MR T2 value of the tibial nerve can be used as a potential non-invasive and quantitative biomarker for the diagnosis of diabetic peripheral neuropathy

Enhancing diabetic foot ulcer prediction with machine learning: A focus on Localized examinations

Background

diabetices foot ulcer (DFU) are serious complications. It is crucial to detect and diagnose DFU early in order to provide timely treatment, improve patient quality of life, and avoid the social and economic consequences. Machine learning techniques can help identify risk factors associated with DFU development.

Objective

The aim of this study was to establish correlations between clinical and biochemical risk factors of DFU through local foot examinations based on the construction of predictive models using automated machine learning techniques.

Methods

The input dataset consisted of 566 diabetes cases and 50 DFU risk factors, including 9 local foot examinations. 340 patients with Class 0 labeling (low-risk DFU), 226 patients with Class 1 labeling (high-risk DFU). To divide the training group (consisting of 453 cases) and the validation group (consisting of 113 cases), as well as preprocess the data and develop a prediction model, a Monte Carlo cross-validation approach was employed. Furthermore, potential high-risk factors were analyzed using various algorithms, including Bayesian BYS, Multi-Gaussian Weighted Classifier (MGWC), Support Vector Machine (SVM), and Random Forest Classifier (RF). A three-layer machine learning training was constructed, and model performance was estimated using a Confusion Matrix. The top 30 ranking feature variables were ultimately determined. To reinforce the robustness and generalizability of the predictive model, an independent dataset comprising 248 cases was employed for external validation. This validation process evaluated the model's applicability and reliability across diverse populations and clinical settings. Importantly, the external dataset required no additional tuning or adjustment of parameters, enabling an unbiased assessment of the model's generalizability and its capacity to predict the risk of DFU.

Results

The ensemble learning method outperformed individual classifiers in various performance evaluation metrics. Based on the ROC analysis, the AUC of the AutoML model for assessing diabetic foot risk was 88.48 % (74.44-97.83 %). Other results were found to be as follows: 87.23 % (63.33 %-100.00 %) for sensitivity, 87.43 % (70.00 %-100.00 %) for specificity, 87.33 % (76.66 %-95.00 %) for accuracy, 87.69 % (75.00 %-100.00 %) for positive predictive value, and 87.70 % (71.79 %-100.00 %) for negative predictive value. In addition to traditional DFU risk factors such as cardiovascular disorders, peripheral artery disease, and neurological damage, we identified new risk factors such as lower limb varicose veins, history of cerebral infarction, blood urea nitrogen, GFR (Glomerular Filtration Rate), and type of diabetes that may be related to the development of DFU. In the external validation set of 158 samples, originating from an initial 248 with exclusions due to missing labels or features, the model still exhibited strong predictive accuracy. The AUC score of 0.762 indicated a strong discriminatory capability of the model. Furthermore, the Sensitivity and Specificity values provided insights into the model's ability to correctly identify both DFU cases and non-cases, respectively.

Conclusion

The predictive model, developed through AutoML and grounded in local foot examinations, has proven to be a robust and practical instrument for the screening, prediction, and diagnosis of DFU risk. This model not only aids medical practitioners in the identification of potential DFU cases but also plays a pivotal role in mitigating the progression towards adverse outcomes. And the recent successful external validation of our DFU risk prediction model marks a crucial advancement, indicating its readiness for clinical application. This validation reinforces the model's efficacy as an accessible and reliable tool for early DFU risk assessment, thereby facilitating prompt intervention strategies and enhancing overall patient outcomes.

Diabetic peripheral neuropathy: pathogenetic mechanisms and treatment

Diabetic peripheral neuropathy (DPN) refers to the development of peripheral nerve dysfunction in patients with diabetes when other causes are excluded. Diabetic distal symmetric polyneuropathy (DSPN) is the most representative form of DPN. As one of the most common complications of diabetes, its prevalence increases with the duration of diabetes. 10-15% of newly diagnosed T2DM patients have DSPN, and the prevalence can exceed 50% in patients with diabetes for more than 10 years. Bilateral limb pain, numbness, and paresthesia are the most common clinical manifestations in patients with DPN, and in severe cases, foot ulcers can occur, even leading to amputation. The etiology and pathogenesis of diabetic neuropathy are not yet completely clarified, but hyperglycemia, disorders of lipid metabolism, and abnormalities in insulin signaling pathways are currently considered to be the initiating factors for a range of pathophysiological changes in DPN. In the presence of abnormal metabolic factors, the normal structure and function of the entire peripheral nervous system are disrupted, including myelinated and unmyelinated nerve axons, perikaryon, neurovascular, and glial cells. In addition, abnormalities in the insulin signaling pathway will inhibit neural axon repair and promote apoptosis of damaged cells. Here, we will discuss recent advances in the study of DPN mechanisms, including oxidative stress pathways, mechanisms of microvascular damage, mechanisms of damage to insulin receptor signaling pathways, and other potential mechanisms associated with neuroinflammation, mitochondrial dysfunction, and cellular oxidative damage. Identifying the contributions from each pathway to neuropathy and the associations between them may help us to further explore more targeted screening and treatment interventions.

