 |
|
||||
| Home | Login | |||||
| About | Editorial board | Articles
|
NSI Publications
|
Search | Instructions | Online Submission | Subscribe | Videos | Etcetera | Contact |
|
||||||||||||||||||||
|
|
|
High resolution ultrasonography of peripheral nerves in diabetic peripheral neuropathy
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.250719
Keywords: Diabetic peripheral neuropathy, high-resolution sonography, nerve conduction study
Diabetes has become one of the largest global health-care problems because of the lifestyle changes. In certain parts of the world like India, the disease has reached epidemic proportions. The disease affects almost all the organs and has complications including cardiovascular disease, nephropathy, retinopathy, and neuropathy, such as diabetic peripheral neuropathy (DPN). The diagnosis of diabetic neuropathy is based primarily on the characteristic symptoms and is confirmed with a nerve conduction study (NCS). With advances in the ultrasound (US) technology and improvement in resolution, imaging of peripheral nerves has become feasible. Entrapment neuropathies, traumatic peripheral nerve injuries, nerve sheath tumors, etc., have been evaluated using high-resolution ultrasonography (HRU). Recently, there has been some interest in evaluating peripheral neuropathy in diabetics using HRU. Clinical and electrodiagnostic criteria have been developed to guide the clinician toward the correct diagnosis. As many peripheral nerves run a superficial course, they can be easily accessible for evaluation by sonography by their predictable anatomic location and characteristic appearance. Using HRU, nerves may be examined over a long course within a few minutes. HRU can assess even very small nerves and fascicular patterns. Nerves appear more echogenic than muscles and less echogenic than tendons. On a longitudinal scan, a nerve appears as multiple hypoechoic bands, corresponding to a neuronal fascicle, separated by hyperechoic lines, corresponding to the perineurium. On transverse scan, a nerve gives rise to the characteristic “honeycomb” appearance with a network of multiple rounded hypoechoic fascicle groups surrounded by hyperechoic peri- and epineurium [Figure 1].[1],[2],[3],[4],[5],[6],[7],[8],[9],[10],[11],[12],[13],[14],[15],[16],[17],[18],[19],[20],[21],[22],[23],[24],[25],[26]
Thirty-seven adults with known diabetes, i.e., 24 males and 13 female patients, presenting with clinical features suggestive of DPN were included in the study. The youngest patient included in the study was 34 years of age, and the oldest was 70 years of age. Forty-five healthy non-diabetic volunteers with no clinical evidence of peripheral neuropathy, who had undergone blood sugar evaluation (i.e., normal fasting and random blood sugar level) in the past 1 year, which was found to be normal, were also selected. Out of the 45 healthy volunteers studied, 25 (55.5%) were males, and 20 (44.4%) were females. The youngest healthy volunteer included in the study was 30 years of age, and the oldest was 68 years of age. The healthy volunteers were drawn from hospital employees and colleagues. There was no significant difference noted in cross-sectional areas (CSAs) of all evaluated peripheral nerves in different genders of healthy volunteers and diabetic patients. The study was conducted over a period of 18 months. The institutional ethics committee and review board approved it, and informed consent was obtained from both the patients and the healthy volunteers.
In patients with diabetic peripheral neuropathy Clinical findings such as tingling, numbness, or any weakness in the hands and feet, and details of the NCS were noted in the patient group. A detailed sonographic examination of the peripheral nerves in healthy volunteers and patients was done in the right upper and lower limb using high resolution 5–18 MHz linear probe on Philips iU22 ultrasound system. All patients and healthy volunteers were examined in the supine position, except for the evaluation of common peroneal nerves (CPNs), where the patients were examined in the prone position. In our study, we preferred HRU of right-sided upper and lower limb peripheral nerves as electrophysiological studies, i.e. NCS, are routinely done on the right side. The peripheral nerves were identified by their predictable anatomic location and characteristic appearance. Nerves appear more echogenic than muscles and less echogenic than tendons. On a longitudinal scan, a nerve appears as multiple hypoechoic bands, corresponding to a neuronal fascicle, separated by hyperechoic lines, corresponding to the perineurium. On transverse scan, a nerve gives rise to the characteristic “honeycomb” appearance with a network of multiple rounded hypoechoic fascicle groups surrounded by hyperechoic peri- and epineurium. Following findings were recorded CSA of the peripheral nerves, including the median and ulnar nerves in right upper limb and the posterior tibial and the common peroneal nerves (CPNs) in the right-lower limb were measured in mm2 in the same anatomical location as in the normal healthy volunteers. In most studies done previously, CSA of median nerve was measured at carpal tunnel, however, we measured median nerve at 5 cm proximal to the wrist joint, the mid-forearm level, and at the elbow joint because the aim was not to evaluate compression neuropathy such as carpal tunnel syndrome. The ulnar nerve was measured at the wrist joint and behind the medial epicondyle. CPN was measured at the neck of the fibula, and the posterior tibial nerve was measured 3 cm above the medial malleolus. The mean CSA of the involved nerves was measured in the two groups at identical positions.[6],[8],[9],[10],[11],[12],[13] All the patients underwent NCS; the findings of studies, including latency (L) and velocity (V), were recorded in the right upper limb and lower limb. The neuropathy was classified as mild or moderate to severe DPN on the basis of L and V assessed by NCS. The CSA measured on USG in diabetic patients was compared with that in the normal healthy volunteers at the same level.[15],[16],[17],[18],[19] The measurements were also correlated with findings of NCS in all patients. The CSA of the examined nerves at various levels was expressed in mean and standard deviation, and using Student's t-test for normal distribution, we compared the difference between the mean values of the healthy volunteers and patients.
