Read The value of the WEST monofilaments in detecting neurosensory deficit caused by hand-arm vibration exposure text version

Health and Safety Executive

The value of the WEST monofilaments in detecting neurosensory deficit caused by hand-arm vibration exposure

Prepared by Health and Safety Laboratory for the Health and Safety Executive 2009

RR712 Research Report

Health and Safety Executive

The value of the WEST monofilaments in detecting neurosensory deficit caused by hand-arm vibration exposure

Kerry Poole & Howard Mason Health and Safety Laboratory Harpur Hill Buxton Derbyshire SK17 9JN

Hand-arm vibration syndrome is categorised according to the Stockholm Workshop scales. This comprises of a rating system for vascular (circulatory system) and sensorineural (touch and sensation) symptoms, which sufferers of this disease can experience. To establish the extent of the sensory component of this disease these scales require the Occupational Health Physician to decide whether they feel an individual has reduced sensory perception. Quantitative tests such as vibrotactile (VPT) and thermal perception threshold (TPT) measurements have been used widely for this, but are generally only available in specialist referral centres. Simple techniques such as monofilaments are cheaper to use and could potentially be used more widely than the specialist quantitative tests. However, it is unclear at present what method of application should be used and how diagnostically useful these are. This study investigates the value of monofilaments in defining neurosensory abnormality caused by excessive exposure to hand-arm vibration by establishing their ability to detect abnormality determined by the standard quantitative tests. This report and the work it describes were funded by the Health and Safety Executive (HSE). Its contents, including any opinions and/or conclusions expressed, are those of the authors alone and do not necessarily reflect HSE policy.

HSE Books

© Crown copyright 2009 First published 2009 All rights reserved. No part of this publication may be

reproduced, stored in a retrieval system, or transmitted

in any form or by any means (electronic, mechanical,

photocopying, recording or otherwise) without the prior

written permission of the copyright owner.

Applications for reproduction should be made in writing to:

Licensing Division, Her Majesty's Stationery Office,

St Clements House, 2-16 Colegate, Norwich NR3 1BQ

or by e-mail to [email protected]

ii

CONTENTS

1 INTRODUCTION......................................................................................... 1

2 METHODS .................................................................................................. 3

2.1 Measurement of vibration perception threshold (VPT) ............................ 3

2.2 Measurement of thermal perception threshold (TPT) .............................. 3

2.3 Measurement of tactile perception using west monofilaments................. 3

2.4 Statistical analysis ................................................................................... 4

3 4 5 RESULTS ................................................................................................... 6

DISCUSSION............................................................................................ 11

REFERENCES.......................................................................................... 14

iii

iv

EXECUTIVE SUMMARY

Hand-arm vibration syndrome is categorised according to the Stockholm Workshop scales. This comprises of a rating system for vascular (circulatory system) and sensorineural (touch and sensation) symptoms, which sufferers of this disease can experience. To establish the extent of the sensory component of this disease these scales require the Occupational Health Physician to decide whether they feel an individual has reduced sensory perception. Quantitative tests such as vibrotactile (VPT) and thermal perception threshold (TPT) measurements have been used widely for this, but are generally only available in specialist referral centres. Simple techniques such as monofilaments are cheaper to use and could potentially be used more widely than the specialist quantitative tests. However, it is unclear at present what method of application should be used and how diagnostically useful these are. This study investigates the value of monofilaments in defining neurosensory abnormality caused by excessive exposure to hand-arm vibration by establishing their ability to detect abnormality determined by the standard quantitative tests. Three widely used testing strategies using the WEST five monofilament screening kit (Random Monofilament test (RMT), Rapid Threshold Test (RTT), Force Choice Test (FCT)) were compared with more complex and sophisticated standard quantitative tests (VPT and TPT). All tests were performed on the pulps of index and little fingers of 115 subjects referred to HSL for a hand-arm vibration syndrome (HAVS) assessment. A significant element of the cohort had sensory HAVS; the proportion of individuals (averaged across both hands) falling in to each of the Stockholm Workshop neurosensory staging categories were 36%, 26%, 36% and 2% for stages SN0 to SN3 respectively. Objectives 1. To establish the relationship between the measures of sensory perception using monofilaments and those measured using standard quantitative tests (VPT and TPT). 2. Using the published normal values for VPT and TPT define `reduced sensory perception' and establish the best cut-off for defining reduced sensation with monofilaments. 3. Establish the sensitivity and specificity for monofilaments in detecting reduced sensory perception.

Main Findings Correlations between each of the tests showed that the best correlation was between VPT measurements at 125Hz and 31.5Hz (rs=0.713). There was a significant correlation between the hot and cold thresholds of the TPT but it was weaker (rs=-

