Journal of Diabetes and Its Complications
Volume 25, Issue 1 , Pages 44-51, January 2011

New diagnostic tests for diabetic distal symmetric polyneuropathy☆

  • Nikolaos Papanas

      Affiliations

    • Outpatient Clinic of the Diabetic Foot in the Second Department of Internal Medicine at Democritus University of Thrace, Greece
    • Corresponding Author InformationCorresponding author. Democritus University of Thrace, 68100 Alexandroupolis, Greece. Tel.: +30 6977 544337; fax: +30 25510 74723.
    • ,
  • Dan Ziegler

      Affiliations

    • Institute for Clinical Diabetology, German Diabetes Center at the Heinrich Heine University, Leibniz Center for Diabetes Research and Department of Metabolic Diseases, University Hospital, Düsseldorf, Germany

Received 22 July 2009; received in revised form 13 September 2009; accepted 28 September 2009. published online 09 November 2009.

Article Outline

Abstract 

Neuropathy needs to be diagnosed early to prevent complications, such as neuropathic pain or the diabetic foot. It is obvious that diagnosis of neuropathy needs to be improved. New peripheral nerve function tests that appear to facilitate diagnosis are now emerging. This review outlines the new tests that have been proposed for the diagnosis of diabetic distal symmetric polyneuropathy, the commonest form of neuropathy in diabetes. New tests are classified into those mainly assessing large-fiber function (tactile circumferential discriminator, steel ball-bearing, and automated nerve conduction study) and those mainly assessing small-fiber function (NeuroQuick and Neuropad). Emerging tests are promising but must be evaluated in prospective studies. Moreover, their cost-effectiveness needs more careful appraisal. The clinician should, therefore, still rely on established modalities to diagnose neuropathy, but wider use of the new tests is expected in the near future.

Keywords: Diabetes mellitus, Diabetic neuropathy, Diabetic foot, Diagnosis, Large fibers, Small fibers

 

Back to Article Outline

1. Introduction 

Neuropathy is an important chronic complication of diabetes and is associated with increased morbidity and mortality (Boulton et al., 2005, Shaw et al., 2003, Várkonyi and Kempler, 2008). Its commonest form is distal symmetric polyneuropathy (Boulton et al., 2005, Shaw et al., 2003, Várkonyi and Kempler, 2008), which is a major cause of diabetic foot ulceration and may lead to chronic pain with reduced quality of life (Boulton et al., 2004, Edmonds, 2004, Ziegler, 2008).

Diagnosis of distal symmetric polyneuropathy rests on clinical examination (Boulton et al., 2005, Shaw et al., 2003, Várkonyi and Kempler, 2008). Ideally, one should assess all nerve functions. In terms of size, this should include both small and large fibers; in terms of function, it should entail autonomic, sensory, and motor modalities (American Diabetes Association and American Academy of Neurology, 1998). In practice, the clinician usually evaluates Achilles tendon reflexes, vibration perception (graded 128 Hz tuning fork), pressure and touch perception (10 g monofilament), pain sensation (e.g., Neurotip), temperature sensation (e.g., Tiptherm rod), and features suggestive of neuropathy (e.g., callus formation and neuropathic foot ulceration) (Feldman et al., 1994, Young et al., 1993). Abnormal findings obtained in standardized clinical examination may be added to form a neuropathy score (Feldman et al., 1994, Young et al., 1993). The most widely used clinical diagnostic tools are the Michigan Neuropathy Screening Instrument, proposed by Feldman et al. (1994), and the Neuropathy Disability Score (NDS), suggested by Young et al. (1993). The former evaluates feet appearance (deformity, dry skin, callus, and fissures), neuropathic ulceration, Achilles tendon reflexes, and vibration perception using a tuning fork (Feldman et al., 1994). The latter evaluates Achilles tendon reflexes, as well as 128-Hz tuning fork, pinprick, and temperature (cold tuning fork) sensation (Young et al., 1993). Further diagnostic tests for neuropathy include the 10-g monofilament [Semmes–Weinstein monofilament (SWMF)] and the vibration perception threshold (VPT). The SWMF assesses pressure/touch sensation. In one of several suggested approaches, it is applied perpendicular to the skin surface on three sites on the plantar aspect of the foot (first and fifth metatarsal head, heel) with enough force to cause it to buckle. The test is abnormal if the SWMF is not felt at any one site on either foot (Abbott et al., 2002). The VPT is assessed with a special device, either the Neurothesiometer or the (older) Biothesiometer. The tractor of the device is applied on the tip of the hallux, and the VPT is measured in volts. The VPT is abnormal when the mean voltage of three measurements exceeds 25 V (Bril & Perkins, 2002). Finally, nerve conduction study (NCS) significantly contributes to the diagnosis of neuropathy, enabling early diagnosis (Olaleye et al., 2001, Rota et al., 2005). However, it requires expensive equipment, is time-consuming, and, therefore, cannot be used as a screening test (Boulton, A.J., et al., 2005).

This review outlines the new peripheral nerve function tests that have been proposed for the diagnosis of diabetic distal symmetric polyneuropathy (Table 1). New tests are classified into those mainly assessing large-fiber function [tactile circumferential discriminator (TCD), steel ball-bearing, and automated NCS] and those mainly assessing small-fiber function (NeuroQuick and Neuropad).

