Diabet. Med. 29, 1553–1555 (2012)
Aims Current National Institute for Health and Clinical Excellence guidelines state that patients with diabetes should have annual examination of their feet to exclude signs of sensory impairment. The VibraTip is a new disposable device producing a vibratory stimulus, which has been developed in order to screen for peripheral sensory neuropathy in diabetes. This study was designed to evaluate the device by assessing intra-rater reliability and comparing the ability of the VibraTip to detect or exclude peripheral sensory neuropathy with other bedside methods.
Methods One hundred and forty-one patients with diabetes (Type 1 or Type 2) were examined for diabetic peripheral sensory neuropathy using a Neurothesiometer, 10-g monofilament, a 128-Hz tuning fork, a Neurotip™ and a VibraTip. The failure to perceive the Neurosthesiometer stimulus at ≥ 25 V in either foot was considered the reference method for the presence of peripheral sensory neuropathy. Receiver operating characteristic curves were produced for each device and the sensitivity, specificity, predictive values and likelihood ratios for the diagnosis of peripheral sensory neuropathy were calculated. Repeat testing with the VibraTip was performed in 18 patients and intra-rater reliability assessed using Cronbach alpha.
Results Analysis of the area under the receiver operating characteristic curves showed that the 10-g monofilament was significantly better than the 128-Hz tuning fork (P = 0.0056) and the Neurotip (P = 0.0022), but was no different from the VibraTip (P = 0.3214). The alpha coefficient for the VibraTip was calculated to be 0.882, indicating good reliability.
Conclusions The VibraTip is a device comparable with the 10-g monofilament and therefore could be considered a useful tool for screening for peripheral sensory neuropathy in diabetes.
Long-term complications of diabetes, such as foot disease, are costly and consume approximately 5% of National Health Care expenditure . Foot disease results largely from the high prevalence of diabetic peripheral sensory neuropathy and its predisposing the patient to foot ulceration . Current National Institute for Health and Clinical Excellence (NICE) guidelines state that patients with diabetes should have annual examination of their feet to include testing of foot sensation using a 10-g monofilament or vibration in order to classify their risk of foot ulceration and implement appropriate preventative measures . Failure to detect ≥ 25 V using a Neurothesiometer (Scientific Laboratory supplies Ltd, Nottingham, UK) is considered to be the ‘gold standard’ of the bedside tests for the diagnosis of peripheral sensory neuropathy, but is limited in use because of the size of the equipment and cost. The VibraTip™ (McCallan Medical, Nottingham, UK) is a new disposable device that has been developed in order to screen for peripheral sensory neuropathy. Supposedly producing a vibratory stimulus more consistent than that of a tuning fork, the VibraTip probe is applied to the patient’s feet once whilst non-vibrating and once whilst vibrating and the patient (with their eyes closed) is asked to indicate when they feel the vibration. To date, only one study has been performed to validate the use of the device, and this reported complete agreement between the perception of vibration from the VibraTip and tuning fork and good agreement between perception of a 10-g monofilament and the VibraTip stimulus . Comparisons with the Neurothesiometer were not included in this study. We have therefore undertaken further analysis of this new device, including an assessment of intra-rater reliability and a comparison of the ability of the VibraTip to identify peripheral sensory neuropathy in comparison with other bed-side methods.
Patients and methods
Patients with Type 1 or Type 2 diabetes were recruited from one secondary care-based specialist diabetes service. All patients gave written informed consent and the study had full ethical approval. Patients with major amputations or minor amputations affecting the proposed sites of testing were excluded.
The VibraTip was applied for approximately 1 s with and without the button depressed at five sites on the foot (1st, 3rd and 5th metatarsal heads on the plantar surface, the hallux pulp, the dorsal surface of the hallux proximal to the nail fold). The procedure was then repeated 2–3 weeks later by the same researcher (NB).
The ability of the patient to detect the 10-g monofilament, a Neurotip™ and vibration using the VibraTip was tested at the same five sites as above on both feet. Detection of ≥ 25 V using the Neurosthesiometer was assessed at the pulp of the hallux and the detection of vibration of the 128-Hz tuning fork at the hallux and medial malleolus. The tests were applied in a random order.
All calculations were performed using Microsoft Office Excel 2007, Predictive Analytics Software (PASW) Statistics 18 (SPSS Inc., Chicago, IL, USA) and NCSS 2007 (NCSS, Kaysville, UT, USA). Intra-rater reliability was assessed using Cronbach’s alpha calculation to give an alpha coefficient. The sensitivity and specificity of each test to detect diabetic peripheral neuropathy was compared with the Neurosthesiometer as criterion standard, and receiver operating characteristic curves constructed.
Eighteen patients (14 men and four women, 72.2% with active or previous ulceration) were studied. The alpha coefficient for the VibraTip was 0.88, indicating good reliability .
One hundred and forty-one [76 men and 65 women, mean age 56.9 (sd ± 14.7) years] patients were studied. The majority (89%) had no history of foot ulcers. The prevalence of neuropathy within the test population was 41%. Receiver operating characteristic curves were produced for each device to establish the number of insensate sites that gave the best sensitivity and specificity for the diagnosis of diabetic peripheral neuropathy using the Neurosthesiometer as the gold standard. The receiver operating characteristic curves (Fig. 1) indicated that ≥ 2 insensate sites from 10 tested across two feet was the best predictor of a diagnosis of peripheral sensory neuropathy for the 10-g monofilament, Neurotip and VibraTip, whereas ≥ 1 insensate site was predictive for the 128-Hz tuning fork.
The sensitivities, specificities, the positive and negative predictive values and the positive and negative likelihood ratios for the prediction of peripheral sensory neuropathy compared with the Neurosthesiometer are shown in Table 1. Analysis of the area under the receiver operating characteristic curve (Fig. 1) showed that the 10-g monofilament was significantly better than the 128-Hz tuning fork (P = 0.0056) and the Neurotip (P = 0.0022), but was no different from the VibraTip (P = 0.3214).
|Device||Sensitivity||Specificity||Positive predictive value||Negative predictive value||Positive likelihood ratio||Negative likelihood ratio|
These results show that, when applied to the 1st, 3rd and 5th metatarsal heads, the hallux pulp and the dorsum of the hallux proximal to the nail fold on each foot, the VibraTip has good intra-rater reliability and could be useful in the diagnosis of peripheral sensory neuropathy in patients with diabetes in the community setting. Although instruments to test vibration are generally only used at one site on the foot, we studied all the sites commonly used for screening for peripheral sensory neuropathy in diabetes in the evaluation and, by constructing receiver operating characteristic curves, we were able to establish the best performance of each instrument compared with the Neurosthesiometer. A total of two or more insensate sites across the two feet should therefore be considered an abnormal result, indicative of peripheral sensory neuropathy. Whilst long-term studies of the use of the VibraTip for the prediction of new foot ulceration are still required, the area under the receiver operating characteristic curve for the VibraTip was not significantly different from the 10-g monofilament in this study, indicating that the two methods are likely to be equally effective clinical practice.
The VibraTip devices were provided free of charge by the manufacturers of the device (McCallan Medical). No funding was received for this study. The authors declare they have no other competing interests.
We acknowledge the help of Mr Michael Seagrave, University of Nottingham with the statistical analysis.