v

0.542). There was some correlation between the monofilament measurements and the standard quantitative tests but this was weak. Reduced sensory perception was determined using published normal values for the standard quantitative tests. The sensitivity and specificity of each monofilament to detect this abnormality for each of the three testing strategies was determined. Overall, there did not appear to be an optimum monofilament that gave both a good sensitivity and specificity for detecting abnormality for a single VPT or TPT test. The extent of abnormality was also investigated in two ways: (1) By determing the number of standard quantitative tests that were abnormal (maximum of four) and (2) By establishing a total z-score for the quantitative tests to establish a range of abnormality. The data suggest that the monofilaments are better at detecting significant abnormality. For example, the best sensitivity and specificity of 78 and 74% was found using a cut-off of 0.2g monofilament (RMT method) to discriminate between those with all four standard quantitative tests abnormal, and those where all tests were normal. The data would suggest that the best cut-off using this set of monofilaments would be around 0.2g. However, there was often an abrupt change in the diagnostic power between 0.2 and 2g, which may suggest that other monofilaments within this range may be more diagnostically useful. The method of application of the monofilaments did appear to affect the sensitivity and specificity of the procedure, with the RMT method appearing slightly better. Recommendations These data suggest that it may be possible to improve the diagnostic value of monofilaments in HAVS assessments to identify neurosensory deficit by either using a monofilament set with a narrower range of forces compared to the current WEST set, or in combination with another screening test (e.g. a question or purdue pegboard). Therefore, further work in this area to establish if it is possible to improve the sensitivity and specificity of this technique is recommended.

vi

1

INTRODUCTION

Hand arm vibration syndrome (HAVS) is recognised as a syndrome with vascular, neurosensory and musculoskeletal components [1-3]. Symptoms include finger blanching, paraesthesia and pain in the hands, loss of manual dexterity and weakness in hand muscles. The diagnosis of HAVS is largely symptom-led, based on anamnestic reporting, taking into account potential exclusory diagnoses. Health surveillance programmes in the UK recommend the use of questionnaires within a medical interview [3]. However, this assessment is often accompanied by a number of physical examinations or clinical manoeuvres to help diagnose HAVS, to eliminate confounding conditions, or to detect non-HAVS conditions associated with vibrating tool use. Health surveillance, which involves making `fitness-to-work' decisions, not only involves the simple diagnosis of HAVS but also the staging of the severity of the syndrome in workers. The Stockholm Workshop Scale is used to classify the severity of staging and has been modified by Lawson and McGeoch (2003) [4] and incorporated in the most recent HSE guidance [3]. A number of authors have suggested that quantitative physiological tests can supplement the medical interview in terms of aiding the reliability of the diagnosis and staging on the neurosensory Stockholm Workshop Scale [5-9]. These tests include twopoint and depth sense discrimination, aesthesiometry, Semmes-Weinstein monofilaments (SWM), sensory perception thresholds to vibration, thermal and electrical current stimuli, hand grip strength and Purdue/nine-hole pegboard tests [1]. These tests vary in complexity in terms of equipment, facilities necessary and test procedure in order to improve the objectivity and precision of the measurement. Such tests for HAVS are usually applied to finger pulps. The tests range from simple screening tests such as SWM monofilaments to more complex tests such as vibrotactile and thermal perception testing (VPT and TPT respectively). While VPT measurements at 31.5Hz and 125 Hz testing frequencies and TPT measurements of thresholds in hot and cold receptors investigate problems with different nerve fibre types and associated receptors in the skin, currently the consensus it that all the tests are necessary, if applied in a HAVS cases, without any proven redundancy [10]. However, these standard quantitative tests are generally only available in specialist referral centres, which can be costly to the occupational health provider. SWM are cheap and easy to use and have been used widely as a screening test in peripheral neurological investigation, such as diabetes [11-20] and recently they have been used more in HAVS health surveillance. However, a recent investigation of the monofilaments in the form of commercially available five monofilament Weinstein Enhanced Sensory TestsTM (WEST) to discriminate between specific Stockholm workshop stages in HAVS cases has shown a relatively weak ability to detect the transitions between Stockholm Workshop stages [21]. However, as staging is a consequence of reported symptoms, clinical examination, quantitative testing and clinical judgement, one may not necessarily expect that a single test would be related to the staging classification. A single 10g monofilament has been applied in diabetic neuropathy with a sensitivity of 90.7% and specificity of 63.8% for abnormality in VPT [19]. In the current work we wished to investigate the usefulness of monofilaments to

1

detect reduced sensory perception as defined by standard quantitative tests (VPT and TPT) in a HAVS referral population. The specific objectives of this work were: 1. To establish the relationship between the measures of sensory perception using monofilaments and those measured using standard quantitative tests (VPT and TPT). 2. Using the published normal values for VPT and TPT define `reduced sensory perception' and establish the best cut-off for defining reduced sensation with monofilaments. 3. Establish the sensitivity and specificity for monofilaments in detecting reduced sensory perception.

2

2

METHODS

The Health and Safety Laboratory (HSL) undertakes HAVS health surveillance assessments for workers exposed to vibration from hand-held tools and referred by their company's occupational health physician/nurse or by their general practitioner. Each patient attending the centre from May 2002 to December 2005 was asked to participate in the study. Ethical approval for this study had been obtained from the HSE Ethical Committee. Vibrotactile and thermal perception tests and SWM measurements were performed within the routine health surveillance assessments, which were largely based on those defined in the Department of Trade and Industry (DTI) miners compensation scheme [4]. The assessment used a standardised physician-led questionnaire, clinical examination and quantitative tests.

