole-body autoradiography in assessing the biodistribution of a long-acting GLP-1 agonist. Their results confirmed the high uptake in the pancreas and kidneys, and consistent localisation in endocrine and gastrointestinal organs.^[52]

These findings validate PET as a powerful modality for evaluating incretin-based drug pharmacokinetics. Overall, the integration of hybrid imaging platforms and innovative multimodal tracers such as PET/fluorescence agents targeting GLP-1R represents a transformative step toward precise diagnostics and therapy monitoring in diabetes and related metabolic disorders.