Ankle reflex and neurological symptom score: a primary level screening method for diabetic peripheral neuropathy

Abstract

Herein, we aimed to develop an easily available and efficient screening method for diabetic peripheral neuropathy (DPN) suitable for primary care settings, emphasizing simplicity, speed, and accuracy. Nerve conduction studies were conducted on 214 patients with diabetes, encompassing the outcomes of five distinct assessments: diabetic neuropathy symptom (DNS), vibration perception threshold (VPT), and nerve screening. The diagnostic accuracy of the VPT and nerve screening was evaluated by comparing them with that of the nerve conduction study. To assess diagnostic efficacy, various combinations were examined, including DNS combined with VPT, pain, temperature, touch, and ankle reflex. The diagnostic performance of DNS was superior to that of the five neurological screening items and VPT, with sensitivity, specificity, and accuracy of 0.68, 0.81, and 0.73, respectively. Among the two combined methods, “DNS + ankle reflex” was identified as having the highest diagnostic value, with an area under the curve, a sensitivity, a specificity, and an accuracy of 0.81, 0.89, 0.70, and 0.80, respectively. Furthermore, a combination of “DNS + ankle reflex + touch + pain + VPT” achieved the best performance among the five combinations, with an area under the curve, sensitivity, specificity, and accuracy of 0.85, 0.93, 0.68, and 0.81, respectively. The combination of DNS, ankle reflex, touch, pain, and VPT methods showed the highest diagnostic value for DPN. However, considering factors including accuracy, time, and economic cost, we recommend using a simpler combination of DNS and ankle reflex for large-scale screening of patients with DPN.

DIABETIC PERIPHERAL NEUROPATHY (DPN) is a prevalent complication of both type 1 and type 2 diabetes, affecting 50%–70% of individuals with diabetes [1, 2]. DPN significantly influences quality of life, worsening individual, societal, financial, and healthcare implications, with an estimated global incidence of 425 million in 2017 [3]. It emerges as a consequence of prolonged hyperglycemia and pathophysiological changes in patients, resulting in neurological impairments [4]. Early symptoms encompass symmetrical pain, numbness, and diminished sensation in the limbs, with the potential for progression to sensory impairment and functional loss. In advanced stages, DPN can culminate in dystrophic muscular atrophy, foot ulcers, gangrene, amputation, and mortality [5]. Therefore, early screening for DPN holds paramount importance.

Despite the presence of numerous screening methods for DPN, there is a lack of unified, convenient, and accurate screening and evaluation approaches. The nerve conduction study (NCS) serves as the gold standard for DPN diagnosis, as it precisely assesses the peripheral nerves’ ability to conduct electrical signals and evaluates their functional status [6]. However, this test has its limitations, including its time-consuming nature, high cost, cumbersome steps, and the need for strong patient cooperation [7]. As a result, its widespread use in primary care and outpatient screening is constrained [8].

Clinical practice commends several DPN screening tools, such as temperature perception, touch, and ankle reflex examinations [9]. However, their sensitivity is limited, which can result in missed diagnoses and subjectivity. In recent years, noninvasive techniques, such as vibration sensory threshold (VPT) examination, have been used for DPN screening. VPT offers a straightforward and quantifiable method for assessing neurosensory dysfunction [10]. Additionally, scale tools, such as the simple diabetic neuropathy symptom (DNS) score, are commonly used for DPN screening. DNS has demonstrated high predictive value and simplicity in screening for diabetic polyneuropathy [11]. However, due to its subjective nature, it is often recommended as a complement to or in combination with other diagnostic methods for diabetic neuropathy [12].

Given the research background, it is evident that individual examination methods have limitations, increasing the risk of missed diagnoses. Combining multiple screening methods has the potential to improve the detection rate of DPN. In this study, we introduce a novel approach that combines DNS with touch, temperature perception, ankle reflex, and VPT. Our goal is to develop a simple, reproducible, and resource-efficient approach with a low rate of missed diagnoses that addresses the primary screening requirements for DPN.

Materials and Methods

Population and measurement methods

In this case-control study, we included inpatients with diabetes admitted between July 2022 and February 2023 to the Department of Endocrinology at the First Affiliated Hospital of the University of Science and Technology, China. All patients met the World Health Organization Diagnostic Criteria for all types of diabetes [13]. The diagnostic criteria for DPN were as follows: (1) meeting the diagnostic criteria for type 2 diabetes mellitus; (2) demonstrating NCS abnormalities in both nerves on neuromyography, with neuropathy attributable to causes other than the neurotoxic effects of drugs, particularly chemotherapy drugs, and nerve damage resulting from metabolic toxins due to renal insufficiency.