2.1 MEASUREMENT OF VIBRATION PERCEPTION THRESHOLD (VPT)

The individual was asked to place the fleshy part of their fingertip over the centre of the vibrating probe at a constant push force of 1N (HVLab, ISVR, University of Southampton). The vibration amplitude is then increased slowly and the subject instructed to depress the response button as soon as they are able to feel the vibration. The vibration amplitude then decreases and the subject is instructed to release the button when they can no longer feel the vibration. This was repeated for 45 seconds. The measurements were performed on the index and little finger of both hands at a frequency of 125Hz. These measurements were repeated at a frequency of 31.5Hz.

2.2

MEASUREMENT OF THERMAL PERCEPTION THRESHOLD (TPT)

The subjects were asked to place the fleshy part of their fingertip over the centre of a metal plate (HVLab, ISVR, University of Southampton). This plate then heats up slowly and the subject was instructed to press a response button as soon as they felt a change in temperature. This was performed six times and the mean warm temperature threshold calculated. This procedure was then repeated but with the plate cooling down so that the cold temperature threshold was measured. The measurements were performed on the index and little finger of both hands.

2.3

MEASUREMENT OF TACTILE PERCEPTION USING WEST MONOFILAMENTS

Tactile sensation levels were measured using the Weinstein Enhanced Sensory TestsTM monofilaments test set which included five standard filaments which apply a standard calibrated force of 0.07, 0.2, 2.0, 4.0 and 200g to the contact surface. These monofilaments were applied perpendicularly to the tip of the index and little finger, and

3

the pressure was increased until the monofilament began to bend which applies a standard force. Subjects under any SWM test were unsighted to the application of the filaments and ask to reply in terms of feeling the stimulus or not, or which of two monofilaments was `heaviest', depending on the testing strategy. Monofilaments were maintained according to the manufacturers recommendations. Three different testing strategies were employed in this study, namely a Random Monofilament Test (RMT), the Rapid Threshold Test (RTT) and a Tactile Test for carpal tunnel syndrome, which is a essentially a force choice test (FCT) between two monofilaments. The latter two testing strategies are detailed with the Weinstein Enhanced Sensory TestsTM monofilaments test set. For the RMT test format each filament was applied in a random order to the test sites and the lowest detectable stimulus was recorded in terms of force for that site. Therefore a subject is categorised as having a tactile sensitivity at a hand-site corresponding to the lowest of the five monofilament forces detected by the individual (five categorical thresholds). The more complex RTT testing strategy allowed for the determination of tactile perception thresholds in-between the individual monofilament forces (9 categorical thresholds 0.07, 0.135, 0.2, 1.1, 2.0, 3.0, 4.0, 102, >200 g). Essentially the method is a modified descending test starting from a readily detectable monofilament and estimates the 50% detection level at near 0% false positives. The FCT involves the subject attempting to correctly identify which of two suprathreshold monofilaments (forces 2.0 & 4.0 g) is the `heaviest' when applied as twenty successive trials of the paired monofilaments in a random but structured pattern. The FCT score is based on the number of correct identifications (e.g. maximum is 20; subjects devoid of tactile sensation should give a mean of 10 (std. dev. around 2.2); scores <10 may indicate attempts to suborn the test or simply guess results). This particular testing strategy is suggested to make guessing or non-compliance transparent.

2.4 STATISTICAL ANALYSIS

All data for the quantitative tests were inspected to establish whether the data were normally distributed. Where appropriate the data were transformed (log or squared) to normalize the data and allow the use of parametric statistical methods. Correlations between the different tests were performed using Spearman's rank correlation. Each of the individual measurements were categorized as normal or abnormal (normality score 0 or 1 respectively) using a cut-off based on the mean plus two standard deviations for VPT and TPT measurements on the fingertip as described by Seah [22]. The same cut-offs were used for all individuals regardless of age as the work of Seah has reported that there is no significant effect of age upon VPT or TPT measurements over the range of 20-65 years. Sensitivity and specificity for detecting abnormality (VPT or TPT) using each of the three monofilament methods were calculated based on these categorizations.

4

Two other categorisations of each measurement site in terms of better or poorer neurosensory function were made. Firstly, the addition of the normality scores for VPT at 31.5 Hz and 125 Hz and TPT cold and hot thresholds, where 0 and 1 reflect normal and abnormal according to Seah [22]. Thus four categories were defined; where a total score of zero reflects all four VPT and TPT perception measurements being normal and a score of four reflects all VPT and TPT measurements being abnormal. The last categorisation did not rely upon a definition of abnormality for the vibrotactile and thermal perception tests, but after applying z-scoring to each of the VPT and TPT parameters, formed a total z-score for each measurement site and then categorized into four equal groups based on quartiles of the total z-score. In this case group 1 were those which reflected the test site with most normal results across the measurements, while group 4 were those testing sites exhibiting the highest levels of sensory abnormality defined by VPT and TPT.

5

3

RESULTS

A total of 115 referrals took part in this study with a mean (standard deviation) age of 44.9 (9.0) years and an age range between 20-64 years. Of these 71/115 and 77/115 were diagnosed with neurosensory HAVS in their dominant and non-dominant hands respectively. The proportion of individuals (averaged across both hands) falling in to each of the Stockholm Workshop neurosensory staging categories were 36%, 26%, 36% and 2% for stages SN0 to SN3 respectively. Each individual had measurements on up to four sites (index and little fingers of both hands) and thus these 115 referrals gave a total of 458 individual measurements for analysis. Table 1 shows the correlation between the various monofilament testing strategies (RTT, RMT, FCT), VPT (125Hz, 31.5 Hz) and TPT (hot and cold thresholds). Correlations between the two testing frequencies for VPT were significantly better than between the hot and cold perception thresholds (p<0.0001). Table 1. Spearman rank correlations between the 7 measurements of sensory perception.