The study included 45 healthy volunteers and 37 patients of diabetic mellitus presenting with symptoms of peripheral neuropathy. In the patient group, the duration of diabetic mellitus varied from 34 to 70 years with a mean age of 53.35 years. Out of the 37 patients studied, 24 (64.86%) were males, and 13 (35.13%) were females. The youngest patient included in the study was 34 years of age, and the oldest was 70 years of age. Forty-five healthy non-diabetic volunteers with no clinical evidence of peripheral neuropathy, who had undergone blood sugar evaluation (i.e., fasting and random blood sugar level) in the past 1 year, which was found to be normal, were also selected. Out of the 45 healthy volunteers studied, 25 (55.5%) were males and 20 (44.4%) were females. The youngest healthy volunteer included in the study was 30 years of age, and the oldest was 68 years of age. The healthy volunteers were drawn from hospital employee and colleagues. There was no significant difference noted in the CSAs of peripheral nerves in different genders of both groups. The mean CSA of the median nerve 5 cm proximal to the wrist crease was 11.61 ± 2.87 mm2 in patients with DPN, which was significantly higher than seen in healthy volunteers where the mean CSA was 7.09 ± 1.49. The mean CSA at the level of mid-forearm was 9.76 ± 2.95 mm2 in patients with DPN, which was also significantly higher than the value obtained in healthy volunteers where the mean CSA was 6.5 ± 1.31 mm2. The mean CSA at the elbow joint was 10.50 ± 2.27 mm2 in patients with DPN, which was significantly higher than the value seen in healthy volunteers where the mean CSA was 6.90 ± 1.43 mm2 [Figure 2].
The mean CSA of the ulnar nerve at the wrist joint was 7.22 ± 1.66 mm2 in patients with DPN, which was significantly higher than the value in healthy volunteers where the mean CSA was 5.56 ± 1.22 mm2. The mean CSA behind the medial epicondyle was 10.33 ± 3.45 mm2 in patients with DPN, which was significantly higher than the value in healthy volunteers where the mean CSA was 6.50 ± 1.41 mm2 [Figure 3].
The mean CSA of the posterior tibial nerve 3 cm above the medial malleolus in a patient with DPN was 7.99 ± 2.15 mm2, which was significantly higher than the value in healthy volunteers where the mean CSA was 5.97 ± 1.83 mm2 [Figure 4].
The mean CSA of CPN at the neck of the fibula was 11.38 ± 4.93 mm2 in patients with DPN, which was significantly higher than the value in healthy volunteers where the mean CSA was 6.75 ± 1.4 mm2 [Figure 5].
The correlation of the mean CSA of patients with peripheral nerve affliction based on the severity of DPN was in accordance with the findings of NCS. Median neuropathy was classified as mild or moderate to severe DPN on the basis of L and V assessed by NCS at the wrist, on the basis of criteria described by Watanabe T et al.[10] The patients with motor nerve conduction V >52 m/s and L <4 ms of median nerve measured at the wrist joint were classified as having a mild DPN, and patients with motor nerve conduction V ≤52 m/s or L ≥4 ms on NCS were classified as having a moderate to severe DPN. In our study, 7 out of 37 patients were included in the mild DPN group, and 30 out of 37 patients were included in the moderate to severe DPN group, on the basis of the criteria defined above. The mean CSA of the median nerve 5 cm proximal to the wrist crease was higher in moderate to severe DPN (12 ± 3.63 mm2) than in patients with mild DPN (9.97 ± 1.92 mm2), but this difference was not statistically significant [Table 1].
