ABSTRACT
Diabetic peripheral neuropathy (DPN) is the most common form of acquired neuropathy. In children with type 1 diabetes, the reported prevalence of DPN varies widely, ranging from 3% to 62%, mainly due to differences in screening methodologies and patient population characteristics. While intraepidermal nerve fiber density assessment via skin biopsy remains the gold standard for detecting small fiber neuropathy, nerve conduction studies are the established diagnostic tool for large fiber involvement. However, several novel and non-invasive diagnostic tools have emerged recently, offering improved screening options for early-stage and subclinical DPN. The frequent presence of asymptomatic neuropathy in pediatric populations, combined with its limited treatment options, underscores the importance of early identification of modifiable risk factors thus reducing the risk of developing clinically significant DPN. This review provides a comprehensive overview of the current evidence on the prevalence, risk factors, and modern diagnostic approaches for DPN in children with diabetes.
Introduction
Diabetic peripheral neuropathy (DPN) is a common complication of diabetes and the most common type of acquired neuropathy. Classically, DPN is defined as the presence of symptoms and/or signs of peripheral nerve dysfunction in patients with diabetes after the exclusion of other causes (1). Typically, this is a chronic complication of diabetes but acute cases also occur (2). Sensory, motor, or autonomic nerves may be affected, often coexisting (1). There is no universally accepted classification for DPN. The American Diabetes Association classified diabetic neuropathies into three main categories: 1) diffuse neuropathy (distal symmetric polyneuropathy and autonomic); 2) mononeuropathy; and 3) radiculopathy or polyradiculopathy (3) (Figure 1). There are two stages of DPN, subclinical and clinical. Subclinical DPN implies neurophysiological changes in nerve function without clinical symptoms. The manifestation of nervous dysfunction characterizes clinical DPN (4).
The most common type of DPN is distal symmetric polyneuropathy (3). It may involve large‐fiber nerves (related to touch, vibration, position perception, and muscle control) and small‐fiber nerves (affected thermal perception, pain, and autonomic function) (Table 1). Most patients, however, have both large‐ and small‐nerve fiber damage. Sensory symptoms include numbness, paresthesia, and neuropathic pain. Symptoms begin in the toes and fingers before progressing in a stocking and then a glove distribution as the disease progresses. In the pediatric population, most patients present with subclinical neuropathy. In the case of clinical manifestation, pain, and dysesthesias are the most common presentations (5).
The US Food and Drug Administration has not approved any casual treatments for DPN. Hence, prevention, reducing risk factors, and finding appropriate screening tests that enable diagnosis at an early stage of the disease are essential (3).
The review aims to summarize the data on the prevalence, diagnostic methods, and risk factors of DPN in children and adolescents, especially in type 1 diabetes (T1D).
Prevalence of Diabetic Neuropathy
The prevalence of DPN in T1D ranges widely from 3% to 62% (6, 7, 8). There is a variation between the results of the studies, which are limited and hard to compare due to differences in age ranges, clinical definitions and diagnostic tests for DPN. The data are summarized in Table 2. The Pittsburgh Epidemiology of Diabetes Complications Study revealed diabetic neuropathy in only 3% of patients under 18 years old, 18% for 18- to 29-year-old patients, and 58% for >30-year-old patients (6). The EURODIAB insulin-dependent diabetes mellitus complications study, one of the largest and most important epidemiological studies on T1D in children and adolescents in Europe, reported diabetic neuropathy in 19% of patients aged 15 to 29 years (9). The Danish Study Group reported subclinical neuropathy in 62% of patients aged 12-27 years, diagnosed by vibration detection threshold (VDT) (10). However, Nelson et al.’s (11) identified neuropathy in 57% of patients using one of the gold standard methods, a nerve conduction study (NCS). The SEARCH for diabetes in youth study (SEARCH) is a multicenter, population-based study designed to assess the incidence, prevalence, and complications of diabetes among children and adolescents in the United States. Data from SEARCH showed that 7% of adolescent patients had evidence of neuropathy, as diagnosed by the Michigan Neuropathy Screening Instrument (MNSI) (12).
Diagnostic Methods
International Society for Pediatric and Adolescent Diabetes recommends that screening for DPN in young people with T1D should start at puberty or from age 11 years if the child has had diabetes for 2-5 years and screening should be repeated annually (13).