Table 1. Sensitivity, specificity, and reproducibility of new diagnostic tests for diabetic distal symmetric polyneuropathy
New tests for the assessment of large-fiber function
New testSensitivity (%)Specificity (%)Reproducibility
TCD92.364.2Not reported
Steel ball-bearing84100Inter-observer: κ=.811
Intra-observer: κ=.749
Automated NCS9282CV<6% (controls)

New tests for the assessment of small-fiber function
New testSensitivitySpecificityReproducibility
Neuropad85–97.8%32–78.5%CV=4.1–5.1%
NeuroQuickAUC with 95% CI of ROC curves: NeuroQuick: 0.764 (0.688–0.840); foot warm TPT: 0.738 (0.664–0.812); foot cold TPT: 0.795 (0.727–0.864)CV=20.4% (controls); CV=8.5% (diabetic patients)

AUC, area under the curve; ROC, receiver operating characteristic.

Clinical examination was used as the gold standard for all tests, except for automated NCS, which was compared with standardized NCS.

Back to Article Outline

2. Peripheral nerves: large and small fibers 

Peripheral nerve fibers may be classified, in terms of their size and the presence or absence of myelin sheath, into large and small fibers (Al-Shekhlee et al., 2002, Chéliout-Héraut et al., 2005, Quattrini et al., 2004). Large fibers have a diameter of 6–12 μm, are protected by a myelin sheath, and mediate ankle reflexes, touch, pressure, vibration, and proprioception. Small fibers may be subdivided into Aδ fibers (myelinated, diameter=1–5 μm) and C fibers (unmyelinated, diameter=0.2–1.5 μm) (Al-Shekhlee et al., 2002, Chéliout-Héraut et al., 2005, Quattrini et al., 2004). Small fibers mediate sensation of temperature and pain, as well autonomic functions (e.g., sweating) (Al-Shekhlee et al., 2002, Chéliout-Héraut et al., 2005, Quattrini et al., 2004). There is evidence that diabetes affects small fibers at an earlier stage (even in subjects with impaired glucose tolerance), before any changes in large fibers (Malik et al., 2005, Smith et al., 2001, Sumner et al., 2003), although some data suggest that small- and large-fiber abnormalities develop in parallel (Yagihashi et al., 2007, Ziegler et al., 1991). The term small-fiber neuropathy (or small-fiber impairment) has been coined to describe neuropathy selectively affecting small fibers and sparing large fibers (Al-Shekhlee et al., 2002, Quattrini et al., 2004).

Available tests for large-fiber function include Achilles tendon reflexes (Feldman et al., 1994, Young et al., 1993), 10 g SWMF (Abbott et al., 2002), 128 Hz tuning fork (Feldman et al., 1994, Young et al., 1993), and VPT (Bril & Perkins, 2002). Available tests to evaluate small-fiber function include pain perception (Neurotip), qualitative temperature perception (Tiptherm rod), thermal perception threshold (TPT), and sudomotor function tests. The Neurotip is attached to the Neuropen device and assesses pinprick sensation (Paisley, Abbott, van Schie, & Boulton, 2002). The sharp and blunt edges of the Neurotip are randomly pressed against the plantar aspect of the hallux, and patients are required to distinguish between the painful and painless stimuli. The test is considered abnormal, if at least two responses out of three readings are wrong (Paisley et al., 2002). The Tiptherm rod is a pen-like device with a plastic cylinder on one end and a metal cylinder on the other end, which is applied on the dorsum of each foot (Viswanathan et al., 2002, Ziegler et al., 2005). Temperature perception is impaired if there are at least two incorrect responses out of three readings on the dorsum of each foot (Ziegler et al., 2005). The TPT is measured with an automatically heated or cooled probe on the dorsum of the foot. Six consecutive measurements for cold and six consecutive measurements for hot are performed, and the TPT is calculated as a mean value (Krämer, Rolke, Bickel, & Birklein, 2004). Sudomotor function tests include the quantitative sudomotor axon reflex test, the sweat imprint, the thermoregulatory test, and the sympathetic skin response (Low, 2004). Such tests are very cumbersome to perform because they require expensive equipment and trained personnel (Low, 2004). Small nerve fiber morphology can nowadays be more precisely studied by a minimally invasive skin biopsy assessing intraepidermal nerve fiber density (Malik et al., 2005, Quattrini et al., 2004, Smith et al., 2005).

Back to Article Outline

3. New tests for the assessment of large-fiber function 

Vileikyte, Hutchings, Hollis, and Boulton (1997) described the TCD (Tacticon Medical Enterprises, West Chester, PA). This is a simple, handheld quantitative sensory testing device that assesses protective sensation. It consists of a cylindrical aluminum disk with eight fixed solid rods (numbered from 0 to 7) of increasing circumference that protrude form the outside (Vileikyte et al., 1997). The circumference of the rods increases from 12.5 to 40 mm. The TCD is applied on the plantar aspect of the hallux, and the patient is asked to distinguish stimuli differing in diameter (Vileikyte et al., 1997). The lowest rod that the patient is able to discriminate from 0 is used as the TCD score (Vileikyte et al., 1997). The authors found that the TCD was very easy to use. There was excellent correlation between TCD and SWMF (P<.0001) or VPT (P<.0001) (Vileikyte et al., 1997). Using clinical examination as gold standard, the TCD yielded 92.3% sensitivity and 64.2% specificity for neuropathy, as well as a 100% sensitivity and a 58.3% specificity for neuropathic ulceration (Vileikyte et al., 1997). In terms of identifying patients at high risk for foot ulcer, there was a 75.2% agreement between TCD and VPT and a 78.9% agreement between TCD and SWMF (Vileikyte et al., 1997). The TCD had the highest sensitivity (100%) in detecting patients with prior ulceration, followed by the SWMF (91.9%) and the VPT (86.5%) (Vileikyte et al., 1997). Specificity was 58.3%, 76.0%, and 79.2%, respectively (Vileikyte et al., 1997). In another study, there was a significant (P<.0001) correlation between VPT and TCD, which was irrespective of age (Maser, Laudadio, Lenhard, & DeCherney, 1997). Agreement between the two modalities was modest (κ=.67 for ≤50 years; κ=.55 for >50 years) (Maser, Laudadio, Lenhard, & DeCherney, 1997).

