Clinical diagnosis of distal diabetic polyneuropathy using neurological examination scores: correlation with nerve conduction studies

Aim The aim of this study was to diagnose diabetic sensorimotor polyneuropathy using neurological examination scores and to correlate the findings with nerve conduction studies (NCS). Patients and methods Thirty patients with type 2 diabetes were included in the study. Detection and grading of neuropathy were carried out based on the Diabetic Neuropathy Symptom (DNS) Score, modified Neuropathy Symptom Score (NSS), Diabetic Neuropathy Examination (DNE), and modified Neuropathy Disability Score (NDS). For the NCS, amplitudes, velocities, and latencies of seven nerves - that is, four motor (median, ulnar, tibial, and common peroneal) and three sensory (median, ulnar, and sural) nerves - were recorded. If the patient had two or more abnormal findings in any of these nerves, the patient was diagnosed as having peripheral sensorimotor neuropathy. Thereafter, the sensitivity, specificity, and diagnostic efficacy of each neurological score were recorded taking NCS as the gold standard. Results Diabetic sensorimotor polyneuropathy was diagnosed clinically and electrophysiologically in 17 patients (56.7%). However, there were nine cases (30%) of subclinical neuropathy. Neurological examination scores were significantly correlated with each other and with individual variables of NCS and the nerve conduction sum score. Taking the NCS as gold standard, DNS, modified NSS, DNE, and modified NDS had 65.4, 61.5, 30.8, and 61.5% sensitivity and 100, 75, 100, and 100% specificity, respectively. Their diagnostic efficacies were 70, 63.3, 40, and 66.7%, respectively. Conclusion Neurological examination scores can detect and grade neuropathy in the majority of cases. However, NCS was accurate for detection of diabetic sensorimotor polyneuropathy, especially for the subclinical neuropathies.


Introduction
Distal symmetric polyneuropathy or diabetic sensorimotor polyneuropathy (DSPN) is a frequent complication of both type 1 and type 2 diabetes [1]. DSPN encompasses a group of clinical and subclinical syndromes with varied etiologies and clinical and laboratorial manifestations, defined by the progressive diffuse or focal degeneration of peripheral somatic and autonomic nerve fibers [2]. measurement of glycosylated hemoglobin (HbA1c) is highly recommended for diabetes management. Maintaining HbA1c levels below 6.5% is a major goal of diabetes management, because HbA1c levels correlate well with diabetes complication risks [8].
The aim of the present study was to diagnose DSPN clinically in patients with type 2 diabetes using neurological examination scores, which are easy to perform, and to correlate these scores with electrophysiological measurements.

Study population
Thirty patients diagnosed with type 2 diabetes (according to the American Diabetes Association criteria) [9] were included in the study. Patients were recruited from the Rheumatology and Rehabilitation department and diabetes outpatient clinic of Internal Medicine department. Ten healthy age-matched and sex-matched participants served as the control group for electrodiagnostic studies. Informed consent was obtained from all participants in the study. The study was approved by the ethics committee of the Faculty of Medicine.
Patients having any evidence of autoimmune diseases (such as rheumatoid arthritis, systemic lupus erythematosus, and scleroderma), cervical and/or lumbar disc lesions, neuropathies of other etiologies (such as heredofamilial polyneuropathy, exposure to neurotoxic drugs, infections, or neuropathies due to other causes or renal failure) were excluded.

Neurological examination scores
We selected four scores on the basis of ease of performance and common use. The scores were DNS, modified NSS, DNE, and modified NDS [6,7].

Diabetic Neuropathy Symptom score
All participants were questioned as regards the presence of symptoms, either positive or negative, suggesting the presence of neuropathy.
The questions were answered with a 'yes' (positive: one point) if a symptom had occurred several times a week during the last 2 weeks, or with a 'no' (negative: no point) if it did not. The patients were questioned as regards the presence of following symptoms: (1) Symptoms of unsteadiness when walking.
(2) Burning, aching pain, or tenderness in legs or feet.
(3) Pricking sensations on legs and feet. (4) Regions of numbness on legs or feet.

Modified Neuropathy Symptom Score
Patients were questioned about the presence or absence of numbness, abnormal hot or cold sensations, tingling sensations, burning pain, irritation from bed clothes in the lower legs and feet, and nocturnal exacerbation of muscular cramps and whether maneuvers could reduce the symptoms. One point for the presence of each of these symptoms was assigned. For the first five symptoms one extra point was added if nocturnal exacerbation was present. The maximum score is 10 points. A score of more than 1 point is defined as positive for PNP [10] (Table 1).

