The reliability of assessment of vibration sense

OBJECTIVES: To assess the reliability of quantitative assessment of vibration sense with a Vibrameter type III. MATERIAL AND METHODS: We examined 111 healthy subjects (21-69 years). For intraobserver reliability, short-term (15 min between measurements) (n=11) and 24-h (n=28) reliability was tested. For interobserver reliability, a second tester performed the second measurement 15 min after the initial test (n=39). We also assessed the independent impact of effects of age, gender and height on vibration thresholds. RESULTS: In our study the intraobserver reliability is good [intraclass correlation coefficients (ICC) ranging from 0.55 to 0.99], whereas the interobserver reliability is moderate (ICC ranging from 0.32 to 0.88). Multiple linear regression analysis showed that age and--to a lesser extent--height was independently associated with the threshold values of the feet, but not with the thresholds of the hands. CONCLUSION: The use of a Vibrameter for measuring vibration thresholds in clinical practice and in multicentre studies is restricted because of the moderate interobserver reliability.