The steel ball-bearing is also a test assessing protective sensation (Papanas, Gries, Maltezos, & Zick, 2006). It is a specially designed ball-bearing made of steel, attached to a plaster (Hansamed, 5×4 cm), using common glue (UHU Kraft). The plaster is applied in the plantar area over the second metatarsal head of each foot, while an empty control plaster is applied on the contralateral foot. Five ball-bearings of progressively increasing diameter (from 1.5 to 3.5 mm) are applied, and the patient is asked to walk barefoot on flat ground (five steps) (Papanas et al., 2006). The smallest ball-bearing that the patient is able to feel defines the ball-bearing score (range=1–6) (Papanas et al., 2006). The ball-bearing score was found to correlate significantly (P=.01) with all other tests of neuropathy (NDS, SWMF, VPT, TPT, Achilles tendon reflexes) (Papanas et al., 2006). Both inter-observer and intra-observer reproducibility were good (κ=.811 and κ=.749, respectively) (Papanas et al., 2006). Using clinical examination (NDS) as gold standard, the cutoff value of 2 for the ball-bearing score had a sensitivity of 84% and a specificity of 100% for the diagnosis of neuropathy. Using a cutoff value of 4 for the ball-bearing score, sensitivity and specificity for detection of patients with prior neuropathic ulceration were 84.6% and 86.1%, respectively (Papanas et al., 2006). Interestingly, the ball-bearing score was the most specific (100%) for neuropathy, followed by the SWMF (90.5%) and the Achilles tendon reflexes (91%) (Papanas et al., 2006). Similarly, specificity of the new test for detection of patients with prior neuropathic ulceration was high (86.1%), second only to the VPT (91%) (Papanas et al., 2006). Papanas et al. (2006) reported that the steel ball-bearing had similar sensitivity but higher specificity (both for neuropathy and for neuropathic ulceration) in comparison to results previously obtained with the TCD (Vileikyte et al., 1997). They added that the TCD appeared easier to use than the steel ball-bearing. Nonetheless, these comparisons need to be interpreted with caution since there is no head-to-head evaluation of the two new tests. An additional advantage of the ball-bearing test is its ability to illustrate the danger of unperceived foot trauma to the patient (Papanas et al., 2006). Thus, according to Papanas et al., the new test may contribute to patient education, which is important for their foot care.

Automated NCS is a further, very promising diagnostic modality (Gozani et al., 2005, Perkins et al., 2006). It combines the advantages of classical NCS and the wide applicability (Gozani et al., 2005, Perkins et al., 2006). Indeed, it is meant to be used by general health care providers, not by specialized technicians (Gozani et al., 2005, Perkins et al., 2006). It has been shown that 1 h of training is sufficient to perform patient examinations (Perkins et al., 2006). Examination is carried out using a portable device (NC-stat, NeuroMetrix, Inc., Waltham, MA) with disposable flexible panels to be applied to the limbs (Gozani et al., 2005, Perkins et al., 2006). There is a preconfigured array of electrodes allowing the automated search for the best-situated electrode that identifies the clearest signal. Examination data are sent by telemetry to centralized laboratory, which faxes the diagnostic report to the clinician within the same day (Gozani et al., 2005, Perkins et al., 2006). Perkins et al. (2006) showed that NC-stat had 92% sensitivity, 82% specificity, and 89% overall accuracy for the diagnosis of neuropathy, as compared to nerve conduction studies performed in a specialist setting. In particular, sural conduction amplitude exhibited a very strong correlation with the reference method (r=.95, P<.0001) (Perkins et al., 2006). There was also a satisfactory quantitative agreement between the two methods (Bland and Altman; mean difference=−1.2 μV) (Perkins et al., 2006). The time needed for patient examination was only 5–6 min, and the median time until receipt of faxed diagnostic report amounted to 40 min (Perkins et al., 2006). Vinik, Emley, Megerian, and Gozani (2004) employed NC-stat to study the median and ulnar nerve and found comparable results similar to those obtained with traditional NCS. The reliability of automated NCS for the study of the median and ulnar nerve has been confirmed by Kong, Gozani, Hayes, and Weinberg (2006). Of note, automated NCS has been demonstrated to be reproducible (Kong, Lesser, Megerian, & Gozani, 2006). In 15 healthy volunteers, most electrophysiological parameters had a coefficient of variance (CV) of less than 6% (Kong, Lesser, et al., 2006). Automated NCS parameters had a similar reproducibility to those obtained with traditional electrophysiologic study (Kong, Lesser, et al., 2006). Reproducibility was highest for F-wave latencies and distal motor latencies, distal sensory latencies, sensory nerve action potential amplitude, and compound motor action potential amplitude (Kong, Lesser, et al., 2006). This reproducibility now remains to be confirmed in diabetic patients as well. Automated NCS has already been used in a large cohort of 1358 diabetic subjects recruited from 28 community clinics in the United States (Vinik, Kong, Megerian, & Gozani, 2006). Among these patients, 75.1% had at least one NCS abnormality and 53.2% had bilateral abnormalities (Vinik et al., 2006). The study concluded that automated NCS may be reliably utilized for nerve conduction confirmation of peripheral neuropathy in the primary care setting (Vinik et al., 2006).