Following the screening process, 214 patients met the inclusion criteria. The participants were enrolled consecutively; there were no missing data. Based on the results of the NCS, which is considered the gold standard, patients were classified into two groups: the DPN group (58.9%) and the non-DPN group (41.1%). This study adhered to the ethical standards outlined in the Declaration of Helsinki and received approval from the Biomedical Ethics Committee of the Hefei Institute of Materials Science, Chinese Academy of Sciences (No. SWYX-Y-2022-22). All participants provided informed consent and had the option to withdraw from the study at any time.

Data collection

By analyzing electronic medical records and relevant examination results, we collected information on patients’ underlying conditions, including age, disease duration, and the presence of complications such as diabetic retinopathy and diabetic nephropathy. Furthermore, we gathered data on various laboratory biochemical indicators, including blood glucose levels, renal function, glycated hemoglobin (HbA1c), and lipid profiles, including serum total cholesterol, triglycerides, creatinine, and high- and low-density lipoprotein cholesterol levels. Furthermore, we obtained information from patients regarding their symptoms, including lower extremity pain, tingling sensations, numbness, and gait instability.

Inspection methods

NCT

An electromyography specialist tested the patients’ nerve conduction function using an electromyometric-evoked potentiometer (KeyPoint6.CH amplifier). The tests were conducted in a quiet indoor environment at a controlled room temperature of 22°C–28°C, with skin temperature maintained above 32°C. Surface electrodes were used for stimulation and recording in NCS, which included motor nerve conduction studies and sensory nerve conduction studies. The motor conduction of the median, common ulnar, common peroneal, and tibial nerves, and the sensory conduction of the median, ulnar, superficial peroneal, and gastrocnemius nerves of the right limb of each subject were measured separately. The analysis indicators included the latency period, amplitude, and conduction velocity (CV) for each nerve. NCS abnormalities were defined as prolonged distal latency, decreased amplitude, or a slowing of the NCV by more than two standard deviations. Following the diagnostic criteria of Dyck et al. [14], NCS abnormalities in both nerves were considered indicative of DPN.

Neuropathy symptom score

DNS comprises the following four symptoms: lower extremity pain, pins and needles, numbness, and gait instability [15]. Each symptom was assigned a score of one, with a maximum total score of four. A score of ≥1 was considered abnormal, suggesting the presence of DNS.

Vibration examination

Sensiometer A, a quantitative sensory detector, was used to assess the vibration sensation on the plantar surface of the foot. The examination involved placing a 450-g vibration probe vertically on the patient’s index or middle finger, gradually increasing the voltage until the patient perceived slight vibrations. Subsequently, the probe was placed on six different areas of the patient’s foot, including both hallux toes; the palmar surface of the 1st, 3rd, and 5th metatarsal heads; the center of the foot; and the heel. Starting from zero, the vibration intensity slowly increased at each location. When the patients perceived vibration, they were instructed to immediately signal it, and the corresponding value was recorded as the VPT in volts (V). To ensure patient attentiveness and comprehension, a zero-stimulation test was randomly conducted. Three consecutive measurements were taken, and the average value was used for the analysis. The diagnostic criteria for VPT were as follows [16]: for individuals aged ≤50 years, an amplitude of >10 V was considered abnormal, and for those aged >50 years, an amplitude of >15 V was considered abnormal.

Temperature perception examination

Temperature perception was assessed using a shallow temperature sensory checker. The instrument had one metal end (providing a cool sensation) and one polyurethane end (providing a warm sensation). Both ends were applied to the skin of the patient’s foot and dorsal feet for testing. The measurement was repeated three times, and the cool temperature perception could not be distinguished as abnormal.

Haptic examination

The haptic sense was evaluated using a haptic examination pen with a 10-g nylon wire end. The procedure was as follows: (1) Placing the nylon wire perpendicular to the skin at various test points and applying pressure to bend the wire by approximately 1 cm. (2) A pause of 2–3 s was applied before measuring the next point. (3) Care was taken to avoid calluses and foot injuries during the measurements. (4) 20 points, including the big, middle, and little toes; 1st, 3rd, and 5th metatarsal heads; plantar center; side of the foot; dorsum; and heel, were measured. If the patient felt an abnormal sensation in 8 or fewer of the 20 points, it was considered indicative of impaired haptic sense. This assessment helped determine the patient’s ability to accurately perceive tactile stimuli.

