Prevalence of neuropathy in patients with impaired glucose tolerance using various electrophysiological tests
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.149393
Source of Support: None, Conflict of Interest: None
Background: Neuropathy is often an associated feature woth long-standing type II diabetes mellitus. Neuropathy may occur even in subjects with impaired glucose tolerance. Objective: To study the prevalence of neuropathy using different electrophysiological techniques in subjects with impaired glucose tolerance (IGT) and no other identifiable cause of neuropathy. Materials and Methods: The study was conducted on 30 age-matched controls and 58 subjects with impaired oral glucose tolerance test (OGTT) attending diabetic awareness. Prediabetes was defined using World Health Organization (WHO) criteria. All subjects had normal glycosylated hemoglobin HbA (1c), vitamin B12 levels, and thyroid function. Neuropathy was evaluated by nerve conduction studies (NCS) performed on one upper and both lower limbs, dorsal sural nerve, medial and lateral planter nerve conductions using conventional techniques. Neuropathy was also evaluated by autononic function tests, and quantitative sensory testing (QST). The subjects were followed up for 4 years. Results: Out of 58 subjects, 19 (32.8%) had neuropathy. Nerve conduction studies showed evidence of neuropathy in 14 (24.13%) subjects, autonomic neuropathy was detected in 8 (13.8%), and QST was found to be abnormal in 16 (27.6%) subjects. Twenty subjects (34.5%) developed diabetes mellitus in the follow-up period. Conclusions: Neuropathy was detected in 32.8% subjects with IGT. Small fiber neuropathy was most common. Of all the three parameters studied, QST was found to be most sensitive technique for the detection of neuropathy. Assessment of medial plantar and dorsal sural NCS increases the sensitivity in the detection of neuropathy.
Keywords: Impaired glucose tolerance, nerve conduction studies, neuropathy, prediabetes, quantitative sensory testing
Diabetes mellitus (DM) and the associated disease outcomes are a cause for concern worldwide. The current global prevalence of DM for all ages has been estimated at 2.8% and is projected to be 4.4 percent by 2030.  The need of the hour is to identify and treat the risk factors that may prevent the onset of DM and minimize morbidity. DM is often associated with peripheral neuropathy (PN) an important cause of foot ulceration, and amputation. Currently, there is a growing focus on the study of early diabetic neuropathy. The Diabetes Control and Complications Trial (DCCT) demonstrated that improved glycemic control can slow the progression of neuropathy.  Recent observational studies suggest an association between impaired glucose tolerance (IGT) and neuropathy. These studies have suggested a predominance of involvement of small nerve fibers in the neuropathy of IGT. , It has been shown that IGT is an indicator of pre-diabetes.
Prediabetes is a condition in which blood glucose levels are higher than normal, but not high enough to be diagnosed as type 2 DM. Pre-diabetes is a known risk factor for overt diabetes and its macrovascular complications. Although neuropathy is an extensively studied complication in patients with DM; however, the neuropathy risk in pre-diabetes has not been well characterized. The potential link between pre-diabetes and neuropathy has been recognized based on clinical observation that many patients with idiopathic neuropathy share phenotypic characteristics of diabetes, such as obesity, hypertension, and dyslipidemia, without having overt diabetes. This led several investigtors to examine the prevalence of pre-diabetes using OGTT. The prevelance of neuropathy in these patients was between 40% and 50%, These observations suggest that a substantial proportion of subjects with IGT exhibit PN and/or neuropathic pain.  But, whether neuropathy already exists in the pre-diabetic stage with IGT is not clear. With this background, this present study was undertaken to determine the prevalence of neuropathy in patients with pre-diabetes with IGT using different electrophysiological techniques who had no other identifiable cause of neuropathy.
The study was conducted in the department of Neurology, at our institute and the study cohort included 58 subjects. We adhered to the good clinical practice guidelines of Helsinki declaration. The protocol was approved by the hospital ethics committee and informed consent was taken from all the subjects. The study subjects were referred from a diabetes health camp conducted at the department of Endocrinology. These subjects were diagnosed to have prediabetes as for the World Health Organization (WHO) criteria.  Age-matched subjects with normal GTT were the controls. The subjects were followed for a 4-year period for development of diabetes.
The subjects were considered prediabetic if the fasting plasma glucose was less than 126 mg/dl and the 2-hour oral glucose tolerance test (OGTT) was between 140-200 mg/dl. Other inclusion were: Normal glycosylated hemoglobin (HbA1c), normal vitamin B12 levels and thyroid function. Serum antinuclear antibodies (ANA) were negative in all of the patients. Individuals who had non-healing skin sores or any foot deformities, history of exposure to agents such as heavy metals, anti-tuberculosis drugs, cancer drugs, opioid drugs were excluded from the study. Individuals who were fasting or had a light breakfast were also excluded from the study. Also, anti-hypertensive medication if any was administered to study participants after the tests were performed.
A detailed clinical examination was done by a neuromuscular expert using neuropathy total symptom score (NTSS).  All patients were clinically normal. Complete clinical neurological assessment was done on all these patients by a neuromuscular specialist. Subjects who had symptoms of neuropathy were excluded from the study. None of the patients included in the study had neurological deficit.