Back to Article Outline

4. New tests for the assessment of small-fiber function 

Ziegler et al. (2005) described the NeuroQuick for early quantitative assessment of small-fiber dysfunction. This is a handheld, microprocessor-operated electronic device (135×58×25 mm) weighing 126 g. It harbors a fan that may be adjusted to rotate at 1 out of 10 different velocities (levels) (Ziegler et al., 2005). Two laser diodes ensure a constant distance from skin at 23 cm during the examination. The patient is tested in the sitting position with eyes shut in a quiet room without airstreams. The test is applied on the dorsum of the feet, and the patient is asked if he feels the airflow on the skin (Ziegler et al., 2005). The NeuroQuick threshold (NQT) is defined as the level at which the airflow is felt (range=1–10; 11 if not felt at all), and the mean of three readings is used for analysis (Ziegler et al., 2005). The CVs to assess reproducibility in controls and diabetic patients were 20.4% and 8.5%, respectively (Ziegler et al., 2005). NQT differed significantly (P<.05) between controls (mean=1.25), diabetic patients without neuropathy (mean=3.09), and diabetic patients with neuropathy (mean=5.68) (Ziegler et al., 2005). NQT correlated significantly (P<.001) with all nerve function tests, and the strongest correlation was with warm TPT (r=.618) and cold TPT (r=.529) (Ziegler et al., 2005). Importantly, the NQT was the test least likely to be normal in the presence of neuropathy. Neuropathic patients showed the following: abnormal NQT with normal warm TPT in 34% (vs. 5% the reverse, P<.05), abnormal NQT with normal cold TPT in 32% (vs. 11% the reverse, P<.05), abnormal NQT with normal Tiptherm rod in 47% (vs. 2% the reverse, P<.05), and abnormal NQT with normal tuning fork in 29% (vs. 10% the reverse, P<.05) (Ziegler et al., 2005). These results suggest that the NeuroQuick is a reliable screening tool for neuropathy, especially small nerve fiber impairment (Ziegler et al., 2005). Moreover, it may be more sensitive for early detection of neuropathy than the other modalities examined (Ziegler et al., 2005). However, this assumption needs to be verified in a prospective study.

More recently, an indicator test for sudomotor function (Neuropad, Miro Verbandstoffe GmbH, Wiehl, Drabenderhöhe, Germany) has been marketed (Zick, Schäper, & Deeters, 2003). This is a patch assessing plantar sweat production by means of a chemical reaction that is manifested as a color change from blue to pink (Waizumi et al., 1990, Young, 2003, Zick et al., 2003). The indicator test contains the blue complex salt anhydrous cobalt II chloride. In the presence of water, each molecule of this salt absorbs six water molecules, and the color changes to pink (Waizumi et al., 1990, Young, 2003). The time required for complete color change is inversely related to skin humidity (Waizumi et al., 1990, Young, 2003, Zick et al., 2003). In the first report, using clinical examination as a gold standard, Neuropad had 90% sensitivity and 74% specificity for the diagnosis of peripheral neuropathy in a mixed patient series of both type 1 and type 2 diabetes (Zick et al., 2003). In a larger study recruiting type 2 diabetic patients, Papanas, Papatheodorou, Christakidis, et al. (2005) also reported 94.4% sensitivity and 69.7% specificity of Neuropad, as compared to the DNI. In line with the symmetric involvement of diabetic neuropathy, there was no difference (P=.99) in time to color change of the indicator test between the right and the left foot (Papanas, Papatheodorou, Christakidis, et al., 2005). The very high sensitivity and moderate specificity of Neuropad for the diagnosis of peripheral neuropathy have now been confirmed in a number of studies (Liatis et al., 2007, Manes et al., 2004, Papanas et al., 2007, Papanas et al., 2008, Quattrini et al., 2008, Shen et al., 2007, Spallone et al., 2009). In these studies, clinical examination (DNI or NDS or other definitions) has been the gold standard against which Neuropad has yielded sensitivity of 85–97.8% and specificity of 32–78.5% (mostly around 65%) (Liatis et al., 2007, Manes et al., 2004, Papanas et al., 2007, Papanas et al., 2008, Quattrini et al., 2008, Shen et al., 2007, Spallone et al., 2009). The invariably lower specificity than sensitivity is due to the fact that Neuropad is abnormal in about one third of patients with clinical examination negative for neuropathy (Papanas et al., 2005, Papanas et al., 2007). It has been proposed that this result may be ascribed to earlier diagnosis of neuropathy by means of Neuropad before conventional clinical signs become positive (Papanas et al., 2005, Papanas et al., 2007), but its specificity in a larger nondiabetic population has not been reported. Neuropad assesses sudomotor function, which is mediated by small fibers, and these may be impaired at a very early stage in diabetes before any large-fiber injury (Malik et al., 2005, Smith et al., 2001, Sumner et al., 2003). This notion, although plausible, must be confirmed in a prospective study. An alternative explanation might be that the 10-min threshold used to define an abnormal Neuropad result could be rather low and might be reconsidered. Recent data from the study by Spallone et al. (2009) support this suggestion. Indeed, the authors reported 85% sensitivity and 32% specificity with the 10-min time threshold and improved specificity (61%) with only slightly diminished sensitivity (80%) with the 15-min time threshold (Spallone et al., 2009). Specificity was further increased to 74% with the 18-min time threshold, but sensitivity was markedly reduced to 60% (Spallone et al., 2009). Based on these observations, a normative database defined by an exact cut-point is required to provide a definitive answer.