Pain examination

Pain perception was evaluated with a handheld steel needle roller using the following procedure: (1) The back of the patient’s hand was tested with a steel needle roller to acquaint the patient with the examination purpose. (2) Care was taken to avoid areas with ulcers, calluses, scars, and necrotic tissues during the procedure. (3) Patients were not allowed to observe the examination process and were asked to report any pain or discomfort experienced. If the patient did not feel any sensation or if the sensation was reduced, it was considered abnormal. This assessment helped evaluate the patient’s ability to accurately perceive and respond to painful stimuli.

Ankle reflex examination

Ankle reflex was assessed with an ankle reflex percussion hammer using the following procedure: The patient was made to lie down with the hip and knee joints slightly flexed; the lower limbs underwent external rotation and abduction; by gently supporting the patient’s soles with the left hand so that the foot was hyperextended, the right percussion hammer was used to strike the Achilles tendon. The normal response was foot flexion toward the plantar surface; if there was no response, it was considered abnormal.

Statistical methods

Statistical analyses were conducted using IBM SPSS (version 25.0; IBM SPSS Inc., Chicago, IL, USA). The Kolmogorov-Smirnov test was used to assess the data distribution. Data following a normal distribution were compared between groups using t-tests, and the results were expressed as mean ± standard deviation (SD). Qualitative variables were analyzed using the chi-square or Fisher’s exact tests. An independent sample t-test was used to compare numerical data between the two groups, with the data presented as percentages. Pearson’s correlation analysis was performed to investigate the relationships among VPT, DNS, and NCS. The sensitivity, specificity, Jordon index, and kappa value (k-value) were calculated using a four-grid table as indicators of evaluation consistency. The receiver operating characteristic curves were generated, and the diagnostic value of different combinations for DPN was compared using the area under the curve (AUC). P < 0.05 was considered statistically significant.

Results

Clinical and laboratory characteristics

A total of 214 patients participated in the study, with 123 (58.9%) diagnosed with DPN. Of the included patients, 59% were male and 41% were female (age range: 20–83 years, average age: 58.85 ± 13.66 years). The average duration of the disease was 10.76 ± 8.66 years. The DPN group exhibited a significantly longer disease duration, higher creatinine levels, and a higher prevalence of retinopathy, nephropathy, smoking history, and ulcer history than the non-DPN group (p < 0.05). However, no significant differences were observed in terms of age; body mass index; serum total cholesterol, triglyceride, creatinine, and high- and low-density lipoprotein cholesterol levels glycated hemoglobin levels; or the proportion of cardiovascular disease between the DPN and non-DPN groups (p > 0.05) (Table 1).

Table 1

Comparison of clinical and laboratory features between patients with and without DPN

	Non-DPN	DPN	p-value
Age (years)	56.56 ± 15.76	59.93 ± 9.90	0.07
Duration of diabetes (years)	7.19 ± 7.16	13.39 ± 8.69	<0.001
BMI	23.92 ± 3.18	23.20 ± 3.18	0.11
Retinopathy	27.4%	72.6%	<0.001
Diabetic nephropathy	20.6%	79.4%	<0.001
Cardiovascular disease	34.6%	65.4%	0.05
History of ulceration	11.8%	88.2%	<0.001
History of smoking	30.9%	69.1%	0.02
Glycated hemoglobin (%)	8.57 ± 2.80	9.22 ± 2.24	0.51
High-density lipoprotein (mmol/L)	1.11 ± 0.24	1.06 ± 0.33	0.69
Low-density lipoprotein (mmol/L)	2.73 ± 0.90	2.63 ± 0.73	0.72
Triglycerides (mmol/L)	1.57 ± 0.68	2.15 ± 2.02	0.35
Total cholesterol (mmol/L)	4.43 ± 1.07	4.41 ± 1.13	0.95
Serum creatinine (mmol/L)	55.71 ± 11.61	91.79 ± 45.79	<0.001

DPN, diabetic peripheral neuropathy; BMI, body mass index.

Correlation analysis between DNS, VPT, and nerve conduction NCS

The results of Pearson’s correlation analysis revealed significant negative correlations between DNS and ulnar nerve motor conduction velocity (MCV), peroneal MCV, tibial MCV, ulnar nerve sensory conduction velocity (SCV), and sural nerve SCV (p < 0.05). Similarly, VPT was significantly negatively correlated with the MCV and SCV of all examined nerves (p < 0.05) (data not shown).

Comparison of the diagnostic performance of common screening methods

Comparing various commonly used screening methods, DNS exhibited the highest diagnostic value, with a sensitivity of 0.68, a specificity of 0.81, and an accuracy of 0.73. Following DNS, the neurological screening of the five items and VPT showed relatively favorable results. Among the five neurological screening items, the ankle reflex demonstrated the best performance (sensitivity = 0.68, specificity = 0.75, accuracy = 0.71, and k-value = 0.42), indicating a moderate level of consistency in diagnosing DPN (Table 2).