Clinical findings suggestive of DPN include a loss of sensation to pinprick, temperature, vibration, and proprioception (5). Neurological examination should consist of reflex testing because loss of ankle reflexes can occur in the early stage of DPN. Weakness of small foot muscles and dorsiflexion is usually observed later (5).
Screening Tests
Touch Sensation Tests
Touch sensation tests assess the integrity of Merkel touch domes and Meissner corpuscles and their associated large-diameter fibers. The Semmes‐Weinstein monofilament test, Neuropen, and the Ipswich touch test can be used to evaluate touch sensation. Frey’s hair test may also allow thr examination of touch perception thresholds (14).
There are different types of monofilament, e.g., 0.5 g, 2 g, 10 g, 50 g, and 200 g. A 10 g monofilament test is the most common, in practice. The Neuropen combines an interchangeable 10 g monofilament and a calibrated sterile Neurotip for assessing pain sensation. The Ipswich touch test is a simple method where the tip of an index finger is used to elicit the sensation of light touch on the tips of the patient’s first, third, and fifth toes for 1-2 seconds. Compared with the 10-g monofilament, the Ipswich Touch Test has a sensitivity and specificity of 76% and 90%, respectively, and can be done at home without equipment (7). Frey’s hairs are based on a method similar to that of a monofilament; touch perception thresholds can be assessed by buckling the hair which corresponds to a predefined force determined by the filament’s mechanical properties (nominal force of 0.026-110 g) with a force ranging from 0.026 g to 110 g (15).
Thermal Perception Testing
Thermal perception testing determines the function of free-nerve endings and their associated unmyelinated and thinly myelinated fibers. It is more valuable to test cold and warm perceptions separately. Only thermal threshold testing is capable of assessing small fiber dysfunction (5, 14, 16).
One of the simple screening methods for testing thermal perception is tip-therm. It is an instrument with two flat sides of a special polymer and a metal alloy. Due to the different thermal conductivity of both materials, the faces of the instruments are perceived differently. Compared to the biothesiometer, tip-therm has high specificity (100%) and sensitivity (97.3%) in diagnosing diabetic neuropathy (16).
Vibratory Sensation Test
Vibration perception testing assesses function in the Meissner, and Pacinian corpuscles, and large-diameter fibers. Typical testing sites are the glabrous skin of the fingers and toes (5, 14, 17). For the vibration perception assessment, a 128-Hz standard tuning fork, vibra tip, Rydel-Seiffer fork, or biothesiometer are used (5).
A 128‐Hz standard tuning fork determines the presence or absence of vibration perception. There are two methods of examination: on-off and timing method. In the first method, a tuning fork is applied to the bony prominence bilaterally situated at the dorsum of the first toe. The patient reports the start and the cessation of the vibration sensation. In the timing method, a tuning fork is applied to the same area, and the patient reports the time at which vibration diminished beyond perception. The results compare to the examiner’s timing of vibration sensation (17).
There is also a pocket-size device, the VibraTip, for assessing vibration perception in DPN screening. It provides a stimulus of 128 Hz (5).
Quantitative methods, including the Rydel‐Seiffer tuning fork (RSTF) and the biothesiometer (5), have been developed to determine VDT. The RSTF allows the testing of various vibration intensities by the patient to assess vibratory discrimination. A triangle and a scale from 0 to 8 imprinted on the weights allow assessment of the vibration threshold (18). After the tuning fork is struck to initiate vibration, its base is applied to a bony prominence. The patient is asked to indicate when the vibration sensation disappears, at which point the examiner records the corresponding value on the triangular scale displayed on the prongs of the tuning fork. Scores range from 8, indicating normal vibration perception, through 5-7 (mild reduction) and 3-4 (moderate impairment), to 0-2, reflecting severe sensory loss or complete absence of vibration perception (19).
The biothesiometer determines VDT on a scale from 0 to 50. It measures VDT by adjusting the amplitude of an electrical vibrator. In contrast to the RSTF, a biothesiometer measures the VDT as the point at which the first sensation of the vibration appears (20). Assessment of VDT using a biothesiometer is quick and reliable. Compared with a tuning fork, the biothesiometer has been reported to be more accurate (21).