Ulnar neuropathy was classified as mild or moderate to severe DPN on the basis of L and V assessed by NCS, as described by David C. Preston a nd Barbara E Shapiro.[25] The patients with motor nerve conduction with V >49 m/s and L <3.4 ms of the ulnar nerve measured at the wrist joint were classified as having a mild DPN, and patients with V ≤49 m/s or L ≥3.4 ms on NCS were classified as having a moderate to severe DPN on the basis of nerve conduction velocity (NCV) values. In our study, 16 out of 37 patients were classified as mild DPN, and 21 out of 37 patients were classified as having a moderate to severe DPN on ulnar NCS. The mean CSA of the ulnar nerve at the wrist crease was higher in moderate to severe DPN (10.72 ± 2.46 mm2) than in the patients with mild DPN (7.28 ± 1.77 mm2). This difference was statistically significant with a P value of < 0.0001 [Table 2].
Patients with a posterior tibial neuropathy were classified as having a mild DPN when a V >42 m/s and a L <4.4 ms of the posterior tibial nerve measured at ankle joint was present. Patients with a V ≤42 m/s or a L ≥4.4 ms on NCS were classified as moderate to severe DPN.[10] In our study, 10 out of 37 patients were classified as having a mild DPN, and 27 out of 37 patients were classified as having a moderate to severe DPN. The mean CSA of posterior tibial nerve 3 cm above the medial malleolus was higher in patients with a moderate to severe DPN (8.38 ± 2.18 mm2) as compared to that in patients with mild DPN (6.95 ± 1.75 mm2). However, this difference was not statistically significant.
Diabetes is a common problem in the world affecting almost all the organs. Although diabetic neuropathy is not life-threatening, it can be extremely debilitating for the patient. The diagnosis of diabetic neuropathy is based primarily on characteristic symptoms and is confirmed with a NCS. HRU of the peripheral nerves is a recent development in neurology. With advances in US transducer technology and improvement in the resolution routine, the use of ultrasound for nerve imaging has become possible. Peripheral nerves run superficially in many cases, and they can be evaluated using HRU. Further, because of their characteristic fascicular pattern, even very small nerves can be studied. On HRU, nerves are more echogenic than muscles but less echogenic than tendons and are less mobile than tendons. High resolution sonography was able to demonstrate median, ulnar, posterior tibial, and CPNs in all the healthy volunteers and patients examined in our study.[1],[2],[26],[27],[28],[29] The mean CSA of the median nerve in our study was significantly higher (P < 0.0001) at all three levels of examination in patients with DPN than the healthy volunteers i.e., 11.61 ± 2.87 mm2 vs. 7.09 ± 1.49 mm2 at 5 cm proximal to the wrist crease, 9.76 ± 2.95 mm2 vs. 6.5 ± 1.31 mm2 at the level of mid-forearm, and 10.50 ± 2.27 mm2 vs. 6.90 ± 1.43 mm2 at the elbow joint. Our findings were in concordance with those of Watanabe T et al., who found that there is a significant increase in the CSA of the median nerve in patients with DPN as compared with the controls.[7],[10] The mean CSA of the median nerve in patients vs. healthy volunteers was 13.5 ± 2.8 mm2 vs. 9 ± 2 mm2, 9.1 ± 2.7 mm2 vs. 7.1 ± 1.6 mm2, and 7.2 ± 2.6 mm2 vs. 7.4 ± 3.2 mm2 at the carpal tunnel, 5 cm proximal to the wrist crease and at the elbow joint, respectively. Pitarokoili et al., also found an increased mean CSA (in mm2) of the median nerve in patients as compared to controls, i.e., 11.21 ± 2. 1 vs. 9.0 ± 1.58 at the carpal tunnel and 7.92 ± 2.14 vs. 6.6 ± 1.6 at mid-forearm.[24] Similar findings have been reported by Afsal M et al.,[27] who found that on high-resolution sonography, there was a diffuse thickening of the peripheral nerve in these patients, and the mean CSA of the median nerve and ulnar nerve was found to be significantly higher in patients with DPN than the normal matched controls. The mean CSA of the median nerve of patients with DPN was 10.5 mm2 at the carpal tunnel, and 8.69 mm2 in the forearm 10 cm proximal to the wrist, whereas the CSA was 7.1 mm2 and 5.89 mm2, respectively at the same levels in controls.[28] When the patients with DPN were classified into the “mild” and “moderate to severe” groups on the basis of motor nerve conduction V and L, the mean CSA of median nerve was higher in patients with moderate to severe DPN (i.e., 12 ± 3.63 mm2) than those with mild DPN (9.97 ± 1.92 mm2), but in our study, the difference was not statistically significant.