Table 2

Diagnostic performance of common DPN screening methods

Common methods	AUC	Kappa value	Sensitivity	Specificity	Accuracy	Yueden Index
DNS	0.74	0.47	0.68	0.81	0.73	0.49
VPT	0.61	0.20	0.39	0.82	0.59	0.21
Pain	0.54	0.07	0.11	0.98	0.56	0.08
Temperature perception	0.56	0.11	0.20	0.92	0.58	0.13
Touch	0.58	0.14	0.12	0.97	0.58	0.16
Ankle reflexes	0.72	0.42	0.68	0.75	0.71	0.43
Nerve pentamajor total	0.73	0.48	0.76	0.71	0.74	0.47

DPN, diabetic peripheral neuropathy; AUC, receiver operating characteristic; DNS, diabetic neuropathy symptom; VPT, vibration perception threshold.

Comparison of the diagnostic performance of multiple combined methods

The diagnostic potential of DNS in DPN and its combination with other methods are presented in Table 3. The findings indicated that the combination of “DNS + ankle reflex” (referred to as 2A) exhibited the highest diagnostic value, with an AUC of 0.81, a sensitivity of 0.89, a specificity of 0.70, an accuracy of 0.80, and a k-value of 0.60, demonstrating consistency with the gold standard. Therefore, when combined with the three methods, the diagnosis of DPN was based on the combination of (2A) as the foundation, along with other methods. The results showed that the combination of (3A) “DNS + ankle reflex + touch” showed optimal diagnostic performance (AUC = 0.83, sensitivity = 0.91, specificity = 0.69, and accuracy = 0.82). Consequently, in the combination of the four methods, the diagnosis of DPN was based on the foundation of 3A combined with other methods. The results demonstrated that the combination of “DNS + ankle reflex + touch + VPT” (referred to as 4A) showed optimal diagnostic performance, with an AUC of 0.84, a sensitivity of 0.92, a specificity of 0.69, and an accuracy of 0.80. Based on the results of the abovementioned four combinations (4A), “DNS + ankle reflex + haptics + VPT” showed the best diagnostic performance; therefore, among the combinations of five methods, (4A) was selected as the basis for the diagnosis of DPN. The results showed that (5A) “DNS + ankle reflex + touch + VPT + pain” had the best diagnostic performance (AUC = 0.85, sensitivity = 0.93, specificity = 0.68, and accuracy = 0.80). Furthermore, the above six methods were combined, namely “DNS + ankle reflex + touch + pain + VPT + temperature perception” for the diagnosis of DPN. The results showed that the diagnostic performance was consistent with the (5A) combination (AUC = 0.85, sensitivity = 0.93, specificity = 0.68, and accuracy = 0.81) (Table 3).

Table 3

Diagnostic values of different combined methods screening DPN

	Name	Compound mode	AUC	Kappa value	Sensitivity	Specificity	Accuracy	Yueden Index
Two combined methods	2A	DNS + Ankle Reflexes	0.81	0.60	0.89	0.70	0.80	0.60
	2B	DNS + VPT	0.79	0.52	0.83	0.68	0.77	0.51
	2C	DNS + Touch	0.77	0.50	0.72	0.79	0.75	0.51
	2D	DNS + Temperature	0.77	0.51	0.76	0.76	0.76	0.51
	2E	DNS + Pain	0.76	0.50	0.71	0.80	0.75	0.51
Three combined methods	3A	DNS + Ankle reflexes + Touch	0.83	0.63	0.91	0.69	0.82	0.61
	3B	DNS + Ankle reflexes + VPT	0.82	0.62	0.90	0.70	0.78	0.60
	3C	DNS + Ankle reflexes + Pain	0.82	0.61	0.90	0.69	0.81	0.59
	3D	DNS + Ankle reflexes + Temperature	0.81	0.58	0.89	0.70	0.81	0.60
Four combined methods	4A	DNS + Ankle reflexes + Touch + VPT	0.84	0.64	0.92	0.69	0.80	0.61
	4B	DNS + Ankle reflexes + Touch + Temperature	0.83	0.58	0.92	0.68	0.81	0.61
	4C	DNS + Ankle reflexes + Touch + Pain	0.83	0.63	0.92	0.68	0.81	0.60
Five combined methods	5A	DNS + Ankle reflexes + Touch + VPT + Pain	0.85	0.64	0.93	0.68	0.80	0.61
	5B	DNS + Ankle reflexes + Touch + VPT + Temperature	0.84	0.58	0.92	0.69	0.81	0.61
Six combined methods	6A	DNS + Ankle reflexes + Touch + VPT + Pain + Temperature	0.85	0.64	0.93	0.68	0.81	0.61

DPN, diabetic peripheral neuropathy; VPT, vibration perception threshold.