Of additional importance, Neuropad has shown excellent reproducibility (Papanas, Papatheodorou, Papazoglou, et al., 2005). Intra-observer reproducibility is high (P=.001), both in patients with and in those without sudomotor dysfunction. Similarly, inter-observer reproducibility is high (P=.001) (Papanas, Papatheodorou, Papazoglou, et al., 2005). Intra- and inter-observer CV ranged between 4.1% and 5.1% (Papanas, Papatheodorou, Papazoglou, et al., 2005).

Specifically, Neuropad is particularly accurate in the diagnosis of small-fiber neuropathy (Papanas, Papatheodorou, et al., 2007). Basing the diagnosis of small-fiber neuropathy on qualitative temperature perception (Tiptherm rod) and pain perception, Papanas, Papatheodorou, et al. (2007) showed 99% sensitivity and 78.3% specificity of Neuropad. Quattrini et al. (2008) demonstrated significant correlations of Neuropad, with parameters of small-fire function, notably cold detection threshold (rs=.394, P=.003) and heat-as-pain perception threshold visual analogue score 0.5 (rs=.279, P=.043). The authors also examined intraepidermal nerve fiber density (in fibers per millimeter) (Quattrini et al., 2008). There was a significant (P=.02) difference between patients with normal and those with abnormal Neuropad result. The latter showed a significant reduction in nerve fiber density, confirming the association between Neuropad and small-fiber impairment (Quattrini et al., 2008).

It has also been shown that time to complete color change of Neuropad differs significantly between healthy controls, diabetic patients without neuropathy, and those with neuropathy (Papanas et al., 2005, Shen et al., 2007). Furthermore, there is a correlation between time to color change of Neuropad and clinical severity of neuropathy. One study showed a significant difference in time to color change between patients with moderate neuropathy (defined as DNI=2.5–4.5) and those with severe neuropathy (DNI=5–8) (mean=14.2 min vs. 32.8 min, P=.003) (Papanas, Papatheodorou, Christakidis, et al., 2005). In another work, neuropathy was diagnosed and staged by clinical examination and NCS, according to the Michigan classification system (Papanas, Giassakis, et al., 2007). A highly significant (P=.001) correlation was demonstrated between time until complete color change of Neuropad and Michigan class of neuropathy (Papanas, Giassakis, et al., 2007). Time until complete color change of the test was also stratified into deciles according to the observed spread of measurements, and new cutoff values that enabled the association of Neuropad results with Michigan neuropathy class were determined (Papanas, Giassakis, et al., 2007). The threshold lower than 530 s until complete color change had 97% sensitivity and 100% specificity for Class 0 neuropathy (Papanas, Giassakis, et al., 2007). Similarly, the other thresholds yielded 93–100% sensitivity and 97–100% specificity for Class 1 to Class 3 neuropathy (Papanas, Giassakis, et al., 2007). Based on such findings, it appears that Neuropad may also prove helpful in assessing the clinical severity of neuropathy. However, use of Neuropad for this purpose is very time-demanding, and it is uncertain, at the moment, whether this promising application can be extensively used in the busy clinic.

Furthermore, Neuropad has been investigated as a potential risk marker of foot ulceration (Papanas et al., 2008). A significant correlation has already been shown between time to color change of the indicator test and VPT (rs=.889, P<.001) (Papanas et al., 2008). Given that VPT is an excellent risk marker of future ulceration, Papanas et al. (2008) considered this correlation promising and suggested that further prospective work is necessitated to confirm or refute the potential utility of Neuropad in the evaluation of the foot at risk for ulceration. These results now appear especially promising, given that sudomotor dysfunction has recently been identified as a powerful independent risk factor (odds ratio=15.3) for foot ulceration (Tentolouris et al., 2009), but whether an abnormal Neuropad response is a predictor of diabetic foot ulcers or amputations has to be determined in long-term prospective studies.

A more practical advantage of Neuropad is its beneficial effect on patient education (Papanas et al., 2005, Papanas et al., 2007, Papanas et al., 2008). Reduced color change of Neuropad has turned out an impressive finding for the patients themselves, prompting a keen interest in the diagnosis of neuropathy, in self-examination, and in suitable shoes (Papanas et al., 2005, Papanas et al., 2007, Papanas et al., 2008). Consequently, an additional advantage of the indicator test is its educational value (Papanas et al., 2005, Papanas et al., 2007, Papanas et al., 2008). The educational value is enhanced by the suitability of Neuropad to be used for self-screening at home (Tentolouris, Achtsidis, Marinou, & Katsilambros, 2008). Tentolouris et al. (2008) have shown that this application of Neuropad is feasible. Impressively, there was excellent agreement (κ=.88) between Neuropad result as read by the physician and the patient (Tentolouris et al., 2008). These results encourage the generalized use of Neuropad as a screening tool for neuropathy with additional educational benefit for the patient. The Neuropad has also started to be used as an objective measure of dry foot skin in research (Banchellini et al., 2008, Eckhard et al., 2007). Finally, it has been used in the study of diabetic autonomic neuropathy (using cardiovascular tests as gold standard), but results have not been consistent (Bilen et al., 2007, Chon et al., 2005, Liatis et al., 2007, Quattrini et al., 2008, Spallone et al., 2009), and this application appears less promising.