New Quantitative Sensory Test
A quantitative sensory test (QST) determines the sensory threshold, which is defined as the minimal energy detected for a particular modality. Using direct patient feedback, QST measures sensory thresholds, such as vibration, cold, warmth, heat, or cold pain. There are two schemes: the method of limits and the method of levels. In the method of limits, a patient indicates as soon as an increasingly strong stimulus is detected or when a decreasing stimulus is no longer felt. In the method of levels, stimuli of defined intensity levels are tested, and the patient is asked to indicate if the stimulus is or is not detected. In the method of limits, the result is dependent on the rate of change of the stimuli and is more variable than the method of levels and will also be affected by a patient’s reaction time (22).
Scoring System
Various clinical scoring systems are used to screen DPNs. They may involve symptom scoring, sign scoring, or both (Table 3) (23, 24, 25, 26, 27, 28, 29, 30). The scoring system is a suitable and easy-to-perform method for the early detection of DPN. The MNSI is a widely used screening tool, including in the pediatric population. However, it has not been formally validated in children. Research in the adult population has shown a range of sensitivity (35-79%) and specificity (65-94%) compared to NCS, depending on the cut-off value used for abnormality in MNSI (23).
Gold Standard
Skin Biopsy
A skin biopsy is a relatively minimally invasive technique and the specimen is taken from the distal calf. Assessing intraepidermal nerve fibers (IENF) counts, lengths, and densities is key when identifying small fiber neuropathy (SFN). IENFs are the last endings of C and A-delta fibers, participating in pain perception and detection of thermal stimuli (31). There are two evaluation methods for IENFs: immunohistochemical (bright-field) or indirect immunofluorescence with and without confocal analysis. Nerve fiber staining is performed with protein gene product 9.5 (PGP9.5), a member of the ubiquitin hydroxylase system, and PGP9.5 is a non-specific pan-neuronal marker (31, 32).
In clinical practice, IENF density (IENFD) is the gold standard for diagnosing conditions with small nerve fiber involvement. Diabetes is the most common cause of SFN, and a decrease in IENFD may be evident in the earliest stage of the disease. Moreover, IENFD can be used to monitor the progression of diabetic SFN, and a decrease in IENFD correlates with the severity of SFN (32).
Mechanoreceptors, innervated by A-beta fiber endings, can also be identified in skin biopsy samples. NCS is a gold standard for diagnosing large fiber neuropathy (LFN). However, skin biopsy allows assessment of early abnormalities of receptors and the distal part of myelinated fibers.
Autonomic involvement can also be assessed through skin biopsy. In DPN, a significant loss of sudomotor (innervating the sweat glands) and pilomotor (controlling errector pili muscle contraction leading to hair follicle erection) nerve fibers has been demonstrated (32).
NCS
NCS is the gold standard for the diagnosis of LFN (5). The Toronto consensus for a firm diagnosis of LFN requires at least one symptom and/or at least one sign of neuropathy and abnormality on NCS (33). During the examination, small pads are inserted into the skin, deliver mild electric shocks, and detect electric signals. Routine NCS includes evaluation of the motor function of the median, ulnar, peroneal, and tibial nerves and the sensory function of the median, ulnar, radial, and sural nerves. Recommended attributes are amplitude, distal latency, distance, conduction velocity, and F‐wave latency. Diagnostic criteria involve the abnormality of one or more attributes in two or more separate nerves to diagnose DPN (34). NCS may not be widely available for routine diagnostic evaluation of DPN. Furthermore, NCS is insensitive for the diagnosis of SFN (35).
Other Methods
Neuropad
A simple indicator test that can detect the early stage of peripheral neuropathy is Neuropad. The Neuropad is applied to the plantar aspect of the first metatarsal head and removed after 10 minutes. It assesses sweat production by the color change of a cobalt II compound from blue to pink. Color change indicates sudomotor dysfunction in DPN, which usually develops before sensory loss. The sensitivity of the test was reported to be 70-83% when validated against the Neuropathy Disability Score, NCS, and VDT (36).
Sudoscan
Sudoscan is a non-invasive, quantitative test to assess sudomotor function using reverse iontophoresis to measure electrochemical skin conductance (ESC) of sweat in the hands and feet. It evaluates the function of small C-fibers that innervate the sweat glands. Degeneration of these fibers results in reduced sweat gland function (37). Validation studies have confirmed good values for sensitivity (70-87.5%) when using foot ESC results to screen for DPN (38). Global collaborative analysis established reference values in healthy subjects for different ethnic groups, ages, and by gender (39).