Comparison of the operation time and practical characteristics of each combination

The results showed that the 2A combination had the shortest operation time (58–71 s). As the number of combination elements increased, the time required for the operation gradually increased (Table 4).

Table 4

Comparison of the practical values of different combination methods

Name	Portfolio projects	Sensitivity	Specificity	Time (s)	Peculiarity
2A	DNS + Ankle reflexes	0.89	0.70	58–71	Fast\reliable
3A	DNS + Ankle reflexes + Touch	0.91	0.69	90–102	High sensitivity\fast
4A	DNS + Ankle reflexes + Touch + VPT	0.92	0.69	382–405	High sensitivity\time-consuming
5A	DNS + Ankle reflexes + Touch + VPT + Pain	0.93	0.68	446–562	Comprehensive\time-consuming

DPN, diabetic peripheral neuropathy; VPT, vibration perception threshold.

Discussion

DPN is an incurable, progressive disease characterized by the demyelination and/or axonal degeneration of peripheral nerves. Once these debilitating lesions manifest, their progression becomes irreversible [17]. Therefore, early screening for DPN is of paramount importance. This study introduces a novel approach to DPN diagnosis by combining assessments based on DNS. Our primary aim was to develop an easy-to-implement screening approach for diabetic DPN, particularly in settings with limited diagnostic resources and varying levels of professional healthcare expertise. We identified five combinations (2A, 3A, 4A, 5A, and 6A) with the highest diagnostic values for DPN. The results indicated that DNS outperformed the neurological screening of five items and VPT. Among these combinations, 5A, which includes DNS, ankle reflex, touch, pain, and VPT methods, demonstrated the highest diagnostic value. However, for large-scale DPN screening, a simpler combination of DNS and ankle reflex (2A) was considered due to its accuracy, efficiency, and cost-effectiveness. Correlation analysis revealed a strong association between DNS and the conduction velocities of the ulnar, tibial, and gastrocnemius nerves.

DNS includes four primary symptoms: lower-extremity pain, pins and needles, numbness, and gait instability. Numbness in the hands and feet is an early sign of DPN and significantly correlates with clinical findings [18]. Sensory complaints related to numbness indicate the involvement of small fiber nerves, which are crucial for maintaining gait and static stability [19]. DNS is a concise and representative symptom score with high sensitivity and specificity that offers a rapid and reproducible assessment [20]. Furthermore, combining neurosymptom scores with noninvasive test results can enhance the diagnostic efficacy of DPN [21]. Therefore, we explored the combination of DNS with other methods to improve diagnostic accuracy.

Numerous studies have shown significant variations in DPN detection rates due to differences in diagnostic methods, criteria, and sample sizes. For instance, a study by Ponirakis et al. [22] detected DPN in 52.2% of 333 patients with diabetes using VPT. Another study involving 1,851 patients with diabetes using the Michigan Peripheral Neuropathy Screening Form found a 41.1% detection rate of DPN; moreover, 58.9% of patients were diagnosed with DPN, and the detection rate of DPN using the five neurological screening items was 55.3% [23]. Among single screening methods, the detection rates were 46.5%, 29.8%, 14.9%, 12.6%, 49.8%, and 49.8% for DNS, VPT, pain, temperature, touch, and ankle reflex, respectively. These findings highlight the variability in DPN detection rates when different screening tools are used. The detection rates of single screening methods being lower than those of confirmed DPN cases are most likely attributable to the involvement of multiple nerve fibers in peripheral nerve damage [1]. Depending solely on a single method for DPN detection carries the risk of overlooking diagnoses, underscoring the need to employ multiple screening approaches.

This study explored five combinations (2A, 2B, 2C, 2D, and 2E) using a combination of two diagnostic methods. Among these combinations, “DNS + ankle reflex” (2A) demonstrated the most robust diagnostic performance, exhibiting higher sensitivity than DNS alone, albeit with a slight decrease in specificity, aligning with the findings of Shehab et al. [24], further confirming the reliability of ankle reflex as a screening tool. The anatomy of the ankle joint may contribute to this, as the ankle reflex is directly innervated by the tibial nerve, which passes beneath the soleus muscle, serving the heel area and the saphenous nerve. Consequently, abnormalities in the ankle reflex directly reflect neuropathy progression [25]. The “DNS + ankle reflex” combination offers high sensitivity and specificity, making it a dependable primary screening method for DPN. When three methods were used in combination, “DNS + ankle reflex + touch” (3A) also demonstrated superior diagnostic performance, with heightened sensitivity and an AUC value of 0.82. The inclusion of touch slightly reduced specificity but increased sensitivity. A previous study indicated that patients testing positive with a 10-g nylon wire had a 7.7-fold higher risk of foot ulcers than those with normal feet [26]. Abnormalities in touch sensation reflect the loss of protective sensation, substantiating the efficacy of the 10-g nylon wire tactile test, to some extent, in diagnosing neuropathy [27]. Therefore, incorporating touch proves advantageous in the diagnosis of DPN.