RMT RTT Corr. Coeff P N Corr. Coeff. P N Corr. Coeff. P N Corr. Coeff. P N Corr. Coeff. P N Corr. Coeff. P N RTT FCT -

FCT

TPT_cold

TPT hot

VPT 125Hz VPT 31.5Hz

0.462 <0.0001 458 -0.119 0.011 458 -0.296 <0.0001 458 0.318 <0.0001 458 0.297 <0.0001 458 0.206 <0.0001 458

TPT cold -

TPT hot -

VPT 125Hz -

-0.277 <0.0001 458 -0.252 <0.0001 458 0.276 <0.0001 458 0.402 <0.0001 458 0.424 <0.0001 458

-

-

-

-

0.149 0.001 458 -0.228 <0.0001 458 -0.243 <0.0001 458 -0.216 <0.0001 458

-

-

-

-0.542 <0.0001 458 -0.387 <0.0001 458 -0.358 <0.0001 458

-

-

0.335 <0.0001 458 0.423 <0.0001 458

-

0.713 <0.0001 458

Table 2 shows the calculated sensitivities and specificities for each monofilament force or correct number of choices (FCT strategy) using the individual VPT and TPT measurements categorized into normal or abnormal using the published data of Seah [22]. The definition of not being able to perceive the 0.2g or lower monofilament as being abnormal, within the RMT testing strategy, appears to show the best discrimination using normal and abnormal as defined across the VPT and TPT tests. However the sensitivity and specificity are both less than 70% suggesting relatively poor discriminate power. The sensitivity drops off significantly if defining the next

6

higher force monofilament as normal. Generally the sensitivities and specificities using the FCT strategy do not suggest a very discriminate test strategy. The more complex RTT test suggests that testing forces between 0.2g and 1.1g may show more discrimation. Table 2. Tabulation of calculated sensitivity (sens) and specificity (spec) according to abnormality defined by the vibrotactile and thermal perception tests for the three monofilament testing strategies. Sensitivity and specificity are calculated either for each monofilament force defined as the lowest perceived monofilament that is considered `normal', with higher monofilament forces considered `abnormal', or the number of correct choices for the FCT method defined as `normal' with lower number of correct choices defined as `abnormal'. TA cold TA hot VPT 31.5Hz VPT 125Hz Monofilaments

Lowest force perceived Normal Abnormal 0.07g [0.2g [2g [4g [200g >0.07g >0.2g >2g >4g >200g Sens 99 57 6 4 0 96 89 82 22 15 7 5 2 2 Spec Sens Spec Sens Spec Sens Spec RMT testing strategy used for monofilaments 7 99 8 99 6 100 7 63 60 67 60 61 66 63 97 6 97 10 98 11 98 99 3 99 5 99 4 99 100 0 100 0 100 0 100 RTT testing strategy used for monofilaments 12 95 12 97 11 98 24 89 25 95 24 94 33 83 35 33 91 91 86 25 89 39 90 43 93 16 95 28 96 29 96 10 97 18 98 18 97 6 97 11 98 12 99 1 99 1 99 2 99 1 99 1 99 2 FCT testing strategy used for monofilaments 34 76 37 77 35 80 54 59 59 61 55 58 65 48 72 51 69 46 75 39 81 42 78 38 83 26 84 31 84 28 12 24 34 91 97 98 99 99 99

[0.07g >0.07g [0.135g >0.135g [0.2g >0.2g [1.1g >1.1g [2g >2g [3g >3g [4g >4g [102g >102g [200g >200g Correct number of choices Normal Abnormal 20 <20 m19 <19 m18 <18 m17 <17 m16 <16

73 52 38 31 25

36 55 67 77 83

Two different ways of categorising the data into four or five sub-cohorts of increasing abnormality were carried out. Firstly by means of summing the number of abnormal VPT or TPT tests (normal=0; abnormal=1) according to Seah [22] for each individual site (table 3) and secondly by adding the z-scores for the four components of VPT and TPT measurements and dividing into quartiles (table 4). Subgroups were numbered with increasing abnormality. Chi-square analysis showed a contingency coefficient of 0.742 (p<0.0001) between these two means of categorisation. Table 3 shows the division for discriminating between those who are `normal' (subgroup 0 i.e. none of the