Corneal Confocal Microscopy
Corneal confocal microscopy (CCM) is a relatively new, non-invasive diagnostic tool for SFN. Nerve fibers from the trigeminal richly innervate the cornea. Corneal nerve fiber density, corneal nerve branch density, corneal nerve fiber length, and inferior whorl length are the main parameters used. A meta‐analysis of over 3,000 participants revealed a reduction in the CCM parameter in patients with DPN compared with healthy controls and those without DPN. In addition, the authors also reported a lesser, but significant, decrease in CCM parameters in diabetic patients without DPN compared with controls, suggesting that CCM may identify subclinical DPN (40). Corneal nerve loss was also observed in patients with only impaired glucose tolerance (41) and in children with T1D (42).
Optical Coherence Tomography Angiography
Optical coherence tomography angiography (OCTA) is a non-invasive tool that enables quantitative assessment of retinal microcirculation. Research highlights the value of OCTA in monitoring neurodegeneration in patients with T1D. The potential of OCTA as an imaging biomarker for evaluating the progression of DPN has been suggested (43).
Risk Factors
Diabetic neuropathy develops as a result of various risk factors, which can be categorized into modifiable and non-modifiable. Identifying and assessing these factors is essential for preventing DPN.
Non-modifiable factors include disease duration, pubertal status, tall stature, ethnicity, and genetic predisposition. Modifiable factors include glycemic control, dyslipidemia, and hypertension.
Non-modifiable Factors
Duration of Diabetes
Long-standing diabetes is a confirmed risk factor for DPN. In SEARCH, in patients with DPN, diabetes duration was significantly longer, but HbA1C values were similar between groups, indicating that longer diabetes duration regardless of metabolic control is a crucial factor in DPN (12).
Tall Stature
In contrast to nephropathy and retinopathy (complications of diabetes, which more often occur in patients with short stature), tall stature was associated with a higher prevalence of neuropathy. Some hypotheses exist, such as patients with longer nerves (and thus a larger total axon surface area) who may be at greater risk for neurologic impairment and have a prolonged time requirement for complete regeneration. The association of height with peripheral neuropathy might be related to the increased hydrostatic pressure in tall patients during standing upright (44).
Puberty
Research has shown that puberty contributes to the development of late metabolic complications of diabetes, such as retinopathy and neuropathy (45). Peripheral nerve function is increasingly impaired during puberty. A cross-sectional study demonstrated an increasing subclinical motor nerve impairment detected during late puberty and after puberty. Reduced sensory nerve conduction velocity (NCV) and sensory nerve action potential amplitude were confirmed. The peripheral nerve fibers of pubertal and postpubertal patients may be the most vulnerable to nerve demyelination and axonal damage caused by poor metabolic control induced by pubertal hormonal changes and poor self-care during puberty (46).
Ethnicity
The prevalence of DPN varies according to ethnicity. Studies of multi-ethnic groups have found a greater prevalence of DPN occurring in Caucasian patients compared to other ethnic groups (47).
Genetic Markers
There are limited data on genetic predisposition to DPN among children with T1D. Single nucleotide polymorphism is a variation at a single position in the DNA sequence among individuals, which can be linked to a higher or lower risk of diseases. A Slovak study confirmed a strong association between the Cytochrome b-245 alpha chain (CYBA) polymorphism rs4673 and DPN in children and adolescents with T1D. CYBA encodes the p22phox protein, a subunit of the nicotinamide adenine dinucleotide phosphate (NADPH) oxidase complex, which plays a key role in the production of reactive oxygen species (48).
A study by Strokov et al. (49) reported that the polymorphic markers Ala(-9)Val in the Superoxide Dismutase 2 (SOD2) gene and Arg213Gly in the SOD3 gene were associated with a higher risk of DPN. SOD2 and SOD3 encode different isoforms of the enzyme SOD, which plays a key role in protecting cells from oxidative stress (49).
Vascular endothelial growth factor (VEGF) is a signaling protein that plays a critical role in angiogenesis. However, it has been reported that the distribution of a VEGF gene polymorphism in the promoter region (-7C/T) differed significantly between patients with T1D with and without neuropathy, and may be implicated in the pathogenesis of DPN (50).