Back to Article Outline

5. New tests: implications for clinical practice 

It is beyond dispute that neuropathy needs to be diagnosed early to prevent further complications, such as neuropathic pain or the diabetic foot (Boulton et al., 2005, Boulton et al., 2004, Edmonds, 2004, Shaw et al., 2003, Várkonyi and Kempler, 2008, Ziegler, 2008). It is also obvious that early detection and appropriate diagnosis of diabetic neuropathy is an important research field and that several diagnostic modalities have been used. Regrettably, neuropathy may still be underdiagnosed in everyday clinical practice (Herman & Kennedy, 2005). Thus, new tests for neuropathy represent an attempt to improve the diagnostic approach to this diabetic complication.

What is the current situation with the new tests? In the authors' opinion, the ideal new test should (a) have a high sensitivity, specificity, and reproducibility; (b) be easy to use; (c) contribute to early diagnosis of neuropathy; and (d) be cost-effective. Neuropad satisfactorily fulfills the criterion of high sensitivity and reproducibility, but its specificity is lower (Liatis et al., 2007, Manes et al., 2004, Papanas et al., 2005, Papanas et al., 2007, Quattrini et al., 2008, Spallone et al., 2009, Zick et al., 2003). In terms of easy applicability, again Neuropad appears to be the most suitable for generalized use. It does not require patient cooperation (Papanas et al., 2005, Papanas et al., 2007) and has already proven suitable for screening purposes (Tentolouris et al., 2008) and patient education (Papanas et al., 2005, Papanas et al., 2007). As regards early detection of polyneuropathy, there is evidence that Neuropad (Papanas et al., 2005, Papanas et al., 2007) and NeuroQuick (Ziegler et al., 2005) hold promise, but this needs to be confirmed in prospective studies. Since NCS is an objective measure of neuropathy and enables diagnosis of the subclinical stage (Olaleye et al., 2001, Rota et al., 2005), it is reasonable that automated NCS may be helpful, but this equally remains to be shown. Regrettably, the cost-effectiveness of the new tests has, to the best of our knowledge, not yet been studied. All in all, diagnosis of neuropathy is extremely important and the new tests represent a step in the right direction. This progress notwithstanding, diagnosis of neuropathy cannot rest solely on the new tests. Instead, the clinician should rely on established modalities but utilize the new tests as diagnostic adjuncts, as dictated by the clinical situation.

Back to Article Outline

6. Conclusions 

Clearly, diagnosis of neuropathy needs to be improved. New tests that appear to facilitate diagnosis are now emerging. New tests may be classified into those mainly assessing large-fiber function (TCD, steel ball-bearing, and automated NCS) and those mainly assessing small-fiber function (NeuroQuick and Neuropad). These tests are promising, but they still need to be evaluated in prospective studies. Moreover, their cost-effectiveness needs more careful appraisal. For the time being, it is safer to base the diagnosis of neuropathy on established modalities. In the near future, more extensive use of the new tests may be anticipated.