Diabetic Neuropathy Examination score
The score contains two items for muscle strength, one pertaining to reflexes and five pertaining to sensation (eight total items). Each item is scored from 0 to 2 (0 is normal and 2 severely disturbed). The maximum score is 16 points. A score greater than 3 points is defined as positive for PNP [11] (Table 2).

Modified Neuropathy Disability Score
The modified NDS can be easily performed in the clinical setting and takes only a minute or two to complete and provides an assessment of the risk for neuropathic ulceration. The score is based on vibration perception, pin-prick sensation, temperature perception, and ankle (Achilles) reflexes. The maximum deficit score is 10, which would indicate complete loss of sensation to all sensory modalities and absent reflexes. A score of 6 or more has been found to indicate an increased risk for foot ulceration [7,12] (Table 3).
The patients were diagnosed as having clinically detectable neuropathy (group I) if DNS was 1 or more, modified NSS was greater than 1, or DNE was greater than 3. Undetectable neuropathy (group II) was diagnosed if DNS was less than 1, modified NSS was 1 or less, or DNE was 3 or less. Modified NSS and modified NDS were used to quantify the severity of the neuropathy [13].
Motor and sensory distal latencies, amplitudes, and nerve conduction velocities were measured in the upper and lower limb nerves. The simplified NCS protocol was followed to record the NCS of the patients [14].
Five nerves were tested (median, ulnar, tibial, common peroneal, and sural nerves). The NCS were used to identify normal or affected nerves. The patients were diagnosed as having polyneuropathy if the value of two or more parameters was abnormal in one nerve, or one parameter was abnormal in any two nerves. Amplitudes, velocities, and latencies of seven nerves -that is, four motor (median, ulnar, tibial, and common peroneal) and three sensory (median, ulnar, and sural) nerves -were recorded. An overall Nerve Conduction Sum score was defined as the number of these five nerves with an abnormal conduction velocity, ranging from 0 (all normal) to 7 (all abnormal) [15].

Grouping of patients on the basis of neurological score and nerve conduction studies
For each neurological score, patients were divided into four groups: true positive, false positive, false negative, and true negative. If neuropathy was present by both clinical examination and NCS, the patient was included in the true-positive group, in the false-positive group if it was present on clinical examination but was absent on NCS, in the false-negative group if it was present on NCS but was absent on clinical examination, and in the true-negative group if it was absent on both testing methods. The sensitivity and specificity of each score were calculated taking NCS as the gold standard. Data presented in Table 8 were used to calculate test performance characteristics. Diagnositic efficacy of each test was derived from the same data.

Statistical analysis
Statistical analysis was performed using Statistical Package for the Social Sciences (SPSS) 16.0. Descriptive statistics were carried out using number and percentage, as well as mean and SD. The ANOVA test was used to compare the difference between more than two group means in interval and ordinal variables. Correlations  were determined using Pearson's correlation coefficient. The level of statistical significance was set at P less than 0.05. Test performance was calculated on the basis of the following equations: Sensitivity (%) = patients with true-positive neuropathy/ patients with true-positive neuropathy + patients with false-negative neuropathy ×100.

Results
The patient characteristics are presented in Table 4. The ages of the studied patients ranged from 32 to 70 years, and the duration of diabetes ranged from 0.5 to 15 years.

Results of neurological examination scores for all patients
The scores (mean ± SD) of all patients, the percentage of patients with abnormal scores, and grading of neuropathic symptoms and deficits are presented in Table 5. Out of 30 patients, nine (30%) had a modified NDS of 6 or more, which indicates an increased risk for foot ulceration.
Severity of neurological symptoms and deficits, graded using modified NSS and modified NDS, was significantly correlated with the disease duration (r = 0.59, P < 0.001; r = 0.66, P < 0.001, respectively), as well as with HbA1c (r = 0.48, P = 0.002 for both).

Results of nerve conduction studies
Comparison between patients with clinically detectable and undetectable neuropathy and the control group showed significant difference between the three groups as regards all parameters of motor NCS except for Data are presented as means ± SD or n (%). ulnar conduction velocity and tibial and peroneal distal latencies (Table 6).
Similarly, comparison between patients with clinically detectable and undetectable neuropathy and the control group showed significant difference between the three groups as regards all parameters of sensory NCS ( Table 7).

Comparison of neurological scores with nerve conduction studies
The comparison of neurological scores with NCS is presented in  Table 9.