Modifiable Factors
Hyperglycemia
Hyperglycemia is a well-known risk factor for DPN. A study in children and youth with T1D confirmed an elevated DPN risk with poor glycemic control. EURODIAB determined the association of DPN with the duration of diabetes and highlights a strong relation with glycemic control independent of diabetes duration (9).
Glycemic Variability
In addition to hyperglycemia, glycemic variability (GV) may be another independent risk factor for diabetic complications, but results are inconsistent. Some studies propose potential pathophysiological mechanisms through which elevated GV contributes to the development of DPN. These include the activation of protein kinase C-dependent NADPH oxidase, which subsequently induces oxidative stress. Furthermore, an oxidative stress and inflammatory response is caused by activating the nuclear factor kappa-light-chain-enhancer of activated B cells pathway (51, 52).
The study by Oberhauser et al. (53), conducted among children and adolescents with T1D, confirmed that high GV was an unexpectedly strong predictor of slowed NCV. However, there are limited studies on the long-term outcomes of GV during the pediatric period, so further prospective research is needed (54).
Obesity
Obesity is an emerging risk factor for neuropathy, independent of hyperglycemia. Obesity may contribute to the development of neuropathy through several mechanisms, including oxidative stress, insulin resistance, metabolic inflammation, and ischemia, all of which can lead to nerve degeneration (55). Fat distribution is likely more important than general obesity in developing neuropathy. Among adolescents with obesity, the association of central adiposity with insulin resistance and inflammation has been confirmed. This effect is observed regardless of body mass index (56). Visceral adipose tissue is a source of pro-inflammatory mediators, including cytokines such as tumor necrosis factor-alpha and interleukin-6, and is associated with elevated levels of the acute-phase reactant C-reactive protein. These cytokines activate microglia and perivascular macrophages, ultimately promoting demyelination and axonal degeneration. Moreover, an excess of free fatty acids, particularly in central obesity, may cause direct damage to neural structures (57). Studies in childhood cohorts reveal a correlation between obesity and increased risk for neuropathy (58).
Dyslipidemia
Dyslipidemia increases the frequency and severity of micro- and macro-vascular complications in T1D. Large clinical studies like EURODIAB and SEARCH have confirmed the relationship between lipid disturbances and peripheral neuropathy. In EURODIAB, higher levels of total and low-density lipoprotein cholesterol (LDL-c) and triglycerides were significantly associated with the cumulative incidence of neuropathy (9). SEARCH found increased triglycerides, LDL-c, and lower levels of high-density lipoprotein cholesterol (HDL-c) as a risk factor for DPN in youth with T1D. The authors suggested that the lower HDL-c could be one of the crucial factors in the pathogenesis of DPN. HDL-c inhibits inflammation process, oxidation, and thrombosis, as well as vasodilatation via endothelial release of nitric oxide. Increasing HDL-c levels through lifestyle modification (dietary modification, aerobic exercise, weight loss) may be one of the therapeutic approaches (12).
Hypertension
Studies confirmed that hypertension was associated with the prevalence of neuropathy in young people with T1D (59). Hypertension is associated with impaired nerve conduction in T1D (60). Glycemic control has to be supported with strict blood pressure control to prevent and delay the onset of DPN. According to EURODIAB the risk factors for DPN in youth with T1D are increased diastolic blood pressure (9). The angiotensin-converting enzyme inhibitor (ACEI) may improve peripheral neuropathy even in normotensive patients with diabetes (61). Hyperglycemia increases tissue angiotensin II, causing oxidative stress, endothelial damage, and vascular changes. These contribute to DPN and can be diminished by blocking the renin-angiotensin system, which may explain the beneficial effect of therapy with ACEIs (62).
Conclusion
DPN is one of the most common complications of diabetes, which can result in foot ulcers and potentially necessitate amputation. The prevalence of neuropathy in patients with T1D differs, based on the screening methods used and the characteristics of the study groups. Nevertheless, the studies reviewed indicate a relatively high occurrence of subclinical neuropathy, highlighting the need for early detection of risk factors to prevent this complication. Of note, therapeutic options are currently very limited, so early screening and modification of risk factors is very important in patients with T1D.