Back to Article Outline

References 

  1. Abbott CA, Carrington AL, Ashe H, Bath S, Every LC, Griffiths J, et al. The North–West Diabetes Foot Care Study: Incidence of, and risk factors for, new diabetic foot ulceration in a community-based patient cohort. Diabetic Medicine. 2002;19:377–384
  2. Al-Shekhlee A, Chelimsky TC, Preston DC. Review: Small-fiber neuropathy. The Neurologist. 2002;8:237–253
  3. American Diabetes Association and American Academy of Neurology . Consensus statement: Report and recommendations of the San Antonio conference on diabetic neuropathy. Diabetes. 1998;37:1000–1004
  4. Banchellini E, Macchiarini S, Dini V, Rizzo L, Tedeschi A, Scatena A, et al. Use of nanotechnology-designed footsock in the management of preulcerative conditions in the diabetic foot: Results of a single, blind randomized study. The International Journal of Lower Extremity Wounds. 2008;7:82–87
  5. Bilen H, Atmaca A, Akcay G. Neuropad indicator test for diagnosis of sudomotor dysfunction in type 2 diabetes. Advances in Therapy. 2007;24:1020–1027
  6. Boulton AJM, Kirsner RS, Vileikyte L. Neuropathic diabetic foot ulcers. The New England Journal of Medicine. 2004;351:48–55
  7. Boulton AJ, Vinik AI, Arezzo JC, Bril V, Feldman EL, Freeman R, et al. Diabetic neuropathies: A statement by the American Diabetes Association. Diabetes Care. 2005;28:956–962
  8. Bril V, Perkins BA. Comparison of vibration perception thresholds obtained with the Neurothesiometer and the CASE IV and relationship to nerve studies. Diabetic Medicine. 2002;19:661–666
  9. Chéliout-Héraut F, Zrek N, Khemliche H, Varnet O, Seret-Begue D, Martinez M, et al. Exploration of small fibers for testing diabetic neuropathies. Joint Bone Spine. 2005;72:412–415
  10. Chon S, Kim YS, Oh S, Woo JT, Kim SW, Kim JW, et al. Assessment of diabetic autonomic neuropathy in type 2 diabetic subjects using Neuropad: A new indicator plaster for detection of disturbed sweat secretion. Diabetologia. 2005;48(Suppl 1):A 361
  11. Edmonds M. The diabetic foot, 2003. Diabetes Metabolism Research and Reviews. 2004;20(Suppl 1):S9–S12
  12. Eckhard M, Lengler A, Liersch J, Bretzel RG, Mayser P. Fungal foot infections in patients with diabetes mellitus—Results of two independent investigations. Mycoses. 2007;50(Suppl 2):14–19
  13. Feldman EL, Stevens MJ, Thomas PK, Brown MB, Canal N, Greene DA. A practical two-step quantitative clinical and neurophysiological assessment for the diagnosis and staging of diabetic neuropathy. Diabetes Care. 1994;17:1281–1289
  14. Gozani SN, Fisher MA, Kong X, Megerian JT, Rutkove SB. Electrodiagnostic automation: Principles and practice. Physical Medicine and Rehabilitation Clinics of North America. 2005;16:1015–1032
  15. Herman WH, Kennedy L. Underdiagnosis of peripheral neuropathy in type 2 diabetes. Diabetes Care. 2005;28:1480–1481
  16. Kong X, Gozani SN, Hayes MT, Weinberg DH. NC-stat sensory nerve conduction studies in the median and ulnar nerves of symptomatic patients. Clinical Neurophysiology. 2006;117:405–413
  17. Kong X, Lesser EA, Megerian JT, Gozani SN. Repeatability of nerve conduction measurements using automation. Journal of Clinical Monitoring and Computing. 2006;20:405–410
  18. Krämer HH, Rolke R, Bickel A, Birklein F. Thermal thresholds predict painfulness of diabetic neuropathies. Diabetes Care. 2004;27:2386–2391
  19. Liatis S, Marinou K, Tentolouris N, Pagoni S, Katsilambros N. Usefulness of a new indicator test for the diagnosis of peripheral and autonomic neuropathy in patients with diabetes mellitus. Diabetic Medicine. 2007;24:1375–1380
  20. Low PA. Evaluation of sudomotor function. Clinical Neurophysiology. 2004;7:1506–1513
  21. Malik RA, Tesfaye S, Newrick PG, Walker D, Rajbhandari SM, Siddique I, et al. Sural nerve pathology in diabetic patients with minimal but progressive neuropathy. Diabetologia. 2005;48:578–585
  22. Manes C, Mikoudi K, Sossidou E, Pigas G, Karagianni D, Skoutas D, et al. Evaluation of a new indicator plaster in identifying diabetic patients at risk of foot ulceration. Diabetologia. 2004;47(Suppl 1):A 376
  23. Maser RE, Laudadio C, Lenhard MJ, DeCherney GS. A cross-sectional study comparing two quantitative sensory testing devices in individuals with diabetes. Diabetes Care. 1997;20:179–181
  24. Olaleye D, Perkins BA, Bril V. Evaluation of three screening tests and a risk assessment model for diagnosing peripheral neuropathy in the diabetes clinic. Diabetes Research and Clinical Practice. 2001;54:115–128
  25. Paisley AN, Abbott CA, van Schie CHM, Boulton AJM. A comparison of the Neuropen against standard quantitative sensory-threshold measures for assessing peripheral nerve function. Diabetic Medicine. 2002;19:400–405
  26. Papanas N, Papatheodorou K, Christakidis D, Papazoglou D, Giassakis G, Piperidou H, et al. Evaluation of a new indicator test for sudomotor function (Neuropad®) in the diagnosis of peripheral neuropathy in type 2 diabetic patients. Experimental and Clinical Endocrinology and Diabetes. 2005;113:195–198
  27. Papanas N, Papatheodorou K, Papazoglou D, Christakidis D, Monastiriotis C, Maltezos E. Reproducibility of a new indicator test for sudomotor function (Neuropad®) in patients with type 2 diabetes mellitus. Experimental and Clinical Endocrinology and Diabetes. 2005;113:577–581
  28. Papanas N, Gries A, Maltezos E, Zick R. The steel ball-bearing test: A new test for evaluating protective sensation in the diabetic foot. Diabetologia. 2006;49:739–743
  29. Papanas N, Giassakis G, Papatheodorou K, Papazoglou D, Monastiriotis C, Christakidis D, et al. Use of the new indicator test (Neuropad) for the assessment of the staged severity of neuropathy in type 2 diabetic patients. Experimental and Clinical Endocrinology and Diabetes. 2007;115:58–61