[10] The CSA of the ulnar nerve was significantly higher (P < 0.0001) at both the levels in DPN patients than in the healthy volunteers, i.e., 7.22 ± 1.66 vs. 5.56 ± 1.22 mm2 at the wrist joint, and 10.33 ± 3.45 vs. 6.50 ± 1.41 mm2 behind the medial epicondyle. Our findings were in concordance with that of Pitarokoili et al.,[24] who found that there was a significant increase in the CSA of the ulnar nerve, and it was greater in patients with DPN than in the controls, i.e., 6.30 ± 1.92 mm2 vs. 5.0 ± 0.94mm2 at the Guyon's canal and 8.34 ± 2.46 mm2 vs. 5.99 ± 1.57 mm2 behind the medial epicondyle.[14],[24] When we classified ulnar neuropathy in patients with DPN into the “mild” and “moderate to severe” groups on the basis of motor nerve conduction V and L described by David C. Preston, Barbara Ellen Shapiro,[25] the mean CSA of ulnar nerve was significantly higher in patients with moderate to severe DPN than in those with mild DPN (P < 0.0001) and this difference was statistically significant. Similar findings were recorded by Afsal M et al.[27] The CSA of the posterior tibial nerve was significantly higher (P < 0.0001) at 3 cm above the medial malleolus than that seen in the healthy volunteers (7.99 ± 2.15 vs. 5.97 ± 1.83 mm2). The CSA of the posterior tibial nerve was <7 mm2 in most of the healthy volunteers, whereas in the majority of patients with DPN, it was >7 mm2. The mean CSA of posterior tibial nerve was higher in patients with moderate to severe DPN than in those with mild DPN. Although the mean CSA of the patients and healthy volunteers examined in our study was lower than the mean CSA measured by Watanabe T et al.,[10] our finding that the mean CSA of the posterior tibial nerve in patients with DPN was significantly larger than the mean CSA in healthy volunteers was concordant with the findings of Watanabe et al.[10] Similar findings were reported by Riazi S et al.,[11] Pitarokoili et al,[24] that the mean CSA of the posterior tibial nerve (P TN) was larger in patients with DPN than in control subjects.[24] Singh also found that the CSA and the maximum thickness of nerve fascicles of the tibial nerve are larger in patients with DPN than in healthy subjects or in diabetic patients with no signs of neuropathy.[29] When the posterior tibial nerve neuropathy in patients with DPN was classified into the “mild” and “moderate to severe” types on the basis of motor nerve conduction V and L, the mean CSA of the posterior tibial nerve 3 cm above medial malleolus was higher in patients with moderate to severe DPN (8.38 ± 2.18 mm2) than that in patients with mild DPN (6.95 ± 1.75 mm2). However, this difference was not statistically significant. Such a correlation was not done in the previous studies. The mean CSA of CPN at the neck of the fibula was significantly higher (P value < 0.0001) in DPN patients than in healthy volunteers i.e., 11.38 ± 4.93 mm2 vs. 6.75 ± 1.4. In majority of the patients with DPN, i.e., 28/37 (75.68%), the mean CSA of the CPN was >9 mm2 at the neck of fibula and was between 9–12 mm2 in 43.24% of patients; and, in majority of the healthy volunteers i.e., 62.22%, the CPN mean CSA was between 6–9 mm2, and it was <9 mm2 in 93.33% healthy volunteers. The study by Pitarokoili et al.,[24] also showed similar results where the mean CSA (in mm2) of the fibular nerve at the fibular head was greater in patients than in the controls (i.e., 12.22 ± 4.97 vs. 7.2 ± 1.89).[24] The correlation of HRU findings with NCS was not possible because this nerve is not routinely evaluated by NCS. Thus, enlargement of nerve size, as measured by CSA, was a common finding in all diabetic patients with neuropathy. Although the CSA of nerves, such as median, ulnar, posterior tibial, and common peroneal, varied with the severity of neuropathy, as determined by NCS, this was not statistically significant. Further studies with larger number of patients are probably required. The study shows the utility of HRU in demonstrating changes of DPN in a large number of upper and lower limb nerves, which have been previously evaluated in a few studies. This has the potential to allow for a non-invasive evaluation of DPN.
HRU is more comfortable and less time-consuming as compared to electrodiagnostic testing. HRU is especially attractive because it allows the evaluation of multiple nerves over a long course. Peripheral nerves can be reliably demonstrated by HRU in both healthy volunteers and diabetic patients. HRU demonstrates an increase in CSAs in all peripheral nerves examined in a patient with DPN as compared to normal healthy volunteers. This difference was statistically significant. As a HRU allows the evaluation of both the nerve and its surrounding structures, it has the potential to demonstrate other lesions masquerading as neuropathy. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2]
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|||||