7

vibrotactile, or thermal perception tests are abnormal) and those who are `significantly abnormal' (subgroup 4 i.e. all four of the vibrotactile and thermal perception tests are abnormal). This leads to the definition of a normal group containing 176 measurements and a significantly abnormal group with 41 measurements. Table 5 shows the calculated sensitivities and specificities for the three monofilament strategies in defining the normal (table 3, subgroup 0) from the significantly abnormal group (table 3, subgroup 4). For the RMT and RTT strategies, the sensitivities are calculated for each of the lowest perceived force (monofilament) as being a breakpoint between `normality' and `significant abnormality'. For the FCT method they are calculated for the total number of correct responses. For example in the RMT strategy, where the lowest perceived force of 0.2g is defined as `normal' and only being able to detect a higher perceived force as suggesting a significantly abnormality, the highest levels of both sensitivity (78%) and specificity (74%) are found, with sensitivity dropping very significantly if the 2g monofilament is defined as normal. For the more complex RTT testing strategy, where intermediate `perceived' forces can be defined for a test site between the 5 monofilaments, there is a strong reciprocal change in the levels of sensitivity and specificity when defining the 0.2g force or the next 1.1g intermediate force as the lower limit of normal perception. Similar results are found in table 6 where the `normal' and significantly `abnormal group' are defined without resorting to published normative values for VPT and TPT measurements. The data in tables 2, 5 and 6 suggests that monofilaments may be better in discriminating between significant neurosensory deficit compared to normality. The more complex RTT testing strategy seems to give higher sensitivities and lower specificities than the RMT strategy; but both may show better discrimination with a set of monofilaments over a smaller force range. The FCT strategy seems to offer poorer relative discrimination between `normal' `and significant abnormality' than the RMT and FCT test.

Table 3. Tabulation of medians and 25-75 percentiles for VPT and TPT measurements categorised by summation of normality or abnormality in the four VPT and TPT measurements.

Subgroup0 All tests normal 0.34 0.20-0.62 0.16 0.11-0.22 39.6 36.9-41.2 26.9 25.0-28.9 176 Subgroup1 Subgroup2 Subgroup3 Subgroup4 All tests abnormal 6.55 2.83-14.91 1.51 1.00-3.29 52.4 49.6-55.0 14.1 5.2-17.3 41

VPT 125 Hz VPT 31.5Hz TPT hot TPT cold

Median IQR Median IQR Median IQR Median IQR N

0.54 0.32-1.13 0.21 0.17-0.28 43.0 39.6-45.7 24.4 21.3-26.7 114

0.88 0.52-1.88 0.28 0.18-0.45 46.4 43.1-51.6 19.9 15.3-23.4 91

3.29 1.51-5.64 0.78 0.58-1.08 47.0 43.9-51.1 21.3 15.8-24.7 36

Subgroup 0 defined as having all tests within normal ranges; subgroup 1 has one test outside normal limits; subgroup 2 has two tests outside normal limits; subgroup 3 has three tests outside normal limits 8

and subgroup 4 had all four tests outside normal limits (using the cut-offs for normality published by Seah [22]). IQR represents the interquartile range (range over which 25-75% of data lies).

Table 4. Tabulation of medians and 25-75 percentiles for VPT and TPT measurements categorised by quartiles of summed z-scores for VPT and TPT measurements.

Subgroup 1 Subgroup 2 Categorised by quartiles of summed o z-scores for VPT and TPT VPT 125 Hz Median 0.23 0.47 IQR 0.15-0.36 0.31-0.74 VPT 31.5Hz Median 0.14 0.19 IQR 0.10-0.17 0.15-0.24 TPT hot Median 38.5 41.1 IQR 36.0-40.3 38.7-43.5 TPT cold Median 28.2 25.2 IQR 26.6-29.3 23.4-26.9 N 114 115 Subgroup 3 0.88 0.50-1.65 0.27 0.20-0.42 43.4 40.6-48.3 23.0 19.4-25.4 115 Subgroup 4 2.89 1.08-6.06 0.75 0.39-1.52 51.0 46.2-53.0 16.5 6.4-23.8 114

Subgroups defined by calculation of summed z-scores for the four tests and dividing the data into four quartiles. Subgroup 1 are those measurements in the first quartile, subgroup 2 contain the measurements in the second quartile, subgroup 3 contain the measurements in the third quartile and subgroup 4 contain the most extreme values in the fourth quartile. IQR represents the interquartile range (range over which 25-75% of data lies).

Table 5. Tabulation of sensitivity (sens) and specificity (spec) for each monofilament testing strategy comparing a normal group (n=176) defined by all four VPT and TPT tests being normal by the data of Seah [22] and a significantly abnormal group (n=41) defined by abnormality in all four vibrotactile and thermal perception tests

RMT strategy Lowest perceived force defined as abnormal >0.07g >0.2g >2g >4g >200g Sens % Spec % RTT strategy Lowest perceived force defined as abnormal >0.07g >0.135g >0.2g >1.1g >2g >3g >4g >102g >200g Sens % Spec % FCT strategy Number of correct choices defined as abnormal <20 <19 <18 <17 <16 <15 Sens % Spec %

100 78 17 10 0

11 74 99 99 100

98 95 90 42 34 22 10 0 0

16 30 43 94 98 99 99 99 99

76 54 44 37 29 24

43 63 75 84 86 92

9

Table 6. Tabulation of sensitivity (sens) and specificity (spec) for each monofilament testing strategy comparing the lowest quartile of total z scores for all four VPT and TPT tests with results for the highest quartile.