Peripheral Neuropathy in Diabetes: What Can MRI Do?

Diabetes peripheral neuropathy (DPN) is commonly asymptomatic in the early stage. However, once symptoms and obvious defects appear, recovery is not possible. Diagnosis of neuropathy is based on physical examinations, questionnaires, nerve conduction studies, skin biopsies, and so on. However, the diagnosis of DPN is still challenging, and early diagnosis and immediate intervention are very important for prevention of the development and progression of diabetic neuropathy. The advantages of MRI in the diagnosis of DPN are obvious: the peripheral nerve imaging is clear, the lesions can be found intuitively, and the quantitative evaluation of the lesions is the basis for the diagnosis, classification, and follow-up of DPN. With the development of magnetic resonance technology, more and more studies have been conducted on detection of DPN. This article reviews the research field of MRI in DPN.

Two-dimensional ultrasound and two-dimensional shear wave elastography on femoral and saphenous neuropathy in patients with type 2 diabetes mellitus

Objective

This study aims to examine the value of two-dimensional (2D) ultrasound and two-dimensional shear wave elastography (2D-SWE) in evaluating femoral nerve (FN) and saphenous nerve (SN) neuropathy in patients with type 2 diabetes mellitus (T2DM).

Materials and methods

According to the diabetic peripheral neuropathy (DPN) diagnostic criteria, 60 patients with T2DM were enrolled and divided into 35 T2DM patients without DPN (non-DPN group) and 25 T2DM patients with DPN (DPN group). The control group consisted of another 15 healthy volunteers. The width, thickness, cross-sectional area (CSA), and perimeter of the FN and SN in the lower extremities were measured with 2D ultrasound. The average shear wave velocity (SWV) of the FN and SN was measured by 2D-SWE. Parameters of the left and right lower extremities were compared in each group, and the above parameters were compared among groups. The correlations between ultrasonographic and laboratory parameters were evaluated, and the independent influencing factors of SWV of the FN and SN were analyzed.

Results

The width, thickness, CSA, perimeter, and SWV of FN and SN in the DPN group were greater than those in the non-DPN groups and control group (P < 0.05). The width, thickness, CSA, perimeter, and SWV of the FN and SN in the non-DPN group were greater than those in the control group (P < 0.05). The CSA of FN was positively correlated with FPG, HbA1c, and TG (r = 0.34–0.69, P < 0.01). The perimeter of FN was positively correlated with FPG, HbA1c, and TG (r = 0.37–0.68, P < 0.01). The perimeter of the FN was negatively correlated with IgF-1 (r = −0.31, P < 0.05). The CSA of the SN was positively correlated with FPG and TG (r = 0.26–0.42, P < 0.05). The perimeter of the SN was positively correlated with FPG and TG (r = 0.37–0.39, P < 0.01). The SWV of FN and SN were positively correlated with FPG and TG (r = 0.35–0.57, P < 0.01; r = 0.43–0.49, P < 0.01). FPG and TG were independent influencing factors of the SWV of the FN and SN (P < 0.05).

Conclusion

2D ultrasound and 2D-SWE could be used to non-invasively, objectively, and accurately evaluate the abnormal changes of the FN and SN in patients with T2DM. It has important clinical significance for the early diagnosis of DPN and the curative effect evaluation.

Applicative value of T2 mapping in evaluating lumbosacral nerve root injury induced by lumbosacral disc herniation

BACKGROUND

To alleviate the damage caused by nerve root entrapment mediated by lumbosacral disc herniation (LDH), an imaging method that allows quantitative evaluation of the lumbosacral nerve injury is necessary.

PURPOSE

To investigate the diagnostic value of magnetic resonance (MR) T2 mapping in nerve root injury caused by LDH.