In the case of the four diagnostic methods being combined, three combinations emerged (4A, 4B, and 4C). Among these, “DNS + ankle reflex + haptics + VPT” (4A) displayed the most remarkable diagnostic efficacy. The correlation analysis revealed that the VPT values were negatively correlated with the MCV and SCV of each nerve in both the upper and lower extremities, suggesting that VPT and NCS reflect the function of small-fiber nerves. VPT evaluates proprioceptive pathways through vibration stimulation and specifically mirrors the conduction pathways of nerve endings beneath the skin [28]. Diabetes-related peripheral neuropathy can impact nerves throughout the extremities. The diagnostic criteria for VPT were set as VPT >10 V for individuals aged ≤50 years and VPT >15 V for those aged >50 years, with sensitivity and specificity of 0.39 and 0.82, respectively, differing from those specified by Mythili et al. [29]. VPT was included as a diagnostic criterion for DPN in the 2010 American Diabetes Association guidelines [30]; however, a specific cutoff point was not provided, leading to inconsistencies in diagnostic performance. Discrepancies between our study findings and those of previous studies may be attributed to factors including sample size, diagnostic criteria, and appropriateness of the VPT cutoff points. Nevertheless, VPT remains a promising screening tool for DPN. Further research with a representative diabetic population is imperative to establish appropriate diagnostic cutoffs and improve diagnostic accuracy.

Combinations incorporating all five methods, 5A and 5B, exhibited the best diagnostic performance for DPN. While pain sensation plays a role in diagnosis, its significance is limited. Pain perception is associated with the activation of receptors in primary afferent fibers, including unmyelinated C fibers and myelinated Aδ fibers. The pain signals travel from the periphery to the dorsal horn of the spinal cord and then through the peripheral nervous system to the brain. The brain, as the central processing unit, integrates and processes information from the peripheral nervous system, coordinating various bodily functions before generating responses to effector organs. A decrease in pain perception partly indicates neuropathy. The 5A combination includes the patient’s symptomatic complaints, covering features of protective sensations like touch and pain. Needle-prick pain and ankle reflexes can reveal abnormalities in small fiber nerves, whereas VPT results help in assessing the severity of DPN. Needle-prick pain can detect abnormalities in small fiber nerves, whereas the ankle reflex directly reflects the status of the tibial motor nerve. Furthermore, the VPT results provide a quantitative assessment of DPN severity. This combination encompasses both sensory and motor fibers of larger myelinated nerves and unmyelinated superficial sensory fibers [31], making it highly valuable for the early diagnosis of DPN. When all six methods were combined, the maximum AUC value remained consistent with that of the 5A combination, indicating no significant change in diagnostic performance. The addition of temperature perception to the 5A combination did not significantly contribute to DPN screening. Previous studies have shown that temperature perception is more relevant than neuropathy in detecting diabetic foot ulcers. Moreover, ambient temperature fluctuations can easily interfere with temperature perception measurement accuracy [32], explaining the relatively low sensitivity of temperature sensation as a diagnostic tool for DPN in the present study.

One of the objectives of this study was to enhance diagnostic sensitivity, reduce missed rates and minimize testing time in the diagnosis of DPN. During the experiment, the practical characteristics and overall diagnostic performance of each combination were assessed, including the time required for each procedure. As the number of elements in the combination increased, sensitivity improved but specificity decreased, and there was a substantial increase in testing time. Although our method may not be ideal for definitive diagnostic purposes owing to its relatively low specificity, its primary goal is to serve as a valuable initial diagnostic tool with high sensitivity and low missed rates, ensuring the timely identification of individuals with underlying DPN for further comprehensive assessment. The high sensitivity of this method significantly reduces the risk of missing potential cases, allowing for timely intervention and management strategies. It is acknowledged that potential misdiagnoses may occur due to the reduced specificity of the method, and the limitations and potential implications of this approach for screening purposes are thoroughly discussed.

Nonetheless, our study has several limitations. The relatively small sample size suggests a need for further expansion of the sample size. Using a cohort study method would be valuable in assessing the screening effect of this combination approach with a larger, more diverse population, enabling more reliable conclusions about its effectiveness in DPN screening. Additionally, future research could explore the potential of integrating machine-learning methods to improve the accuracy of DPN diagnosis.