RMT strategy Lowest perceived force defined as abnormal >0.07g >0.2g >2g >4g >200g Sens Spec RTT strategy Lowest perceived force defined as abnormal >0.07g >0.135g >0.2g >1.1g >2g >3g >4g >102g >200 Sens Spec FCT strategy Number of correct choices defined as abnormal <20 <19 <18 <17 <16 <15 Sens Spec

100 71 11 4 0

15 71 100 100 100

99 96 91 37 30 19 12 2 2

22 39 51 97 100 100 100 100 100

78 62 53 46 33 26

50 68 82 90 91 95

10

4

DISCUSSION

Classification and categorisation of hand-arm vibration syndrome is achieved by using the Stockholm Workshop Scales [3, 23, 24]. These have been recently adapted and published within the current HSE guidance for health surveillance for HAVS [3]. Within the scale for the sensory component of this disease it is necessary to determine whether sensory perception is reduced. This may be achieved by using standard quantitative tests such as vibration and thermal perception thresholds (VPT and TPT), which have been used widely in this area [4], have been widely investigated to establish the factors affecting precision of measurement [25-27] and, for which, standard protocols have been developed [28]. However, this type of testing has limited availability and can be costly as it is generally only available within specialist referral centres. The current work investigates the usefulness of WEST monofilaments, which are easy and cheap to use, to define reduced sensory perception in HAVS referrals. Single point perception using monofilaments have been widely reported in screening for neuropathy in diabetes [11, 14, 29], leprosy [30-32] and other conditions involving nerve injury and repair [33, 34]. Monofilament testing on the hands for carpal tunnel syndrome and other upper limb nerve entrapments [35] and, increasingly for HAVS using the WEST monofilament kit has been reported. However, their diagnostic value or place within health surveillance for the latter condition has not been fully evaluated. The original Semmes-Weinstein monofilament kit consisted of 20 filaments with force ranging from 0.0045g to 200g force, while the WEST kit uses only 5 monofilaments distributed between 0.075-200g [36]. A recent investigation, using a similar cohort to that in this study, had concluded that using the WEST five monofilament set was not particularly helpful in defining transitions between key Stockholm Workshop stages [21]. However, Stockholm Workshop staging is a clinical judgement based upon symptomology, medical history, clinical examination and quantitative testing and therefore one may not necessarily expect that a single test would relate favourably with this. The role of monofilaments in HAVS diagnosis may be in helping to define a reduction in sensory perception that would then be used with all the other relevant information to help inform staging. It has been shown that a single monofilament (10g) can be used in diabetic neuropathy with a sensitivity of 90.7% and a specificity of 63.8% for detecting neuropathy as detected by VPT [19]. In the present study we wished to investigate if there was such a filament that could be used to define a reduction in sensory perception measured by standard quantitative tests used in HAVS diagnosis (VPT and TPT). Referral for VPT and TPT testing had been built into the HAVS assessment strategy for the British coalminers compensation scheme [4, 10] and subsequent guidance for health surveillance in the UK [3]. Interestingly there has been no HAVS-based evidence proffered that within the four tests (VPT at 31.5Hz and 125 Hz and TPT cold and warm thresholds) there is redundancy or superiority of tests. In fact current wisdom is that both VPT and TPT measurements are valuable [10]. Therefore in defining subgroups where significant vibration-induced neurosensory deficit may be encountered, we have summed effects in the four VPT/TPT outcomes with equal weighting, although one

11

method involves using published normative values for the tests [22] and the other solely relies on z-scores of test data within the study population, which includes individuals with a diagnosis of HAVS. Of the two monofilament procedures (RMT and RTT) aiming to define the lowest detectable force applied to the skin, one (RTT) is more complex and capable of defining a perceived intermediate force between two applied monofilaments. Differences in calculated sensitivities and specificities for each of the common monofilament forces when applying the RMT and RTT strategies seems to suggest that the method of application of monofilaments could influence their diagnostic power. The third monofilament strategy (FCT), which used a repeated force choice to identify the `heaviest' between two different detectable monofilaments, showed poorer discrimination between normality and abnormality in the VPT/TPT tests. This form of an FCT test seems to have little value in this context. It appeared that the RMT and RTT test may be better at indicating significant vibration induced neurosensory deficit defined by all four VPT/TPT outcomes, rather than isolated abnormality in a single VPT/TPT outcome. The RMT strategy suggests that definition of the 0.2g monofilament is best at helping define normal neurosensory function from significant deficit in workers exposed to hand-arm vibration defined by VPT and TPT measurements. But there is evidence from sensitivities and specificities calculated defining both the ability to detect the 0.2 g or lower monofilament and the next higher applied force as normal, that better diagnostic power may be gained by using monofilaments intermediate 0.2g to 2g for the RMT strategy and 0.2g and 1.1 g for the RTT strategy. In the full Semmes-Weinstein monofilament set there are an additional 3 monofilaments between 0.2g and 2g. It may be useful to further investigate the usefulness of monofilaments with a more restricted range of forces around 0.2g to optimise the sensitivity and specificity. The RMT monofilament strategy suggests that the inability to detect a 0.2g monofilament is suggestive of significant vibration-induced neurosensory deficit and has the highest sensitivity and specificity of 74% and 73% respectively. Use of likelihood ratios for tests rather than sensitivity and specificity can be more informative in terms of how good a diagnostic test is and the influence of the test on the pre-test probability of poor outcome. LR+ (positive likelihood ratio) and LR- (negative likelihood ratio) are calculated as 2.70 and 0.36 respectively. These values suggest that an abnormal result (cannot detect a force down as low as 0.2g) would only have moderate value in helping ruling-in significant abnormality (an LR+ of greater than 10 rules in a poor outcome/disease; an LR- of less than 0.1 virtually rules out the disease). For example in assessing a possible HAVS case where the physician considered pretesting that there was 50% probability of significant neurosensory deficit, an abnormal monofilament RMT test (>0.2g force monofilament) would only increase the post-test probability of significant neurosensory deficit to 80%. A negative result would reduce the pre-test probability by about half, i.e. a post-test probability of around 22%. In cases where the pre-test probability of significant deficit in a worker was considerably lower e.g. 5%, being used more as a screening tool in vibration exposed cohorts, the same abnormal test result would only give an increase in post-test probability to around 30% i.e. the chances that this worker has significant neurosensory deficit was only