MATERIAL AND METHODS

A total of 70 patients with unilateral sciatic nerve pain and 35 healthy volunteers were divided into three groups: LDH with nerve root entrapment; LDH without nerve root entrapment; and 35 healthy volunteers. All participants underwent 3.0-T MR with T1-weighted (T1W) imaging, T2-weighted (T2W) imaging, and T2-mapping images. T2 was measured and observed with the left and right nerve roots of the L4-S1 segments in healthy volunteers; the differences between the three groups were compared. T2 and the relaxation rate of nerve root injury were analyzed.

RESULTS

T2 showed significant differences among the three groups (F = 89.494; P = 0.000), receiver operating characteristic curve revealed that the T2 relaxation threshold was 79 ms, the area under curve (AUC) area was 0.86, sensitivity was 0.77, and specificity was 0.74; the T2 relaxation rate was 1.06, the AUC area was 0.88, sensitivity was 0.74, and specificity was 0.85.

CONCLUSION

T2 mapping could quantitatively evaluate the nerve root injury with lumbar disc degeneration. Hence, it can be used for the clinical evaluation of nerve root entrapment caused by LDH.

Multiparametric quantitative MRI for the evaluation of dysthyroid optic neuropathy

OBJECTIVE

To evaluate the ability of quantitative MRI parameters for predicting dysthyroid optic neuropathy (DON).

METHODS

We retrospectively collected and analyzed the clinical features and 3.0 T MRI data of 59 patients with Graves orbitopathy (GO), with (n = 26) and without DON (n = 33). We compared MRI quantitative parameters, including the modified muscle index (mMI), proptosis, volume of intra-orbital fat, mean apparent diffusion coefficient value, and T2 value of the optic nerve among patients with and without DON. A logistic regression analysis was performed to identify the risk factors associated with DON. Moreover, we performed a receiver operating characteristic curve analysis and decision curve analysis to evaluate the diagnostic performance of the identified parameters for DON.

RESULTS

We studied 118 orbits (43 and 75 with and without DON, respectively). The mMI and mean T2 value of the optic nerve were significantly greater in orbits with DON (p < 0.001). A greater mMI at 21 mm (odds ratio (OR), 1.039; 95% confidence interval (CI): 1.019, 1.058) and higher mean T2 value of the optic nerve (OR, 1.035; 95% CI: 1.017, 1.054) were associated with a higher risk of DON. A model combining the mMI at 21 mm and mean T2 values for the optic nerve effectively predicted DON in patients with GO, with a sensitivity and specificity of 95.3% and 76%, respectively.

CONCLUSION

A quantitative MRI parameter combining the mMI at 21 mm and mean T2 value of the optic nerve can be an effective imaging marker for identifying DON.

KEY POINTS

• Patients with GO and DON had greater mMI than those without DON. • Optic nerves in patients with DON demonstrated an increased T2 value. • The quantitative MRI parameter combining the mMI at 21 mm and mean T2 value of the optic nerve is the most effective method for diagnosing DON.

Magnetic Resonance Imaging as a Biomarker in Diabetic and HIV-Associated Peripheral Neuropathy: A Systematic Review-Based Narrative

Background: Peripheral neuropathy can be caused by diabetes mellitus and HIV infection, and often leaves patients with treatment-resistant neuropathic pain. To better treat this condition, we need greater understanding of the pathogenesis, as well as objective biomarkers to predict treatment response. Magnetic resonance imaging (MRI) has a firm place as a biomarker for diseases of the central nervous system (CNS), but until recently has had little role for disease of the peripheral nervous system. Objectives: To review the current state-of-the-art of peripheral nerve MRI in diabetic and HIV symmetrical polyneuropathy. We used systematic literature search methods to identify all studies currently published, using this as a basis for a narrative review to discuss major findings in the literature. We also assessed risk of bias, as well as technical aspects of MRI and statistical analysis. Methods: Protocol was pre-registered on NIHR PROSPERO database. MEDLINE, Web of Science and EMBASE databases were searched from 1946 to 15th August 2020 for all studies investigating either diabetic or HIV neuropathy and MRI, focusing exclusively on studies investigating symmetrical polyneuropathy. The NIH quality assessment tool for observational and cross-sectional cohort studies was used for risk of bias assessment. Results: The search resulted in 18 papers eligible for review, 18 for diabetic neuropathy and 0 for HIV neuropathy. Risk of bias assessment demonstrated that studies generally lacked explicit sample size justifications, and some may be underpowered. Whilst most studies made efforts to balance groups for confounding variables (age, gender, BMI, disease duration), there was lack of consistency between studies. Overall, the literature provides convincing evidence that DPN is associated with larger nerve cross sectional area, T2-weighted hyperintense and hypointense lesions, evidence of nerve oedema on Dixon imaging, decreased fractional anisotropy and increased apparent diffusion coefficient compared with controls. Analysis to date is largely restricted to the sciatic nerve or its branches. Conclusions: There is emerging evidence that various structural MR metrics may be useful as biomarkers in diabetic polyneuropathy, and areas for future direction are discussed. Expanding this technique to other forms of peripheral neuropathy, including HIV neuropathy, would be of value. Systematic Review Registration: (identifier: CRD 42020167322) https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=167322.