  30. Papanas, N., Papatheodorou, K., Papazoglou, D., Christakidis, D., Monastiriotis, C., & Maltezos, E. (2007). The new indicator test (Neuropad): A valuable diagnostic tool for small-fiber impairment in patients with type 2 diabetes. The Diabetes Educator, 33, 257–258, 260, 262.
  31. Papanas N, Papatheodorou K, Papazoglou D, Monastiriotis C, Christakidis D, Maltezos E. A comparison of the new indicator test for sudomotor function (Neuropad) with the vibration perception threshold and the clinical examination in the diagnosis of peripheral neuropathy in subjects with type 2 diabetes. Experimental and Clinical Endocrinology and Diabetes. 2008;116:135–138
  32. Perkins BA, Grewal J, Ng E, Ngo M, Bril V. Validation of a novel point-of-care nerve conduction device for the detection of diabetic sensorimotor polyneuropathy. Diabetes Care. 2006;29:2023–2027
  33. Quattrini C, Jeziorska M, Tavakoli M, Begum P, Boulton AJ, Malik RA. The Neuropad test: A visual indicator test for human diabetic neuropathy. Diabetologia. 2008;51:1046–1050
  34. Quattrini C, Jeziorska M, Malik RA. Small fiber neuropathy in diabetes: Clinical consequence and assessment. The International Journal of Lower Extremity Wounds. 2004;3:16–21
  35. Rota E, Quadri R, Fanti E, Isoardo G, Poglio F, Tavella A, et al. Neurophysiological findings of peripheral neuropathy in newly diagnosed type II diabetes mellitus. Journal of the Peripheral Nervous System. 2005;10:348–353
  36. Shaw JE, Zimmet PZ, Gries FA, Ziegler D. Epidemiology of diabetic neuropathy. In:  Gries FA,  Cameron NE,  Low PA,  Ziegler D editor. Textbook of diabetic neuropathy. Stuttgart: Thieme; 2003;p. 64–82
  37. Shen J, Cao Y, Han YJ, Luo XR, Xie CH, Li JM, et al. Sweat function evaluation for early diagnosis of diabetic peripheral neuropathy. Nan Fang Yi Ke Da Xue Xue Bao. 2007;27:1210–1212
  38. Smith AG, Ramachandran P, Tripp S, Singleton JR. Epidermal nerve innervation in impaired glucose tolerance and diabetes-associated neuropathy. Neurology. 2001;57:1701–1704
  39. Smith AG, Howard JR, Kroll R, Ramachandran P, Hauer P, Singleton JR, et al. The reliability of skin biopsy with measurement of intraepidermal nerve fiber density. Journal of the Neurological Sciences. 2005;228:65–69
  40. Spallone V, Morganti R, Siampli M, Fedele T, D'Amato C, Cacciotti L, et al. Neuropad as a diagnostic tool for diabetic autonomic and sensorimotor neuropathy. Diabetic Medicine. 2009;26:686–692
  41. Sumner CJ, Sheth S, Griffin JW, Cornblath DR, Polydefkis M. The spectrum of neuropathy in diabetes and impaired glucose tolerance. Neurology. 2003;60:108–111
  42. Tentolouris N, Achtsidis V, Marinou K, Katsilambros N. Evaluation of the self-administered indicator plaster Neuropad for the diagnosis of neuropathy in diabetes. Diabetes Care. 2008;31:236–237
  43. Tentolouris N, Marinou K, Kokotis P, Karanti A, Diakoumopoulou E, Katsilambros N. Sudomotor dysfunction is associated with foot ulceration in diabetes. Diabetic Medicine. 2009;26:302–305
  44. Várkonyi T, Kempler P. Diabetic neuropathy: New strategies for treatment. Diabetes Obesity and Metabolism. 2008;10:99–108
  45. Vileikyte L, Hutchings G, Hollis S, Boulton AJM. The tactile circumferential discriminator: A new, simple screening device to identify diabetic patients at risk of foot ulceration. Diabetes Care. 1997;20:623–626
  46. Vinik AI, Emley MS, Megerian JT, Gozani SN. Median and ulnar nerve conduction measurements in patients with symptoms of diabetic peripheral neuropathy using the NC-stat system. Diabetes Technology and Therapeutics. 2004;6:816–824
  47. Vinik AI, Kong X, Megerian JT, Gozani SN. Diabetic nerve conduction abnormalities in the primary care setting. Diabetes Technology and Therapeutics. 2006;8:654–662
  48. Viswanathan V, Snehalatha C, Seena R, Ramachandran A. Early recognition of diabetic neuropathy: Evaluation of a simple outpatient procedure using thermal perception. Postgraduate Medical Journal. 2002;78:541–542
  49. Waizumi K, Masuda H, Ohtaki H, Tsukamoto K, Sunagawa I. In situ observations of the phase transition among Cobalt (II) dichloride hydrates and crystal structures of the tetra- and hexahydrates. Bulletin of the Chemical Society of Japan. 1990;63:3426–3433
  50. Yagihashi S, Yamagishi S, Wada R. Pathology and pathogenetic mechanisms of diabetic neuropathy: Correlation with clinical signs and symptoms. Diabetes Research and Clinical Practice. 2007;77(Suppl 1):S184–S189
  51. Young JA. Cobalt (II) chloride hexahydrate. Journal of Chemical Education. 2003;80:610
  52. Young MJ, Boulton AJM, MacLeod AF, Williams DRR, Sonksen PH. A multicentre study of the prevalence of diabetic peripheral neuropathy in the United Kingdom hospital clinic population. Diabetologia. 1993;36:150–154
  53. Zick R, Schäper T, Deeters U. Early detection of peripheral neuropathy. Measurement of perspiration in the diabetic foot. Klinikarzt. 2003;32:192–194
  54. Ziegler D. Painful diabetic neuropathy: Treatment and future aspects. Diabetes Metabolism Research and Reviews. 2008;24(Suppl 1):S52–S57
  55. Ziegler D, Mayer P, Mühlen H, Gries FA. The natural history of somatosensory and autonomic nerve dysfunction in relation to glycaemic control during the first 5 years after diagnosis of type 1 (insulin-dependent) diabetes mellitus. Diabetologia. 1991;34:822–829
  56. Ziegler D, Siekierka-Kleiser E, Meyer B, Schweers M. Validation of a novel screening device (NeuroQuick) for quantitative assessment of small fiber dysfunction as an early feature of diabetic polyneuropathy. Diabetes Care. 2005;28:1169–1174

☆ Conflicts of interest: Both authors have received honoraria for participation in advisory board meetings of Miro, the manufacturer of Neuropad. D.Z. was also a co-owner of a patent on the NeuroQuick.

PII: S1056-8727(09)00097-X

doi:10.1016/j.jdiacomp.2009.09.006

Journal of Diabetes and Its Complications
Volume 25, Issue 1 , Pages 44-51, January 2011

Article Tools