Relation between clinical scores and nerve conduction sum score
In all patients, the nerve conduction sum score ranged from 1 to 7 according to the involved nerves.
There were significant correlations of the DNS, modified NSS, DNE, and modified NDS with the nerve conduction sum score (r = 0.71, P < 0.001; r = 0.58, P < 0.001; r = 0.66, P < 0.001; r = 0.73, P < 0.001, respectively) (Fig. 1). It can be interpreted that the more the neurological scores had deteriorated, the higher were the chances of further deranged nerve conduction sum score.

Discussion
The disease process of diabetes causes alterations in the normal nerve functions, which can be reflected either when performing neurological examination, or during electrophysiological testing of the patient. The neurological scores and the electrophysiological studies are used for the diagnosis of the sensorimotor neuropathy. The relations between physiology and pathophysiology emphasize the close interdependence between electrophysiological studies and clinical findings [16].
In the present study, we selected the four neurological scoring systems (DNS, modified NSS, DNE, and modified NDS), which were common and easy to perform, and compared them with NCS, which also has similar advantages. Out of 30 patients with type 2 DM, DSPN was diagnosed clinically (using neurological examination scores) and electrophysiologically (using NCS) in 17 patients (56.7%). However, there were nine cases (30%) of subclinical neuropathy. Subclinical neuropathy indicates the state of electrophysiologically verified neuropathy with the absence of subjective and objective neurological signs. It occurs in about 20% of patients with diabetes [17].
Studies on prevalence of neuropathy in type 2 diabetes had widely differing results, varying from 15 to 50%. The wide variability was attributed to differences in patient sample, diagnostic methods, and criteria adopted for diagnosis [18,19]. Studies that used NCS as a diagnostic marker also reported higher prevalence of neuropathy [20].
The higher prevalence of neuropathy in the present study may be due to the use of NCS for diagnosis of neuropathy, which is a more sensitive method.
In diabetic patients, correlations between various neuropathy tests and scores have previously been reported [13,21]. An association between NSS and NDS has been observed [22].  In the present study, the neurological examination scores significantly correlated with each other, which were similar to results of Meijer et al. [6]. At the Correlation between neurological examination scores and nerve conduction sum.

Figure 2
Correlation between disease duration and nerve conduction sum.
same time, the neurological examination scores were significantly correlated with individual variables of NCS. This is in accordance with the study by Meijer et al. [23], in which both scores strongly correlated with electrodiagnostic studies . In this study, we added the nerve conduction sum score, which ranged from 1 to 7 based on the involved nerves, and there were significant correlations with DNS, modified NSS, DNE, and modified NDS (r = 0.71, P <0.001; r = 0.58, P < 0.001; r = 0.66, P < 0.001; r = 0.73, P < 0.001, respectively).  [24] supports the fact that subclinical neuropathy can be detected with NCS. They proved that NCS was more accurate in the detection of neuropathy compared with clinical examination, especially in the subclinical group, although the latter also has its role in the detection of neuropathy.
In Pakistan, Niazi et al. [25] evaluated diabetic polyneuropathy by performing electrodiagnostic study on 41 patients. Although clinical examination was carried out in detail, no statistical comparison was made between the clinical findings and NCS. However, it was suggested that these studies are capable of diagnosing diabetic neuropathy even before clinical manifestations, which has been proved in our results.
We assessed the sensitivity and specificity of the four scores taking NCS as the gold standard. DNS was found to be the most sensitive test (65.4%), and DNS, DNE, and modified NDS had equal specificity (100%). DNS and modified NDS had a better diagnostic efficacy (70 and 66.7%, respectively). According to Asad et al. [26], NDS was found to be the most sensitive test (92%) and DNE had the highest specificity (81%). NDS and NSS had a better diagnostic efficacy (76.67%), but DNS sensitivity was 64%, which was similar to that reported in our result. This difference may be due to different geographic area and different races being examined.
Meijer et al. [6] also assessed the validity of DNS score against NSS. They did not use NCS. Their results as regards sensitivity and specificity were different from ours. They found high correlation between the two testing methods [6].
The results of both neurological scores and NCS in the present study confirmed the previous findings of greater involvement of the peripheral nervous system in diabetic patients with prolonged disease duration and elevated HbA1c [27]. Moreover, the severity of neurological symptoms and deficits was significantly correlated with the disease duration (r = 0.59, P < 0.001; r = 0.66, P < 0.001, respectively), as well as with HbA1c (r = 0.48, P = 0.002 for both).
In conclusion, neurological examination scores can detect and grade neuropathy in majority of cases. However, NCS was accurate for the detection of DSPN, especially subclinical neuropathies. Therefore, we can use each, NCS or neurological examination scores, for detecting and grading diabetic neuropathy and using both gives us a better chance for earlier diagnosis.

Financial support and sponsorship
Nil.

Conflicts of interest
There are no conflicts of interest.