Medtronic Keypoint machine (Denmark) was used for nerve condiction studies (NCS) using surface electrodes with >45-min acclimatization (room temperature 22-24°C). Examinations were performed as per the established clinical practice. Conventional NCS were done in one upper and both lower limbs: Median, ulnar, bilateral common peroneal, posterior tibial, sural nerves, dorsal sural nerves, and medial and lateral plantar nerves. Neuropathy was considered when abnormal NCS were present in ≥2 nerves.  NCS was considered abnormal if there was reduction of Compound Muscle Action Potential / Sensory Nerve Action Potential (CMAP/SNAP) amplitudes, reduction in conduction velocity, increased distal latencies more than 2 standard deviations.
Tests for evaluation of autonomic function tests (AFT) were: Heart rate response to deep breathing (HR DB ), valsalva maneuver, standing (30:15 ratio), blood pressure response to standing and sympathetic skin response (SSR). Abnormality in ≥2 tests were considered as abnormal AFT. 
Quantitative sensory testing (QST) was performed by Computer Assisted Sensory Examination System IV (CASE IV-WR Medical). QST was performed at the dorsal surface of great toe. Vibration detection threshold (VDT) was estimated at maximum allowed stimulation of 576.60 μm and tested at 4, 2, and 1 stepping algorithm with null stimuli.  Thresholds ≥95 th percentile for age were taken as abnormal. Cooling detection threshold (CDT) was evaluated at the dorsal surface of the foot. The maximum allowed stimulation was 9 o C for 10 sec. with starting temperature of 30 o C and ramp rate of 4.0 o C. at 4, 2, and 1 stepping algorithm with null stimuli. Heat pain threshold (HPT) was evaluated at the dorsal surface of the foot. Maximum stimulation was fixed at 45 o C for 10 sec., with baseline skin temperature of 34 o C, measured on scale of 1 to 10 where one was considered as the least discomfort or pain and 10 as most severe pain. It was assessed by testing with non-repeating ascending stepping algorithms with null stimuli. 
The significance of the variables were calculated using Chi square test (P < 0.05). The correlation between different electrophysiological tests was studied using Pearsons correlation co-efficient method.
The study cohort included 42 males and 16 females. The mean age was 53.8 years and body mass index varied from <25 to >30 kg/m 2 . The demographic characteristics are shown in Tab le 1.
Of the enrolled subjects, 10 had history of arthralgia especially knee joint (ANA negative), 10 had a history of hypertension, and 6 gave history of smoking. Lipid profile was abnormal in 16 subjects, 10 of them with neuropathy. Two subjects had both hypertriglyceridemia and low high-density lipoprotein (HDL) cholesterol values and the rest had only low HDL abnormality. It was also observed family history of DM was present in 28 (48.3%) subjects, 12 of them had neuropathy. Of the 58 subjects, 56 (96.6%) subjects had body mass index (BMI) <30. All subjects were followed for 4 years and 20 (34.5%) subjects developed DM and needed treatment [Table 1].
Of rhw 58 subjects with IGT, 19 (38.2%) had neuropathy. Nerve condiction studies showed evidence of neuropathy in 14 (24.1%), 5 (8.6%) had NCS abnormalities in ≥2 nerves, 9 (15.5%) had medial plantar and dorsal sural nerve involvement, and 6 (10.3%) had NCS abnormalities in one nerve combined with medial plantar or dorsal sural nerve [Table 2].
Autonomic neuropathy was demonstrated in 8 (13.8%) subjects: HR DB abnormalities in 4 (6.89%); valsalva maneuver, in 5 (8.6%); standing 30:15 ratio in 2 (3.5%); and SSR in 3 (5.2%) [Table 2]. Prevalence of neuropathy as assessed by QST was seen in 16 (27.59%) subjects. Of them, VDT was abnormal in 11 participants, CDT in 16, and HPT in 10 subjects [Table 2].
There was no correlation between NCS/QST/AFT abnormalities using Pearsons correlation coefficient method [Figure 1]a-c.
Neuropathy was earlier thought to be a result of advanced DM. However, recent reports suggest that neuropathy can occur early in the stage of pre-diabetes. Also 10% of diabetic patients have PN at the time of the initial diagnosis, suggesting an early axonal injury in the course of glucose intolerance.  Prediabetes is an independent predictor of conversion to type 2 DM and most can be identified by doing fasting glucose. Both Impaired fasting glucose (IFG) and IGT are risk factors for type 2 DM and risk is even greater when IFG and IGT occur together. During 8 year follow-up period, 15/37 (40.5%) prediabetic subjects developed DM.  In the present, study, 20 (34.5%) developed DM which is comparable to previous report.
It is important to identify these cases as they pose a therapeutic challenge and also appropriate treatment cannot be instituted without an accurate diagnosis. Evaluation of IGT and its association with neuropathy is important as it was projected by 2011 there would be 77.2 million people with prediabetes in India.  There is evidence that neuropathy occurs earlier in the course of hyperglycemia than other microvascular complications. The Diabetes Prevention Program (DPP) has demonstrated that intensive diet and exercise counseling slows down the progression of IGT to DM. 