12

about 1 in 3. Therefore, this analysis suggests that if identification with high probability of significant vibration-induced neurosensory deficit defined by more complex TPT/VPT referral testing was sought by simple, non-referral office-type testing procedure, either ways of improving monofilament testing as described above need to be investigated or current monofilament testing needs to be combined with some additional quantitative test or questionnaire outcome. In conclusion, the use of WEST monofilaments to define the lowest perceivable force applied to a fingertip may have some limited value in defining neurosensory deficit in vibration exposed subjects which is substantiated by the current practise of VPT and TPT referral testing. Currently, random application (RMT) of the five WEST monofilaments and defining abnormality as not being able to detect an applied force of 0.2g or lower has the best, but limited diagnostic power. Use of five monofilaments with a restricted force range especially above the 0.2g level could lead to better definition of, and possibly superior diagnostic power with either the RMT or RTT testing strategy. Combination of monofilament testing with additional screening tests may be able to offer a better probability of correctly identifying those with significant vibration-induced neurosensory deficit without the need for referral VPT and TPT testing. Definition of a subject with significant vibration induced neurosensory deficit may help the occupational health physician to stage the subject according to the Stockholm workshop scale when combined with other information including symptom history, occupational history and clinical examination.

13

5

REFERENCES

1. Mason, H. and K. Poole, Evidence review on the clinical testing and management of individuals exposed to hand-transmitted vibration. 2004, Faculty of Occupational Medicine: London. p. 1-208.

2. Health and Safety Executive, Hand-Arm Vibration. HSG(88). 2001: HSE Books, Suffolk. 3. Health and Safety Executive, Guidance on regulations: Hand -arm vibration. The control of vibration at work regulations. L140. L140. 2005, Sudbury, Suffolk: HSE Books. 1144. 4. Lawson, I.J. and K.L. McGeoch, A medical assessment process for a large volume of medico-legal compensation claims for hand-arm vibration syndrome. Occup Med (Lond), 2003. 53(5): p. 302-8. 5. McGeoch, K. and W. Gilmour, Sensorineural Objective Tests in the Assessment of Hand-Arm Vibration Syndrome. Occupational and Environmental Medicine., 1994. 51(1): p. 57-61. 6. Pelmear, P.L. and R. Kusiak, Clinical assessment of hand-arm vibration syndrome. Nagoya J Med Sci, 1994. 57 Suppl: p. 27-41. 7. Lawson, I.J. and K.L. McGeoch, Objective tests exist to aid diagnosis of hand-arm vibration syndrome. BMJ, 1996. 313(7065): p. 1148. 8. Kent, D.C., et al., Clinical evaluation of hand-arm-vibration syndrome in shipyard workers: sensitivity and specificity as compared to Stockholm classification and vibrometry testing. Conn Med, 1998. 62(2): p. 75-83. 9. Allen, J., S. McGrann, and K. McKenna, Use of questionnaire screening for vibration white finger in a high risk industrial population. International Archives of Occupational and Environmental Health, 2002. 75(1-2): p. 37-42. 10. McGeoch, K.L., et al., Use of sensorineural tests in a large volume of medico-legal compensation claims for HAVS. Occup Med (Lond), 2004. 54(8): p. 528-34. 11. Kumar, S., et al., Semmes-Weinstein monofilaments: a simple, effective and inexpensive screening device for identifying diabetic patients at risk of foot ulceration. Diabetes Res Clin Pract, 1991. 13(1-2): p. 63-7. 12. Mayfield, J.A. and J.R. Sugarman, The use of the Semmes-Weinstein monofilament and other threshold tests for preventing foot ulceration and amputation in persons with diabetes. J Fam Pract, 2000. 49(11 Suppl): p. S17-29. 13. Meijer, J.W., et al., Diabetic neuropathy examination: a hierarchical scoring system to diagnose distal polyneuropathy in diabetes. Diabetes Care, 2000. 23(6): p. 750-3. 14. Pham, H., et al., Screening techniques to identify people at high risk for diabetic foot ulceration: a prospective multicenter trial. Diabetes Care, 2000. 23(5): p. 606-11.