Selective blood-nerve barrier leakiness with claudin-1 and vessel-associated macrophage loss in diabetic polyneuropathy

Diabetic polyneuropathy (DPN) is the most common complication in diabetes and can be painful in up to 26% of all diabetic patients. Peripheral nerves are shielded by the blood-nerve barrier (BNB) consisting of the perineurium and endoneurial vessels. So far, there are conflicting results regarding the role and function of the BNB in the pathophysiology of DPN. In this study, we analyzed the spatiotemporal tight junction protein profile, barrier permeability, and vessel-associated macrophages in Wistar rats with streptozotocin-induced DPN. In these rats, mechanical hypersensitivity developed after 2 weeks and loss of motor function after 8 weeks, while the BNB and the blood-DRG barrier were leakier for small, but not for large molecules after 8 weeks only. The blood-spinal cord barrier remained sealed throughout the observation period. No gross changes in tight junction protein or cytokine expression were observed in all barriers to blood. However, expression of Cldn1 mRNA in perineurium was specifically downregulated in conjunction with weaker vessel-associated macrophage shielding of the BNB. Our results underline the role of specific tight junction proteins and BNB breakdown in DPN maintenance and differentiate DPN from traumatic nerve injury. Targeting claudins and sealing the BNB could stabilize pain and prevent further nerve damage. KEY MESSAGES: • In diabetic painful neuropathy in rats: • Blood nerve barrier and blood DRG barrier are leaky for micromolecules. • Perineurial Cldn1 sealing the blood nerve barrier is specifically downregulated. • Endoneurial vessel-associated macrophages are also decreased. • These changes occur after onset of hyperalgesia thereby maintaining rather than inducing pain.

Thyroid-Associated Ophthalmopathy: Preliminary Study Using T2 Mapping to Characterize Intraorbital Optic Nerve Changes Before Dysthyroid Optic Neuropathy

OBJECTIVE

To evaluate the performance of T2 mapping in detecting intraorbital optic nerve (ON) changes in patients with thyroid-associated ophthalmopathy (TAO) before the onset of dysthyroid optic neuropathy (DON).

METHODS

Thirty-five patients with TAO and without DON (21 active, 14 inactive) and 21 healthy controls (HCs) were enrolled. Magnetic resonance imaging-derived parameters of T2 relaxation time (T2RT) at the intraorbital ON, extraocular muscle (EOM), orbital fat, exophthalmos, summed thickness of EOMs, orbital fat thickness, and clinical variables were compared. Correlations between T2RT at the ON and other variables were assessed.

RESULTS

Patients with TAO showed significantly higher T2RTs at the intraorbital ON than HCs (P < .001). Patients with active TAO had significantly higher T2RTs than those with inactive TAO and HCs (P < .001). Differences between patients with inactive TAO and HCs were insignificant (P > .05/3). T2RT at the intraorbital ON was positively correlated with clinical activity score, modified NOSPECS score, T2RT at EOM, exophthalmos, and summed thickness of EOMs in the TAO group (P ≤ .003) and negatively correlated with visual acuity (P = .033) and visual field indices (P = .030) in patients with active TAO. A T2RT cutoff of 82.9 ms for the intraorbital ON distinguished active TAO and healthy eyes optimally (area under the curve, 0.800; sensitivity, 85.7%; specificity, 64.3%).

CONCLUSION

T2RT detects disturbance in the intraorbital ON in patients with TAO, especially active TAO, before DON develops. T2 mapping has a potential for noninvasive evaluation of ON changes in patients with TAO.

T2 mapping of the distal sciatic nerve in healthy subjects and patients suffering from lumbar disc herniation with nerve compression

OBJECTIVE

To measure T2 values for magnetic resonance neurography (MRN) of the healthy distal sciatic nerve and compare those to T2 changes in patients with nerve compression.