Conclusion

The combination of DNS, ankle reflex, touch, VPT, and pain demonstrated the most optimal diagnostic performance for DPN, offering a comprehensive assessment criterion with high sensitivity. Nonetheless, its drawback resides in the lengthy testing procedure, rendering it less suitable for screening large populations. Therefore, we recommend considering the adoption of the “DNS + ankle reflex” combination method as a potential primary screening approach for DPN, particularly in resource-constrained settings with varying levels of healthcare expertise. This proposed approach is specifically tailored to address the critical challenge of the early identification of DPN cases on a broader scale, considering the prevalent issues in resource-limited healthcare settings. This method ensures precision, reduces testing time and costs, and alleviates the workload of medical personnel.

Authors’ Contributions

Conceptualization

XM, ML, and ZM.

Data curation

HX, TS, and LL.

Investigation

SL and YS.

Software

XM, LL, and ZM.

Supervision

XM, YS, and ZM.

Manuscript drafting and revision

XM, ML, SL, YS, and ZM.

The manuscript has been thoroughly read and approved by all authors.

Conflict of Interest

The authors declare no conflicts of interest.

Ethics Approval

This study was approved by the Biomedical Ethics Committee of the Hefei Institute of Materials Science, Chinese Academy of Sciences (No. SWYX-Y-2022-22).

Funding

This research was supported by the National Key R&D Program of China (Grant No.2022YFC2010200) and the Anhui Province Major Project, Cloud-Edge Collaborative Self-service Smart Health Hut R&D and Demonstration Application (No. 202103A07020004).