14

15. Resnick, H.E., et al., Diabetes, peripheral neuropathy, and old age disability. Muscle Nerve, 2002. 25(1): p. 43-50. 16. Ziegler, D., P. Mayer, and F.A. Gries, Evaluation of thermal, pain, and vibration sensation thresholds in newly diagnosed type 1 diabetic patients. J Neurol Neurosurg Psychiatry, 1988. 51(11): p. 1420-4. 17. Ziegler, D., et al., Assessment of small and large fiber function in long-term type 1 (insulin-dependent) diabetic patients with and without painful neuropathy. Pain, 1988. 34(1): p. 1-10. 18. Ziegler, D., et al., The natural course of peripheral and autonomic neural function during the first two years after diagnosis of type 1 diabetes. Klin Wochenschr, 1988. 66(21): p. 1085-92. 19. Paisley, A.N., et al., A comparison of the Neuropen against standard quantitative sensory-threshold measures for assessing peripheral nerve function. Diabet Med, 2002. 19(5): p. 400-5. 20. Kamei, N., et al., Effectiveness of Semmes-Weinstein monofilament examination for diabetic peripheral neuropathy screening. J Diabetes Complications, 2005. 19(1): p. 47-53. 21. Harris-Roberts, J., K. Poole, and H. Mason, Semmes-Weinstein monofilaments in the staging of HAVS. Report MU/06/02. 2006. 22. Seah, S.A. and M.J. Griffin, Normal values for thermotactile and vibrotactile thresholds in males and females. Int Arch Occup Environ Health, 2008. 81(5): p. 535-43. 23. Brammer, A. and W. Taylor, Sensorineural Stages of Hand-Arm Vibration Syndrome. Scandinavian Journal of Work, Environment and Health, 1987. 13(4): p. 279-283. 24. Gemne, G., et al. Stockholm Workshop 94. Hand-arm vibration syndrome: Diagnostics and quantitative relationships to exposure - Proceedings. 1995. Stockholm: Arbete Och Halsa. 25. Whitehouse, D. and M. Griffin, A comparison of vibrotactile thresholds obtained using different diagnostic equipment: the effect of contact conditions. International Archives of Occupational and Environmental Health, 2002. 75(1-2): p. 85-9. 26. Maeda, S. and M.J. Griffin, A comparison of vibrotactile thresholds on the finger obtained with different equipment. Ergonomics, 1994. 37(8): p. 1391-406. 27. Harada, N. and M.J. Griffin, Factors influencing vibration sense thresholds used to assess occupational exposures to hand transmitted vibration. Br J Ind Med, 1991. 48(3): p. 18592. 28. Lindsell, C. and M. Griffin, Standardised diagnostic methods for assessing components of the hand-arm vibration syndrome. 1997, ISVR: Southampton. 29. McGill, M., et al., Possible sources of discrepancies in the use of the SemmesWeinstein monofilament. Impact on prevalence of insensate foot and workload requirements. Diabetes Care, 1999. 22(4): p. 598-602. 30. van Brakel, W.H., et al., Evaluation of sensibility in leprosy--comparison of various clinical methods. Lepr Rev, 1994. 65(2): p. 106-21.

15

31. Kuipers, M. and T. Schreuders, The predictive value of sensation testing in the development of neuropathic ulceration on the hands of leprosy patients. Lepr Rev, 1994. 65(3): p. 253-61. 32. Feenstra, W., et al., Can people affected by leprosy at risk of developing plantar ulcers be identified? A field study from central Ethiopia. Lepr Rev, 2001. 72(2): p. 151-7. 33. Jerosch-Herold, C., Assessment of sensibility after nerve injury and repair: a systematic review of evidence for validity, reliability and responsiveness of tests. J Hand Surg [Br], 2005. 30(3): p. 252-64. 34. Jerosch-Herold, C., A study of the relative responsiveness of five sensibility tests for assessment of recovery after median nerve injury and repair. J Hand Surg [Br], 2003. 28(3): p. 255-60. 35. Amirjani, N., et al., The impact of ulnar nerve compression at the elbow on the hand function of heavy manual workers. Neurorehabil Neural Repair, 2003. 17(2): p. 118-23. 36. Weinstein, S., Fifty years of somatosensory research: from the Semmes-Weinstein monofilaments to the Weinstein Enhanced Sensory Test (WEST). J Hand Ther, 1993. 6(1): p. 11-22.

16

17

Published by the Health and Safety Executive

06/09

Health and Safety Executive

The value of the WEST monofilaments in detecting neurosensory deficit caused by hand-arm vibration exposure

Hand-arm vibration syndrome is categorised according to the Stockholm Workshop scales. This comprises of a rating system for vascular (circulatory system) and sensorineural (touch and sensation) symptoms, which sufferers of this disease can experience. To establish the extent of the sensory component of this disease these scales require the Occupational Health Physician to decide whether they feel an individual has reduced sensory perception. Quantitative tests such as vibrotactile (VPT) and thermal perception threshold (TPT) measurements have been used widely for this, but are generally only available in specialist referral centres. Simple techniques such as monofilaments are cheaper to use and could potentially be used more widely than the specialist quantitative tests. However, it is unclear at present what method of application should be used and how diagnostically useful these are. This study investigates the value of monofilaments in defining neurosensory abnormality caused by excessive exposure to hand-arm vibration by establishing their ability to detect abnormality determined by the standard quantitative tests. This report and the work it describes were funded by the Health and Safety Executive (HSE). Its contents, including any opinions and/or conclusions expressed, are those of the authors alone and do not necessarily reflect HSE policy.

RR712

www.hse.gov.uk

Information

The value of the WEST monofilaments in detecting neurosensory deficit caused by hand-arm vibration exposure

26 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

192208


You might also be interested in

BETA
The value of the WEST monofilaments in detecting neurosensory deficit caused by hand-arm vibration exposure