In this study, neuropathy was found in 32.8% subjects. An earlier study has shown a prevelance of 44.4%.  Evidence suggests that IGT or IFG may be associated with early painful small fiber neuropathy involving the epidermal unmyelinated nerve fibers.  In a series of patients with idiopathic axonal neuropathy in many neurology clinics, there is a high prevalence of IGT. In spite of extensive studies, it is unclear whether impaired glucose metabolism causes diabetic polyneuropathy. The results vary considerably in different studies. The controversy whether IGT represents a mechanism of neuropathy or whether it is a covariant condition nerve fibers instead, continues to be a debate. However, recently, it was shown that the prevalence of retinopathy, nephropathy, and neuropathy was not high in those with IGT, IFG IA1C.  Neuropathy was evaluated using NCS, AFT, and QST. NCS is considered to be gold standard, accurate and reliable indicators of neuropathy. It was reported that even with normal NCS findings of lower extremity sensory nerves, there can be NCS changes in the more distal portions of these nerves, e. g., dorsal sural, medial dorsal cutaneous, and medial planter nerves. , With this background, in this study, NCS was also performed on dorsal sural and medial planter nerve to increase the diagnostic sensitivity of NCS. It was demonstrated that small fiber neuropathy is more common in individuals with IGT as reported by previous studies. , It was found that out of the three parameters studied NCS detected neuropathy in 14 (24.1%) subjects. AFT was performed to detect neuropathy as there are earlier reports that IGT and early diabetic neuropathy is associated with mild autonomic neuropathy. This study included both sympathetic and parasympathetic components of AFT. Neuropathy was detected in 8 (13.8%) of the subjects by this method that is similar to earlier study.  Previous studies mentioned that QST is an important non-invasive tool for the investigation of diabetic sensory polyneuropathy. It is a reliable way of assessing large and small sensory nerve fiber function. The tests are capable of assessing individual sensory modalities like perception thresholds for vibration (VDT), cold (CDT), and heat stimuli (heat pain). VDT results are believed to represent large fiber sensory nerve, whereas CDT and heat pain, small fiber sensory nerve function. It is believed that QST is a psychophysical test and lacks the objectivity of NCS. The results are subject to changes owing to distraction, boredom, mental fatigue, drowsiness, or confusion.  The above criteria should be considered before selecting subects for QST. QST using CASE IV device is an appropriate screening tool and endpoint measure for multicenter trials of neuropathy in IGT. The device was developed by Dyck and colleagues. This has undergone a number of modifications in machine specifications and algorithms.  In this study, QST using CASE IV device was demonstrated to be a more appropriate screening tool for IGT neuropathy as QST detected neuropathy in 16 (27.6%) participants. Among the QST tests, CDT was found to be abnormal in more number of participants (27.6%), compared to abnormalities in VDT (18.9%) and HPT (17.2%). It was observed that CDT was more reproducible than HPT. This may be due to the fact that the mean age of the participants was 53.8 years and that sensitivity to HPT decreases at the dorsal surface of the foot at such age. It is known that small myelinated and unmyelinated A-delta fibers carry temperature sensation. CDT was evaluated at 4, 2, and 1 stepping algorithm with null stimuli. This algorithm is less time consuming, a positive response to null stimuli prompts reinstruction and retesting the subject. Subjects with pre-diabetes can show various forms of diabetic neuropathy: Peripheral neuropathy,  small-fiber neuropathy,  and autonomic neuropathy.  As there is no correlation between peripheral large fiber neuropathy (NCS) and small fiber neuropathy (QST and AFT) both neuropathies can occur independently.
Hyperglycemia, abdominal obesity, or BMI > 30, elevated blood pressure, elevated triglycerides, and reduced HDL cholesterol constitute the metabolic syndrome. Any three of these five components confer the diagnosis. Several studies have shown that BMI > 30 is a risk factor for prediabetes and DM. However, prediabetes is seen in non-obese individuals (BMI < 30 kg/m 2 ). , In our study, 96.6% had BMI < 30. Low HDL cholesterol has been found to be a risk factor for developing neuropathy as described previously. 
It can be inferred from this study that QST can be a more useful technique than NCS as it may allow earlier diagnosis of neuropathy in IGT. But, QST tests are dependent on subjective responses and therefore have a high interobserver variability and poor reproducibility. It is, therefore, suggested to combine QST with NCS and with patient's symptoms to arrive at a diagnosis. This study can be further extended to a larger sample size. Also, skin biopsy can be included as a measure of neuropathy.
The study demonstrated that a considerable proportion of subjects with IGT exhibit subclinical, asymptomatic PN especially of small fibers. Dorsal sural nerve conduction and medial plantar nerve conduction study should be included in the protocol. Family history of diabetes may predispose to prediabetes neuropathy. Subjects with dyslipidemia and BMI <30 had a higher prevalence of neuropathy. Early detection of such signs in at-risk population is important as significant number can convert into DM. The progression of the same can be reduced by lifestyle changes.
We acknowledge the support given by Dr. Usha to carry out this study.
[Table 1], [Table 2]