References

• 1   Burgess  J,  Frank  B,  Marshall  A,  Khalil  RS,  Ponirakis  G, et al. (2021) Early detection of diabetic peripheral neuropathy: a focus on small nerve fibres. Diagnostics (Basel) 11: 165.
• 2   Morales Vidal  S,  Morgan  C,  McCoyd  M,  Hornik  A (2012) Diabetic peripheral neuropathy and the management of diabetic peripheral neuropathic pain. Postgrad Med 124: 145–153.
• 3   Cho  NH,  Shaw  J,  Karuranga  S,  Huang  Y,  da Rocha Fernandes  J, et al. (2018) IDF Diabetes Atlas: Global estimates of diabetes prevalence for 2017 and projections for 2045. Diabetes Res Clin Pract 138: 271–281.
• 4  The Diabetes Control and Complications Trial Research Group (1995) The effect of intensive diabetes therapy on the development and progression of neuropathy. Ann Intern Med 122: 561–568.
• 5   Biddinger  KR,  Amend  KJ (2004) The role of surgical decompression for diabetic neuropathy. Foot Ankle Clin 9: 239–254.
• 6   Vinik  AI,  Nevoret  ML,  Casellini  C (2015) The new age of sudomotor function testing: a sensitive and specific biomarker for diagnosis, estimation of severity, monitoring progression, and regression in response to intervention. Front Endocrinol (Lausanne) 6: 94.
• 7   Perkins  B,  Bril  V (2014) Electrophysiologic testing in diabetic neuropathy. Handb Clin Neurol 126: 235–248.
• 8   Hohman  T,  Cotter  M,  Cameron  N (2000) ATP-sensitive K+ channel effects on nerve function, Na+, K+ ATPase, and glutathione in diabetic rats. Eur J Pharmacol 397: 335–341.
• 9   American Diabetes  Association (2018) Microvascular complications and foot care: standards of medical care in diabetes—2018. Diabetes Care 41: S105–S118.
• 10   Rådman  L,  Gunnarsson  LG,  Nilsagård  Y,  Nilsson  T (2016) Neurosensory findings among electricians with self-reported remaining symptoms after an electrical injury: a case series. Burns 42: 1712–1720.
• 11   Meijer  J,  Smit  A,  Sonderen  E,  Groothoff  J,  Eisma  W, et al. (2002) Symptom scoring systems to diagnose distal polyneuropathy in diabetes: the diabetic neuropathy symptom score. Diabet Med 19: 962–965.
• 12   Laupacis  A,  Wells  G,  Richardson  WS,  Tugwell  P (1994) Users’ guides to the medical literature: V. How to use an article about prognosis. JAMA 272: 234–237.
• 13   Alberti  KG,  Zimmet  PZ (1998) Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 15: 539–553.
• 14   Dyck  PJ,  Karnes  JL,  Daube  J,  O’brien  P,  Service  FJ (1985) Clinical and neuropathological criteria for the diagnosis and staging of diabetic polyneuropathy. Brain 108: 861–880.
• 15   Meijer  JWG,  Bosma  E,  Lefrandt  JD,  Links  TP,  Smit  AJ, et al. (2003) Clinical diagnosis of diabetic polyneuropathy with the diabetic neuropathy symptom and diabetic neuropathy examination scores. Diabetes Care 26: 697–701.
• 16   Young  MJ,  Breddy  JL,  Veves  A,  Boulton  AJ (1994) The prediction of diabetic neuropathic foot ulceration using vibration perception thresholds: a prospective study. Diabetes Care 17: 557–560.
• 17   Chinese Diabetes  Society (2018) Guidelines for the prevention and control of type 2 diabetes in China (2017 Edition). Chin J Pract Inter Med 38: 292–344.
• 18   Sloan  G,  Selvarajah  D,  Tesfaye  S (2021) Pathogenesis, diagnosis and clinical management of diabetic sensorimotor peripheral neuropathy. Nat Rev Endocrinol 17: 400–420.
• 19   Fang  F,  Wang  YF,  Gu  MY,  Chen  H,  Wang  DM, et al. (2013) Pedobarography—a novel screening tool for diabetic peripheral neuropathy? Eur Rev Med Pharmacol Sci 17: 3206–3212.
• 20   Mayfield  JA,  Reiber  GE,  Sanders  LJ,  Janisse  D,  Pogach  LM (2003) Preventive foot care in people with diabetes. Diabetes Care 26: S78–S79.
• 21   Zhang  Q,  Liang  XC (2021) Advances in noninvasive methods for screening and evaluating diabetic peripheral neuropathy. Acta Acad Med Sin 43: 124–129 (In Chinese).
• 22   Ponirakis  G,  Elhadd  T,  Al Ozairi  E,  Brema  I,  Chinnaiyan  S, et al. (2022) Prevalence and risk factors for diabetic peripheral neuropathy, neuropathic pain and foot ulceration in the Arabian Gulf region. J Diabetes Investig 13: 1551–1559.
• 23   Hazari  A,  Mishra  V,  Agouris  I (2023) Clinical evaluation of painful diabetes peripheral neuropathy in Type 2 Diabetes patients: Lessons from expatriates in the United Arab Emirates. Diabetes Metab Syndr 17: 102832.
• 24   Shehab  D,  Al-Jarallah  K,  Abraham  M,  Mojiminiyi  O,  Al-Mohamedy  H, et al. (2012) Back to basics: ankle reflex in the evaluation of peripheral neuropathy in type 2 diabetes mellitus. QJM 105: 315–320.
• 25   Gylfadottir  SS,  Weeracharoenkul  D,  Andersen  ST,  Niruthisard  S,  Suwanwalaikorn  S, et al. (2019) Painful and non-painful diabetic polyneuropathy: clinical characteristics and diagnostic issues. J Diabetes Investig 10: 1148–1157.
• 26   Armstrong  DG,  Lavery  LA,  Vela  SA,  Quebedeaux  TL,  Fleischli  JG (1998) Choosing a practical screening instrument to identify patients at risk for diabetic foot ulceration. Arch Intern Med 158: 289–292.
• 27  Perkins BA, Bril V (2023) Clinical diagnosis of diabetic peripheral neuropathy. In: Tesfaye S, Gibbons CH, Malik RA, Veves A (eds) Diabetic neuropathy: advances in pathophysiology and clinical management (3rd). Humana Press, Cham, Switzerland: 67–92.
• 28   Deshpande  N,  Metter  EJ,  Ling  S,  Conwit  R,  Ferrucci  L (2008) Physiological correlates of age-related decline in vibrotactile sensitivity. Neurobiol Aging 29: 765–773.
• 29   Mythili  A,  Kumar  KD,  Subrahmanyam  K,  Venkateswarlu  K,  Butchi  RG (2010) A comparative study of examination scores and quantitative sensory testing in diagnosis of diabetic polyneuropathy. Int J Diabetes Dev Ctries 30: 43–48.
• 30   Tesfaye  S,  Boulton  AJ,  Dyck  PJ,  Freeman  R,  Horowitz  M, et al. (2010) Diabetic neuropathies: update on definitions, diagnostic criteria, estimation of severity, and treatments. Diabetes Care 33: 2285–2293.
• 31   Vinik  AI,  Nevoret  ML,  Casellini  C,  Parson  H (2013) Diabetic neuropathy. Endocrinol Metab Clin North Am 42: 747–787.
• 32   Ibrahim  A (2017) IDF Clinical practice recommendation on the diabetic foot: a guide for healthcare professionals. Diabetes Res Clin Pract 127: 285–287.