MATERIALS AND METHODS

Twenty-one healthy subjects and five patients with sciatica due to disc herniation underwent MRN using a T2-prepared turbo spin echo (TSE) sequence of the distal sciatic nerve bilaterally. Six and one of those healthy subjects further underwent a commonly used multi-echo spin-echo (MESE) sequence and magnetic resonance spectroscopy (MRS), respectively.

RESULTS

T2 values derived from the T2-prepared TSE sequence were 44.6 ± 3.0 ms (left) and 44.5 ± 2.6 ms (right) in healthy subjects and showed good inter-reader reliability. In patients, T2 values of 61.5 ± 6.2 ms (affected side) versus 43.3 ± 2.4 ms (unaffected side) were obtained. T2 values of MRS were in good agreement with measurements from the T2-prepared TSE, but not with those of the MESE sequence.

DISCUSSION

A T2-prepared TSE sequence enables precise determination of T2 values of the distal sciatic nerve in agreement with MRS. A MESE sequence tends to overestimate nerve T2 compared to T2 from MRS due to the influence of residual fat on T2 quantification. Our approach may enable to quantitatively assess direct nerve affection related to nerve compression.

Radiomics of peripheral nerves MRI in mild carpal and cubital tunnel syndrome

OBJECTIVE

To assess the discriminative power of radiomics of peripheral nerves at 1.5T MRI, using common entrapment neuropathies of the upper limb as a model system of focal nerve injury.

MATERIALS AND METHODS

Radiomics was retrospectively done on peripheral nerve fascicles on T1-weighted 1.5T MRI of 40 patients with diagnosis of mild carpal (n = 25) and cubital tunnel (n = 15) syndrome and of 200 controls. Z-score normalization and Mann-Whitney U test were used to compare features of normal and pathological peripheral nerves. Receiver operating characteristic analysis was performed.

RESULTS

A total of n = 104 radiomics features were computed for each patient and control. Significant differences between normal and pathological median and ulnar nerves were found in n = 23/104 features (p < 0.001). According to features classification, n = 5/23 features were shape-based, n = 7/23 were first-order features, n = 11/23 features were classified as gray level run length matrix. Nine of the selected features showed an AUC higher that 0.7: minimum AUC of 0.74 (95% CI 0.61-0.89) for sum variance and maximum AUC of 0.90 (95% CI 0.82-0.99) for zone entropy.

CONCLUSION

Features analysis demonstrated statistically significant differences between normal and pathological nerve. The results suggested that radiomics analysis could assess the median and ulnar nerve inner structure changes due to the loss of the fascicular pattern, intraneural edema, fibrosis or fascicular alterations in mild carpal tunnel and mild cubital tunnel syndromes even when the nerve cross-sectional area does not change.

T2 mapping of lumbosacral nerves in patients suffering from unilateral radicular pain due to degenerative disc disease

OBJECTIVE

Lumbosacral radicular syndrome (LRS) is a very common condition, often requiring diagnostic imaging with the aim of elucidating a structural cause when symptoms are longer lasting. However, findings on conventional anatomical MRI do not necessarily correlate with clinical symptoms, and it is primarily performed for the qualitative evaluation of surrounding compressive structures, such as herniated discs, instead of to evaluate the nerves directly. The present study investigated the performance of quantitative imaging by using magnetic resonance neurography (MRN) in patients with LRS.

METHODS

Eighteen patients (55.6% males, mean age 64.4 ± 10.2 years), with strict unilateral LRS matching at least one dermatome and suspected disc herniation, underwent high-resolution 3-T MRN using T2 mapping. On T2 maps, the presumably affected and contralateral unaffected nerves were identified; subsequent regions of interest (ROIs) were placed at preganglionic, ganglionic, and postganglionic sites; and T2 values were extracted. Patients then underwent an epidural steroid injection (ESI) with local anesthetic agents at the site of suspected nerve affection. T2 values of the affected nerves were compared against the contralateral nerves. Furthermore, receiver operating characteristics were calculated based on the measured T2 values and the responsiveness to ESI.

RESULTS

The mean T2 value was 77.3 ± 1.9 msec for affected nerves and 74.8 ± 1.4 msec for contralateral nerves (p < 0.0001). In relation to ESI performed at the site of suspected nerve affection, MRN with T2 mapping had a sensitivity/specificity of 76.9%/60.0% and a positive/negative predictive value of 83.3%/50.0%. Signal alterations in affected nerves according to qualitative visual inspection were present in only 22.2% of patients.

CONCLUSIONS

As one of the first of its kind, this study revealed elevated T2 values in patients suffering from LRS. T2 values of lumbosacral nerves might be used as more objective parameters to directly detect nerve affection in such patients.